27

FOOD
SAFETY
AND
QUALITY
SERIES
ISSN 2415-1173

FAO/WHO
BACKGROUND DOCUMENT
ON THE RISKS AND BENEFITS
OF FISH CONSUMPTION
FAO/WHO
BACKGROUND DOCUMENT
ON THE RISKS AND BENEFITS
OF FISH CONSUMPTION

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS

WORLD HEALTH ORGANIZATION
ROME, 2024
Required citation:
FAO & WHO. 2024. FAO/WHO background document on the risks and benefits of fish consumption.
Food Safety and Quality Series, No. 27. Rome. https://doi.org/10.4060/cd1548en

The designations employed and the presentation of material in this information product do not imply the expression
of any opinion whatsoever on the part of the Food and Agriculture Organization of the United Nations (FAO) or
the World Health Organization (WHO) concerning the legal or development status of any country, territory, city
or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The mention of specific
companies or products of manufacturers, whether or not these have been patented, does not imply that these have
been endorsed or recommended by FAO or WHO in preference to others of a similar nature that are not mentioned.

The views expressed in this information product are those of the author(s) and do not necessarily reflect the views
or policies of FAO or WHO.

ISSN 2415-1173 [Print]

ISSN 2664-5246 [Online]

ISBN 978-92-5-138944-7 [FAO]

ISBN (WHO) 978-92-4-009688-2 (electronic version)
ISBN (WHO) 978-92-4-009689-9 (print version)
© FAO and WHO, 2024

Some rights reserved. This work is made available under the Creative Commons Attribution-NonCommercial-
ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo/legalcode).

Under the terms of this licence, this work may be copied, redistributed and adapted for non-commercial purposes, provided
that the work is appropriately cited. In any use of this work, there should be no suggestion that FAO or WHO endorses
any specific organization, products or services. The use of the FAO or WHO logo is not permitted. If the work is adapted,
then it must be licensed under the same or equivalent Creative Commons licence. If a translation of this work is created, it
must include the following disclaimer along with the required citation: “This translation was not created by the Food and
Agriculture Organization of the United Nations (FAO) or the World Health Organization (WHO). Neither FAO nor WHO
is responsible for the content or accuracy of this translation. The original English edition shall be the authoritative edition.

Disputes arising under the licence that cannot be settled amicably will be resolved by mediation and arbitration as described in
Article 8 of the licence except as otherwise provided herein. The applicable mediation rules will be the mediation rules of the
World Intellectual Property Organization http://www.wipo.int/amc/en/mediation/rules and any arbitration will be conducted
in accordance with the Arbitration Rules of the United Nations Commission on International Trade Law (UNCITRAL).

Third-party materials. Users wishing to reuse material from this work that is attributed to a third party, such as tables, figures or
images, are responsible for determining whether permission is needed for that reuse and for obtaining permission from the copyright
holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user.

Sales, rights and licensing. FAO information products are available on the FAO website (www.fao.org/publications)
and can be purchased through [email protected]. Requests for commercial use should be submitted via:
www.fao.org/contact-us/licence-request. Queries regarding rights and licensing should be submitted to: [email protected].

Cover photos (from left to right): © FAO/Luis Tato, © FAO/Giulio Napolitano

Layout: Tomaso Lezzi
PREPARATION OF THE DOCUMENT
Since the last FAO/WHO (Food and Agriculture Organization of the United
Nations/World Health Organization) Expert Consultation on the Risks and Benefits
of Fish Consumption was held in 2010, new evidence has become available in this
arena, and the two organizations decided, in agreement with the Codex Committee
on Fish and Fishery Products, to update the conclusions and recommendations
of the previous consultation. The Norwegian Institute of Marine Research was
commissioned to conduct a systematic literature review to inform the 2023 Joint
FAO/WHO Expert Consultation on the Risks and Benefits of Fish Consumption,
held in Rome. The report resulting from this expert consultation will be published
as a separate document.
Esther Garrido Gamarro, of the Secretariat, coordinated the work, with support
from Jogeir Toppe, Juliana De Oliveira Mota, Vittorio Fattori, Moez Sanaa, Markus
Lipp, Molly Ahner and Angeliki Vlachou, also of the Secretariat. The document was
edited by Dianne Berest. Layout was provided by Tomaso Lezzi.

iii
 iv
© FAO/Ines Gonsalves
CONTENTS

Preparation of the document..................................................................................................iii

Contributors.......................................................................................................................... xvi
Acknowledgements..............................................................................................................xvii
Abbreviations......................................................................................................................xviii
Executive summary................................................................................................................ xx
Background..........................................................................................................................xxiii

CHAPTER 1
INTRODUCTION....................................................................................................... 1
1.1 Fish as food................................................................................................................... 1
1.2 Benefits of fish consumption....................................................................................... 1
1.2.1 Nutrients........................................................................................................... 1
1.2.2 Health benefits.................................................................................................. 2
1.3 Risks of fish consumption............................................................................................ 2
1.3.1 Dioxins and dioxin-like polychlorinated biphenyls..................................... 2
1.3.2 Methylmercury................................................................................................. 5
1.3.3 Selenium and methylmercury......................................................................... 6

CHAPTER 2
METHODS ............................................................................................................... 9
2.1 Methods for the review “Evidence of health benefits from fish consumption”... 10
2.1.1 2022 VKM report: Benefit-and-risk assessment of fish in the
Norwegian diet............................................................................................... 10
2.1.2 Literature search strategy.............................................................................. 11
2.1.3 Literature search............................................................................................. 13
2.1.4 Selection of articles from the literature search............................................. 14
2.1.5 Quality assessment (risk of bias)................................................................... 14
2.1.6 Extraction of data from the included literature........................................... 17
2.1.7 Summary of the literature reviewed............................................................. 17
2.1.8 Weight of evidence of the literature.............................................................. 18
2.2 Methods for the review “Toxic effects of dioxins and dl- polychlorinated
biphenyls”................................................................................................................... 19
2.2.1 Literature search strategy.............................................................................. 19
2.2.2 Literature search............................................................................................. 21
2.2.3 Selection of studies from the literature search............................................. 21
2.2.4 Quality assessment (risk of bias)................................................................... 23
2.2.5 Extraction of data from the literature........................................................... 24
2.2.6 Summarization of the included literature..................................................... 25
2.3 Methods for the review “Toxic effects of MeHg”................................................... 25
2.3.1 Literature search strategy.............................................................................. 25
2.3.2 Literature search............................................................................................. 26
2.3.3 Selection of studies from the literature search............................................. 27

v
 2.3.4 Quality assessment (risk of bias)................................................................... 29
2.3.5 Extraction of data from the included literature........................................... 29
2.3.6 Summarization of the literature.................................................................... 30
2.4 Methods for the review “The role of selenium with regard to the health
effects of MeHg”......................................................................................................... 30
2.4.1 Literature search strategy.............................................................................. 30
2.4.2 Literature search............................................................................................. 31
2.4.3 Selection of studies from the literature search............................................. 32
2.4.4 Quality assessment (risk of bias)................................................................... 33
2.4.5 Extraction of data from the literature........................................................... 33
2.4.6 Summarization of the literature included..................................................... 33
2.4.7 Weight of evidence of the literature.............................................................. 34
2.5 Methods for the review “Occurrence data for MeHg, dioxins and dl-PCBs”...... 34
2.5.1 Literature search strategy.............................................................................. 34
2.5.2 Literature search............................................................................................. 34
2.5.3 Selection of studies from the literature search............................................. 36
2.5.4 Quality assessment......................................................................................... 36
2.5.5 Extraction of data........................................................................................... 37
2.5.6 Data from public databases........................................................................... 38

CHAPTER 3
RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW “EVIDENCE
OF HEALTH BENEFITS FROM FISH CONSUMPTION” 43
3.1 Literature search and quality assessment.................................................................. 43
3.1.1 Allergy and immunology............................................................................... 46
3.1.2 Birth and growth outcomes........................................................................... 46
3.1.3 Bone health..................................................................................................... 47
3.1.4 Cancer............................................................................................................. 48
3.1.5 Cardiovascular diseases and outcomes......................................................... 49
3.1.6 Type 2 diabetes............................................................................................... 49
3.1.7 Neurodevelopment and neurological disorders.......................................... 50
3.1.8 Mortality......................................................................................................... 51
3.1.9 Overweight and obesity................................................................................. 52
3.2 Results and summarization of the included literature............................................. 53
3.2.1 Allergy and immunology............................................................................... 53
3.2.2 Birth and growth outcomes........................................................................... 65
3.2.3 Bone health..................................................................................................... 72
3.2.4 Cancer............................................................................................................. 78
3.2.5 Cardiovascular diseases and outcomes......................................................... 90
3.2.6 Type 2 diabetes............................................................................................. 112
3.2.7 Neurodevelopment and neurological disorders........................................ 119
3.2.8 Mortality....................................................................................................... 127
3.2.9 Overweight and obesity in adults............................................................... 137
3.3 Final weight of evidence for “Health benefits of fish consumption”.................. 141

vi
CHAPTER 4
RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW “TOXIC EFFECTS
OF DIOXINS AND dl-PCBs”...................................................................................147
4.1 Literature search and quality assessment................................................................ 147
4.1.1 Systematic reviews........................................................................................ 147
4.1.2 Primary studies............................................................................................. 148
4.2 Results and summarization of the literature included........................................... 148
4.2.1 Chloracne and other dermal effects............................................................ 148
4.2.2 Reproductive effects (including organs)..................................................... 148
4.2.3 Female reproductive effects......................................................................... 155
4.2.4 Birth outcomes............................................................................................. 155
4.2.5 Thyroid disease and thyroid hormones..................................................... 156
4.2.6 Type 2 diabetes and obesity......................................................................... 160
4.2.7 Cardiovascular effects.................................................................................. 163
4.2.8 Hepatic disorders and digestive effects...................................................... 163
4.2.9 Effects on the immune system.................................................................... 163
4.2.10 Effects on the nervous system..................................................................... 164
4.2.11 Effects on teeth and bones........................................................................... 166
4.2.12 Cancer........................................................................................................... 166
4.2.13 Other effects................................................................................................. 169

CHAPTER 5
RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW “TOXIC EFFECTS
OF MeHg”............................................................................................................171
5.1 5.1 Literature and quality assessment......................................................................???
5.1.1 Systematic reviews........................................................................................ 171
5.1.2 Primary studies............................................................................................. 172
5.2 Results and summarization of the literature included........................................... 173
5.2.1 Neurological outcomes................................................................................ 173
5.2.2 Cardiovascular outcomes............................................................................ 189
5.2.3 Growth.......................................................................................................... 195
5.2.4 Other health outcomes................................................................................ 201

CHAPTER 6
RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW “THE ROLE
OF SE WITH REGARD TO THE HEALTH EFFECTS OF MeHg”.....................................213
6.1 6.1 Literature and quality assessment......................................................................???
6.1.1 Human studies.............................................................................................. 213
6.1.2 Animal studies.............................................................................................. 214
6.2 Results and summarization of the literature included........................................... 215
6.2.1 Description of the literature........................................................................ 215
6.2.2 Cardiovascular outcomes............................................................................ 216
6.2.3 Oxidative stress............................................................................................ 218
6.2.4 Immune system............................................................................................ 220
6.2.5 Reproduction................................................................................................ 222

vii
 6.2.6 Thyroid hormones....................................................................................... 223
6.2.7 Birth outcomes............................................................................................. 224
6.2.8 Neurodevelopment and cognition.............................................................. 225
6.2.9 Vision function............................................................................................. 227
6.2.10 Motor function............................................................................................. 229
6.3 Final weight of evidence for the review “The role of Se with regard
to the health effects of MeHg”................................................................................ 230

CHAPTER 7
RESULTS AND SUMMARIZATION OF “OCCURRENCE DATA FOR MeHg AND,
DIOXINS AND dl-PCBs”........................................................................................233
7.1 Literature search....................................................................................................... 233
7.2 Quality assessment of the literature included from the literature search............ 233
7.3 Overview of concentration data for total Hg, MeHg, dioxins and
dioxin-like polychlorinated biphenyls.................................................................... 234
7.3.1 Overview of concentration data from the literature search...................... 235
7.3.2 Overview of concentration data from the database of the
European Food Safety Authority............................................................... 235
7.4 Mercury in finfish and shellfish from different regions around the world......... 236
7.4.1 Mercury in finfish from the literature review............................................ 236
7.4.2 Mercury in finfish from data from the European Food
Safety Authority........................................................................................... 251
7.4.3 Mercury in finfish, summary....................................................................... 259
7.4.4 Mercury in shellfish from the literature review......................................... 259
7.4.5 Mercury in shellfish from data from the European Food
Safety Authority........................................................................................... 266
7.4.6 Mercury in shellfish, summary................................................................... 272
7.5 Dioxins and dioxin-like polychlorinated biphenyls in finfish and shellfish
from different regions around the world................................................................ 273
7.5.1 Dioxins and dioxin-like polychlorinated biphenyls in finfish from
the literature review..................................................................................... 273
7.5.2 Dioxins and dioxin-like polychlorinated biphenyls from data from
the European Food Safety Authority......................................................... 278
7.5.3 Dioxins and dioxin-like polychlorinated biphenyls in finfish,
summary........................................................................................................ 289
7.5.4 Dioxins and dioxin-like polychlorinated biphenyls in shellfish
from the literature review............................................................................ 290
7.5.5 Dioxins and diolin-like polychlorinated biphenyls in shellfish
from data from the European Food Safety Authority.............................. 291
7.5.6 Dioxins and dioxin-like polychlorinated biphenyls in shellfish,
summary........................................................................................................ 298

REFERENCES.......................................................................................................301

v iii
APPENDICES

1. TERMS OF REFERENCE PROVIDED BY FAO/WHO...............................................329

2. QUALITY ASSESSMENT TOOLS (RISK OF BIAS)..................................................331

3. EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION............................336

4. TOXIC EFFECTS OF DIOXINS AND dl-PCBs.......................................................419

5. TOXIC EFFECTS OF MeHg................................................................................438

6. SE AND MeHg..................................................................................................450

7. OCCURRENCE DATA OF MeHg AND DIOXINS AND dl-PCBs.................................475

ix
 TABLES

1.1 Dioxins and dioxin-like polychlorinated biphenyls with toxicity equivalence

factor (TEF) set by the World Health Organization in 2005...................................... 4
2.1 Population, intervention, comparison, outcomes and study design for the
literature search on “Evidence of health benefits from fish consumption”............. 12
2.2 Categories of health outcomes included in the literature search on “Evidence
of health benefits from fish consumption”.................................................................. 13
2.3 Inclusion and exclusion criteria for the literature search on “Evidence of health
benefits of fish consumption”....................................................................................... 15
2.4 Criteria of the World Cancer Research Fund for weighing the evidence................. 18
2.5 Population, exposure, comparison, outcomes and study designs (PECOS)
for the literature search on “Toxic effect of dioxins and dl-PCBs”.......................... 20
2.6 Inclusion and exclusion criteria for the literature search on “Toxic effects of
dioxins and dl-PCBs”.................................................................................................... 22
2.7 Risk-of-bias categories.................................................................................................. 24
2.8 Overall risk-of-bias judgment...................................................................................... 24
2.9 Population, exposure, comparison, outcomes and study designs (PECOS)
for the literature search on “Toxic effects of MeHg”................................................. 26
2.10 Inclusion and exclusion criteria for the literature search on “Toxic effects of
MeHg”............................................................................................................................ 28
2.11 Population, intervention, comparison, outcomes and study designs for literature
search on “The role of selenium with regard to the health effects of MeHg”.......... 31
2.12 Inclusion and exclusion criteria for the literature search on “The role of selenium
with regard to the health effects of MeHg”................................................................. 32
2.13 Inclusion and exclusion criteria for the literature search on “Occurrence data for
MeHg, dioxins and dl-PCBs”....................................................................................... 35
2.14 Questions for quality assessment of articles, adapted from the European Food
Information Resource................................................................................................... 37
2.15 Inclusion and exclusion criteria for data, from the European Food Safety
Authority database........................................................................................................ 38
3.1 Results of the literature search on “Evidence of health benefits from fish
consumption”................................................................................................................. 45
3.2 Results from the systematic reviews from the literature search on “Allergy
and immunology”.......................................................................................................... 60
3.3 Summary of results from original primary studies (cohort studies) included
from the literature search on “Birth and growth outcomes”..................................... 70
3.4 Summary of results from original primary studies (intervention studies)
included from the literature search on “Bone health” for “Evidence of health
benefits of fish consumption”....................................................................................... 75
3.5 Summary of results from original primary studies (cohort studies) included
from the literature search on “Bone health”............................................................... 76
3.6 Summary of results from systematic reviews included from the literature
search on “Cancer”........................................................................................................ 80

x
3.7 Summary of results from original primary studies (cohort studies) included
from the literature search on “Cancer”........................................................................ 84
3.8 Summary of results from systematic reviews included from the literature
search on “CVD outcomes”......................................................................................... 96
3.9 Summary of results from original primary studies (cohort studies) included
from the literature search in the theme “CVD outcomes”...................................... 103
3.10 Summary of results from systematic reviews included from the literature
search on “Type 2 diabetes”........................................................................................ 114
3.11 Summary of results from original primary study (cohort study) included
from the literature search on “Type 2 diabetes”....................................................... 118
3.12 Summary of results from original primary study (intervention study) included
from the literature search on “Neurodevelopment and neurological disorders”..... 125
3.13 Summary of results from original primary study (cohort study) included
from the literature search on “Neurodevelopment and neurological disorders”..... 126
3.14 Summary of results from systematic reviews from the literature search on
“Mortality”................................................................................................................... 132
3.15 Summary of results from original primary studies (cohort studies) included
from the literature search on “Mortality”................................................................. 135
3.16 Summary of results from original primary studies (cohort studies) from the
literature search on “Overweight and obesity”........................................................ 139
3.17 Summary of final weight of evidence for “Evidence of health benefits of fish
consumption”............................................................................................................... 141
4.1 Overview of primary studies with reproductive outcomes for the review of
“Health effects of dioxins and dl-PCBs”................................................................... 152
4.2 Overview of primary studies with birth weight and other outcomes for the
review of “Health effects of dioxins and dl-PCBs”.................................................. 157
4.3 Overview of primary studies with thyroid disease and thyroid hormone
outcomes for the review of “Health effects of dioxins and dl-PCBs”.................... 159
4.4 Overview of primary studies with type 2 diabetes and obesity outcomes for
the review of “Health effects of dioxins and dl-PCBs”........................................... 162
4.5 Overview of primary studies with nervous system outcomes for the review
of “Health effects of dioxins and dl-PCBs”.............................................................. 165
4.6 Overview of primary studies with cancer outcomes for the review of
“Health effects of dioxins and dl-PCBs”................................................................... 168
5.1 Overview of systematic reviews with neurological outcomes for the review
of “Health effects of MeHg”...................................................................................... 175
5.2 Overview of primary studies with neurological outcomes for the review of
“Toxic effects of MeHg”............................................................................................. 182
5.3 Overview of systematic reviews with cardiovascular outcomes for the review
of “Health effects of MeHg”...................................................................................... 191
5.4 Overview of primary studies with cardiovascular outcomes for the review
of “Health effects of MeHg”...................................................................................... 194
5.5 Overview of systematic reviews with growth outcomes for the review of
“Health effects of MeHg”........................................................................................... 197
5.6 Overview of primary studies with growth outcomes for the review of
“Health effects of MeHg”........................................................................................... 199

xi
 5.7 Overview of systematic reviews with other health outcomes for the review
of “Health effects of MeHg”...................................................................................... 206
5.8 Overview of primary studies with other health outcomes (n = 15) for the
review of “Health effects of MeHg”.......................................................................... 208
6.1 Distributions of the studies and publications included per investigated country..... 215
6.2 Studies included for the health outcome “cardiovascular outcomes”
for the review “Se and MeHg”................................................................................... 217
6.3 Studies included for the health outcome “oxidative stress” for the review
“Se and MeHg”............................................................................................................ 219
6.4 Studies included for the health outcome “immune system” for the review
“Se and MeHg”............................................................................................................ 221
6.5 Studies included for the health outcome “reproduction” for the review
“Se and MeHg”............................................................................................................ 222
6.6 Studies included for the health outcome “thyroid hormones” for the review
“Se and MeHg”............................................................................................................ 223
6.7 Studies included for the health outcome “birth outcomes” for the review
“Se and MeHg”............................................................................................................ 224
6.8 Studies included for the health outcome “neurodevelopment and cognition”
for the review “Se and MeHg”................................................................................... 226
6.9 Studies included for the health outcome “vision function” for the review
“Se and MeHg”............................................................................................................ 228
6.10 Study included for the health outcome “motor function” for the review
“Se and MeHg”............................................................................................................ 229
6.11 Summary of final weight of evidence for the “Role of Se with regard to the
health effects of MeHg”.............................................................................................. 230
7.1 Mean total mercury (THg) and methylmercury (MeHg) levels
(mg/kg wet weight) in muscle tissue of species of farmed finfish from
different inland regions (FAO areas)......................................................................... 237
7.2 Mean total Hg (THg) and methylmercury (MeHg) levels in muscle tissue
of species of farmed finfish from different marine regions (FAO areas),
in mg/kg wet weight.................................................................................................... 238
7.3 Mean total mercury (THg) levels in different genera of wild-caught finfish
from inland regions (FAO areas), in mg/kg wet weight.......................................... 239
7.4 Mean methylmercury (MeHg) levels in genus of wild-caught finfish from
different inland regions (FAO areas), in mg/kg wet weight................................... 242
7.5 Mean total mercury (THg) levels in tissue of different genera of wild-caught
finfish from different marine regions (FAO areas), in mg/kg wet weight............. 244
7.6 Mean methylmercury (MeHg) levels in wild-caught finfish from different
marine regions (FAO areas), by genus, in mg/kg wet weight................................. 250
7.7 Mean total mercury (THg) and methylmercury (MeHg) levels in muscle
tissue of species or species groups of farmed and wild-caught finfish from
inland waters (all regions), in mg/kg wet weight..................................................... 252
7.8 Mean total mercury (THg) levels in muscle tissue of species or species
groups of wild-caught, farmed or unspecified finfish from different inland
regions (FAO areas), in mg/kg wet weight............................................................... 253

xii
7.9 Mean total mercury (THg) and methylmercury (MeHg) levels in muscle
tissue of species or species groups of farmed finfish from different marine
(or unspecified) regions (FAO areas), in mg/kg wet weight.................................... 254
7.10 Mean total mercury (THg) levels in muscle tissue of species or species
groups of wild-caught (or unspecified) finfish from different marine
(or unspecified) regions (FAO areas), in mg/kg wet weight.................................... 256
7.11 Mean total mercury (THg) and methylmercury (MeHg) levels in muscle tissue
of species or species groups of wild-caught (or unspecified) finfish from marine
or unspecified areas around the world (all regions), in mg/kg wet weight............ 258
7.12 Mean total mercury (THg) and methylmercury (MeHg) levels in species
of farmed shellfish from inland waters, in mg/kg wet weight................................ 260
7.13 Mean total mercury (THg) levels in species of farmed shellfish from different
marine regions (FAO areas), in mg/kg wet weight.................................................. 260
7.14 Mean total mercury (THg) and methylmercury (MeHg) levels in tissues of
different species of wild-caught (or unknown) shellfish from different inland
regions (FAO areas), in mg/kg wet weight............................................................... 262
7.15 Mean total mercury (THg) and methylmercury (MeHg) levels in tissue of
different genera of wild-caught shellfish from marine waters (all regions),
in mg/kg wet weight.................................................................................................... 263
7.16 Mean total mercury (THg) levels in tissue of different genera of wild-caught
shellfish from different marine regions (FAO areas), in mg/kg wet weight.......... 265
7.17 Mean total mercury (THg) levels in species or species groups of farmed
shellfish from inland and marine waters, in mg/kg wet weight............................. 267
7.18 Mean total mercury (THg) levels in species or species groups of wild-caught
(or unspecified) shellfish from inland waters (all regions), in mg/kg wet weight.... 267
7.19 Mean total mercury (THg) levels in species or species groups of wild-caught
(or unspecified) shellfish from different inland regions (FAO areas),
in mg/kg wet weight.................................................................................................... 268
7.20 Mean total mercury (THg) and methylmercury (MeHg) levels in species
or species groups of wild-caught (or unspecified) shellfish from marine waters
(all regions), in mg/kg wet weight.............................................................................. 270
7.21 Mean total mercury (THg) levels in species or species groups of wild-caught (or
unspecified) shellfish from different marine (or unspecified) regions (FAO areas),
in mg/kg wet weight.................................................................................................... 271
7.22 Mean concentrations of sum dioxins, sum dioxin-like polychlorinated biphenyls
(dl-PCBs) and sum dioxins and dl-PCBs in farmed finfish from different inland
and marine regions (FAO areas) (ng toxic equivalent quotient/kg wet weight)...... 274
7.23 Mean concentrations of sum dioxins, sum dioxin-like polychlorinated biphenyls
(dl-PCBs) and sum dioxins and dl-PCBs in species of wild finfish from different
inland regions (FAO areas) (ng toxic equivalent/kg wet weight)............................. 275
7.24 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in muscle tissue of wild
finfish from different marine regions (FAO areas), by genus (ng toxic
equivalent quotient/kg wet weight)........................................................................... 276
7.25 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in whole body tissue of
species of wild finfish from different marine regions (FAO areas)
(ng toxic equivalent quotient/kg wet weight) .......................................................... 278

xiii
 7.26 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in muscle tissue of different
species or species groups of farmed finfish from inland waters (all regions)
(ng toxic equivalent quotient/kg wet weight)........................................................... 279
7.27 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in muscle tissue of different
species or species groups of farmed finfish from marine waters (all regions)
(ng toxic equivalent quotient/kg wet weight)........................................................... 280
7.28 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in muscle tissue of different
species or species groups of wild-caught (or unspecified) finfish from different
inland regions (FAO areas) (ng toxic equivalent quotient/kg wet weight) .......... 281
7.29 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in muscle tissue of different
species or species groups of wild (or unspecified) finfish from marine or
unspecified waters (all regions ) (ng toxic equivalent quotient/kg wet weight)........ 282
7.30 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in muscle tissue of different
species or species groups of wild-caught (or unspecified) finfish from different
marine (or unspecified) regions (FAO areas) (ng toxic equivalent
quotient/kg wet weight).............................................................................................. 284
7.31 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in tissue of farmed shellfish
from different marine regions (FAO areas) (ng toxic equivalent
quotient/kg wet weight) ............................................................................................. 290
7.32 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in tissue of wild shellfish
from different inland regions (FAO areas) (ng toxic equivalent
quotient/kg wet weight).............................................................................................. 291
7.33 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in tissue of wild shellfish
from different marine regions (FAO areas) (ng toxic equivalent
quotient/kg wet weight).............................................................................................. 292
7.34 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in various species or species
groups of farmed shellfish from marine waters (all regions) (ng toxic
equivalent quotient/kg wet weight)........................................................................... 293
7.35 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in various species or species groups
of wild (or unspecified) shellfish from inland waters (all regions)
(ng toxic equivalent quotient/kg wet weight)........................................................... 294
7.36 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in various species or species
groups of wild-caught (or unspecified) shellfish from different inland regions
(FAO areas) (ng toxic equivalent quotient/kg wet weight)..................................... 294
7.37 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls
(dl-PCBs) and the sum of dioxins and dl-PCBs in various species or species
groups of wild-caught (or unspecified) shellfish from marine or unspecified
waters (all regions) (ng toxic equivalent quotient/kg wet weight).......................... 295
7.38 Mean concentrations of dioxins, dioxin-like polychlorinated biphenyls (dl-PCBs)
and the sum of dioxins and dl-PCBs in various species or species groups of
wild-caught (or unspecified) shellfish from different marine (or unspecified)
regions (FAO areas) (ng toxic equivalent quotient/kg wet weight).......................... 296

xiv
FIGURES

3.1 Flow diagram for the literature review “Evidence of health benefits
from fish consumption”................................................................................................ 44
4.1 Flow diagram for the literature review “Toxic effects of dioxins and
dl-PCBs”...................................................................................................................... 149
5.1 Flow diagram for the review “Toxic effects of MeHg”............................................ 172
6.1 Flow diagram for the review “The role of Se with regards to the health
effects of MeHg”.......................................................................................................... 214
7.1 Flow diagram for the review “Occurrence data or MeHg, and dioxins
and dl-PCBs in fisheries and aquaculture products”................................................ 234

xv
 CONTRIBUTORS

This background document was prepared by the Institute of Marine Research

(IMR), in Bergen, Norway. The following authors were involved in the work:
Research director: Gro-Ingunn Hemre
Project leaders: Amund Maage and Robin Ørnsrud
Project coordinator: Synnøve Næss Sleire
Authors (in alphabetical order): Inger Aakre, Atabak Azad, Annette Bernhard,
Lisbeth Dahl, Carey Donald, Even Fjære, Sylvia Frantzen, Rita Hannisdal,
Hilde Elise Heldal, Quang Ho Tri, Marian Kjellevold, Tanja Kögel, Kai Kristoffer Lie,
Amund Maage, Maria Wik Markhus, Lene Secher Myrmel, Bente M. Nilsen, Ole
Jakob Nøstbakken, Josef Daniel Rasinger, Monica Sanden, Synnøve Næss Sleire,
Elin Sørhus, Martin Wiech, Yiou Mike Zhu and Jannike Øyen.
Funding: FAO and IMR (through the Norwegian Ministry of Trade, Industry
and Fisheries). The authors declare no competing interests.

SECRETARIAT:
Esther Garrido Gamarro, FAO, Italy
Jogeir Toppe, FAO, Italy
Vittorio Fattori, FAO, Italy
Juliana de Oliveira Mota, WHO, Switzerland
Moez Sanaa, WHO, Switzerland
Molly Ahern, FAO, Italy
Markus Lipp, FAO, Italy
Angeliki Vlachou, FAO, Italy

xv i
ACKNOWLEDGEMENTS

FAO and WHO would like to express their appreciation to the Norwegian Institute
of Marine Research for their collaborative spirit and openness throughout the
preparation of this publication.

xvii
 ABBREVIATIONS

ADHD attention deficit hyperactivity disorder

ASD autism spectrum disorder
BMI body mass index
CHD coronary heart disease
CI confidence interval
COPD chronic obstructive pulmonary disease
CVD cardiovascular disease
DHA docosahexaenoic acid
dl-PCBs dioxin-like polychlorinated biphenyls
EPIC European Prospective Investigation into Cancer and Nutrition
EFSA European Food Safety Authority
FAO Food and Agriculture Organization of the United Nations
FFQ food frequency questionnaire
GDM gestational diabetes mellitus
HOMA2-IR homeostatic model assessments for insulin resistance
HOMA2-B homeostatic model assessments for beta-cell function
HR hazard ratio
I2 measurement of heterogeneity
IQ intelligence quotient
JECFA Joint FAO/WHO Expert Committee on Food Additives
LCn3PUFA long-chain n-3 polyunsaturated fatty acid
LOD limit of detection
LOQ limit of quantification
MI myocardial infarction
N number (of)
OCDD octachlorodibenzodioxin
OHAT Office of Health Assessment and Translation
PCB polychlorinated biphenyl
PCDD polychlorinated dibenzo-p-dioxin
PCDD/F polychlorinated dibenzo‐p‐dioxin and dibenzofuran

xv iii
PCDF polychlorinated dibenzofuran
PECOS Population, exposure, comparison, outcomes and study designs
PICOS Population, intervention, comparison, outcomes and study designs
PUFA polyunsaturated fatty acid
RCT randomized controlled trial
RR risk ratio
T2D type 2 diabetes
TCDD 2,3,7,8-tetrachlorodiobenzo-p-dioxin
TEF toxic equivalent factor
TEQ toxic equivalent quotient
TWI tolerable weekly intake
VKM Norwegian Scientific Committee for Food and Environment
WCRF World Cancer Research Fund
WHO World Health Organization

xix
 EXECUTIVE SUMMARY

The report of the first Joint FAO/WHO Expert Consultation on the Risks and Benefits
of Fish Consumption was published in 2010. Since then, new literature, data and
information on the subject have become available. As such, FAO and WHO decided to
generate a background report consisting of a comprehensive literature review, followed
by an expert consultation, to update the report with new scientific evidence.
This background document aims to provide scientific evidence about the risks
and benefits of fish consumption in order to update the 2010 Report of the Joint
FAO/WHO Expert Consultation on the Risks and Benefits of Fish Consumption.
To provide new scientific evidence, five extensive literature reviews were conducted,
focusing on the following five topics:
1. evidence of health benefits from fish consumption;
2. toxic effects of dioxins and dioxin-like polychlorinated biphenyls (dl-PCBs)
(from studies published since 2010);
3. toxic effects of methylmercury (MeHg) (from studies published since 2010);
4. the role of selenium (Se) with regard to the health effects of MeHg;
5. occurrence data for MeHg, dioxins and dl-PCBs in fishery and aquaculture
products (from studies published since 2010).
The reviews followed a systematic approach, performing a systematic literature
search and implementing elements from systematic literature reviews. A systematic
review attempts to identify, appraise and synthesize all the empirical evidence
that meets prespecified eligibility criteria to answer a specific research question.
When conducting systematic reviews, systematic methods are used, aiming to
minimize bias in order to produce more reliable findings to inform decision-making.
In this background document, a combined method was applied integrating
primary research, systematic reviews and other relevant, available risk and benefit
assessments. In addition, the available literature was summarized. The PRISMA
guidelines for searching, selecting and reporting on literature were followed during
the research and preparation of this document.

1. EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION

To evaluate the evidence of health benefits from fish consumption, the literature
reviewed comprised the report, Benefit and risk assessment of fish in the Norwegian
diet, published by The Norwegian Scientific Committee for Food and Environment
(VKM) in 2022, as well as literature (systematic reviews and original primary studies)
included from a systematic literature search. The VKM report was based on an
extensive systematic literature review evaluating the epidemiological evidence for
associations between fish consumption and health outcomes.

xx
A systematic literature search was performed in the databases PubMed, Web of Science
and Cochrane Library. The initial search resulted in 39 092 records. After screening the
titles and abstracts, 791 records were assessed in full text. Further, 127 records were quality
assessed with risk-of-bias tools. Thus, the final review included 1 risk-and-benefits
assessment (VKM, 2022), 22 systematic reviews, and 47 original primary studies.
The literature reviewed was further categorized into the following health outcomes:
allergy and immunology, birth and growth outcomes, bone health, cancer, cardiovascular
diseases and outcomes, type 2 diabetes, neurodevelopment and neurological disorders,
mortality, and overweight and obesity. A final weight of evidence was performed, using
the criteria of the 2018 World Cancer Research Fund (WCRF) report, Judging the
evidence, to grade the evidence for the different health outcomes into the categories:
convincing (strong evidence); probable (strong evidence); limited, suggestive;
limited, no conclusion; and substantial effect of risk unlikely (strong evidence).

2. TOXIC EFFECTS OF DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS

(FROM STUDIES PUBLISHED IN THE LAST TEN YEARS)
To evaluate the evidence of toxic effects of dioxins and dl-PCBs published in the
last ten years, the literature consisted of the report, Risk for animal and human
health related to the presence of dioxins and dioxin‐like PCBs in feed and food,
published by the European Food Safety Authority (EFSA, 2018) and literature
(original primary studies) included from a systematic literature search.
The systematic literature search was performed in the databases PubMed and
Web of Science. The initial search resulted in 2 770 records. After screening the
titles and abstracts, 396 records were assessed in full text. Further, 81 records were
quality assessed with risk-of-bias tools. Thus, the final review of the literature
included one risk assessment (EFSA, 2018) and 20 original primary studies.
The literature reviewed was further categorized into the following health outcomes:
chloracne and other dermal effects, reproductive effects, female reproductive effects,
birth outcomes, thyroid disease and thyroid hormones, type 2 diabetes and obesity,
cardiovascular effects, hepatic disorders and digestive effects, effects in the immune
system, effects in the nervous system, effects in teeth and bones, cancer, and other
effects. A final weight of evidence was not performed, as only literature published
after 2010 was included, and therefore it was not possible to give an overall grading
of the available evidence.

3. TOXIC EFFECTS OF METHYLMERCURY (FROM STUDIES PUBLISHED

IN THE LAST TEN YEARS)
To evaluate the evidence of toxic effects of MeHg published in the last ten years,
the literature consisted of the EFSA 2012 MeHg report, Scientific Opinion on the
risk for public health related to the presence of mercury and MeHg in food; the 2022
VKM report; and literature (original primary studies) included from a systematic
literature search.

xxi
 The systematic literature search was performed in the databases PubMed and
Web of Science. The initial search resulted in 1 929 records. After screening the titles
and abstracts, 396 records were assessed in full text. Further, 100 records were quality
assessed with risk-of-bias tools. Thus, the final review included 2 risk assessments
(EFSA, 2015 and VKM, 2022), 16 systematic reviews, and 20 original primary studies.
The literature reviewed was further categorized into the following health outcomes:
neurological outcomes, cardiovascular outcomes, growth, and other health outcomes.
A final weight of evidence was not performed, as only literature published after 2010 was
included, and thus it was not possible to give an overall grading of the available evidence.

4. THE ROLE OF SELENIUM WITH REGARD TO THE HEALTH EFFECTS

OF METHYLMERCURY
To evaluate the evidence of the role of Se with regard to the health effects of MeHg,
the literature review included original primary studies identified from a systematic
literature search.
The systematic literature search was performed in the databases PubMed and
Web of Science. The initial search resulted in 1 154 records. After screening the
titles and abstracts, 234 records were assessed in full text. Further, 47 records
were quality assessed with risk-of-bias tools. Finally, 45 original human primary
studies were included in the final literature review. The literature reviewed was
further categorized into the following health outcomes: cardiovascular outcomes,
oxidative stress, immune system, reproduction, thyroid hormones, birth outcomes,
neurodevelopment and cognition, vision function, and motor function. A final weight
of evidence using the report, Judging the evidence (WCRF, 2018), was performed,
grading the evidence for the different health outcomes into the categories: convincing
(strong evidence); probable (strong evidence); limited, suggestive; limited,
no conclusion; or substantial effect of risk unlikely (strong evidence).

5. OCCURRENCE DATA FOR METHYLMERCURY, DIOXINS AND DIOXIN-LIKE

POLYCHLORINATED BIPHENYLS IN FISHERY AND AQUACULTURE PRODUCTS
(FROM STUDIES PUBLISHED IN THE LAST TEN YEARS)
To evaluate the data for MeHg, dioxins and dl-PCBs in fishery and aquaculture products
published in the last ten years, data were obtained from the WHO GEMS/Food
database and the ESFA Chemical Monitoring database (both public databases) and
from literature included from a systematic literature search.
A systematic literature search was performed in the database Web of Science. The initial
search resulted in a total of 6 851 records. After screening the titles and abstracts, 2 306 records
were assessed in full text. Further, 1 252 records were quality assessed. As a result, the
final review of the literature included 554 original primary articles. Concentration data
for total Hg, MeHg and/or dioxins and dl-PCBs were extracted from these articles,
along with necessary metadata and compiled in an Excel spreadsheet. Further, data on
concentrations of MeHg, dioxins and dl-PCBs obtained from the literature review and
the data from the EFSA database, were treated and compiled separately.

xxii
BACKGROUND
At its thirty-eighth session, held in 2006, the Codex Committee on Food Additives
and Contaminants asked the Codex Alimentarius Commission to seek scientific
advice from FAO and WHO on the risks and benefits of fish consumption;
specifically, to compare the health benefits of fish consumption with the risks
associated with the contaminants MeHg and dioxins (defined here to include
polychlorinated dibenzo-p-dioxins [PCDDs], polychlorinated dibenzofurans
[PCDFs] and dioxin-like polychlorinated biphenyls [dl-PCBs]), which may be
present in fish. The request was driven by growing public concern regarding the
presence of chemical contaminants in fish and, at the same time, the increasing
clarity of the multiple nutritional benefits of including fish in the diet. In response
to the request, from 25 to 29 January 2010, FAO and WHO conducted the Expert
Consultation on the Risks and Benefits of Fish Consumption to review data on
the levels of nutrients and specific chemical contaminants (MeHg and dioxins) in
a range of fish species and to review scientific literature covering the risks and
benefits of fish consumption. The review considered the risks and benefits for
specific endpoints, including sensitive groups of the population. The findings of the
Expert Consultation were published in the Report of the Joint FAO/WHO Expert
Consultation on the Risks and Benefits of Fish Consumption (FAO and WHO,
2010). Since then, new literature, data and information on the subject have become
available. As such, FAO and WHO decided to conduct another comprehensive
literature review and prepare a background document of the findings, in order to
update the 2010 report with new scientific evidence.

xxiii
 xxi v
© FAO/Ines Go
CHAPTER 1
INTRODUCTION

1.1 FISH AS FOOD

Global fishery and aquaculture production are at a record high, and the sector will
play an increasingly important role in providing food and nutrition into the future
(FAO, 2022). Global fish consumption has more than doubled in the last 50 years,
and annual per capita consumption reached more than 20 kg in 2020 (FAO, 2022).
In addition, fish is an important food source and a part of cultural food traditions
of many people globally (FAO and WHO, 2010).
Unless otherwise stated, the term “fish” is here defined as finfish (vertebrates)
and other seafood (aquatic invertebrates, including crustaceans, molluscs and
echinoderms, most frequently consumed by humans), whether of marine or
freshwater origin, farmed or wild (FAO, 2020). Aquatic mammals, reptiles, seaweed,
algae and aquatic plants are considered outside the scope of this background
document and, thus, are excluded from it (FAO, 2020). Furthermore, sustainability
issues and environmental impacts are also considered outside the scope of this
background document.

1.2 BENEFITS OF FISH CONSUMPTION

1.2.1 NUTRIENTS
Fish is dietary source of several important nutrients, including high-quality
proteins, marine long-chain n-3 polyunsaturated fatty acids, or LC n-3 PUFAs
(eicosapentaenoic acid and docosahexaenoic acid), vitamin A, vitamin D, vitamin
B12, iodine, iron, selenium (Se), and zinc (Byrd et al., 2021). The benefits of fish
consumption have been related to the intake of these essential micronutrients and
fatty acids, as consuming seafood can potentially reduce micronutrient deficiencies
(Golden et al., 2021). Globally, fish and seafood intake provides about 7 percent
of all proteins and 17 percent of animal protein, and up to more than 50 percent of
animal protein in several countries in Africa and Asia (FAO, 2022).

1
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

1.2.2 HEALTH BENEFITS

Until recently, most research on the health benefits of fish consumption has focused
on the beneficial effects of proteins and marine LC n-3 PUFAs (FAO and WHO,
2010). In recent decades, however, the potential health benefits of micronutrients
from fish consumption have received more attention (Golden et al., 2021).
Recent estimates propose that, globally, over half of preschool-aged children and
two-thirds of women of reproductive age have micronutrient deficiencies, including
deficiencies in vitamin A, iron, iodine and vitamin D (Stevens et al., 2022). Micronutrient
deficiencies increase the risk of morbidity and mortality worldwide (Micha et al.,
2020), and consumption of fish and seafood can potentially improve human health by
supplying essential micronutrients (Golden et al., 2021). Moreover, by replacing the
intake of foods such as red and processed meats, fish consumption might further reduce
the risk of diet-related non-communicable diseases (Golden et al., 2021).
Previous risk assessments and guidelines for fish consumption have mostly pointed
to the potential health benefits of fish consumption related to a reduced risk of
cardiovascular disease and a lower risk of suboptimal neurodevelopment (FAO and
WHO, 2010; EFSA, 2014a).

1.3 RISKS OF FISH CONSUMPTION

In addition to considering the health benefits of fish consumption, however, the
fact that fish may contain undesirable components that may impact human health
must also be considered. Many contaminants are ubiquitously present in the
environment and may be present in fish and seafood, depending on factors such as
geography and the age and trophic level of aquatic animals. Of particular concern
are substances such as dioxins, dioxin-like polychlorinated biphenyls (dl-PCBs)
and methylmercury (MeHg).

1.3.1 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS

Dioxins and dl-PCBs are members of a group of compounds that are classified as
persistent organic pollutants, as they are characterized by their persistence in the
environment and their potential to bioaccumulate and biomagnify in the food chains.
Dioxins and dl-PCBs are often grouped together due to their ability to bind to the
aryl hydrocarbon receptor and, therefore, share similar toxicological properties.
They are often considered together within the context of public health concerns
(JECFA 2002, EFSA 2022).

1.3.1.1 Dioxins
Dioxins are a collective term for the chlorine-containing groups of polychlorinated
dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs).
Their diverse chemical structures depend on the number of chlorine atoms
(1 to 8) and their position in the rings. There are 75 different PCDDs and 135

2
C H A P T E R 1 . I N T R O DUCTION

different PCDFs, also termed different “congeners” (EFSA, 2011). Dioxins enter
the environment mainly as byproducts of industrial processes, but also from
natural processes, such as volcanic eruptions and forest fires (Kanan et al., 2018).
Of the many PCDD and PCDF congeners, 17 of them (7 PCDDs and 10 PCDFs)
are considered relevant to public health because of their persistence and toxic effects
(Van den Berg et al., 2006).

1.3.1.2 Dioxin-like polychlorinated biphenyls

Polychlorinated biphenyls (PCBs) are a group of organochlorine compounds,
consisting of two compound benzene rings that are substituted with chlorine atoms.
Their chemical structures depend on the number and positions of the chlorine atoms
(between 1 and 10) at the two rings, which makes for 209 different compounds,
or congeners (EFSA, 2011). Twelve of the PCB congeners have structures and
toxicological properties that are similar to those of dioxins. These 12 congeners are
therefore termed dioxin-like PCBs, or dl-PCBs. All PCBs are synthetic compounds
that were produced for numerous industrial applications, until they were banned in
the 1980s and scheduled to be phased out in most countries through the Stockholm
Convention. Because of these regulations, the emissions of both dioxins and
dl-PCBs have been reduced by more than 90 percent since 1987. However, due to their
chemical stability, they are very persistent in the environment (Mozaffarian et al., 2006).

1.3.1.3 Toxicity of dioxins and dioxin-like polychlorinated biphenyls

The potential negative health effects of dioxins and dl-PCBs include endocrine,
immune, developmental, reproductive and carcinogenic effects (JECFA, 2002;
EFSA, 2018). The mechanism of toxicity is proposed to be similar for all the
congeners of dioxins and dl-PCBs; that is, through activating the aryl hydrocarbon
receptor causing the same type of biochemical and adverse effects mediated by the
aryl hydrocarbon receptor (JECFA, 2002; EFSA, 2018). The congeners differ in
the number and position of their chlorine atoms, which affects the affinity of each
congener to the aryl hydrocarbon receptor, leading to differences in their toxic
potential. Because of the variation in toxicity, toxic equivalency factors (TEFs) were
developed to express the toxicities of dioxins and dl-PCBs (Van den Berg et al., 2006).
The TEF values published by the World Health Organization (WHO) in 2005 and
are given in Table 1.1.1
Among the congeners of dioxins and dl-PCBs, 2,3,7,8-tetrachlorodiobenzo-p-dioxin
(TCDD) was defined as the most toxic congener and was assigned a TEF of 1.
All other congeners were assigned TEF values based on their toxicity relative to that
of TCDD. As dioxins and dl-PCBs can be present together, for instance in foods,
the total dioxin-like toxicity of existing mixtures can be calculated by summing
the products of each single congener’s concentration and its corresponding TEF,
resulting in toxic equivalent quotients (TEQ) (Van den Berg et al., 2006).

1
The 2005 TEF values set by WHO have recently been updated at an expert meeting held by WHO in October 2022. The outcome of this
expert consultation was published in 2023, after the development of this background document (DeVito et al. 2024).

3
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 1.1 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS WITH TOXICITY EQUIVALENCE
FACTOR (TEF) SET BY THE WORLD HEALTH ORGANIZATION IN 2005

COMPOUND TEF
Polychlorinated dibenzo-p-dioxins (PCDDs) (n = 7)
2,3,7,8-TCDD 1
1,2,3,7,8-PeCDD 1
1,2,3,4,7,8-HxCDD 0.1
1,2,3,6,7,8-HxCDD 0.1
1,2,3,7,8,9-HxCDD 0.1
1,2,3,4,6,7,8-HpCDD 0.01
OCDD 0.0003
Polychlorinated dibenzofurans (PCDFs) (n = 10)
2,3,7,8-TCDF 0.1
1,2,3,7,8-PeCDF 0.03
2,3,4,7,8-PeCDF 0.3
1,2,3,4,7,8-HxCDF 0.1
1,2,3,6,7,8-HxCDF 0.1
1,2,3,7,8,9-HxCDF 0.1
2,3,4,6,7,8-HxCDF 0.1
1,2,3,4,6,7,8-HpCDF 0.01
1,2,3,4,7,8,9-HpCDF 0.01
OCDF 0.0003
Dioxin-like polychlorinated biphenyls (dl-PCBs) (n = 12)
Non-ortho PCBs (n = 4)
PCB-77 0.0001
PCB-81 0.0003
PCB-126 0.1
PCB-169 0.03
Mono-ortho PCBs (n = 8)
PCB-105 0.00003
PCB-114 0.00003
PCB-118 0.00003
PCB-123 0.00003
PCB-156 0.00003
PCB-157 0.00003
PCB-167 0.00003
PCB-189 0.00003

Notes: TEF: toxicity equivalence factor, TCDD: 2,3,7,8-tetrachlorodiobenzo-p-dioxin, PeCDD: pentachlorodibenzo-P-dioxin, HxCDD:
hexachlorodibenzo-p-dioxin, HpCDD: heptachlorodibenzo-p-dioxin, OCDD: octachlorodibenzo-p-dioxin, TCDF: tetrachlorodibenzofuran,
PeCDF: pentachlorodibenzofuran, HxCDF: hexachlorodibenzofuran, HpCDF: heptachlorodibenzofuran, PCB: polychlorinated biphenyls.
Source: Van den Berg, M., Birnbaum, L.S., Denison, M., De Vito, M., Farland, W., Feeley, M., Fiedler, H., Hakansson, H., Hanberg, A. and Haws,
L. 2006. The 2005 World Health Organization re-evaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like
compounds. Toxicological Sciences, 93(2): 223-241.

4
C H A P T E R 1 . I N T R O DUCTION

1.3.1.4 Dioxins and dioxin-like polychlorinated biphenyls in the environment

Dioxins and dl-PCBs are found throughout the environment, but the compounds are
lipid-soluble and therefore tend to accumulate in organisms, where the compounds
are mainly bound to lipids. As dioxins biomagnify in the food chain, the highest
concentrations are generally found in fatty tissues of top predators. Because of their
chemical stability, once dioxins enter the human body, they last a long time and are
stored in fat tissue. The half-life of dioxins in the human body has been estimated
at 7 to 11 years (Kerger et al., 2007).

1.3.1.5 Main food sources of dioxins and dioxin-like polychlorinated biphenyls

More than 90 percent of human exposure to dioxins and dl-PCBs is through
food consumption (WHO, 2016). The highest concentrations are found in
organisms located high up in the food chain, and particularly in fat of animal origin
(VKM, 2022). The main food sources of dioxins and dl-PCBs are fish and seafood,
meat, eggs, and milk and dairy products (EFSA, 2018).

1.3.1.6 Tolerable intakes of dioxins and dioxin-like polychlorinated biphenyls

In 2018, ESFA established a tolerable weekly intake (TWI) for dioxins and dioxin-like
PCBs in food of 2 picograms per kg of body weight (EFSA, 2018). This new TWI is
seven times lower than the previous TWI set in 2001 by the European Commission’s
former Scientific Committee on Food. According to EFSA, this updated TWI should
protect against effects on semen quality, lower sex ratio of sons to daughters, higher
levels of thyroid-stimulating hormone in newborns and developmental enamel
defects on teeth.

1.3.2 METHYLMERCURY
Mercury (Hg) is a toxic non-essential element and metal that is naturally present
in the earth’s crust. It is distributed in the environment by both natural and
anthropogenic processes, including volcanic eruptions, erosion, mining, coal
incineration and other industrial processes. Hg cycles between the atmosphere,
water, ocean, biota and land, where it undergoes complex transformations between
the different forms of Hg (ATSDR, 2022). Humans are exposed to Hg during
these biogeochemical cycles, and this may result in various health implications
(ATSDR, 2022).
Depending on the chemical state, Hg toxicity and toxicokinetics vary. The chemical
forms of Hg can be categorized into three groups; metallic or elemental Hg (Hg0),
inorganic Hg compounds (Hg22+ and Hg2+), and organic Hg compounds (EFSA, 2012;
ATSDR, 2022). Organic Hg is combined with carbon covalently. Methylmercury
(MeHg) is the most common form of organic Hg found in the environment, and
the form of Hg of most concern, due to its toxicological properties (Guangliang et
al., 2012).

5
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

1.3.2.1 Main food sources

Humans are exposed to MeHg predominantly through fish and seafood consumption
(Bradley et al., 2017). In fish, 80 to 100 percent of total Hg in fish muscle is present in
the form of MeHg, and 100 percent presence of MeHg is often assumed to account
for a conservative approach in exposure estimates (EFSA, 2012; VKM, 2014).
MeHg is primarily produced by methylation of inorganic Hg as a result of
microbial activity of microorganisms in aquatic systems. MeHg content in fish and
seafood species varies extensively and depends on a variety of factors including
species, trophic level, size, age and diet (Sheehan et al., 2014). MeHg accumulates
in organisms and biomagnifies along the aquatic food chain, reaching the highest
concentrations in animals at the highest trophic levels. Thus, long-lived predatory
seafood species contain the highest concentrations of MeHg (National Research
Council, 2000; Guangliang et al., 2012).

1.3.2.2 Toxicity and health effects of methylmercury

All chemical forms of Hg are toxic to humans, although toxicity depends on several
factors including the chemical form of Hg, exposure route, dose, age at exposure and
duration of exposure (Berlin et al., 2014). The toxic effects of Hg affect several body
compartments and studies have shown adverse effects on the central and peripheral
nervous system, kidneys, cardiovascular system, liver, gastrointestinal system and
immune system (Berlin et al., 2014). However, the primary targets of MeHg toxicity
are the brain and the central nervous system (Antunes dos Santos et al., 2016).

1.3.2.3 Tolerable intakes

In 2003, the Joint Food and Agriculture Organization of the United Nations
(FAO)/WHO Expert Committee on Food Additives (JECFA) established a
provisional TWI for MeHg of 1.6 µg/kg of body weight (FAO/WHO, 2004).
In 2012, EFSA established a TWI for MeHg of 1.3 µg/kg of body weight (EFSA,
2012). Both JECFA’s provisional TWI and ESFA’s TWI are based on the most
sensitive toxicological endpoint, which was considered to be neurodevelopmental
outcomes after prenatal exposure to MeHg.

1.3.3 SELENIUM AND METHYLMERCURY

While fish and seafood consumption is the main source of MeHg exposure (EFSA, 2012),
fish and seafood are also rich in micronutrients, including vitamins and minerals.
In particular, fish and seafood are major sources of selenium (Se).
The most sensitive toxicological endpoints for MeHg toxicity were proposed
to be brain damage in developing foetus and young children (FAO and WHO,
2004; EFSA, 2012). Therefore, concern has been raised for populations with
high exposure to MeHg. However, the epidemiological studies investigating
populations with moderate exposure to MeHg from fish consumption in pregnancy
and neurodevelopmental outcomes in offspring have shown conflicting results,

6
C H A P T E R 1 . I N T R O DUCTION

reporting both negative and null effects of increased MeHg exposure

(Grandjean et al., 1997: Crump et al., 1998; Davidson et al., 1998; Myers et al.,
2003; Strain et al., 2012; van Wijngaarden et al., 2013b). One proposed explanation
of the results showing no adverse effect of increased MeHg exposure from fish
consumption, was counteraction by the high levels of Se in fish, as Se may have a
protective effect against MeHg toxicity (Cuvin-Aralar et al., 1991; Chapman et al.,
2000; Raymond et al., 2004).
Selenium is an essential trace element, the main function of which, in the human
body, is as part of enzymes such as glutathione peroxidase, which has an important
role in protecting against oxidative stress, and in the enzyme iodothyrodine
5’-deiodinase, which is important for the metabolism of the thyroid hormones
(EFSA, 2014b). EFSA has set a dietary reference value for Se of 70 µg/day for adults
and an additional 15 µg for lactating women (EFSA, 2014b). However, EFSA did
not consider the possible protective effect of Se on MeHg toxicity in this dietary
reference value. An upper limit for Se intake was proposed at 255 µg/day, and it is
assumed that no consumers with a normal diet will exceed this (EFSA, 2023).
The proposed protective effect of Se against MeHg toxicity was claimed to be
achieved by influencing the transport, bioavailability, speciation and detoxification
processes (Raymond et al., 2004). Several animal studies suggest that Se can reduce
MeHg toxicity when both are simultaneously present in the diet, particularly in
seafood (Stillings et al., 1974; Ohi et al. 1980; Bjerregaard et al., 2012; Mellingen et
al., 2022).
The molecular basis for this proposed effect is the very strong binding between
Se and Hg – one million times stronger that the binding between Hg and sulphur.
This binding might be part of the basis both for the toxic effects of MeHg and the
protective effect of Se (Timmerman et al., 2021). Se and Hg may form complexes in
blood and, thereby, decrease the availability of both. MeHg is proposed to interfere
with or block the function of enzymes containing Se, and extra dietary Se can mitigate
this blocking. Additionally, direct binding of Se and Hg in nodules is proposed
as a mechanism to inactivate MeHg. This will normally require demethylation of
MeHg to inorganic Hg, as shown to occur in mammals with very high levels of
both Hg and Se, such as the bottlenose dolphin (Marumoto et al., 2022). To ensure
Se sufficiency for its essential functions, Se requirements may therefore increase
with MeHg exposure.
Though mechanistic in vitro studies and animal studies have shown potential
effects of Se against MeHg toxicity, very few human studies have been conducted
on this question.

7
 8
© FAO/Massimo Berruti
CHAPTER 2
METHODS

To provide new scientific evidence, five extensive literature reviews were conducted
focusing on the following topics:
> evidence of health benefits from fish consumption;
> toxic effects of dioxins and dl-PCBs (published in the last ten years);
> toxic effects of MeHg (published in the last ten years);
> the role of Se with regard to the health effects of MeHg;
> occurrence data for MeHg, dioxins and dl-PCBs in fishery and aquaculture
products (published in the last ten years).
The reviews followed a systematic approach, including a systematic literature search
and elements from systematic literature reviews. A systematic literature review
attempts to identify, appraise and synthesize all the empirical evidence that meets
prespecified eligibility criteria to answer a specific research question (Cochrane
Library, 2023). When conducting systematic reviews, systematic methods are
selected, aiming to minimize bias and produce more reliable findings to inform
decision-making (Cochrane Library, 2023). The following steps are generally
followed in a systematic review (Page et al., 2021):
1. determination of the scope and research question;
2. definition of inclusion and exclusion criteria;
3. development of a literature search strategy;
4. performance of literature search in relevant databases;
5. selection of the literature by screening of titles and abstracts;
6. selection of the literature by screening of full texts;
7. quality assessment of the selected literature;
8. extraction of data and results from the selected literature;
9. synthesizing and summarizing of evidence from the selected literature.
Generally, systematic reviews include the primary research that is available
(Clarke, 2011), although some reviews also summarize the available systematic
reviews (Aromataris et al., 2015). The literature review presented here uses a
combined method, integrating primary research, systematic reviews, and other

9
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

relevant risk and benefit assessments. In addition to reviewing the literature,

the literature was summarized. The review was conducted according to the PRISMA
guidelines for searching, selecting and reporting on literature (Page et al., 2021).

2.1 METHODS FOR THE REVIEW “EVIDENCE OF HEALTH BENEFITS FROM

FISH CONSUMPTION”
The objective of this literature review was to evaluate the evidence of health benefits
from fish consumption. The review included a systematic literature search, quality
assessment of the literature, summarizing the available literature, and weighing the
evidence.
In June 2022, the Norwegian Scientific Committee for Food and Environment
published the report, Benefit and risk assessment of fish in the Norwegian diet
(VKM, 2022). The report included a systematic literature review (including both
systematic reviews and primary articles) of fish consumption and relevant health
outcomes. The report was included in this literature review as it was considered
sufficiently comprehensive for the evaluation of evidence of health benefits from
fish consumption. In addition to selecting this report, a comprehensive literature
search was performed. The following sections provide a comprehensive description
of the VKM report, followed by a description of the methods of the literature search
and further steps.

2.1.1 2022 VKM REPORT: BENEFIT-AND-RISK ASSESSMENT OF FISH

IN THE NORWEGIAN DIET
The Norwegian Scientific Committee for Food and Environment (VKM) carries
out independent risk assessments for the Norwegian Food Safety Authority
(Mattilsynet) and the Norwegian Environment Agency (Miljødirektoratet). The
VKM comprises independent experts who conduct independent risk assessments.
In 2022, the VKM published the report, Benefit and risk assessment of fish in the
Norwegian diet.2 The report was based on an extensive systematic literature review
conducted to evaluate the epidemiological evidence for associations between fish
consumption and health outcomes. The review covered both original primary
studies and systematic reviews. The health outcomes included were: cardiovascular
disease (CVD); congenital heart disease (CHD); myocardial infarction; stroke;
heart failure; heart fibrillation; venous thrombosis; neurodevelopment in children;
mental disorders in children (such as autism spectrum disorder [ASD] and attention
deficit hyperactivity disorder [ADHD]); cognition and cognitive decline in adults
(including Alzheimer’s disease and dementia); depression in adults; type 2 diabetes
(T2D); weight/overweight in children and adults; bone health; birth outcomes
(such as preterm birth, small-for-gestational-age, low birth weight, birth weight

2
A protocol for the methodology of the report was published in 2020 after a public consultation, and the work was conducted according
to the protocol (VKM, 2020).

10
C H A P T E R 2 . M E T H ODS

[continuous], birth length and head circumference [continuous]); asthma and allergy
(especially in children); multiple sclerosis; rheumatoid arthritis; vaccine response
and semen quality/male fertility.
The systematic literature review followed the established guidelines for performing
systematic reviews (PRISMA or JBI) for searching, selecting and reporting on the
literature. The review, performed in the databases Medline, Embase and PsycINFO,
included a systematic literature search, including the search string for all the health
outcomes. The search resulted in 27 182 papers (26 384 primary studies and 798
systematic reviews, after removal of duplicates).
These papers were further screened by two blinded reviewers, using predefined
inclusion and exclusion criteria. After screening the papers, 409 papers (346 primary
studies and 63 systematic reviews) were included for quality assessment using risk-
of-bias tools. Based on quality assessment, 100 papers (76 primary studies and 24
systematic reviews) were excluded, leaving 309 papers (270 primary studies and
39 systematic reviews) included in the review for data extraction. In addition,
pooled estimates were calculated from the primary studies included. This entailed
calculating a summary risk ratio (RR) with 95 percent confidence interval (CI) for
binary disease outcomes in relation to the highest versus lowest fish intake. The
pooled estimates (summary RR) were further compared to previous meta-analyses
that were included from the literature search. Finally, the weight of evidence was
based on the criteria from the World Cancer Research Fund (WCRF).
The literature search and inclusion and exclusion criteria were based on the work
of the VKM in their 2022 report, which was graded with the quality-assessment
tool AMSTAR-2 as “high quality” (Appendix 3, Table A3.35) and was therefore
considered sufficiently comprehensive to answer the research question “evidence
of health benefits of fish consumption”.
As the literature search string used in the 2022 VKM report focused on Norway, for
this background report, the search string was extended to a global setting including
all fish and seafood species according to the major species produced in aquaculture
and marine capture production (FAO, 2020).

2.1.2 LITERATURE SEARCH STRATEGY

The search strategy was developed using a population, intervention, comparison,
outcome and study designs, or PICOS, table (Table 2.1) for the research question
“What is the evidence of health benefits from fish consumption?”.

11
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 2.1 POPULATION, INTERVENTION, COMPARISON, OUTCOMES AND STUDY DESIGN TABLE FOR THE
LITERATURE SEARCH ON “EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

INTERVENTION/
POPULATION COMPARISON HEALTH OUTCOMES STUDY DESIGNS
EXPOSURE
General population1 Fish consumption No consumption or > Allergy and Systematic reviews
Including both sexes The term “fish” lower fish consumption immunology > Meta-analysis
and all ages (foetuses, is here defined as > Birth and growth > Cochrane reviews
infants, children, finfish (vertebrates) outcomes > Systematic reports/
adolescents, adults). and other seafood > Bone health reviews without a
(aquatic invertebrates, > Cancer meta-analysis
including crustaceans, > Cardiovascular > Umbrella reviews
molluscs and disease > Randomized
echinoderms), whether > Neurodevelopment controlled trials
of marine or freshwater and neurological Non-randomized
origin, farmed or wild. disorders intervention trials
> Mortality
Prospective cohort
> Obesity
studies
Note: 1 The following conditions will also be considered part of the general population: type 2 diabetes, overweight and obesity,
musculoskeletal disorders, malnutrition and nutritional deficiencies.

2.1.2.1 Population
The target population of the literature search is the general population, including
both sexes, foetuses and infants, children, adolescents and adults.

2.1.2.2 Intervention/exposure
The intervention is in term of fish consumption. The term “fish” is here defined as
finfish (vertebrates) and other seafood (aquatic invertebrates, including crustaceans,
molluscs and echinoderms), whether of marine or freshwater origin, farmed or
wild. Marine mammals and algae, as well as sustainability issues and environmental
impacts, are considered outside the scope of the report.

2.1.2.3 Comparison
The group of participants with fish consumption are compared to a group with no
fish consumption or lower fish consumption.

2.1.2.4 Health outcomes

Table 2.2 lists the categories of health outcomes defined to investigate the evidence
of the health benefits of fish consumption. The health outcomes were selected based
on established knowledge about fish and seafood consumption and health outcomes
relevant to nutrients for which fish and seafood are good sources.

12
C H A P T E R 2 . M E T H ODS

TABLE 2.2 CATEGORIES OF HEALTH OUTCOMES INCLUDED IN THE LITERATURE SEARCH

ON “EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

HEALTH OUTCOME DEFINITION

Allergy and immunology Allergy, asthma/wheezing, eczema, multiple sclerosis, allergic rhinitis, rheumatoid arthritis,
inflammation, immunodeficiency
Birth and growth outcomes Birth weight, premature birth, small for gestational age
Bone health Osteoporosis, osteopenia, rickets, osteomalacia, bone density, bone resorption,
demineralization, vitamin D deficiency
Cancer Cancer or tumours
Cardiovascular diseases and Heart or cardiac diseases, failure, event, arrest, infarct
outcomes (CVD) Atherosclerosis, stenosis, restenosis
Ischemia, thromboses, thromboembolism, tachycardia, arrythmia, fibrillation
Stroke, cardiac death, transient ischemic attack
Brain haemorrhage, brain infarct
Dental health Dental or teeth malformation, enamel changes and defects
Type 2 diabetes Type 2 diabetes
Neurodevelopment and neurological Child development, cognitive function, neurodevelopment, intelligence quotient,
disorders attention deficit hyperactivity disorder (ADHD), attention deficit disorder (ADD), Asperger,
schizophrenia, Alzheimer's, Parkinson’s, depression
Mortality Mortality, death, death rate
Overweight and obesity Overweight or obesity defined by body mass index (BMI)

2.1.2.5 Study designs

The study designs and types of articles included systematic reviews (meta-analysis,
Cochrane reviews, systematic reviews and umbrella reviews). For primary studies,
randomized controlled trials (RCTs) and prospective cohort studies were included.

2.1.3 LITERATURE SEARCH

The literature search was performed on 29 Nov. to 3 Dec. 2021, in the databases
PubMed, Web of Science and Cochrane Library. The search strategy for each
database is given in Appendix 3 (Table A3.1).
The search terms included a combination of the following:
> fish and seafood: other typical words for fish and seafood, specific fish and
seafood species;
> study type: systematic reviews, meta-analyses, umbrella reviews, RCTs, trials,
prospective cohort studies;
> dietary intake: specification that the literature investigated consumption, eating
or intake of fish and seafood;
> health outcomes: relevant health outcomes included in literature search.
In Web of Science the literature search was performed under the field “topic search”
which included search words within title, abstract, author keyword and keywords

13
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

plus. In PubMed, the literature search was performed including “all fields”. In
Cochrane Library, the literature search was performed under the field of “Title
Abstract Keyword”.
All records identified in the searches from each of the databases were imported into
the reference manager program EndNote. Duplicates were removed in EndNote by
comparing articles with similar “title” and “authors”.

2.1.4 SELECTION OF ARTICLES FROM THE LITERATURE SEARCH

Title and abstract screening of articles from the literature search was performed blind
by two reviewers using Rayyan, a web-based tool for selecting studies in systematic
reviews (Ouzzani et al., 2016). The selection of articles was based on predefined
inclusion and exclusion criteria (see Table 2.3), which were developed based on
the research question and a PICOS diagram. Full-text articles were retrieved when
the information given in the title or abstract seemed to fulfil the inclusion criteria.
After screening the articles by title and abstract, the blinding was removed, and the
two reviewers discussed any conflicting decisions regarding the papers that were
included and excluded.
The articles selected via abstract and title screening were further reviewed in full
text according to the same inclusion and exclusion criteria.
To avoid reporting duplicate publications, the systematic reviews and primary studies
that were included in the 2022 VKM report were excluded for further assessment in
the literature search. Furthermore, the primary studies that were already included
in the systematic reviews included in the literature search, were also excluded to
avoid duplicate reporting of publications.

2.1.5 QUALITY ASSESSMENT (RISK OF BIAS)

2.1.5.1 Systematic reviews

All the systematic reviews that fulfilled the inclusion criteria were quality-assessed
using AMSTAR 2 (Shea et al., 2017). AMSTAR 2 is a tool used to assess the
methodological quality of systematic reviews and meta-analyses, which consists
of 16 questions (see Appendix 2, Table A2.1). Each systematic review was quality-
assessed independently by two reviewers, after which they discussed their findings
and determined which reviews would be included. If the two reviewers were unable
to agree on the inclusion or exclusion of a systematic review, a third member in the
review team helped make a final determination.

14
C H A P T E R 2 . M E T H ODS

TABLE 2.3 INCLUSION AND EXCLUSION CRITERIA FOR THE LITERATURE SEARCH
ON “EVIDENCE OF HEALTH BENEFITS OF FISH CONSUMPTION”

CRITERIA FOR INCLUSION CRITERIA FOR EXCLUSION

> Studies investigating fish consumption in relation to one > Studies investigating fish consumption without any relation
or more health outcomes defined in the literature search to any of the health outcomes defined in the literature search
strategy strategy
> Study design: > Studies investigating exposure to LCn3PUFAs or dietary
> systematic reviews: meta-analysis (from systematic supplements only
reviews), Cochrane reviews, systematic reports/reviews > Dietary pattern studies
without meta-analysis, umbrella reviews > Studies investigating contaminants: PCBs, POPs, metals
> randomized controlled trials > Study design:
> non-randomized intervention trials > retrospective cohort studies
> prospective cohort studies > case-control studies
> Population: > cross-sectional studies
> general population, including both sexes and all ages > animal studies
(foetuses, infants, children, adolescents and adults). > in vitro studies
> The following conditions will also be considered as > non-systematic reviews (e.g. narrative reviews and
part of the general population: type 2 diabetes, obesity, scoping reviews).
musculoskeletal disorders, malnutrition and nutritional > systematic reviews including only results from
deficiencies, and nutritional deficiency diseases (goitre, retrospective cohort studies, case-control studies, cross-
osteoporosis, rickets, osteomalacia, anaemia). sectional studies, animal studies or in vitro studies.
> Population: Specific patient groups (exceptions are specified
in the inclusion criteria).
> Non-English-language articles
> Publication types: book chapters, case stories, letters to the
editor, posters, abstracts

AMSTAR 2 is not intended to generate an overall score, but rather to indicate an

overall level of confidence in the results of the review, as follows:
> high (zero or one non-critical weakness): The systematic review provides an
accurate and comprehensive summary of the results of the available studies that
address the question of interest.
> moderate (more than one non-critical weakness): The systematic review
has more than one weakness, but no critical flaws. It may provide an accurate
summary of the results of the available studies that were included in the review.
> low (one critical flaw with or without non-critical weaknesses): The review
has a critical flaw and may not provide an accurate and comprehensive summary
of the available studies that address the question of interest.
> critically low (more than one critical flaw with or without non-critical
weaknesses): The review has more than one critical flaw and should not be relied
on to provide an accurate and comprehensive summary of the available studies.
Papers graded “high” or “moderate” were included, while papers graded “low” or
“critically low” were excluded for further assessment.

15
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

2.1.5.2 Primary studies

All the primary studies that fulfilled the inclusion criteria were quality assessed
for risk of bias. This assessment was performed using the quality-assessment tools
developed for Nordic Nutrition Recommendations (NNR, 2014), which were further
adapted for the VKM report. These tools were chosen as they were appropriate for
nutritional studies assessing fish consumption as the exposure with different health
outcomes. Specific tools for RCTs and prospective cohort studies were used in
this case. All the questions included in the quality-assessment tools for RCTs and
prospective cohort studies are given in Appendix 2, Table A2.2 and Table A2.3.
The quality-assessment tool for RCTs and trials consists of 6 domains and 23
questions. The quality-assessment tool for prospective cohort studies consists of 7
domains and 22 questions. Each primary study was quality assessed independently
by two reviewers, after which they discussed and determined if the study was to be
included or not. If the two reviewers were unable to agree, a third member of the
review team was called in to assist in making a final decision. A prespecified Excel
template, including all the questions used, was completed by each reviewer when
conducting the quality assessment of each systematic review. Based on the answer
to each question, the primary study was graded as A (high quality), B (moderate
quality) or C (low quality) (VKM, 2022):
> A (high quality): The results from studies that have an acceptably low level of
bias are considered valid. These studies adhere mostly to the commonly held
concepts of high quality including the following: a comprehensive study design;
clear description of the participants, setting, interventions, and control group(s);
appropriate measurement of outcomes; appropriate statistical and analytical
methods and reporting; less than 30 percent dropout (depending on the length
of the study, see the quality-assessment tool for RCTs in Appendix 2) or over
50 percent participation rate for prospective cohort studies; clear reporting of
dropouts; and no obvious bias. Where appropriate, studies must provide a valid
estimation of food intake/nutrient exposure, from dietary assessments and/or
biomarkers with a reasonable range of measurement error, and justification for
approaches to control for confounding in the design and analyses.
> B (moderate quality): Studies may have some bias, but not sufficient to invalidate
the results. They do not meet all the criteria in category “A”. They have some
deficiencies, but none likely to cause major bias. The study may be missing
information, making it difficult to assess limitations and potential problems.
> C (low quality): Studies have significant bias that may invalidate the results.
These studies have serious errors in design, analysis or reporting; there are large
amounts of missing information, or discrepancies in reporting.
Papers graded A or B were included, while papers graded C (low quality) were
excluded for further assessment.

16
C H A P T E R 2 . M E T H ODS

2.1.6 EXTRACTION OF DATA FROM THE INCLUDED LITERATURE

Relevant prespecified data were extracted from the systematic reviews and primary
studies selected and recorded in premade data-extraction forms. The data extraction
was performed by one person, and the extracted data was double-checked by
another person.

2.1.6.1 From systematic reviews

The data extracted from the systematic reviews included:
> reference details: reference, year, title;
> outcome: main outcome and specific outcome groups;
> population: type of population and number of participants;
> study information: study design and number of studies included;
> information regarding fish intake;
> overall results: including results from meta-analysis, if applicable;
> overall conclusion.

2.1.6.2 From primary studies

The data extracted from the primary studies reviewed included:
> reference details: reference, trial or study name, geography, year of sampling;
> study information: study type, study duration and follow-up time;
> participants: number of participants, age (years) at exposure assessment, sex;
> measurements and intake of fish consumption;
> for RCTs and trials: information regarding intervention, duration, dose and
control group;
> outcome: measurement of outcome and type of outcome;
> overall results;
> overall conclusion.

2.1.7 SUMMARY OF THE LITERATURE REVIEWED

The results from the 2022 VKM report, the systematic reviews and primary studies
were summarized according to each health outcome. For the 2022 VKM report,
this included summarizing the literature included in their report, the effect of fish
consumption on the specific health outcome, and the final weight of evidence. From
the systematic reviews and primary studies from the literature search, the extracted
data from the data extraction tables were summarized with the most relevant
information for each health outcome.

17
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

2.1.8 WEIGHT OF EVIDENCE OF THE LITERATURE

In the final weight of evidence of the literature reviewed, the weight of evidence for
the specific health outcomes in the 2022 VKM report was compared with that of the
systematic reviews and primary studies from the literature search, and a final weight
of evidence was computed. The 2022 VKM report used the weight-of-evidence scale
of the WCRF (2018a), where evidence is graded as “convincing (strong evidence)”,
“probable (strong evidence)”, “limited, suggestive”, “limited, no conclusion”, or
“substantial effect of risk unlikely (strong evidence)”. Table 2.4 provides the criteria
for each WCRF weight-of-evidence grading. (The outcome “cancer” is replaced
with “outcome” as several health outcomes were included here).

TABLE 2.4 CRITERIA OF THE WORLD CANCER RESEARCH FUND FOR WEIGHING THE EVIDENCE

WEIGHT-OF-EVIDENCE DESCRIPTION
GRADING
Evidence strong enough to support a judgement of a convincing causal (or protective) relationship,
which justifies making recommendations designed to reduce risk of an outcome. The evidence is robust
enough to be unlikely to be modified in the foreseeable future as new evidence accumulates. All the
following are generally required:
> evidence from more than one study type;
> evidence from at least two independent cohort studies;
> no substantial unexplained heterogeneity within or between study types or in different populations
CONVINCING relating to the presence or absence of an association, or direction of effect;
(STRONG EVIDENCE)
> good quality studies to exclude with confidence the possibility that the observed association results
from random or systematic error, including confounding, measurement error and selection bias;
> presence of a plausible biological gradient (“dose-response”) in the association. Such a gradient
needs not be linear or even in the same direction across the different levels of exposure, so long as
this can be explained plausibly.
> Strong and plausible experimental evidence, either from human studies or relevant animal models,
that typical human exposures can lead to relevant health outcomes.
Evidence strong enough to support a judgement of a probable causal (or protective) relationship, which
generally justifies recommendations designed to reduce the risk of an outcome. All the following criteria
are generally required:
> evidence from at least two independent cohort studies, or at least five case-control studies;
PROBABLE > no substantial unexplained heterogeneity between or within study types in the presence or absence
(STRONG EVIDENCE)
of an association, or direction of effect;
> good quality studies to exclude with confidence the possibility that the observed association results
from random or systematic error, including confounding, measurement error and selection bias;
> evidence for biological plausibility.
Evidence that is too limited to permit a probable or convincing causal judgement but is suggestive of
a direction of effect. The evidence may be limited in amount or by methodological flaws, but shows a
generally consistent direction of effect. This judgement is broad, and includes associations where the
evidence falls only slightly below that required to infer a probably causal association through those
where the evidence is only marginally strong enough to identify a direction of effect. This judgement is
LIMITED — very rarely sufficient to justify recommendations designed to reduce the risk of cancer; any exceptions
SUGGESTIVE to this require special, explicit justification. All the following criteria are generally required:
> evidence from at least two independent cohort studies or at least five case-control studies;
> the direction of effect is generally consistent though some unexplained heterogeneity may be
present;
> evidence for biological plausibility.

18
C H A P T E R 2 . M E T H ODS

TABLE 2.4 CRITERIA OF THE WORLD CANCER RESEARCH FUND FOR WEIGHING THE EVIDENCE (cont.)
WEIGHT-OF-EVIDENCE DESCRIPTION
GRADING
Evidence is so limited that no firm conclusion can be made. This judgement represents an entry
level and is intended to allow any exposure for which there are sufficient data to warrant Panel
consideration, but where insufficient evidence exists to permit a more definitive grading. This does not
necessarily mean a limited quantity of evidence. A body of evidence for a particular exposure might
be graded “limited — no conclusion” for several reasons. The evidence may be limited by the amount
of evidence in terms of the number of studies available, by inconsistency of direction of effect, by
LIMITED — methodological flaws (for example, lack of adjustment for known confounders), or by any combination
NO CONCLUSION
of these factors.
When an exposure is graded “limited — no conclusion”, this does not necessarily indicate that
the Panel has judged that there is evidence of no relationship. With further good-quality research,
any exposure graded in this way might in the future be shown to increase or decrease the risk of an
outcome. Where there is sufficient evidence to give confidence that an exposure is unlikely to have an
effect on an outcome risk, this exposure will be judged “substantial effect of risk unlikely”.
Evidence is strong enough to support a judgement that a particular food, nutrient or physical activity
exposure is unlikely to have a substantial causal relation to cancer outcomes. The evidence should be
robust enough to be unlikely to be modified in the foreseeable future as new evidence accumulates. All
the following criteria are generally required:
> evidence from more than one study type;
> evidence from at least two independent cohort studies;
SUBSTANTIAL EFFECT > summary estimate of effect close to 1.0 for comparison of high versus low exposure categories;
ON RISK UNLIKELY
(STRONG EVIDENCE) > no substantial unexplained heterogeneity within or between study types or in different populations;
> good-quality studies to exclude, with confidence, the possibility that the absence of an observed
association results from random or systematic error, including inadequate power, imprecision or
error in exposure measurement, inadequate range of exposure; confounding and selection bias.
> absence of a demonstrable biological gradient (“dose-response”);
> absence of strong and plausible experimental evidence, from either human studies or relevant
animal models, that typical human exposure levels lead to relevant outcomes.
Source: WCRF (World Cancer Research Fund). 2018. Judging the evidence. Continuous Update Project Expert Report 2018.
https://www.wcrf.org/wp-content/uploads/2021/02/judging-the-evidence.pdf.

2.2 METHODS FOR THE REVIEW “TOXIC EFFECTS OF DIOXINS AND

DL- POLYCHLORINATED BIPHENYLS”
The objective of the review was to evaluate new data (published after the 2010 FAO/WHO
report on the risks and benefits of fish consumption) on toxic effects of dioxins
for all population groups. This was conducted through a systematic search, quality
assessment and the summarizing of the available literature on this topic.

2.2.1 LITERATURE SEARCH STRATEGY

Dioxins have previously been risk assessed by different authorities. The most recent
assessment was conducted by EFSA in June 2018, with the results set forth in the
report, Risk for animal and human health related to the presence of dioxins and
dioxin-like PCBs in feed and food (EFSA, 2018). The assessment was conducted
through a comprehensive systematic literature search, including studies published
between 1 July 1998 and 5 July 2016. This report was evaluated for use as a starting

19
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

point in the current evaluation, using the quality-assessment tool for systematic
reviews, AMSTAR 2 (see Section 2.1.5.1). The AMSTAR 2 score was “high”
(judgement given in Appendix 4, Table A4.6), and, as such, the EFSA report was
used as a basis for the years 2010 to 2016. In addition, the extensive literature search
protocol and search strategy developed by EFSA (Vedrine et al., 2018), was used and
updated for the years 2016 to 2022. To avoid duplicate reporting of publications,
only studies published from 5 July 2016 onwards were included in the further search.
The search covered literature published since 2010 and was based on the following
PECOS table (Table 2.5), aiming to identify and characterize the toxic effects of
dioxins and dl-PCBs in all human population groups.

TABLE 2.5 POPULATION, EXPOSURE, COMPARISON, OUTCOMES AND STUDY DESIGNS (PECOS) TABLE
FOR THE LITERATURE SEARCH ON “TOXIC EFFECT OF DIOXINS AND dl-PCBs”

INTERVENTION/
POPULATION COMPARISON HEALTH OUTCOMES STUDY DESIGNS
EXPOSURE
General population, all Exposure of dioxins and Lower exposure of > Toxic and adverse Cohort studies
population groups dl-PCBs dioxins effects on human Case-control studies
Including both sexes, health (any health
Cross-sectional studies
all ages (foetuses, outcome or endpoint)
Systematic reviews and
infants, children,
meta-analyses
adolescents, and
adults) in all countries
Note: dl-PCB: dioxin-like polychlorinated biphenyl

2.2.1.1 Population
The target population of the literature search is the general population, including all
population groups, both sexes and all ages (foetuses, infants, children, adolescents
and adults) in all countries.

2.2.1.2 Exposure
The intervention or exposure is dioxin exposure. This includes all routes of exposure
(dietary, dermal, inhalation and transplacental exposure). Studies were included if
dioxins were measured in human tissues (including by bioassays) or if the total
dietary exposure to the following target compounds was estimated:
> 17 polychlorinated dibenzo‐p‐dioxins and dibenzofurans (PCDD/Fs) and 12
dl-PCBs  
> 17 PCDD/Fs  
> 12 dl-PCBs  
> 17 PCDD/Fs plus non-ortho PCBs, at least one PCB being PCB-126  
> TCDD (when dominating the TEQs, as in the Seveso incident) or any of the
individual target congeners that contribute to a substantial part of the TEQs.

20
C H A P T E R 2 . M E T H ODS

2.2.1.3 Comparison
The group of participants with exposure to dioxins was compared to a group with
lower exposure to dioxins.

2.2.1.4 Outcomes
Toxic and adverse effects on human health (any health outcome or endpoint) were
included.

2.2.1.5 Study designs

Human studies of cohort, case-control and cross-sectional studies were included.
Studies with any duration and any number of subjects were included. Also,
systematic reviews and meta-analyses were included.

2.2.2 LITERATURE SEARCH

The literature search was performed on 13 December 2021. The searches were
performed in two databases, namely PubMed and Web of Science Core Collection,
focusing on literature published since 2010. The search strategy for each database
is provided in Appendix 4, Table A4.1 and Table A4.2.
The search terms included a combination of the following:
> dioxins: TCDD, dioxin, PCB, PCDD and PCDF;
> population: human, women, men, children;
> outcome: adverse effect or health outcome and dioxins measures in human tissues
and waste products;
> study design: specific study designs included.
In Web of Science, the literature search was performed under the field “topic
search”, which included search words within title, abstract, author keyword and
keywords plus. In PubMed, the literature search included search words within title
and abstract. The literature searches were restricted to type of publication, human
studies, language (English), and date of publication, in accordance with the literature
search criteria (Table 2.5).
All identified records from the searches in each of the databases were imported into
the reference manager program, EndNote. Duplicates were removed in EndNote
by comparing articles with similar “title” and “authors”.

2.2.3 SELECTION OF STUDIES FROM THE LITERATURE SEARCH

The studies included from the literature search were further assessed by screening
the titles and abstracts. This was performed independently by two reviewers using
Rayyan, a web-based tool for selecting studies in systematic reviews (Ouzzani et
al., 2016). Study selection was based on the predefined inclusion and exclusion
criteria (Table 2.6), which were developed based on the research question and a

21
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

PECOS table. Full-text articles were retrieved when the information given in the
title or abstract seemed to fulfil the inclusion criteria. After screening the papers by
title and abstract, the blinding was removed, and the two reviewers discussed any
conflicting decisions regarding studies that were included or excluded. The studies
selected via abstract and title screening were further reviewed in full text according
to the same inclusion and exclusion criteria.

TABLE 2.6 INCLUSION AND EXCLUSION CRITERIA FOR THE LITERATURE SEARCH ON
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

CRITERIA FOR INCLUSION CRITERIA FOR EXCLUSION

> Studies investigating exposure to dioxins in relation to one or > Studies on mono-ortho PCBs only
more health outcomes in a human population > Studies on non-dioxin-like PCBs (indicator)
> Studies including all routes of exposure (dietary, dermal, > Studies on mixtures in which the contribution from the target
inhalation, transplacental) compounds does not allow the calculation of TEQs.
> Studies where dioxins were measured in human tissues > Studies on gene expression or drug metabolizing activity or
(including by bioassays), or studies in which the total dietary levels only
exposure to the following target compounds was estimated: > Study design:
> - 17 PCDD/Fs and 12 dl-PCBs > animal studies
> - 17 PCDD/Fs > in vitro studies
> - 12 dl-PCBs > other reviews (narrative reviews)
> - 17 PCDD/Fs plus non-ortho PCBs, at least one PCB being > Publication types: expert opinions, case studies, editorials,
PCB-126 letters to editors, PhD theses, abstracts, conference
> - TCDD (when dominates the TEQs, as in the Seveso proceedings
incident) or any of the individual target congeners that > Non-English-language articles
dominates the TEQs
> Study designs:
> systematic reviews
> randomized controlled trials
> non-randomized intervention trials
> cohort studies
> case-control studies
> cross-sectional studies
> case series/case reports
> Population:
> general population, all population groups
> both sexes, all ages (foetuses, infants, children,
adolescents, and adults) in all countries
> Published from 5 July 2016 onwards

Note: dl-PCB: PCDD/Fs: polychlorinated dibenzo‐p‐dioxins and dibenzofurans, PCB: polychlorinated biphenyl, dl-PCB: dioxin-like
polychlorinated biphenyl, TCDD: 2,3,7,8-tetrachlorodiobenzo-p-dioxin, TEQ: toxic equivalent quotient.

The systematic reviews retrieved in full text from the literature search included only
a few relevant primary studies not already included in EFSA, 2018 (that is, published
after 2016). Therefore, all the systematic reviews were excluded, except the EFSA 2018
Dioxin report (the excluded systematic reviews are listed in Appendix 4, Table A4.3).
Primary studies published after 2016 were included for individual quality assessment.

22
C H A P T E R 2 . M E T H ODS

2.2.4 QUALITY ASSESSMENT (RISK OF BIAS)

2.2.4.1 Systematic reviews

The risk assessment report of dioxins and dl-PCBs from EFSA 2018 was quality
assessed using the quality-assessment tool for systematic reviews, AMSTAR 2. More
information regarding the AMSTAR 2 tool is provided in Section 2.1.5.1 and in
Appendix 2, Table A2.1.

2.2.4.2 Primary studies

All the studies that fulfilled the inclusion criteria were quality assessed using the Risk of
Bias Rating Tool for Human and Animal studies of the US National Toxicology Program’s
Office of Health Assessment and Translation (OHAT) (Appendix 2, Table A2.4)
(Rooney et al., 2014; OHAT, 2015; OHAT, 2019). The OHAT tool, which was also
used in the EFSA (2018) report on dioxins and dl-PCBs in feed and food, consists
of 11 different questions or domains, where each question is applicable to specific
study designs. The tool can be used for the following study designs:
> experimental animal;
> human controlled trial (randomized controlled trials and non-randomized
experimental studies);
> cohort studies;
> case-control studies;
> cross-sectional studies;
> case series/reports.
For cohort, case-control and cross-sectional studies and case series or reports, seven
questions are applicable; while for experimental animal studies, nine questions
are applicable. The questions are grouped under six types of bias (confounding,
selection, performance, attrition/exclusion, detection and selective reporting). Each
risk-of-bias question is given a score based on a four-point scale, where the reviewer
chooses between four different options, based on the quality of the study (Table 2.7):
> (++): Definitely low risk of bias
> (+): Probably low risk of bias
> (-) or (not reports [NR]): Probably high risk of bias
> (--): Definitely high risk of bias.

23
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 2.7 RISK-OF-BIAS CATEGORIES

RISK-OF-BIAS RATING EXPLANATION

(++) Definitely low risk of bias There is direct evidence of low risk-of-bias practices.
Probably low risk of bias There is indirect evidence of low risk-of-bias practices OR it is deemed that deviations
(+) from low risk-of-bias practices for these criteria during the study would not appreciably
bias results, including consideration of direction and magnitude of bias.
(-) Probably high risk of bias/not There is indirect evidence of high risk-of-bias practices OR there is insufficient
NR reported information (e.g. not reported [NR]) provided about relevant risk-of-bias practices.
(- -) Definitely high risk of bias There is direct evidence of high risk-of-bias practices.
NA Not applicable Not relevant for this study.

Some of the risk-of-bias questions were considered key questions (Appendix 2,

Table A2.4). The individual bias rating for each question was then combined by an
algorithm and translated into an overall tier for each study (Table 2.8 and Table 2.8)
(OHAT 2019).

TABLE 2.8 OVERALL RISK-OF-BIAS JUDGMENT

TIER (TOTAL SCORE) CRITERIA FOR CLASSIFICATION

1 (low risk of bias) All key questions are scored (+) or (++) AND (+) or (++) for more than half the non-key questions
2 (moderate risk of bias) All combinations not falling under tier 1 or 3
3 (high risk of bias) All key questions are scored (- -) or (-) AND (- -) or (-) for more than half the non-key questions

Each paper was quality assessed independently by two reviewers using OHAT, after
which the reviewers discussed their assessments and agreed on a final decision for
each paper. If the two reviewers could not agree on a final decision, a third member
of the review team was included to help make a final determination. A prespecified
Excel template, including all the bias questions, was completed by each reviewer
when conducting the quality assessment of each paper. Only Tier 1-rated articles
were considered for data extraction and summarizing.

2.2.5 EXTRACTION OF DATA FROM THE LITERATURE

For the extraction of data from the primary studies included, a modified version
of the EFSA extraction template was used (EFSA, 2018), including the following
information:
> study or trial name;
> number of participants;
> funding source;
> study design and time frame;
> sex, age, ethnicity and socioeconomic variables;
> geography;
> compounds of dioxins or dl-PCBs measures, type of tissues measured, validation

24
C H A P T E R 2 . M E T H ODS

of the method;
> levels measured in human tissue or estimated dietary intake;
> outcome measured (health category and specific outcomes measured) and
method used to measure health outcome;
> confounders assessed;
> measures of effect between exposure and outcome (including statistical test and
treatment of variables);
> dose–response effect, if applicable.

2.2.6 SUMMARIZATION OF THE INCLUDED LITERATURE

As previously described, the risk assessment by EFSA (2018) was used as a basis
for summarizing the available literature. Thus, the summaries consist of a short
summary of the findings and conclusions by EFSA, supplemented with a brief
description of the studies identified through the present systematic literature search.
All studies deemed Tier 1 quality according to the risk-of-bias assessment were
within the endpoint health categories suggested by EFSA. As such, this report
summarizes the result according to the categories used by EFSA in their 2018 report:
chloracne and other dermal effects, reproductive effects (including male reproductive
effects, female reproductive effects and birth outcomes), thyroid disease and thyroid
hormones, T2D and obesity, cardiovascular effects, hepatic disorders and digestive
effects, effects on the immune system, effects on the nervous system, effects on teeth
and bones, cancer and other effects.
A weight-of-evidence approach was not given as the review of “Toxic effects of
dioxins and dl-PCBs” only included literature published after 2010. As such, it was
not possible to compute an overall grading of the available evidence.

2.3 METHODS FOR THE REVIEW “TOXIC EFFECTS OF MeHg”

The objective of the review was to evaluate new data (published after the 2010 FAO/
WHO report on the risks and benefits of fish consumption) on the toxic effects of
MeHg for all population groups. The evaluation was conducted through a systematic
search, quality assessment, and summarizing of the available literature.

2.3.1 LITERATURE SEARCH STRATEGY

The search strategy was developed using a PECOS table (Table 2.9) aiming to
identify and characterize toxic effects of MeHg in all human population groups,
focusing on literature published since 2010.

25
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 2.9 POPULATION, EXPOSURE, COMPARISON, OUTCOMES AND STUDY DESIGNS (PECOS) TABLE
FOR THE LITERATURE SEARCH ON “TOXIC EFFECTS OF MeHg”

POPULATION EXPOSURE COMPARISON OUTCOMES STUDY DESIGNS

General population, all Exposure to MeHg Reference groups Toxic and adverse effects Cohort studies
population groups assumed to have a on human health (any Case-control studies
Including both sexes, lower exposure to MeHg health outcome) are Cross-sectional studies
all ages (foetuses, than exposed groups included. Randomized controlled
infants, children, trials or non-randomized
adolescents, and trials
adults), and pregnant Systematic reviews
and lactating women, Meta-analysis (if
in all countries included in a systematic
review)

2.3.1.1 Population
The target population of the literature search was the general population, including
all population groups, both sexes, all ages (foetuses, infants, children, adolescents and
adults), and pregnant and lactating women, in all countries (if reported in English).

2.3.1.2 Exposure
The intervention or exposure is in terms of MeHg exposure from ingestion (dietary
intake) or transplacental exposure. As the focus was on studies on MeHg exposure
from fish consumption, the review excluded studies targeting occupational or
environmental Hg only and studies assessing dermal absorption or inhalation of
MeHg.

2.3.1.3 Comparison
Participants with MeHg exposure were compared to individuals with lower MeHg
exposure.

2.3.1.4 Outcomes
Toxic and adverse effects on human health (any outcome) were included.

2.3.1.5 Study designs

Human RCTs or trials; cohort, case-control and cross-sectional studies; systematic
reviews and meta-analyses were included. Any study duration and any number of
subjects in the studies were included.

2.3.2 LITERATURE SEARCH

The literature search for individual primary studies was performed on 15 December
2021. The literature search was performed in two databases, namely PubMed and
Web of Science Core Collection, focusing on literature published since 2010.

26
C H A P T E R 2 . M E T H ODS

The search strategy for each database is given in Appendix 5, Table A5.1 and Table A5.2.
The search terms included a combination of the following:
> MeHg: methylmercury, MeHg, methyl-Hg, CH3Hg;
> study type: specified study design in the inclusion criteria: cohort studies, case-
control studies, cross-sectional studies
In Web of Science, the literature search was performed under the field “topic
search”, which included search words within title, abstract, author keyword and
keywords plus. In PubMed, the literature search included search words within title
and abstract.
All identified records from the searches in each of the databases were imported into
the reference manager program, EndNote (version 20), and duplicate entries were
removed comparing articles with similar “title” and “authors”.

2.3.2.1 Systematic reviews

The search terms used for systematic reviews were identical to those used for finding
relevant primary studies. This literature search was also limited to papers published
since 2010. The search strategy for the updated literature search is given in Appendix 5,
Table A5.1.

2.3.3 SELECTION OF STUDIES FROM THE LITERATURE SEARCH

The studies included from the literature search were further assessed by screening
the titles and abstracts. This was performed independently by two reviewers using
Rayyan, a web-based tool for selecting studies in systematic reviews (Ouzzani
et al., 2016). Studies were selected based on the predefined inclusion and exclusion
criteria (Table 2.10), which were developed based on the research questions and
the information provided in the PECOS diagram (Table 2.9). After screening the
papers by title and abstract, blinding in Rayyan was removed, and the two reviewers
discussed with a third reviewer any conflicting decisions regarding the included or
excluded papers. Subsequently, full texts of the records that fulfilled the inclusion
criteria based on the initial screening were retrieved. The full-text records were
further reviewed in detail according to the same inclusion and exclusion criteria
before proceeding to quality assessment.

27
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 2.10 INCLUSION AND EXCLUSION CRITERIA FOR THE LITERATURE SEARCH ON
“TOXIC EFFECTS OF MeHg”

CRITERIA FOR INCLUSION CRITERIA FOR EXCLUSION

> Studies investigating MeHg exposure in relation to one or > Studies on mercury (Hg) only (unless the study specifically
more health outcomes in a human population described Hg exposure in relation to seafood or fish intake)
> Ingestion (dietary) or transplacental exposure of MeHg > Studies including populations exposed to Hg in working
> Study designs: environments (e.g. miners, dentists)
> systematic reviews > Studies including populations from contaminated areas
> randomized controlled trials (mines, factories) unless data on MeHg exposure was
> non-randomized intervention trials provided and seafood/fish as exposure source was specifically
> cohort studies named
> case-control studies > Studies in which only levels of Hg or MeHg were measured
> cross-sectional studies in human tissues and adverse health effects were not
> Population: considered OR if health effects were reported but tissue or Hg
> general population, all population groups concentration was not measured in blood/hair/toenails
> both sexes, all ages (foetuses, infants, children, > Dermal absorption or inhalation of mercury
adolescents and adults), and pregnant and lactating > Studies where MeHg is only a confounding or stratifying
women, in all countries. variable
> Published from 1 January 2010 onwards > Studies on gene expression or drug metabolizing activity/
levels only
> Comparison with existing health-based guidance values
(HBGV) only
> Study designs:
> animal studies
> in vitro studies
> case reports
> qualitative studies
> other reviews (narrative reviews)
> Publication types: Expert opinions, case studies, editorials,
letters to editors, PhD theses, abstracts, conference
proceedings
> Non-English-language articles

As both systematic reviews and primary studies were included in the literature
review, there was some overlap between the primary studies identified in our
literature search and the primary studies included in the systematic reviews that
were identified in the literature search. To avoid overlap and double reporting of
the primary studies, primary studies from the literature search that were already
included in one of the systematic reviews were excluded from the review. Primary
studies that were excluded for this reason are listed in Appendix 5 (Table A5.5).
Subsequently, systematic reviews and original primary studies (not included in
earlier reviews) were subjected to quality assessment with risk of bias.

28
C H A P T E R 2 . M E T H ODS

2.3.4 QUALITY ASSESSMENT (RISK OF BIAS)

2.3.4.1 Systematic reviews

All systematic reviews that fulfilled the inclusion criteria were quality assessed using
AMSTAR 2 (Shea et al., 2017). Information regarding the quality-assessment tool
AMSTAR 2 is provided in Section 2.1.5.1 and in Appendix 2 (Table A2.1).

2.3.4.2 Primary studies

All primary studies that fulfilled the inclusion criteria (and were not already included
in systematic reviews assigned an AMSTAR 2 rating of high or medium) were
quality assessed using the quality-assessment tool OHAT Risk of Bias Rating Tool.
Information regarding the quality-assessment tool is given in Section 2.2.4.2 and
in Appendix 2 (Table A2.4). Only studies graded Tier 1 were included for further
assessment.

2.3.5 EXTRACTION OF DATA FROM THE INCLUDED LITERATURE

Relevant prespecified data were extracted from the systematic reviews and primary
studies included in the review, into premade data-extraction forms. The extracted
data was double-checked by a different person than the person who extracted the
data.

2.3.5.1 Systematic reviews

The data extracted from the systematic reviews included:
> reference details: reference, year, title;
> outcome: main outcome and specific outcome groups;
> population: type of population and number of participants included;
> study information: study design and number of studies included;
> information regarding MeHg/Hg measurements and MeHg/Hg levels;3
> overall results: including results from meta-analysis if applicable;
> overall conclusion.

2.3.5.2 Primary studies

The data extracted from the primary studies included:
> reference details: reference, trial or study name, geography, year of sampling;
> study information: study design, study duration and follow-up time;
> participants: type of population, number of participants in the study, age (years)
at exposure assessment, sex;

3
Most studies measured Hg, not specifically MeHg. Measurement of Hg is used as a proxy of MeHg exposure.

29
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

> measurements (type of specimen, analytical method) and levels of Hg;

> measurement of fish intake;
> outcome: measurement of outcome and type of outcome assessed; specific
outcome and overall outcome measured;
> overall results;
> overall conclusion.

2.3.6 SUMMARIZATION OF THE LITERATURE

The systematic reviews and primary studies from the literature search covered
a diverse range of health outcomes associated with exposure to MeHg. These
were grouped into four primary domains, namely (i) neurological outcomes, (ii)
cardiovascular outcomes, (iii) growth, and (iv) other outcomes. Narrative summaries
were prepared for each systematic review and primary study, organized into these
same four domains.
In addition to the findings extracted from the systematic reviews and primary
studies, the main findings of two reports – Scientific Opinion on the risk for public
health related to the presence of mercury and methylmercury in food (EFSA,
2012) and Benefit and risk assessment of fish in the Norwegian diet (VKM, 2022)
– were provided for the respective health outcomes. Both reports included reviews
of literature related to the health effects of MeHg exposure. In their 2012 report,
ESFA considered original primary studies only (original primary studies published
between 2004 and 2012), while, in their 2022 report, VKM included a review of
systematic reviews published since 2012.
A weight-of-evidence approach was not given, as the review of “Toxic effects of
MeHg” only included literature published after 2010, and as such it was not possible
to compute an overall grading of the available evidence.

2.4 METHODS FOR THE REVIEW “THE ROLE OF SELENIUM WITH REGARD
TO THE HEALTH EFFECTS OF MeHg”
The objective of the review was to evaluate the role of Se with regard to the health
effects of MeHg. The evaluation was conducted through a systematic search, a
quality assessment, and by summarizing the available literature and weighing the
evidence.

2.4.1 LITERATURE SEARCH STRATEGY

The search strategy was developed based on the research question “What is the role
of Se with regard to the health effects of MeHg?”. The search strategy, defined in a
PICOS table, is set forth in Table 2.11. Studies investigating the role of Se on MeHg
health effects were included.

30
C H A P T E R 2 . M E T H ODS

As few human studies have assessed the role of Se with regard to the health effects of
MeHg, it was decided to include both human and animal studies (mammalian models
systems) in the literature search. The animal studies included from the literature
search were only used as background information for mechanistic and biologically
plausible evidence in the context of the human studies included. Only the human
studies were quality assessed (assessment of risk of bias) and systematically weighed
for evidence.

TABLE 2.11 POPULATION, INTERVENTION, COMPARISON, OUTCOMES AND STUDY DESIGNS TABLE FOR
LITERATURE SEARCH ON “THE ROLE OF SELENIUM WITH REGARD TO THE HEALTH EFFECTS
OF MeHg”

INTERVENTION/
POPULATION COMPARISON OUTCOMES STUDY DESIGNS
EXPOSURE
Human studies: Effects and No exposure to Se or Health outcomes related Human studies:
> General population, measurement of Se on lower exposure to Se to adverse effects of > randomized controlled
including both MeHg health effects. MeHg exposure or toxicity trials
sexes and all > non-randomized
ages (foetuses, intervention trials
infants, children, > cohort studies
adolescents and > case-control studies
adults). > cross-sectional
> Patient groups studies
Animal studies: Animal studies (including
> Mammalian model trials with Se and MeHg
systems, including exposure from ingestion
rodents (rats, mice, (dietary intake)
guinea pigs), dogs
and monkeys

2.4.2 LITERATURE SEARCH

The literature search was performed on 13 December 2021. The literature search was
performed in two databases, namely, PubMed and Web of Science Core Collection,
with no restrictions regarding when the studies were conducted. The search strategy
for each database is given in Appendix 5, Table A5.1.
The search terms included a combination of the following:
> selenium: selenium or HgSe;
> MeHg: methylmercury, MeHg, CH3Hg;
> exposure and outcome: health, benefits, effect, toxic, intervention, uptake,
accumulation, excretion, clearance, response, exposure, stressor, interaction;
> type of population and study group: human, mammal, patient, rat, mouse, rodent,
guinea pig, dog, monkey, men, man, women, woman, child, infants, toddler,
infant, foetus.
All identified records from the searches in each of the databases were imported into
the reference manager program, EndNote (version 20). Duplicates were further
removed in EndNote by comparing articles with similar “title” and “authors”.

31
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

2.4.3 SELECTION OF STUDIES FROM THE LITERATURE SEARCH

The studies included from the literature search were further assessed by screening
the titles and abstracts. This was performed independently by two reviewers using
Rayyan, a web-based tool for selecting studies in systematic reviews (Ouzzani et al.,
2016). The selection of studies was based on the predefined inclusion and exclusion
criteria (Table 2.12), which were developed based on the research questions and the
information provided in the PICOS table (Table 2.11). After screening the papers by
title and abstract, blinding in Rayyan was removed, and the two reviewers discussed
with a third reviewer any conflicting decisions regarding the inclusion or exclusion
of papers. Subsequently, full texts of the records that fulfilled the inclusion criteria
based on the initial screening were retrieved. The full-text records were further
reviewed in detail according to the same inclusion and exclusion criteria before
proceeding to quality assessment.

TABLE 2.12 INCLUSION AND EXCLUSION CRITERIA FOR THE LITERATURE SEARCH
ON “THE ROLE OF SELENIUM WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

CRITERIA FOR INCLUSION CRITERIA FOR EXCLUSION

> Studies investigating the role of selenium with regard to the > Studies conducted on invertebrates
health effects of MeHg. > In vitro studies
> Measured health outcome (e.g. neurodevelopment, > Measurement only of concentrations of Hg and Se in animals
intelligence quotient [IQ], sperm quality, cardiovascular or humans, and not related to health effects of Hg/MeHg
disease) > Studies only measuring indirect health effects (e.g.
> Measured concentrations of Hg/MeHg and Se (e.g. in blood, demethylation of MeHg)
serum, plasma, hair, nails, cord blood) or estimated dietary > Studies in which the effect of Se was not assessed separately
intake of Hg/MeHg and Se but in combination with other covariates or confounders.
> Population:
> Human studies:
» general population, including both sexes and all ages
(foetuses, infants, children, adolescents and adults).
» patient groups
> Animal studies:
» mammalian model systems, including rodents (rats,
mice, guinea pigs), dogs and monkeys.
> Study designs:
> Human studies:
» randomized controlled trials
» non-randomized intervention trials
» cohort studies
» case-control studies
» cross-sectional studies
» case studies/reports
> Animal studies (including trials with Se and MeHg
exposure from ingestion [dietary intake])

32
C H A P T E R 2 . M E T H ODS

2.4.4 QUALITY ASSESSMENT (RISK OF BIAS)

All studies that fulfilled the inclusion criteria were quality-assessed using the OHAT
quality-assessment tool. Information regarding the OHAT quality-assessment tool
is provided in Section 2.2.4.2 and Appendix 2, Table A2.4. Only studies graded Tier
1 or Tier 2 were included for further assessment.

2.4.5 EXTRACTION OF DATA FROM THE LITERATURE

Relevant prespecified data were extracted from the included studies into premade
data-extraction forms in Excel. The data included:
> reference details: reference, trial or study name, geography, year of sampling;
> study information: study type, study duration and follow-up time;
> participants: number of participants in the study, age (years) at exposure
assessment, sex;
> measurements and levels of Hg/MeHg exposure;
> measurements and levels of Se exposure;
> measurements of health outcomes;
> overall results: the role of Se with regard to the health effects of MeHg;
> overall conclusion.
The data extraction was performed by one person, and the extracted data was double
checked by another person.

2.4.6 SUMMARIZATION OF THE LITERATURE INCLUDED

The studies included from the literature search covered a diverse range of health
outcomes where Se and MeHg exposure were measured. These were grouped into
different categories of health outcomes, namely:
> cardiovascular;
> oxidative stress;
> immune system;
> reproduction;
> thyroid hormones;
> birth;
> neurodevelopment and cognition;
> vision function;
> motor function.
The studies included for each health outcome were summarized together. In
addition, the animal studies were summarized together with the relevant health
outcomes, though the animal studies were not included in the weight of evidence.

33
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

2.4.7 WEIGHT OF EVIDENCE OF THE LITERATURE

For a final grading of the available literature, the weight-of-evidence approach was
used, following the guidelines from the WCRF (2018a), where the evidence is graded
“convincing (strong evidence)”, “probable (strong evidence)”, “limited, suggestive”,
“limited, no conclusion”, or “substantial effect of risk unlikely (strong evidence)”.
Information regarding the tool used for weighing of evidence is given in Section 2.1.8.

2.5 METHODS FOR THE REVIEW “OCCURRENCE DATA

FOR MeHg, DIOXINS AND dl-PCBs”
The objective of the review was to evaluate and collect new data for MeHg, dioxins
and dl-PCBs in fishery and aquaculture products published in the last ten years
(2011 to 2021). The data were obtained by conducting a systematic literature search
in the database Web of Science and by extracting data from public databases (WHO’s
GEMS/Food database and EFSA’s Chemical Monitoring database).

2.5.1 LITERATURE SEARCH STRATEGY

The search strategy was developed according to the objective of the systematic
review. Inclusion and exclusion criteria are given in Table 2.13.
For occurrence data on MeHg, analysed data of MeHg or total Hg were included.
The objective was to include data on MeHg; however, studies analysing MeHg are
relatively rare, and total Hg is often used as a proxy for MeHg. Therefore, data on
total Hg were also included.
For occurrence data on dioxins and dl-PCBs, data were included if results were
reported for either: all 29 congeners, all 17 dioxin-congeners, or all 12 dl-PCBs.
Concentration data could only be included if they were reported as upper-bound
TEQ values (WHO, 2005), or in a form that could be transformed to upper-bound
TEQ values (Table 2.13).

2.5.2 LITERATURE SEARCH

Specific literature searches were conducted for Hg or MeHg and for dioxins on 22
November 2021. The literature search was performed in the database Web of Science.
The search strategy is given in Appendix 7, Table A7.1 and Table A7.2. The search
terms included a combination of the following:
> contaminant measured:
> Hg OR mercury OR MeHg OR Me-Hg OR “methyl Hg” OR “methyl-Hg”
OR “methylmercury” OR “methyl-mercury”
> dioxin* OR furan* OR PCDD* OR PCDF* OR “polychlorinated
dibenzodioxin*” OR “polychlorinated dibenzofuran*” OR TEQ;

34
C H A P T E R 2 . M E T H ODS

TABLE 2.13 INCLUSION AND EXCLUSION CRITERIA FOR THE LITERATURE SEARCH
ON “THE ROLE OF SELENIUM WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

CRITERIA FOR INCLUSION CRITERIA FOR EXCLUSION

> Types of species included: > Type of samples excluded:
> finfish (vertebrates) and shellfish (aquatic invertebrates); > marine mammals and algae;
> marine and freshwater origin; > data from feeding trials or exposure studies;
> farmed and wild; > processed samples;
> raw, unprocessed samples; > data where the name of the species or genus is not given;
> edible part analysed: finfish: muscle or whole fish; > data on dry-weight or fat-weight basis, where it was not
shellfish: muscle meat, soft tissue, whole body, or edible possible to recalculate to fresh-weight concentrations.
parts; > Sampling year before 2011 or year of sampling not provided
> individual or composite samples; > Data where quality control of the analytical method was not
> concentrations given on fresh-weight basis and, if not, appropriately described.
fresh weight; concentrations must be possible to calculate
based on dry-weight concentrations and dry-matter
content or lipid-weight concentrations and fat content.
> Occurrence data (analysed data) of:
> MeHg or Hg;
> dioxins (polychlorinated dibenzodioxins and
polychlorinated dibenzofurans) and dioxin-like PCBs
(non-ortho and mono-ortho PCBs). Analysed either all 29
congeners, or all 17 dioxin-congeners, or all 12 dl-PCBs.
Concentration data must be reported as upper-bound
TEQ-values (WHO, 2005), or in a form that can be
transformed to upper-bound TEQ-values.
> Data from samples collected in 2011 or later
> Data must include names of species, which part(s) of the
animal were analysed, and from which geographical area the
species were sampled
> The data had to be analysed in accredited laboratories, or by
a method validated by performance testing (e.g. proficiency
testing or use of appropriate reference material).

Notes: PCB: polychlorinated biphenyl, TEQ: toxic equivalent quotient, WHO: World Health Organization

> fish and seafood species: several typical words for fish and seafood, specific fish
and seafood species;
> measurement of contaminant: concentration* OR level* OR measure* OR
amount* OR value* OR content* OR determin*.
In Web of Science, the literature search was performed with the complete search
string within the search fields title (TI), abstract (AB), and author keyword (AK).
The literature searches were restricted by type of publication (excluding review
articles, proceedings papers, meeting abstracts, corrections, editorial materials,
letters, retracted publications and news items), language (including only publications
in English), and date of publication (1 January 2011 through 31 December 2021).
All identified records from the Web of Science searches were imported into the
reference manager program, EndNote. Duplicates were removed in EndNote by
comparing articles with similar “title” and “authors”.

35
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

2.5.3 SELECTION OF STUDIES FROM THE LITERATURE SEARCH

Screening of titles and abstracts was performed blind by two reviewers using
Rayyan, a web-based tool for selecting studies in systematic reviews (Ouzzani et
al., 2016). The selection of studies was based on predefined inclusion and exclusion
criteria (Table 2.13), which were developed based on the research question. Full-text
articles were included when the information given in the title or abstract seemed
to fulfil the inclusion criteria. After screening the articles by title and abstract, the
blinding was removed, and the two reviewers discussed any conflicting decisions
regarding studies to be included or excluded, consulting a third reviewer as needed.
When consensus was reached on all the studies, full-text articles were retrieved for
all included studies. The full-text articles were further reviewed according to the
inclusion and exclusion criteria.

2.5.4 QUALITY ASSESSMENT

For the quality assessment, a quality-assessment template was prepared using questions
adapted from a quality-assessment system used by the European Food Information
Resource (Oseredczuk et al., 2009). Using this template, the articles were assessed
according to questions in six different categories, as shown in Table 2.14.
Each question was answered with “yes”, “no” or “n.a.” (not applicable). Only
articles that met all the required criteria for inclusion (that is, for which the answer
to all strictly required questions was “yes”) were included. Questions marked by
an asterisk in Table 2.14 were not considered strictly required, and articles could be
included even if the answer was “no” for these two questions.
Because only few articles reported data for dioxins and dl-PCBs, slightly less strict
criteria were used for inclusion of such articles. Specifically, for articles reporting
data for dioxins and dl-PCBs, the following three criteria were modified, as follows:
> Sampling year missing: accepted if the article was published after 2015 (assuming
sampling year 2011 or later for publications after 2015).
> Data given on dry-weight, and dry-matter (or moisture) content not provided:
Accepted. In such cases, an approximate dry-matter content of 25 percent for
both finfish and shellfish was used for calculating wet-weight results.
> Analytical method validation not described: accepted if the method was described
and considered appropriate, even if information about method validation (quality
assessment/quality control) was missing.
The Excel template of the quality assessment is given in Appendix 7 (Table A7.3),
together with the results of the screening and quality assessment for all full-text
articles.

36
C H A P T E R 2 . M E T H ODS

TABLE 2.14 QUESTIONS FOR QUALITY ASSESSMENT OF ARTICLES, ADAPTED FROM THE EUROPEAN FOOD
INFORMATION RESOURCE
1. Food description
> Was year of capture provided?
> Was the food source or main ingredient provided? Species Latin name
> Was the part of animal provided?
> Was analysed portion described and is it clear if food was analysed with or without inedible portion?
> Was the extent of heat treatment/other processing provided?
> Was information on geographic origin of the food provided?
> If results given on dry weight: Was the moisture content of the sample measured and the result provided?
> If dioxin results given on lipid weight only, was fat content provided?
2. Component identification
> Is the component described unambiguously?
> Is the unit unequivocal?
> Is the matrix unit unequivocal?
3. Sampling plan
> Was the number of primary samples provided?
4. Number of analytical samples
> Was the number of analytical samples provided?
5. Sample handling
> Were the samples homogenized? *
6. Analytical method and performance
> Were analytical sample replicates tested? *
> Was the laboratory accredited for this method? (Stop here if the answer is yes, if no answer next questions).
> Was the analytical method used in the source appropriate? Was the method validated by one or more of the following:
> Was the method validated by an in-house validation study?
> Was the method validated by performance testing (PT schemes, proficiency testing)?
> Was the method validated by a collaborative study?
> Was the method validated using an appropriate reference material (analyte, concentration, matrix)?
> Was the method validated by a recovery study leading to a relative standard deviation ≤ 20 percent?

Note: Questions marked by an asterisk (*) were not considered strictly required, and the article could be included even if the answer was
“no” for these two questions.
Source: Oseredczuk, M., Salvini, S., Roe, M. & Moller, A. 2009. EuroFIR Workpackage 1.3., Task group 4. Guidelines for Quality Index
Attribution to original data from scientific literature or reports for EuroFIR data interchange. https://www.eurofir.org/wp-admin/wp-
content/uploads/Deliverables/EuroFIR_Quality_Index_Guidelines.pdf. Accessed 26th October 2022.

2.5.5 EXTRACTION OF DATA

For the articles included, the concentrations and metadata were extracted and
compiled in an Excel data sheet (Appendix 7, Table A7.4). The data retrieved
included: year(s) of sampling, geographic origin, species identification, subgroup
(finfish/shellfish), tissue (fillet, whole fish, muscle, soft tissue, edible parts), wild/
farmed, sampling method (field sampling, purchased from fishermen or market),
habitat (river, lake, ocean or estuary), number of primary and analytical samples,
sample type (individual or pooled), matrix unit (wet weight or dry weight), mean
value, median value, minimum value and maximum value. Where the results were
given in dry weight, results for dry-matter content were also included. None of the
articles included had data given on the basis of lipid weight.

37
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

2.5.6 DATA FROM PUBLIC DATABASES

FAO and EFSA were contacted on 26 October 2021 to request access to data on
total Hg, MeHg, dioxins and dl-PCBs in seafood from 2011 onwards. Data on Hg
in food (including seafood) were received from EFSA on 21 June and 14 October
2022, and data on dioxins and dl-PCBs in food were received on 24 August and 11
November 2022. Via FAO/WHO, access was granted to GEMS/Food data on Hg,
MeHg, dioxins and dl-PCBs in seafood. Data on Hg and MeHg were accessed 15
September 2022, and data on dioxins and dl-PCBs were accessed 28 September 2022.

2.5.6.1 Extraction of relevant data from European Food Safety Authority datasets
The data from EFSA were assessed using the exclusion and inclusion criteria shown
in Table 2.15.

TABLE 2.15 INCLUSION AND EXCLUSION CRITERIA FOR DATA, FROM THE EUROPEAN FOOD SAFETY
AUTHORITY DATABASE

CRITERIA FOR INCLUSION CRITERIA FOR EXCLUSION

> Type of samples included: > Type of samples excluded:
> finfish and shellfish; > non-seafood;
> marine and freshwater origin; > fish liver, fish roe, fish offal, fish oil, meal from fish or
> farmed and wild; mollusc (feed), fish oils, other seafood products for feed;
> raw, unprocessed samples; > processed or preserved products (dried, canned,
> edible part analysed: finfish (muscle meat); shellfish marinated/pickled, smoked, etc.);
(edible parts); > samples where name of species was not given or sample
> individual or composite samples. description was too general (e.g. fish meat, marine fish,
> Occurrence data (analysed data) of: molluscs, crustaceans);
> MeHg or total Hg; > samples analysed for inorganic mercury only;
> dioxins (polychlorinated dibenzodioxins and > samples missing information about whether data were
polychlorinated dibenzofurans) and dioxin-like PCBs given on fresh-weight or lipid-weight basis and samples
(non-ortho and mono-ortho PCBs); analysed either all 29 where it was not possible to recalculate fresh-weight
congeners, or all 17 dioxin-congeners, or all 12 dl-PCBs; values (fat content missing);
concentration data reported as raw data that can be > samples analysed for dioxins or dioxin-like PCBs, for
transformed to upper-bound TEQ-values; which less than 17 dioxin-congeners or less than 12 dl-
> concentrations given on fresh-weight basis and if not, PCBs were reported;
fresh-weight concentrations must be possible to calculate > Sampling year before 2011.
based on dry-weight concentrations and dry-matter
content or lipid-weight concentrations and fat content.
> Data from samples collected in the year 2011 or later
> Data including name of species or group of species and
information about geographical sampling area.

Notes: PCB: polychlorinated biphenyl, dl-PCB: dioxin-like polychlorinated biphenyl, TEQ: toxic equivalent quotient.

38
C H A P T E R 2 . M E T H ODS

2.5.6.2 Processing of data included from the European Food Safety Authority
dataset
Samples where results were given on fat-weight basis were converted to fresh
weight using the fat content of the sample.
Most data for dioxins and dl-PCBs in the EFSA dataset were reported as raw
concentrations, and only a small subset of the samples also had concentrations
reported as upper-bound TEQ-values for sum dioxins and dl-PCBs. For
consistency, the individual raw congener concentrations were used for all
samples. TEQ values for individual congeners were calculated using the WHO
2005 TEF values for each congener (Van den Berg et al., 2006). Upper-bound
sums of 17 dioxins and furans (sum dioxin), 12 dl-PCBs (sum dl-PCB) and all
29 dioxins and dl-PCBs (sum dioxin and dl-PCB) were calculated using limit
of quantification (LOQ) values (or limit of detection [LOD] values when LOQ
value was not given) for congeners with concentrations below the LOQ (or
LOD). Three samples of blue mussel, three samples of Atlantic salmon, one
sample of Japanese seabass, one sample of sea bream, and one sample of trout
with very high concentrations of upper-bound sum dioxins were considered
outliers and excluded from the dataset because the data showed that the high
concentrations were caused by incorrectly reported (much too high) LOQ values
for the dioxin congeners. The results for these samples originated from a single
laboratory in a single year.
Three samples (blue whiting, clams, shrimps and prawns) were excluded from
the dataset because of extreme values of LOQ for total Hg (10–70 mg/kg).
These extreme LOQ values may have been reported with an erroneous unit of
measurement (µg/kg values reported as mg/kg).
The EFSA dataset contained information about geographic origin either as a
description of marine area, inland water area or country of origin for all included
samples. This information was used to assign the samples to specific major
fishing areas as classified by FAO (FAO-areas). In cases where a specific FAO
area could not be assigned (for instance, because the geographic information
was given in very broad terms or the country of origin was bordered by more
than one FAO area), the samples were categorized as unspecified (for example,
“Pacific Ocean, unspecified” or “France, unspecified”).
The samples were categorized as inland or marine based on the geographic
information given. In cases where the exact geographic location was unclear, the
samples were categorized as inland or marine when sample descriptions elsewhere
in the dataset indicated “freshwater fish” or “marine fish”, respectively. Where
no such information was found, the samples were categorized as “unspecified”.
Most of the samples originated from marine waters, and it is probable that most
of the “unspecified” samples were from marine waters rather than from inland
waters. As such, data for samples from marine and unspecified waters were
combined in the final tables.

39
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

In order to categorize samples as wild or farmed, finfish/shellfish species

information and information about sampling point (SAMPPOINT_ID)
given in the EFSA dataset was used. Samples with SAMPPOINT_ID given
as “Aquaculture”, “Breeding”, “Farm”, “Hatchery”, “Rearing of animals” or
“Slaughterhouse” were categorized as farmed finfish/shellfish, while samples
with SAMPPOINT_ID given as “Fishery activities”, “Fishing”, “Fishing and
processing”, “Game handling establishment”, “Hunting” or “Natural habitat”
were categorized as wild finfish/shellfish. In addition, some finfish species or
species groups which are generally known not to be used in aquaculture were
categorized as wild fish. Many of the remaining samples were, however, sampled
at border-control posts; from distribution, wholesale or retail sale; during
processing activities or storage; in restaurants or in other parts of the value
chain, giving no information indicating whether the samples were from wild
or farmed finfish/shellfish. These samples were categorized as “unspecified”,
except for some samples where information about the origin of the samples
from aquaculture were found in other parts of the dataset. Most of the samples
were from wild finfish/shellfish, and most of the unspecified samples are thus
likely to be wild finfish/shellfish. Data for wild and “unspecified” samples were,
therefore, combined in the final results.
The dataset with all the included samples from EFSA is given in Appendix 7,
Table A7.5.

2.5.6.3 Extraction of relevant data from the Global Environment Monitoring

System/Food datasets of the World Health Organization
Relevant data from the GEMS/Food datasets were extracted after removal of
the following types of samples:
> samples collected before 2011;
> samples of processed products (such as breaded haddock and raw breaded
squid rings) or samples where the food state (FoodStateName) was unknown;
> samples where the sample description was too general (for instance, fish
muscles);
> samples from marine mammals;
> samples analysed for inorganic Hg only.
After removal of the samples according to these exclusion criteria, there were no
samples left to include from the GEMS/Food dataset on dioxins and dl-PCBs.
For the GEMS/Food dataset on Hg and MeHg, 5 826 samples were included,
5 304 samples were analysed for total Hg and 522 samples were analysed for
MeHg. The dataset contained very little information about the geographic origin
of the samples, and only three different region codes (“PAHO”, “SEARO” and
“WPRO”) were given in the dataset. In the WHO region Codelist, “SEARO”
and “WPRO” were found to be acronyms for Southeast Asia Region and Western
Pacific Region, respectively, whereas the region code “PAHO” was not listed in

40
C H A P T E R 2 . M E T H ODS

the WHO region Codelist. Since the geographic origin of the samples was given
only within these very wide geographical regions, it was not possible to assign
the samples to their respective FAO areas. Information needed to categorize
the samples as farmed or wild was also missing in the dataset. Because of these
limitations, and because this dataset contained a relatively small number of
samples compared to the data from the literature review and the EFSA database,
these data were not processed further in this study. The dataset with all the
included samples is, however, available in Appendix 7, Table A7.6.

41
 42
© FAO/Massimo Berruti
CHAPTER 3
RESULTS AND
SUMMARIZATION OF
THE LITERATURE REVIEW
“EVIDENCE OF HEALTH
BENEFITS FROM FISH
CONSUMPTION”

3.1 LITERATURE SEARCH AND QUALITY ASSESSMENT

Literature searches for the systematic review on the “Evidence of health benefits from
fish consumption” were performed in the databases PubMed, EMBASE and Cochrane,
with no lower limit for the year of publication. A literature search was performed
for each category of health outcomes that were defined (Table 2.2). Figure 3.1
provides a flow diagram of the results from the literature searches regarding all the
themes in “Evidence of health benefits from fish consumption”. The literature searches
resulted in 39 092 records. Of these, 5 401 duplicate records were removed, leaving
33 691 records that were assessed via title and abstract screening in Rayyan. Based
on title and abstract screening, 32 895 records were excluded, and 791 records were
assessed in full text. In the full-text assessment, 665 records were excluded for further
assessment based on criteria, and 127 records (one risk/benefits assessment [the 2022
VKM report], 32 systematic reviews and 94 primary studies) were quality assessed
with risk-of-bias tools. As a result, 12 systematic reviews and 60 primary studies
were excluded owing to the low quality of the studies. Thus, the final review included
one risk/benefits assessment (the 2022 VKM assessment), 22 systematic reviews and
47 primary studies. The results from the literature search for each theme of health
outcomes are given in Table 3.1 and further summarized in the following paragraphs.

43
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

FIGURE 3.1. FLOW DIAGRAM FOR THE REVIEW ON THE “EVIDENCE OF HEALTH BENEFITS FROM FISH
CONSUMPTION”

IDENTIFICATION OF STUDIES IDENTIFICATION OF STUDIES

VIA DATABASES VIA OTHER METHODS

Records identified from Records removed before Records identified from:

IDENTIFICATION

database searching (PubMed, screening: Reports

Web of Science and Cochrane): Duplicate records removed (n = 1, VKM 2022)
(n = 47 252) in EndNote (n = 5 416)

Records screened by title Records excluded by eligibility

and abstract (n = 41 836) criteria (n = 40 443)
SCREENING

Full-text articles assessed Records excluded by eligibility

for eligibility (n = 1 328) criteria or inclusion in other
systematic reviews (n = 1 187)

Quality assessment (risk of bias) Records excluded after quality

of full-text articles (n = 141): assessment (risk of bias)
 Systematic reviews (n = 34) Systematic reviews (n = 12)
 Primary studies (n = 107) Primary studies (n = 60)
INCLUDED

Risk benefits assessment available (n = 1)

Systematic reviews and meta-analyses included in final review (n = 22)
Primary studies included in final review (n = 47)

Source: The figure was prepared based on Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow,
C.D., Shamseer, L. et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.
BMJ, 372: n71. https://doi.org/10.1136/bmj.n71

44
 TABLE 3.1 RESULTS OF THE LITERATURE SEARCH ON “EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

2. NUMBER OF 3. DUPLICATE RECORDS 5. RECORDS EXCLUDED

1. TOTAL HITS (WEB 4. TOTAL RECORDS
DUPLICATES REMOVED REMOVED DURING BASED ON CRITERIA 6.RECORDS SOUGHT FOR
HEALTH OUTCOME OF SCIENCE, PUBMED, SCREENED IN RAYYAN
IN ENDNOTE BEFORE ABSTRACT SCREENING DURING ABSTRACT FULL-TEXT SCREENING
COCHRANE) (1– [2+3])
SCREENING IN RAYYAN SCREENING IN RAYYAN
Allergy and immunology 5 136 551 113 4 472 4 314 158
Birth and growth outcomes 5 973 369 5 5 599 5 564 35
Bone health 693 86 14 593 549 44
Cancer 3 831 710 82 3 039 2 998 41
Cardiovascular disease 3 717 426 58 3 233 3 067 166
Dental health 19 2 0 17 17 0
Type 2 diabetes 2 223 290 173 1 760 1 700 60
Neurodevelopment and
8 612 1 277 4 7 331 7 214 117
neurological disorders
Mortality 3 793 583 59 3 151 3 026 125
Overweight and obesity 5 095 534 65 4 496 4 458 38
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Total number of records 39 092 4 828 573 33 691 32 902 784

45
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.1.1 ALLERGY AND IMMUNOLOGY

Literature searches for the theme “Allergy and immunology” were performed in
PubMed, EMBASE and Cochrane, with no lower limit for the year of publication.
The literature searches resulted in 5 136 records. Before screening, 608 duplicates
were removed in EndNote and 113 duplicates were removed in Rayyan. Thus, 4
472 records were screened by title and abstract, using the online screening tool
Rayyan. As a result of this assessment, based on inclusion and exclusion criteria, 4
314 records were excluded. Thus, 158 records were assessed for full-text screening.
Of these, 16 records were systematic reviews and 142 records were primary studies.

3.1.1.1 Systematic reviews

Sixteen reviews were assessed in full-text after title and abstract screening. Of
these, two reviews were excluded, based on inclusion and exclusion criteria, and six
reviews were excluded as they were already assessed in the risk/benefits assessment
in the 2022 VKM report (see list of excluded reviews in Appendix 3, Table A3.2).
Thus, eight systematic reviews were assessed for risk of bias using the tool AMSTAR
2, according to the overall confidence in the results of the systematic review. One
review was graded “high” confidence, four were graded “moderate” confidence,
one was graded “low” confidence, and two were graded “critically low” confidence
(judgement and references given in Appendix 3, Table A3.20). The reviews graded
“low” or “critically low” were excluded. As such, after the risk-of-bias assessment,
five systematic reviews were included for further assessment in this review (Pattison
et al., 2004; Di Giuseppe et al., 2014; Netting et al., 2014; Ierodiakonou et al., 2016;
and Venter et al., 2020).

3.1.1.2 Primary studies

After the title and abstract screening, 142 primary studies remained and were
assessed in full text. Of these, 78 primary studies were excluded, based on inclusion
and exclusion criteria, and 42 primary studies were excluded as they were already
assessed in one of the systematic reviews included in this review (see list of
excluded primary studies in Appendix 3, Table A3.3). Thus, 22 primary studies
were quality assessed with the risk-of-bias tool. All 22 primary studies were graded
“C (low quality)” with the risk-of-bias tool and therefore were excluded for further
assessment (Appendix 3, Table A3.21).

3.1.2 BIRTH AND GROWTH OUTCOMES

Literature searches for the theme “Birth and growth” were performed in PubMed,
EMBASE and Cochrane, with no lower limit for the year of publication.
The literature searches resulted in 5 973 records. Before screening, 369
duplicates were removed in EndNote and 5 duplicates were removed in Rayyan.
Thus, 5 599 records were assessed in title and abstract screening using the online
screening tool, Rayyan. As a result of this screening, and based on inclusion and

46
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

exclusion criteria, 5 564 records were excluded in Rayyan. Thus, 35 records remained
and were assessed in full text. Of these, one record was a systematic review, and 34
records were primary studies.

3.1.2.1 Systematic reviews

One review was assessed in full text after title and abstract screening. This review,
by Hibben et al. (2019), was also excluded in this screening, based on inclusion and
exclusion criteria (Appendix 3, Table A3.4).

3.1.2.2 Primary studies

After the title and abstract screening, 34 primary studies were assessed in full text.
Of these, 11 were excluded based on inclusion and exclusion criteria and 20 were
excluded as they were already assessed in one of the systematic reviews included (see
list of excluded primary studies in Appendix 3, Table A3.5). Thus, three primary
studies were quality assessed with the risk-of-bias tool. Two studies were graded “B
(moderate quality)”, while one study was graded “C (low quality)” (see Appendix
3, Table A3.22). The study graded “C” was excluded, leaving two primary studies,
which were included for further assessment in this review (Oken et al., 2004; Zhao
et al., 2022).

3.1.3 BONE HEALTH

Literature searches for the theme “Bone health” were performed in PubMed,
EMBASE and Cochrane, with no lower limit for the year of publication. The
literature searches resulted in 693 records. Before screening, 86 duplicates were
removed in EndNote and 14 duplicates were removed in Rayyan. Thus, 593 records
were assessed through title and abstract screening, using the online screening tool,
Rayyan. In the title and abstract screening, 549 records were excluded based on
inclusion and exclusion criteria. Thus, 44 records were assessed in full-text screening.
Of these, 7 records were systematic reviews and 37 records were primary studies.

3.1.3.1 Systematic reviews

Seven reviews were assessed in full text after the title and abstract screening. Of
these, five were excluded based on inclusion and exclusion criteria and two were
excluded as they were already assessed in the 2022 VKM risk/benefit assessment (see
list of excluded reviews in Appendix 3, Table A3.6). Thus, no systematic reviews
were further quality assessed with risk of bias or included for further assessment
in this review.

3.1.3.2 Primary studies

After the title and abstract screening, 37 primary studies were assessed in full text.
Of these, 17 were excluded based on inclusion and exclusion criteria and nine were
excluded as they were already assessed in one of the systematic reviews included

47
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

(see list of excluded primary studies in Appendix 3, Table A3.7). Thus, 11 primary
studies were quality assessed with the risk-of-bias tool. Four studies were graded
“B (moderate quality)”, while seven were graded “C (low quality)” (Appendix 3,
Table A3.23). The study graded “C” was excluded for further assessment, leaving
four primary studies that were included for further assessment in this review (Hirota
et al., 2005; Lucey et al., 2008; Thacher et al., 2015 and Tong et al., 2020).

3.1.4 CANCER
Literature searches for the theme “Cancer” were performed in PubMed, EMBASE
and Cochrane, with no lower limit for the year of publication. The literature searches
resulted in 3 831 records. Before screening, 710 duplicates were removed in EndNote
and 82 duplicates were removed in Rayyan. Thus, 3 039 records were assessed
through title and abstract screening, using the online screening tool, Rayyan. As a
result, 2 998 records were excluded in Rayyan, based on inclusion and exclusion
criteria. Thus, 41 records were assessed through full-text screening. Of these, 21
were systematic reviews and 20 were primary studies.

3.1.4.1 Systematic reviews

After the title and abstract screening, 21 reviews were assessed in full text. Of these,
18 were excluded based on inclusion and exclusion criteria (see list of excluded
reviews in Appendix 3, Table A3.8). Thus, three systematic reviews were assessed
for risk-of-bias, using the tool AMSTAR 2 according to the overall confidence in
the results of the systematic review. Two reviews were graded “high” confidence,
and one was graded “moderate” confidence (judgement and references given in
Appendix 3, Table A3.24). Thus, all the three systematic reviews were included for
further assessment in this review (Jayedi et al.; 2020, Kazemi et al., 2021 and Gao
et al., 2022).

3.1.4.2 Primary studies

After the title and abstract screening, 20 primary studies were assessed in full text.
Of these, eight were excluded based on inclusion and exclusion criteria and two were
excluded as they were already assessed in one of the systematic reviews included in
this review (see list of excluded primary studies in Appendix 3, Table A3.9). Thus,
ten primary studies were quality assessed with the risk-of-bias tool. All ten studies
were graded “B (moderate quality)” (Appendix 3, Table A3.25), and, as such, ten
primary studies were included for further assessment in this review (Etemadi et al.,
2018; Outzen et al.; 2018, Ma et al., 2019; Aglago et al., 2020; Bradbury et al., 2020;
Cai et al., 2020; Makiuchi et al., 2020; Zamani et al., 2020; Dianatinasab et al., 2021
and Hermans et al., 2021).

48
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.1.5 CARDIOVASCULAR DISEASES AND OUTCOMES

Literature searches for the theme “Cardiovascular diseases and outcomes” were
performed in PubMed, EMBASE and Cochrane, with no lower limit for the year
of publication. The literature searches resulted in 3 717 records. Before screening,
426 duplicates were removed in EndNote and 58 were removed in Rayyan. Thus,
3 233 records were assessed through title and abstract screening using the online
screening tool, Rayyan. In this screening, based on inclusion and exclusion criteria,
3 067 records were excluded, leaving 166 records, which were assessed in full text.
Of these, 16 were systematic reviews, and 150 were primary studies.

3.1.5.1 Systematic reviews

After the title and abstract screening, 16 reviews were assessed in full text. Of these, 13
reviews were excluded based on inclusion and exclusion criteria (see list of excluded
reviews in Appendix 3, Table A3.10). Thus, three systematic reviews were assessed
for risk of bias, using the tool AMSTAR 2, according to the overall confidence in the
results of the systematic review. Two reviews were graded “moderate” confidence,
and one was graded “critically low” confidence (judgement and references given in
Appendix 3, Table A3.26). The review graded “critically low” was excluded, leaving
two systematic reviews, which were included for further assessment in this review
(Mente et al., 2009 and Chowdhury et al., 2012).

3.1.5.2 Primary studies

After the title and abstract screening, 150 primary studies were assessed in full
text. Of these, 57 primary studies were excluded based on inclusion and exclusion
criteria and 83 were excluded as they were already assessed in one of the systematic
reviews included (see list of excluded primary studies in Appendix 3, Table A3.11).
Thus, ten primary studies were quality assessed with the risk-of-bias tool. All ten
studies were graded “B (moderate quality)” (Appendix 3, Table A3.27) and, as
such, all ten were included for further assessment in this review (Frost et al., 2005;
Matheson et al., 2009; Lajous et al., 2013; Gammelmark et al., 2016; Venø et al.,
2018; Lasota et al., 2019; Tong et al., 2019; Acosta et al., 2021; Petermann-Rocha et
al., 2021 and Zhong et al., 2021).

3.1.6 TYPE 2 DIABETES

Literature searches for the theme “Type 2 diabetes” (T2D) were performed in
PubMed, EMBASE and Cochrane, with no lower limit for the year of publication.
The literature searches resulted in 2 223 records. Before screening, 290 duplicates
were removed in EndNote and 173 duplicates were removed in Rayyan. Thus, 1 760
records were assessed through title and abstract screening, using the online screening
tool, Rayyan. In this screening, 1 700 records were excluded based on inclusion and
exclusion criteria. Thus, 60 records remained for full-text screening. Of these, 19
were systematic reviews and 41 were primary studies.

49
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.1.6.1 Systematic reviews

After the title and abstract screening, 19 reviews were assessed in full text. Of
these, six were excluded based on inclusion and exclusion criteria and four were
excluded as they were already assessed in the 2022 VKM risk/benefit assessment
(see list of excluded reviews in Appendix 3, Table A3.12). Thus, nine systematic
reviews were assessed for risk of bias, using AMSTAR 2, according to the overall
confidence in the results of the systematic review. Five reviews were graded “high”
confidence, two were graded “moderate” confidence, and two were graded “critically
low” confidence (judgement and references given in Appendix 3, Table A3.28).
The reviews graded “critically low” were excluded, leaving seven systematic reviews,
which were included for further assessment in this review (Wallin et al., 2012; Wu
et al., 2012; Xun et al., 2012; Zheng et al., 2012; Zhou et al., 2012; Zhang et al., 2013
and Muley et al., 2014).

3.1.6.2 Primary studies

After the title and abstract screening, 41 primary studies were assessed in full
text. Of these, 24 were excluded based on inclusion and exclusion criteria and
16 were excluded as they were already assessed in one of the systematic reviews
already included (see list of excluded primary studies in Appendix 3, Table A3.13).
Thus, one primary study was quality assessed with the risk-of-bias tool and this
was graded as “B (moderate quality)” (Appendix 3, Table A3.29). After the risk-
of-bias assessment, only this primary study was included for further assessment in
this review (Chen et al., 2020).

3.1.7 NEURODEVELOPMENT AND NEUROLOGICAL DISORDERS

Literature searches for the theme “Neurodevelopment and neurological disorders”
were performed in PubMed, EMBASE and Cochrane, with no lower limit for
the year of publication. The literature searches resulted in 8 612 records. Before
screening, 1 277 duplicates were removed in EndNote and 4 duplicates were
removed in Rayyan. Thus, 7 331 records were assessed through title and abstract
screening, using the online screening tool, Rayyan. As a result of this screening,
7 214 records were excluded in Rayyan based on inclusion and exclusion criteria.
Thus, 117 records were assessed in full text. Of these, 23 were systematic reviews,
and 94 were primary studies.

3.1.7.1 Systematic reviews

After the title and abstract screening, 23 reviews were assessed in full text. Of these,
nine were excluded based on inclusion and exclusion criteria and 11 were excluded
as they were already assessed in the 2022 VKM risk/benefit assessment (see list of
excluded reviews in Appendix 3, Table A3.14). Thus, three systematic reviews were
assessed for risk of bias with the tool AMSTAR 2, according to the overall confidence
in the results of the systematic review. Two reviews were graded “low” confidence,

50
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

and one was graded “critically low” confidence (judgement and references given
in Appendix 3, Table A3.30). The reviews graded “low” or “critically low” were
excluded, and as such, none of the systematic reviews were included for further
assessment in this review.

3.1.7.2 Primary studies

After the title and abstract screening, 94 primary studies were assessed in full text.
Of these, 22 primary studies were excluded based on inclusion and exclusion criteria
and 58 were excluded as they were already assessed in one of the systematic reviews
included (see list of excluded primary studies in Appendix 3, Table A3.15). Thus, 14
primary studies were quality assessed with the risk-of-bias tool. Two studies were
graded “B (moderate quality)”, while ten were graded “C (low quality)” (Appendix
3, Table A3.31). The studies graded “C” were excluded for further assessment,
leaving two primary studies to be further assessed in this review (Mesirow et al.,
2017 and Al-Ghannami et al., 2019).

3.1.8 MORTALITY
Literature searches for the theme “Mortality” were performed in PubMed, EMBASE
and Cochrane, with no lower limit for the year of publication. The literature searches
resulted in 3 793 records. Before screening, 583 duplicates were removed in EndNote
and 59 duplicates were removed in Rayyan. Thus, 3 151 records were assessed
through title and abstract screening, using the online screening tool, Rayyan. After
this screening, 3 026 records were excluded in Rayyan based on inclusion and
exclusion criteria, leaving 125 records for full-text screening. Of these, 24 were
systematic reviews and 101 were primary studies.

3.1.8.1 Systematic reviews

After the title and abstract screening, 24 reviews were assessed in full text. Of these,
eight were excluded based on inclusion and exclusion criteria and 11 were excluded
as they were already assessed in the 2022 VKM risk/benefit assessment (see list of
excluded reviews in Appendix 3, Table A3.16). Thus, five systematic reviews were
assessed for risk of bias using AMSTAR 2, according to the overall confidence in
the results. Three reviews were graded “moderate” confidence, and two were graded
“low” confidence (judgement and references given in Appendix 3, Table A3.31).
The reviews graded “low” were excluded, and three systematic reviews remained for
further assessment in this review (He et al., 2004; Geelen et al., 2007 and Szymanski
et al., 2010).

3.1.8.2 Primary studies

After the title and abstract screening, 101 primary studies were assessed in full text.
Of these, 40 were excluded based on inclusion and exclusion criteria and 45 were
excluded as they were already assessed in one of the systematic reviews included

51
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

in this review (see list of excluded primary studies in Appendix 3, Table A3.17).
Thus, 16 primary studies were quality assessed using the risk-of-bias tool. Five
studies were graded “B (moderate quality)”, while 11 studies were graded “C (low
quality)” (Appendix 3, Table A3.33). The studies graded as “C” were excluded
for further assessment, leaving five primary studies for further assessment in this
review (Walda et al., 2002; Iso et al., 2006; Streppel et al., 2008; Pertiwi et al., 2021
and Sun et al., 2021).

3.1.9 OVERWEIGHT AND OBESITY

Literature searches for the theme “Overweight and obesity” were performed in
PubMed, EMBASE and Cochrane, with no lower limit for the year of publication.
The literature searches resulted in 5 095 records. Before screening, 534 duplicates
were removed in EndNote and 65 were removed in Rayyan. Thus, 4 496 records
were assessed through title and abstract screening, using the online screening tool,
Rayyan. As a result of this screening, 4 458 records were excluded in Rayyan based
on inclusion and exclusion criteria, leaving 38 records to be assessed in full text: 3
systematic reviews and 35 primary studies.

3.1.9.1 Systematic reviews

After the title and abstract screening, three reviews were assessed in full text. One
review was excluded based on inclusion and exclusion criteria, and two reviews were
excluded as they were already assessed in the 2022 VKM risk/benefit assessment
(see list of excluded reviews in Appendix 3, Table A3.18). Thus, no systematic
reviews were further quality assessed with risk-of-bias tools or included for further
assessment in this review.

3.1.9.2 Primary studies

After the title and abstract screening, 35 primary studies were assessed in full
text. Of these, 20 primary studies were excluded based on inclusion and exclusion
criteria and 8 were excluded as they were already assessed in one of the systematic
reviews included (see list of excluded primary studies in Appendix 3, Table A3.19).
Thus, seven primary studies were quality assessed with the risk-of-bias tool. Three
studies were graded “B (moderate quality)”, while four studies were graded “C (low
quality)” (see Appendix 3, Table A3.34). The studies graded “C” were excluded
for further assessment, leaving three primary studies that were included for further
assessment in this review (Smith et al., 2015; Tørris et al., 2017 and Beulen et al.,
2018).

52
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2 RESULTS AND SUMMARIZATION OF THE INCLUDED LITERATURE

For each health outcome, the results of the review “Evidence of health benefits from
fish consumption” are divided into the following sections: i) summary from the 2022
VKM report; ii) summary from the systematic reviews included from the literature
search; and iii) summary from the primary studies included from the literature search.
Moreover, for each health outcome, an overall summary of all the literature included
– the 2022 VKM report, the systematic reviews and the primary studies – was made.
Finally, a final weight-of-evidence analysis was performed for each health outcome
(summary provided in Section 3.3). An overview of the literature included in the final
weight of evidence for each health outcome is given in Appendix 3, Table A3.35.

3.2.1 ALLERGY AND IMMUNOLOGY

The literature included in the theme “Allergy and immunology” includes results
from the 2022 VKM report, and five systematic reviews from the literature search.
No original primary studies were included from the literature search for this theme.

3.2.1.1 Summary of the findings on “Allergy and immunology” in the VKM report,
Benefit and risk assessment of fish in the Norwegian diet
The 2022 VKM report, Benefit and risk assessment of fish in the Norwegian
diet, summarized the evidence of an association between fish consumption and
the outcomes of “allergic rhinitis”, “allergic sensitization in children”, “asthma in
children”, “eczema in children”, “multiple sclerosis”, and “rheumatoid arthritis”.

3.2.1.1.1 Allergic rhinitis in children

3.2.1.1.1.1 Description of the studies included
The 2022 VKM report includes two systematic reviews and one pooled analysis
investigating the association between fish intake and allergic rhinitis in children as
an outcome. The associations of fish intake during pregnancy and allergic rhinitis
reported in the pooled analysis by Stratakis et al. (2017) (RR = 1.01, 95% CI: 0.99,
1.03) and in the systematic reviews by Malmir et al. (2021) (RR = 0.91, 95% CI:
0.75, 1.09) and by Zhang et al. (2017) (RR = 0.95, 95% CI: 0.62-1.45), were not
statistically significant. The systematic review by Zhang et al. (2017) also included
the outcome of early introduction of fish and risk of allergic rhinitis in children,
finding a reduced risk with higher fish intake (RR = 0.54, 95% CI: 0.36, 0.81).
The VKM report included five original primary studies, including one pooled
analysis, with allergic rhinitis in children or adolescents as an outcome in relation
to either maternal or child fish intake. All studies had a prospective observational
design (birth cohort, or cohort based on intervention study) with rhinitis or rhino
conjunctivitis as the outcome. The body of evidence on allergic rhinitis consisted
of European studies and one US cohort. Most of the included studies reported no
associations between fish consumption and risk of allergic rhinitis.

53
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.1.1.1.2 Conclusion, weight of evidence

The evidence that maternal fish intake during pregnancy affects the risk of rhinitis
in the offspring was graded “limited, no conclusion” since no overall statistically
significant differences were found in one pooled analysis and two meta-analyses,
and no clear dose-response relationship was found.
The evidence that child fish intake in the first year protects against allergic rhinitis
was graded “limited, no conclusion”. The conclusion was based on one meta-
analysis and three primary studies on early fish introduction.

3.2.1.1.2 Allergic sensitization in children

3.2.1.1.2.1 Description of the studies included
The 2022 VKM assessment included four primary studies regarding allergic
sensitization in children in relation to fish intake by mothers during pregnancy
(two studies) or child intake (two studies). All four studies included were European
birth cohort studies. Sensitization was diagnosed as high serum immunoglobulin
(Ig) E to any food or inhalant allergen, or as a positive skin prick test (SPT) at ages
2 to 8 years.
The VKM report also included two systematic reviews on maternal fish intake in
pregnancy and risk of allergic sensitization.
The four primary studies and two systematic reviews were used to evaluate
the evidence for an association of fish intake with the development of allergic
sensitization in children.

3.2.1.1.2.2 Conclusion, weight of evidence

The evidence that maternal fish intake during pregnancy affects the risk of allergic
sensitization to food or inhalant allergens was graded “limited, no conclusion” by
VKM.
The evidence that child fish intake protects against sensitization was also graded
“limited, no conclusion”.

3.2.1.1.3 Asthma in children

3.2.1.1.3.1 Description of the studies included
The VKM assessment included eight primary studies, three with results on
pregnancy intake, two on intake during lactation, and five on intake in children. All
studies were prospective observational studies (birth cohort, nested case-control, or
cohort based on intervention study). The body of evidence on asthma consisted of
European studies and one US cohort. All studies included total fish intake. From the
primary studies included, VKM calculated summary RR. No significant association
of maternal intake during pregnancy and asthma in school-age children was reported
by VKM (RR = 1.16, 95% CI: 0.88, 1.54). Further, no conclusions can be drawn
regarding a differential effect of fatty and lean fish intake during pregnancy.

54
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

VKM also mentioned three meta-analyses of maternal intake and infant intake of
fish and risk of child asthma.
The evidence of an association between fish intake in the lactation period (two
studies) and risk of asthma was also too limited to make any conclusion.
The evidence of an association between fish intake in infancy and risk of asthma
was also limited. Further, the results showed no association or protective association
between fish intake and asthma in the first year of life.

3.2.1.1.3.2 Conclusion, weight of evidence

The evidence indicating effects of maternal fish intake during pregnancy on the risk
of asthma in the offspring was graded “limited, no conclusion”. The conclusion
was based on no statistically significant findings in one pooled analysis, two meta-
analyses, and a summary analysis conducted by VKM, as no clear dose–response
relationship was found.
No conclusion could be drawn for an association between child fish intake and asthma
since studies remain limited with inconsistent results of fish intake and the outcome.

3.2.1.1.4 Fish intake and eczema in children

3.2.1.1.4.1 Description of the studies included
The VKM assessment included two meta-analyses covering maternal intake (high–
low) of fish during pregnancy and infant intake of fish and the risk of child eczema.
The VKM assessment included nine primary studies with eczema in children as the
outcome, one with results on pre-pregnancy intake, seven on pregnancy intake,
one on intake during lactation, and three on intake in children. One study was a
community-based lifestyle intervention with a control cohort, while eight studies
were cohorts.
The primary studies had a skewed geographic distribution, with one study from
Asia (Japan) and nine studies from Europe (France, Germany, the Kingdom of the
Netherlands, Norway, Poland, Spain, Scotland and Sweden).
Two meta-analyses of maternal intake of fish during pregnancy and risk of child
eczema mentioned in the VKM report found borderline statistically significant
associations on the protective side. Analysis of the combined meta-analyses and
primary studies by VKM showed weak association. Furthermore, no conclusions
could be drawn regarding a differential effect of fatty and lean fish.
Prospective studies of fish intake in children and the risk of eczema were reported by
VKM. Associations were protective for intake in the first year of life, but not later.

3.2.1.1.4.2 Conclusion, weight of evidence

The evidence that maternal fish consumption during pregnancy reduces the
risk of eczema was graded “limited, suggestive”. Evidence on fish intake in the
lactation period and risk of eczema (one study) was too limited for a conclusion.

55
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

No conclusions could be drawn for the effects of fatty fish or lean fish due to
limited evidence.
The evidence that fish intake in infants reduces the risk of eczema was graded
“limited, suggestive”. Protective associations were found for intake in the first
year of life, but not later. Associations with eczema at 8 and 12 years of age were
attenuated when restricted to analyses of children without early symptoms of
allergic disease (one study), suggesting an influence of disease-related modification
of exposure.

3.2.1.1.5 Fish intake and multiple sclerosis

3.2.1.1.5.1 Description of the studies included
The VKM assessment included two primary studies (case control) with occurrence
of multiple sclerosis as the outcome. The studies overlapped partially as they used
data from the Swedish Epidemiological Investigation of Multiple Sclerosis (EIMS)
study.
The VKM assessment also included one meta-analysis (based on case-control
studies) on fish intake and the risk of multiple sclerosis.
VKM did not calculate a summary RR based on the two identified studies as they
partially overlapped. The overall results from the primary studies showed higher risk
among those with the lowest intake. This applied to both fatty and lean fish, when
analysed separately. The association with multiple sclerosis was found independent
of vitamin D status (mediation analysis).
In one meta-analysis (Rezaeizadeh et al., 2020), increased intake of fish was
associated with a significantly decreased risk of multiple sclerosis (RR = 0.77, 95%
CI: 0.64, 0.92).

3.2.1.1.5.2 Conclusion, weight of evidence

A protective association between fish intake and the risk of multiple sclerosis was
graded “limited, suggestive” by VKM due to uncertain mechanisms and the evidence
base consisting of case-control studies only.

3.2.1.1.6 Rheumatoid arthritis

3.2.1.1.6.1 Description of the studies included
The VKM assessment included one systematic review and six primary studies (two
case-control and four prospective cohort studies) that included the outcome of
rheumatoid arthritis incidence.
The primary studies were performed in Denmark, France, Greece, Sweden and the
United States of America.
The systematic review included in the VKM report (which included 5 cohort and
5 case-control studies) showed a significant 11 percent lower risk of rheumatoid
arthritis among those with high versus low intake of fish, but not when limited to

56
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

cohort studies. Stronger associations were found in case-control studies.

In three primary cohort studies, the summary RR from the three cohort studies
showed no statistically significant association.

3.2.1.1.6.2 Conclusion, weight of evidence

The evidence was graded “limited, suggestive” for a protective effect of fish
consumption on the risk of rheumatoid arthritis.

3.2.1.2 Summary of the findings on “Allergy and immunology” in the systematic

reviews included from the literature search
Five systematic reviews were included from the literature search on the theme
“Allergy and immunology”. Similar to the outcomes reported in the 2022 VKM
report, the outcomes of the systematic reviews were summarized and analysed in
the following categories:
> allergic rhinitis: two systematic reviews assessing the association with maternal
fish intake or early fish introduction in children;
> allergic sensitization in children: one systematic review describing the association
between early fish introduction and the outcome in children;
> asthma in children: two systematic reviews describing the association between
maternal fish intake and the outcome in children;
> eczema in children: two systematic reviews describing the association between
maternal fish intake and the outcome in children;
> multiple sclerosis: no further systematic reviews identified;
> rheumatoid arthritis: two systematic reviews assessing the association between
fish intake and the risk of rheumatoid arthritis.
The summary of the results from the systematic reviews assessed are summarized
in Table 3.2. The primary studies included in the systematic reviews were mainly
prospective cohort studies or randomized controlled trials. In three systematic
reviews (Pattison et al., 2004; Netting et al., 2014 and Venter et al., 2020), data were
summarized narratively and not pooled for meta-analysis. In two systematic reviews
(Di Giuseppe et al., 2014 and Ierodiakonou et al., 2016), the outcomes were reviewed
through meta-analysis. The results from the systematic reviews are summarized in
the following sections according to the outcomes.

3.2.1.2.1 Allergic rhinitis in children

Two systematic reviews (Ierodiakonou et al., 2016 and Venter et al., 2020) estimating
the association between maternal fish intake or early fish introduction and allergic
rhinitis were included. A description of these systematic reviews, including main
outcome, population, type of studies and overall results, is provided in Table 3.2.
One systematic review (Ierodiakonou et al., 2016) included four prospective cohort
studies for estimating the association between early dietary introduction of fish and
allergic rhinitis. The other study (Venter et al., 2020) included two observational

57
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

studies of maternal fish intake in association with allergic rhinitis. Fish and seafood
intake was assessed based on different quantitative food frequency questionnaires.
In the maternal fish intake analysis of Venter et al., 2020, heterogeneity was found
between studies. Fatty fish were associated with reduced risk of allergic rhinitis in
one study, while, in the other study, all fish were associated with increased risk of
allergic rhinitis.
The meta-analysis described by Ierodiakonou et al., 2016 found a protective
association between early dietary introduction of fish and allergic rhinitis, based
on four prospective cohort studies.

3.2.1.2.2 Allergic sensitization in children

One systematic review (Ierodiakonou et al., 2016) describes the association between
early fish introduction and the risk of allergic sensitization. Table 3.2 summarizes
this systematic review, including the main outcome, population, type of studies
included and overall results. The study included five prospective cohort studies
on children with fish introduced into their diet in the first year of age. There was
low- to very low-certainty evidence that early fish introduction was associated with
reduced allergic sensitization from the five prospective cohort studies combined.

3.2.1.2.3 Asthma in children

Two systematic reviews (Netting et al., 2014 and Venter et al., 2020) describing the
association between maternal fish intake and asthma in children were included.
Netting et al. (2014) combined six studies, including one randomized control study,
one retrospective cohort study and four prospective cohort studies. Venter et al.
(2020) included one randomized control, two observational studies of maternal fish
intake in association with asthma in children less than three years of age and four
observational studies of maternal fish intake in association with asthma in children
at age three years and above. Different quantitative food frequency questionnaires
were commonly used to derive fish intake.
Netting et al. (2014) did not pool data for meta-analysis. Most studies showed no
association between maternal food intake and asthma or wheezing in children.
In one RCT included in Venter et al. (2020), maternal fish was not associated with
asthma or wheezing in children. In most of the observational studies included,
reduced offspring asthma (in children less than three years of age) was associated
with higher intake of total fish and fatty fish. For children three years of age and
above, higher maternal intake of fatty fish in pregnancy was associated with reduced
offspring asthma/wheezing outcomes in the observational studies. In contrast,
higher intake of fish sticks was associated with an increased offspring risk of asthma/
wheezing in the observational studies.

58
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.1.2.4 Eczema in children

Two systematic reviews (Netting et al., 2014 and Venter et al., 2020) were found
to describe an association between maternal fish intake and eczema. Table 3.2
summarizes the systematic reviews, including main outcome, population, type of
studies included and overall results. Venter et al. (2020) included four randomized
controlled trials, consisting of one study with salmon and three studies with fish oil.
The authors also included four observational studies. Netting et al. (2014) included
ten studies: one randomized control, one retrospective cohort and eight prospective
cohort studies. Maternal dietary intake was obtained from different quantitative
food intake frequency questionnaires.
The data in both systematic reviews were summarized narratively and not pooled for
meta-analysis. From the combined studies described by (Venter et al., 2020), effect
of fish consumption during pregnancy on eczema was unclear among the studies.
Similarly, most studies showed no association between maternal food intake and
allergy outcomes as described by (Netting et al., 2014).

3.2.1.2.5 Multiple sclerosis

No systematic reviews were identified for the outcome multiple sclerosis.

3.2.1.2.6 Rheumatoid arthritis

Two systematic reviews (Pattison et al., 2004 and Di Giuseppe et al., 2014) describing the
association between fish intake and rheumatoid arthritis were included (see Table 3.2).
In one systematic review, described by Pattison et al. (2004), eight case-control
and six prospective cohort studies were included, but only two studies reported
an association between rheumatoid arthritis onset and fish consumption. Studies
were hospital-based (inpatients and outpatients) or population-based. Fish intake
was obtained from food frequency questionnaires. The other study described by
Di Giuseppe et al. (2014) showed a dose–response meta-analysis to evaluate fish
consumption and the risk of rheumatoid arthritis, including four case-control and
three prospective cohort studies. Participants were hospital-based or population-
based. Fish consumption was expressed as servings per week.
In one systematic review, reported by Pattison et al. (2004) the results could not be
pooled due to the heterogeneity of the study designs. A protective effect of higher
consumption of fish on the risk of rheumatoid arthritis was reported by Pattison et
al. (2004), however, the review included a small number of studies, with variation
in study design. In the meta-analysis conducted by Di Giuseppe et al. (2014), the
results showed a decreased risk of developing rheumatoid arthritis for each one
serving of fish per week (RR = 0.96, 95% CI: 0.91, 1.01).

59
60
TABLE 3.2 RESULTS FROM THE SYSTEMATIC REVIEWS FROM THE LITERATURE SEARCH ON “ALLERGY AND IMMUNOLOGY”

AUTHOR, YEAR FISH AND RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS OVERALL CONCLUSION
STUDY TITLE SEAFOOD INTAKE (AMSTAR 2)
Lerodiakonou et al., 2016 Early fish introduction Introduction of fish in n = 5, prospective Fish introduction Meta-analysis: Risk of sensitization There was low to very low Moderate
Timing of Allergenic Food in infancy and risk of infancy and followed up cohort studies before age 6 to 9 to any allergen: OR = 0.75, 95% CI: certainty evidence that
Introduction to the Infant allergic sensitization during childhood months. 0.64, 0.88. early fish introduction
Diet and Risk of Allergic n = 14 193 Fish intake measured Meta-analysis: Risk of sensitization could reduce the risk of
or Autoimmune Disease: by questionnaires. to any food: allergic sensitization in
Studies from Finland,
A Systematic Review and children.
Germany and Sweden OR = 0.52, 95% CI: 0.37, 0.73.
Meta-analysis
Early fish introduction Introduction of fish in n = 4, prospective Fish introduction Meta-analysis: Risk of allergic rhinitis There was some evidence
in infancy and risk of infancy and followed-up cohort studies. before age 6 to 12 at age ≤4 y: OR = 0.59, 95% CI: 0.4, that early introduction
allergic rhinitis during childhood months 0.87. of fish could reduce the
n = 12 781 Fish intake measured Meta-analysis: Risk of allergic rhinitis risk of allergic rhinitis in
by questionnaires. at age 5–14 y: OR = 0.68, 95% CI: children.
Studies from Finland,
Norway and Sweden 0.47, 0.98.

Venter et al., 2020 Maternal fish intake Children of age <3 y as n = 2, observational Semi-quantitative Higher intake of fish and fatty The effects of maternal Moderate
Dietary factors during during pregnancy and risk an outcome studies food frequency fish were associated with reduced fish consumption in
pregnancy and atopic of asthma/wheezing in questionnaires offspring asthma/wheezing in the pregnancy and the effects
outcomes in childhood: A children observational studies. on child outcomes of
systematic review from Children of age of 3 y or n = 4, observational Higher intake of fatty fish was asthma, allergy and
the European Academy above as an outcome. studies associated with reduced offspring eczema are unclear and
of Allergy and Clinical asthma/wheezing outcomes in the no firm conclusion could
Immunology included observational studies. be drawn.

Higher intake of fish sticks was

associated with an increased
offspring risk of asthma/wheezing in
the observational studies.
Maternal fish intake Children from 18 n = 2, observational Fatty fish were associated with
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

during pregnancy and months to 7 years studies reduced risk of allergic rhinitis
risk of allergic rhinitis in (study 1), and 3 years to while all fish (unspecified type)
children 8 years (study 2). were associated with increased
risk of allergic rhinitis. However,
heterogeneity was found between
studies.
Maternal fish intake Children from 6 months n = 4, observational In one study, intake of fatty fish and
during pregnancy and risk onwards studies shellfish was positively associated
of eczema in children with offspring risk of eczema; while in
the other four studies, maternal intake
of fish was associated with a reduced
risk of developing atopic dermatitis.
 TABLE 3.2 RESULTS FROM THE SYSTEMATIC REVIEWS FROM THE LITERATURE SEARCH ON “ALLERGY AND IMMUNOLOGY” (cont.)

AUTHOR, YEAR FISH AND SEAFOOD RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS OVERALL CONCLUSION
STUDY TITLE INTAKE (AMSTAR 2)
Netting et al., 2014 Maternal fish intake in >40 000 children n = 10 studies: Maternal fish intake The studies investigated the outcome The findings suggest Moderate
Does maternal diet during pregnancy and risk of n = 1 RCT in pregnancy of eczema for different age groups: that there might
pregnancy and lactation eczema in children 3 months to 1 year: One study found be a connection
n= 1 retrospective
affect outcomes in no clear link between eczema in between maternal fish
cohort study
offspring? A systematic children and how much fish their consumption during
n = 8 prospective cohort pregnancy and a lower
review of food-based mothers ate during pregnancy. Two
studies risk of eczema in children.
approaches studies found that daily consumption
of about 30 g of fish by their mothers However, the results vary
during pregnancy and increased between studies and
consumption (daily versus less age groups, and some
frequently) showed protective studies didn’t account for
effects against eczema. However, other factors that could
no adjustments of the results for influence the outcomes.
potential confounders were applied in
these studies.
2 years to less than 3 years: In one
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

study, when mothers consumed

around 30 grams of fish daily during
pregnancy or ate fish more frequently
(every day compared to less often),
their children had a lower risk of
eczema. However, these studies did
not consider other factors that might
affect the results.
5 years: In a study involving 5-year-
old children, if mothers ate fish at
least once a week during pregnancy,
their children had a reduced risk of
eczema.
Maternal fish intake in n = 5 studies: Maternal fish intake The results from the studies were not Overall, while there
pregnancy and risk of n = 1 RCT in pregnancy consistent across the board. Some were indications of a
asthma/wheezing in studies found a potential protective potential protective
n = 4 prospective cohort
children effect of maternal fish intake on effect of maternal fish
studies
the risk of asthma and wheezing in consumption during
children, suggesting that higher fish pregnancy against the
consumption during pregnancy might risk of asthma and
be associated with a lower risk of wheezing in children,
these respiratory issues. However, the results were not
other studies did not find a significant consistent across all
association. studies.

61
62
TABLE 3.2 RESULTS FROM THE SYSTEMATIC REVIEWS FROM THE LITERATURE SEARCH ON “ALLERGY AND IMMUNOLOGY” (cont.)

AUTHOR, YEAR FISH AND SEAFOOD RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS OVERALL CONCLUSION
STUDY TITLE INTAKE (AMSTAR 2)
Pattison et al., 2004 Investigating a possible One study: Hospital- n = 2 case-control Food frequency Two studies reported association Protective effect of Moderate
The role of diet in effect of individual based inpatients and studies questionnaires between rheumatoid arthritis onset higher consumption
susceptibility to components of diet outpatients (n = 168 and fish consumption. of fish on the risk of
rheumatoid arthritis: a and the development of patients vs n = 137 In one study, the association was rheumatoid arthritis was
systematic review rheumatoid arthritis (RA) controls) men and significant only for high consumption reported. However, it was
women aged 24-89 y. of broiled or baked fish (≥ 2 servings/ inconclusive due to the
One study: Population- week), and a stronger effect was small number of studies
based, female seen in seropositive compared to available and due to
participants aged seronegative cases. variation in study design.
15–64 y (n = 324 cases
vs. 1 243 controls)
Di Giuseppe et al., 2014 Association between fish Studies were from n = 7 studies: FFQs (fish For each one serving per week Results from the Moderate
Fish consumption and consumption and risk of Denmark, Greece, 4 case-control studies consumption was increment in fish consumption, the meta-analysis showed
risk of rheumatoid rheumatoid arthritis (RA) Sweden and the United expressed as servings RR (95% CI) of RA was 0.96 (0.91, an inverse (but not
3 prospective cohort
arthritis: a dose-response States. per week) 1.01). The risk of RA was 20 to 24 significant) association
studies
meta-analysis n = 174 701 included in percent lower for 1 to 3 servings per between fish consumption
analyses with n = 3 346 week of fish and RA.
cases with rheumatoid (RR (95% CI) = 0.76 (0.57, 1.02) as
arthritis compared to no consumption.

Note: n: numbers, OR: odds ratio, CI: confidence interval, FFQ: food frequency questionnaire, RCT: randomized controlled trial, RR: risk ratio, RA: rheumatoid arthritis
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.1.3 Final weight of evidence for “Allergy and immunology”

A final weight of evidence for the outcome “Allergy and immunology” was based on
the 2022 VKM report and the systematic literature search. An overview of the total
amount of literature included in the final weight of evidence is given in Appendix 3,
Table A3.35.

3.2.1.3.1 Allergic rhinitis in children

The evidence for associations of maternal fish intake and early fish introduction
with child rhinitis are based on the 2022 VKM report, two systematic reviews and
three primary studies.
In the maternal fish intake analysis described by the systematic review (Venter
et al., 2020), heterogeneity was found between studies. In addition, associations
between fish intake during pregnancy and allergic rhinitis assessed by VKM were
not statistically significant. The evidence that maternal total, fatty and lean fish
intake during pregnancy affect the risk of rhinitis in the offspring was therefore
graded “limited, no conclusion”.
There was lower-certainty evidence that early fish introduction was associated with
reduced allergic rhinitis from the meta-analysis by (Ierodiakonou et al., 2016). In
the VKM report, very limited data was found to describe protective associations
of infant fish intake and allergic rhinitis. Only intake in the first year of life was
associated with a reduced risk of rhinitis in one of two studies, while intake at later
ages (2 to 8 years) was not included by VKM. The evidence that child fish intake in
the first year protects against rhinitis is therefore graded “limited, no conclusion”.

3.2.1.3.2 Allergic sensitization in children

The evidence that maternal fish intake during pregnancy affects the risk of allergic
sensitization to food or inhalant allergens is graded “limited, no conclusion” based
on two primary studies and two systematic reviews included in the VKM assessment.
The evidence that child fish intake protects against sensitization was also graded
“limited, no conclusion”. This evidence was based on two studies included in the
VKM assessment and one systematic review included from the literature search.

3.2.1.3.3 Asthma in children

The evidence indicating effects of maternal fish intake during pregnancy on the risk
of asthma in the offspring was graded “limited, no conclusion”. This conclusion
was based on no significant findings in the two systematic reviews included from
the literature search and two meta-analyses included in the VKM report.
No conclusion could be drawn for an association between child fish intake and
asthma, and the evidence was graded “limited, no conclusion”. This was based
on limited primary studies with inconsistent results of fish intake and outcomes
reported in the VKM report.

63
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.1.3.4 Eczema in children

In the VKM report, the evidence that maternal fish consumption during pregnancy
reduced the risk of eczema was graded “limited, suggestive”. This was based on two
meta-analyses that found borderline associations on the protective side (Malmir et
al., 2021: RR=0.93, 0.84-1.03 and Zhang et al., 2017: RR=0.84, 0.69-1.01). Maternal
fish consumption was not associated with eczema in the offspring in a recent meta-
analysis by Malmir et al. (2021). However, in a dose–response meta-analysis Malmir
et al. found that maternal fish consumption during pregnancy was weakly associated
with a lower risk of eczema.
Further, according to the two systematic reviews (Netting et al., 2014 and Venter et
al., 2020) from the systematic literature search, the effect of fish consumption during
pregnancy on eczema is unclear. Those reviews showed no association between
maternal food intake and allergy outcomes. However, such studies were not pooled
for meta-analysis. Venter et al. found inconsistency of direction of effect. Thus, the
evidence from the studies was not sufficient to identify any protective effect from
maternal fish consumption during pregnancy on the risk of eczema. As such, the
final weight of evidence that maternal fish consumption during pregnancy reduces
the risk of eczema was graded “limited, no conclusion”.
The evidence that fish intake in infants reduces the risk of eczema was graded
“limited, suggestive” based on the primary studies assessed by VKM. Protective
associations were found for intake in the first year of life, but not later.

3.2.1.3.5 Multiple sclerosis

Based on the assessment by VKM, a protective association of fish intake with the risk
of multiple sclerosis was graded “limited, suggestive” due to uncertain mechanisms
and the evidence base consisting of case-control studies only.

3.2.1.3.6 Rheumatoid arthritis

With regard to an association between fish consumption and rheumatoid arthritis,
VKM identified six studies and one meta-analysis, including a dose-response
analysis. In the meta-analysis identified by VKM, based on case-control and
cohort studies, total fish intake was associated with reduced risk of rheumatoid
arthritis; but the association was not statistically significant when only including
cohort studies. Furthermore, VKM’s summary RR was not statistically significant
for cohort studies. A protective effect of higher consumption of fish on the risk of
rheumatoid arthritis was reported by one systematic review (Pattison et al., 2004);
however, the review included a small number of studies and there were variations
in the study design. The results of the meta-analysis by Di Giuseppe et al. (2014),
based on four case-control and three prospective cohort studies, showed an inverse
association between fish consumption and rheumatoid arthritis.
Evidence was too limited to conclude on the association between the intake of fish
and rheumatoid arthritis. Thus, the evidence was graded “limited, no conclusion”
for a protective effect of fish consumption on the risk of rheumatoid arthritis.

64
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.2 BIRTH AND GROWTH OUTCOMES

The literature included on the theme “Birth and growth outcomes” includes results
from the 2022 VKM report, and two original primary studies from the literature
search. No systematic reviews were included from the literature search.

3.2.2.1 Summary of the findings on “Birth and growth outcomes” in the VKM report,
Benefit and risk assessment of fish in the Norwegian diet
In the 2022 VKM report, Benefit and risk assessment of fish in the Norwegian
diet, the authors summarized the evidence of an association between maternal
fish consumption in pregnancy and the outcomes of “preterm birth”, “small for
gestational age”, “birth weight”, “birth length”, “birth head circumferences” and
“low and high birth weight”.

3.2.2.1.1 Preterm birth

3.2.2.1.1.1 Description of the studies included
The VKM assessment included 1 systematic review and meta-analysis (Zhao et al.,
2021) which included 11 observational studies (prospective cohort and case-control
studies) (n = 36 391 participants) investigating the association between maternal
fish consumption in pregnancy and the risk of preterm birth. In the systematic
review, the summary odds ratio (OR) (95% CI) of highest versus lowest intake of
total fish showed no clear effect of fish consumption: 0.90 (0.72, 1.14). Though in
linear analyses (including 7 cohort studies and 4 675 participants), a dose–response
increment of consumption of 45 g fish/day was associated with a 16 percent lower
risk of preterm birth (summary OR = 0.84, 95% CI: 0.71, 1.01). No clear effects
were found for lean or for fatty fish intake in the meta-analysis by Zhao et al., 2021.
The VKM assessment included 11 original primary studies (10 single studies and 1
pooled analysis) that investigated the association between maternal fish consumption
and the risk of preterm birth (< gestational week 37). All studies were prospective
birth cohorts, except for two – a case-control study and a retrospective cohort study.
VKM calculated a summary RR of the primary studies in relation to the highest
versus lowest intake of total fish, which showed a direction of a protective effect of
fish consumption (RR = 0.89, 95% CI: 0.77, 1.04), but with significant heterogeneity
(Pheterogeneity = 0.001). VKM also calculated a summary RR for the studies, separating
intake into lean (n = 3 cohort studies) and fatty fish (n = 4 cohort studies), though no
clear effects were found (lean fish: RR = 0.83, 95% CI: 0.64, 1.08, Pheterogeneity = 0.44);
fatty fish: RR = 0.89, 95% CI: 0.70, 1.13, Pheterogeneity = 0.29).

3.2.2.1.1.2 Conclusion, weight of evidence

In the final weight of evidence, the VKM gave more weight to the dose–response
results included. Because of the evidence for biological plausibility and a dose–
response association, the evidence of a protective effect of maternal total fish intake
in pregnancy on the risk of preterm birth was graded “probable”. No conclusions
could be drawn for fatty or lean fish intake, and these outcomes were graded
“limited, no conclusion”.

65
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.2.1.2 Small for gestational age

3.2.2.1.2.1 Description of the studies included
The VKM assessment included one systematic review and meta-analysis (Zhao et
al., 2021), which included nine observational studies (prospective cohort studies and
case-control studies) (n = 2 146 participants) investigating the association between
maternal fish consumption in pregnancy and risk of being small for gestational age.
In the systematic review, the summary OR (95% CI) of highest versus lowest intake
of total fish showed no clear effect of fish consumption: 0.79 (0.59, 1.06) (including
11 cohort studies and 2 146 participants). Though in linear analyses (including
7 cohort studies and 1 360 participants) a dose-response increment of 45 g/day of
fish consumption was associated with a 16 percent lower risk of being born small
for gestational age (summary OR = 0.84, 95% CI: 0.71, 0.98 Pheterogeneity = 0.04).
No clear effects were found on lean or fatty fish intake in the meta-analysis.
The VKM assessment included nine original primary studies (eight single studies
and one pooled analysis) that investigated the association between maternal fish
consumption and the risk of being born small for gestational age. Five of the studies
were birth cohorts, while four were case-control studies. VKM calculated a summary
RR of the studies in relation to the highest versus lowest intake of total fish which,
showed no clear effect of fish consumption (RR = 1.02, 95% CI: 0.80, 1.30), with
significant heterogeneity (Pheterogeneity = 0.02). VKM also calculated a summary RR
for the effect of intake prior to pregnancy, which showed a protective effect of total
fish consumption (n = 3 case-control studies, summary RR = 0.73, 95% CI: 0.61,
0.88, Pheterogeneity = 0.38).
No clear effects were found on lean or fatty fish intake in the studies included in
the 2022 VKM assessment.

3.2.2.1.2.2 Conclusion, weight of evidence

Based on the results from the systematic review and the calculated summary RR for
fish consumption prior to pregnancy, there was some evidence that maternal fish
intake could protect against being born small for gestational age. The evidence of
the association between a protective effect of maternal total fish consumption and
the risk of small for gestational age was therefore graded “limited, suggestive”. No
conclusions could be drawn for fatty or lean fish intake, and these outcomes were
graded “limited, no conclusion”.

3.2.2.1.3 Birth weight

3.2.2.1.3.1 Description of the studies included
The VKM assessment included 13 original primary studies (12 single studies and 1
pooled analysis) that investigated the association between maternal fish intake in
pregnancy and birth weight as a continuous variable. All studies were prospective
birth cohorts. VKM did not calculate a summary RR of high versus low intake
because of the large heterogeneity between the studies. Of the single studies, the

66
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

comparison of the association between the highest fish consumption group and birth
weight versus that of the lowest fish consumption group was rather unclear. In the
pooled analysis study (including 13 European cohort studies), a 15.2 g increase in
birth weight in the highest versus the lowest intake of fish categories was reported.

3.2.2.1.3.2 Conclusion, weight of evidence

The evidence that maternal total, fatty and lean fish intake in pregnancy increases
birth weight was graded “limited, suggestive (positive association)” by VKM.

3.2.2.1.4 Low and high birth weight

3.2.2.1.4.1 Description of the studies included
The VKM assessment included one systematic review and meta-analysis (Zhao
et al., 2021), which included 11 observational studies (birth cohorts) (n = 26 823
participants) investigating the association between maternal fish consumption in
pregnancy and risk of low birth weight. In the systematic review, the summary OR
(95% CI) of highest versus lowest intake of total fish showed a protective effect
of fish consumption: 0.78 (0.61, 1.00). In addition, VKM included a linear analysis
(including 7 cohort studies and 869 participants) where a dose–response increment
of 45 g/day fish consumption was associated with a 35 percent lower risk of low
birth weight (summary OR = 0.65, 95% CI: 0.47, 0.90, Pheterogeneity = 0.04). The meta-
analysis found no clear effects on lean and fatty fish intake.
The VKM assessment included ten original primary studies (nine single studies and one
pooled analysis) that investigated the association between maternal fish consumption
and risk of low and high birth weight. All studies were prospective birth cohorts. VKM
calculated a summary RR of the studies in relation to the highest versus lowest intake of
total fish, which showed a direction of a protective effect of fish consumption, although
this was not statistically significant (RR = 0.86, 95% CI: 0.66, 1.13, Pheterogeneity = 0.15).
No conclusions could be drawn from the studies regarding fatty or lean fish intake.
The outcome “high birth weight” was only assessed in the pooled analyses by
Leventakou et al., 2014. They found a small increase in the risk of high birth weight
with the highest versus lowest total intake of fish. The effects of lean and fatty fish
were similar (measured on a continuous scale).

3.2.2.1.4.2 Conclusion, weight of evidence

In the final weight of evidence, the VKM assessment gave more weight to the dose–
response results included. Because of the evidence for biological plausibility and a
dose–response association, the evidence of a protective effect of maternal total fish
intake in pregnancy on the risk of low birth weight was graded “probable”. No
conclusions could be drawn for fatty or lean fish intake, and these outcomes were
graded “limited, no conclusion”.
In the final judgement by the VKM assessment, the evidence that maternal total,
fatty and lean fish consumption in pregnancy increased the risk of high birth weight
was graded “limited, suggestive”.

67
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.2.1.5 Birth length and head circumference

3.2.2.1.5.1 Description of the studies included
The VKM assessment included seven primary studies that investigated the
association between maternal total fish intake in pregnancy and birth length and
head circumference (one study only included birth length). Five of the studies
included were prospective birth cohorts, one was a cohort-based community trial,
and one was a retrospective cohort study. Six studies were conducted in Europe,
while one was conducted in the United States. All studies presented estimates of
total fish consumption, except for one study that presented intake of canned tuna.
Five studies also presented intake of subcategories of lean and fatty fish.
VKM did not calculate a summary RR of high versus low intake because of the
large heterogeneity between the studies. For intake of total fish in pregnancy and
the relation to birth length, one study reported increased birth length in the highest
intake category compared to the lowest, while the rest of the studies reported no
association. For total fish intake in pregnancy and the relation to head circumference,
two studies reported larger, and one study reported smaller head circumference in
the highest compared to the lowest intake category, while the rest of the studies
reported no association. For lean and fatty fish intake, the studies reported no clear
association with either birth length or head circumference.

3.2.2.1.5.2 Conclusion, weight of evidence

The current evidence of an association between maternal total fish intake in
pregnancy and birth length did not support an association, and the evidence was
concluded to be “limited, suggestive (no association)”. For head circumference, the
evidence was inconsistent and was graded “limited, no conclusion”. No conclusions
could be drawn for fatty or lean fish intake, and these outcomes were graded
“limited, no conclusion” for both birth length and head circumference.

3.2.2.2 Summary of the findings on “Birth and growth outcomes” in primary studies
included in the literature search
3.2.2.2.1 Description of the primary studies
Two primary studies were included from the literature search in the “Birth and
growth outcomes” category. Table 3.3 gives an overview of the results from the
two studies, including author, title, study type, study population, measurement of
fish and seafood consumption, measurement of outcome, overall results and overall
conclusion. Both studies are prospective cohort studies. In the study by Oken et al.,
2004, participants were enrolled from 1999 to 2002 in Project Viva in Massachusetts,
in the United States, to collect data on gestational diet, pregnancy outcomes and
offspring health. The study by Zhao et al., 2022 was a prospective analysis of data
from the Tongji Birth cohort in Wuhan, China from 2018 to 2021.

68
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.2.2.2 Description of study population

The study by Oken et al. included 2 109 women. The women’s ethnicity reflected the
diversity of the source population, including 16 percent Black, 7 percent Hispanic,
and 6 percent Asian American subjects, ranging from 14 to 44 years of age.
The study by Zhao et al. included 2 149 pregnant women aged 18 to 45 years at
baseline, but the analysis in the present study includes data from 1 701 mother–infant
pairs.

3.2.2.2.3 Description of fish consumption

In Oken et al., a semiquantitative FFQ, including intake of “canned tuna fish”,
“shrimp, lobster, scallops, clams”, “dark meat fish, e.g., mackerel, salmon, sardines,
bluefish, swordfish”; and “other fish, e.g., cod, haddock, halibut” was used to
evaluate the maternal diet in relation to the outcomes.
Dietary intake was assessed through face-to-face interviews using a modified version
of a semiquantitative FFQ in the study by Zhao et al. The modified FFQ consisted
of 74 food items, including 4 fish items (saltwater fish, freshwater fish, prawn and
crabs, and molluscs).

3.2.2.2.4 Results from the primary studies included

In the study by Oken et al., the frequency of fish consumption during pregnancy
showed a trend towards an inverse association with birth weight and foetal growth
and was not associated with length of gestation. One unadjusted analysis, an increase
in first-trimester fish consumption from less than one serving per month to more
than two servings per week, was associated with a decrease in birth weight from 3
487 g to 3 452 g. After multivariable adjustment, there was still a suggestion of an
inverse association between frequency of fish intake and foetal growth, although
this relation was statistically significant only during the first trimester.
In the study by Zhao et al., higher intake of freshwater fish during pregnancy
showed a reduced risk of newborns born small for gestational age. The association
also remained significant after adjusting for relevant confounders. No significant
associations were observed between total fish, saltwater fish and shellfish intake and
risk of newborns born small for gestational age.

69
70
TABLE 3.3 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “BIRTH AND GROWTH OUTCOMES”

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND SEAFOOD
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Zhao et al., 2022 2018 to 2021 n = 2 149 pregnant Dietary intake including Neonatal characteristics From the lowest to highest B The study found
Tongji Birth cohort Cohort study women fish consumption was were obtained from quintiles of freshwater supportive evidence
n = 1 701 mother–infant assessed through face- hospital obstetric records, fish intake, the crude ORs that moderate intake
Wuhan, China More than 13th or less
pairs were included in to-face interviews using including gestational age, for SGA were: of freshwater fish in
than 28th week of
this analysis a modified version of a sex of the infant, and Q1: reference mid-pregnancy is related
gestation until birth
semiquantitative FFQ. birth weight. Newborns to a lower risk of SGA in
à 29 to 14 weeks Cases (n = 1 701) Q2: 0.63 (95 % CI: 0.35,
The four fish items were were classified as SGA Chinese pregnant women.
Maternal age, years, 1.11)
saltwater fish, freshwater if their birth weight was
median (IQR) 28.7 Q3: 0.75 (95 % CI: 0.43,
fish, prawns and crabs, below the 10th percentile
(26.8–30.8) 1.31)
and molluscs. Prawns of the gestational age
Sex of infant, male, n and crabs and molluscs distribution of the Q4: 0.50 (95 % CI: 0.26,
(percent) = 895 (52.6) were combined into one Chinese population. 0.93)
category, collectively Q5: 0.64 (95 % CI 0.35,
called shellfish. 1.13)
Total fish, g/day, median P for trend = 0.206).
(IQR) 23.9 (10.9-43.6)
No significant
Freshwater fish, g/ associations were
day, median (IQR) 12.1 observed between total
(4.3-26.4) fish, saltwater fish, and
Saltwater fish, g/day, shellfish intake and risk
median (IQR) 0 (0-3.6) of SGA.
Shellfish, g/day, median
(IQR) 5.0 (1.3-11.3)
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
 TABLE 3.3 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “BIRTH AND GROWTH OUTCOMES” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND SEAFOOD
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Oken et al., 2004 1999 to 2002 n = 2 128 pregnant Semiquantitative food Birth weight in grams On unadjusted B Frequency of fish
Project Viva Birth cohort women who delivered a frequency questionnaires was obtained from the analysis, an increase consumption during
live infant including intake of hospital medical record in first-trimester fish pregnancy showed a
From mean gestational
n = 2 109 (99 percent) “canned tuna fish Length of gestation in consumption from less trend towards an inverse
age 10.6 weeks until birth
completed at least one (3–4 oz.)” (1 oz. = 28.3g); days was calculated by than one serving per association with birth
dietary questionnaire, “shrimp, lobster, scallops, subtracting the date of month to more than weight and foetal growth
clams (1 serving)”; “dark the last menstrual period two servings per week but was not associated
n = 1 797 1st trimester
meat fish, e.g., mackerel, from the date of delivery. was associated with with length of gestation.
n = 1 663 2nd trimester salmon, sardines, a decrease in z value
Birth weight for
n = 2 070 3rd trimester bluefish, swordfish from 0.22 to 0.16 and a
gestational age was
Maternal age ranged from (3–5 oz.)”; and “other decrease in birth weight
determined by using as
14 years to 44 years. fish, e.g., cod, haddock, from 3 487 g to 3 452 g.
a reference a combined
halibut (3–5 oz.).” After multivariable
1999–2000 US natality
The lowest group reported adjustment, there was
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

data set.
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

consuming none/<1 still a suggestion of

This method adjusts
serving of seafood per an inverse association
for gestational age and
month; the remaining between frequency of fish
provides a normal z value
subjects were divided into intake and foetal growth,
measuring distance of
tertiles, with the highest although this relation was
the birth weight from
intake group, more than statistically significant at
the median for a given
two servings per week, the conventional standard
gestational age.
used as the referent. of P < 0.05 for only the
first trimester.
No indication of a relation
with length of gestation.
No association of
seafood intake with the
dichotomous outcomes
low birth weight, small
for gestational age, and
preterm delivery.

Notes: All studies are prospective cohort studies. Q: quartile, IQR: interquartile range, SGA: small for gestational age, OR: odds ratio, FFQ: food frequency questionnaire, CI: confidence interval.

71
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.2.3 Final weight of evidence for “Birth and growth outcomes”

A final weight of evidence for the theme “Birth and growth outcomes” was based on
the 2022 VKM report and the systematic literature search. An overview of the literature
included in the final weight of evidence is provided in Appendix 3, Table A3.35.
The summary of the final weight of evidence for “Birth and growth outcomes” was
based on the judgement from the 2022 VKM report and the two original primary
studies included. The inclusion of the primary studies did not change the conclusion
of the 2022 VKM report.
Thus, the final weight of evidence for the association between maternal total fish
consumption in pregnancy is graded:
> preterm birth: “Probable (protective)”;
> small for gestational age: “Limited, suggestive (protective)”;
> birth weight (continuous scale): “Limited, suggestive (positive association)”;
> low birth weight: “Probable (protective)”;
> high birth weight: “Limited, suggestive (positive association)”;
> birth length: “Limited, no conclusion”;
> head circumference: “Limited, no conclusion”.
For maternal lean and fatty fish intake in pregnancy, all outcomes were graded
“limited, no conclusion”, except for birth weight (continuous scale), which was
graded “limited, suggestive (positive association)”, and risk of high birth weight,
which was graded “limited, suggestive (positive association).

3.2.3 BONE HEALTH

The literature included in the theme “Bone health” includes results from the 2022
VKM report and four original primary studies originating from the literature search.
No systematic reviews were included.

3.2.3.1 Summary of the findings on “Bone health” in the VKM report, Benefit and risk
assessment of fish in the Norwegian diet
The 2022 VKM assessment included eight primary studies (one case-control study
and seven prospective cohort studies), which included the outcomes of hip fracture,
bone mineral density in the femoral neck, or total hip. The study populations were
from Asia, Europe and the United States. VKM calculated a summary RR from four
of the prospective cohort studies for incident hip fracture in relation to the highest
versus lowest intake of total fish. This result suggested that a high intake of fish,
compared to a lower intake of fish, may lower the RR of hip fracture (RR = 0.70,
95 % CI: 0.55, 0.88).
VKM also included one systematic review (Sadeghi et al., 2019) that investigated the
effects of total fish intake on the risk of hip fractures. The meta-analysis in Sadeghi
et al. suggested a borderline significant protective effect of total fish intake.

72
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

In conclusion, from the VKM report, the evidence that high total fish consumption
may lower the risk of hip fractures was graded “limited, suggestive”. No conclusion
could be drawn for fatty fish or lean fish as only one study was included.

3.2.3.2 Summary of the findings on “Bone health” in primary studies included in the
literature search
3.2.3.2.1 Description of the primary studies included
Four primary studies were included under the category “bone health”. A description
of the studies, including study name, design, time period, study population, intake
of fish consumption and overall results, can be found in Table 3.4 and Table 3.5.
Two of the primary studies were RCTs (Lucey et al., 2008 and Tong et al., 2020) and
two were prospective cohort studies (Hirota et al., 2005 and Thacher et al., 2015).

3.2.3.2.2 Description of study population

One RCT included 276 adult men and women from 20 to 40 years of age (Lucey et
al., 2008), while the other included 96 children under the age of 5 years with calcium
deficiency rickets (Thacher et al., 2015). One of the prospective cohort studies was
conducted on children (n = 548), with a follow-up time of 5 years (Hiorta et al.,
2005), while the other was conducted among adults (n = 54 898) and lasted from
1993 until 2016 (Tong et al., 2020).

3.2.3.2.3 Description of fish consumption

In the RCTs, one intervened with ground fish, using limestone in the second study
arm (Thacher et al., 2015), while the other study had salmon, fish oil and cod as
interventions (Lucey et al., 2008). In Hirota et al., questionnaires and interviews
regarding several food groups, including fish and small fish, were applied and carried
out among the participants; while in Tong et al., questionnaires were used to classify
participants into fish eaters, meat eaters, vegetarians and vegans.

3.2.3.2.4 Results from the primary studies

3.2.3.2.4.1 Calcium deficiency rickets
In the Thacher et al. study conducted with children with calcium-deficiency rickets,
treatment with calcium as either ground fish or limestone for 6 months healed rickets
in the majority of the children. There was no control group in this study.

3.2.3.2.4.2 Bone turnover and bone loss

In the RCT by Lucey et al., different interventions with fish and fish oil were
examined for attenuation of bone turnover and bone loss in overweight adults on
a weight-loss diet. The inclusion of fish (salmon or cod) or fish oil in the diet
was unable to attenuate the effect of weight loss on bone turnover. No significant
differences were seen in the different bone turnover biomarkers between the
different intervention groups.

73
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.3.2.4.3 Bone status

The study by Hirota et al. on Japanese adolescent girls and boys found that annual
increase in bone status in girls aged 10 or 11 years was associated positively with
increased intake of fish, fruit, vegetables, soybeans and milk products and associated
negatively with preference for meat. They conclude that increased intake of fish
could improve bone status (Hirota et al., 2005). Japanese traditionally eat the
whole body of small fish, which contains more calcium, vitamin D, magnesium
and proteins, compared to fish filet, which may be the reason for the beneficial
effect on bone status.

3.2.3.2.4.4 Hip fracture

In the European Prospective Investigation into Cancer and Nutrition (EPIC)
Oxford study conducted by Tong et al., fish eaters, vegans and vegetarians were
compared to meat eaters. The researchers found that fish eaters had a higher risk of
hip fractures compared with meat eaters. These risk differences were likely partially
due to their lower body mass index (BMI), and possibly to lower intake of calcium
and protein.

74
 TABLE 3.4 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (INTERVENTION STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “BONE HEALTH”
FOR “EVIDENCE OF HEALTH BENEFITS OF FISH CONSUMPTION”

NUMBER OF INTERVENTION AND

PARTICIPANTS IN THE MEASUREMENT
AUTHOR, YEAR CONTROL GROUP
STUDY TYPE STUDY (N) AND INTAKE OF
STUDY TITLE INFORMATION MEASUREMENT OF
STUDY DURATION AGE (YEARS) FISH AND SEAFOOD OVERALL RESULTS RISK OF BIAS
REGION, COUNTRY REGARDING OUTCOME
AND FOLLOW-UP TIME AT EXPOSURE CONSUMPTION AT
YEAR OF SAMPLING INTERVENTION,
ASSESSMENT BASELINE DURATION, DOSE
SEX (PERCENT, MEN)
Lucey et al., 2008 Randomized controlled n = 276 men and women At baseline, seafood Intervention: Participants Serum osteocalcin and No significant differences B
SEAFOODplus YOUNG trial 20–40 years intake was assessed were randomly assigned bone-specific alkaline were seen in the different
study 8 weeks by a validated food to 1 of 4 groups: phosphatase were bone turnover biomarkers
43 percent men
frequency questionnaire Group 1: control measured. between the different
Iceland, Spain and Ireland
(FFQ). Dietary intake (sunflower oil), Serum C-terminal intervention groups.
2004-2005 was assessed by 2-day
Group 2: 150 g cod 3 telopeptide of type I
weighed food records collagen was measured.
times/week for 8 weeks,
before baseline (habitual Urinary N-telopeptides
diet). Group 3: 150 g salmon 3
of type I collagen was
times/week for 8 weeks,
measured.
Group 4: 3 g fish oil /day
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

for 8 weeks.
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Control: Sunflower oil, 6

capsules per day.
Thacher et al., 2015 Randomized trial n = 96 children Breastfeeding history, Ground fish: 10 g twice/ Radiographic score, Of the 88 children who B
Nigeria 14 or 24 weeks Age: Group 1 limestone: usual dairy product day for 14 weeks. assessed by radiographic completed the study,
median (p25–p75) 28 intake, were assessed. Limestone: 1.75 g twice healing was defined as 29 (66 percent) in the
1998–2000
(19–42) months. daily for 24 weeks. achieving a score of 1.5 ground fish group and
or less on a 10-point 24 (55 percent) in the
Group 2, ground fish: No control group.
scale. limestone group achieved
median (p25–p75) 42
the primary outcome
(27–59) months.
of a radiographic score
Sex: male/female ratio, of 1.5 or less within 6
Limestone group: 16/29, months (P = 0.39). The
Ground fish group: 23/28 mean radiographic score
improved from 6.2 ± 2.4
to 1.8 ± 2.2 in the ground
fish group and from 6.3
± 2.2 to 2.1 ± 2.4 in the
limestone group (P = 0.68
for group comparison).

75
76
TABLE 3.5 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “BONE HEALTH”

NUMBER OF
STUDY TYPE MEASUREMENT
AUTHOR, YEAR PARTICIPANTS IN THE
YEAR OF SAMPLING, AND INTAKE OF MEASUREMENT
STUDY TITLE STUDY (N) OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
STUDY DURATION AND FISH AND SEAFOOD OF OUTCOME
REGION, COUNTRY AGE (YEARS)
FOLLOW-UP TIME CONSUMPTION
SEX (PERCENT, MEN)
Hirota et al., 2005 Prospective cohort study n = 548 Questionnaires and Bone measurement Associated factors with B The study in Japanese
Japan 1995–1999 262 girls + 286 boys interviews regarding dairy Bone status of the os the initial bone status of adolescent girls and boys
consumption determined calcis was measured with 10- and 11-year-olds: found that increased
Follow-up time: 5 years Age: 10–15 years
how often subjects quantitative ultrasound Girls (n = 114): Intake intake of fish could
Girls: mean (SD) 12.4±1.6 consumed any type of improve bone status.
(QUS) using the Achilles of small fish: Pearson’s
Boys: 12.3±1.5 milk, cheese or yogurt. A1000 ultrasonometer. correlation coefficient: r =
Sex: 52.2 percent men Inquiries were also made Achilles measures SOS 0.22; P = 0.024
about other traditional (speed of sound in meters Boys (n = 112): no
sources of calcium, such per second), and BUA association same test
as soybeans, curdled soy (broad-band ultrasound (Table 4).
protein (tofu), fermented attenuation in decibels
soy (Natto), green Associated factors with
per megahertz), a
leafy vegetables, other yearly change of bone
measure of frequency-
vegetables, seaweeds, status from age from
dependent attenuation
fish, and small fish. 10–11 years to 11–12
of the ultrasound wave
years:
Intake (frequency, times passing through the heel.
or dishes/week) mean Stiffness index (SI), a Girls (n = 114); Increased
(SD): variable derived from a fish intake was positively
combination of SOS and associated with bone
Girls - of fish: 3.4 (1.7)
BUA, was calculated by status. Pearson’s
Girls - of small fish: 2.0 correlation coefficient =
the analysis software
(1.5) 0.35, P< 0.001)
according to the equation:
Boys - of fish: 3.4 (1.7) 0.67 BUA + 0.28 SOS) Boys: Increased intake of
Boys - of small fish: 2.4 420. small fish was associated
(2.0) (p<0.05 more than with bone status.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

girls) Pearson’s correlations

coefficient = 0.10, P =
0.045.
 TABLE 3.5 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “BONE HEALTH” (cont.)

NUMBER OF
STUDY TYPE MEASUREMENT
AUTHOR, YEAR PARTICIPANTS IN THE
YEAR OF SAMPLING, AND INTAKE OF MEASUREMENT
STUDY TITLE STUDY (N) OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
STUDY DURATION AND FISH AND SEAFOOD OF OUTCOME
REGION, COUNTRY AGE (YEARS)
FOLLOW-UP TIME CONSUMPTION
SEX (PERCENT, MEN)
Tong et al., 2020 Prospective cohort study n = study population Questionnaires asking if Outcomes were identified Fish eaters had B Fish eaters had higher
EPIC-Oxford study Recruitment 1993–2001 included a minimum of participants included fish through linkage to increased risk of hip risk of hip fractures
54 898 in the diet. The dietary hospital records or fracture compared compared with meat
the United Kingdom Follow-up in 2010
participants (in analyses questions were used to death certificates with meat eaters, even eaters. These risk
of Great Britain and Follow-up with record
for total fractures), classify participants until mid-2016. The after adjusting year of differences were likely
Northern Ireland linkage in 2016
of whom 30 391 had into four groups at outcomes were total recruitment, ethnicity, partly due to their lower
Average of 17.6 years of repeated measures of diet baseline and follow-up: fracture and site-specific Townsend deprivation BMI, and possibly to lower
follow-up 14 years later meat eaters, fish eaters, fractures identified by index, physical activity, intakes of calcium and
vegetarians and vegans. the relevant 9th or 10th smoking, alcohol protein.
revisions of the World consumption, dietary
Health Organization’s supplement use, in
International women menopausal
Classification of Diseases status, hormone
(ICD-9/ICD-10) codes. replacement therapy,
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

BMI, dietary calcium

intake and protein intake,
showing an adjusted
hazard ratio (HR) (95 %
CI) of 1.25 (1.01, 1.55).

Note: SD: standard deviation, EPIC: European Prospective Investigation into Cancer and Nutrition, BMI: body mass index.

77
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.3.3 Final weight of evidence for “Bone health”

A final weight of evidence for the theme “Bone health” was based on the 2022 VKM
report and the systematic literature search. An overview of the literature included
in the final weight of evidence is given in Appendix 3, Table A3.35.
The weight of evidence is only considered for the outcome of hip fracture, since
that is the only outcome evaluated by VKM, and the other outcomes included
only involved one study, which was considered too little to grade the weight of
evidence. Using estimates from four studies, VKM graded the evidence that high
fish consumption may lower the risk of hip fractures “limited, suggestive”. No
significant heterogeneity was found between the studies, where the estimates were
on the protective side. In the one primary study on hip fracture in the current review
by Tong et al. (2020), a higher risk of hip fracture was seen in fish eaters compared
with meat eaters; however, these risk differences may be due to lower BMI in the
fish-eater group. Therefore, the evidence of a “limited, suggestive” protective effect
of fish intake on hip fracture remains.

3.2.4 CANCER
The literature included in the theme “Cancer” includes results from the report of
World Cancer Research Fund on Diet, Nutrition, Physical Activity and Cancer,
published in 2018 (WCRF, 2018b), and three systematic reviews and ten original
primary studies originating from the literature search.

3.2.4.1 Summary of the findings on “Cancer” in the World Cancer Research Fund
report
The information referenced is derived from the third expert report from the World
Cancer Research Fund and the American Institute for Cancer Research, Diet,
Nutrition, Physical Activity and Cancer: a Global Perspective.
The 2018 WCRF report defines fish as any of various cold-blooded, aquatic
vertebrates, having gills, commonly fins, and typically an elongated body covered
with scales, as well as shellfish. Cantonese-style salted fish is part of the traditional
diet consumed by people living in the Pearl River Delta region in Southern China.
It is prepared with less salt than is used in Northern China, allowed to ferment, and
eaten in a decomposed state.
The 2018 WCRF report concludes that there was “strong evidence” for a probable
increased risk of nasopharyngeal cancer from increased intake of Cantonese-style
salted fish, and “limited, suggestive” evidence for a decreased risk of liver and
colorectal cancer from increased total fish intake. For other cancer outcomes, no
conclusion could be made.

78
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.4.2 Summary on the findings on “Cancer” in the systematic reviews included from
the literature search
A comprehensive literature search was conducted to identify systematic
reviews examining the relationship between fish intake and the risk of cancer.
Three systematic reviews that met our inclusion criteria and that were published from
2018, after the WCRF report was published, were identified in the search: Jayedi et
al., 2020; Kazemi et al., 2021 and Gao et al., 2022. All three studies included a meta-
analysis. Table 3.6 provides a comprehensive overview of these studies, presenting
details including outcomes investigated, population and participant information,
study design and included studies, time period, study population characteristics,
fish consumption intake, as well as the overall results and conclusions derived from
these systematic reviews.
One of the systematic reviews identified covered site-specific cancer risk (Jayedi
et al., 2020), while the other two covered the risk of pancreatic cancer (Gao et
al., 2022) and breast cancer (Kazemi et al., 2021). All systematic reviews included
participants from the general population, encompassing both males and females,
except for Kazemi et al., which only included women. According to Gao et al., no
substantial relationship was found between fish intake and the increase of pancreatic
cancer risk. Their analysis encompassed 22 studies (11 case-control studies and 11
cohort studies), which investigated the association between fish intake and pancreatic
cancer risk. The findings from these studies collectively suggested no significant
association between fish consumption and the risk of developing pancreatic cancer.
When comparing the highest and lowest levels of fish intake, the pooled RR was
1.00 (95% CI: 0.93, 1.07), indicating no substantial difference in risk. Kazemi et al.
investigated the association between fish intake and breast cancer. Their analysis
incorporated 17 studies, including case-cohort studies, nested case-control studies,
cohort studies and randomized control trials. The main findings from these studies
revealed no significant association between each additional 100 g/day increase in
fish intake and breast cancer risk (RR = 1.0, 95 % CI: 0.93, 1.08). Furthermore, there
was no evidence of a nonlinear dose–response relationship (P-nonlinearity, 0.39).
The study findings suggest that fish intake does not have a substantial impact on
the risk of developing breast cancer.
Jayedi et al. analysed 120 prospective cohort studies and found moderate-quality
evidence suggesting an inverse association between fish consumption and the risk of
liver cancer (summary RR = 0.65, 95% CI: 0.48, 0.87). They also found low-quality
evidence for a positive association between fish consumption and the risk of myeloid
leukaemia and gastric cancer. However, no significant associations were observed
for cancers at other sites.

79
80
TABLE 3.6 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “CANCER”

OUTCOME (MAIN POPULATION NUMBER OF INCLUDED OVERALL RESULTS

INFORMATION
AUTHOR, YEAR OUTCOME AND AND NUMBER OF STUDIES AND TYPE (INCLUDE RISK OF BIAS
REGARDING FISH OVERALL CONCLUSION
TITLE SPECIFIC PARTICIPANTS OF STUDY DESIGNS META-ANALYSIS (AMSTAR 2)
AND SEAFOOD INTAKE
OUTCOME GROUPS) INCLUDED INCLUDED IF APPLICABLE)
Gao et al., 2022 Investigated associations General population. n = 25 for all FFQ Meta-analysis: When the Overall, the results Moderate
Poultry and Fish Intake between poultry and Two studies only with n = 22 including fish highest and lowest fish achieved in this study
and Pancreatic Cancer fish consumption and females, one only with intake were compared, further clarify that fish
11 case-control studies
Risk: A Systematic pancreatic cancer (PC) males, and the rest of the pooled RR was 1.00 intake has no substantial
and 11 cohort studies
risk the studies included both (95% CI: 0.93–1.07), with relationship with the
Review and Meta-Analysis
males and females no apparent heterogeneity incidence of pancreatic
(P = 0.27, I2 = 14 cancer.
n = 1 367 330
percent).
Subanalysis: In studies
without adjusted energy
intake, fish intake
negatively correlated
with PC risk (RR = 0.84,
95% CI: 0.72, 0.99),
while studies adjusted for
energy intake showed no
correlation.
Jayedi et al., 2020 Meta-analyses of General population aged n = 120 primary studies FFQs, diet history, 24-hour Moderate quality of For site-specific cancers, High
Fish Consumption and the observational studies 18 years or older Prospective cohort studies dietary recalls, and evidence for the relation moderate-quality
Risk of Chronic Disease: evaluating the n = 135 971 dietary records. between fish consumption evidence was found that
Meta-analyses of
association of fish and the risk of liver higher fish consumption
An Umbrella Review of observational studies that
consumption with the risk cancer (summary RR was associated with a
Meta-Analyses of combined prospective,
of chronic disease. for each 100-g/day lower risk of liver cancer.
Prospective Cohort retrospective, and
Fish consumption and increment: 0.65; 95% CI: There was also an inverse
Studies cross-sectional studies in
site-specific cancer risk 0.48, 0.87). There was association for prostate
their analyses were also
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

also low-quality evidence cancer mortality, but the

eligible.
for the inverse association quality of the evidence
of fish consumption and was rated low. For
the risk of prostate cancer cancers at other sites,
mortality, as well as for no significant inverse
the positive association of associations were found
fish consumption and the and the quality of the
risk of myeloid leukaemia evidence was rated low or
and gastric cancer. Fish very low.
consumption was not
associated with the risk
of cancers at other sites.
 TABLE 3.6 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “CANCER” (cont.)

OUTCOME (MAIN POPULATION NUMBER OF INCLUDED OVERALL RESULTS

INFORMATION
AUTHOR, YEAR OUTCOME AND AND NUMBER OF STUDIES AND TYPE (INCLUDE RISK OF BIAS
REGARDING FISH OVERALL CONCLUSION
TITLE SPECIFIC PARTICIPANTS OF STUDY DESIGNS META-ANALYSIS (AMSTAR 2)
AND SEAFOOD INTAKE
OUTCOME GROUPS) INCLUDED INCLUDED IF APPLICABLE)
Kazemi et al., 2021 Summarize the Females over 18 years
Intake of Various Food associations between of age.
Groups and Risk of Breast food groups and risks of n = N/A
breast cancer
Cancer: A Systematic
Review and Dose- Investigated the
Response association of fish with
breast cancer
Meta-Analysis of
Prospective Studies
n = 17 studies Validated FFQ No association was To ameliorate the High
Includes cohort, observed for each cancer risk, fish and
case-cohort, nested additional 100-g/day poultry represent good
case-control studies, increase of fish substitutes for red
and follow-up studies of (RR = 1.0, 95% CI: 0.93, meat in the dietary
randomized controlled 1.08; P-heterogeneity composition.
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

trials = 0.19. As in the present meta-

There was no evidence of analysis, fish had no
a nonlinear significant association
dose-response with the risk of breast
association cancer.
(P-nonlinearity = 0.39; n
= 11 studies). The risk of
breast cancer increased
by approximately 10
percent with increasing
intake of fish, up to 110
g/day.

Note: N/A: not applicable, FFQ: food frequency questionnaire, RR: risk ratio, CI: confidence interval

81
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.4.3 Summary of the findings on “Cancer” in primary studies included from the
literature search
3.2.4.3.1 Description of the primary studies
Ten primary studies were included with cancer as an outcome and fish intake as the
exposure. Table 3.7 describes the studies, including study name, study type, number
of participants, measurements of seafood consumption and outcome, overall results,
risk of bias and overall conclusion.
All ten studies were prospective cohort studies with geographic distribution,
including the EPIC cohort with ten countries around Europe, in addition to
other cohorts from Denmark, Japan, the Kingdom of the Netherlands, the United
Kingdom of Great Britain and Northern Ireland, and the United States.

3.2.4.3.2 Description of study population

The number of study participants varied between the included studies, ranging
from 26 749 to 521 324 participants. Median follow-up time, the mean age of the
participants, and the number of events varied between the studies.

3.2.4.3.3 Description of fish consumption

Fish consumption was assessed by questionnaires or 24-hour dietary assessment.
Fish intake was reported either as frequency or amount.

3.2.4.3.4 Results from the primary studies

3.2.4.3.4.1 Colorectal cancer

Three of the primary studies examined the association between fish intake and
colorectal cancer. Aglago et al. (2020) used data from the EPIC cohort (521 324
participants, follow-up time – 14.9 years) and found that regular consumption
of fish at the recommended levels was associated with a lower risk of colorectal
cancer. Etemadi et al. (2018) also found an association between fish intake and a
decreased risk of colorectal cancer. This study consisted of three US-based cohorts
(407 270 participants in all) and had a follow-up time of 13.8 years. The UK study by
Bradbury et al. (2020) found no association between colorectal cancer and total fish
intake. This study had less than half the follow-up time (5.7 years) in comparison
to the two other studies, and included 475 581 participants.

3.2.4.3.4.2 Other types of cancer

Three of the studies included, covering different types of cancer, reported an
association between fish intake and a lower risk for cancer, including hepatocellular
carcinoma, upper gastrointestinal cancer and bladder cancer. One study from two
US prospective cohorts (Ma et al., 2019) on hepatocellular carcinoma in 142 857
participants with 32 years of follow-up, suggested an inverse association of fish
intake with hepatocellular carcinoma. One study on upper gastrointestinal cancer

82
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

risk (Zamani et al., 2020), involving a US-based cohort from six states, with 468
952 participants and with 15.5 years of follow-up, reported an association between
non-fried fish intake and a lower risk for head and neck cancer and oesophageal
adenocarcinoma. One study (Dianatinasab et al., 2021), including 11 cohorts from
different European countries with 518 545 participants and 11.3 years of follow-up,
found an inverse association between total fish intake and bladder cancer risk in
men, but not in women.
Four studies were included on different types of cancer. All reported no association
between fish intake and site-specific cancer, including lung cancer, cancer of unknown
primary, biliary tract cancer and prostate cancer. One study from Japan (Cai et al.,
2020) investigated the association between fish intake and lung cancer risk in 73 187
participants with 16 years of follow-up time and reported no association between
fish intake and lung cancer. Another study, from the Kingdom of the Netherlands,
(Hermans et al., 2021) with 120 852 participants and 20.3 years of follow-up, found
no association between fish intake and the risk for cancer of unknown primary.
One study from a Japan-based cohort (Makiuchi et al., 2020) on biliary tract cancer,
including 98 663 participants, reported no association between fish intake and biliary
tract cancer risk. One study in a Danish cohort (Outzen et al., 2018) involving 26
749 men with prostate cancer, with 19 years of follow-up, reported no association
between any type of fish intake and the risk of prostate cancer.

83
84
TABLE 3.7 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “CANCER”

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Aglago et al., 2020 1992-2014 n = 521 324 Fish consumption Cases of incident cancer Risk of CRC: B Regular consumption of
European Prospective Cohort study Cases (n = 6 291: mean assessed by a validated of unknown primary Total intake of fish: fish (total fish, fatty fish,
Investigation into Cancer age 57 years and 43 centre-specific were identified through Quintile 5 vs. 1: lean fish and shellfish), at
Median follow-up time:
and Nutrition (EPIC) percent men questionnaire. regional cancer registries HR = 0.88, 95% CI: 0.80, recommended levels, was
14.9 years.
cohort Total fish and shellfish or via a combination 0.96; Ptrend = .005 associated with a lower
Colorectal cancer (CRC) Non-cases (n = 469 869):
intake at baseline, mean of methods, including Fatty fish: risk of cancer of unknown
10 European countries Mean age 51 years and
(SD): health insurance records, primary.
(Denmark, France, 30% men Quintile 5 vs. 1:
pathology registries Overall, weekly intake
Germany, Greece, Italy, Cases: 39 (35) g/day HR = 0.90, 95% CI: 0.82,
and active follow-up of 100–200 g of fatty or
Netherlands [Kingdom Non-cases: 37 (36) g/day 0.98; Ptrend = .009
of participants and lean fish was associated
of the], Norway, Spain, (P< 0.001). relatives. Cases of cancer Lean fish: with a 7 percent lower
Sweden, the United
of unknown primary Quintile 5 vs. 1: risk of cancer of unknown
Kingdom of Great Britain
were defined according primary.
and Northern Ireland) HR = 0.91, 95% CI: 0.83,
to the International
1.00; Ptrend = .016
Classification of Diseases
for Oncology (ICD-O).
Bradbury et al., 2020 2006–2010 n = 475 581 Touchscreen Prevalent and incident Total fish intake and risk B No association was found
Diet and colorectal Cohort study (219 329 men and 256 questionnaire cancer cases were of cancer (HR, 95% CI) between colorectal cancer
cancer in UK Biobank: a 252 women) 40–69 years Subsample of identified through linkage Reference: < once/week and total fish intake.
Average of 5.7 years
prospective study at recruitment ercent men to cancer and death (n = 165)
follow-up participants (n = 175
= 46.12 percent registries.
the United Kingdom Colorectal cancer 402) completed at least vs.
of Great Britain and one online 1.0–1.9 times/week (n =
Northern Ireland 24-hour dietary 1 007): 0.98 (0.83,1.16)
assessment 2.0–2.9 times/week (n =
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Total fish intake: 643): 0.98 (0.82,1.16)

< once/week; ≥3 times/week (n = 761
1.0–1.9 times/week; cases): 0.95 (0.80,1.13)
2.0–2.9 times/week; The HR (95% CI) for each
≥3 times/week 25-g/day increment in
fish was 0.96 (0.86, 1.07).
Ptrend = 0.470
 TABLE 3.7 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “CANCER” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Cai et al., 2019 1995–2013 Cohort I, 73 187 participants Dietary assessment was Cases were identified Fish intake and risk of B Fish intake was not
Association between meat 1998–2013 Cohort (32 934 men and 40 253 performed using the Food from major local hospitals lung cancer: associated with lung
and saturated fatty acid II, median 16.0-year women) 45 percent men Frequency Questionnaire in the study area and Men: Q1 (reference) vs. cancer risk in either
intake and lung cancer follow-up (FFQ) by data linkage with Q4 intake: HR = 1.09, women or men.
45–74 years
risk: The Japan Public Lung cancer population-based 95% CI: 0.90, 1.33.
1 315 (901 men and 414 cancer registries.
Health Center-based Women: Q1 (reference)
women) cases of lung Death certificate
prospective study vs. Q4 intake: HR = 1.01,
cancer information was used as
Japan 95% CI: 0.76, 1.34.
a supplementary source
of data. Women, Q4, model 3, HR:
1.01 (0.76–1.34).
Dianatinasab et al., 2021 Cohort studies 518 545 participants Dietary data were Each study ascertained Marginally non-significant B Inverse association
The association Median follow-up: 11.3 167 095 (32 percent) men obtained using a self- incident bladder association between total between total fish
between meat and fish years and 351 444 (68 percent) administered or trained- cancer cases, defined fish and fish products and fish products
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

consumption and bladder women interviewer-administered to include all subjects with risk of bladder consumption and bladder
Bladder cancer
cancer risk: a pooled food frequency with urinary bladder cancer comparing highest cancer risk in men, but no
2 848 bladder cancer
analysis of 11 cohort questionnaire (FFQ) that neoplasms according with lowest tertile (HR = association was found in
cases and 515 697
studies was validated on either to the International 0.89, 95% CI 0.63, 1.25, women.
non-cases
food groups and/or energy Classification of P-trend = 0.369).
Bladder cancer Epidemiology Mean age: 60.6 (± 7.3) intake. Diseases for Oncology Inverse association
and Nutritional for cases and 52.5 (ICD-O-3 code C67) between total fish
Determinants consortium (± 10.1) for non-cases using population-based and fish products
(BLEND) cancer registries, health consumption and
These studies originated insurance records, or bladder cancer risk in
from 11 countries, Europe: medical records men comparing highest
European Prospective with lowest tertile (HR
Investigation into Cancer = 0.79, 95% CI: 0.65,
and Nutrition cohort 0.97, P-trend = 0.04)
studies (EPIC), Denmark, was observed, but no
France, Germany, Italy, association was found
Netherlands (Kingdom in women comparing
of the), Norway, Spain, highest with lowest tertile
Sweden, the United (HR = 1.07, 95% CI 0.76,
Kingdom of Great Britain 1.51, P-trend = 0.658).
and Northern Ireland;
Netherlands cohort study
(NLCS); North America:
VITamins and Lifestyle
cohort study (VITAL)

85
86
TABLE 3.7 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “CANCER” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Etemadi et al., 2018 Three cohorts 407 270 participants NCI-Diet History State cancer registries, Fish intake was B Fish intake was
Anatomical subsite can AARP 1995–2011 Age (years) Questionnaire (DHQ) medical record associated with associated with a
modify the association PLCO 1993–2001–2009 AARP 50–71 years abstraction, Linkage to decreased risk of total decreased risk of total
between meat and AHS 1993–1997–2013 PLCO 55–74 years the cancer registries. colorectal cancer (HR = colorectal cancer
meat compounds Overall median follow-up AHS mean age from The cancer endpoints 0.79, 95% CI 0.68, 0.89)
and risk of colorectal of 13.8 years baseline in relation to were defined, based on (P< 0.001).
adenocarcinoma: Colorectal cancer subsites quintiles of red meat first primary diagnosis, HR (95% CI) given for
Findings from three large intake was 56.9 (Q1), by anatomic site each 50 g/1 000 kcal per
US cohorts 53.5 (Q2), 52.2 (Q3), 51.6 and histologic codes day increased intake in
(Q4), 50.7 (Q5) of the International adjusted models
Three US-based studies;
Sex (percent men) Classification of Diseases
NIH-AARP Diet and Health AARP 327 183 for Oncology, third edition
Study (AARP), Prostate, participants
Lung, Colorectal and (191 925 men and 135
Ovarian Cancer Screening 258 women)
Trial (PLCO), Agricultural PLCO 49 850 participants
Health Study (AHS) (23 761 men and 26 089
women)
AHS - Licensed pesticide
applicators (farmer and
commercial applicators)
and spouses of farmer
and commercial
applicators – 30 237
individuals (16 295 men
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

and 13 942 women)

 TABLE 3.7 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “CANCER” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Hermans et al., 2021 Case–cohort design 120 852 participants Self-administered Record linkage to the No associations were B No associations were
Meat consumption and Cases were derived from 899 CUP cases questionnaire on diet Netherlands Cancer observed between fish observed between fish
cancer of unknown the full cohort, while the Based on the distribution Registry and the Dutch consumption and CUP risk consumption and CUP
55–69 years
primary (CUP) risk: results number of person years of the sub-cohort, Pathology Registry Increased CUP risk, but risk.
Total of 92 389 women
from The Netherlands at risk for the full cohort participants were it was not statistically
and 50 468 men (55
cohort study on diet and was estimated from compared using quartiles significant (Q4 vs. Q1: HR
percent)
cancer a subcohort of 5 000 (Q), increments of 25 g/ = 1.25, 95% CI 0.99,1.57,
The Netherlands cohort participants who were day for fish consumption Ptrend = 0.29)
study on diet and cancer randomly sampled from Fish intake: CUP cases vs
(NLCS) the full cohort at baseline subcohort members; 14.1
in 1986 g/day vs. 12.9 g/day
1986–2006
20.3 years of follow-up
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Cancer of unknown
primary (CUP) is a
metastasised cancer for
which no primary lesion
could be identified during
life.
Ma et al., 2019 Two prospective cohorts n = 142 857 Validated Medical records and Fish intake was inversely B Suggestive inverse
Meat intake and risk 1980–2012 semiquantitative food- pathological reports, the associated with HCC risk association of fish with
of hepatocellular frequency questionnaire National Death Index for (HR = 0.70, 95% CI 0.47, HCC risk
Up to 32 years of Total of 92 389 women
carcinoma in two large (FFQ) 1980, 1984, all deaths attributable to 1.05, Ptrend = 0.10)
follow-up and 50 468 men (55
US prospective cohorts of 1986 and every 4 years liver cancer. Suggestive inverse
Hepatocellular carcinoma percent)
women and men thereafter in the NHS association of fish.
(HCC) NHS: 1976, women,
The Nurses” Health Study In the HPFS, dietary The substitution of
30–55 years
(NHS) and the of women information collected in poultry or fish for 1 SD
HPFS: 1986, men, 40–75 1986 and every 4 years
and men of processed red meat
years thereafter using similar
The Nurses” Health Study intake was associated
163 incident HCC cases FFQs. with a decrease in risk of
(NHS) and the Health
(87 women and 76 men) Nine possible HCC (HR = 0.79, 95% CI
Professionals Follow-up
Study (HPS) intake-frequency 0.61, 1.02).

US cohorts responses, ranging from

“never” to “more
than 6 times a day”.

87
88
TABLE 3.7 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “CANCER” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Makiuchi et al., 2019 Cohort study 98 663 participants: Food frequency Active patient notification Fish consumption was not B Fish was not associated
Relationship between 1995 and 1999, until 43177 men (43.8 percent) questionnaire (FFQ) from local major significantly associated with biliary tract cancer
Meat/Fish Consumption 2012 and 49 323 women, 45 to The validity of the FFQ hospitals in the study with biliary tract cancer risk in men and women
and Biliary Tract Cancer: 74 years for the assessment of area and data linkage risk in either men or
followed-up for 607 757.0
The Japan Public Health meat consumption was with population-based women.
person-years in men and 217 male and 162 female
Center–Based Prospective evaluated using 14- or cancer registries. Death Q1 vs. Q4 fish intake:
728 820.3 person-years BTC cases
Study 28-day dietary records certificate information.
in women Men: HR = 1.39, 95% CI
The Japan Public Health as the gold standard. 0.92, 2.08.
Biliary tract cancer (BTC)
Center-based Prospective Reproducibility of the
Women: HR = 1.04, 95%
Study (JPHC Study) FFQ was evaluated
CI 0.68, 1.60.
by administering two
Japan
questionnaires, 1 year
apart.
Baseline fish intake: Men
Q1; 35.7 g, Q2: 63.9 g,
Q3: 92.6g, Q4: 143.8 g.
Women Q1: 35.1 g, Q2:
62.7 g, Q3: 89.5 g, Q4:
135.1 g.
Outzen et al., 2018 Prospective cohort study 26 749 men, 1 690 Food-frequency Record linkage to the Overall, no association B No association between
Fish consumption and 1993–1997 until prostate cancer cases questionnaire (FFQ) and a Danish Cancer Registry, was found between any any type of fish intake
prostate cancer risk and 2012–2013 50–64 years lifestyle questionnaire Danish Pathology type of fish intake and the and risk of total prostate
mortality in a Danish Both the total fish Register, Danish Causes risk of total or high-grade cancer
19 year-follow-up period
cohort study intake and the intake of of Death Registry prostate
(1993–2012)
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Danish “Diet, Cancer and subtypes of fish (lean and cancer.

Prostate cancer
Health” cohort fatty fish) were analysed Association between total
as either continuous fish intake and prostate
Denmark
(increment: 25 g/day) or cancer
categorical (quartiles)
Q4 vs. Q1: adjusted RR =
variables.
1.12, 95% CI 0.97,1.29,
Total fish intake baseline: P-trend 0.06
Cohort: 41.9 g Per 25 g/day, adjusted:
(12.7–99.0) adjusted RR = 1.04, 95%
CI 1.00,1.09.
 TABLE 3.7 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “CANCER” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Zamani et al., 2020 Prospective cohort study 468 952 participants Dietary data was Death records, linkage Compared with the B Non-fried fish intake is
Dietary Polyunsaturated conducted in 6 states 275 948 male (58.8 collected at baseline with state cancer lowest quintile of total associated with lower
Fat Intake in Relation 1995–2011, study percent) using a 124-item, registries fish/shellfish intake, HNC and EA risk.
to Head and Neck, duration and 15.5 years self-administered food the highest quintile Fish/shellfish intake
193 004 females
Esophageal, and Gastric median follow-up frequency questionnaire was associated with a was associated with a
Total HNC; n = 2 453 (FFQ) developed and 20 percent lower risk of
Cancer Incidence in the Upper gastrointestinal 20 percent to 27 percent
National Institutes of EA; n = 855 validated by the National HNC (HR = 0.80, 95% lower risk of HNC and EA.
cancer risk
Health–AARP Diet and oesophageal squamous Cancer Institute to assess CI 0.69, 0.91; Ptrend
incidence of head the frequency and portion = 0.0002, adjusted
Health Study cell carcinoma (n = 267)
and neck cancer sizes of foods P-trend = 0.001), which
National Institutes of (HNC), oesophageal gastric cancer (cardia:
The FFQ has been appeared to be primarily
Health–AARP Diet and adenocarcinoma (EA), n = 603; non-cardia: n
validated using two 24- due to consumption of
Health Study oesophageal squamous = 631)
hour recalls as criterion fish high in n-3 PUFAs
the United States cell carcinoma, and (HR = 0.76, 95% CI
instruments.
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

gastric cancer
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

0.66, 0.87; P-trend <

Total fish intake (g/day):
0.0001, adjusted P-trend
Q1 ≤5.70, Q2 5.71–10.04,
= 0.0006) and nonfried
Q3 10.05–16.91, Q4
fish (HR = 0.83, 95%
16.92–27.74, Q5
CI 0.71, 0.96; P-trend <
27.75–821.41
0.0001, adjusted Ptrend
= 0.0006).
Note: SD: standard deviation; FFQ: food frequency questionnaire; y: years; HR: hazard ratio; CI: confidence interval; Q: quartile; PLCO: Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial; AHS:
Agricultural health study; NIH: National Institutes of Health; CUP: cancer of unknown primary risk; HNC: head and neck cancer; EA: oesophageal adenocarcinoma; PUFA: polyunsaturated fatty acid

89
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.4.4 Final weight of evidence for “Cancer”

A final weight of evidence for the theme “Cancer” was based on the 2022 VKM
report, the 2018 WCRF report and the systematic literature search. An overview
of all the literature included in the final weight of evidence is given in Appendix 3,
Table A3.35.
The weight of evidence for the association between dietary fish intake and cancer was
based on the 2018 WCRF report, primary studies and systematic reviews included
in the evaluation. The overall weight of evidence of total fish intake associated
with pancreatic cancer and breast cancer was graded “limited, no conclusion.”
Furthermore, the weight of evidence from total fish intake associated with liver
cancer and colorectal cancer was graded “limited, suggestive” for a protective effect.
Lastly, the association of Cantonese-style salted fish intake with nasopharyngeal
cancer was graded “strong evidence” for an increased risk, based solely on the
WCRF report.

3.2.5 CARDIOVASCULAR DISEASES AND OUTCOMES

The literature included in the theme “CVD outcomes” includes results from the 2022
VKM report as well as two systematic reviews and ten primary studies originating
from the literature search.

3.2.5.1 Summary of the findings on “Cardiovascular diseases and outcomes” in the

VKM report, Benefit and risk assessment of fish in the Norwegian diet
VKM summarized the evidence of an association between fish consumption and
the outcomes of total CVD incidence, coronary heart disease (CHD) incidence,
secondary prevention, myocardial infarction (MI) incidence, stroke incidence,
ischemic stroke, haemorrhagic stroke, heart failure, atrial fibrillation and venous
thromboembolism.

3.2.5.1.1 Total cardiovascular disease incidence

3.2.5.1.1.1 Description of the studies included
The VKM assessment included eight primary studies for the outcome of total
CVD incidence. All the primary studies were prospective cohort studies. The diet
exposures included total, fatty and lean fish intakes. VKM calculated a summary
RR from the primary studies.
The summary RR found a non-significant protective association between total
fish intake and CVD incidence (RR = 0.94, 95% CI 0.86, 1.02) with significant
heterogeneity (p = 0.01). The heterogeneity was not further explained. No clear
effects were found for fatty fish intake (RR = 0.93, 95% CI 0.73, 1.19) nor for lean
fish intake (RR = 1.01, 95% CI 0.90, 1.14).

90
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.5.1.1.2 Conclusion, weight of evidence

The evidence was graded “limited, suggestive” for a protective effect of total fish
intake on CVD incidence, and “limited, no conclusion” for an effect of fatty or lean
fish on CVD incidence.

3.2.5.1.2 Coronary heart disease

3.2.5.1.2.1 Description of the studies included
The VKM assessment included three systematic reviews, two umbrella reviews and
ten primary studies for the outcome CHD. All the primary studies were prospective
cohort studies. The diet exposures included total, fatty and lean fish, and shellfish
intakes.
One of the systematic reviews, Zhang et al. (2020), also found a significant protective
association between total fish intake and CHD incidence (RR = 0.91, 95% CI 0.84,
0.97), with moderate heterogeneity (I2 = 47.4 percent). The other systematic review,
Bechthold et al. (2019), found a non-significant protective association (RR = 0.94,
95% CI 0.88, 1.02), with moderate heterogeneity (I2 = 52.0 percent). Both systematic
reviews performed meta dose-response analyses. Zhang et al. found that a 20 g/
day increment in fish intake was associated with a 4 percent reduction in CHD
incidence and mortality. Bechthold et al. showed that an increment in fish intake
of approximately 250 g/day was associated with a reduction of approximately 15
percent in the risk of CHD.
The summary RR of the primary studies found a significant protective association
between total fish intake and CHD incidence (RR = 0.89, 95% CI 0.81, 0.98). There
was significant heterogeneity among the primary studies (p = 0.002), but no report
of a statistically significant adverse effect.
VKM’s summary RR for fatty fish indicated a protective association between fatty
fish intake and CHD incidence (RR = 0.93, 95% CI 0.83, 1.04), while there was no
association for lean fish (RR = 0.99, 95% CI 0.93,1.05).

3.2.5.1.2.2 Conclusion, weight of evidence

The VKM assessment graded the evidence “probable” for a protective effect of total
fish intake on CHD incidence, “limited, suggestive” for a protective effect of fatty
fish, and “limited, suggestive” for no effect of lean fish on CDC incidence.

3.2.5.1.3 Myocardial infarction

3.2.5.1.3.1 Description of the studies included
The VKM assessment included one systematic review and eight primary studies for
the outcome of myocardial infarction. The diet exposures included total, fatty and
lean fish intakes.
The systematic review, Jayedi et al. (2019), showed a protective association
(RR = 0.73, 95% CI 0.59, 0.87) of total fish consumption; however,

91
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

high heterogeneity was reported (I2 = 72 percent). The systematic review also
performed a meta linear dose–response analysis and found that a 15 g/day increment
in fish consumption reduced by 4 percent the risk of myocardial infarction
(RR = 0.96, 95% CI 0.94, 0.99).
From the primary studies, VKM calculated a summary RR. All the studies were
prospective cohort studies, except for one, which was a nested case-control study.
The summary RR reported by VKM suggested no association between total fish
intake and incidence of myocardial infarction (RR = 0.96, 95% CI 0.82, 1.12) with
borderline statistic significant heterogeneity (p = 0.051). The summary RR reported
by VKM suggested a slightly non-significant protective effect of fatty fish (RR =
0.93, 95% CI 0.82, 1.05), Pheterogeneity = 0.37), but no effect of lean fish (RR = 1.04,
95% CI 0.94, 1.14), Pheterogeneity = 0.54).

3.2.5.1.3.2 Conclusion, weight of evidence

The evidence was graded “limited, suggestive” for a protective effect of total fish
intake on incidence of myocardial infarction, “limited, suggestive” for a protective
effect of fatty fish, and “limited, suggestive” for no effect of lean fish.

3.2.5.1.4 Total stroke

3.2.5.1.4.1 Description of the studies included
The VKM assessment included four systematic reviews and 19 primary studies for
the outcome total stroke. The diet exposures included total, fatty and lean fish, and
shellfish intakes.
Two of the systematic reviews found moderate to significant protective associations
with low to moderate heterogeneity. Both Zhao et al. (2019) and Bechthold et al.
(2019) found no departure from linearity and the risk of total stroke incidence
decreased by 12 to 14 percent, with an increment of 100 g/day of total fish intake.
VKM calculated a summary RR from the primary studies. The summary RR
suggested a significant protective association of total fish intake on total stroke
incidence (RR = 0.92, 95% CI 0.89, 0.95), without significant heterogeneity
(p = 0.67).
VKM also found a non-significant protective association of fatty fish intake on total
stroke incidence (RR = 0.92, 95% CI 0.80, 1.05) without significant heterogeneity
(p = 0.21). The same was found in one of the systematic reviews. In addition,
VKM reported a non-significant protective association of lean fish intake on total
stroke incidence (RR = 0.95, 95% CI 0.89, 1.01), without significant heterogeneity
(p = 0.51).
In the systematic review, by Zhao et al. (2019), intake of shellfish was also
investigated. The systematic review found no association of shellfish intake on total
stroke incidence (hazard ratio (HR) = 0.96, 95% CI 0.83, 1.11).

92
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.5.1.4.2 Conclusion, weight of evidence

The evidence was graded “probable” for a protective effect of total fish intake on
total stroke incidence and “limited, suggestive” for a protective effect of fatty and
lean fish on total stroke incidence.

3.2.5.1.5 Ischemic stroke

3.2.5.1.5.1 Description of the studies included

The VKM assessment included three systematic reviews and eight primary studies
for the outcome of ischemic stroke. The diet exposures included total fish intake.
VKM calculated a summary RR from the primary studies. The summary RR
indicated a protective effect (RR = 0.93, 95% CI 0.86, 1.02), without significant
heterogeneity (p = 0.27). This result was consistent with the meta-analysis by Zhao
et al. (2019), which found: (RR = 0.96, 95% CI 0.89, 1.03).

3.2.5.1.5.2 Conclusion, weight of evidence

The evidence was graded “limited, suggestive” for a protective effect of total fish
intake on ischemic stroke incidence. This was based on the calculated summary RR
from the primary study and the similar results from the meta-analysis.

3.2.5.1.6 Haemorrhagic stroke

3.2.5.1.6.1 Description of the studies included
The VKM assessment included three systematic reviews and eight primary studies
for the outcome haemorrhagic stroke. The diet exposures included total fish intake.
VKM calculated a summary RR from the primary studies. The summary RR
indicated a non-significant protective effect (RR = 0.88, 95% CI 0.71, 1.09), without
significant heterogeneity (p = 0.29). The systematic review by Zhao et al. (2019)
found a significant protective association of total fish intake on haemorrhagic stroke
(RR = 0.88, 95% CI 0.80, 0.96), with low heterogeneity (I2 = 0 percent). In addition,
Zhao et al. identified a linear dose–response relation between total fish intake and
haemorrhagic stroke incidence.

3.2.5.1.6.2 Conclusion, weight of evidence

The evidence was graded “limited, suggestive” for a protective effect of total fish
intake on haemorrhagic stroke incidence.

3.2.5.1.7 Atrial fibrillation

3.2.5.1.7.1 Description of the studies included
For the outcome of atrial fibrillation, the VKM assessment included one systematic
review and five primary studies on total fish intake, four primary studies on fatty
fish intake, and three primary studies on lean fish intake.

93
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

For total fish consumption, VKM calculated a summary RR from the five studies
for the highest versus lowest intake of total fish in relation to atrial fibrillation. The
summary RR for total fish suggested an adverse association (RR = 1.06, 95% CI 1.00,
1.13), without significant heterogeneity (Pheterogeneity = 0.66). The systematic review,
Li et al. (2017), based on six prospective cohort studies, indicated no association
for the outcome atrial fibrillation, high versus low total fish consumption; (RR =
1.01, 95% CI 0.94, 1.09).
For fatty fish consumption, VKM calculated a summary RR from the four studies
for the highest versus lowest intake of fatty fish in relation to atrial fibrillation. The
summary RR for fatty fish was not statistically significant (RR = 1.26, 95% CI 0.80,
1.97), with significant heterogeneity (Pheterogeneity < 0.001).
For lean fish consumption, VKM calculated a summary RR from the three studies
for the highest versus lowest intake of total fish in relation to atrial fibrillation. The
summary RR for lean fish suggested a protective association (RR = 0.85, 95% CI
0.73, 0.99), without significant heterogeneity (Pheterogeneity = 0.39).

3.2.5.1.7.2 Conclusion, weight of evidence

In conclusion, the evidence by VKM was graded “limited, suggestive” for a potential
adverse effect of total fish intake on the risk of atrial fibrillation. For fatty fish
consumption, the evidence was graded “limited, no conclusion”, while for lean fish
consumption, the evidence was graded “limited, suggestive” for a protective effect.

3.2.5.1.8 Heart failure

3.2.5.1.8.1 Description of the studies included
The VKM assessment included one systematic review and eight primary studies
for the outcome heart failure. The diet exposures included total, fatty and fried
fish intake. Overall, fish intake was associated with a significantly lower risk of
heart failure in one of the four studies. In the two studies on fried and non-fried
fish, findings were consistent: intake of non-fried (including baked/boiled) fish was
associated with lower risk and fried fish with a higher risk of heart failure. In the two
studies from Sweden, fatty fish was associated with a statistically significant lower
risk (except for the highest intake category) in women, but not in men.

3.2.5.1.8.2 Conclusion, weight of evidence

In conclusion, the evidence by VKM that consumption of fish reduces the risk of
heart failure was graded “limited, suggestive”.

3.2.5.1.9 Venous thromboembolism

3.2.5.1.9.1 Description of the studies included
The VKM assessment included three primary studies for the outcome of venous
thromboembolism. The diet exposures included total fish intake. The study from
Norway, Hansen-Krone et al. (2014), found no significant association between fish

94
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

intake for dinner (fatty or lean) and risk of venous thromboembolism in participants
who did not take fish oil supplements (stratified analysis). The US study, Lutsey et
al. (2009), reported a statistically significant adverse association for the highest versus
lowest intake level, and the UK study, Zhang et al. (2021b), reported a statistically
significant protective association.

3.2.5.1.9.2 Conclusion, weight of evidence

In conclusion, the evidence by VKM that consumption of fish reduces the risk of
venous thromboembolism was graded “limited, no conclusion”.

3.2.5.2 Summary of findings on “Cardiovascular diseases and outcomes” from the

systematic reviews included from the literature search
Two systematic reviews were included from the literature search for “CVD
outcomes”. The two systematic reviews, Mente et al. (2009) and Chowdhury et
al. (2012), are summarized in further detail in Table 3.8. These systematic reviews
included the outcomes of total CHD, and total stroke.

3.2.5.2.1 Coronary heart disease

The systematic review, Mente et al., included 29 prospective cohort studies and three
RCTs. Because the information on the studies was not available, it was assumed that
their exposure was on total fish intake. Mente et al. found a protective association
of total fish intake on CHD incidence, based on the prospective cohort studies
(RR = 0.81, 95% CI 0.70, 0.92), but found no effect, based on the RCTs. They also
performed a dose–response analysis using the p values for trend of their studies,
but found no linear relation.

3.2.5.2.2 Total stroke

In the systematic review, Chowdhury et al.(2012), the authors included
21 prospective cohort studies with total fish intake as the exposure. They
found moderate to significant protective association with total stroke incidence
(2–4 servings/week versus ≤ 1 serving/week: RR = 0.94, 95% CI 0.90, 0.98);
≥ 5 versus ≤ 1 serving/week: RR = 0.88, 95% CI 0.81, 0.96), with no evidence of
heterogeneity (I2 = 20–22 percent, p > 0.05). They also conducted a meta linear
dose–response analysis and found that an increment of two servings/week of total
fish (serving size not specified) could reduce the risk of total stroke by 4 percent
(95% CI 1–7).

95
96
TABLE 3.8 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “CVD OUTCOMES”

AUTHOR, YEAR FISH AND RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS OVERALL CONCLUSION
STUDY TITLE SEAFOOD INTAKE (AMSTAR 2)
Chowdhury et al., 2012 Total and cause-specific General populations, n = n = 29 prospective cohort Standardized categories RR (95% CI): 2–4 Fish consumption Moderate
Association between fish cerebrovascular disease 67 5048 studies of fish consumption (2–4 servings/week vs ≤1: 0.94 is moderately but
consumption, long chain times/week and ≥5 times/ (0.90–0.98) (18 studies); significantly associated
omega 3 fatty acid, and week), compared with a RR (95% CI): ≥5 with a reduced risk of
risk of cerebrovascular reference category (≤1/ servings/week vs ≤1 0.88 incident cerebrovascular
disease: systematic week). [0.81–0.96] (8 studies) disease.
review and meta-analysis Dose–response meta-
analysis (18 studies)
shows an increment of 2
servings/week of any fish
was associated with a 4
percent reduced risk of
cerebrovascular disease
(95% CI 1–7%). All 21
studies showed the RR
(95% CI) when comparing
participants in the
highest with the lowest
category of fish intake
was 0.88 (0.84–0.93). In
a subset of studies, the
RR for white fish types
was 1.03 (0.90–1.19) and
for fatty fish types was
0.84 (0.72–0.98).

Notes: CHD: congenital heart disease, RR: risk ratio, RCT: randomized controlled trial, FFQ: food frequency questionnaire.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
 TABLE 3.8 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “CVD OUTCOMES” (cont.)

AUTHOR, YEAR FISH AND RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS OVERALL CONCLUSION
STUDY TITLE SEAFOOD INTAKE (AMSTAR 2)
Mente et al., 2009 All CHD All populations, n = 363 Total fish Intake of fish was Higher intake of fish Moderate
A systematic review of 228 n = 29 prospective cohort significantly associated was associated lower
the evidence supporting studies with lower risk of CHD incidence of total CHD.
a causal link between RR (95% CI) = (0.81
dietary factors and (0.7–0.92)). Stratification:
coronary heart disease FFQ (0.78 [0.66–0.90)
vs Food Record (1.21
[0.18–2.24)], Men (0.85
[0.70–1.01]) vs Women
(0.77 [0.51–1.02]) vs
Both (0.78 [0.63–0.94]);
the United States (0.80
[0.70–0.95]) vs Europe
(0.87 [0.66–1.07]) vs
Asia (0.74 [0.47–1.01]);
Primary prevention (0.83
[0.73–0.93]) vs secondary
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

prevention (0.45
[0.12–0.79]).
n = 3 RCTs Total fish Association not No evidence of an effect
significant: RR (95% CI): in RCTsxxyyzz
1.12 (0.66–1.59)

Notes: CHD: congenital heart disease, RR: risk ratio, RCT: randomized controlled trial, FFQ: food frequency questionnaire.

97
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.5.3 Summary of findings on “Cardiovascular diseases and outcomes”

from the primary studies included from the literature search
Ten primary studies were included from the literature search for “CVD outcomes”.
Detailed information regarding the studies, including study design, participants, fish
intake, outcome, and overall results are given in Table 3.9. Further, each of specific
outcomes according to the outcomes in the 2022 VKM assessment are summarized
individually in the following sections.

3.2.5.3.1 Total cardiovascular disease incidence

3.2.5.3.1.1 Description of the primary studies included
Two primary studies were included from the literature search for the outcome “total
CVD incidence”. Petermann-Rocha et al. (2021) conducted a prospective cohort study
on 422 791 participants from the United Kingdom of Great Britain and Northern
Ireland, including males and females. The participants were categorized as meat-eaters,
fish-eaters, fish- and poultry-eaters and vegetarians. “Fish” referred to total fish, and
the meat-eaters had fish in the diet. The outcome, total CVD incidence, was defined
to include CHD, myocardial infarction, total stroke and heart failure.
Zhong et al. (2021) conducted a substitution analysis on six prospective cohort
datasets. The cohorts had 29 682 participants from the United States, including males
and females. Five dietary exposures (whole eggs, processed meat, unprocessed red
meat, poultry and fish) were categorized. The analysis substituted one serving per
week (1 serving = 85 g) of one dietary exposure for another. The outcome, CVD
incidence, was defined to include CHD, total stroke, heart failure and CVD death.

3.2.5.3.1.2 Results from the primary studies included

Petermann-Rocha et al. (2021) found a significant protective association between
total fish intake and CVD incidence (HR = 0.93, 95% CI 0.88, 0.97).
Zhong et al. (2021) found that substituting one serving per week of fish for eggs
and processed meat significantly reduced risks of CVD incidence (HR (95% CI)
= 0.98 (0.96, 0.99) and 0.96 (0.94, 0.99), respectively), and the reduced risks further
increased if the substitution was one serving per day.

3.2.5.3.2 Coronary heart disease

3.2.5.3.2.1 Description of primary studies included
Five primary studies were included from the literature search for the outcome
“CHD” (Table 3.9): Matheson et al., 2009; Lajous et al., 2013; Tong et al., 2019;
Acosta et al., 2021 and Petermann-Rocha et al., 2021.
Acosta et al. conducted a prospective cohort study on 26 900 participants from
Sweden, including males and females. “Fish” included both fish and shellfish. The
outcome was total atherosclerotic cardiovascular disease, which was defined as a
diagnosis of CHD, ischemic stroke, coronary artery disease and peripheral artery

98
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

disease (that is, as incidence in the original study).

Lajous et al. conducted a prospective cohort study on 79 569 participants from the
United States, including males and females. “Fish” was defined as the sum of canned
tuna, dark fish, white fish and others (such as breaded fish). Thus, the exposure was
treated as total fish intake.
Matheson et al. conducted a prospective cohort on 13 355 participants from the
United States, including males and females. Shellfish consumption was the dietary
exposure (Table 3).
Tong et al. conducted a prospective cohort study on 48 188 participants from the
United Kingdom of Great Britain and Northern Ireland, including males and
females. The participants were categorized as meat-eaters, fish-eaters and vegetarians
(combined with vegans).
The study by Petermann-Rocha et al. is described in Section 3.2.5, Cardiovascular
diseases and outcomes.

3.2.5.3.2.2 Results from the primary studies included

Acosta et al. (2021) found total fish and shellfish intake had a significant protective
effect against ACVD incidence (HR = 0.95, 95% CI 0.93, 0.98).
Lajous et al. found no association between total fish intake and CHD incidence
for males (RR = 1.03, 95% CI 0.90, 1.15), but significant protective association for
females (RR (95% CI) = 0.87 (0.76-0.98), ≥ 2 servings/week versus 0 times/week).
Further, they found that the reduced risk increases with higher intake for females
(RR = 0.73, 95% CI 0.57, 0.94, ≥ 3 servings/week versus 0 times/week).
Matheson et al. found no association between shellfish intake and CHD incidence
(HR = 0.96, 95% CI 0.80, 1.16).
Petermann-Rocha et al. found a significant protective association of total fish intake
and CHD incidence (HR = 0.79, 95% CI 0.70, 0.88). A similar association was found
in Tong et al. (HR = 0.87, 95% CI 0.77, 0.99).
Tong et al. found a significant protective effect on ischemic heart disease, and that
fish-eaters has lower rates of ischemic heart disease than meat-eaters (HR = 0.87,
95% CI 0.77, 0.99).

99
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.5.3.3 Myocardial infarction

3.2.5.3.3.1 Description of the primary studies included
Three primary studies were included from the literature search for the outcome
“Myocardial infarction”: Gammelmark et al., 2016; Tong et al., 2019 and Petermann-
Rocha et al., 2021. (Table 3.9).
Gammelmark et al. conducted a prospective cohort study on 55 547 participants
from Denmark, including males and females. “Fish” included fatty and lean fish,
based on the content of ω-3 fatty acids (threshold = 1 g/100 g). Tong et al. conducted
a prospective cohort study on 48 188 participants from the United Kingdom of
Great Britain and Northern Ireland, including males and females. The Petermann-
Rocha et al. study was described in the section of CVD. The participants were
categorized as meat-eaters, fish-eaters and vegetarians (combined with vegans).

3.2.5.3.3.2 Results from the primary studies included

Gammelmark et al. found a non-significant protective association of fatty fish intake
on myocardial infarction incidence for both genders (HRMales = 0.88, 95% CI 0.77,
1.00), and HRFemales = 0.78, 95% CI 0.63, 0.96), but no association for lean fish intake
(HRMales = 1.05, 95% CI 0.91, 1.20), and HRFemales = 0.93, 95% CI 0.75, 1.14). They
found no dose–response relation between fatty fish intake and myocardial infarction
incidence. Petermann-Rocha et al. found a significant protective association of total
fish intake on myocardial infarction incidence (HR = 0.70, 95% CI 0.56, 0.88), while
Tong et al. found no association (HR = 1.00, 95% CI (0.78, 1.26).

3.2.5.3.4 Total stroke

3.2.5.3.4.1 Description of the primary studies included
Two primary studies were included from the literature search for the outcome “total
stroke” (Table 3.9): Tong et al., 2019 and Petermann-Rocha et al., 2021. The study by
Petermann-Rocha et al. was described in the section of CVD, and the study by Tong
et al. is described in Section 3.2.5, Cardiovascular diseases and outcomes. However,
both primary studies used FFQ with dichotomized terms (such as yes/no). While
Petermann-Rocha et al. conducted the FFQ at the baseline and specified the period
(for instance, “Have you consumed fish over the previous year?”), Tong et al. used
a more general question, without specifying the period (such as “Do you eat fish?”)
and sent out a follow-up FFQ in 2010 including identical questions.

3.2.5.3.4.2 Results from the primary studies

(Petermann-Rocha et al., 2021) found a significant protective association on total
stroke incidence (HR = 0.79, 95% CI (0.63, 0.98), while (Tong et al., 2019) found
no association (HR = 1.14, 95% CI 0.94, 1.38). However, in both included studies,
no frequency of consumption was given, and the meat-eater group in both primary
studies had fish in their diet. This might have caused heterogeneity.

100
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.5.3.5 Ischemic stroke

3.2.5.3.5.1 Description of the primary studies included
Two primary studies were included from the literature search for the outcome
“ischemic stroke” (Table 3.9). Venø et al. (2018) conducted a prospective cohort
study on 55 338 participants from Denmark, including both males and females. A
substitution analysis was conducted by substituting fish for red meat or poultry at
150 g/week. The exposure included total, fatty and lean fish (threshold between fatty
and lean fish = 1 g/100 g of ω-3 polyunsaturated fatty acid [PUFA]). The study by
Tong et al. (2019) was described in the section on CHD.

3.2.5.3.5.2 Results from the primary studies

Venø et al. found no association between total fish intake and total ischemic stroke
incidence, but lower risk of subtypes of ischemic stroke. Tong et al. also found no
association (HR = 1.05, 95% CI 0.80, 1.39).

3.2.5.3.6 Haemorrhagic stroke

3.2.5.3.6.1 Description of the primary study included
One primary study, Tong et al., 2019, was included from the literature search for
the outcome “haemorrhagic stroke” (Table 3.9). This study was described in Section
3.2.5, Cardiovascular diseases and outcomes.

3.2.5.3.6.2 Results from the primary study

Tong et al. found no association of total fish intake on haemorrhagic stroke incidence
(HR = 1.12, 95% CI 0.78, 1.61).

3.2.5.3.7 Atrial fibrillation

3.2.5.3.7.1 Description of the primary study included
One primary study was included from the literature search for the outcome “atrial
fibrillation” (Table 3.9). The study, Frost et al. (2005), was a prospective cohort
study from the Danish Diet, Cancer and Health Study, including 47 949 participants.
The participants included in the study were both men and women from the general
adult population. The study used FFQ to estimate food and fish consumption,
relying on the Danish food composition tables. The intake of n-3 from fatty fish was
categorized based on the frequency of consumption of herring, mackerel, sardine,
trout and salmon, which are readily available and commonly consumed in Denmark.

3.2.5.3.7.2 Results from the primary study included

Frost et al. found no significant association between n-3 consumption from fatty fish
and the risk of atrial fibrillation or flutter in their study. However, when comparing
the highest quantile of intake with the reference category (quantile 1), an increased
risk (HR = 1.34, 95% CI (1.02, 1.76) of atrial fibrillation was found.

101
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.5.3.8 Heart failure

3.2.5.3.8.1 Description of the primary study included
One primary study was included from the literature search for the outcome “heart
failure” (Table 3.9). Petermann-Rocha et al. (2021) has already been described in
the section on total CVD (3.2.5.3.1.1 Description of the primary studies included).

3.2.5.3.8.2 Results from the primary study included

Petermann-Rocha et al. found that after a median follow-up period of 8.5 years,
individuals who consumed fish had a lower risk of heart failure compared to meat-
eaters, with a HR of 0.78 (95% CI 0.63-0.97) after adjusting for confounding factors.
In contrast, there was no significant difference in the risk of heart failure between
individuals who consumed fish or poultry compared to meat-eaters, with a HR of
0.94 (95% CI 0.74, 1.20).

3.2.5.3.9 Peripheral arterial disease

3.2.5.3.9.1 Description of the primary study included
One primary study (Lasota et al., 2019) was included from the literature search for
the outcome “Peripheral arterial disease” (Table 3.9). This was a prospective cohort
study from the Danish Diet, Cancer and Health Study, including 54 597 participants,
consisting of both men and women from the general adult population. The study
used FFQ for estimating food and fish consumption, relying on the Danish food
composition tables.

3.2.5.3.9.2 Results from the primary study included

Lasota et al. (2019) suggest that substituting poultry and red meat with fish,
whether lean or fatty, is associated with a reduced risk of peripheral arterial disease.
Specifically, the study found that replacing red meat with total fish, particularly
fatty fish, was linked to a lower risk of peripheral arterial disease. Lower risk of
peripheral arterial disease was observed when fatty fish replaced unprocessed red
meat (RR = 0.89, 95% CI 0.79, 1.00), processed red meat (RR = 0.88, 95% CI 0.78,
1.01), or lean fish (RR = 0.90, 95% CI 0.76, 1.07).

102
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES”

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Acosta et al., 2021 Prospective cohort study 26 990 participants were The measurement and Risk of incident The study found that higher B The study found that
Malmö Diet and Cancer 1991-1996, median included in the final intake of fish and seafood atherosclerotic intake of fish and shellfish higher intake of fish
Study (MDCS) follow-up time: 21.1 years analysis. were assessed using a cardiovascular disease was associated with a and shellfish was
Incident of atherosclerotic 7-day food diary and a (ACVD), which was reduced risk of atherosclerotic associated with a reduced
Sweden
cardiovascular disease 168-item food frequency defined as the composite cardiovascular disease risk of atherosclerotic
(n = 5 858; mean 61.8 questionnaire that endpoint of coronary (ACVD) (HR 0.95 per SD cardiovascular disease
years, males: 51.1 included foods regularly artery disease, all-cause increment, 95% CI 0.93–0.98; (ACVD)
percent) consumed in the past ischemic stroke, carotid p = 0.001).
year. Complementary artery disease, and
No incident (21 132;
information was peripheral arterial
mean 56.1 years, males:
gathered through 1-hour disease.
34.5 percent)
interviews. The intake
of fish and shellfish was
reported in grams per
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

week.
Frost et al., 2005 Prospective cohort study n = 47 949 The measurement of Incidence of atrial Adjusted hazard ratio (HR) B The overall conclusion
The Danish Diet, Cancer, 1993–1997 follow-up Median age: 55.6 years fish intake in this study fibrillation or flutter. (95% CI) of atrial fibrillation of the study was that
and Health Study time: The follow-up time was based on a detailed or flutter in quantiles of n-3 the intake of n-3
n = 22 528 men
for the study mentioned semi-quantitative PUFAs from fish consumption: polyunsaturated fatty
Denmark (47 percent)
in the text is not explicitly food frequency Quantile 1 (reference): 0 acids from fish was not
n = 25 421 women questionnaire (FFQ). Quantile 2: 0.86 (0.65, 1.15) associated with a reduced
stated. Mean follow-up:
(53 percent) The study participants Quantile 3: 1.08 (0.82, 1.42) risk of atrial fibrillation
5.7 years
were asked to fill in a Quantile 4: 1.01 (0.77, 1.34) or flutter in this cohort of
questionnaire about the Quantile 5: 1.34 (1.02, 1.76) Danish men and women.
type and frequency of fish P for trend = 0.006 However, the highest
consumption, and the quantile of intake had an
daily intake of specific increased risk compared
foods and nutrients was to the reference category.
computed from the FFQ
for each participant with
the use of the software
program FOODCALC.

103
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES” (cont.)

104
NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Gammelmark et al., 2016 Prospective cohort study n = 55 547 Fish and seafood The outcome of interest The paper reports that there B The study found that
The Danish Diet, Cancer 1993–1997 Cases: 3 028 consumption was in this study was incident was a consistent inverse high intake of fatty fish
and health study quantified using a myocardial infarction association between high was inversely related
Median follow-up time: Cases, men: 2 136 (70.5
detailed and validated (MI). fish intake and incident to incident myocardial
Denmark 17 years percent)
food frequency myocardial infarction (MI). infarction (MI) in both
Median age: 57.7 years
questionnaire (FFQ). When comparing the highest men and women when
Cases women: 892 (29.9
Different species of fish and the lowest quintile of comparing the highest
percent)
were categorized as either fatty fish intake, there was a and lowest quintiles.
Median age: 59.3 years
lean or fatty depending on 12 percent lower RR of MI in However, a clear dose-
their content of n-3 PUFA, men (hazard ratio (HR) 0·88; response relationship
below or above 1 g/100 95% CI 0·77, 1·00) and a 22 could not be established,
g, respectively. Fatty percent lower HR in women and the test for trends
fish mainly comprised (HR 0–78; 95% CI 0·63, 0·96), across quintiles was not
herring, salmon, trout and after adjustments. statistically significant
mackerel, whereas lean in the adjusted analyses.
fish comprised mainly Lean fish was not
plaice, flounder and cod. associated with MI. The
Intake was measured in study supports the current
grams per day. view that consumption
of fatty fish may protect
against MI.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Lajous et al., 2013 2 prospective cohort n = 79 569 The measurement The outcome of the study The risk ratios for CHD B Increasing fish
The Health Professionals studies: 32.4 percent men of fish and seafood was the risk of coronary when comparing the risk consumption during
Follow-Up Study and the n = 25 979 men: 67.6 percent women consumption in the heart disease (CHD). The if everyone had consumed midlife may lower the risk
Nurses’ Health Study 1990–2008 study was conducted researchers estimated at least 2 servings of fish of coronary heart disease
Two prospective US
and using a 127-food item CHD risk under different per week with the risk if no (CHD) in women but not
the United States cohorts: 25 797 men in
n = 53 772 women: semiquantitative food hypothetical interventions one consumed fish during men. The researchers
the Health Professionals
1986–2008 frequency questionnaire. on fish consumption the follow-up periods were estimated CHD risk under
Follow-Up Study and 53
This questionnaire during mid- and later 1.03 (95% CI 0.90, 1.15) for different hypothetical
772 women in the Nurses’
was first sent to life in two prospective US men and 0.87 (95% CI 0.76, interventions on fish
Health Study. At baseline,
participants in the cohorts of 25 797 men in 0.98) for women. The results consumption during
the average age of male
Health Professionals the Health Professionals suggest that increasing fish mid- and later life in two
participants was 56.5
Follow-Up Study in 1986, Follow-Up Study and 53 consumption to at least 2 prospective US cohorts,
years (standard deviation,
to participants in the 772 women in the Nurses’ servings per week in mid- or and they observed
9.3). The average age of
Nurses’ Health Study in Health Study. later life may lower CHD risk 1 865 incident CHD
female participants at
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

1984 and 1986, and to all in women but not in men. cases among men (in
baseline was
participants every 4 years 1990–2008) and 1 891
52.1 years (standard
afterward. CHD cases among women
deviation, 7.1). The study
(in 1986–2008).
collected data on risk
factors and disease every
2 years and on diet every
4 years from both cohorts.

105
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES” (cont.)

106
NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Lasota et al., 2019 Prospective cohort study n = 54 597 The measurement of fish The measurement of The overall results of B The conclusion of the
Diet, Cancer and Health Established between 1993 Median of 13.6 years of and seafood consumption outcome in this study this study suggest that study was that a higher
Study and 1997 follow up was conducted using was the incidence of substituting poultry and red intake of fish and a lower
a software program peripheral arterial disease meat with fish, whether total, intake of poultry or red
Denmark Median of 13.6 years of 47 percent males: 25 725
based on Danish food (PAD). lean or fatty, is associated meat were associated
follow up 53 percent females:
composition tables. The with a lower risk of peripheral with a lower risk of
28 872
participants completed arterial disease. Specifically, incident peripheral
questionnaires about the study found that replacing arterial disease.
their dietary intake. red meat with total fish and Specifically, replacing red
Substitution of meat or especially fatty fish was meat with total fish and
poultry with fish associated with a lower risk of especially fatty fish was
PAD. The replacement of lean found to be associated
fish with fatty fish showed a with a lower risk of PAD,
similar association. although the results
No associations with incident were only borderline
PAD could be statistically significant.
The replacement of lean
demonstrated when lean fish
fish with fatty fish showed
replaced poultry, unprocessed
a similar association.
red meat or processed No associations with the
red meat. In addition, no risk of PAD were found in
association models.
was observed when fatty fish
replaced poultry. However,
when fatty fish replaced
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

unprocessed red meat (0.89;

95%
CI 0.79–1.00), processed
red meat (0.88; 95% CI
0.78–1.01), or lean fish (0.90;
95% CI 0.76–1.07), lower
risks
of PAD were found, although
these were not statistically
significant.
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Matheson et al., 2009 Prospective cohort study n = 13 355 Shellfish consumption Incidence of CHD In total, 1 382 suffered B No association was
Atherosclerosis Risk in 1987–1989 (baseline) 45–64 years assessed with a CHD events; 41.8 percent found between shellfish
Communities study, a through December 31, modified version of a male, 58.2 percent female. consumption and the
41.8 percent men
cohort of middle-aged 2001. previously validated Reference group: Low intake risk of adverse coronary
and elderly adults FFQ and participants vs.: heart disease events.
Median follow up not
were departed into three Medium intake: hazard ratio Possibly due to generally
the United States given
categories: (HR) [95% CI] 0.89 [0.79- low consumption in the
Low: almost never ate 1.00] (unadjusted) and 0.96 cohort, and potentially
shellfish [0.8-1.16] (adjusted); less healthy way of
Medium: ate shellfish one preparing shellfish among
High intake: HR [95% CI] 0.91
to three times per month the cohort population
[0.80-1.03] (unadjusted) and
High: ate shellfish once a compared with that from
0.98 [0.82-1.18] (adjusted)
week or more another existing study.
Baseline intake:
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Low: 62 percent
Medium: 28 percent
High: 9 percent
Petermann-Rocha et al., Prospective cohort study n = 422 791 Diet type assessed with a Incident and fatal 106 690 (24 percent) B Compared with meat-
2021 2006–2010 (baseline) up 37–73 years touch-screen FFQ. event due to CVD, developed CVD and 6 580 eaters, fish eaters
General population to June 2020 in England Baseline: 94.7 percent IHD, MI, stroke and died from CVD. Fish eaters had a lower risk of
55.4 percent women
recruited by UK Biobank and March 2017 in Wales meat-eater; fish-eater; HF (International had lower risks (in HR) of CVD several cardiovascular
and Scotland. vegetarian. Classification of Disease 0.93, IHD 0.79, MI 0.70, stroke outcomes–incident
the United Kingdom
[ICD]s Tenth revision). 0.79 and HF 0.78 (adjusted); CVD, IHD, MI, stroke,
of Great Britain and Median follow-up: 8.5
no association between diets and HF–independent of
Northern Ireland years (CVD incidence) and
and CVD mortality. confounders. People who
9.3 years (CVD mortality).
ate poultry and fish, did
not have a lower risk.

107
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES” (cont.)

108
NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Veno et al., 2018 Prospective cohort study n = 55 338 Diet was assessed using Incident IS and 1 879 IS occurred, with B Fish intake was not
The Diet, Cancer and 1993–1997 50–64 years old a validated 192-item FFQ, subtypes of IS: an subtypes large artery associated with risk of
Health Cohort Study together with a lifestyle acute disturbance of atherosclerosis (319), total IS but is associated
Median follow-up: 13.5 47.6 percent men
questionnaire, and a focal or global cerebral cardioembolism (102), with a lower risk of
Denmark years.
physical examination. function with symptoms small-vessel occlusion subtypes.
Substitutions of 150 g/ lasting more than 24 (844), other ethology (98)
week fish for 150 g/week hours. Cases identified and undetermined ethology
of red meat or poultry according to ICD-8 or (516). No association between
ICD-10 total IS incidence and fish
replacement. Lower rate of
large artery atherosclerosis:
HR 0.78 [0.67–0.90]
(processed) and 0.87 [0.75–
0.99] (unprocessed red meat).
Higher rate of cardioembolism
when poultry replaced total
fish (1.42 [1.04–1.93]). Lower
rate of small-vessel occlusion
when unprocessed red meat
was replaced by fatty fish
(0.88 [0.77–0.99]).
Tong et al., 2019 Prospective cohort study n = 48188 Diet groups determined Outcomes include 2 820 cases of IHD and 1 B Fish eaters had lower
EPIC-Oxford cohort 1993–2001. Recruitment at 35–59 by FFQ: meat eaters ischaemic heart 072 cases of total stroke risks of ischaemic heart
(general practice) years (M, include fish), fish disease, including acute recorded. After adjustment, disease than meat eaters.
the United Kingdom Median follow-up:
old and 20+ years old eaters (F, no meat at all), myocardial infarction, fish eaters had lower rates of
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

of Great Britain and 18.1 years (follow-up

(postal recruitment) vegetarian (V, including total stroke including ischaemic heart disease than
Northern Ireland questionnaire sent in
vegan). ischaemic stroke and meat eaters (0.87 [0.770.99]).
2010, returned between 76.6 percent women
Also, a validated haemorrhagic stroke. No significant differences
2010-2013)
semiquantitative FFQ on between diet groups for the
dietary intake over the risk of myocardial infarction
past year. or ischemic stroke
 TABLE 3.9 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH IN THE THEME “CVD OUTCOMES” (cont.)

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT, MEN)
Zhong et al., 2021 Prospective cohort study n = 29682 Dietary intake determined Incident CVD (including 6 963 incident CVD cases B Substituting fish for eggs
Six prospective cohort 1985–2016. 53.7±15.7 years old by a validated diet history CHD, stroke, heart failure and 8 875 deaths recorded. and processed meat was
studies or FFQ. and CVD death); all Substituting (1 serving per associated with lower
Median follow-up: 19.0 55.6 percent women
events adjusted by each week) eggs with fish (+nuts, risks of incident CVD. The
the United States (14.1-23.7) years
original cohort. legumes or whole grains) was reduced risks increase
associated with 2–3 percent from 1 serving/week to 1
lower risks and 0.4–0.7 serving/day.
percent lower absolute risks
for incident CVD; (substitution
=1 serving per day) 15–21
percent lower RR and 3.4–4.8
percent lower absolute risks.
Substituting processed meat
with fish (+nuts, legumes or
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

whole grains) was associated

with 4–5 percent lower RR
and 0.9–1.2 percent lower
absolute risks for incident
CVD (1/week); 28–34 percent
lower RR and 7.5–9.0 percent
lower absolute risks (1/day).
Other substitutions with fish
are not significant.

Notes: SD: standard deviation, CI: confidence interval, HR: hazard ratio, PUFA: polyunsaturated fatty acid, IS: ischemic stroke, IHD: ischemic heart disease, CVD: cardiovascular disease, HF: heart failure,
RR: risk ratio.

109
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.5.4 Final weight of evidence for “Cardiovascular diseases and outcomes”

A final weight of evidence for the theme “CVD outcomes” was based on the 2022
VKM report and on the systematic literature search. An overview of the literature
included in the final weight of evidence is given in Appendix 3, Table A3.35.

3.2.5.4.1 Total cardiovascular disease incidence

There is evidence from more than two independent, high-quality cohort studies of
the association between total fish intake and reduced risk of total CVD. Among the
primary studies and the 2022 VKM report, the direction of association is generally
consistent on the protective side, with high heterogeneity. The biological plausibility
is evident with a dose-response relation. However, dose-response analysis alone
does not contribute to an upgrading factor.
In conclusion, the evidence was graded “limited, suggestive” for a protective effect
of total fish intake on total CVD incidence. Due to fewer studies of the incidence
of fatty and lean fish on total CVD incidence than studies on the incidence of total
fish intake, and as the evidence was not significant, the evidence was graded “limited,
no conclusion” for an effect of fatty and lean fish intakes on total CVD incidence.

3.2.5.4.2 Coronary heart disease

Among the literature from the 2022 VKM report and the systematic reviews and
primary studies from the literature search, the direction of association is generally
consistent on the protective side, but with high heterogeneity. There is evidence
of this from more than two independent and high-quality cohort studies. The
biological plausibility is evident with a dose–response relation.
In conclusion, the evidence is graded “probable” for a protective effect of total fish
intake on CHD incidence. Due to fewer studies of fatty and lean fish, as well as
shellfish, the evidence was graded “limited, suggestive” for a protective effect of fatty
fish intake on CHD incidence; and the evidence was graded “limited, suggestive”
for no effect of lean fish intake, and “limited, no effect” for any effect of shellfish
intake on CHD incidence.

3.2.5.4.3 Myocardial infarction

In the 2022 VKM report and the primary studies included from the literature search,
the direction of association is generally consistent on the protective side, with
some exceptions. In conclusion, the evidence is graded “limited, suggestive” for a
protective effect of total fish intake on myocardial infarction incidence. There were
a few studies on the effect of fatty and lean fish on myocardial infarction incidence,
and the evidence was graded “limited, suggestive” for a protective effect of fatty
fish on myocardial infarction incidence and “limited, suggestive” for no effect of
lean fish on myocardial infarction incidence.

110
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.5.4.4 Total stroke

Among the included literature from the 2022 VKM report, and the systematic
reviews and primary studies from the literature search, the direction of association
is generally consistent on the protective side and moderate heterogeneity was
reported. There is evidence from more than two independent and good-quality
cohort studies on total fish intake. The biological plausibility is evident with a dose–
response relation. In conclusion, the evidence is graded “probable” for a protective
association for total fish consumption on total stroke incidence. The evidence is
graded “limited, suggestive” for a protective effect of fatty and lean fish intake, and
“limited, no conclusion” for any effect of shellfish intake on total stroke incidence,
due to fewer studies.

3.2.5.4.5 Ischemic stroke

There is evidence from more than two independent and good-quality cohort
studies on the effect of total fish intake on ischemic stroke. Among the studies, the
direction of association is generally consistent on the protective side with moderate
heterogeneity, which was explained. No dose–response analysis was performed.
In conclusion, the evidence is graded “limited, suggestive” for a protective effect of
total fish intake on ischemic stroke incidence.

3.2.5.4.6 Haemorrhagic stroke

There is evidence from more than two independent and good quality cohort studies
on the effect of total fish intake on haemorrhagic stroke. Among the studies, the
direction of association is generally consistent on the protective side with low to
moderate heterogeneity. There is biological plausibility with a possible linear dose–
response relation.
In conclusion, the evidence is graded “limited, suggestive” for a protective effect of
total fish intake on haemorrhagic stroke incidence.

3.2.5.4.7 Atrial fibrillation

In conclusion, the evidence by VKM was graded “limited, suggestive” for a potential
adverse effect of total fish intake on the risk of atrial fibrillation. For fatty fish
consumption, the evidence was graded “limited, no conclusion”; while for lean
fish consumption, the evidence was graded “limited, suggestive” for a protective
effect. The literature search included one additional study, which did not change
the weight of evidence conclusions from VKM. Thus, the final weight of evidence
for the outcome atrial fibrillation are graded “limited, suggestive” for a potential
adverse effect of total fish intake, “limited, no conclusion” for fatty fish intake, and
“limited, suggestive” for a protective effect of lean fish intake.

111
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.5.4.8 Heart failure

VKM based its conclusion on eight studies that examined heart failure as an outcome.
These studies were distinct and did not overlap. The overall trend of the findings
consistently leaned towards a protective effect, with low heterogeneity observed.
Furthermore, there is supporting evidence for the biological plausibility of these
effects.
The summary RR conducted by VKM suggested that there is evidence indicating
a protective effect of fish intake on heart failure, except when the fish is fried. In
summary, the available evidence suggested that consuming fish may reduce the risk
of heart failure, although this conclusion is graded "limited, suggestive". In our
search, we found one additional study by Petermann-Rocha et al., (2021), however
this study does not alter VKM’s conclusion. The overall evidence supporting the
notion that fish consumption reduces the risk of heart failure remains graded
"limited, suggestive".

3.2.5.4.9 Venous thromboembolism

No additional studies were found in the search, and the evidence regarding fish
consumption and reduced risk of venous thromboembolism remains graded
"limited, no conclusion."

3.2.5.4.10 Peripheral arterial disease

The search resulted in one study evaluating the risk of peripheral arterial disease
and fish intake. A lower risk of peripheral arterial disease was observed when fatty
fish replaced unprocessed meat, processed meat and fish. Due to the small number
of studies available and potential methodological modifications, the evidence is
graded "limited - no conclusion" for the protective effect of fish and fatty fish on
the risk of PAD.

3.2.6 TYPE 2 DIABETES

The literature included for the health outcome “Type 2 diabetes” includes results
from the 2022 VKM report and seven systematic reviews and one original primary
study originating from the literature search.

3.2.6.1 Summary of the findings on “Type 2 diabetes” in the VKM report, Benefit and
risk assessment of fish in the Norwegian diet
The report included 1 umbrella review, 3 systematic reviews and 16 primary studies
for the health outcome T2D.
In the two systematic reviews, which included a meta-analysis (Schwingshackl et
al., 2017 and Namazi et al., 2019), no strong evidence for protective or adverse
associations between total fish intake and T2D were observed. In the third systematic
review, which included a federated meta-analysis, an adverse modest association was

112
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

found between total fish, fatty fish and lean fish intake and T2D incidence in women,
but not in men (Pastorino et al., 2021).
The summary RR of the primary studies on total fish intake and T2D showed no
significant association (RR =1.04, 95% CI 0.96, 1.14) and heterogeneity was high.
The summary RR of the primary studies (n =7) on fatty fish intake and T2D proved
a protective association (RR = 0.88, 95% CI 0.78, 0.99), and the heterogeneity was
borderline significant. For lean fish intake, no significant association was observed,
but significant heterogeneity was found.
In conclusion, the evidence of the association between total fish or fatty fish
consumption and the risk of T2D was graded “limited, no conclusion”, and the
association between lean fish intake and T2D was graded “limited, suggestive” (no
association).

3.2.6.2 Summary of the findings on “Type 2 diabetes” from systematic reviews

included from the literature search
Seven systematic reviews and meta-analyses were included (Table 3.10) (Wallin et
al., 2012; Wu et al., 2012; Xun et al., 2012; Zheng et al., 2012; Zhou et al., 2012;
Zhang et al., 2013 and Muley et al., 2014). All seven included prospective cohort
studies with T2D as an outcome. The participants were men and women from the
general population. Total fish intake was included in six of the studies, whereas
fatty fish, lean fish and shellfish was included in two studies. In addition to overall
risk estimates, analyses stratified by geographic region were performed in five of
the studies.
The results from all meta-analyses that included total fish intake showed that there
were no associations between total fish intake and risk of T2D. However, in all
stratified analyses, a higher total fish intake was associated with a higher risk of
T2D for Western populations, whereas the opposite was observed for the Asian
population.
In one meta-analysis, fatty fish intake was associated with lower risk of T2D.
However, no significant association was observed in the other meta-analysis
investigating fatty fish intake separately. No significant associations were found
between the intake of lean fish or shellfish and T2D.

113
 TABLE 3.10 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “TYPE 2 DIABETES”

114
AUTHOR, YEAR FISH AND OVERALL RISK OF BIAS
OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS
STUDY TITLE SEAFOOD INTAKE CONCLUSION (AMSTAR 2)
Muley et al., 2014 Type 2 diabetes General adult population n =10 for fish intake Information on fish intake Results from meta-analyses Fatty fish intake, Moderate
ALA, Fatty Fish or > 18 years old. n = 7 for total fish was obtained by FFQ showed that the pooled but not lean fish or
Marine n-3 Fatty Acids Total n = 679 763 n = 2 for fish + shellfish and given in mg/day or estimate for fatty fish intake shellfish, was related
for Preventing DM?: A n = 4 for fatty fish portions/week was associated with a to decreased risk
Systematic Review and n = 3 for lean fish reduced risk of T2D (RR = of T2D.
Meta-Analysis Prospective cohort studies 0.89, 95% CI 0.80, 0.98)
with a heterogeneity of I2 =
0, p = 0.028. No significant
association was found
between the risk of T2D and
lean fish (RR = 1.02, 95% CI
1.03, 1.12) or shellfish (RR =
0.89, 95% CI 0.70, 1.13)
Wallin et al., 2012 Type 2 diabetes General adult population n = 10 for total fish 11 studies used self- High degree of heterogeneity Total fish intake was Moderate
Fish consumption, dietary Total n = 527 441 intake based on 13 cohort administered FFQ, 5 between the 13 studies associated with an
long-chain n-3 fatty studies studies used interview- included (I2 = 81.3 percent, p increased risk of T2D
T2D cases = 24 082
acids, and risk of type Prospective cohort studies administered FFQ. Intakes < 0.001) and results across in the United States
2 diabetes: systematic were given as servings/ studies were not combined and with a decreased
review and meta-analysis week. in one overall risk estimate risk in Asia, and
of prospective cohort but divided into geographic no association was
studies regions. In the United States, observed for Europe.
an increased risk of T2D was
observed for an increment
with one serving of total fish
per week (RR = 1.05, 95%
CI 1.02, 1.09). In Europe,
no significant associations
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

were observed (RR = 1.03,

95% CI 0.96, 1.11). In Asia a
decreased risk was observed
(RR = 0.98, 95% CI 0.97,
1.00)
 TABLE 3.10 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “TYPE 2 DIABETES” (cont.)

AUTHOR, YEAR FISH AND OVERALL RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS
STUDY TITLE SEAFOOD INTAKE CONCLUSION (AMSTAR 2)
Wu et al., 2012 Type 2 diabetes General adult population For total fish intake, 13 Intake based on FFQ. RR for 100g/day. For total Total fish intake Moderate
Omega-3 fatty acids and studies included. Intakes were given as fish/seafood intake, no was not associated
incident type 2 diabetes: Prospective cohort studies g/day. significant association was with increased
n for fish intake = 481
a systematic review and observed (RR = 1.12, 95% CI or decreased risk
489
meta-analysis 0.94, 1.34). of T2D in total.
T2D cases: 20 830 However, when
In stratified analyses, a
decreased risk of T2D was geographic regions
observed in Asia (RR = 0.89, were investigated,
95% CI 0.81, 0.98) and an total fish intake
increased risk was observed in was related to lower
the United States/Europe (RR risk of T2D in Asia
= 1.38, 95% CI 1.13, 1.70). and higher risk in
Substantial heterogeneity (I2 the United States /
around 80%) was observed Europe.
among the studies included.
Xun et al., 2012 Type 2 diabetes General adult population 9 studies included (12 Fish intake based on FFQ. Total fish consumption (never Overall, no significant Moderate
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Fish consumption and n = 438 214 independent cohorts) Intakes were given as or less than once/month vs association between
incident diabetes. Prospective cohort studies servings/week. 2–4 times/week) was not total fish intake and
associated with incident T2D was observed.
Meta-analysis
diabetes (RR = 1.00, 95% CI However, when
0.85, 1.18). geographic regions
In stratified analyses, lower were investigated,
risk of T2D was observed in total fish intake was
Eastern countries (Asia) but related to lower risk
not in Western countries. The of T2D in Eastern
heterogeneity was good (I2 = countries but not in
83.7%, p=0.000). Western countries.

Zhang et al., 2013. Type 2 diabetes General adult population 10 studies included Fish intake obtained by No significant association Overall, no significant Moderate
Fish and marine omega-3 n = 549 955 Prospective cohort studies FFQ. Reported as g/day. between total fish intake and association between
polyunsaturated fatty T2D was observed (RR = 1.04, total fish intake and
acid consumption and 95% CI 0.89, 1.20). There was T2D was observed.
incidence of type 2 significant study heterogeneity However, when
diabetes: a systematic (I2 83%, P < 0.00001). geographic regions
review and meta-analysis Stratified analyses showed were investigated,
a beneficial effect of fish total fish intake was
consumption on T2D risk in related to lower risk
the Asian population, but not of T2D in the Asian
in the Western population. population.

115
 TABLE 3.10 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS INCLUDED FROM THE LITERATURE SEARCH ON “TYPE 2 DIABETES” (cont.)

116
AUTHOR, YEAR FISH AND OVERALL RISK OF BIAS
OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS
STUDY TITLE SEAFOOD INTAKE CONCLUSION (AMSTAR 2)
Zheng et al., 2012.
Marine N-3
polyunsaturated fatty
acids are inversely
associated with risk of
type 2 diabetes in Asians:
a systematic review and
meta-analysis.
Zhou et al., 2012. Type 2 diabetes General adult population 6 publications included (9 Fish intake obtained by Lowest vs highest category Higher total Moderate
Association of fish and n = 367 757 cohort studies). FFQ. Reported as g/day or of total fish intake showed no fish intake was
n-3 fatty acid intake Prospective cohort studies categories. significant association for the associated with a
with the risk of type 2 risk of T2D (RR = 1.14, 95% modest higher risk of
diabetes: a meta-analysis CI 0.97, 1.34) with substantial T2D in linear analysis
of prospective studies. heterogeneity between the but not in the
studies (I2 = 79%, p< 0.001). categorical analysis.
The dose-response relation
was also tested in linear
analysis. For total fish intake,
a significantly higher risk
of T2D was observed (RR =
1.04, 95% CI 1.02, 1.05). No
between-study heterogeneity
was observed (I2 = 0.00%, p
= 0.421).

Notes: FFQ: food frequency questionnaire, T2D: type 2 diabetes, RR: risk ratio, CI: confidence interval, P: P-value
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.6.3 Summary of the findings on “Type 2 diabetes” from primary studies included
from the literature search
3.2.6.3.1 Description of the primary study
One primary study by Chen et al. (2020) was included (Table 3.11). The study is a
prospective cohort study, including three subcohorts with participants living in the
Ommoord District of Rotterdam, the Kingdom of the Netherlands.

3.2.6.3.2 Description of study population

A total of 6 813 participants ≥ 45 years of age, without diabetes at baseline, were
included. The cohort comprised both men and women, with 41.4 percent men. The
mean (standard deviation) age at baseline was 65.4 (11.3) years. During a median
follow-up time of 7.2 years, 643 T2D cases were documented.

3.2.6.3.3 Description of fish consumption

At baseline, dietary intake was obtained by a validated semi-quantitative 170-item
FFQ. For follow-up, a validated semiquantitative 289-item FFQ was used. The
participants had a median (25-75th percentile) daily intake of proteins from fish of
2.9 (0.6-5.7) g/day.

3.2.6.3.4 Results from the primary studies included

In multivariable models, intake of protein from fish was associated with an increased
risk of T2D (HR = 1.65, 95% CI 1.30,2.10). In sensitivity analyses by age, sex or
waist circumferences, the results were similar to those of the main analysis.

117
 TABLE 3.11 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDY (COHORT STUDY) INCLUDED FROM THE LITERATURE SEARCH ON “TYPE 2 DIABETES”

118
NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT MEN)
Chen et al., 2020 Prospective cohort study m = 6 813 Dietary intake obtained At baseline and follow-up, Protein from fish intake was B An increased risk of type
The Rotterdam Study 1993–2014 Type 2 diabetes cases using a validated type 2 diabetes was associated with increased risk 2 diabetes was observed
= 643 semiquantitative 170- identified from general of type 2 diabetes (HR = 1.65, with intake of protein
Rotterdam, the Kingdom
Follow-up time: Median item FFQ. practitioners, structured 95% CI 1.30,2.10). from fish.
of the Netherlands
7.2 years. Protein from fish: median home interviews,
(25th–75th percentile) – pharmacy dispensing
Mean (SD) 65.4 (11.3)
2.9 (0.6–5.7) g/day. records and follow-up
years.
examination at the
41.4 percent men. research centre. Type 2
diabetes was defined
according to World Health
Organization guidelines.

Notes: SD: standard deviation, FFQ: food frequency questionnaire, HR: hazard ratio, CI: confidence interval
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.6.4 Final weight of evidence for “Type 2 diabetes”

A final weight of evidence for the theme “Type 2 diabetes” (T2D) was based on
the 2022 VKM report, the 2018 WCRF report and the systematic literature search.
An overview of the literature included in the final weight of evidence is given in
Appendix 3, Table A3.35.
In conclusion, the overall evidence of total fish and fatty fish intake associated with
T2D was graded “limited, no conclusion”. In addition, the evidence of lean fish
intake associated with T2D was graded “limited, suggestive (no association)”.

3.2.7 NEURODEVELOPMENT AND NEUROLOGICAL DISORDERS

The literature included in the theme “Neurodevelopment and neurological
disorders” includes results from the 2022 VKM report and two original primary
studies originating from the literature search. No systematic reviews were included.

3.2.7.1 Summary of the findings on “Neurodevelopment and neurological disorders” in

the VKM report, Benefit and risk assessment of fish in the Norwegian diet
The VKM report summarizes the evidence of an association between fish
consumption and the outcomes of “maternal fish intake and neurodevelopmental
outcomes in children”, “child fish intake and neurodevelopment in children”, “fish
intake and neurocognitive and psychiatric endpoints in adults”, and “fish intake and
depression and other psychiatric symptoms in adults”.

3.2.7.1.1 Maternal fish intake and neurodevelopmental outcomes in children

3.2.7.1.1.1 Description of the studies included
The VKM assessment included one systematic review (Hibbeln et al., 2019) and 22
original primary studies investigating the association between maternal fish intake
and neurodevelopmental outcomes in children.
The systematic review from Hibbeln et al. (2019) included 29 prospective
cohort studies (106 237 mother–child pairs), which evaluated the relationship
between maternal seafood consumption in pregnancy or during lactation and
neurodevelopmental outcomes in the infant. The systematic review concluded that
there was moderate and consistent evidence that maternal seafood consumption in
pregnancy is associated with improved neurocognitive development of offspring as
compared to eating no seafood. However, the evidence does not meet the criteria
for “strong evidence”, as there is a lack of RCTs.
Of the 22 primary studies, two were RCTs involving lean fish consumption and the
rest were prospective birth cohort studies. The majority of the studies incorporated
in the analysis originated from European countries, including Denmark, Italy,
Norway, the Kingdom of the Netherlands, Spain, Sweden and the United Kingdom
of Great Britain and Northern Ireland. Five studies were conducted in the United
States, while one study focused on findings from Japan. These studies encompass

119
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

a range of large-scale cohorts, such as the Avon Longitudinal Study of Parents

and Children; the Spanish Childhood and Environment project; Project Viva;
Generation R; the Norwegian Mother, Father and Child Cohort Study and the
Danish National Birth Cohort. Additionally, smaller studies with more limited
sample sizes were also included. Most of the studies reported total fish intake, which
included a summarization of all fish. Some studies also had subanalyses including
lean and fatty fish.

3.2.7.1.1.2 Conclusion, weight of evidence

In summary, the direction of associations was generally consistent towards a
protective effect of maternal fish consumption. In addition, there was limited,
unexplained heterogeneity and evidence for biological mechanisms. However,
the neurodevelopmental domains, age at assessment and assessment tools for
neurodevelopment varied substantially from study to study.
In conclusion, the evidence for a protective association between maternal total fish
consumption and child neurodevelopment was graded “limited, suggestive”. There
were fewer studies of fatty fish and lean fish than of total fish. As such, the evidence
for the effects of fatty fish and lean fish on child neurodevelopment was graded
“limited, no conclusion”.

3.2.7.1.2 Child fish intake and neurodevelopmental outcomes in children

3.2.7.1.2.1 Description of the studies included
The VKM assessment included one systematic review (Hibbeln et al., 2019) and ten
primary studies investigating child fish intake and child neurodevelopment outcomes.
The systematic review by Hibbeln et al. included six RCTs, four prospective cohort
studies and nine case-control studies, including 25 960 children altogether. The
authors concluded that there was moderate and consistent evidence indicating
that seafood consumption during childhood has beneficial associations with
neurocognitive outcomes.
Of the primary studies, four RCTs and two prospective cohort studies studied fatty
fish intake and cognition in children aged 4–18 years. Two RCTs and one prospective
cohort reported findings on child fish intake and mental health from birth until
18 years of age. One prospective study reported findings on child fish intake and
early child development (≤3 years). Two RCTs were from Norway, one was from
Denmark and one was from Germany. The prospective studies were from Sweden
and the United Kingdom of Great Britain and Northern Ireland. All studies included
both girls and boys and two studies stratified the analyses by sex.
Two publications involving child cognition reported associations between total fish
intake in adolescents at 15 years of age and school grades at 16 years and cognitive
abilities at 18 years. Both studies reported significantly better scores with higher
fish intake for all outcomes. One of the RCTs in preschool children significantly
reported larger improvements in the fish-intervention group compared to the control

120
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

group in two subtests for the cognitive outcomes, but not for the 15 remaining
comparisons. The other RCT study in preschool children reported no significant
effects of the fish intervention compared to meat on any outcome. In four of the
studies reporting findings from RCTs, estimates were presented unadjusted and
adjusted for dietary compliance.
For the mental health outcomes, one RCT in preschool children reported no
significant effect on scores of the Strengths and Difficulties Questionnaire. The RCT
in school-aged children reported significant difference between groups in 2 out of 17
included subtests; while the RCT in adolescents reported a protective effect of fish on
emotional problems and total problems in the dichotomized scores, but no such effects
for the four remaining outcomes and no protective effect when scores were used on
a continuous scale. The one prospective study reported no significant associations
between child total fish intake at three years and mental health problems.

3.2.7.1.2.2 Conclusion, weight of evidence

In the four prospective cohort studies on the association between total child
fish intake and neurodevelopment, two suggested a beneficial association in all
included comparisons and one in parts of the comparisons, while two reported no
significant associations. In these prospective studies, the age at outcome assessment
ranged substantially (from 15 months to 18 years) and, although the two studies in
adolescents reported protective associations, overall conclusions were limited by
the difference in outcome measures. The direction of the effects and associations
are generally consistent towards protection from fish intake with little unexplained
heterogeneity and there was evidence for biological mechanisms between child fish
intake and neurodevelopment (biological plausibility).
There was evidence from four independent RCTs involving fatty fish interventions
and four independent prospective cohort studies on total fish intake. Findings were
limited by multiple included outcome measures, multiple methods of outcome
measurement, multiple comparisons, and substantial age differences at outcome
assessment. In conclusion, the evidence was graded “limited, suggestive” for
child fish consumption (total and fatty fish) benefitting neurodevelopment. There
were few studies focusing on lean fish and, as such, this was graded “limited, no
conclusion”.

3.2.7.1.3 Fish intake and neurocognitive and psychiatric endpoints in adults

3.2.7.1.3.1 Description of the studies included
To investigate the association between fish intake and neurocognitive and psychiatric
endpoints in adults, the outcomes included were incidence of dementia and
Alzheimer’s disease, risks or symptoms of cognitive decline, and general cognition.
The VKM assessment included four systematic reviews investigating the association
between fish intake and cognitive decline in adults. These included the outcomes
of risk of dementia and Alzheimer’s disease. All four systematic reviews showed
that a higher consumption of fish was associated with a lower risk of dementia and

121
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Alzheimer’s disease:
> Kosti et al., 2022: Highest versus lowest intake of fish:
> Risk of dementia: n = 9 prospective cohort studies. Summary RR = 0.80,
95% CI: 0.69, 0.93.
> Risk of Alzheimer’s disease: n = 7 prospective cohort studies. Summary RR
= 0.74, 95% CI: 0.63, 0.87.
> Bakre et al., 2018: Consumed fish (or consumed fish at a higher level) compared
with those who did not eat fish (or consumed fish at a lower level):
> Risk of dementia: n = 15 prospective cohort studies and cross-sectional
studies. Summary RR = 0.80, 95% CI: 0.74, 0.87.
> Risk of Alzheimer’s disease: n = 7 prospective cohort studies and cross-
sectional studies. Summary RR = 0.73, 95% CI: 0.65, 0.82.
> Zeng et al., 2017: Highest versus lowest intake of fish:
> Risk of dementia: n = 6 prospective cohort studies. Summary RR = 0.86,
95% CI: 0.73, 1.02
> Risk of Alzheimer’s disease: n = 7 prospective cohort studies. Summary RR
= 0.80, 95% CI: 0.65, 0.97.
> Zhang et al., 2015: Increment of 1 serving/week of fish:
> Risk of dementia: n = 4 prospective cohort studies. Summary RR = 0.95,
95% CI: 0.90, 0.99
> Risk of Alzheimer’s disease: n = 5 prospective cohort studies. Summary RR
= 0.93, 95% CI: 0.90, 0.95.
The VKM assessment included 24 primary studies in the evaluation of fish intake
and risk of neurocognitive and psychiatric endpoints in adults. All the studies were
prospective cohort studies, most originating from European countries, but some
also from the United States and from Asian countries. All studies included total fish
as the exposure, while one study also separated into fatty fish. VKM calculated a
summary RR of the three outcomes: risk of dementia, risk of Alzheimer’s disease
and risk of cognitive decline:
> Risk of developing dementia: n = 5 prospective cohort studies, highest versus
lowest intake of total fish. Summary RR = 0.85 (95% CI: 0.75, 0.96) without
significant heterogeneity (Pheterogeneity = 0.37).
> Risk of developing Alzheimer’s disease: n = 4 prospective cohort studies, highest
versus lowest intake of total fish. Summary RR = 0.95 (95% CI: 0.84, 1.08)
without significant heterogeneity (Pheterogeneity = 0.34).
> Risk of cognitive decline: n = 8 prospective cohort studies, highest versus lowest
intake of total fish. Summary RR = 0.81 (95% CI: 0.73, 0.89) with significant
heterogeneity (Pheterogeneity = 0.004).

3.2.7.1.4 Conclusion, weight of evidence

122
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Based on the systematic reviews and primary studies, VKM concluded that there is
evidence that total fish intake reduces the risk of dementia, Alzheimer’s disease and
cognitive decline. This was determined as there was no substantial heterogeneity
between the studies and there was evidence for several plausible mechanisms, and
in view of reported dose–response relationships from one meta-analysis.
In conclusion, the evidence that consumption of total fish reduces the risk of
dementia, Alzheimer’s disease and cognitive decline was graded “probable”.
The number of studies conducted on fatty fish and lean fish was lower than those
considering total fish. Thus, the available evidence on the impact of fatty fish and
lean fish on the risk of dementia, Alzheimer’s disease and cognitive decline was
graded "limited, no conclusion."

3.2.7.1.5 Fish intake and depression in adults

3.2.7.1.5.1 Description of the studies included
The VKM assessment included four systematic reviews with meta-analyses
investigating the association between fish intake and depression and other psychiatric
symptoms in adults. The results of the four systematic reviews varied, though most
of them pointed to a protective effect of higher fish consumption being associated
with lower risk of developing depression:
> Matison et al. 2021: Fish intake. n = 3 prospective cohort studies and risk of
depression in adults >45 years. Summary RR = 1.00, 95% CI: 0.80, 1.26.
> Yang et al., 2018: Highest versus lowest intake of fish. n = 10 prospective cohort
studies in adults. Summary RR = 0.89, 95% CI: 0.80, 0.99.
> Grosso et al., 2016: Highest versus lowest intake of fish. n = 10 prospective
cohort studies in adults. Summary RR = 0.83, 95% CI: 0.70, 0.97.
> Li et al., 2016: Highest versus lowest intake of fish. n = 10 prospective cohort
studies in adults. Summary RR = 0.84, 95% CI: 0.75, 0.94.
The VKM assessment included 13 primary studies in the evaluation of the association
between fish consumption and the risk of depression and other psychiatric symptoms
in adults. Of these, 10 studies assessed the risk of depression, and three assessed
postpartum depression. All 13 primary studies were prospective cohort studies. Four
were conducted in Europe, four in the United States, one in Australia, and four in
Asian countries. All the studies included the association of total fish consumption
with the outcomes, while one study also included fatty fish consumption. Based on
the primary studies, VKM calculated summary RRs for the outcomes of depression
and postpartum depression:
> Risk of developing depression: n = 8 prospective cohort studies. Summary RR =
0.88 (95% CI: 0.79, 0.98) without significant heterogeneity (Pheterogeneity = 0.64).
> Risk of developing postpartum depression: n = 2 prospective cohort studies,
highest versus lowest intake of fish in pregnancy. Summary RR = 0.79 (95% CI:
0.66, 0.95) without significant heterogeneity (Pheterogeneity=0.14).

123
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.7.1.6 Conclusion, weight of evidence

Based on the systematic reviews and primary studies, VKM concluded that there was
some evidence that fish intake can have a protective effect on developing depression
in adults. The direction of the associations was generally consistent towards a
protective effect and the heterogeneity was low. However, the plausible mechanisms
were not fully explained, and the dose–response relationship was not given.
In conclusion, the evidence that consumption of total fish reduces the risk of
depression in adults and post-partum depression in adults was graded “limited,
suggestive”. There were fewer studies of fatty fish and lean fish than studies of total
fish and, as such, the evidence was graded “limited, no conclusion” for the effects
of fatty and lean fish on adult depression and postpartum depression.

3.2.7.2 Summary of the findings on “Neurodevelopment and neurological disorders”

from the primary studies included from the literature search

3.2.7.2.1 Description of the primary studies

Two primary studies, one RCT study by Al-Ghannami et al. (2019) and one
birth cohort study by Mesirow et al. (2017), were included under the category
“Neurodevelopment and neurological disorders”. A description of the studies,
including study name, design, time period, study population, intake of fish
consumption and overall results, can be found in Table 3.12 and in Table 3.13.

3.2.7.2.2 Description of study population

The Al-Ghannami et al. RCT study included n =132 children (mean age, 9.5 years) in
Oman. The Mesirow et al. birth cohort study included 5 727 mother–child pairs from
the Avon Longitudinal Study of Parents and Children study in the United Kingdom
of Great Britain and Northern Ireland, where fish consumption was investigated at
32 weeks gestation and children were followed up between 3 and 13 years.

3.2.7.2.3 Description of fish consumption

In the RCT study, the children were randomized to receive either daily supplements
with omega-3 (DHA) or a daily lunch comprising 100 g of lightly grilled fish
sandwich for 12 weeks. In the birth cohort, fish consumption was measured from
a validated FFQ at gestational week 32.

3.2.7.2.4 Results from the primary studies

The RCT study with fish or omega-3 supplements in children investigated the effects
on several behavioural and cognitive functions using standardized tests. The 12 weeks
of intervention with fish oil versus fish meals demonstrated a better effect of fish oil
supplementation than fish meals on cognitive and behavioural functioning in children.
The birth cohort study showed that a diet low in fish in pregnancy was associated
with early-onset persistent conduct problems, co-occurring difficulties in early
adolescence (parent-reported) and increased emotional difficulties.

124
 TABLE 3.12 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDY (INTERVENTION STUDY) INCLUDED FROM THE LITERATURE SEARCH
ON “NEURODEVELOPMENT AND NEUROLOGICAL DISORDERS”

NUMBER OF MEASUREMENT AND INTERVENTION AND

PARTICIPANTS IN
STUDY TYPE INTAKE CONTROL GROUP
THE STUDY (N)
AUTHOR, YEAR STUDY DURATION OF FISH AND INFORMATION MEASUREMENT OVERALL
AGE (YEARS) OVERALL RESULTS RISK OF BIAS
REGION, COUNTRY AND FOLLOW-UP SEAFOOD REGARDING OF OUTCOME CONCLUSION
AT EXPOSURE
TIME CONSUMPTION INTERVENTION,
ASSESSMENT AT BASELINE DURATION, DOSE
SEX (PERCENT, MEN)
Al-Ghannami et al., Randomized open-label n = 132 (66 in each No information on diet 12-week intervention Main outcomes were The mean (SE) score B 12 weeks of
2019 trial 12 weeks of intervention group) at baseline, but groups period during weekdays behavioural (ADHD symptoms) for the Vanderbilt intervention with
Muscat, Oman intervention with fish Mean (SD) age: 9.5 had similar levels of when school was in and cognitive function. Three Assessment Scales fish oil versus
oil versus fish meals (0.5) years PUFAs (EPA, DHA, DPA) session. tests were used to assess measuring symptoms fish meals
at baseline. Fish meal group: Daily various domains of cognitive of ADHD for the fish- demonstrated
40.9% boys in the fish
Fish oil group: Sum lunch comprising 100 function: verbal ability, oil group was 22.1 (2) a better effect
oil group, 45.5% boys
PUFAs mean (SD) 4.8 g of lightly grilled fish learning and remembering, for the pretest and of fish oil
in the fish meal group
(1.7) sandwich (grouper, and executive functioning: the 23 (2.4) at post-test. supplementation
sea beam, king fish, verbal fluency test to examine For the fish-meal than fish meals
Fish meal group: 4.7
emperor or snapper). lexical ability and initiation group the score was on cognitive
(1.8)
100 g grilled fish was speed of verbal responses; the 25.7 (2.2) at pretest and behavioural
Buschke Selective Reminding and 18 (2.1) at functioning
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

estimated to provide
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

150–200 mg DHA. Test to tap into immediate post-test. The mean in children in
recall (working memory and difference of change Oman.
Fish oil group: capsules
attentional capacity) and a between the groups
containing 403 mg
trail-making test was used was significant (P <
DHA daily during lunch
as part B. 0.001).
break.
A standardized Arabic version Cognitive function:
of the Vanderbilt Assessment Only the trail-making
Scales-Teacher Assessment test for executive
Scale was used to examine functioning showed
behavioural and emotional a difference between
functioning. the groups. Median
(IQR) difference
between pre- and
post-intervention was
61.5 (19.3, 103.2) in
the fish-oil group and
24.5 (-15.2, 74.7) in
the fish-meal group
(P = 0.005).

Notes: SD: standard deviation, PUFA: polyunsaturated fatty acid, EPA: eicosapentaenoic acid, DHA: docosahexaenoic acid, DPA: docosapentaenoic acid, ADHD: attention deficit hyperactivity disorder, SE:
standard error, IQR: interquartile range

125
 TABLE 3.13 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDY (COHORT STUDY) INCLUDED FROM THE LITERATURE SEARCH ON “NEURODEVELOPMENT

126
AND NEUROLOGICAL DISORDERS”

NUMBER OF
REFERENCE AUTHOR, PARTICIPANTS
YEAR FISH AND SEAFOOD
STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
TRIAL OR STUDY NAME INTAKE
AGE (YEARS)
REGION, COUNTRY SEX (PERCENT MEN)
Mesirow et al., 2017. Birth cohort study n = 5 727 mother–child Dietary data were Early-onset persistent Compared to low CP, mothers B Prenatal and postnatal
ALSPAC study pairs. collected from a validated conduct problems (EOP of EOP children consumed less diets low in fish were
Children 3 years old at FFQ. Mothers’ intake at CP) were created using fish (P < 0.01). For EOP, less associated with an EOP
the United Kingdom
baseline. Follow up 4–13 32 weeks gestation and parent-reported SDQ than two servings of fish/week CP trajectory and co-
of Great Britain and
years report of what the mother conduct problem scale. was associated with increased occurring difficulties in
Northern Ireland
fed her child at emotional difficulties. early adolescence.
38 months of age (3
years).

Notes: ALSPAC: Avon Longitudinal Study of Parents and Children, SDQ: Strengths and Difficulties Questionnaire, FFQ: food frequency questionnaire
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.7.3 Final weight of evidence for “Neurodevelopment and neurological disorders”

A final weight of evidence for the theme “Neurodevelopment and neurological
disorders” was based on the 2022 VKM report and the systematic literature search.
An overview of the literature included in the final weight of evidence is given in
Appendix 3, Table A3.35.
In the VKM report, the outcomes “maternal fish intake and neurodevelopment in
children”, “child fish intake and neurodevelopment in children”, “neurocognitive
and psychiatric endpoints in adults” and “depression in adults” were evaluated and
the evidence was graded. No additional systematic reviews were identified and only
two primary studies were identified in addition to the VKM report. The two primary
studies included the outcomes of child mental health (parent-reported persistent
problems) and behavioural and cognitive function in children. When comparing the
evaluation of the weight of evidence in the primary studies with that of the VKM
report, the primary studies did not change this conclusion.
Therefore, in conclusion, the final weight of evidence for the association between
“maternal total fish intake and neurodevelopment in children” and “child total and
fatty fish intake and neurodevelopment in children” is graded “limited, suggestive”.
The evidence for the association between “total fish consumption and neurocognitive
and psychiatric endpoints in adults (dementia, Alzheimer’s disease and cognitive
decline) is graded “probable, suggestive”. The evidence of the association between
“total fish consumption and depression and post-partum depression in adults” is
graded “limited, suggestive”.

3.2.8 MORTALITY
3.2.8.1 Summary of the findings on “Mortality” in the VKM report, Benefit and risk
assessment of fish in the Norwegian diet
3.2.8.1.1 Description of the literature included
The VKM assessment included nine systematic reviews and meta-analyses (three
umbrella reviews and six meta-analyses) on the association between fish intake
and mortality. A significant inverse association between fish intake and all-cause
mortality was concluded from four meta-analyses and one meta-analysis among T2D
patients. An inverse association was also found between fish intake and mortality
from cardiovascular disease (CVD) and coronary heart disease (CHD).
The VKM assessment included 25 primary studies of fish intake and all-cause
mortality, as well as cause-specific mortality, in addition to five primary studies,
including patient-based populations with CVD/CHD/MI, and three primary studies
with diabetes populations.
The primary studies included in the VKM report were from several different countries,
including Australia; China; China, Hong Kong SAR; Denmark; Finland; the Islamic
Republic of Iran; Italy; Japan; the Kingdom of the Netherlands; Norway; Spain; Sweden;
the United Kingdom of Great Britain and Northern Ireland and the United States.

127
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.8.1.2 Weight of evidence

3.2.8.1.2.1 Mortality from Alzheimer’s disease
The VKM assessment included two primary studies on fish intake and mortality
from Alzheimer’s disease. The summary RR (RR = 0.76, 95% CI 0.53, 1.09) was
not statistically significant and the evidence for an association between fish intake
and mortality from Alzheimer’s was graded “limited, no conclusion”.

3.2.8.1.2.2 Mortality from cardiovascular disease

The VKM assessment included 20 primary studies on fish intake and mortality from
CVD. Eighteen studies included participants from the general population while
two were on patients with prior CVD or high risk of CVD from vascular disease,
or having T2D.
The summary RR for total fish and CVD mortality indicated a protective association
for the highest versus lowest intake (RR = 0.92, 95% CI: 0.86, 0.98). For intake of
fried fish (two studies), VKM’s summary RR suggested a small, increased risk of
CVD mortality (RR = 1.03, 95% CI: 0.99, 1.07). For non-fried fish (three studies),
VKM’s summary RR suggested no association with CVD mortality (RR = 0.89, 95%
CI: 0.67, 1.19). For total fish intake in the patient-based populations, VKM’s high–
low summary RR for CVD mortality in patients with a CVD history or at high risk
of CVD (one publication, three studies) suggested a statistically significant, lower
risk (RR = 0.84, 95% CI: 0.77, 0.92) without significant heterogeneity (Pheterogeneity
= 0.66).
Lower CVD mortality for the highest intakes of total fish (18 studies) was indicated
by VKM according to the summary RR for primary studies. In one primary study
with patients with T2D, a statistically significant protective association was observed.
VKM reported evidence for an inverse dose–response relation from two independent
meta-analyses as an upgrading factor.
VKM concluded that the published evidence suggested a protective association
between fish intake and CVD mortality that was statistically significant. The
evidence was graded “probable” for a protective effect of fish consumption on
CVD mortality in the general population.

3.2.8.1.2.3 Mortality from total heart disease

The VKM assessment included two primary studies on mortality from all heart
conditions as the outcome, both from the United States. One study demonstrated a
protective association, while the other study demonstrated no significant association.
VKM concluded that the evidence that fish intake was associated with mortality
from all heart diseases was graded “limited, no conclusion”.

3.2.8.1.2.4 Mortality from coronary heart disease

The VKM assessment included 22 primary studies on mortality from CHD as
the outcome, three studies included CHD mortality in patient-based populations

128
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

– survivors of CHD or MI or populations with T2D. VKM’s high–low summary

RR, based on 18 studies, indicated lower CHD mortality for high overall fish
intakes (RR = 0.91, 95% CI: 0.82, 1.01). The estimate was borderline statistically
significant without significant heterogeneity (Pheterogeneity = 0.16). VKM’s high–low
summary RR for CHD mortality in patients with coronary artery disease was based
on two studies of secondary prevention with few cases and did not suggest an
association with total fish intake (RR = 1.03, 95% CI: 0.58, 1.84), Pheterogeneity = 0.99).
A dose–response meta-analysis was also included where an increase in fish intake
by 20 g/day was associated with a 4 percent reduction in CHD mortality and a
suggested threshold with no further risk reduction above 60 g/day fish intake.
The evidence was graded “probable” for a protective effect of fish intake on CHD
mortality. Two pooled studies with results on fatty and lean fish were included,
where the high–low summary RR (95% CI) for fatty fish was 0.94 (0.81–1.10)
without significant heterogeneity, and 0.95 (0.75–1.21) for lean fish, also without
significant heterogeneity. The evidence was graded “limited, no conclusion” for the
effects of fatty and lean fish on CHD mortality.

3.2.8.1.2.5 Mortality from myocardial infarction

The VKM assessment included five primary studies on mortality from MI. The summary
RR indicated significantly lower MI mortality for the highest versus lowest intakes
(RR = 0.63, 95% CI: 0.46, 0.85). Heterogeneity was significant (Pheterogeneity = 0.01),
but all estimates were consistent in the direction of the association. The evidence
that consumption of total fish reduces MI mortality was graded “probable”.

3.2.8.1.2.6 Mortality from stroke and stroke sub-types

The VKM assessment included 12 studies (prospective, observational in the general
population on the mortality from total stroke/CVD). Among the studies, eight were
based on study populations in Asia and five on populations in the United States,
most included both men and women. The summary RR reported from VKM for
fish overall intake indicates lower mortality of total stroke (RR = 0.86, 95% CI:
0.81, 0.90), without significant heterogeneity (Pheterogeneity = 0.64).
For subtypes of stroke, the haemorrhagic stroke mortality RR, based on six studies,
was statistically significant (RR = 0.86, 95% CI: 0.78, 0.96, Pheterogeneity = 0.64),
indicating lower mortality for the highest overall fish intake. The summary RR
for ischemic stroke was not statistically significant (RR = 0.92, 95% CI: 0.82,1.03,
Pheterogeneity = 0.36).
The evidence that consumption of fish reduces stroke mortality was graded
“probable”. For subtypes of stroke, the evidence was graded “limited, suggestive”
for a protective effect of total fish intake on both ischemic stroke and haemorrhagic
stroke mortality.

129
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.8.1.2.7 Mortality from type 2 diabetes

The VKM assessment included four publications on cause-specific mortality from
T2D (prospective, observational design), all from China or the United States and
including both men and women. Three of the studies were from general population
groups, while one study was in a population with T2D.
The summary estimates from the three studies from the general population were not
statistically significant (summary RR = 0.92, 95% CI: (0.59, 1.43) with borderline
significant heterogeneity (Pheterogeneity = 0.048). The evidence was graded “limited,
no conclusion” for an effect of total fish intake on T2D mortality.

3.2.8.1.2.8 All-cause mortality

The VKM assessment included 23 primary studies (prospective, observational
design) on the association between fish intake and all-cause mortality. A statistically
significant protective association (RR = 0.93, 95% CI: 0.90, 0.97), and significant
heterogeneity (Pheterogeneity < 0.001) was found for overall fish consumption based
on these publications.
Regarding overall fish intake in patients with previous CVD or at high risk of
CVD, the RR suggested a statically significant protective association based on four
prospective studies (RR = 0.83, 95% CI: 0.76, 0.90) without significant heterogeneity
(Pheterogeneity = 0.50). For overall fish intake in subpopulations with T2D, the RR for
all-cause mortality (RR = 0.95, 95% CI: 0.90, 1.01), based on five studies, suggested
a protective association as well.
For intake of fried fish (high–low intake) in relation to all-cause mortality (three
studies), VKM suggested a potentially small increased risk based on the summary
RR (RR = 1.02, 95% CI: 1.00, 1.03, Pheterogeneity = 0.74). For non-fried fish (four
studies), the summary RR by VKM suggested a protective association that was
borderline statistically significant (RR = 0.93, 95% CI: 0.86, 1.00, Pheterogeneity = 0.16).
In conclusion, the evidence was graded “probable” for a protective effect of fish
consumption on all-cause mortality in the general population. The evidence for
an effect of fatty and lean fish on all-cause mortality was graded “limited, no
conclusion”, based on one study.
3.2.8.2 Summary of the findings on “Mortality” from the systematic reviews
included from the literature search
Three systematic reviews (He et al., 2004; Geelen et al., 2007 and Szymanski et
al., 2010) were included under the category “mortality”. A summary of the main
outcome, population, studies included, seafood intake, overall results and conclusion
is provided in Table 3.14.
The systematic review by Szymanski et al. (2010) included a meta-analysis of fish intake
and prostate cancer-specific mortality, including four cohort studies with 49 661 men
and 740 fatal prostate cancers. The overall results demonstrated that high consumption
of fish was associated with a significant reduction in prostate cancer-specific mortality.

130
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

The meta-analyses by Geelen et al. (2007) investigated the association between fish
consumption and colorectal cancer mortality, including four cohort studies ranging
from 3 158 to 265 118 persons (both men and women) in the different cohorts. This
meta-analysis demonstrated no evidence of an association between fish consumption
and colorectal cancer mortality.
He et al. (2004) performed a meta-analysis of 14 cohorts with 222 354 persons (men
and women), investigating the association between fish intake and CHD mortality.
This meta-analysis demonstrated that each 20 g increase in fish intake per day was
associated with a lower risk for CHD mortality, concluding an inverse association
between fish intake and fatal CHD.

131
 TABLE 3.14 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS FROM THE LITERATURE SEARCH ON “MORTALITY”

132
AUTHOR, YEAR FISH AND OVERALL RISK OF BIAS
OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS
STUDY TITLE SEAFOOD INTAKE CONCLUSION (AMSTAR 2)
Szymanski et al., 2010 Prostate cancer and Adult men n=4 Median fish intake from High consumption of fish Total fish intake was Moderate
Fish consumption and prostate cancer-specific 4 cohort studies (n = 49 Cohort studies (n = 4) the four studies: was associated with a associated with 63%
prostate cancer risk: a mortality 661) on cancer-specific - 1.3 times/month significant 63% reduction reduction in prostate
review and meta-analysis mortality - Moderate part in fatal disease (prostate cancer–specific
- 0.5 servings/week cancer-specific mortality) (RR mortality
740 fatal prostate
- 3.25 times/week = 0.37, 95% CI: 0.18, 0.74; P
cancers
= 0.005)
In univariate meta-regression
analysis, larger studies
reported a weaker inverse
association with prostate-
cancer mortality (P = 0.15).
With stratification on the
number of study participants,
studies (15, 37) with > 17
000 participants showed fish
consumption to have a 34%
protective association (RR
= 0.66, 95% CI: 0.43, 1.01),
although the results were
not significant at the 0.05
level. Studies (23, 28) with
< 7 000 participants showed
a significant pooled risk
reduction of 80% (RR = 0.20,
95% CI: 0.09, 0.43).
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
 TABLE 3.14 SUMMARY OF RESULTS FROM SYSTEMATIC REVIEWS FROM THE LITERATURE SEARCH ON “MORTALITY” (cont.)

AUTHOR, YEAR FISH AND OVERALL RISK OF BIAS

OUTCOME POPULATION STUDY INFORMATION OVERALL RESULTS
STUDY TITLE SEAFOOD INTAKE CONCLUSION (AMSTAR 2)
Geelen et al., 2007 Colorectal cancer Not reported in the 4 cohort studies on Exposure definition in the The pooled RR (95% CI) for No evidence of an Moderate
Fish Consumption, mortality systematic review for colorectal cancer four studies included: the highest compared with association between
n-3 Fatty Acids, and mortality. mortality - Fish: daily consumption the lowest fish consumption fish consumption
Colorectal Cancer: Findings of the four - Fish: number of times/ category was 1.02 and colorectal cancer
A Meta-Analysis of primary studies included: month (0.90–1.16). mortality was found.
Prospective Cohort - n = 265 118 (male, n = - Boiled fish
Studies 12 226; female, n = 142 - Fish: number of times/
857) adults aged 40 and week
above were followed up
for 17 years (1966–1982)
and age standardized
mortality rates for cancer
of each site
- n = 17 633 white males
aged 35 and older, 20
years of follow-up, 120
colon cancer and 25
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

rectal cancer deaths

- In Hokkaido by analysing
n = 1 524 men and n =
1 634 women separately,
aged 40 and over
- n = 45 181 men and n
= 62 643 women aged
40–79 years enrolled in
the Japan Collaborative
Cohort Study
He et al., 2004 Coronary heart disease Women and men (>16) 13 cohorts (from 11 Fish consumption RRs (95% CI) for CHD Fish consumption is Moderate
Accumulated evidence (CHD) mortality 222 364 individuals with studies) was standardized and mortality were 0.89 (0.79, inversely associated
on fish consumption and an average 11.8 years of categorized into 5 1.01) for fish intake 1 to 3 with fatal CHD.
coronary heart disease follow-up intervals: “never or 1/ times per month, 0.85 (0.76,
mortality: a meta-analysis month”, “1 to 3/month”, 0.96) for once per week, 0.77
(3 032 CHD deaths)
of cohort studies “1/week”, “2 to 4/week”, (0.66, 0.89) for 2 to 4 times
and “≥5/week.” per week, and 0.62 (0.46,
0.82) for 5 or more times per
week. Each 20-g/day increase
in fish intake was related to a
7% lower risk of CHD mortality
(P for trend = 0.03)

Notes: RR: relative risk, CHD: coronary heart disease, CI: confidence interval

133
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.8.3 Summary of the findings on “Mortality” from the primary studies included from
the literature search

3.2.8.3.1 Descriptions of the primary studies

Five primary studies were included (Walda et al., 2002; Iso et al., 2006; Streppel et
al., 2008; Pertiwi et al., 2021 and Sun et al., 2021). Table 3.15 describes the studies,
including study name, type, number of participants, measurements of seafood
consumption and outcome, overall results, risk of bias and overall conclusion.
All five studies were prospective cohort studies with the geographical distribution
including Finland, Italy, the Kingdom of the Netherlands and the United States.

3.2.8.3.2 Description of study population

The number of study participants varied between the studies, ranging from 1 373
to 366 048. Median follow-up time and the mean age of the participants also varied.
The number of fatal events varied from a low of 62 fatal to a high of 1 877.

3.2.8.3.3 Description of fish consumption

Dietary intake was assessed by FFQs, the cross-check dietary history method or
by 24-hour dietary recall. Fish intake was reported either as frequency or amount.

3.2.8.3.4 Results from the primary studies

3.2.8.3.4.1 Cardiovascular disease mortality
Four of the five primary studies included evaluated the association between seafood
intake and mortality from CVD. Iso et al. (2006) found no association between
sudden cardiac death or fatal CHD and fish intake. Similarly, Sun et al. (2021) found
no association between seafood intake and CVD mortality. However, the study by
Pertiwi et al. (2021), demonstrated a lower risk of CHD mortality, but not with
CVD mortality. The study by Streppel et al. (2008) also found a lower risk for CHD
death with long-term fish consumption, in addition to lower risk for sudden cardiac
death with fatty fish consumption. In agreement with this, Yamagishi et al. (2019)
reported higher mortality from total aortic disease with low/seldom fish intake.
Overall, two studies reported no association between CVD mortality and fish
intake, while three studies reported a lower risk for CVD mortality with fish intake.

3.2.8.3.4.2 Chronic obstructive pulmonary disease mortality

One study (Walda et al., 2002) investigated chronic obstructive pulmonary disease
(COPD) death and found no association between seafood intake and COPD death.

3.2.8.3.4.3 All-cause mortality

Two studies, (Pertiwi et al., 2021 and Sun et al., 2021), found no association between
fish intake and all-cause mortality.

134
 TABLE 3.15 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “MORTALITY”

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT MEN)
Iso et al., 2006 Prospective cohort study n = 41 578 Dietary intake assessed Cardiovascular disease Fish intake was not B Intake of fish was not
The Japan Public Health 1990–1992 to 2001 Fatal coronary events by FFQ at two different registered at main associated with fatal associated with fatal
Center-Based (JPHC) = 62 time points (1990 + hospitals in the region coronary heart disease or coronary heart disease
Maximum 11-year follow-
study Cohort 1 Sudden cardiac death 1995) by medical records sudden cardiac death. HR (CHD) or sudden cardiac
up time
= 37 Fish intake at baseline; reviewed by physicians. (95% CI) for quintile 5 vs 1 death.
Japan
40–59 years lowest quintile: once per For fatal myocardial in multivariable-adjusted The low number of cases,
week (median 23 g/day), infarctions and sudden models: Sudden cardiac respectively 62 and 37,
27 053 men, 27 435
highest quintile: 8 times cardiac deaths: a death: 1.14 (0.36, 3.63), fatal and thus low statistical
women
per week (median 180 systematic search for coronary events 1.08 (0.42, power, may have
g/day) death certificates was 2.76). influenced the results.
performed.
Pertiwi et al., 2021 Prospective cohort study n = 4 067 Dietary intake assessed Information on CVD Total fish consumption B Total fish and oily fish
Alpha Omega Cohort Baseline 2002–2006, Coronary heart disease by a validated 203-item and deaths obtained inversely associated with CHD intake was associated
follow-up to 2018 deaths (CHD) = 515 FFQ. from national mortality (HR (95% CI) = with lower risk of CHD
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

Netherlands (Kingdom
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

(median follow-up time of Cardiovascular disease Total fish intake: median mortality registries 0.73, (0.54, 0.99) for >20–40 mortality, but not with
of the)
12 years) (CVD) deaths = 834 (IQR) 14 (5 to 20 g/day. and the International vs ≤5 g/day). Finding for oily CVD and all-cause
Mean (SD) 69.0 (5.6) Oily fish intake: 5 (1 to Classification of Diseases. fish was similar (HR (95% mortality.
years 79.2% men 11 g/day) CI) 0.72 (0.54, 0.95) for >11
vs <1 g/day. No associations
were observed for CVD or all-
cause mortality.
Streppel et al., 2008. Prospective cohort study. n = 1 373 Dietary intake collected Causes of death were Long-term fish consumption B Long-term fish
The Zutphen Study Baseline 1960 + a new CHD deaths = 348 by the cross-check dietary ascertained by a (cumulative average), consumption lowered the
cohort included in 1985. Sudden coronary deaths history method, conducted clinical epidemiologist average 22 g/day had a risk of CHD death. Fatty
Netherlands (Kingdom
Follow-up until 2000. = 66 by dieticians (assessed and coded according 22% lower CHD death risk, fish consumption lowered
of the)
every 5th year). to the Eighth revision recent fish consumption was the risk of sudden cardiac
Mean (SD) age (1960): 49
Total fish intake: Mean of the International not associated with CHD death.
(6), 1985: 71 (5).
range from 16 to 21 g/ Classification of Disease. death. Fatty fish intake (no
100% men vs yes) was associated with
day. Lean and fatty fish
also included. decreased risk of sudden
coronary death (multivariate
model: HR = 0.46, 95% CI:
0.27, 0.78).
Long-term fatty fish
consumption, average 7 g/
day, 54% lower sudden
cardiac death risk.

135
 TABLE 3.15 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) INCLUDED FROM THE LITERATURE SEARCH ON “MORTALITY” (cont.)

136
NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT MEN)
Sun et al., 2021 Prospective cohort study n = 17 295 Dietary intake assessed Death status determined No associations between B No associations between
The National Health and Included 2003 to 2012. All-cause mortality = by two 24-hour dietary using the NHANES Public an increase in seafood seafood consumption and
Nutrition Examination 1 076 recall. Use Linked Mortality File, consumption of 1 oz- deaths.
Follow-up until 31
Survey (NHANES) CVD deaths = 181 HR for an increase in based on the results of equivalent per day and
December 2015.
Mean (SD) age 45.9 seafood consumption of a probabilistic match all-cause (HR = 0.84, 95% CI:
the United States
(17.1) years. 1 oz (28 g) equivalent per between NHANES and the 0.66, 1.07) and CVD-related
day increase National Death Index. mortality (HR = 0.89, 95% CI:
46.7% men
0.54, 1.47).
Walda et al., 2002 Prospective cohort study. n = 2 917 Dietary intake assessed Information on cause No associations between fish B No associations between
A study from three Baseline 1970 Chronic obstructive by cross-check dietary of death determined intake and COPD mortality (RR seafood consumption and
European countries. Follow-up 20 years (1990) pulmonary disease history method. by two investigators (95% CI) for highest vs lowest COPD death.
(COPD) death = 73 Mean fish intake, Finland who reviewed clinical tertile 1.02 (0.59–1.78).
Finland, Italy and
Age 50–69 years 40 (47) g/day, Italy 20 records from family
Netherlands (Kingdom
100% men (21) g/day, Netherlands doctors, specialists
of the) (two Finnish,
(Kingdom of the) 17 (19) and relatives. Primary
two Italian and one
g/day mortality coded according
Netherlands cohort of the
to the International
Seven Countries Study are
Classification of Diseases
involved)
(ICD) of the WHO.

Notes: CHD: coronary heart disease, FFQ: food frequency questionnaire, IQR: interquartile range, HR: hazard ratio, CI: confidence interval, SD: standard deviation, NHANES: The National Health and Nutrition
Examination Survey (NHANES), WHO: World Health Organization, COPD: chronic obstructive pulmonary disease
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.8.4 Final weight of evidence for “Mortality”

A final weight of evidence for the theme “Mortality” was based on the 2022 VKM
report and the systematic literature search. An overview of the literature included
in the final weight of evidence is given in Appendix 3, Table A3.35.
The weight of evidence for all-cause mortality and disease-specific mortality were
based on the VKM report and on primary studies and systematic reviews included in
the evaluation. In conclusion, the overall evidence of total fish intake associated with
mortality from Alzheimer’s disease, total heart disease, T2D, colorectal cancer and
prostate cancer were all graded “limited, no conclusion”. In addition, the evidence
of fatty fish and lean fish intake associated with all-cause mortality and mortality
from CHD was also graded “limited, no conclusion”. Furthermore, the evidence
from total fish intake associated with ischemic stroke and haemorrhagic stroke were
graded “limited, suggestive” for a protective effect. Lastly, the association of total
fish intake with all-cause mortality and mortality from CVD, CHD, MI and stroke
were graded “probable” for a protective effect.

3.2.9 OVERWEIGHT AND OBESITY IN ADULTS

The literature included in the theme “Overweight and obesity” includes results from
the 2022 VKM report and from three original primary studies originating from the
literature search. No systematic reviews were included.

3.2.9.1 Summary of the findings on “Overweight and obesity” in the VKM report,
Benefit and risk assessment of fish in the Norwegian diet
VKM conducted a comprehensive literature search and performed an analysis of
a systematic review and three primary prospective cohort studies to investigate
the relationship between fish intake and body weight in adults. The studies were
conducted across diverse geographic regions, encompassing Asia, Europe and the
United States.
The systematic review contained three studies (including two prospective studies
on abdominal obesity in adults) that showed that higher total fish intake was related
to reduced abdominal obesity. However, one study on the risk of developing
overweight/obesity showed no association with fish intake.
As there were few studies and due to heterogenous presentation of results, no
summary RR was calculated on the basis of the primary studies. One study,
which included a large number of participants, reported no association between
fish consumption and abdominal obesity, while another study found a protective
association.
In conclusion, from the VKM report, the association between fish intake and adult
body weight was graded “limited, no conclusion”, considering that there were
few studies, reporting different endpoints, and that the results showed weak or no
associations between fish consumption and weight gain (general or abdominal obesity).

137
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

3.2.9.2 Summary of the findings on “Overweight and obesity” in primary studies

included from the literature search
3.2.9.2.1 Description of the primary studies
Three primary studies on the association between fish and seafood intake and
overweight and obesity (Smith et al., 2015; Tørris et al., 2017 and Beulen et al., 2018)
were identified and met the inclusion criteria. A description of the three studies,
including study name, design, time period, population, intake of fish consumption
and overall results, is provided in Table 3.16.
Two of the studies are prospective cohort studies conducted in Europe (Spain) and the
United States. The third study is an epidemiological, population-based study comprising
several cross-sectional surveys, primarily involving participants from Norway.

3.2.9.2.2 Description of study population

All three studies included healthy participants from the general adult population.
Two studies, Beulen et al. (2018) and Tørris et al. (2017), included both men and
women, while one study, Smith et al. (2015), had separate data sets including only
women (n= 2 datasets) and men (n = 1 dataset).

3.2.9.2.3 Description of fish consumption

All three studies used validated FFQs for estimating fish and food consumption.
Beulen et al. assessed the daily substitution of one portion of red meat with
oily fish or white fish and how this affects weight gain. Smith et al. assessed the
intake of proteins from seafood in relation to other protein foods and long-term
weight changes. Tørris et al. assessed the intake of lean fish or fatty fish and waist
circumference in both men and women.

3.2.9.2.4 Results from the primary studies

Beulen et al. examined the daily substitution of one portion of red meat with oily
fish and white fish. The effects on body weight were estimated using generalized
equations. The results showed weight changes of up to -0.75 kg (95% CI: -1.13,
-0.38) and -0.87 kg (95% CI: -1.17, -0.56) for oily fish and white fish, respectively.
In Smith et al., dose-response analyses were performed using generalized models.
It was found that increasing seafood consumption by one serving per day was
associated with a weight loss of 0.45 kg (95% CI: -0.83, -0.09 kg) when glycaemic load
was simultaneously increased. However, when glycaemic load was simultaneously
decreased, weight loss increased to 1 kg (95% CI: -1.24, -0.76 kg).
Tørris et al. investigated the association between lean and fatty fish consumption
and waist circumference. The study found that consuming lean fish once a week
or more was significantly associated with decreased waist circumference (-1.15,
95% CI: -1.96 to -0.35). On the other hand, consuming fatty fish was significantly
associated with increased waist circumference for both genders (women: 0.97, 95%
CI: 0.29 to 1.65; men: 0.6, 95% CI: 0.01 to 1.18). Lean fish consumption in particular
seemed to be associated with reducing waist circumference.

138
 TABLE 3.16 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) FROM THE LITERATURE SEARCH ON “OVERWEIGHT AND OBESITY”

NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT MEN)
Beulen et al., 2018 2003–2009 n = 6 942 Semiquantitative 137- Anthropometric Daily substitution of one B Reductions in red meat
Prevención con Dieta Prospective cohort study Mean age: 67 years, 47% item FFQ measurements per year. portion of red meat with white consumption coupled with
Mediterránea (PREDIMED) obesity at baseline. Continuous outcome: body meat, oily fish or white fish respective increases in
Median follow-up time:
(Prevention through the weight showed weight changes up to: white meat or fish would
4.8 years Control group: advice on
Mediterranean diet) -0.64 kg (95% CI: -0.94, lead to less weight gain.
following a low-fat diet. Dichotomized outcomes:
-0.35),
Spain Non-control group: cut-off of body weight (a
-0.75 kg (95% CI: -1.13,
Mediterranean diet change ≥10%), incidence
-0.38) and -0.87 kg (95% CI:
supplemented with either (increasing to a BMI ≥30
-1.17, -0.56), respectively.
extra-virgin olive oil or kg/m2) and reversion of
mixed nuts. obesity (decreasing to a
BMI <30 kg/m2).
Smith et al., 2015 Prospective cohort study In total, n = 120 784 Validated food-frequency -Protein foods and Negative association between B Seafood intake was
3 prospective US cohorts: Follow-up: 4 years, 16 healthy participants, questionnaires. glycaemic load every changes in protein seafood negatively associated
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

including 46 994 in the 4 years using food and long-term weight change: with long-term weight
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

Nurses’ Health Study years and 24 years. Mean (SD) seafood intake
(NHS), Nurses’ Health NHS, 47 928 in the NHS II, at baseline (servings/ frequency questionnaires. Mean (95% CI) (kg) gain.
Study II (NHS II), and and 25 862 in the HPFS. day*, See Smith et al., -Weight change every NHS: -0.77(-0.88, -0.66)
Health Professionals Baseline: 2015, Supplemental 4 years (p<0.0001)
Follow-Up Study (HPFS) NHS: 1 976, female, Table 2): NHS II: -0.78(-0.93, -0.64)
age (mean+SD): 48.9 NHS: 0.34 (0.13) (p<0.0001)
the United States
±2.7 years HPFS: -0.54(-0.67, -0.41)
NHS II: 0.27 (0.2)
Weight: 64.0 ±4.1 kg, (p<0.0001)
HPFS: 0.37 (0.15) Pooled: -0.70(-0.85, -0.54)
BMI: 23.7 ±1.4 kg/m2
NHS II:1 989, female, age Mean 4-year change: (p<0.0001)
(mean ±SD): 37.7 ±3.2 mean (95% CI) (Smith et al., 2015,
years NHS: -0.02 (-0.2, 0.14) Supplemental Table 4).
Weight: 62.6 ±7.7 kg, - A 1-serving/day increase in
NHS II: -0.01 (-0.11, 0.09)
BMI: 23.0 ±2.4 kg/m2 seafood was associated with
HPFS: -0.01 (-0.17, 0.14) 0.45 kg (CI: -0.83, -0.09 kg)
HPFS: 1 986, male, age
(mean ±SD): 47.3 ±2.7 weight loss when glycaemic
years load was simultaneously
increased, yet 1kg (CI:
Weight: 79.4 ±4.5 kg,
-1.24, -0.76 kg) weight loss
BMI: 24.8 ±1.1 kg/m2
when glycaemic load was
Weight gain during
simultaneously decreased.
follow-up:
NHS: 1.1 kg/4 years
NHS II:2.1/4 years
HPFS: 0.7/4 years

139
 TABLE 3.16 SUMMARY OF RESULTS FROM ORIGINAL PRIMARY STUDIES (COHORT STUDIES) FROM THE LITERATURE SEARCH ON “OVERWEIGHT AND OBESITY” (cont.)

140
NUMBER OF
AUTHOR, YEAR PARTICIPANTS FISH AND
STUDY TITLE STUDY INFORMATION IN THE STUDY (N) OUTCOME OVERALL RESULTS RISK OF BIAS OVERALL CONCLUSION
SEAFOOD INTAKE
REGION, COUNTRY AGE (YEARS)
SEX (PERCENT MEN)
Tørris et al., 2017 Prospective cohort study Tromsø Study 4 Fish consumption -Physical variables: waist Lean fish consumption once a B Fatty and lean fish
Data from the Norwegian 13-year follow-up period. (1994–1995): n = 23 907, assessed by a validated circumference (WC) and week or more was significantly consumption likely
Tromsø Study 26–69 years. centre-specific blood pressure (BP). associated with decreased influence MetS differently.
Tromsø Study 4:
Tromsø Study 6 questionnaire. -Non-fasting blood future MetS (women: -0.05, Lean fish consumption
Tromsø, Norway 1994–1995
(2007–2008): n = 12 981, The nutrients were samples: triglycerides 95% CI: -0.09 to -0.01, men: seems to be associated
Tromsø Study 6:
30–87 years. computed based on (TG), HDL-cholesterol -0.1, 95% CI: -0.15 to -0.05), with beneficial changes in
2007–2008
Baseline characteristics the food frequency (HDL-C), and blood decreased TG (women: -0.04, the MetS components.
of the participants questionnaire (FFQ). glucose (BG). 95% CI: -0.08 to -0.00, men:
(1994–1995): Almost 80% of the Metabolic score (MetS) -0.11, 95% CI: -0.17 to
n = 23 907 participants reported ranging from 0 to 5 -0.06), and increased HDL-
Age (mean ±SD): 44.1 lean fish consumption at (abdominal obesity, cholesterol (women: 0.03,
±11.5 years, 48% men dinner once or more per increased TG, decreased 95% CI: 0.01 to 0.05, men:
week, while 64% reported HDL-C, hypertension and 0.04, 95% CI: 0.02 to 0.05),
BMI:25.1 ±3.8 kg/m2
consuming processed hyperglycaemia) was whereas decreased WC (-1.15,
38% reported as daily
fish and 37% reported performed. 95% CI: -1.96 to -0.35) and
smokers.
consuming fatty fish. BP (SBP: -0.86, 95% CI:
-1.66 to -0.06; DBP: -0.63,
95% CI: -1.18 to -0.07) was
identified only for men (age-
adjusted models).
Fatty fish consumption was
significantly associated with
increased WC for both genders
(women: 0.97, 95% CI: 0.29 to
1.65, men: 0.6, 95% CI: 0.01
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

to 1.18) and increased HDL-C

levels in men (0.02, 95% CI:
0.00 to 0.03).

Notes: BMI: body mass index, CI: confidence interval, SD: standard deviation
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

3.2.9.3 Final weight of evidence for “Overweight and obesity”

A final weight of evidence for the theme “Overweight and obesity” was based on the
2022 VKM report and the systematic literature search. An overview of the literature
included in the final weight of evidence is given in Appendix 3, Table A3.35.
Three more primary studies that complemented the VKM 2022 assessment were
discovered through the literature search; however, each study reported a different
outcome, such as substitution, long-term weight gain and waist circumference.
The present body of evidence regarding a link between fish consumption and adult
body weight was rated "limited, no conclusion" by VKM due to a limited number
of papers and inconsistent methods of measuring the outcome. The three additional
studies were not sufficient to change VKM’s conclusion, and, as such, the association
between fish intake and obesity is rated "limited, no conclusion".

3.3 FINAL WEIGHT OF EVIDENCE FOR “HEALTH BENEFITS OF FISH

CONSUMPTION”
A summary of the final weight of evidence for the different outcomes included in
the chapter “Evidence of health benefits of fish consumption” is given in Table 3.17.
More information of the data for the weight of evidence can be found in each
specific outcome section in Section 3.2. The weight of evidence was based on a
summarization of the available evidence from previously published risk-benefit
assessment reports (WCRF, 2018b and VKM, 2022) and the systematic literature
search, including systematic reviews and original primary studies. An overview of
the literature included in the final weight of evidence for each health outcome is
given in Appendix 3, Table A3.35.

TABLE 3.17 SUMMARY OF FINAL WEIGHT OF EVIDENCE FOR “EVIDENCE OF HEALTH BENEFITS
OF FISH CONSUMPTION”

CONCLUSION
HEALTH OUTCOME FISH INTAKE “WEIGHT OF EVIDENCE”1
ALLERGY AND IMMUNOLOGY
Allergic rhinitis in children Maternal total fish intake in pregnancy Limited, no conclusion
Early fish introduction Limited, no conclusion
Allergic sensitization in children Maternal total fish intake in pregnancy Limited, no conclusion
Child total fish intake Limited, no conclusion
Asthma in children Maternal total fish intake in pregnancy Limited, no conclusion
Maternal fatty fish intake in pregnancy Limited, no conclusion
Maternal lean fish intake in pregnancy Limited, no conclusion
Eczema in children Maternal total fish intake in pregnancy Limited, no conclusion
Child total fish intake Limited, suggestive (protective for intake
in the first year of life, but not later)
Multiple sclerosis Total fish intake Limited, suggestive (protective)
Rheumatoid arthritis Total fish intake Limited, no conclusion

141
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 3.17 SUMMARY OF FINAL WEIGHT OF EVIDENCE FOR “EVIDENCE OF HEALTH BENEFITS
OF FISH CONSUMPTION” (cont.)
CONCLUSION
HEALTH OUTCOME FISH INTAKE “WEIGHT OF EVIDENCE”1
BIRTH AND GROWTH OUTCOMES
Preterm birth Maternal total fish intake in pregnancy Probable (protective effect)
Maternal fatty and lean fish intake in Limited, no conclusion
pregnancy
Maternal lean fish intake in pregnancy Limited, no conclusion
Small for gestational age Maternal total fish intake in pregnancy Limited, suggestive (protective)
Maternal fatty fish intake in pregnancy Limited, no conclusion
Maternal lean fish intake in pregnancy Limited, no conclusion
Birth weight Maternal total fish intake in pregnancy Limited, suggestive (protective)
Maternal fatty fish intake in pregnancy Limited, suggestive (protective)
Maternal lean fish intake in pregnancy Limited, suggestive (protective)
Low birth weight Maternal total fish intake in pregnancy Probable (protective effect)
Maternal fatty fish intake in pregnancy Limited, no conclusion
Maternal lean fish intake in pregnancy Limited, no conclusion
High birth weight Maternal total fish intake in pregnancy Limited, suggestive (increased risk)
Maternal fatty fish intake in pregnancy Limited, suggestive (increased risk)
Maternal lean fish intake in pregnancy Limited, suggestive (increased risk)
Birth length Maternal total fish intake in pregnancy Limited, no conclusion
Maternal fatty fish intake in pregnancy Limited, no conclusion
Maternal lean fish intake in pregnancy Limited, no conclusion
Head circumference Maternal total fish intake in pregnancy Limited, no conclusion
Maternal fatty fish intake in pregnancy Limited, no conclusion
Maternal lean fish intake in pregnancy Limited, no conclusion
BONE HEALTH
Hip fracture Total fish intake Limited, suggestive (protective)
CANCER
Liver cancer Total fish intake Limited, suggestive (protective)
Liver cancer Total fish intake Limited, suggestive (protective)
Colorectal cancer Total fish intake Limited, suggestive (protective)
Nasopharyngeal cancer Cantonese-style salted fish2
Strong evidence (increased risk)
Pancreatic cancer Total fish intake Limited, no conclusion
Breast cancer Total fish intake Limited, no conclusion
CARDIOVASCULAR DISEASES
Total cardiovascular disease Total fish intake Limited, suggestive (protective effect)
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, no conclusion
Coronary heart disease Total fish intake Probable (protective effect)
Fatty fish intake Limited, suggestive (protective effect)
Lean fish intake Limited, suggestive (no effect)
Myocardial infarction Total fish intake Limited, suggestive (protective effect)
Fatty fish intake Limited, suggestive (protective effect)
Lean fish intake Limited, suggestive (no effect)

142
C H A P TE R 3 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”

TABLE 3.17 SUMMARY OF FINAL WEIGHT OF EVIDENCE FOR “EVIDENCE OF HEALTH BENEFITS
OF FISH CONSUMPTION” (cont.)
CONCLUSION
HEALTH OUTCOME FISH INTAKE “WEIGHT OF EVIDENCE”1
CARDIOVASCULAR DISEASES
Total stroke Total fish intake Probably (protective effect)
Fatty fish intake Limited, suggestive (protective effect)
Lean fish intake Limited, suggestive (protective effect)
Ischemic stroke Total fish intake Limited, suggestive (protective effect)
Haemorrhagic stroke Total fish intake Limited, suggestive (protective effect)
Atrial fibrillation Total fish intake Limited, suggestive (adverse effect)
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, suggestive (protective effect)
Heart failure Total fish intake Limited, suggestive (protective effect)
Venous thromboembolism Total fish intake Limited, no conclusion (protective effect)
Peripheral arterial disease Total fish intake Limited, no conclusion
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, no conclusion
TYPE 2 DIABETES
Type 2 diabetes Total fish intake Limited, no conclusion
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, suggestive (no association)
NEURODEVELOPMENT AND NEUROLOGICAL DISEASES
Neurodevelopment in children Maternal total fish intake in pregnancy Limited, suggestive (protective)
Maternal fatty fish intake in pregnancy Limited, no conclusion
Maternal lean fish intake in pregnancy Limited, no conclusion
Child total fish intake Limited, suggestive (protective)
Child fatty fish intake Limited, suggestive (protective)
Child lean fish intake Limited, no conclusion
Neurocognitive and psychiatric endpoints Total fish intake Probable (protective effect)
in adults (dementia, Alzheimer’s disease Fatty fish intake Limited, no conclusion
and cognitive decline)
Lean fish intake Limited, no conclusion
Depression and post-partum depression Total fish intake Limited, suggestive
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, no conclusion
MORTALITY
Alzheimer’s disease mortality Total fish intake Limited, no conclusion
Cardiovascular disease (CVD) mortality Total fish intake Probable (protective)
Total heart disease mortality Total fish intake Limited, no conclusion
Coronary heart disease (CHD) mortality Total fish intake Probable (protective)
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, no conclusion
Myocardial infarction (MI) mortality Total fish intake Probable (protective)
Stroke mortality Total fish intake Probable (protective)
Stroke subtypes: ischemic stroke- and Total fish intake Limited, suggestive (protective)
haemorrhagic stroke mortality
Type 2 diabetes mortality Total fish intake Limited, no conclusion

143
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 3.17 SUMMARY OF FINAL WEIGHT OF EVIDENCE FOR “EVIDENCE OF HEALTH BENEFITS
OF FISH CONSUMPTION” (cont.)
CONCLUSION
HEALTH OUTCOME FISH INTAKE “WEIGHT OF EVIDENCE”1
MORTALITY
Colorectal cancer mortality Total fish intake Limited, no conclusion
Prostate cancer-specific mortality Total fish intake Limited, no conclusion
All-cause mortality Total fish intake Probable (protective)
Fatty fish intake Limited, no conclusion
Lean fish intake Limited, no conclusion
OBESITY
Obesity in adults Total fish intake Limited, no conclusion

Notes: 1 Final weight of evidence is based on the World Cancer Research Fund grading system (WCRF, 2018 and WCRF,
2018a). 2 Cantonese-style salted fish is part of the traditional diet of people living in the Pearl River Delta region in
southern China. It is prepared with less salt than is used in northern China and allowed to ferment.

144
145
 146
© FAO/Harsha Vadlamani
CHAPTER 4
RESULTS AND
SUMMARIZATION OF THE
LITERATURE REVIEW
“TOXIC EFFECTS OF
DIOXINS AND dl-PCBs”

4.1 LITERATURE SEARCH AND QUALITY ASSESSMENT

Literature searches for the systematic review on the “Toxic effects of dioxins
and dl-PCBs” were performed in the databases PubMed and Web of Science.
A flow diagram of the results from the literature searches is given in Figure 4.1.
The literature searches in Web of Science and PubMed resulted in 2 770 records.
Of these, 372 duplicate records were identified and removed, leaving 2 398 records,
which were screened by title and abstract using the online screening tool, Rayyan.
As a result of the screening, 264 duplicates were removed, and 1 736 records were
excluded based on inclusion and exclusion criteria. Thus, 396 records (33 systematic
reviews and 363 primary articles) were assessed in full text.

4.1.1 SYSTEMATIC REVIEWS

In the current literature search, the EFSA Dioxin Report from 2018 was included, as
it was considered sufficiently comprehensive for the evaluation of the toxic effects
of dioxins. To avoid reporting duplicate publications, only studies published from
5 July 2016 onwards were included further. (All the remaining systematic reviews
identified in literature search were excluded for further assessment, either because
they were not relevant according to the criteria or they consisted of studies published
before 2016 [see Appendix 4, Table A4.3]). The EFSA report was quality assessed
with the risk-of-bias tool AMSTAR 2 (Appendix 4, Table A4.6) and graded “high”
according to the overall confidence in the results.

147
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

4.1.2 PRIMARY STUDIES

A total of 363 primary studies were assessed in full text after title and abstract
screening. Of these, 271 studies were excluded based on inclusion and exclusion
criteria during the full-text assessment (Appendix 4, Table A4.4) and 12 were
excluded as they were already assessed in the EFSA 2018 systematic review
(Appendix 4, Table A4.5). Thus, 80 primary studies were quality assessed with
the OHAT risk-of-bias tool (Appendix 4 Table A4.7). Based on the risk of bias
assessment, 20 primary studies were graded Tier 1, 51 primary studies were graded
Tier 2, and 9 studies were graded Tier 3. Only studies graded Tier 1 were included
for further assessment in the final review.

4.2 RESULTS AND SUMMARIZATION OF THE LITERATURE INCLUDED

The 2018 EFSA report and the primary studies included from the literature search
(rated Tier 1) are summarized in the following sections with regard to the relevant
health outcomes.

4.2.1 CHLORACNE AND OTHER DERMAL EFFECTS

The most unequivocal effect of dioxins is chloracne. Chloracne is a cystic and
hyperkeratotic skin disorder caused by high exposure to TCDD. EFSA considers
chloracne the most reliable and specific indicator of acute TCDD toxicity; however,
its occurrence has only been shown following accidental, deliberate or occupational
high-dose exposure. EFSA therefore deemed it of little pertinence for risk assessment
of background exposure. No further studies were found reporting on chloracne in
the literature search.

4.2.2 REPRODUCTIVE EFFECTS (INCLUDING ORGANS)

In addition to the EFSA report, nine primary studies were found in the literature
search investigating the outcome of reproductive effects. A summarization of these
studies is given in Table 4.1.
As mentioned in the EFSA report, changes in sex hormones were not considered
to be an adverse health effect. However, as they can contribute to a mechanistic
explanation for the effect on reproduction, they were still reported. We found
eight additional studies in our literature search investigating changes in different
sex hormones following dioxin and dl-PCB exposure. These outcomes were not
considered adverse health effects, but the studies are summarized in Table 4.1. In
short, the studies showed conflicting results, and no conclusion can be made on the
effect on sex hormone regulation following dioxin and dl-PCB exposure.

148
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

FIGURE 4.1. FLOW DIAGRAM FOR THE LITERATURE REVIEW “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

IDENTIFICATION OF STUDIES IDENTIFICATION OF STUDIES

VIA DATABASES VIA OTHER METHODS

Records identified from Records removed before Records identified from:

IDENTIFICATION

database searching (n = 2 770): screening: Reports

Web of Science (n = 1 577) Duplicate records removed (n = 1): EFSA, 2018
PubMed (n = 1 193) (n = 372)

Records screened by title Records excluded by (n = 2 000):

and abstract (n = 2 398) Duplicates (n = 264)
Criteria (n = 1 736)
SCREENING

Full-text articles assessed Records excluded by criteria

for eligibility (n = 396): (n = 315):
 Systematic reviews (n = 33)  Systematic reviews (n = 32)
 Primary studies (n = 363)  Primary studies (n = 283)

Quality assessment (risk of bias) Records excluded after quality

of full-text articles (n = 81) assessment (risk of bias)
 Systematic review (n = 1) Tier 2 primary studies (n = 51)
INCLUDED

 Primary studies (n = 80) Tier 3 primary studies (n = 9)

Systematic review included in final review (n = 1) (EFSA, 2018)

Tier 1 rated primary studies included in final review (n = 20)

Source: Prepared by the authors based on: Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow,
C.D., Shamseer, L. et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ,
372: n71. https://doi.org/10.1136/bmj.n71.

4.2.2.1 Effects on semen quality

The most sensitive dioxin-induced health endpoint concerning semen quality was
reduced sperm concentration, although total count, motility, morphology, volume
and viability have also been assessed. In their evaluation of epidemiological studies,
EFSA found seven studies on male reproduction as an outcome. Three of these were
rated Tier 1, while four were rated Tier 2. One additional study was found in the
literature search, which was rated Tier 1. The EFSA considered associations between
exposure to TCDD during infancy/prepuberty and impaired semen quality to be
causal. This was based on the weight of evidence from epidemiological observational
studies and from corroborating experimental animal studies. The EFSA report

149
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

weighted particularly two studies from the Seveso cohort (Mocarelli et al., 2008
and Mocarelli et al., 2011), and one from the Russian children’s study (Mínguez-
Alarcón et al., 2017). These studies showed a particularly sensitive period of effect
from infancy to prepuberty, and the most pronounced effect was reduced sperm
concentration after exposure to dioxins.
The EFSA based the TWI for dioxin and dl-PCB on the no-observed-adverse-effect
serum level for PCDD/F (measured in WHO toxic equivalent quotients, or WHO-
TEQ) of 7.0 pg WHO-TEQ/g fat from the Russian children’s study, in which
reduced semen concentration was the main health outcome (Mínguez-Alarcón et
al., 2017). It should be noted that the Russian children’s study did not show any
association between sperm concentration and total TEQ of dioxins and dl-PCBs
up to a quartile level of 47.8 pg WHO-TEQ/g fat, nor any associations of dl-PCB
with decreased sperm concentration.
Complementing the search from EFSA, one more study was found in the literature
search on the relationship between dl-PCB and semen quality (Paul et al., 2017).
This was a case-control study, where a group of men with low sperm quality (cases;
n = 24) was compared to a group of men with normal sperm quality (controls; n =
26). In this study, individuals with low sperm quality exhibited significantly higher
levels of non-ortho PCBs (949.49 ± 624.97 pg/g lipid; p = 0.020) and total dl-PCBs
(7029.96 ± 3023.97 pg/g lipid; p = 0.028; 22.52 ± 21.2 pg WHO-TEQ/g lipid) than
the control group (508.40 ± 324.44 pg/g lipid and 4805.92 ± 2205.02 pg/g lipid 14.00
± 10.82 pg WHO-TEQ/g lipid, respectively). However, following a multivariate
regression, only semen volume was found to be significantly affected by sum dl-
PCB.
In conclusion, the additional study does not conflict with the EFSA conclusion on
causality between PCDD/Fs and reduced sperm quality since only dl-PCBs were
measured in the additional study. No clear association between total dl-PCBs and
sperm quantity was found, which is in accordance with the observational studies
presented by EFSA.

4.2.2.2 Cryptorchidism
Cryptorchidism is the failure of the testicles to descend to the bottom of the scrotum
during development. EFSA found two nested case-control studies dealing with this
endpoint, in which one of these found no association between placenta levels of
dioxins and dl-PCBs and cryptorchidism, and the other study did find associations
to sum PCDD/F levels in subcutaneous adipose tissue biopsies, but only in the
adjusted analysis. However, due to weight of evidence, EFSA concluded that these
two studies did not provide sufficient evidence for an effect of dioxins and dl-PCBs
on cryptorchidism. No additional studies were found in the literature search to
further assess this health outcome.

150
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

4.2.2.3 Pubertal development

EFSA found seven studies on pubertal development, three rated Tier 1 and four rated
Tier 2. Pubertal development was defined using the Tanner staging system (Marshall
et al., 1970), which includes: genital development, pubic hair growth, auxiliary hair
growth, testicular volume, and age at first ejaculation. EFSA determined that the
three Tier 2-rated studies contained too low sample size for certain conclusions. The
three Tier 1-rated studies were all from one cohort (the Russian children’s study),
and these reported a dose-related association between serum TCDD and delayed
puberty onset. However, these studies also showed high correlation between total
TEQ and organochlorine pesticides, which also can impact the timing of puberty
onset. Taken together, EFSA deemed the interpretation of a causal relationship
between pubertal development and total TEQ difficult, and therefore concluded
that there was insufficient information to conclude on the effects of dioxins and dl-
PCBs on pubertal development. No further studies on pubertal development were
found in the literature search.

151
 TABLE 4.1 OVERVIEW OF PRIMARY STUDIES WITH REPRODUCTIVE OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS

152
OF DIOXINS AND dl-PCBs”

AUTHOR, YEAR STUDY DIOXIN AND

STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Boda et al., 2017 Cross-sectional Reproduction: Oestradiol Pregnant women (n 17 OCDD and PCDF In generalized linear models (β (95% CI): Reduction in testosterone levels of girls and boys:
Viet Nam study and testosterone = 210) enrolled and measured in breast milk > Negative association between 1,2,3,6,7,8-HxCDD and testosterone concentrations in girls
followed-up with their (lipid adjusted). (β= -0.320 (-0.538, -0.101).
newborns (n = 162) Mean (SD) TCDD: 2.2 (2.1) > Reduction in testosterone levels in umbilical cord blood for boys with 2,3,7,8-TCDD ≥ 5.5 pg/g
pg/g TEQ lipid compared to levels <1 pg/g
Mean (SD) PCDD/F: 8.8 > Reduction in testosterone levels in umbilical cord blood for girls with 2,3,7,8-TCDD 1-3 pg/g
(1.6) pg/g TEQ lipid compared to levels <1 pg/g
> Reduction in testosterone levels in umbilical cord blood for girls with TEQ-PCDD/Fs between 9
and 12 pg/g lipid compared to levels < 1 pg/g lipid.
In conclusion, increased exposure of dioxins in breast milk was associated with decreased
concentrations of testosterone in cord blood of newborns.
Lambertino et al., Cross-sectional Reproduction: LH and n = 87 post- All 29 congeners measured In Pearson’s correlation coefficients:
2020 study FSH hormones menopausal women in blood (lipid adjusted). > Sum TEQs negatively associated with LH concentrations: r = -0.26
the United States (NHANES) Mean (95% CI) Ln sum TEQ > Mono-ortho PCBs negatively associated with LH concentrations: r = -0.22
0.11 (0.09, 0.13) pg/g > A doubling of sum TEQ was associated with a decrease in LH of 11.9% (CI: -21.3, -21.4%)
(P = 0.03).
Luong et al., 2018 Cross-sectional Reproduction: n = 42 men (mean 17 PCDD/Fs and 4 non- Significant findings in the study:
Viet Nam study Reproductive hormones; (SD) age: 41 (10) ortho PCBs, measured in > Negative correlation between PCDF and testosterone: r = -0.376, P = 0.022
FSH, LH, progesterone, years) blood (lipid adjusted). > Negatic correlation between PCBs and testosterone: r = -0.339, P = 0.04
prolactin, oestradiol, Mean (SD) sum PCDD/F > Strong positive correlation between PCDD/Fs and prolactin: r = 0.458, P = 0.004.
testosterone and non-ortho PCB TEQ: > Strong correlation between sum PCDD/F and PCBs: r = 0.445, P = 0.006
37.8 (2.1) pg/g lipid In conclusion, higher dioxin levels were associated with lower testosterone and higher prolactin
levels in men.
Miyashita et al., Prospective birth Reproduction: n = 183 mother–child All 29 congeners measured In multivariate linear regression: No associations for PCDD/F, sex-dependent association between
2018 cohort Progesterone, pair in blood (lipid adjusted) non-ortho PCB and ratio testosterone/oestradiol β (95% CI) = -0.18 (-0.39, 0.03) interaction (P
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Japan oestradiol, testosterone, in pregnancy (2nd or 3rd = 0.018), sex-dependent association between total DLC and ratio testosterone/oestradiol β (95%
androstenedione, trimester). CI) = -0.22 (-0.54, 0.10) (P = 0.049), association between non-ortho PCB and DHEA beta 0.27 (CI
DHAEA, cortisol, Total TEQ median (IQR): 0.01-0.54) (P < 0.05).
cortisone, SHBG, 14.5 (10.4–18.6) pg/g
prolactin, LH, FSH, lipids
Inhibin B, insulin-like
factor 3.
 TABLE 4.1 OVERVIEW OF PRIMARY STUDIES WITH REPRODUCTIVE OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS
OF DIOXINS AND dl-PCBs” (cont.)

AUTHOR, YEAR STUDY DIOXIN AND

STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Oanh et al., 2018 Case-control study Reproduction: Cortisol, Cases: n = 35 17 PCDD/Fs in breast milk Many single congeners were correlated to changes in hormones and enzymes measured in
Viet Nam cortisone, 17-OH-P4, mother-child pairs in (lipid adjusted) children. But for Total TEQ PCDD/Fs the following correlation coefficients were found for each
progesterone, DHEA, hotspot area Mean (SD) TEQ total hormone/enzyme: DHEA r -0.41 p value < 0.001, Androstenedione r 0.35 p value < 0.001,
androstenedione, Controls: n = 50 PCDDs/Fs: testosterone r -0.45 p value < 0.001, CYP lyase r 0.36 p value 0.004, 17β-HSD r -0.52 p value
testosterone, 3β-HSD, mother–child pairs in < 0.001, 3β-HSD r -0.31 p value < 0.001. Further, a multiple regression showed for sum TEQ for
> Cases (hotspot): 10.6
17βHSD, CYP17 lyase in non-sprayed area DHEA β of -0.41 p value < 0.001, for Androstenedione β 0.28 p value 0.005, testosterone β -0.46
(1.4) pg/g
children. p value < 0.001.
Children followed up > Controls (non-sprayed):
at age 5 years 3.4 (1.5) pg/g
Oyama et al., 2021 Prospective case- Reproduction: Cases: n = 45 17 PCDD/Fs in breast milk Same as at age 5, lower testosterone levels in hotspot children (both boys, -66.7%; and girls,
Viet Nam control study Cortisol, Cortisone, mother–child pairs in (lipid adjusted) -45.7%), more significant in boys. Calculated 17β-HSD activity lower in only boys. DHEA higher
17-hydroxyprogesterone, hotspot area in hotspot area, but only in boys. DHEA level decreases at ages 3 and 5 in girls, but recovered to
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

Mean (SD) TEQ total

progesterone, Controls: n = 51 PCDDs/Fs: normal range by age of 7 (DHEA levels significantly increased in boys at age 7). A-dione levels
testosterone, DHEA, mother–child pairs in increased in boys at age 5 in hotspot. At age 7, no differences for boys compared to control. For
> Cases (hotspot): 10.8
estrone, oestradiol. non-sprayed area girls, at age 5, A-dione levels now decreased and no differences between hotspot and control.
(1.4) pg/g
enzyme activity of In boys, the serum testosterone level and 17β-HSD activity had a strong inverse correlation with
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

Followed-up 4 to 16 > Controls (non-sprayed):

3β-hydroxysteroid the TEQ total PCDD/Fs level in the breast milk (r = −0.47, p = 0.001, and r = −0.62, p < 0.001,
weeks after birth, 3.2 (1.5) pg/g
dehydrogenase respectively). However, this was not seen in the girls (r = −0.21, p = 0.139, and r = −0.18, p =
and at 1, 3, 5 and 7
(3β-HSD), 0.22, respectively). On the other hand, a positive correlation (r = 0.49, p < 0.001, and r = 0.56,
years old
17β-hydroxysteroid p < 0.001) was observed between the progesterone and TEQ total PCDD/Fs in boys and girls. In
dehydrogenase boys, DHEA and CYP17 lyase were specifically observed to have positive correlations (r = 0.5, p
(17β-HSD), and < 0.001, and r = 0.4, p < 0.01) with the TEQ total PCDD/Fs in breast milk. However, this was not
cytochrome P450 observed in girls (r = 0.12, p = 0.39, and r = −0.01, p = 0.972). The cortisol, cortisone and 17-OH
17,20-lyase (CYP17 progesterone concentrations were not significantly correlated with the dioxin congeners in either
lyase) in children. the boys or girls (data not shown).
Paul et al., 2017 Case-control study Reproduction: Sperm Cases: n = 24 men 12 dl-PCB measured in Individuals with altered semen parameters exhibited significantly higher levels of non-ortho PCBs
Spain concentration, volume, with low semen serum (lipid adjusted). (949.49 ± 624.97 pg/g lipid; p = 0.020) and total Dl-PCBs (7029.96 ± 3023.97 pg/g lipid; p =
percent motile quality Cases (low-semen-quality 0.028) than in the control group (508.40 ± 324.44 pg/g lipid and 4805.92 ± 2205.02 pg/g lipid,
sperm, and percent Controls: n = 26 men group): Mean (SD) total respectively). But this was only significant for the individual congener PCB 105 (p = 0.031). For
morphologically normal with normal semen TEQ 22.52 (21.2) pg/g. the low-semen-quality group, negative significant correlation between PCB 126 in serum, the
sperm quality most toxic dioxin-like PCB, and viability (r = −0.645; p = 0.013). Moreover, sperm morphology
Controls (normal-semen-
was positively correlated with two non-ortho PCBs, PCB 77 (r = 0.671; p = 0.009) and PCB 81 (r =
quality group): Mean (SD)
0.552; p = 0.041). Finally, positive correlations between sperm volume and PCB 118 (r = 0.556; p
total TEQ 14.00 (10.82)
= 0.039), total mono ortho PCBs (r = 0.583; p = 0.029) and total Dl-PCBs (r = 0.593; p = 0.025)
pg/g.
were found.

153
 TABLE 4.1 OVERVIEW OF PRIMARY STUDIES WITH REPRODUCTIVE OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS

154
OF DIOXINS AND dl-PCBs” (cont.)

AUTHOR, YEAR STUDY DIOXIN AND

STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Shi et al., 2020 Cross-sectional Reproduction: n = 78 men (mean 17 PCDD/Fs measured in In general, linear models the significant findings were:
China study Testosterone, [SD] age: 66 [5] years) serum (lipid adjusted). Adjusted means (95% CI) by quartiles of dioxins and dl-PCBs:
dihydrotestosterone living in an e-waste Mean (SD) TEQ PCDD/Fs:
DHEA: PCDFs TEQ: Quartile 1 (reference: <3.80 pg/g): 1 447 (1 114, 1 780) vs. quartile 2 (3.80 –
(DHT), region in China 19.8 (13.8) pg/g
6.31 pg/g): 1 933 (1 568, 2 298) (P < 0.05). No difference with Q3 or Q4.
dehydroepiandrosterone
(DHEA), DHEA: PCDD/Fs TEQ: Quartile 1 (reference: < 8.57 pg/g): 1 360 (1 043, 1 678) vs. quartile 2 (8.57
androstenedione –15.11 pg/g): 1 966 (1 666, 2 326) (P < 0.01). No difference with Q3 or Q4.
(A-dione), 3β-HSD: TCDD: Quartile 1 (reference: < 1.30 pg-TEQ/g): 496 (326, 665) vs. quartile 2 (1.30 – 1.67
3β-hydroxysteroid pg-TEQ/g): 719 (545, 893) (P < 0.05), and vs. quartile 4 (≥ 2.64 pg-TEQ/g): 807 (618, 996) (P <
dehydrogenase 3β-HSD 0.01).
Sun et al., 2017 Case-control study Reproduction: Cases: n = 50 from 17 PCDD/Fs and 4 non- No significant association between serum hormones and any congener after adjustment in
Viet Nam Testosterone, dioxin hotspot ortho PCB, measured in multiple linear regression. But unadjusted simple correlation found effect for certain congeners
dihydrotestosterone Controls: n = 48 from serum (lipid adjusted) and certain hormones.
(DHT), non-sprayed regions Sums and TEQ not given
dehydroepiandrosterone but calculated in ng TEQ/g
(DHEA), oestradiol, lipid.
3β-hydroxysteroid Cases mean: 34.0 ng TEQ/g
dehydrogenase 3β-HSD Controls mean: 10.6 ng
TEQ/g

Notes: OCDD: octachlorodibenzodioxin, PCDF: polychlorinated dibenzofurans, SD: standard deviation, TEQ: toxic equivalent quotient, NHANES: National Health and Nutrition Examination Survey, CI:
confidence interval, PCDD/Fs: polychlorinated dibenzo‐p‐dioxins and dibenzofurans, r: spearman correlation coefficient, LH: Luteinizing hormone, P: P-value, PCB: polychlorinated biphenyls, TCDD:
2,3,7,8-tetrachlorodiobenzo-p-dioxin, HxCDD: hexachlorodibenzo-p-dioxin, FSH: follicle-stimulating hormone, DHAEA: dehydroepiandrosterone , SHBG: sex hormone binding globulin.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

4.2.3 FEMALE REPRODUCTIVE EFFECTS

The main endpoints referenced by EFSA on female reproductive effects were on
endometriosis, pubertal development, and other effects on female reproduction. No
further studies on female reproduction were found in the literature search.

4.2.3.1 Endometriosis
Twelve studies addressing endometriosis were assessed by EFSA, one of which was
a prospective cohort study, while the rest were cross-sectional case-control studies.
No dose-response was observed in the prospective cohort study, and EFSA found
limitations in the cross-sectional studies. Therefore, EFSA concluded that the studies
were insufficient to conclude on the association between serum levels of dioxin and
dl-PCB, and endometriosis.

4.2.3.2 Pubertal development

Four studies addressing pubertal development were assessed by EFSA, and no
association between dioxin and dl-PCB exposure and female pubertal development
was found.

4.2.3.3 Other effects on female reproduction

EFSA reported on three studies that investigated the effect of dioxins and dl-PCBs
on the menstrual cycle. EFSA reported that the results gave no consistent support for
an association between exposure levels and irregular menstrual cycles. Furthermore,
EFSA reported on four studies addressing the effect of dioxins and dl-PCBs on
either time to pregnancy, ovarian function, leiomyomas, or age at menopause. EFSA
concluded that, since there was only one study per outcome, the evidence was
insufficient.

4.2.4 BIRTH OUTCOMES

For birth outcomes, EFSA divided the main effects into sex ratio, birth weight
and other birth outcomes. One further study was found in the literature search
addressing birth weight as an outcome (Table 4.2).

4.2.4.1 Sex ratio

EFSA reported on four studies investigating the difference in sex ratio of children
born to parents exposed to dioxins and dl-PCBs. EFSA rated three of these as Tier
2 studies, and one as a Tier 1 study. Although there were some uncertainties in the
back-calculation of the levels of dioxins in the parents, a pattern of reduced sex
ratio (number of newborn males divided by total births) was observed across three
different cohorts. EFSA therefore concluded that the decreased sex ratio in response
to dioxin exposure likely is causal. No further studies were found in the literature
search reporting on this outcome.

155
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

4.2.4.2 Birth weight and other outcomes

EFSA assessed 18 studies on birth weight and other outcomes (12 Tier 2 studies and
6 Tier 1 studies). The other outcomes assessed were: Yusho disease, gestational age,
child head circumference, birth defects, parity, spontaneous abortion, preterm delivery,
pregnancy loss, congenital anomalies and infant death. EFSA concluded that the
studies were inconclusive and could not be used for risk assessment. One additional
study (Kobayashi et al., 2017) was found which investigated an association between
birth weight and total dioxin exposure with regard to polymorphism in three dioxin-
metabolizing enzymes as a factor. The genes assessed were AhR, CYP1A1 (Cytochrome
P450 1A1) and GST (Glutathione S-Transferase). This study only found a significant
correlation between birth size and dioxin exposure, when combined with the GST null
genotype. However, this study, taken together with the findings from the EFSA report,
still shows inconsistent data, rendering it difficult to determine a conclusion of effect.

4.2.5 THYROID DISEASE AND THYROID HORMONES

The EFSA report concluded that there was insufficient evidence for an association
with thyroid function/disease in adults from studies resulting from accidental exposure
or incidents (high exposure to TCDD or PCDD/F and dl-PCBs). One study from
the EFSA report (Baccarelli et al., 2008), provided relatively strong support for a
causal association between prenatal exposure to TCDD and increased neonatal blood
thyroid stimulating hormone (TSH) concentration, indicating possible subclinical
hypothyroidism in highly exposed children from Seveso. Studies with background
exposure in newborns or children did not suggest any adverse effects on thyroid function
in children. Four further studies were found in the literature search addressing the
outcome of thyroid hormone function. A summarization of these is given in Table 4.3.

4.2.5.1 Studies in adults

Two studies from the literature search assessed thyroid hormone function in adults
(Table 4.3). One study (Li et al., 2019) assessed maternal thyroid hormone associations
in mothers from the LUPE cohort (n = 99 breast milk samples) by investigating
the association between background exposure to seventeen PCBs and five PCDD/
PCDFs and levels of total thyroxine (T4), triiodothyronine (T3), and reverse T3 (rT3).
The study found that total T3 was significantly inversely associated with OCDD.
The study suggested that dioxin levels in breast milk may be associated with maternal
thyroid disruption, but that the study has limited power due to the small sample size.
The other study included from the literature search (Li et al., 2018) assessed total
concentrations of T4, T3 and rT3 in placenta samples from mothers who gave birth to
boys only (n = 58) and background exposure of dioxins and dl-PCBs (also measured
in placenta). In that study, T4 was inversely associated with TCDD and positively
associated with 1,2,3,4,6,7,8-HpCDF; T3 was positively associated with TCDF and
PCDF; and rT3 was positively associated with PCB 81, PCDF and 2,3,4,6,7,8-HxCDF.
The study suggested that dioxin exposure is associated with thyroid hormone levels
in placenta, but that it has limited power due to the small sample size.

156
 TABLE 4.2 OVERVIEW OF PRIMARY STUDIES WITH BIRTH WEIGHT AND OTHER OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF DIOXINS AND dl-PCBs”

AUTHOR, YEAR STUDY DIOXIN AND

STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Kobayashi et al., Prospective birth Reproduction: birth Pregnant women All 29 congeners measured In multiple linear regression: Regression coefficient divided in polymorphisms groups:
2017 cohort study weight, length, (n = 356) enrolled in maternal blood (lipid Mothers carrying the GSTM1 null genotype showed a tenfold increase in TEQ, associated with
Japan head circumference, and followed up with adjusted). a decrease in birth weight of β = -345 g (95% CI -584, -105).
polymorphism in their newborns Mean TEQ level: 17.5 pg/g Mothers carrying the CYP1A TT/TC genotype showed a significant decreasing effect on birth
AhR, CYP1A1 and (n = 148) lipid. weight β = -202 g (95% CI -387, -17)
glutathione-S-
transferase

Notes: AhR: aryl hydrocarbon receptor, TEQ: toxic equivalent quotient, CI: confidence interval.
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

157
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

4.2.5.2 Studies in newborns or children

Two studies from the literature search assessed thyroid hormone function in
newborns or children (Table 4.3). One study investigated the association between
low background exposure of 29 congeners of PCDD/PCDF and dl-PCBs in
mothers (n = 386) at gestational week 30 and thyroid hormones (TSH and free
T4[FT4]) in infants (n = 410) and mothers (Baba et al., 2018). The mothers were
recruited at a hospital in Hokkaido, Japan and median whole blood total TEQ
was 13.8, with a range of 3.42 to 43.4 pg/g lipid. There was a positive association
between mother dioxin TEQ and neonatal FT4. Total dioxin-TEQ and coplanar
PCBs were positively associated with neonatal FT4, and the association was stronger
in boys. Several PCDD/F and PCB isomers were also positively associated with
neonatal FT4. Total PCBs or non-dioxin-like PCBs were not associated with any
maternal or neonatal thyroid hormones. The results suggest that perinatal exposure
to background-level of dioxin-like substances increases neonatal FT4, especially in
boys. The other study included from the literature search (Warner et al., 2020a),
reported on the association between mothers (n = 383) in the Seveso Women’s
Health Study exposed to high concentrations of TCDD from the factory explosion,
and thyroid hormones (TSH, FT4, free T3 [FT3] and total T4) in children (n = 288
female, n = 282 male). Maternal serum TCDD was estimated at pregnancy (14.4
[6.4-33.3 ppt]) based on an initial measurement of 60.2 (28.4 – 156 ppt). Maternal
TCDD was associated with lower FT3 and FT4 in offspring.
The study by Baba et al. (2018) provided evidence for an association between
prenatal low/moderate exposure and increased neonatal blood FT4. However, no
firm conclusions could be made on the association between thyroid hormones and
exposure of dioxins in children (inconsistency of direction of effect). In conclusion,
the additional studies on thyroid hormones are in line with the EFSA conclusion
that studies with background exposure to TCDD, other PCDDs, PCDFs or dl-
PCBs do not suggest any adverse effects on the thyroid.

158
 TABLE 4.3 OVERVIEW OF PRIMARY STUDIES WITH BIRTH WEIGHT AND OTHER OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF DIOXINS AND dl-PCBs”

AUTHOR, YEAR STUDY DIOXIN AND

STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Baba et al., 2019 Prospective birth Thyroid hormone Pregnant women (n All 29 congeners measured In multiple linear regression (β, 95% CI): Effect between maternal PCDD/F and neonatal FT4
Japan cohort study function: TSH and FT4 = 514) enrolled and in maternal blood (lipid (β = 0.224, 95% CI: 0.016, 0.433), and between dl-PCB and neonatal FT4 (β = 0.206 [0.034,
followed up with their adjusted). 0.378]). This effect was higher for boys compared to girls.
infants (n = 410). Median maternal total Maternal non-ortho PCBs were positively associated with maternal FT4 (β = 0.185, P = 0.023).
(range) TEQ: 13.8 In conclusion, the study suggested that perinatal exposure of dioxins and dl-PCBs increases
(3.42–43.4) pg/g lipid. neonatal FT4 concentrations, especially in boys.
Li et al., 2018 Nested case-control Thyroid disease and n = 58 placenta 14 PCDD/Fs and 35 PCBs T4 was inversely associated with 2378-TeCDD, and positively associated with
Denmark study thyroid hormones: Total samples collected measured in placenta 1234678-HpCDF; T3 was positively associated with 2378-TeCDF and 12378-PeCDF; and
T4, T3, rT3 in placenta from mothers of boys (lipid adjusted). rT3 was positively associated with PCB 81, 12378-PeCDF, and 234678-HxCDF (p<0.05 and
born with: Geometric mean: PCDD p<0.01).
Cases TEQ: 12.6 and PCB TEQ:
(cryptorchidism):
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

2.41
n = 28
Controls:
(cryptorchidism):
n = 30
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

Li et al., 2019 Cohort study Thyroid disease and n = 99 mothers 17 PCDD/Fs and five PCBs Total T3 was inversely associated with OCDD (95% CI: −0.20, −0.003)
Germany thyroid hormones: Total including breast milk measured in breast milk
T4, T3, rT3 in placenta samples two months (lipid adjusted).
after birth Sum total PCBs: Mean
(range): 9 090 (2 222–20
225) pg/g
Sum total PCDD/Fs: Mean
(range): 35.7 (0.00–115)
pg/g
Warner, Rauch and Prospective birth Thyroid disease and Women (40 years of Median (IQR) maternal Effect measured in adjusted β (95% CI). Compared to the lowest quartile (Q1), maternal TCDD
Ames et al., 2020 cohort (SWHS study) thyroid hormones: TSH, age) exposed during 1976 serum TCDD: 50.2 was associated with lower free T3 (Q2: adj-β = −0.13; Q3: adj-β = −0.22; Q4: adj-β = −0. 14;
Italy FT3, FT4 Seveso explosion in (28.4-156) ppt (lipid p-trend = 0.02). In participants with high thyroid antibody status, inverse associations between
1976. Children who adjusted) maternal initial serum TCDD and free T3 were significantly stronger than in participants with
were born after the Median (IQR) estimated normal antibody status (p-interaction = 0.02). Positive association between maternal initial
explosion in 1976 of TCDD at pregnancy serum TCDD and TSH concentrations in participants with high thyroid antibody status (Q2:
the pregnant women (estimated from measured adj-β = 11.4%; Q3: adj-β = 49.0%; Q4: adj-β = 105.5%; p-trend < 0.01) but not in those
were included and value in 1976): 14.1 participants with normal antibody status. Similar results were found for TCDD estimated at
followed from age (6.4–33.3) ppt (lipid pregnancy.
2–17 years (n = 570) adjusted)

Notes:TEQ: toxic equivalent quotient, PCDD/Fs: polychlorinated dibenzo‐p‐dioxins and dibenzofurans, PCDD: polychlorinated dibenzo-p-dioxin, PCB: polychlorinated biphenyls, dl-PCB: dioxin-like polychlorinated
biphenyls, TCDD: 2,3,7,8-tetrachlorodiobenzo-p-dioxin.

159
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

4.2.6 TYPE 2 DIABETES AND OBESITY

As both T2D and obesity belong to the cluster of metabolic conditions, which are
closely associated and represent risk factors for the development of CVD metabolic
syndrome, the two outcomes were evaluated and presented together by EFSA in
their 2018 assessment. EFSA included in their assessment 20 epidemiological studies
investigating associations between exposure to PCDD/Fs and dl-PCBs and T2D
and obesity. Fourteen of the studies were rated Tier 2 and six were rated Tier 1.
Several methodological weaknesses were pointed out for the majority of studies
included in the assessment. Uncertainties in the results were described with regards
to the possibility of reverse causality in cross-sectional studies, but also flawed
study design or reporting (very low number of participants or cases, undefined
study populations, poor quality of outcome assessment, selection bias). Four studies,
highlighted by EFSA as the most convincing evidence, showed mixed results, and
EFSA concluded that the studies on T2D and obesity were inconclusive and could
not be used as a basis for the risk assessment.
Three additional studies were found in the literature search investigating associations
between exposure to dioxins and obesity or diabetes. The studies are described in the
following sections and details are shown in Table 4.4. The three studies investigate
different outcomes. Thus, evidence presented in these studies may not be sufficient
to change the conclusion made by EFSA.

4.2.6.1 Prenatal exposure in the Seveso Second Generation Study

Two studies of the Seveso Second Generation Study were included (Warner et al.,
2019 and Warner et al., 2020b). Warner et al., 2019 reported on metabolic outcomes
of 611 children (n = 314 girls, n = 297 boys) born to 402 mothers of the Seveso
Women’s Health Study, who had been exposed to high concentrations of TCDD
during or before their childbearing years from the factory explosion in 1976. In
utero TCDD exposure was estimated from the body burden measured in sera soon
after the accident or during follow-ups (1996 or 2008), using a first-order kinetic
model with a half-life adjusted to the initial dose, age and other co-variates. In
addition, the assessed outcome parameters were tested towards maternal initial
TCDD burden in sera. The outcome parameters recorded in the offspring were
BMI, waist circumference, elevated triglyceride levels, total cholesterol, HDL-C,
glucose in plasma from a fasting blood draw and blood pressure measurement. The
outcome parameters were classified and combined to assess development or presence
of metabolic syndrome. The study found sex-dependent associations between the
initial maternal TCDD levels and the cardiometabolic outcomes. A tenfold increase
in initial maternal TCDD concentration was inversely associated with BMI in girls,
but not boys. In contrast, in boys only, initial maternal TCDD was associated with
an increased risk for metabolic syndrome. Results for TCDD estimated at pregnancy
were comparable.
In the second study included in the literature, Warner et al., 2020b, the authors
investigated associations between maternal TCDD exposure and glucose metabolism

160
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

in the same study population. However, analysis was restricted to their children of
18 years and older to exclude variation from decreased insulin sensitivity during
puberty, rendering 426 adult children born to 303 mothers, who were included in the
analysis. The relationship of maternal TCDD burden (estimated as explained above)
and different endpoints (serum insulin and plasma glucose from a fasted blood
draw and the computer-based homeostatic model assessments for insulin resistance
[HOMA2-IR] and beta-cell function [HOMA2-B]) were assessed. In line with the
previous study, sex-specific effects were reported. The maternal TCDD burden
estimated at pregnancy was inversely associated with serum insulin and HOMA2-B
among daughters, but not among sons. Similar effect modification was observed for
TCDD estimated at pregnancy and HOMA2-IR. However, as reported by Warner
et al., 2019, the associations observed between serum insulin and HOMA2-B in
female offspring suggested to be mediated by BMI.

4.2.6.2 Metabolic effects in pregnant women

One study (Liu et al., 2019) investigated the association between dioxin-like
compounds and the prevalence of gestational diabetes mellitus (GDM) and with a
nested case-control study in a population of pregnant Chinese women. The main
aim of the study was to establish human-based relative-effect potencies. As this is
not relevant for the risk assessment of dioxins, the results on relative-effect potencies
are not discussed. There were 439 pregnant women recruited in the study. GDM
was diagnosed based on an oral glucose tolerance test and fasting and postprandial
blood glucose levels. Seventy-seven diagnosed cases and 2 pair-matched controls per
case (154 controls in total) were included in the study (overall study participation
rate – 53 percent of initially recruited population). An association with obesity was
not apparent for either exposure or GDM outcome. Seventeen PCDD/Fs and 12
dl-PCBs were measured in serum. The median (interquartile range) of total TEQ
for GDM and controls were 9.9 (range: 7.3–13.2) and 6.9 (range: 5.8–9.2) pg/g
lipids, respectively. The association between total TEQ exposure and GDM risk
was calculated from prevalence in the interquartile ranges. Significantly increased
prevalence for GDM was found in Q3 and Q4 with quartile levels of total TEQ
≥7.72 pg/g lipid. In addition, an association was found between per SD increase in
total TEQ and increase risk in GDM (OR = 2.12, 95% CI: 1.57, 2.86).

161
 TABLE 4.4 OVERVIEW OF PRIMARY STUDIES WITH TYPE 2 DIABETES AND OBESITY OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF DIOXINS AND dl-PCBs”

162
AUTHOR, YEAR STUDY DIOXIN AND
STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCBs EXPOSURE
Liu et al., 2019 Prospective nested Type 2 diabetes and Cases with All 29 congeners measured Total TEQ cases: Median (IQR): 9.9 (7.3-13.3) g/g
Viet Nam case-control study obesity: Gestational gestational diabetes in blood (lipid adjusted). Total TEQ controls: Median (IQR): 6.9 (5.8-9.2) g/g
diabetes mellitus (GDM) mellitus: n = 77 Total TEQ whole group In adjusted analyses (BMI and foetal sex) of risk of GDM and association with total TEQ: OR
and fasting blood Controls: n = 154 median (IQR): 7.72 (95% CI): 2.12 (1.57, 2.86) per change in SD of total TEQ.
glucose (6.14-10.36) pg/g lipids When departed into quartile levels of total TEQ (pg/g):
Q1 (<6.14 pg/g): Reference
Q2 (6.14-7.72 pg/g): 2.04 (0.78, 5.35)
Q3 (7.72-10.36): 4.02 (1.61, 10.06)
Q4 (≥ 10.36): 7.74 (3.10, 19.29)
Ptrend <0.001
Warner, Rauch and Prospective birth Type 2 diabetes and Women (40 years of Median (IQR) maternal Effect measured in adjusted β (95% CI). TCDD estimated at pregnancy was inversely
Brambilla et al., cohort obesity: Glucose, insulin age) exposed during 1976 serum TCDD: 53.1 associated with insulin (adj β = −1.24 μIU/mL, 95% CI: −2.38, −0.09) and HOMA2-B (adj β =
2020 (SWHS study) HOMA2-IR, HOMA2-B, Seveso explosion in (25.1-112) ppt (lipid −10.2% decrease, 95% CI: −17.8, −1.9) among daughters, but not sons (insulin: adj β = 0.57
Italy 1976. Children who adjusted) μIU/mL, 95% CI: −0.84, 1.98, P for interaction = 0.04; and HOMA2-B: adj-β = 0.8% increase,
were born after the Median (IQR) estimated 95% CI: −10.7, 13.9, P for interaction = 0.11). Similar effect modification was observed for
explosion in 1976 of TCDD at pregnancy TCDD estimated at pregnancy and HOMA2-IR (P for interaction = 0.13). The observed
the pregnant women (estimated from measured associations in daughters showed evidence of mediation by BMI, which we have previously
were included and value in 1976): 20.6 (9.4- found to be associated with prenatal TCDD exposure in female offspring.
followed up at age 18 47.1) ppt (lipid adjusted)
or older (n = 426)
Warner et al., 2019 Prospective birth Type 2 diabetes and Women (40 years of Median (IQR) maternal A tenfold increase in initial maternal TCDD concentration was inversely associated with BMI in
Italy cohort obesity: BMI, metabolic age) exposed during 1976 serum TCDD: 51.0 girls (adj-β = −0.99 kg/m2), but not boys. In contrast, in boys only, initial maternal TCDD was
(SWHS study) syndrome (MetS) based Seveso explosion in (24.4–108) ppt (lipid associated with increased risk for MetS (adj RR = 2.09, 95% CI 1.09, 4.02). Results for TCDD
on waist circumference, 1976. Children who adjusted) estimated at pregnancy were comparable.
fasting plasma TAG, were born after the Median (IQR) estimated
HDL-cholesterol, total explosion in 1976 of TCDD at pregnancy
cholesterol, glucose and the pregnant women (estimated from measured
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

blood pressure. were included and value in 1976): 10.6

followed from age 2 to (5.5–25.3) ppt (lipid
17 years (n = 611) adjusted)

Notes:TEQ: toxic equivalent quotient, PCDD: polychlorinated dibenzodioxins, PCDD/Fs: polychlorinated dibenzo-p-dioxins and dibenzofurans, PCB: polychlorinated biphenyls, dl-PCBs: dioxin-like polychlorinated
biphenyls, BMI: body mass index, SWHS: Seveso Women's Health Study, IQR: interquartile range, TCDD: 2,3,7,8-tetrachlorodiobenzo-p-dioxin, CI: confidence interval, TAG: triacyl glycerides
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

4.2.7 CARDIOVASCULAR EFFECTS

EFSA reported on 12 epidemiological studies covering the effects of dioxins and
dl-PCBs on different cardiovascular endpoints, such as: cardiovascular mortality,
heart disease, ischemic heart disease and hypertension. Ten studies were rated
Tier 2, while two studies were rated Tier 1. EFSA pointed to one study strongly
supporting increased cardiovascular risk after exposure to TCDD (Steenland et al.,
1999), however, this study was conducted on workers with very high occupational
exposure (serum TCDD > 1000 pg/g fat). Studies assessing lower exposures to
dioxins and dl-PCBs were, according to EFSA, inconclusive in finding associations
between cardiovascular risk and dioxin or dl-PCB exposure. No further studies were
found in the literature search on cardiovascular risk factors after dioxin exposure.

4.2.8 HEPATIC DISORDERS AND DIGESTIVE EFFECTS

EFSA reported on five studies that compared blood PCDD/F levels of occupationally
or accidentally exposed cohorts and potential non-cancer hepatic and digestive
disorders or abnormal function. Three of the studies were rated Tier 1. No further
studies were found in the extensive literature search. In summary, based on the few
existing studies, EFSA concluded that there is no evidence for an association of
hepatic or digestive diseases with prolonged accidental or occupational exposure
to PCDD/Fs.

4.2.9 EFFECTS ON THE IMMUNE SYSTEM

EFSA reported on 14 studies, which did not present strong evidence for an effect
of dioxin exposure on the immune system at adolescence or adulthood. Only one
of these studies (Dinse et al., 2016) was rated Tier 1, and this study examined an
association between autoimmunity and serum concentrations of PCDD/Fs and dl-
PCBs in a cross-sectional study with 4 340 participants from the National Health
and Nutrition Examination Survey cohort. No associations were observed in this
study. Further, the studies in adults were retrospective and had shortcomings related
to exposure evaluation and confounding factors. Moreover, the studies measured
different immune parameters, making it difficult to compare them. In addition to the
studies in adults, EFSA reported on eight studies on exposure during development.
Half of these studies were given a Tier 1 rating, and these are summarized here.
Weisglas-Kuperus et al. (2000) investigated immunological and clinical parameters in
children. After primary vaccination against mumps, measles and rubella, associations
with antibody levels were observed for the non-dl-PCBs in blood. PCDD/Fs and
dl-PCBs were not examined. PCDD/Fs and dl-PCBs were measured in milk but did
not show a relation with the levels of antibodies or illness parameters investigated.
Van Den Heuvel et al. (2002) performed a study on 207 adolescents and found a
decreased prevalence of allergic responses with increasing exposure. (Nagayama
et al., 2007) assessed lymphocyte subsets in peripheral blood of 92 children and
correlated this to the concentration of organochlorine compounds in their mothers’

163
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

breast milk. They found an association of exposure to PCDD/Fs and dl-PCBs and
an increased ratio of CD4/CD8 cells, as well as an increased percentage of CD3
cells. However, EFSA concluded that the clinical relevance of this finding is unclear.
Miyashita et al. (2011) correlated prenatal exposure to dioxin-like compounds
with allergies and infections during infancy (birth cohort n = 514). Higher levels
of exposure in males were found to be associated with increased incidence of otitis
media. In addition, the authors reported a minor significant association with allergies
during infanthood. Since this study measured an extended number of outcomes, the
possibility for type 1 errors was considered likely.
According to EFSA, the available studies did not provide sufficient evidence for
an association between PCDD/Fs or dl-PCBs and effects on the immune system.
No further studies were found in the literature search on the effects on the immune
system after dioxin exposure.

4.2.10 EFFECTS ON THE NERVOUS SYSTEM

EFSA reported on 15 studies in children and 7 occupational studies in adults for the
outcome “effects on the nervous system”. Various neurodevelopmental outcomes
at different ages were investigated in children, but few outcomes were assessed
consistently in more than one cohort or at similar age. No association or conflicting
results were observed between dioxins and cognitive and motor development in
infants and children (> 11 years old). In a Norwegian maternal-exposure cohort
study, there was a small increase in OR for negative effects on grammar skills in
children. Girls had increased OR for moderate to severe language delay and reduced
communication skills. Results from occupational cohorts indicated a persistent
association between TCDD exposure and different neurophysiological outcomes.
However, back calculation to original exposures based on levels in blood sampled
decades after the exposure are highly uncertain. Also, the occupational studies suffer
from lack of participants. Thus, EFSA concluded that the available information
was not sufficient to form a basis for the risk assessment on neurodevelopment in
children. There was insufficient information to draw conclusions on effects on the
nervous system after exposure in adult life. There is also no apparent mode of action.
In addition to the EFSA report, one additional study was found in the literature search
on neurodevelopment in children (Table 4.5). In this study, neurodevelopmental
outcomes were measured at 2 years of age, showing an association between neonatal
electroencephalography brain signals with increased TCDD exposure (Pham et al.,
2021). However, a direct association between gaze behaviour and TCDD was not
reported.
Due to limited novel evidence following the EFSA report, EFSA’s conclusion
still stands: There is not sufficient evidence to conclude on neurodevelopmental
outcomes.

164
 TABLE 4.5 OVERVIEW OF PRIMARY STUDIES WITH NERVOUS SYSTEM OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF DIOXINS AND dl-PCBs”.

AUTHOR, YEAR STUDY DIOXIN AND

STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Pham et al., 2021 Prospective birth Nervous system: Intra- n = 51 mother–infant 17 PCDD/Fs measured in Perinatal dioxin exposure, particularly TCDD exposure above 2.8 pg/g, influenced relative EEG
Viet Nam cohort uterine brain activity pairs breast milk sample one power values mainly in the intra-burst-interval part of the trace alternant pattern in the quiet
(measured by EEG 2 month after birth (lipid sleep stage. In intra-burst-intervals, decreased frontal delta power and increased frontal and
days after birth). 2 adjusted). parietal alpha power values in the left hemisphere and temporal beta power values in the right
years after birth: gaze Mean (SD) TEQ PCDD/Fs: hemisphere were associated with increased TCDD exposure, with significant dose–response
behaviour, cognitive, 7.9 (1.6) pg/g relationships.
language and motor
skills measured by
Bayley-III

Notes: dl-PCB: dioxin-like polychlorinated biphenyl, EEG: electroencephalogram, PCDD/Fs: polychlorinated dibenzo‐p‐dioxins and dibenzofurans, TCDD: 2,3,7,8-tetrachlorodiobenzo-p-dioxin, SD: standard
deviation, TEQ: toxic equivalent quotient.
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

165
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

4.2.11 EFFECTS ON TEETH AND BONES

EFSA assessed three studies investigating associations between dioxins and dl-
PCBs and teeth development from three different population groups. A dose–
response relationship was observed for tooth enamel hypomineralization or enamel
defects. The hypomineralization of permanent teeth is likely to be causally related
to postnatal exposure. Increased incidents of hypodontia were also associated
with exposure to TCDD serum levels. Tooth defects have also been reported to
be the most sensitive response to TCDD in animal models. Unlike teeth, bone
is undergoing a constant remodelling process. Limited evidence from one cohort
indicates associations between PCDD/F and dl-PCB exposure and some changes
in bone parameters.
Although, in the previous EFSA report, the Panel on Contaminants in the Food
Chain concluded that exposure to PCDD/Fs via breast milk may lead to increased
enamel defects in children, these data were deemed less suitable for dose–response
assessment. No further studies on the effect of dioxins on teeth and bones were
identified in the literature search.

4.2.12 CANCER
EFSA divided the studies on cancer into two categories: industrial accidents or
contamination incidents and occupational exposure. For industrial accidents and
contamination incidents, EFSA included five studies. Two were rated Tier 1, and
three were rated Tier 2. For occupational exposure, 17 studies were included, five
of which were rated Tier 1 and twelve were rated Tier 2. In their summary, EFSA
reported that many studies showed a positive association between dioxin exposure
and all cancers combined. However, there was no clear link to any specific site, nor
did the studies show a clear dose–response relationship. Therefore, EFSA deemed
these studies unsuitable for risk assessment.
Two further studies were found in the literature search on the effects of dioxins
and dl-PCBs on cancers (Table 4.6). Koual et al. (2019) conducted a case-control
study with 91 participants, in which 38 displayed metastatic breast cancer and
53 displayed non-metastatic breast cancer. All 29 congeners were measured in
adipose tissue. The study found a positive association between TCDD concentration
in adipose tissue and risk of metastasis in patients with a BMI ≥25 kg/m2 (p-value
0.03). However, most results showed no, or weak, association between dioxins,
PCBs, or groups thereof and metastasis of breast cancer. This study was conducted
on all cancer patients and explored only the prevalence of metastasis and tumour size.
Furthermore, the study was small (n = 91). The second study (Lim et al., 2017), also a
case-control study, investigated prostate cancer risk in a Korean cohort. The authors
randomly selected 1 879 participants from a cohort of 159 844. Of these, 256 controls
and 110 cases (diagnosed with prostate cancer) were used for the analyses. Twelve
dl-PCBs were assessed, and a hazard ratio for TEQ was found at 1.40 (1.21–1.62).
However, the hazard ratio for dl-PCBs did not show a positive association.

166
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

The two additional studies from the literature search showed low causality, had a
low number of observations, and did not change the conclusion made by EFSA.
As such, it was concluded that the studies found do not display a clear association
between dioxins and dl-PCBs and cancers.

167
 TABLE 4.6 OVERVIEW OF PRIMARY STUDIES WITH CANCER OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF DIOXINS AND dl-PCBs”

168
AUTHOR, YEAR STUDY DIOXIN AND
STUDY DESIGN HEALTH OUTCOME RESULTS AND CONCLUSIONS
COUNTRY PARTICIPANTS dl-PCB EXPOSURE
Koual et al., 2019 Case-control study Cancer: Metastasis in Cases: n = 38 All 29 congeners measured TCDD concentration in adipose tissue was positively associated with the risk of metastasis
France breast cancer metastatic breast in blood (lipid adjusted). in patients with a BMI≥ 25 kg/m2: (OR [95% CI]: 4.48 [1.32, 20.7] P = 0.03). Furthermore,
cancer Metastatic median (IQR) the concentrations of TCDD and PCB 126, 118 and 123 in adipose tissue were positively
Controls: n = 53 total TEQ: 22.5 (16.9–30.8) associated with lymph node metastasis and tumour size.
non-metastatic breast pg/g.
cancer Non-metastatic median
(IQR) total TEQ: 24.1
(16.1–35.2) pg/g

Lim et al., 2017 Case-control study Cancer: Prostate cancer Cases: n = 110 12 dl-PCBs, measured in Hazard ratio (95% CI) for TEQ: 1.40 (1.21, 1.62) (P = 0.073)
Republic of Korea Controls: n = 256 blood (lipid adjusted). Hazard ratio (95% CI) for dl-PCBs: 1.39 (0.89, 2.19) (P = 0.146)
Levels not given, only the In conclusion, the findings suggested a possible role of dl-PCBs in the aetiology of prostate
association with health cancer.
outcome.

Notes: IQR: interquartile range, TEQ: toxic equivalent quotient, dl-PCB: dioxin-like polychlorinated biphenyl, BMI: body mass index, OR: odds ratio, CI: confidence interval, TCDD: 2,3,7,8-tetrachlorodiobenzo-
p-dioxin, PCB: polychlorinated biphenyls.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 4 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF DIOXINS AND dl-PCBs”

4.2.13 OTHER EFFECTS

In addition to the effects described above, EFSA found two studies (Michalek
et al., 2001 and Gupta et al., 2006) which could not be attributed to any of the
aforementioned groups. These studies were included in the EFSA report as they
met the criteria. However, since risk assessment should not be based on single
studies, they were not included in the assessment. Both studies were conducted on
veterans from Viet Nam (Operation Ranch Hand). Michalek et al. (2001) reported
on haematological changes in response to TCDD, and no clear effects were found.
Gupta et al. (2006) showed that higher TCDD levels were associated with decreased
risk of benign prostatic hyperplasia. No additional studies were found in the
literature search.

169
 170
© FAO/Fahad Al Dhuhli
CHAPTER 5
RESULTS AND
SUMMARIZATION
OF THE LITERATURE
REVIEW “TOXIC
EFFECTS OF MeHg”

5.1 LITERATURE AND QUALITY ASSESSMENT

Literature searches for the review on the “Toxic effects of MeHg” were performed
in the databases PubMed and Web of Science. A flow diagram of the results from the
literature searches is given in Figure 5.1. The parallel literature searches performed in
Web of Science and PubMed resulted in 3 248 records. Of these, 1 319 were found to be
duplicate records. Following duplicate removal in Endnote, 1 929 records were subjected
to title and abstract screening using Rayyan. Based on the predefined inclusion and
exclusion criteria (Table 2.10), 1 731 records were excluded and 198 records (comprising
43 systematic reviews and 155 primary articles) were retained for full-text screening.

5.1.1 SYSTEMATIC REVIEWS

A total of 43 systematic reviews were assessed in full text after title and abstract
screening. Of these, nine additional records were excluded based on the inclusion
and exclusion criteria (Appendix 5, Table A5.3), leaving 34 systematic reviews that
were assessed with the AMSTAR 2 risk-of-bias tool. Following the assessment
with AMSTAR 2, the studies were graded into categories according to the overall
confidence in the results from the review (Appendix 5, Table A5.5). Four studies
were graded “high” and 12 were graded “moderate”. These were included for further
assessment in the report. Thirteen studies were graded “low” and five studies were
graded “critically low”. These were excluded for further assessment.

171
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

FIGURE 5.1. FLOW DIAGRAM FOR THE REVIEW “TOXIC EFFECTS OF MeHg”

IDENTIFICATION OF STUDIES IDENTIFICATION OF STUDIES

VIA DATABASES VIA OTHER METHODS

Records identified from Records removed before Records identified from:

IDENTIFICATION

database searching (n = 3 248): screening: • Reports (n = 2): EFSA 2012

• Web of Science (n = 1 261) • Duplicate records removed and VKM 2022
• PubMed (n = 1 929) (n = 1 319)

Records screened by title Records excluded by criteria

and abstract (n = 1 929) (n = 1 731)
SCREENING

Full-text articles assessed for Records excluded by criteria

eligibility (n = 198): (n = 98):
• Systematic reviews (n = 43) • Systematic reviews (n = 9)
• Primary studies (n = 155) • Primary studies exclude
by criteria (n = 45)
• Primary articles covered in
included systematic review
(n = 44)

Quality assessment (risk of bias) Records excluded after quality

of full-text articles (n = 100): assessment (risk of bias):
• Systematic review (n = 34) • Systematic reviews (n = 18)
• Primary studies (n = 66) • Primary studies (n = 10)
INCLUDED

Systematic reviews included in final review (n = 16)

Primary studies included in final review (n = 56)
Reports included in final review (n = 2)

Source: The figure was prepared based on Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow,
C.D., Shamseer, L. et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.
BMJ, 372: n71. https://doi.org/10.1136/bmj.n71

5.1.2 PRIMARY STUDIES

A total of 155 primary studies were assessed in full text after title and abstract screening.
Of these, 45 primary studies were excluded based on the inclusion and exclusion
criteria (Appendix 5, Table A5.4) and 44 primary studies were excluded for further
assessment as they were already covered in one of the systematic reviews included
(Appendix 5, Table A5.6). This left 66 primary articles, which were subjected to
risk-of-bias assessment using the OHAT risk-of-bias tool (Appendix 5, Table A5.7).

172
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

Based on the risk-of-bias assessment, 56 primary studies were graded Tier 1, 10 were
graded Tier 2, and 0 were graded Tier 3. Only the 56 studies graded Tier 1 were
included in the final assessment.

5.2 RESULTS AND SUMMARIZATION OF THE LITERATURE INCLUDED

The systematic reviews and original articles included in this extensive literature
review covered a diverse range of health outcomes associated with exposure to
MeHg. These were grouped into four primary domains: (i) neurological outcomes,
(ii) cardiovascular outcomes, (iii) growth and (iv) other outcomes. Narrative
summaries of the findings from the systematic reviews and original primary studies
are reported in the following sections. In addition, summaries of the reports from
the 2012 EFSA and the 2022 VKM assessments are provided.

5.2.1 NEUROLOGICAL OUTCOMES

The following sections provide short narrative summaries of the neurological
outcomes obtained from the systematic reviews and original articles excluded from
the literature review.

5.2.1.1 Systematic reviews

Six articles were identified that reviewed the effects of Hg exposure in relation to
neurological outcomes Table 5.1). Three of the systematic reviews (Jafari et al.,
2017, Saghazadeh et al., 2017 and Zhang et al., 2021a) considered autism spectrum
disorders, and a fourth (Yoshimasu et al., 2014) considered both autism and ADHD.
One systematic review (Puty et al., 2019) assessed neurological effects in adults and
another (Hibbeln et al., 2019) concerned neurodevelopment in children exposed
prenatally to Hg.
The relationship between autism spectrum disorders and Hg in blood was explored
in three systematic reviews. Each of these conducted a meta-analysis of case-control
studies and came to similar conclusions, stating that in subjects with autism spectrum
disorders, higher concentrations of Hg were detected when compared to the
respective control groups. The authors note that the observed differences were small,
that there was evidence of heterogeneity across the studies included in the respective
meta-analysis, and that the study designs employed limited causal inference in
relation to Hg exposure and autism. One of the systematic reviews (Yoshimasu et
al., 2014) considered both autism and ADHD. The authors reviewed four studies
and obtained inconsistent results, with two of the four studies suggesting that Hg
contamination in fish-consuming populations may be related to an increased risk
of autism or ADHD.
One systematic review (Hibbeln et al., 2019), surveyed 25 studies for the associations
of prenatal Hg exposure and neurocognition. This review found six studies that
reported null associations between maternal Hg levels and neurocognition, while

173
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

seven studies (45 957 mother–infant pairs) were found to report paradoxical findings
indicating that higher levels of Hg seemed to have beneficial effects in relation to
neurocognitive development. Overall, Hibbeln et al. concluded that no net adverse
neurocognitive outcomes were reported among offspring at the highest ranges of
seafood intakes, despite associated increases in Hg exposures.
The final review concerning neurological effects (Puty et al., 2019) focused
on disorders in relation to dietary Hg exposure in adults living in areas with
environmental exposures. The review reached no conclusion among the six studies
reviewed, citing low level of evidence and high risk of bias.

174
 TABLE 5.1 OVERVIEW OF SYSTEMATIC REVIEWS WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION

TITLE OUTCOME STUDIES DETECTED
Yoshimasu et al., 2014 Neurological; Total n = 4 studies - Meta-analysis: Summary OR from two of the included studies for the Hg exposure from fish consumption might be
A meta-analysis of the ADHD and autism involving MeHg evaluation of a relationship between Hg exposure and ADHD (OR = 1.60, 95% related to an increased risk of ADHD, but the
evidence on the impact spectrum disorders exposure from CI: 1.10, 2.33). The two other studies included found no effect. results were not consistent. The results suggest
of prenatal and early (ASD) fish consumption: that prenatal or early infant mercury exposures
infancy exposures to Cohort studies (n by vaccination did not have serious adverse
mercury on autism = 2), case-control effects on increasing risks of ASD or ADHD, while
and attention deficit/ study (n = 1) and environmental mercury exposure caused by air
hyperactivity disorder in cross-sectional pollution or maternal fish consumption might
“TOXIC EFFECTS OF MeHg”

childhood study (n = 1) promote incidence of ASD and ADHD, respectively.

Jafari et al., 2017 Neurological; ASD Total n = 44: Case- See results from Meta-analysis: Subjects with ASDs had higher Hg concentrations
The association between control studies (n meta-analysis. The Hg level in whole blood (n = 16 studies with 1 239 cases vs. n = 1 039 compared to healthy subjects. However, the study
mercury levels and = 44) controls: Mean difference: 0.43, 95% CI: 0.12, 0.74, P = 0.007); design limited any causal conclusion.
autism spectrum the Hg level in red blood cells (n = 5 studies with 251 cases vs. n = 915
disorders: A systematic controls, Mean difference: 1.61, 95% CI: 0.83, 2.38, P < 0.001); and
review and meta- the Hg level in brain (n = 3 studies with 16 cases vs. n = 20 controls: Mean
analysis difference: 0.61 ng/g, 95 % CI, 0.02, 1.19, P = 0.043) was significantly higher
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

in ASD patients than in healthy subjects, whereas mercury level in hair (n = 23

studies with 1 038 cases vs. n = 933 controls: Mean difference: −0.14 mg/g,
95% CI: −0.28, −0.01, P = 0.039) was significantly lower in ASD patients than
in healthy subjects.
The mercury level in urine was not significantly different between ASD patients
and healthy subjects (n = 8 studies with 491 cases vs. n = 417 controls: Mean
difference: 0.51 mg/g creatinine, 95% CI: −0.14, 1.16, P = 0.121).
Saghazadeh et al., 2017 Neurological; ASDs Total n = 38: Case- See results from Meta-analysis: Autism spectrum disorders patients vs. controls: Standardized ADS patients had higher Hg concentrations
Systematic review control studies (n meta-analysis. mean difference (95% CI) (P-value): compared to control for some specimens; however,
and meta-analysis = 38) Hg blood, serum, plasma (overall) (n = 629 vs. n = 476): 1.32 (0.51, 2.13) the differences were rather small.
links autism and toxic (P = 0.001)
metals and highlights Hg blood, serum, plasma (without outlier) (n = 584 vs. n = 431): 0.43 (-0.064,
the impact of country 0.92) (P = 0.089)
development status: Hg blood (overall) (n = 547 vs. n = 422): 1.42 (0.47, 2.37) (P= 0.003)
Higher blood and Hg blood (without outlier) (n = 502 vs. n = 377): 0.19 (-0.28, 0.66) (P = 0.422)
erythrocyte levels for Hg erythrocyte: (n = 216 vs. n = 184): 1.56 (0.42, 2.70) (P= 0.007)
mercury and lead, and Hg hair (overall): (n = 1092 vs. n = 973): 0.49 (-0.056, 1.03) (P = 0.079)
higher hair antimony, Hg hair (without outlier) (n = 948 vs. n = 829): 0.082 (-0.20, 0.36) (P = 0.570)
cadmium, lead, and Hg hair (developed and transition countries): (n = 653 vs. n = 537): -0.18
mercury (-0.45, 0.092) (P = 0.195)
Hg hair (developing countries): (n = 295 vs. n = 292): 0.77 (0.31, 1.23)
(P = 0.001)
Hg urine (n = 111 vs. n = 229): 0.25 (-0.11, 0.36) (P = 0.306)

175
 TABLE 5.1 OVERVIEW OF SYSTEMATIC REVIEWS WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg” (cont.)

176
AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION
TITLE OUTCOME STUDIES DETECTED
Zhang et al., 2021 Neurological; ASDs Total n = 12: Case- See results from Meta-analysis: Overall standardized mean difference in Hg concentrations Children with ASDs had some higher levels of Hg
Trace elements in control studies meta-analysis. between ASDs and healthy controls: 1.37 (95% CI: 0.46, 2.28). compared to healthy controls; however, there was
children with autism (n = 12) evidence of heterogeneity.
spectrum disorder: A
meta-analysis based on
case-control studies
Hibbeln et al., 2019: Neurological; Total n = 25 (but Pregnant women: Prenatal exposure: Six studies reported null associations between maternal Consuming seafood during pregnancy and
Relationships between Neurodevelopment; included n = Range mean hair mercury levels and neurocognition, while seven studies (45 957 mother–infant childhood was likely beneficial and clearly not
seafood consumption Intelligence/ 44 articles for Hg 0.-6.9 ppm pairs) reported seemingly paradoxical findings; that higher levels of mercury adverse to neurocognition. No net adverse
during pregnancy Cognition overall seafood had beneficial relationships to neurocognitive development. Eight of the neurocognitive outcomes were reported among
and childhood and assessment): studies had mean hair Hg exposures above 1.1 ppm, and all the studies had offspring at the highest ranges of seafood intakes,
neurocognitive RCTs (n = 4), participants with exposures that were many times higher than the reference despite associated increases in mercury exposures.
development: Two prospective cohort dose. None of the studies reported any net adverse effects on neurocognitive
systematic reviews studies (n = 33), development from seafood consumption in any amount.
case-control Childhood: Two studies reported measurements of mercury exposure in the
studies (n = 5) children. Although mercury levels were not reported in the remainder of the
studies, it is highly likely that greater seafood consumption in those studies
resulted in higher mercury exposures. Since no study reported net adverse
outcomes from seafood consumption in children, it is unlikely that mercury
exposure from seafood was associated with substantive neurocognitive harms.
Puty et al., 2019 Neurological; Total n = 6: Case- Hair Hg The studies suggested alterations related to the psychosensory, motor and The study could not demonstrate an association of
Association between neurotoxicity control studies (n concentrations coordination system, as well as motor speech, hearing, visual impairment, MeHg and neurological alterations.
methylmercury in adults; = 4) and cross- given in two of the mood alterations and loss of intelligent quotient. However, two of the six
environmental exposure neurobehavioral sectional studies studies (mean): studies presented a high risk of bias, with methodological problems related to
and neurological alterations, (n = 2) 8.8 and 2.6 ppm. confounding factors, and all studies presented evidence levels ranging from
disorders: A systematic memory, very low to low.
review communicative
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

and cognitive
disorders, motor
and visual
dysfunctions.

Notes: OR: odds ratio, ADHD: attention deficit hyperactivity disorder, ASD: autism spectrum disorder.
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

5.2.1.1.1 Primary studies

The literature search identified 31 original articles that assessed neurological effects
associated with dietary Hg exposure (Table 5.2). Of these, nineteen reported that
one or more outcomes measured were associated with Hg, whether positively or
negatively. Six of the studies were cross-sectional studies, while 25 were cohort
studies (of which 23 were pregnancy cohorts that specifically included at least one
pre- or perinatal time point). Among the 25 cohort studies, 13 unique cohorts were
identified, indicating that multiple articles were published using data from the same
established cohorts. With four articles each, the “Seychelles Child Development
Study” and the “Arctic Cord Blood Monitoring Program/Nunavik Environmental
Contaminants” were found to be the two most called-upon cohorts in the articles
assessed.
The types of neurological effects investigated were primarily neurodevelopment
in children (28 articles), followed by two articles concerning depression in
adults, and one article concerning neurotoxicity in adults. Under the subheading
“Neurodevelopment”, the outcomes assessed were sorted into ten different
subcategories: behaviour, brain morphology, intelligence/cognition, memory and
learning, mental and psychomotor development, motor function, nerve signalling,
reactions and reflexes, speech and language, and visual-motor integration. In the
following sections, the findings are described for depression and neurotoxicity
and, subsequently, summaries are provided concerning the ten categories of
neurodevelopment. Sixteen of the articles included in this section focused on more
than one of the neurodevelopmental categories listed above. As such, the outcomes
of these articles are described under multiple section headings.

5.2.1.1.2 Depression
Two cross-sectional studies, Berk et al., 2014 and Rossa-Roccor et al., 2021,
considered associations of Hg exposure with depression in adults. Both studies used
data from the National Health and Nutrition Examination Survey study, conducted
in the United States. Rossa-Roccor et al. reported no association between MeHg
and depression, and reported that all levels of MeHg in blood samples were below
the US Environmental Protection Agency’s reference dose of 5.8 µg/L. Berk et al.
reported an inverse association between Hg and depression; however, the levels of
Hg in blood were not provided. Thus, neither of the two studies provided evidence
that Hg is associated with increased depression in adults.

5.2.1.1.3 Neurotoxicity in adults

Nakamura et al. (2014) performed a cross-sectional analysis of adults living in towns
with traditional whaling practices. Hair MeHg levels were 8.2 times higher than the
general Japanese population, at 14.9 pg/g (geometric mean), 1.1–101.9 (range), with
a sample size of 194. The authors found no correlations between hair Hg levels and
any of the neurological outcomes measured, including motor function, reactions
and reflexes, nerve signalling, hearing loss and brain morphology.

177
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

5.2.1.1.4 Neurodevelopment
Twenty-eight articles were identified that investigated neurodevelopmental
outcomes in children. Nineteen of the articles reported one or more significant
associations with Hg, whether positive or negative. Brief descriptions of the articles
reporting significant findings are provided in the following sections (sorted by the
ten neurodevelopmental categories). The health outcomes for ASD and ADHD
were topics of systematic reviews (above), but no additional original articles were
found that were not included in the systematic reviews.

5.2.1.1.4.1 Behaviour
Eleven articles that included behaviour as an outcome potentially associated with
exposure to Hg were identified. Of these, nine reported no associations with adverse
behaviour outcomes (Cao et al., 2010; Marques et al., 2011; Van Wijngaarden et al.,
2013a; Boucher et al., 2014a; Strain et al., 2015; Golding et al., 2016a; Golding et al.,
2016b; Barbone et al., 2020 and Myers et al., 2020). Concerning the remaining two
studies, Ng et al. (2013) and Al-Saleh et al. (2016), Ng et al. stratified participants
according to whether these were carriers of the Apolipoprotein E epsilon 4 allele,
a risk factor associated with Alzheimer’s disease. Among carriers of the allele, the
authors reported that prenatal Hg exposure was associated with all tested aspects of
development, but especially, with the social developmental quotient as a behaviour
outcome. For their part, Al-Saleh et al. investigated developmental delays using
two composite metrics that each included behaviour outcomes as one of several
outcomes. Significant effects were not parsed out for the different outcome
categories; therefore, an association with behaviour specifically cannot be discerned.

5.2.1.1.4.2 Brain morphology

One article was identified that assessed brain morphology in children and
its association with Hg (Nišević et al., 2019). This study reported two brain
morphometrics (cerebellum length and superior frontal gyrus) to be different
between exposed (n = 19, cord blood Hg > 5.8 µg/L) and unexposed (n = 29, cord
blood Hg < 5.8 µg/L) groups. The authors did not suggest whether the measured
differences were to be considered adverse outcomes or not.

5.2.1.1.4.3 Intelligence/cognition
Ten articles considered intelligence and/or cognition and associations with Hg in
children: Cao et al., 2010; Ng et al., 2013; Van Wijngaarden et al., 2013a; Boucher
et al., 2014a; Choi et al., 2014; Wang et al., 2014; Jacobson et al., 2015; Debes et al.,
2016; Al-Saleh et al., 2020 and Llop et al., 2020.
Three of these articles (Jacobson et al., Debes et al. and Al-Saleh et al.) reported
negative associations between Hg exposure and child intelligence specifically. Jacobson
et al. reported an association of prenatal Hg exposure with poorer performance on a
school-age intelligence quotient assessment. Children with cord Hg ≥ 7.5 µg/L were
four times as likely to have an intelligence quotient score < 80, the clinical cut-off for

178
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

borderline intellectual disability. Debes et al. reported a follow-up study on a Faroe

Islands birth cohort, at age 22. Within a battery of over 30 tests, significant negative
associations were primarily found in tests that assessed crystalized intelligence/
verbal comprehension. Finally, Al-Saleh et al. reported an association between lower
intelligence scores with both higher infant urine Hg, and higher mothers’ blood MeHg.
Two of the ten articles assessing intelligence and/or cognition however, Wang et al.,
2014 and Llop et al., 2020, found positive associations with Hg levels. Llop et al.,
2020 reported that higher hair total mercury (THg) was associated with all scales of
an assessment that considered general cognition. Wang et al. also reported positive
associations with Hg in a suite of assessments that included intelligence/cognition.
The authors of both studies suggested that the positive effects related to higher Hg
levels may be caused by a higher fish or seafood intake.

5.2.1.1.4.4 Memory and learning

Six studies were identified that investigated associations between Hg exposure and
memory and learning: Boucher et al., 2014a; Choi et al., 2014; Orenstein et al., 2014;
Wang et al., 2014; Debes et al., 2016 and Llop et al., 2020. Three of these studies
reported significant negative effects in memory and learning. Boucher et al. used
several assessments and reported that higher prenatal Hg was associated with poorer
performance on the A-not-B test – a test that depends on working memory and
is an early measurement of executive function. Choi et al. reported that prenatal
exposure was associated with a deficit in recall for short delay at 7 years of age, and
there were marginal associations with verbal learning. In both studies, the significant
associations were strengthened when adjusted for seafood consumption or n-3 fatty
acids. Orenstein et al. reported an adverse relationship between low-level prenatal
Hg and memory and learning, particularly for visual memory.
Two of the studies in this category reported positive associations with low levels
of Hg: Wang et al., 2014 and Llop et al., 2020 (see Section 5.2.1.1.4.3 Intelligence/
cognition). In the sixth and final study addressing memory and learning, Debes et al.
reported no effects relating to memory and learning specifically, but they reported
effects in intelligence/cognition (again see Section 5.2.1.1.4.3).

5.2.1.1.4.5 Mental and psychomotor development

Nine articles measured mental and psychomotor development and the association
with prenatal Hg exposure: Cao et al., 2010; Boucher et al., 2014a; Strain et al., 2015;
Rothenberg et al., 2016; Kim et al., 2018; Tatsuta et al., 2018; Nišević et al., 2019;
Castriotta et al., 2020 and Rothenberg et al., 2021. While this outcome category is
similar to intelligence/cognition (Section 5.2.1.1.4.3), in the present literature review
the studies were summarized under mental and psychomotor development if the
researchers used the Bayley Scales of Infant Development II. The Bayley Scales of
Infant Development II contains two parts, the mental developmental index and
the psychomotor developmental index. Five of the nine articles identified in this
category reported no significant results.

179
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Four studies, all of which were pregnancy cohort studies, reported significant results.
Kim et al. (2018) found a relationship between Hg in early pregnancy and both
motor and psychomotor development at 6 months of age, after adjusting for fish
intake, but not at 12, 24 or 26 months. Rothenberg et al. (2016) and Rothenberg et
al. (2021) published studies concerning a Chinese population whose primary MeHg
intake was from rice, not fish. At both 12 months (Rothenberg et al., 2016) and 35
months of age (Saghazadeh et al., 2017), lower scores on the mental developmental
index were associated with low levels of MeHg after adjusting for confounders. The
associations the authors reported were at MeHg levels well below most prior studies
among seafood consumers that also used the Bayley scales. Finally, Tatsuta et al.
(2018) reported that cord blood Hg was not associated with mental and psychomotor
development; however, when the data were stratified by child gender, in boys only,
Hg exposure was found to be associated with lower psychomotor development.

5.2.1.1.4.6 Motor function

Eleven articles assessed associations of Hg exposure with motor function: Marques
et al., 2011; Ng et al., 2013; Van Wijngaarden et al., 2013a; Choi et al., 2014; Wang
et al., 2014; Al-Saleh et al., 2016; Boucher et al., 2016; Debes et al., 2016; Golding
et al., 2016; Barbone et al., 2020 and Llop et al., 2020. Two of these studies reported
associations of impaired motor function with higher prenatal Hg exposure. Boucher
et al. (2016) showed in a pregnancy cohort of Inuit children which were followed up
at 11 years of age that Hg cord blood levels were associated with effects on fine motor
function before adjusting for confounders. Total blood Hg levels at 11 years of age also
were associated with effects on fine motor function, both with and without adjusting
for confounders. Barbone et al. (2020) reported that children whose mothers’ hair
MeHg was greater than 1 ppm were more likely to perform as expected or poorer
than expected in a fine motor-control assessment than children whose mothers’ hair
MeHg was below 1 ppm. Another three studies also report an association between Hg
and motor function. As already observed in the studies summarized in the sections
on memory and learning and intelligence/cognition, Wang et al. (2014) and Llop
et al. (2020) also reported that positive associations for neurological development
generally (which may include motor function) were observed with higher Hg levels
due to fish consumption. Finally, Al-Saleh et al. (2016) reported developmental delays
in a composite metric that included motor function as one of several outcomes; but
significant effects were not parsed out for the different outcome categories.

5.2.1.1.4.7 Nerve signalling

Two studies were identified focusing on nerve signalling: Boucher et al., 2010
and Yorifuji et al., 2013. Both were pregnancy cohort studies and both reported
significance with prenatal Hg exposures. Boucher et al. (2010) reported that prenatal
Hg was associated with altered attentional mechanisms that modulated early
processing of sensory information in Inuit children at 11 years of age. Yorifuji et al.
(2013) reported in Faroese children aged 3.75 years, that higher Hg concentrations,
particularly in maternal hair, were associated with prolonged latencies of visual
evoked potentials, which is a measure of optical nerve signalling to the brain.

180
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

5.2.1.1.4.8 Reactions and reflexes

The one article identified that measured reactions and reflexes, Marques et al. (2011),
reported no significant associations between Hg and these endpoints.

5.2.1.1.4.9 Speech and language

Nine studies were identified that investigated the outcome speech and language:
Marques et al., 2011; Ng et al., 2013; Choi et al., 2014; Wang et al., 2014; Al-Saleh et
al, 2016; Debes et al., 2016; Barbone et al., 2020; Llop et al., 2020 and Young et al.,
2020. Three of these studies reported significant effects associated with Hg exposure
(Choi et al., Debes et al. and Llop et al.). This outcome is related to both intelligence/
cognition and memory and learning. As such, overlapping results may be found
among these categories. Debes et al. report negative associations between prenatal
Hg and verbal comprehension, the Boston Naming Test, synonyms and antonyms
test, and the California Verbal Learning Test. Choi et al. reported marginal, yet
significant associations with verbal learning. In contrast, Llop et al. reported a
positive association between Hg and the verbal portion of an abilities assessment;
although the association was weaker after adjusting for fish intake. Finally, in a
fourth study, Al-Saleh et al. reported developmental delays in a composite metric
that includes speech and language as one of several outcomes. In that study, the type
of neurological effects was not parsed out for the different outcome categories, as
with behaviour and motor function.

5.2.1.1.4.10 Visual-motor integration

One study was identified that measured visual-motor integration: Al-Saleh et al.,
2020. This outcome is similar to, and may overlap with, psychomotor development,
motor function, nerve signalling, and reactions AND reflexes. This outcome was
reported individually because of the specificity of the metric used – the Beery Visual-
Motor Integration assessment. In a cross-sectional study of children aged 5 to 8
years, Al-Saleh et al. reported a decrease in visual–motor integration with higher
infant urine Hg levels. However, a positive association was found between visual–
motor skill and MeHg in mothers’ and infants’ hair.

181
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg”

182
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Rossa-Roccor et al., Cross-sectional Neurological, Adults, pregnant MeHg in blood: median 0.36 µg/L, IQR 0.14–0.99 Low-dose MeHg did not seem to be associated with
2021 Depression: women were depression in this study.
the United States Depressive excluded, age >18
symptoms years; n = 3 930
(56% female)
Berk et al., 2014 Cross-sectional Neurological, Adults, median THg in blood: levels not provided There was an inverse association between mercury
the United States Depression: age 46, (IQR range and depressive symptoms observed (contrary to
Depressive 31–63); n = 15 other pollutants tested). May be explained by a
symptoms 140 (51% female) protective role for fish consumption.
Nakamura et al., 2014 Cross-sectional Neurological, Adults, residents THg in hair: median 17.8 pg/g, range 1.1–102 Multivariate regression analysis demonstrated
Japan Neurotoxicity in of coastal town no significant correlations between hair mercury
adults: Motor where whaling is a levels and neurological outcomes. These findings
function tradition, age 20 to suggested that sufficient Se intake might be one of
Reactions and 85 years; n = 194 causes of the absence of adverse effects of MeHg
reflexes (40% female) exposure in this study.
Nerve signalling
Hearing loss
Brain morphology
Myers et al., 2020 Pregnancy cohort Neurological, Children, born in Prenatal: THg in maternal hair: mean 6.8 ppm, SD 4.5 At 107 months of age in the Seychelles, no clear
Seychelles Neurodevelopment: the Seychelles, Postnatal: THg in child hair at 107 months: mean 6.1 ppm, SD 3.5 pattern of adverse association was found between
Behaviour age: birth to 107 behaviours measured by the Child Behaviour
months Checklist and either prenatal or postnatal MeHg
(= 8.9 years); exposure at the levels achieved by consuming a
n = 643 (gender diet high in fish.
not provided)
Castriotta et al., 2020 Pregnancy cohort Neurological, Children, age: 40 THg in maternal blood: mean 3.4 ng/g, SD 3.8 There was no clear relation between maternal Hg
Neurodevelopment: months, ranging THg in cord blood: 5.6 ng/g, SD 4.9
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Italy (northern Adriatic exposure in pregnancy and adverse effects on

Sea) Mental and from 38 to 42 THg in breast milk: 0.36 ng/g, SD 1.49 children’s neurodevelopment at 40 months of age.
psychomotor months; n = 456
development (48% female)
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg” (cont.)

AUTHOR, YEAR HEALTH STUDY

STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Kim et al., 2018 Pregnancy cohort Neurological, Children, of women THg in maternal blood; GM (10th percentile; 90th percentile): Prenatal Hg exposure during early pregnancy was
Republic of Korea Neurodevelopment: (> 18 years) in Early pregnancy: 3.3 µg/L (1.81; 5.91) adversely associated with early neurodevelopment
Mental and early pregnancy Late pregnancy: 3.0 µg/L (1.68; 5.57) at 6 months, after adjusting for fish and n-3, n-6
psychomotor (< 20th week) Cord blood: 5.1 µg/L (2.94; 8.93) fatty acid intake.
development living in targeted
locations (Seoul,
Cheonan and
Ulsan), age 20
“TOXIC EFFECTS OF MeHg”

weeks of gestation
until 3 years; n = 1
251 (48% female)
Marques et al., 2011 Cross-sectional Neurological, Children, THg in hair: mean 4.33 µg/g, SD 1.7 Multiple regression analysis showed that children’s
Brazil (Rio Madeira Neurodevelopment: transitioning hair mercury had no impact on Gesell Development
Basin, Amazon) Behaviour from a traditional Scores, but some variables did interact significantly
Motor function lifestyle, age 1–59 with specific domains (motor and language
Reactions and months; development).
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

reflexes n = 249 (gender

Speech and not provided)
language
Boucher et al., 2010 Pregnancy cohort Neurological, Children, Inuit, THg in cord blood: mean 21.5 µg/L, range 1.8–99.3 These data suggested that prenatal MeHg exposure
Canada (Nunavik) Neurodevelopment: age at birth, 11 THg in child blood mean 4.69 µg/L, range 0.2–28.1 alters attentional mechanisms modulating early
Nerve signalling years, mean (range processing of sensory information.
10.2–12.9); n =
118 (65% female)
Boucher et al., 2016 Pregnancy cohort Neurological, Children, Inuit, THg in cord blood: mean 21.4 µg/L, range 1.0–99.3 Hg levels at school age were associated with
Canada (Nunavik) Neurodevelopment: age at birth and THg in child blood at 11 years: mean 4.8 µg/L, range 0.1–34.1 poorer fine motor function.
Motor function 11 years; n = 265
(52% female)
Boucher et al., 2014 Pregnancy cohort Neurological, Children, Inuit, THg in cord blood: mean 22.5 µg/L, SD 16.6, range 2.4–97.3 Prenatal Hg was associated with poorer
Canada (Nunavik) Neurodevelopment: age at birth, performance on A-not-B, which depends on
Behaviour 6.5 months, 11 working memory and is believed to be a precursor
Intelligence/ months, and 11 measurement of executive function; Hg was not
cognition years; n = 60–94 associated with BSID-II assessment. Statistical
Memory and (36% female) control for seafood nutrients also led to the
learning detection of stronger adverse associations between
Mental and prenatal Hg exposure and cognitive function in
psychomotor previous studies.
development

183
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg” (cont.)

184
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Jacobson et al., 2015 Pregnancy cohort Neurological, Children, Inuit, THg in cord blood: mean 21.8 µg/L, SD 17.5, range 1.0–99.3 Data suggest an association of prenatal mercury
Canada (Nunavik) Neurodevelopment: age at birth and THg in maternal hair: mean 4.9 μg/g, SD 2.8, range 1.4–15.1 exposure with poorer performance on a school-age
Intelligence/ 8.6–14.3 (mean assessment of IQ. The association was seen at
Cognition 11.3); n = 70–282 levels in the range within which many US children
(51% female) of Asian-American background are exposed.
Rothenberg et al., 2021 Pregnancy cohort Neurological, Children, rural, THg in maternal hair: median 0.40 μg/g, range 0.08–1.7 In conclusion, for young children living in rural
China (Daxin County) Neurodevelopment: age: at birth, 12, China, where rice consumption is the primary
Mental and and 36 months; source of MeHg exposure, a biomarker of prenatal
psychomotor n = 391 (49% MeHg exposure was associated with decrements
development female) in cognitive function assessed between 12 and
36 months of age. These associations were
demonstrated at MeHg exposure levels well below
most prior studies among seafood consumers, in
which the Bayley scales were assessed.
Rothenberg et al., 2016 Pregnancy cohort Neurological, Children, age: THg in maternal hair: GM 0.47 µg/g, range 0.078–1.7 For 12-month-old offspring living in rural China,
China (Daxin County) Neurodevelopment: followed-up at MeHg in maternal hair: GM 0.26 µg/g, range 0.048–1.4 prenatal methylmercury exposure was associated
Mental and birth and 12 with statistically significant decrements in offspring
psychomotor months; n = 270 cognition, but not psychomotor development. In
development (53% female) rural China, where 85% of mothers ingested rice
daily and 41% of mothers rarely or never ingested
fish, statistically significant inverse associations
were observed for 12-month-old offspring between
MDI scores and log10 hair THg, after adjustment
for fish/shellfish ingestion, rice ingestion,
maternal energy intake, and maternal/offspring
characteristics.
Nisevic et al., 2019 Pregnancy cohort Neurological, Children, age: THg in cord blood, exposed > 5.8 µg/L This analysis demonstrated the existence of
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Croatia and Italy Neurodevelopment: followed-up at THg in cord blood, unexposed < 5.8 µg/L morphological brain changes in newborns which
Brain morphology birth and at birth were prenatally exposed at mercury concentrations
Mental and and 18 months old; above 5.8 µg/L blood. There was no demonstrated
psychomotor n = 19–257 (48% correlation of THg concentration in umbilical cord
development female) blood with the neurodevelopmental scores at the
age of 18 months.
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg” (cont.)

AUTHOR, YEAR HEALTH STUDY

STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Debes et al., 2016 Pregnancy cohort Neurological, Adults, age: 22 THg in cord blood: mean 22.91 µg/L The results from age 22 years suggested that
Faroe Islands Neurodevelopment: years (sampling THg in maternal hair: mean 4.24 µg/g cognitive deficits associated with prenatal
Intelligence/ also done at birth, THg child blood, age 22: mean 2.53 µg/L methylmercury exposure remained through young
cognition 7 years and 14 THg in child hair, age 22: mean 0.68 µg/g adult age, with effect sizes somewhat lower than
Memory and years); n = 814 those observed at ages 7 and 14 years.
learning (51% female)
Motor function
Speech and
“TOXIC EFFECTS OF MeHg”

language
Choi et al., 2014 Pregnancy cohort Neurological, Children, age: Hg in cord blood: geometric mean 21.4 µg/L, range 1.90–101.8 Prenatal exposure to methylmercury was
Faroe Islands Neurodevelopment: followed-up at Hg in maternal hair: geometric mean 4.10 μg/g, range 0.32–16.3 associated with deficits at school age in domains
Intelligence/ birth and 7 years; known to be sensitive to this neurotoxicant, with
cognition n = 176 (50% associations being strengthened after fatty acid
Memory and female) adjustment.
learning
Motor function
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

Speech and
language
Barbone et al., 2020 Pregnancy cohort Neurological, Children, of THg in maternal hair; THg in child hair; MeHg in maternal hair; and MeHg in These pilot findings are suggestive of an
Italy (northern Adriatic Neurodevelopment: mothers residing child hair: Summary data are not clearly presented association between children’s fine motor skills
Sea) Behaviour in coastal fishing and their prenatal methylmercury exposure from
Motor function towns, age 3 maternal fish consumption.
Speech and months and 18–30
language months; n = 53
(47% female)
Tatsuta et al., 2018 Pregnancy cohort Neurological, Children. One Urban area: These findings suggested prenatal exposure to low
Japan (Tohoku Region) Neurodevelopment: cohort from an THg in cord blood: median 10.0 ng/g, 5–95 percentile range 4.2–22.4 levels of methylmercury may have adverse effects
Mental and urban area and THg in maternal hair: median 2.0 µg/g, 5–95 percentile range 0.9–4.4 on child development, especially in boys.
psychomotor one cohort from Coastal area:
development a coastal area, THg in cord blood: median 16.0 ng/g, 5–95 percentile range 5.6–39.3
followed-up at THg in maternal hair: median 2.6 µg/g, 5–95 percentile range 0.9–6.0
birth and age 18 THg in breast milk: median 0.8 ng/g, 5–95 percentile 0.1–1.8
months (range, 17
to 24); n = 1016
(48% female)

185
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg” (cont.)

186
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Young et al., 2020 Pregnancy cohort Neurological, Children, born in THg in maternal hair: mean 6.8 ppm, SD 4.6, range <1–26.7 No adverse associations between articulatory
Seychelles Neurodevelopment: Seychelles, age at THg in child hair: mean 6.5 ppm, SD 3.3, range <1–24.8 and phonologic speech skills and prenatal
Speech and birth and 5.5 years; MeHg exposure were detected. The findings of
language n = 544 (gender this investigation are compatible with previous
not provided) developmental assessments of Seychellois children
that have indicated no adverse effects of prenatal
MeHg exposure from fish consumption.
van Wijngaarden et al., Pregnancy cohort Neurological, Children, born in THg in maternal hair: mean 6.9 ppm There were no adverse associations between
2013 Neurodevelopment: Seychelles, age THg in child hair: mean 10.3 ppm prenatal MeHg and any of the measured endpoints.
Seychelles Behaviour prenatal until 19 The findings continue to provide no evidence
Intelligence/ years of age; for an adverse effect of prenatal MeHg exposure
cognition n = 553 (not on development in [a Seychellois] cohort that
Motor function provided) consumes fish daily.
Strain et al., 2015 Pregnancy cohort Neurological, Children, born MeHg in maternal hair: mean 3.92 ppm, range 0–31.66 No overall adverse association between prenatal
Seychelles Neurodevelopment: in Seychelles, MeHg exposure and neurodevelopmental outcomes
Behaviour age birth and at 20 months of age.
Mental and 1.66 years; n = 1
psychomotor 265 (gender not
development provided)
Al-Saleh et al., 2016 Cross-sectional Neurological, Children, age THg in maternal urine: mean 1.73 µg/L In summary, both the DDST-II and PEDS tests
Saudi Arabia Neurodevelopment: 2–12 months; n THg in maternal hair: mean 1.79 µg/g dw showed some evidence of developmental delays
Behaviour = 245–944 (48% THg in breastmilk: mean 0.97 µg/L in infants that were related to some measures of
Motor function female) THg in maternal blood: mean 0.88 µg/L Hg exposure, despite the low levels of Hg and
Speech and THg in infants’ urine: mean 0.90 µg/L regardless of the infant’s breastfeeding status.
language THg in infants’ hair: mean 1.17 µg/g dw
MeHg in maternal hair: mean 0.20 µg/g dw
MeHg in infants’ hair: mean 0.14 µg/g dw
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Al-Saleh et al., 2020 Cross-sectional Neurological, Children, who THg child urine: median 0.359 µg/L, range 0.010– 5.641 The results showed that early exposure to mercury
Saudi Arabia Neurodevelopment: had in a previous MeHg child hair: median 0.182 µg/L, range 0.003– 4.470 measured in infants’ urine had an adverse
Intelligence/ study failed association with the performance of children on a
cognition neurodevelopment nonverbal IQ test and on the integration of their
Visual-motor screening tools visual and motor abilities. Whereas methylmercury
integration and/or had in mother’s blood was inversely associated with
elevated mercury, children’s nonverbal IQ, methylmercury in the hair
age 5–8 years, of mothers and their infants was associated with
mean 6.81, SD enhanced visual-motor skills.
0.64; n = 82 (56%
female)
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg” (cont.)

AUTHOR, YEAR HEALTH STUDY

STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Ng, et al., 2013 Pregnancy cohort Neurological, Children, age 2 THg in cord blood: 80 measurements were <12 µg/L, and 88 measurements Prenatal Hg exposure was associated with
Taiwan Neurodevelopment: years; n = 168 (44 were >12 µg/L significant adverse effects on cognition, social and
Behaviour percent female) whole neurodevelopment development quotients
Intelligence/ among subjects who have at least one Ap oe e4
cognition allele. The inter-population difference prevalence
Motor function of the alleles might be one of explanations for
Speech and the inconsistent findings related to prenatal Hg
language exposure and neurodevelopment.
“TOXIC EFFECTS OF MeHg”

Golding, et al., 2016 Pregnancy cohort Neurological, Children, age THg in maternal blood: median 1.86 µg/L, range <LOD-12.8 There were no adverse effects of maternal prenatal
(Prenatal mercury Neurodevelopment: 47 months, 81 mercury levels on the behaviour of the offspring.
exposure and offspring Behaviour months, 7–8 years, A similar lack of relationship was found when the
behaviour in childhood 10–11 years, analyses were confined to those offspring whose
and adolescence) 11–12 years, 13 mothers had eaten fish in pregnancy, and no
the United Kingdom years, 16–17 consistent differences were found between the fish
of Great Britain and years; n = 1 599–2 and non-fish eaters.
Northern Ireland (Avon) 776 (gender not
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

provided)
Golding, et al., 2016 Pregnancy cohort Neurological, Children, age 6, THg in prenatal blood: range 0.17–12.76 µg/L No evidence of adverse associations was found
(Associations between Neurodevelopment: 18, 30, and 42 between maternal prenatal blood mercury and child
prenatal mercury Behaviour months; n = 2 development between 6 and 42 months of age.
exposure and early child Motor function 394–3 264 (gender
development in the not provided)
ALSPAC study)
the United Kingdom
of Great Britain and
Northern Ireland (Avon)
Orenstein et al., 2014 Pregnancy cohort Neurological, Children, whose THg in maternal hair: mean 0.6 μg/g, range, 0.3–5.1 These results support an adverse relationship
the United States Neurodevelopment: mother’s residence between low-level prenatal MeHg exposure and
(New Bedford, Memory and is close to a childhood memory and learning, particularly visual
Massachusetts) learning PCB-contaminated memory.
harbour, age
perinatal and 8
years (range, 7–11
years); n = 393
(50% female)

187
 TABLE 5.2 OVERVIEW OF PRIMARY STUDIES WITH NEUROLOGICAL OUTCOMES FOR THE REVIEW OF “TOXIC EFFECTS OF MeHg” (cont.)

188
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Cao et al., 2010 Cohort Neurological, Children, living in THg in pre-, post-, treatment and placebo group blood means were between At the present postnatal MeHg exposure level of US
the United States Neurodevelopment: poverty, who had 0.52 and 0.65 µg/L, with 95% CIs ranging from 0.49 to 0.61 children, adverse effects on children’s cognition and
(Philadelphia, PA; Behaviour blood Pb levels behaviour were not detectable, since they did not
Newark, NJ; Cincinnati, Intelligence/ between 20–44 appear in a longitudinal study among children living
OH; and Baltimore, MD) cognition mg/dL, age 2, 5 in poverty in the United States.
Mental and and 7 years; n =
psychomotor 531–767 (approx.
development 44% female)
Yorifuji et al., 2013 Pregnancy cohort Neurological, Children, age birth THg in cord blood: GM 22.8 µg/L, IQR 13.7–41.2 The present study found that higher mercury
Faroe Islands Neurodevelopment: and 3.75 years; THg in maternal hair: GM 4.6 µg/L, IQR 2.7–8.2 concentrations were associated with the prolonged
Nerve signalling n = 139 (52% latencies of visual-evoked potentials, in particular
female) higher maternal hair mercury was associated with
the prolonged N145 latency.
Llop et al., 2020 Pregnancy cohort Neurological, Children, age THg in hair: GM 0.98 µg/g, 95% CI 0.94–1.03, SD 1.42 There was a positive effect of Hg on the
Spain Neurodevelopment: 4.1–6.4 years; neurophysiological development for all measured
Intelligence/ n = 1251 (53% scales. However, the effect was attenuated when
cognition female) adjusted by children’s fish intake or theoretical
Memory and scenarios of low precision in fish intake and Hg
learning measurements.
Motor function
Speech and
language
Wang et al., 2014 Cohort Neurological, Children, urban MeHg in blood: GM 0.56 µg/L, 95% CI 0.52–0.59 The relatively low MeHg exposure in US
the United States Neurodevelopment: inner-city children school-aged children from this population (lead
(Philadelphia, PA; Intelligence/ exposed to lead, exposed) had no detectable adverse effect on
Newark, NJ; Cincinnati, cognition age 12 months neuropsychological development. The positive
through 7 years; n
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

OH and Baltimore, MD) Memory and associations observed between MeHg and
learning = 613–780 (45% neurodevelopment may reflect exposure to
female) beneficial polyunsaturated fatty acids from seafood.
Motor function
Speech and
language

Notes: THg: total mercury; BSID II: Bayley Scales of Infant Development-Second Edition; IQR: interquartile range; IQ: intelligence quotient; DDST-II: Denver Developmental Screening Test, version II; PEDS:
Parents’ Evaluation of Developmental Status; PCB: polychlorinated biphenyl; Pb: Lead; CI: confidence interval; GM: geometric mean; SD: standard deviation.
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

5.2.1.2 Previous reports

In their 2012 report, EFSA summarized recent literature by the time of exposure.
First, for prenatal exposures, findings were reported by region. In 14 years of
data from the Faroe Islands, results consistently pointed to a detrimental effect
of Hg on some neurological outcomes. In Seychelles, no consistent findings were
found, although the newest reports do report a negative association between Hg
and neurological endpoints, when maternal fish intake is included as a confounder.
Among other regions, few studies found associations at Hg levels that were lower
than those reported in the Faroe Islands or Seychelles. As such, no additional
conclusions were drawn. EFSA did not reach any conclusions regarding postnatal
Hg exposure and neurological effects in childhood, primarily because of inconsistent
findings. Finally, studies addressing neurotoxicity in adults revealed no relevant
associations with Hg exposures at low levels.
The 2022 VKM report considered two systematic reviews on autism and/or AHDH,
and their associations with Hg. Both reviews (one of which – Yoshimasu et al.,
2014 – is summarized in Section 5.2.1.1 Systematic reviews) were inconclusive. A
meta-analysis of inhibitory control in humans was also inconclusive for MeHg. The
remaining reviews on neurological outcomes in VKM 2022 are summarized in the
systematic reviews described previously.

5.2.2 CARDIOVASCULAR OUTCOMES

5.2.2.1 Systematic reviews
The literature search identified four systematic reviews related to Hg and
cardiovascular disease (Table 5.3). Two of them (Chowdhury et al., 2018 and Hu et
al., 2021) studied the association between Hg and various cardiovascular endpoints
in a general adult population. The two other (Hu et al., 2018 and Gallego-Vinas et
al., 2019) examined indicators related to cardiovascular disease, with one focusing
on blood pressure and hypertension in a general adult population (Hu et al.) and
the other focusing on blood pressure in children and adolescents (Gallego-Vinas et
al.). Of the four reviews, three generally concluded that there was no association
with Hg.
A systematic review by Chowdhury et al. (2018), which investigated the risk
of cardiovascular disease in relation to exposure to toxic environmental metal
contaminants, included 37 articles, of which nine were related to Hg. Three of the
articles were cohort studies and six were case-control studies. The authors concluded
that Hg exposure was not associated with cardiovascular disease. Hu et al. (2021)
reviewed 14 studies to investigate the effects of Hg exposure on cardiovascular
disease and mortality. Nine of these studies were cohort studies, four were case-
control studies, and one was a cross-sectional study. The author concluded that
exposure to Hg was associated with certain cardiovascular endpoints, which are
summarized in Table 5.3.

189
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Gallego-Vinas et al. (2019) reviewed eight articles in a systematic review on Hg and

blood pressure among children and adolescents. The review comprised five cohorts,
one cross-sectional study and one case-control study. The results of the studies were
inconsistent, and the authors stated that no conclusion could be established. Hu et
al. (2018) reviewed 29 studies, including 27 cross-sectional studies, one cohort study
and one case-control study, in a meta-analysis assessing the effects of Hg exposure
on blood pressure and hypertension. A significant positive association between Hg
and hypertension and between Hg and blood pressure was identified.

190
 TABLE 5.3 OVERVIEW OF SYSTEMATIC REVIEWS WITH CARDIOVASCULAR OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION

TITLE OUTCOME STUDIES DETECTED
Chowdhury et al., 2018 Cardiovascular; Total n = 9: Blood Hg ranging Meta-analysis: Cardiovascular disease (n = 11 410): RR (95% CI): 0.94 (0.66– Mercury exposure was not associated with any
Environmental toxic Total n = 3 cohort from 0.004 to 3.5 1.36). Coronary heart disease (n = 9 169): RR (95% CI): 0.99 (0.65–1.49) cardiovascular risk.
metal contaminants and cardiovascular studies and µg/L.
risk of cardiovascular and coronary heart n = 6 case-control
disease: systematic disease studies
review and meta-
analysis
Gallego-Viñas et al., Cardiovascular; Total n = 8: Prenatal exposure 4/7 studies found a positive association between chronic Hg exposure and
“TOXIC EFFECTS OF MeHg”

The results of the studies were inconsistent. As

2018 Blood pressure n = 6 cohort (maternal): blood pressure. Three of these studies assessed Hg exposure in prenatal life. such, with the current available scientific evidence,
Chronic mercury studies, Arithmetic mean of it cannot be concluded that an association exists.
exposure and blood n = 1 case-control total Hg:
pressure studies, Range hair Hg:
in children and n = 1 cross- 565–700 µg/kg.
adolescents: a sectional study Blood Hg: 1.86
systematic review µg/L
Cord blood:
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

22.6–31.77 µg/L
Postnatal exposure
(child):
Arithmetic mean of
total Hg:
Blood Hg: 0.10–8.1
µg/L
Hair Hg: 960 µg/kg

191
 TABLE 5.3 OVERVIEW OF SYSTEMATIC REVIEWS WITH CARDIOVASCULAR OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg” (cont.)

192
AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION
TITLE OUTCOME STUDIES DETECTED
Hu et al., 2018 Cardiovascular; Total n = 29: n = 11 studies Meta-analysis: The pooled OR for hypertension, comparing the highest and The association between Hg exposure and the
Mercury Exposure, Blood pressure and n = 1 cohort study were conducted at lowest mercury exposure categories, was 1.35 (95% CI: 0.99, 1.83) for prevalence of hypertension was non-linear,
Blood Pressure, hypertension n = 1 case-control low to moderate populations with hair mercury ≥2 µg/g in comparison with the OR of 1.12 (95% with no association in populations exposed to
and Hypertension: A study, and mercury exposure CI: 0.82, 1.52) for populations with hair mercury <2 µg/g. A nonlinear dose– low-to-moderate mercury (hair Hg <2 µg/g)
Systematic Review and n = 27 cross- levels (mean Hg response relationship with an inflection point at 3 µg/g was identified, for both and evident association in populations exposed
Dose–response Meta- sectional studies concentration hypertension and systolic blood pressure. Hair Hg concentration higher than to high mercury (hair Hg ≥2 µg/g). However,
analysis of the highest 2 µg/g is associated with a 59% increase in OR for hypertension, an increase the interpretation of a causal association of Hg
exposure group ≤ of 2.20 mmHg and 1.24 mmHg in systolic blood pressure and diastolic blood exposure and hypertension is limited by the
2 µg/g in hair or pressure, respectively. cross-sectional design of original studies. Current
equivalent). n = evidence suggests that hair Hg concentration of
18 studies were 2–3 µg/g might be considered the threshold of
conducted at high Hg’s toxic effect on hypertension. Heterogeneity
mercury exposure was observed for Hg species and exposure groups
levels (mean Hg across different studies.
concentration
of the highest
exposure group
>2 µg/g in hair or
equivalent).
Hu et al., 2021 Cardiovascular; Total n = 14: Highest exposure Meta-analysis: Chronic exposure to Hg was associated with an
Mercury exposure, Cardiovascular n = 9 cohort levels in the Total cardiovascular disease (n = 10 351): RR = 0.93, 95% CI: 0.80, 1.08). increased risk of all-cause mortality and fatal/
cardiovascular disease, disease and studies, studies varied Ischemic heart disease (n = 18 312): RR = 1.21, 95% CI: 0.98, 1.50. Stroke non-fatal ischemic heart disease. The risk of
and mortality: A mortality; Fatal/ n = 4 case-control from hair Hg (n = 18,428): RR = 1.03, 95% CI: 0.87,1.23. multiple cardiovascular endpoints starts to increase
systematic review and non-fatal studies, and concentrations of 1 All-cause mortality (n = 3 254): RR = 1.21, 95% CI: 0.90, 1.62. consistently at a hair Hg concentration of 2 μg/g.
dose-response meta- ischemic heart n = 1 cross- to 30 µg/g.
A J-shaped relationship between Hg exposure and fatal/non-fatal
analysis disease, stroke, sectional studies
cardiovascular outcomes was observed, with turning points at 1 μg/g hair
all cardiovascular
Hg for ischemic heart disease and 2 μg/g for stroke and all cardiovascular
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

disease,
disease.
cardiovascular
disease mortality,
all-cause mortality.
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

5.2.2.2 Primary studies

The search identified four primary studies assessing the effect of Hg on outcomes
related to cardiovascular disease (Table 5.4). Of these studies, two (Tajik et al., 2016
and Chan et al., 2021) reported a significant association, while the other two (Chen
et al., 2018 and Tajik et al., 2018) found no association. One study reported on the
association between Hg and ischemic stroke, while the other three studies looked
at indicators related to cardiovascular disease.
A cohort study conducted by Chen et al. (2018) did not support the posited
hypothesis that Hg exposure at low-to-moderate levels was associated with incidence
of ischemic stroke within a population of adults above 44 years of age.
Chan et al. (2021) found that in children, prenatal MeHg exposure, but not recent
MeHg exposure, was adversely associated with cardiac autonomic function, as
measured by heart-rate variability. No associations were found between postnatal
MeHg exposure and heart-rate variability or postnatal MeHg exposure and blood
pressure. Tajik et al. (2016) studied the association of serum long-chain n-3 PUFA
and hair Hg with resting heart rate, peak heart rate during exercise and heart rate
recovery after exercise in men of 42 to 60 years. Higher hair Hg content was
associated with a trend towards lower peak heart rate after adjusting for the long-
chain n-3 PUFA. Hair Hg concentration was not associated with resting heart rate
or heart-rate recovery. Further, Tajik et al. (2018) found no association between hair
Hg concentrations and exercise cardiac power in men 42 to 60 years.

193
 TABLE 5.4 OVERVIEW OF PRIMARY STUDIES WITH CARDIOVASCULAR OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg

194
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Chen et al., 2018 Case-cohort Cardiovascular, Adults, older than THg in serum: median 0.03 µg/dL The present study does not support the hypothesis
the United States Cardiovascular: 44 years; Blacks that mercury exposure is associated with the
(southeastern states) Ischemic stroke were intentionally incidence of ischemic stroke within a population
oversampled, age with low-to-moderate level of exposure.
mean 65 years;
SD 9.4; n = 2 494
(55% female)
Chan et al., 2021 Pregnancy cohort Cardiovascular, Children, age 8.1 Cord blood Hg: mean 50.12 nmol/L, SD 23.9 Prenatal Hg exposure is adversely associated with
Hong Kong Cardiovascular: years ± 0.9; n = Child blood Hg: mean 15.94 nmol/L, SD 9.94 cardiac autonomic function as measured by heart
cardiac autonomic 604 (45% female) rate variability. In comparison, no associations were
function and blood found between postnatal MeHg exposure and heart-
pressure rate variability and blood pressure.
Tajik et al., 2016 Cohort Cardiovascular, Adults, men, age THg in hair: mean 1.94 μg/g, range 0 to 15.67 Hair Hg was not associated with the measured
Finland Cardiovascular: 42–60 years; n = 1 cardiac outcomes. Higher circulating concentrations
exercise cardiac 672 (0% female) of long-chain n-3 PUFA, mainly a marker of
power; VO2 max; fish consumption in this study population, were
maximal systolic associated with higher exercise cardiac power and
blood pressure VO2 max in middle-aged and older men.
during exercise
Tajik et al., 2018 Cohort Cardiovascular, Adults, men free THg in hair: mean 1st quartile 1.09 µg/g, mean 4th quartile 2.74. Higher hair Hg content showed a trend towards
Finland Cardiovascular: of cardiovascular lower peak heart rate after adjusting for the long-
resting heart rate, disease, age chain n-3 PUFA (P trend = 0·05), but this only
peak heart rate 42–60; n = 1 008 slightly attenuated the associations of the serum
during exercise and (0% female) long-chain n-3 PUFA with heart rate.
heart-rate recovery
after exercise
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

5.2.2.3 Previous reports

EFSA, 2012 included ten articles on Hg and cardiovascular disease. Two studies
(Guallar et al., 2002 and Virtanen et al., 2005) found an increased risk of acute
coronary event or myocardial infarction with higher Hg concentrations. Two
studies (Ahlqwist et al., 1999 and Yoshizawa et al., 2002) did not find a statistically
significant association between myocardial infarction and Hg concentrations. Two
newer studies (Wennberg et al., 2007 and Mozaffarian et al., 2011) did not indicate
any association between stroke and Hg exposure. Three studies (Hallgren et al.,
2001; Wennberg et al., 2011 and Bergdahl et al., 2013) showed associations between
Hg and decreased risk of myocardial infarction. Finally, one study (Mozaffarian et
al., 2011) showed no association between Hg and the risk of cardiac disease.
The 2012 EFSA assessment included 17 articles on indicators related to cardiovascular
disease: Dórea et al., 2005; Pedersen et al., 2005; Vupputuri et al., 2005; Fillion et
al., 2006; Valera et al., 2008; Bautista et al., 2009; Choi et al., 2009; Valera et al.,
2009; Lim et al., 2010; Yaginuma-Sakurai et al., 2010; Valera et al., 2011a; Valera
et al., 2011b; Nielsen et al., 2012; Olsén et al., 2012 and Valera et al., 2013. Of
these, 14 articles focused on blood pressure, 1 on hypertension, 7 on hearth-rate
variability, 1 on vasodilating function, 1 on brainstem auditory-evoked potentials, 1
on carotid intima-media thickness, and 1 on vasodilation function. EFSA concluded
that the observations related to myocardial infarction, heart-rate variability and
possible blood pressure were of potential importance; however, the results were
not conclusive. EFSA specified that it is important to take the beneficial effects of
fish consumption into account when studying the associations between Hg and
cardiovascular outcomes.
The 2022 VKM assessment identified one systematic review on Hg and cardiovascular
disease (Hu et al., 2021). This study was also included in the systematic reviews and
has been summarized in a prior section. The 2022 VKM assessment identified two
reviews on blood pressure and hypertension, which were also included in the review
and are summarized in a prior section (Hu et al., 2018 and Gallego-Vinas et al., 2019).
The 2022 VKM assessment identified two reviews on heart rate variability
(Gribble et al., 2015 and Karita et al., 2018). Neither of these was included in the
present literature analysis due to being rated “low” and “very low” confidence in
the risk-of-bias assessment (Appendix 5, Table A5.6). According to the 2022 VKM
report, neither of these two studies found associations between autonomic heart
control and MeHg exposure.

5.2.3 GROWTH
5.2.3.1 Systematic reviews
The literature search conducted identified one recent systematic review (Dack
et al., 2021) (Table 5.5) on Hg and prenatal growth (birth weight, birth length
and head circumference), including 11 prospective and 16 cross-sectional studies,
comprising participants from 17 countries. According to the authors, many of the

195
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

studies included showed no strong evidence of an effect. Inverse associations with

birth weight were reported in a significant minority with the highest mean Hg
concentrations. Overall, the systematic review by Dack et al. did not identify strong
evidence that Hg exposure led to impaired prenatal growth, but some evidence of
a negative association with Hg was noted.

196
 TABLE 5.5 OVERVIEW OF SYSTEMATIC REVIEWS WITH GROWTH OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION

TITLE OUTCOME STUDIES DETECTED
Dack et al., 2021 Growth; Birth Total n = 27: Blood levels: Birth weight: A total of 27 studies with inconsistent results, but populations Most studies showed no strong evidence of an
Mercury and Prenatal outcomes; Foetal n = 11 prospective Range mean Hg: with high Hg exposure and high-quality studies were more likely to report a effect, but a significant minority report inverse
Growth: A Systematic growth outcomes: studies, and n = 0.91–4.9 µg/L. negative association. associations with birth weight, particularly studies
Review Birth weight, birth 1 cross-sectional Hair levels: Birth length: 10/14 studies found no evidence of association. of populations with the highest mean mercury
length, and head studies Range mean Hg: Head circumference: 13/14 studies found no evidence of association. concentrations.
circumference 320–607 µg/kg
Umbilical cord:
Range mean Hg:
“TOXIC EFFECTS OF MeHg”

0.91-21.0 µg/L
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

197
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

5.2.3.2 Primary studies

Six articles considering the association of Hg exposure with growth were identified
(Table 5.6). Three cross-sectional studies assessed measurements at birth only (García-
Esquinas et al., 2013; Miyashita et al., 2015 and Tang et al., 2016). Two of the three cohort
studies addressed prenatal growth using ultrasound measurements (Drouillet-Pinard
et al., 2010 and Ballester et al., 2018), and one of these (Drouillet-Pinard et al.) combined
it with measurements at birth. The third cohort study (Papadopoulou et al., 2021)
followed the participating children from one month to eight years of age. Among all six
growth studies, four report at least one association between Hg and a growth outcome.
Among the three cross-sectional studies on newborns, one reported an association
between Hg and growth (Tang et al., 2016), while two did not (García-Esquinas
et al., 2013 and Miyashita et al., 2015). No associations were observed between
Hg concentrations in cord, maternal and paternal blood and weight, length or
Apgar scores in newborns in Spain (García-Esquinas et al., 2013). In newborns
in Japan, no associations between Hg concentrations and birth weight, length,
chest circumference and head circumference were observed (Miyashita et al.,
2015). For newborns from certain areas of China, Tang et al. (2016) concluded
that their results suggested that Hg exposure had potential adverse effects on birth
outcomes, as they found that birth weight was significantly reduced in the second
tertile of Hg concentration in umbilical cord serum after adjusting for all covariates.
No association was found in multivariable linear regression for Hg.
Regarding prenatal growth in the cohort studies, Ballester et al. (2018) followed
a Spanish pregnancy cohort and found that prenatal Hg exposure did not affect
foetal parameters investigated at 12, 20 and 34 weeks of gestation, except for
being associated with a small reduction of biparietal diameter early in pregnancy.
Drouillet-Pinard et al. (2010) assessed prenatal and growth measurements at birth
in a pregnancy cohort in France. In this study, a negative association was observed
between maternal Hg exposure and biparietal diameter, measured at 20 to 24 weeks
of gestation. Otherwise, no association between maternal level of total hair Hg and
ultrasound measures or newborn anthropometric measures were found.
In the only cohort study identified following postnatal growth (Papadopoulou et al.,
2021), Norwegian mother–child pairs were enrolled. High Hg exposure (top decile)
was associated with a reduction in children’s weight growth trajectory, compared
to lower exposure, especially for girls.
In summary, two studies found no association between prenatal exposure and
growth in newborns, while one study suggested a potential association of Hg in
cord serum with birth weight in newborns. For prenatal growth, both studies found
a small reduction of biparietal diameter in early to mid-pregnancy but no association
with other measured foetal growth parameters. The only longitudinal study on post-
natal growth found an association between high Hg exposure and the weight-growth
trajectory in children. However, for seafood consumption, the authors concluded
that no detrimental effects were found. The results of the individual studies overall
were similar to those reported in the systematic review included in this section.

198
 TABLE 5.6 OVERVIEW OF PRIMARY STUDIES WITH GROWTH OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

AUTHOR, YEAR HEALTH STUDY

STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Garcia-Esquinas et al., Cross-sectional Growth, Foetal Children, of THg in cord blood: GM 6.7 µg/L, IQR 5–11 No associations were observed between [mercury]
2013 and child growth: mother/father/ THg in maternal blood: GM 3.9 µg/L, IQ range 2.4–6.8 levels and newborn’s weight, length or Apgar
Spain (Madrid) newborn size and newborn sets, THg in paternal blood: mean 5.4 µg/L, IQ range 3.5–9.1 (appearance, pulse, grimace, activity, respiration)
health status where the birth scores. The findings do not support an association
occurred at a between total mercury in cord blood and foetal
public hospital, health.
age at birth; n =
112 (gender not
“TOXIC EFFECTS OF MeHg”

provided)
Miyashita, et al., 2015 Cross-sectional Growth, Foetal Children, newborn, THg in maternal hair: range 0.24–4.73 µg/g There were no associations between the
Japan (Hokkaido) and child age at birth; n = concentrations of hair Hg and newborn
growth: newborn 367 (53% female) anthropometric measurements of birth
anthropometric weight, length, chest circumference and head
measurements, circumference in the multiple linear regression
babies born small models, with or without adjustment for factors.
for gestational age Although, the incidence of babies born small for
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

(SGA) gestational age by weight may reduce with higher

concentrations of Hg in hair.
Tang, et al., 2016 Cross-sectional Growth, Foetal Children, THg in cord blood serum: median 21.94 µg/L, IQR: 15.10, 27.64 Results suggested that Pb and Hg exposure
China (Shengsi Island) and child Chinese and fish has potential adverse effects on birth outcomes
growth: infant consuming, age in Chinese fish consumers from Yangtze River
anthropometric at birth; n = 103 outlet and Hangzhou Bay estuary regions. The
measurements, (42% female) second tertile Hg concentration in the umbilical
including birth cord serum was significantly correlated with birth
weight, length weight after adjusting for all covariates, but not the
and head third tertile. From multivariable linear regression
circumference for Hg, no significant associations were found
between exposure and birth weight, height, head
circumference, and gestation age in both models
adjusted for different covariates
Ballester, et al., 2018 Pregnancy cohort Growth, Foetal Children, age: THg in cord blood: GM 8.2 µg/L The longitudinal assessment of the association of
Spain and child growth: prenatal at 12, prenatal Hg exposure with foetal growth showed
longitudinally 20 and 34 weeks a small reduction of biparietal diameter early
measured foetal gestation; n = 1 in pregnancy, but no significant changes were
biometry (foetal 552-1 560 (48% observed in other foetal parameters.
growth) female)

199
 TABLE 5.6 OVERVIEW OF PRIMARY STUDIES WITH GROWTH OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

200
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Drouillet-Pinard et al., Pregnancy cohort Growth, Foetal and Children, age THg in maternal hair: median 0·52 µg/g, IQR 0.30–0.82, SD 2.6 In conclusion, the data do not support a detrimental
2010 child growth: foetal 20–24 and 30–34 THg in child hair: median 0·38 µg/g, IQR 0.30–0.43, SD 0.32 effect of low maternal Hg contamination on
France (Nancy and growth weeks of gestation, birth weight or other newborn anthropometric
Poiters) and birth; n = measurements. In the whole sample of women,
109–156 (45% there was no association between maternal level of
female total hair Hg and ultrasound measures as well as
newborn anthropometric measures.
Papadopoulou et al., Pregnancy cohort Growth, Foetal and Children, age 1 THg in maternal blood: median 1.03 mg/L, 0.96 IQR, range 0.003–12.68 High prenatal mercury exposure (top decile
2021 child growth: child month to 8 years; compared to the rest) was associated with a
Norway body mass n = 2277 (49% reduction in the child’s weight–growth trajectory
index trajectories female) (Estimates ranging from -130 g (95% CI = -247,
-12 g) at 18 months to -608 g (95% CI = -1.102,
-113 g) at 8 years. Overall, the findings do
not provide evidence of detrimental effects of
seafood consumption on child growth from birth to
childhood.

Notes: THg: total mercury; GM: geometric mean, IQR: interquartile range, IQ: intelligence quotient, SD: standard deviation.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

5.2.3.3 Previous reports

In the 2012 EFSA report, growth was addressed under “Developmental toxicity
other than neurotoxicity and immunotoxicity”. Of the six studies identified by
EFSA, three did not find an association between birth outcomes and Hg: Lucas
et al., 2004; Lederman et al., 2008 and Drouillet-Pinard et al., 2010. Of the three
studies that did find an association, one (Lee et al., 2010) found an association only
for a certain genotype of the mother; one (Cace et al., 2011) found an association for
cerebellum width, but not length; and one (Ramon et al., 2009) found an association
for birth weight and size for gestational age.
The 2022 VKM assessment included two systematic reviews on “Foetal growth and
birth outcomes” in their report. These reviews were also identified in the literature
search. However, one of the reviews, Saavedra et al., 2022, was graded “low” in
confidence in the risk-of-bias analysis and was excluded for further assessment and,
thus, was not mentioned earlier (Appendix 5, Table A5.6). This review concluded,
based on four studies, that Hg exposure was associated with lower birth weight. The
authors noted that Hg toxicity may sometimes be mitigated, for instance through
PUFAs in the maternal diet.

5.2.4 OTHER HEALTH OUTCOMES

5.2.4.1 Systematic reviews, primary studies and previous reports
For the topics described in the following sections (5.2.4.1.1 through 5.2.4.1.11), only
a limited number of articles could be identified for any one outcome in the literature
review. Therefore, original articles, reviews and findings from earlier reports are
described together, categorized by outcome. The main results from the group of
“other health outcomes” are summarized in Table 5.7 for the systematic reviews
and in Table 5.8 for the primary studies.

5.2.4.1.1 Diabetes and metabolic syndrome

The literature search identified one systematic review considering diabetes and
metabolic syndrome together, one systematic review on metabolic syndrome, and
two primary studies on T2D and metabolic syndrome.
In the systematic review by Roy et al. (2017), on both diabetes and metabolic
syndrome, the associations found in the 24 articles included were weak to
moderate, ranging from no association to an OR of 7.35 (1.73–31.1). Several of the
studies included found an association between increased Hg exposure and risk of
diabetes and metabolic syndrome. However, the associations were reported not to
be consistent between studies and the results from the longitudinal studies were
conflicting. The authors stated that, based on a weight of evidence approach, an
association between total Hg concentration in various matrices and the development
of either diabetes mellitus or metabolic syndrome seems to exist, but sufficient
evidence for a causal relationship was not found.

201
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Reviewing studies on metabolic syndrome, Xu et al. (2021) performed a meta-

analysis on the results of 11 studies and found that exposure to Hg was associated
with a higher prevalence of metabolic syndrome.
Tsai et al. (2019) conducted a cross-sectional study entitled Nutrition and Health
Survey in Taiwan on T2D and Hg exposure in the general population. The authors’
findings showed a significant association between elevated Hg concentration in
red blood cells and T2D prevalence. It was stated that future research, particularly
longitudinal cohort studies with suitable specimens, were needed in order to verify
the findings.
Stratakis et al. (2020) assessed the association between maternal Hg exposure
and metabolic syndrome in children. An aggregate metabolic syndrome score (a
metric considering waist circumference; systolic and diastolic blood pressures; and
levels of triglyceride, high-density lipoprotein cholesterol and insulin) was put in
association with maternal Hg exposure and high maternal Hg exposure was found
to be associated with an unfavourable metabolic profile in children.
In summary, one original article (Tsai et al., 2019) suggested an association between
Hg and diabetes and another (Stratakis et al., 2020) suggested an association with
metabolic syndrome. In addition, the existing systematic reviews identified several
studies that found an association. However, in one review it was stated that the
associations were not consistent and did not provide sufficient evidence for a causal
relationship to be established, and the other review argues for more studies in order
to be able to adjust adequately for appropriate confounders.
The 2012 EFSA assessment identified one original article (Park et al., 2009) on
metabolic syndrome, which found an association with Hg exposure.

5.2.4.1.2 Immune system

Four primary studies were identified on the possible association of Hg exposure
and immune system functions. Two of the studies (Emeny et al., 2019 and Kindgren
et al., 2019) found no association of Hg exposure with endpoints connected to the
immune function, while two others (Miyake et al., 2011 and Carrasco et al., 2021)
found indications for an association in some endpoints.
Miyake et al. (2011) assessed the occurrence of wheeze and eczema symptoms related
to allergic disorders in children of Japan in a pregnancy cohort study, finding that
neither maternal nor children’s hair Hg levels were related to the risk of wheeze
or eczema. Carrasco et al. (2021) investigated prenatal and postnatal Hg exposure
and respiratory health in a pregnancy cohort study in Spain and reported that no
association was found.
Emeny et al. (2019) investigated endpoints. including respiratory infections and
symptoms, eczema and occurrence of allergies in infants in the United States, and
related the parameters investigated to THg in the mothers’ toenails. Higher toenail
Hg in fish-consuming mothers was reported to be related to a greater RR of lower-
tract respiratory infections and respiratory symptoms of their child at 9–12 months

202
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

of age, while a reduced RR was reported to be observed among infants 0 to 4

months of age. The authors also reported that there was little to no evidence of an
association between toenail Hg and upper-respiratory infection, allergy or eczema
at any age interval.
To assess the association of heavy metal exposure with autoimmunity and the
development of juvenile idiopathic arthritis, Kindgren et al. (2019) conducted a
prospective birth cohort study in Sweden. The authors found that Hg in cord blood
was significantly higher in the juvenile idiopathic arthritis group than in the controls,
but not in the antinuclear antibodies group, and concluded that moderate exposure
to heavy metals during pregnancy and early childhood may affect the immune
system.
The 2012 EFSA assessment identified seven original articles on “immunotoxicity”.
These include Miyake et al., 2011 and Carrasco et al., 2021 (described earlier in
this section). Three further articles found no adverse association of Hg in relation
to antinuclear antibodies (Alves et al., 2006), antibodies against vaccine toxoids
(Heilmann et al., 2010), and asthma and atopic dermatitis (Grandjean et al., 2010).
One study investigating immunological status (Belles-Isles et al., 2002) found an
association for a subset of naive T-cells, and another study (Nyland et al., 2011b)
found an association with total immunoglobulins, but not with IgG. Finally, one
study (Park et al., 2011) provided evidence for an association of Hg with atopic
dermatitis.

5.2.4.1.3 Reproduction
One systematic review and one original article were identified considering the effect
of Hg exposure on reproduction generally. The systematic review (Sirohi et al.,
2021) investigated the association between environmental exposures to endocrine-
disrupting chemicals and endometriosis. Only one study within this review was
identified to report on the association between serum Hg levels and endometriosis,
and no association was found. Mocevic et al. (2013) conducted a cross-sectional
study on male partners of pregnant women in Greenland, Poland and Ukraine,
evaluating the association between serum levels of reproductive hormones and THg
in blood. The authors reported that no evidence was found that environmental Hg
exposure had effects on biomarkers of male reproductive health in the considered
population and exposures.
The 2012 EFSA assessment identified five studies on “reproductive toxicity”. Three
of these – investigating pre-term birth (Xue et al., 2007), reproductive hormones and
anovulation (Pollack et al., 2011) and endometriosis and uterine myomas (Jackson et
al., 2008) – reported not to have found any association. The remaining two studies
assessed semen parameters, and while one reported to not have found evidence
for an association with Hg exposure (Rignell-Hydbom et al., 2007), the second
one reported an association with particularly head and midpiece defects and some
motion characteristics (Choy et al., 2002).

203
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

5.2.4.1.4 Vision
In a cross-sectional study on visual acuity, and its association with Hg exposure,
Fillion et al. (2011), investigated adults from fish-eating communities in Brazil (the
Caruso project). According to the authors, the concentration of Hg in hair may be
associated with visual acuity loss in older people, but not in younger people.
The 2012 EFSA assessment identified one article (Lemire et al., 2010) that looked
at vision as an outcome. The study found that individuals with a blood Hg above
the 25th percentile had a higher prevalence of age-related cataracts, compared
to individuals below the 25th percentile, though the finding was not statistically
significant for the oldest age group (>65 years of age). Further, it was stated that the
results needed to be interpreted with caution due to the low number of participants
and cases included in said article.

5.2.4.1.5 Osteoporosis
Two articles were identified in the present literature review which assessed the
association between Hg exposure and osteoporosis by measuring bone mineral
density and THg in blood in participants of the Korea National Health and
Nutrition Examination Survey. Cho et al. (2012) studied postmenopausal women
in said cohort, Kim et al. (2016) considered middle-aged men. According to both
studies, the risk for osteoporosis was reduced with higher blood Hg concentrations.

5.2.4.1.6 Multiple sclerosis

The literature search identified one systematic review (Sarihi et al., 2021) on heavy
metal concentrations (including Hg) in multiple sclerosis patients. Of the 16 articles
included in the systematic review, data from six studies (one cohort and five case-
control studies, all from Italy or Iran) were subjected to a meta-analysis focusing
on Hg. The authors reported that no differences in Hg exposure between multiple
sclerosis patients and healthy controls were found.

5.2.4.1.7 Hypertension and renal disease

The literature search identified one original study (Sanders et al., 2019 – a cross-
sectional study on children from the United States) investigating the association
of exposure to different metals, including Hg, on hypertension and renal disease.
Survey-weighted single chemical analyses showed an increase in the estimated
glomerular filtration of 0.6 percent (95% CI: 0.1, 1.0) with each decile increase in
urine Hg. No significant association with blood Hg was observed. The authors
concluded that metals, including As, Pb, Hg, Cd and their combinations, may affect
renal parameters, but stated that potential reverse causation cannot be ruled out due
to the cross-sectional study design.

5.2.4.1.8 Thyroid hormones

The literature search identified a meta-analysis addressing the association between
Hg exposure and thyroid hormone levels (Hu et al., 2021). The analysis included

204
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“TOXIC EFFECTS OF MeHg”

13 studies, 6 of which were prospective cohort studies and 7 of which were cross-
sectional studies. The meta-analysis indicated that exposure to Hg in blood could
significantly correlate with the levels of TSH, T4, and FT4 in the general population.
However, the authors did not state if the observed changes might cause adverse
effects.

5.2.4.1.9 Pulmonary function

Pan et al. (2020) conducted a cross-sectional study among Chinese children to assess
the association between blood Hg levels and pulmonary function. After adjusting
for confounders, no association was found.

5.2.4.1.10 Cancer
Two articles, Zidane et al., 2019 and Rhee et al., 2020, considered the association
between Hg exposure and cancer. Zidane et al. conducted a case-control study on
thyroid cancer in French Polynesia measuring THg in fingernails and reported
that the concentrations were not associated with thyroid cancer risk. Rhee et al.
investigated skin cancer in adults in the United States in a cross-sectional study,
reporting higher blood THg and MeHg levels to be associated with a higher
prevalence of non-melanoma skin cancer.

5.2.4.1.11 Sexual maturation

One article was identified in the literature search (De Craemer et al., 2017), which
assessed the association between Hg exposure and sexual maturation in Belgian
adolescents participating in the Flemish Environment and Health Study. The authors
reported that no significant effect of Hg exposure on hormone levels in blood was
observed. Among several other physical maturation endpoints assessed in this study,
it was reported that only female breast development was significantly associated
with hair Hg, and the authors stated that the results for Hg should be interpreted
with caution.

205
 TABLE 5.7 OVERVIEW OF SYSTEMATIC REVIEWS WITH OTHER HEALTH OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

206
AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION
TITLE OUTCOME STUDIES DETECTED
Hu et al., 2021 Other; thyroid Total n = 13: The Hg content Meta-analysis: This meta-analysis indicates that exposure to Hg in
Association between hormones; thyroid Prospective cohort of most studies TSH (n = 10 930): Effect size (ES) (95% CI): 0.48 (0.18, 0.78). blood could significantly correlate with the levels of
mercury exposure and stimulating studies (n = 6) and included was T3 (n = 8 216): ES (95% CI): 0.20 (-0.37, 0.77). TSH, T4 and FT4 in the general population.
thyroid hormones levels: hormone (TSH), cross-sectional defined as low by fT3 (n = 8 821): ES (95% CI): 0.20 (-0.40, 0.80).
A meta-analysis triiodothyronine studies (n = 7) reported evidence, T4 (n = 6 809): ES (95% CI): -0.02 (-0.02, -0.01).
(T3) and free T3, which indicated fT4 (n = 8,726): ES (95% CI): 0.47 (0.11, 0.82)
thyroxine (T4) and a blood exposure
free T4 threshold of
approximately
equivalent to 4–5
µg/L.
Roy et al., 2017 Other; diabetes; Total n = 24: Not mentioned The associations found in the articles were weak to moderate, ranging from no Increased total Hg exposure may augment the risk
Is mercury exposure type 2 diabetes, Prospective cohort association at all to an OR of 7.35 (1.73–31.1). Several of the studies found an of diabetes and metabolic syndrome, but the lack
causing diabetes, gestational studies (n = 4, association between increased Hg exposure and risk of diabetes and metabolic of consistency of the epidemiological evidence
metabolic syndrome and diabetes, insulin nested case- syndrome, but the associations were not consistent between studies and the prevents the inference of a causal relationship.
insulin resistance? A resistance control studies (n results from the longitudinal studies were conflicting.
systematic review of the and metabolic = 3) and cross-
literature syndrome sectional studies
(n = 17)
Sarihi et al., 2021 Other; multiple Total n = 6: See results from Meta-analysis: No clear differences in Hg concentrations in
Toxic heavy metal sclerosis Cohort study meta-analysis. Cases (n = 296) vs. controls (n = 361) in pooled estimates of the weighted multiple sclerosis patients vs. healthy controls.
concentrations in (n = 1) and case- mean differences on Hg concentrations: Weighted mean differences (95% CI):
multiple sclerosis control studies -0.14 (-0.77, 0.49).
patients: A systematic (n = 5)
review and meta-
analysis
Sirohi et al., 2021 Other; Total n = 1: Not mentioned Only one study (Pollack et al., 2014) reported an association between serum
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

No association between Hg exposure and

Environmental endometriosis Cohort study (n Hg levels and endometriosis. The study did not find a statistically significant endometriosis was found.
exposures to endocrine = 1) positive association between serum and urinary levels of Hg
disrupting chemicals
(EDCs) and their role
in endometriosis: a
systematic literature
review
 TABLE 5.7 OVERVIEW OF SYSTEMATIC REVIEWS WITH OTHER HEALTH OUTCOMES FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg” (cont.)

AUTHOR, YEAR HEALTH NUMBER OF LEVEL OF Hg RESULTS OVERALL CONCLUSION

TITLE OUTCOME STUDIES DETECTED
Xu et al., 2021: Other; metabolic Total n = 11: Study Blood Hg ranging Meta-analysis: Hg concentrations were higher in metabolic syndrome patients Exposure to Hg was associated with prevalence of
Associations between syndrome type not given from 0.50 to 65.0 compared to controls: Pooled estimated effect sizes = 1.26 (95% CI: 1.06, metabolic syndrome.
metabolic syndrome and µg/L 1.48).
four heavy metals: A
systematic review and
meta-analysis

Notes: CI: confidence interval; OR: odds ratio.

“TOXIC EFFECTS OF MeHg”
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

207
 TABLE 5.8 OVERVIEW OF PRIMARY STUDIES WITH OTHER HEALTH OUTCOMES (N = 15) FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg”

208
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Tsai, et al., 2019 Cross-sectional Other outcomes, Adults, age THg red blood cells in non-type 2 diabetes: GM 13.21 ppb, 95% CI 12.42–14.04 Findings showed that elevated Hg in red blood
Taiwan Province of type 2 diabetes: for non-type 2 THg red blood cells in type 2 diabetes: GM 18.95 ppb, 95% CI 15.66–22.93 cells is significantly associated with type 2 diabetes
China plasma glucose diabetes: 41.13 ± prevalence.
levels after fasting 14.64 years and
and/or treatment for type 2 diabetes:
with hypoglycaemic 55.37 ± 12.87
medication years; n = 646
(51% female)
Stratakis, et al., 2020 Pregnancy cohort Other outcomes, Children, age at THg in maternal blood: median 2.5 µg/L, IQR 1.5-4.2 High maternal mercury exposure was associated
France, Greece, Norway, metabolic health birth until age 6 to with a metabolic profile in children. However,
Spain, the United and inflammation: 12 years; n = 805 moderate fish intake was associated with
Kingdom of Great metabolic (44% female) improvements in the metabolic health of children.
Britain and Northern syndrome score for
Ireland children

Miyake, et al., 2011 Pregnancy cohort Other outcomes, Children, age THg in maternal hair: median 1.52 µg/g, range 0.26–6.05 Neither maternal nor children’s hair mercury levels
Japan (Osaka) immune system: prenatal 2–9 THg in child hair: median 1.38 µg/g, range 0.13–9.51 were related to the risk of wheeze or eczema after
allergic disorders, months gestation; adjustment for potential confounding variables.
wheeze and and 16–24
eczema months and 29–39
months; n = 582
(47% female)
Carrasco, et al., 2021 Pregnancy cohort Other outcomes, Children, age at THg in cord blood: GM 8.23 µg/L, IQR 9.00 No associations were found between prenatal and
Spain immune system: birth and 4.4 years THg in child hair: GM 0.97 µg/L, IQR 1.04 postnatal Hg exposure and respiratory and allergy
respiratory health ± 0.2; n = 1 347–1 problems, although these associations could be
868 (48% female) modulated by diet and other pollutants, especially
during prenatal period.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION
 TABLE 5.8 OVERVIEW OF PRIMARY STUDIES WITH OTHER HEALTH OUTCOMES (N = 15) FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg” (cont.)

AUTHOR, YEAR HEALTH STUDY

STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Emeny, et al., 2019 Pregnancy cohort Other outcomes, Infants, of women THg in maternal toenails: GM 0.02 µg/g, range 0.00007–1.19 In-utero exposure to MeHg might increase the risk
the United States immune system: in gestational of lower (but not higher) respiratory infections and
(New Hampshire) respiratory week 24–28 if respiratory symptoms in the first year of life. There
infections or they reported was little to no evidence of associations of toenail
respiratory using water Hg with upper respiratory infection, allergy or
symptoms, from a private, eczema at any interval.
atopic dermatitis unregulated well
(eczema) or allergy in their home
“TOXIC EFFECTS OF MeHg”

in the preceding since their last

four months menstrual period
and were not
planning a change
in residence prior
to delivery, age
4–12 months;
n = 706 (49%
female)
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

Kindgren, et al., 2019 Pregnancy cohort Other outcomes, Children, age THg in cord blood Hg (µg/L): median (range) Concentrations of Hg (and Al, Cd, Li) in cord blood
Sweden and retrospective immune system: 16 years; n = 40 Controls: 0.20 (0.20–0.53) were significantly higher in the juvenile idiopathic
case-control juvenile idiopathic controls, and 42 Antinuclear antibodies: 0.29 (0.20–0.65) arthritis-group than in controls. Concentrations of
arthritis, and cases of juvenile juvenile idiopathic arthritis: 0.30 (0.20–1.01) Hg in cord blood were not significantly higher in the
antinuclear idiopathic arthritis antinuclear antibodies-group compared to controls.
antibodies (ANA) Moderate exposure to heavy metals, including Hg,
associated with fish consumption, during pregnancy
and early childhood may cause effects on the
immune system of the offspring.
Mocevic, et al., 2013 Cross-sectional Other outcomes, Adults, male THg in blood (ng/mL): median (range) Results show evidence that environmental mercury
Greenland, Poland and reproduction: partners of Greenland: 9.2 (0.2–385.8) exposure in Greenlandic and European men with
Ukraine semen quality pregnant women, Poland: 1.0 (0.2–6.4) median whole blood concentration up to 10 ng/
and serum levels age mean (range): Ukraine: 1.0 (0.2–4.9) ml has adverse effects on biomarkers of male
of reproductive 29.5 (18.0–51.3); reproductive health.
hormones n = 529 (0%
female)

209
 TABLE 5.8 OVERVIEW OF PRIMARY STUDIES WITH OTHER HEALTH OUTCOMES (N = 15) FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg” (cont.)

210
AUTHOR, YEAR HEALTH STUDY
STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Fillion, et al., 2011 Cross-sectional Other outcomes, Adults, in THg in hair: median 11.5 mg/g, range 1.0–57.9 Hair Hg may be associated with visual acuity
Brazil (Lower Tapajos vision: visual fish-eating THg in blood: median 41.8 mg/L, range 1.7–179.3 in older persons. No association was observed
River basin) acuity communities, age THg in plasma: median 6.9 mg/L, range 0.2–30.9 between Log Hair Hg and near visual acuity loss in
mean (range): younger people, whereas for older persons, there
35.8 (15–66); n = was a highly significant association. Log Hair Hg
243 (female 52%) showed positive association with distant visual
acuity loss, but did not reach significance level (P
<0.05).
Cho, et al., 2012 Cross-sectional Other outcomes, Adults, THg in blood: median 3.74 µg/L, IQR <2.67–5.23 High blood Hg levels were associated with a lower
South Korea osteoporosis: bone postmenopausal risk of having osteoporosis in postmenopausal
mineral density women, age mean women.
between 60 and
64 years; n = 481
(100% female)
Kim, et al., 2016 Cross-sectional Other outcomes, Adults, men, age THg in blood: median 5.337 µg/L, IQR 3.347–8.104 Data showed that high blood Hg levels may be
South Korea osteoporosis: one 61 years ±0.2; positively associated with increased total hip and
mineral density n = 1 190 (0% femur neck bone mineral density, but do not offer
female) significant protection against fragility-associated
fractures.
Sanders, et al., 2019 Cross-sectional Other outcomes, Children, age THg in blood: mean 0.68 µg/L, SD 1.56 The findings suggest metals, including As, Pb,
the United States hypertension and mean 15.4 (SD THg in urine: mean 0.41 µg/L, SD 0.94 Hg, Cd and their combinations, may affect renal
renal disease: 2.3); n = 2709 parameters. Survey-weighted single chemical
kidney function (48% female) analyses showed estimated glomerular filtration
rate was 0.6% (95% CI: 0.1, 1.0) higher with
each decile increase in urine Hg. There was no
significant association with blood Hg.
Pan, et al., 2020 Cross-sectional Other outcomes, Children, local and THg in blood: mean 1.41 µg/L, 10th percentile 0.77, 95th percentile was 2.95
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

There was no association between blood Hg

China (Wuxi City) pulmonary native-born, age concentrations and pulmonary function in children.
function: forced mean 12.5; n =
expiratory volume 221 (47% female)
in 1 s and forced
vital capacity
 TABLE 5.8 OVERVIEW OF PRIMARY STUDIES WITH OTHER HEALTH OUTCOMES (N = 15) FOR THE REVIEW OF “HEALTH EFFECTS OF MeHg” (cont.)

AUTHOR, YEAR HEALTH STUDY

STUDY DESIGN MERCURY CONCENTRATIONS RESULTS AND CONCLUSIONS
COUNTRY OUTCOME PARTICIPANTS
Zidane, et al., 2019 Case-control Other outcomes, Cases of thyroid THg in fingernails, cases: mean 1.10 µg/g, 95% CI 0.96-1.24 Hg (and other non-essential trace elements)
French Polynesia cancer: thyroid cancer and a THg in fingernails, control: mean 1.20 µg/g, 95% Cl 0.90-1.30 fingernail concentrations were not associated to
cancer risk control group differentiated thyroid cancer risk.
(closest in terms
of date of birth,
otherwise random),
-, age younger
than 56; n = 229
“TOXIC EFFECTS OF MeHg”

cases, 249 controls

(gender not
provided)
Rhee, et al., 2020 Cross-sectional Other outcomes, Adults, age >20 THg in blood: mean 1.6 µg/L, SD 2.4 Higher blood THg and MeHg levels were associated
the United States cancer: skin cancer years; n = 29413 IHg in blood: mean 0.3 µg/L, SD 0.4 with a higher prevalence of nonmelanoma skin
(52% female) MeHg in blood: mean 1.3 µg/L, SD 2.4 cancer.
De Craemer, et al., 2017 Cross-sectional Other outcomes, Adolescents, age THg in hair: mean 0.169 µg/g, 95% CI 0.157–0.181 No significant effect on hormone levels in blood
Belgium sexual maturation: 14–15 years; MeHg in hair: mean 0.112 µg/g, 95% CI 0.104–0.121 was observed. For the other maturation endpoints,
C H A P TE R 5 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW

sex hormone n = 138–311 only female breast development was significantly

levels in males (balanced gender associated with H-MeHg (OR= 0.635, 95% CI:
and genital ratio) 0.411, 0.951) and H-THg (OR= 0.744, 95% CI:
development in 0.527, 0.973). Authors are cautious to draw
males and females conclusions.

Notes: THg: total mercury; GM: geometric mean, CI: confidence interval, ppb: parts per billion, IQR: interquartile range, SD: standard deviation, OR: odds ratio.

211
 212
© FAO/Alisa Suwanrumpha
CHAPTER 6
RESULTS AND
SUMMARIZATION OF
THE LITERATURE REVIEW
“THE ROLE OF SE WITH
REGARD TO THE HEALTH
EFFECTS OF MeHg”

6.1 LITERATURE AND QUALITY ASSESSMENT

Literature searches for the review on “The role of Se with regard to the health effects
of MeHg” were performed in the databases PubMed and Web of Science. A flow
diagram of the results from the literature searches is given in Figure 6.1.
The parallel literature searches performed in Web of Science and PubMed resulted in
1 154 records. Of these, 249 were found to be duplicate records. Following duplicate
removal in Endnote, 905 records were subjected to title and abstract screening using
Rayyan. Based on the predefined inclusion and exclusion criteria (Table 2.12), 671
records were excluded and 234 records (comprising 115 human studies and 119
animal studies) were retained for full-text screening.

6.1.1 HUMAN STUDIES

A total of 115 records of the human studies were assessed in full text. Of these,
68 were excluded based on the inclusion and exclusion criteria (Appendix 6,
Table 6.2). Thus, 47 records were assessed for risk of bias with the OHAT tool. Of
the 47 articles included for full-text quality assessment, 28 articles were rated Tier 1,
17 articles were rated Tier 2, and 2 articles were rated Tier 3 (Appendix 6, Table 6.3).
The studies graded Tier 3 were excluded, leaving 45 records for further assessment.

213
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

FIGURE 6.1. FLOW DIAGRAM FOR THE REVIEW “THE ROLE OF SE WITH REGARDS
TO THE HEALTH EFFECTS OF MeHg”

IDENTIFICATION OF STUDIES
VIA DATABASES

Records identified from database Records removed before

IDENTIFICATION
searching (n = 1 154): screening:
• Web of Science (n = 792) • Duplicate records removed
• PubMed (n = 362) (n = 249)

Records screened by title Records excluded by

and abstract (n = 905) Eligibility criteria (n = 671)

Full-text articles assessed

SCREENING

for eligibility (n = 234)

• Human studies (n = 115)
• Animal studies (n = 119)1

Human studies (n = 115) Human studies:

Records in full text excluded
by eligibility criteria (n = 68)

Human studies: Human studies:

Quality assessment (risk of bias) Records excluded after quality
of full-text articles (n = 47) assessment (risk of bias)
INCLUDED

(n = 2)

Human studies:
Primary studies included in final review (n = 45)

Notes: 1 Animal studies were included in the literature search and further used for background information for mechanistic
and biologically plausible evidence in the context of the included human studies. However, the animal studies were not
quality assessed or included in the weighing of the evidence of the studies.
Source: Prepared by authors based on Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D.,
Shamseer, L. et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ, 372: n71.
https://doi.org/10.1136/bmj.n71

6.1.2 ANIMAL STUDIES

A total of 119 animal studies were included and found relevant from the literature
search. These studies were further used for background information for mechanistic and
biologically plausible evidence. However, they were not quality assessed (risk-of-bias
assessment) nor was their evidence included in the weight-of-evidence assessment.

214
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

6.2 RESULTS AND SUMMARIZATION OF THE LITERATURE INCLUDED

6.2.1 DESCRIPTION OF THE LITERATURE
Forty-five human primary studies were included in the final review investigating the
role of Se with regard to the health effects of MeHg. A detailed description of the
human studies included is given in Appendix 6, Table A6.4, including study design,
study participants, measurements and exposure of Hg and Se, outcome assessment,
and results and conclusions regarding the effect of Se on MeHg toxicity.
The 45 articles comprised 34 different studies (some studies were published in several
articles) from 12 countries in Asia, Europe, North America and South America
(Table 6.1). Of the 45 included articles, 23 studies were cohort studies, 6 were case-
control studies, and 16 were cross-sectional studies.

TABLE 6.1 DISTRIBUTIONS OF THE STUDIES AND PUBLICATIONS INCLUDED

PER INVESTIGATED COUNTRY

COUNTRY NUMBER OF STUDIES NUMBER OF ARTICLES

Brazil 2 6
Canada 6 7
China 1 1
Faroe Islands 1 2
Japan 4 4
Korea 1 2
Poland 1 2
Slovenia and Croatia 1 1
Sweden 5 3
Taiwan Province of China 2 2
the United Kingdom of Great Britain and Northern Ireland 1 4
the United States 9 11

The reported results and the information given in the 45 articles varied. Although all
the studies measured Se and Hg levels in different human tissues, not all studies could
assess the potential effect of Se on Hg toxicity. Fifteen of the studies did not find
any association between Hg and the investigated health outcome. Thus, although Se
was measured, the studies could not assess the potential effect of Se on Hg toxicity.
In addition, four of the studies had not specifically studied the effects of Se on Hg
toxicity, even though they had measured both Se and Hg concentrations. (A summary
of the studies is given in Appendix 6, Table A6.4). This section does not describe in
detail the studies that could not assess the potential role of Se on MeHg exposure.
Thus, 26 studies (13 cohort studies, 1 case-control study, and 12 cross-sectional
studies) are described in further detail in the following results sections.

215
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

The health outcomes measured in the studies varied, and the results from the studies
are summarized within the following health outcomes:
> cardiovascular outcomes
> oxidative stress
> immune system
> reproduction
> thyroid hormones
> prenatal somatic development
> neurodevelopment and cognition
> vision function
> motor function
The results from the animal studies are communicated briefly with the goal of
coupling human findings to mechanisms, but those studies were not quality assessed
and are therefore not used for weighing the evidence.

6.2.2 CARDIOVASCULAR OUTCOMES

Five studies investigated the effects of Se on MeHg toxicity for cardiovascular
outcomes. A general summary of the studies is given in Table 6.2. The five studies
included three cohort studies (in four publications) (Bélanger et al., 2008b; Park et
al., 2016; Hu et al., 2017 and Park et al., 2017) and one cross-sectional study (Ayotte
et al., 2011).
Of these, two cohort studies (in three publications) (Park et al., 2016; Hu et al., 2017
and Park et al., 2017) and the cross-sectional study (Ayotte et al., 2011) showed an
effect of Se on MeHg toxicity for cardiovascular outcomes, while one cohort study
(Bélanger et al., 2008b) showed unclear effects (see Table 6.2).
In addition to the five studies that found an effect of Hg on cardiovascular outcomes
and measured the potential effect of Se, eight additional studies were found in the
literature search including cardiovascular outcomes (Appendix 6 Table A6.4). One
study did not study effects of Se on MeHg toxicity (Emanuele et al., 2017), and
seven studies (Engström et al., 2011; Mozaffarian et al., 2011; Wennberg et al., 2011;
Mozaffarian et al., 2012; Park et al., 2013; Gregory et al., 2016 and Chen et al.,
2018) did not show toxicity of Hg on the investigated parameters and thereby
could not investigate the modification of such toxicity by Se. As Se may alleviate
cardiovascular toxicity of Hg, the studies that did not find an effect of Hg could
have had Se as a confounder. Of those studies reporting no Hg toxicity, some had
reported lower mean levels of Hg in blood and toenails than reported in the four
studies shown in Table 6.2: For toenails, the concentration range was 0.21–0.31 µg/g
as median (Mozaffarian et al., 2011 and Mozaffarian et al., 2012) and for blood the
geometric mean was 1.03 µg/L (Park et al., 2013). The other studies could not be
compared for Hg concentrations, as they measured Hg in other matrices, namely
erythrocytes, urine, serum or hair.

216
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

TABLE 6.2 STUDIES INCLUDED FOR THE HEALTH OUTCOME “CARDIOVASCULAR OUTCOMES”
FOR THE REVIEW “SE AND MeHg”

STUDIED GROUP,
OTHER STUDY CONCLUSIONS
TIER, STUDY Hg AND Se
AUTHOR, YEAR CHARACTERISTICS EFFECT OF Se ON MeHg
DESIGN, EXPOSURE AND STATISTICAL TOXICITY 1
APPROACH
Ayotte et al., 2011 Tier 1, Blood geometric Canadian Inuit, Se intake offsets Hg exerted
cross-sectional mean: Nunavik, high inhibition of PON1 activity, a
study, n = 869, 37 Hg 11 µg/L seafood diet predictor for coronary heart disease.
years, Se 300 µg/L Regression analysis Effect
45%
Hu et al., 2017 Tier 2, cohort study, Blood median Hg, Canadian Inuit, High Se and low Hg had the lowest
n = 2169, 42 years 7.8 µg/L Nunavut ++, aged prevalence of cardiovascular
39% Se 280 µg/L >18 years outcomes (except stroke).
Two concentrations Effect
for each Hg and Se
Park et al., 2016; Tier 1, cohort study, Toenail mean Koreans from Positive association between toenail
Park et al., 2017 2 n = 501, 45 years Hg 0.4 µg/g Yeungnam area >35 Hg and
47% mean/median years > metabolic syndrome, which
Se 0.7/1.0 µg/g Regression analysis was weaker at higher
Se concentrations;
> higher risk of
hyper-LDL-cholesterolemia and
dyslipidemia, non-significant
at high toenail selenium levels
>0.685 µg/g.
Effect
Bélanger et al., Tier 2, cohort study, Blood mean Canadian sports Fishing season improved
2008 b n = 31, 47 years Hg 21.9 nmol/L fishermen cardiovascular health, not
100% before fishing season Two different doses homocysteine. Role of Se unclear.
of seafood in same Likely dominating confounder:
population Seasonal lifestyle with more exercise
and fresh air.
Not designed to investigate effect
of Se.
Unclear effect

Notes: PON1: paraoxonase-1; LDL: low-density lipoprotein. 1 The effects of Se on MeHg toxicity were assessed and
categorized as: “effect”, “indirect effect”, “unclear effect” and “no effect”. 2 These two articles are about the same
study, but refer to different outcomes.

6.2.2.1 Weight of evidence for “Cardiovascular outcomes”

Although some studies showed an effect of Se on the toxic effects of Hg on
cardiovascular outcomes, only a small number of studies were found, and different
outcomes were investigated in specific population groups. In addition, the potential
risk of confounding factors cannot be ruled out. Therefore, in conclusion, when
weighing the evidence according to the criteria from the WRCF, there is “limited,
suggestive” evidence of the effect of Se on MeHg toxicity on cardiovascular
outcomes.

217
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

This conclusion was based on:

> There is evidence from at least two independent cohort studies (Park et al., 2016;
Hu et al., 2017 and Park et al., 2017).
> There is evidence from more than one study type, though one study (Ayotte
et al., 2011) was a cross-sectional study, and therefore was not weighed heavily
compared to the cohort studies.
> There is low unexplained heterogeneity within or between study types or in
different populations relating to the presence or absence of an association, or
direction of effect. The studies were deemed low risk of bias (one cohort study
ranked Tier 1), and included both males and females and different ethnic groups.
> The presence of a plausible biological gradient (“dose–response”) in the
association: One study was a small two-factorial study investigating high and
low Se combined with high and low Hg. Most of the difference was found
between high Se/low Hg and low Se/high Hg (Hu et al., 2017), while another
study (Ayotte et al., 2011) found that metabolic syndrome was associated
positively with Hg, but that this was dependent on the concentration of Se.

6.2.2.2 Animal studies on “Cardiovascular outcomes”

There were no animal studies with the outcome of cardiovascular diseases included
in the literature search.

6.2.3 OXIDATIVE STRESS

Five studies investigated the effects of Se on MeHg toxicity for the outcome of
oxidative stress. A general summary of the studies is given in Table 6.3. The studies
included two cohort studies (Bélanger et al., 2008b and Kuras et al., 2019) and two
cross-sectional studies (in three publications) (Bélanger et al., 2006; Bélanger et
al., 2008a and Karimi et al., 2016) (Table 6.3). One cross-sectional study (Karimi
et al.) showed an effect of Se on Hg toxicity, while the remaining studies showed
only indirect or less clear effects. Those with unclear results had a low number of
participants (< n = 100 participants). Three studies were rated Tier 1, while one study
was rated Tier 2. There was no directional heterogeneity between the studies, as
there was no study that did not find effects of Se on Hg toxicity regarding oxidative
stress (Table 6.3).
In addition to the five studies that found an effect of Hg on the oxidative stress
outcomes and measured the potential effect of Se, two additional studies were found
in the literature search including oxidative stress: Rocha et al., 2016 and Kuras et al.,
2019. Rocha et al. did not find any toxicity of Hg and was not included for further
assessment, while Kuras et al. did not investigate modulation of Hg effects by Se.
Kuras et al., however, did show correlations of Hg with selenoprotein-P and small
oxidants in blood.

218
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

TABLE 6.3 STUDIES INCLUDED FOR THE HEALTH OUTCOME “OXIDATIVE STRESS” FOR THE REVIEW
“SE AND MeHg”

STUDIED GROUP,
TIER, TYPE, OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg AND Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF Se ON Hg TOXICITY 1
APPROACH
Karimi et al., 2016 Tier 1, cross- Blood Long Island the Association of high Hg from fish
sectional study, mean United States, avid consumption and GSH:GSSG
n = 268, 48 years, Hg 8 µg/L seafood eaters (reduced glutathione:oxidized
42% median Regression analysis glutathione), which is indicative for
Se 240 µg/L oxidative stress, was slightly less
pronounced in those with highest Se
in blood. Possible slight alleviation
of Hg-induced stress.
Effect
Belanger et al., 2006 Tier 1, cross- Blood mean Inuit > Concentration of plasma
Belanger et al., sectional study, Hg 21 µg/L Regression analysis homocysteine was negatively
2008a2 n = 99, 43 years, Se 636 µg/L predicted by Se, but dietary Hg
28% showed no association.
> Low LDL oxidation may be partly
explained by high blood Se
status, that might reduce the
deleterious effects of MeHg on
cardiovascular health.
Indirect effect
Bélanger et al., Tier 2, cohort study, Blood mean Canadian sports Fishing season improved
2008b n = 31, 47 years, Hg 4.4 µg/L before fishermen cardiovascular health, not
100% fishing season Two different doses homocysteine. Role of Se unclear.
and 7.1 µg/L after of seafood in same Not designed to investigate effect
(recalculated from population of Se.
molar to mass
Indirect effect
based)
Se blood mean 243
µg/L before and 248
µg/L after fishing
season.
Kuras et al., 2019 Tier 1, cohort study, Blood median Hg 1 Polish men Both Hg and Se increase in an
n = 67, 41 years, µg/L, hair 0.3 µg/g, Regression analysis intervention with high fish intake.
100% urine 0.23 µg/L. Increase in TBARS as an effect of
Plasma median Se fish intake but little evidence for
82 µg/L; urine 17 direct action of Se on GSH-Px. No
µg/L. direct evidence for a protective
effect of Se.
Se and selenoproteins mainly
investigated as biomarker of Hg
effect.
Indirect effect

Notes: LDL: low-density lipoprotein; TBARS: thiobarbituric acid-reactive substances; GSH-Px: glutathione peroxidase.
1
The effect of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”, “unclear
effect” and “no effect”. 2 These two articles are about the same study, but refer to different outcomes.

219
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

6.2.3.1 Weight of evidence for “Oxidative stress”

In conclusion, when weighing the evidence according to the criteria from the WCRF,
this was found to be “limited evidence, no conclusion” regarding the effect of Se on
Hg toxicity outcomes on oxidative stress. This conclusion was based on:
> There is direct evidence from only one cross-sectional study, and only indirect
evidence from more than one study type.
> There is no direct, only indirect, evidence from two independent cohort studies.
> The number of participants in the cohort studies was low (< n = 100).
The one cross-sectional study (Karimi et al., 2016) with a moderate number of
participants (n = 268), rated Tier 1, shows a slight effect, and four studies (Bélanger et
al., 2006; Bélanger et al., 2008; Bélanger et al., 2008b and Kuras et al., 2018) indicate
the same direction of effect, but without providing direct evidence.

6.2.3.2 Animal studies on “Oxidative stress”

Further corroboration on the outcomes of oxidative stress comes from animal
studies (Grotto et al., 2008; Grotto et al., 2009; Joshi et al., 2014; Orct et al., 2015
and Moniruzzaman et al., 2021). A mice study that was included in the literature
search (Balthrop et al., 1985) showed that intraperitoneal injections of Se produced
an increase in glutathione and glutathione-S-transferase activity in Se-deficient
animals only. However, when followed by MeHg injections, the negative effect of
MeHg on glutathione-S-transferase was reduced by the administered Se. However,
Se supplementation may either alleviate or augment the effects of MeHg, depending
on their doses and combinations (Jin et al., 2012).

6.2.4 IMMUNE SYSTEM

Five studies investigated the effects of Se on MeHg toxicity for the outcome of
biomarkers of the immune system. A general summary of the studies is given in
Table 6.4.
The five studies included two cross-sectional studies: Hui et al., 2016 and Nyland
et al., 2011a. When investigating the effect of Se on Hg toxicity on the immune
system, Hui et al. found an effect, while Nyland et al. did not. Both studies were
rated Tier 2 in the quality assessment conducted with the OHAT risk-of-bias tool.
In addition to the two studies that found an effect of Hg on immune system
outcomes and measured the potential effect of Se, an additional study was found
in the literature search: Monastero et al. (2017). This study did not find an effect
of Hg toxicity.

220
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

TABLE 6.4 STUDIES INCLUDED FOR THE HEALTH OUTCOME “IMMUNE SYSTEM” FOR THE REVIEW
“Se AND MeHg”

STUDIED GROUP,
TIER, TYPE, OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg and Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF Se ON Hg TOXICITY 1
APPROACH
Hui et al., 2016 Tier 2, Median Hg Chinese local birth Small but significant association
cross-sectional cord blood: 9 µg/L cohort between mercury and IL-10
study, blood: 3 µg/L Regression analysis concentration, more pronounced
n = 608, 8 years Median Se when selenium and cord blood
54% blood: 92 µg/L mercury were low. Se seems to
alleviate Hg toxicity. The clinical
significance is unclear.
Effect
Nyland et al., 2011 Tier 2, cross- Median Hg Brazil Tapajos River Antinuclear (ANA) and antinucleolar
sectional study, hair: 14 µg/g basin population, (ANoA) autoantibody levels and
n = 232, 15–78 blood: 54 µg/L affected by MeHg eight cytokines in serum samples,
years (no mean plasma: 9 µg/L from gold extraction. e.g. (IL)-6, (IFN)-γ, (IL-4), and IL-
provided), 48% urine: 3 µg/L Regression analysis 17, were investigated. Se status was
Se concentrations not associated with any changes in
not provided. ANA and did not modify associations
between Hg and ANA titers.
No effect

Notes: IL: interleukin; IFN: interferon 1 The effect of Se on MeHg toxicity were assessed and departed into the categories of
“effect”, “indirect effect”, “unclear effect” and “no effect”.

6.2.4.1 Weight of evidence for “Immune system”

In conclusion, when weighing the evidence according to the WCRF criteria, very
“limited evidence, no conclusion” was found of the effect of Se on Hg toxicity
outcomes on the immune system. This conclusion was based on:
> There were no cohort studies investigating this hypothesis, both studies were
cross- sectional and were graded Tier 2.
> There may be a countereffect of Se on the effect of Hg on IL-10, with uncertain
clinical significance. Other immune system markers were not found to be
associated.
The data on effects of Se on Hg toxicity on the immune system was too scarce
for conclusions; however, there are animal studies that demonstrate effects and
encourage further investigation in humans.

6.2.4.2 Animal studies on “Immune system”

In a study of mice (Li et al., 2014a), proliferation of T and B lymphocytes was investigated
upon administration of Hg and Se with drinking water, where – at low Hg concentration
– Se exhibited protective effects on Hg-induced suppression, while – at greater Hg
concentration – immunotoxicity induced by high concentrations of Se became evident,
with protective effects at a Se/Hg molar ratio of 1:1, witnessing antagonistic effects
between Se and Hg. Another mice study (Li et al., 2014b) investigating redox-mediated
immune suppression by MeHg found a protective effect of Se administered by diet.

221
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

6.2.5 REPRODUCTION
Two studies, a case-control study (Maeda et al., 2019) and a cross-sectional study
(Ai et al., 2019), investigated the effects of Se on MeHg toxicity for the outcome of
reproduction. A general summary of the studies is given in Table 6.5. Both studies
were graded Tier 2. The cross-sectional study (Ai et al., 2019) did not show an effect,
while the case-control study (Maeda et al., 2019) showed an effect. The study finding
an effect bases this claim on a small difference in Se and Hg blood levels. Hg blood
levels in these studies were compared between pregnant and infertile groups, where
a higher level of Hg was found in the infertile group.

TABLE 6.5 STUDIES INCLUDED FOR THE HEALTH OUTCOME “REPRODUCTION” FOR THE REVIEW
“Se AND MeHg”

STUDIED GROUP,
TIER, TYPE, OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg and Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF Se ON Hg TOXICITY 1
APPROACH
Maeda et al., 2019 Tier 2, case-control Mean blood Hg Women from The Se and Se/Hg ratio were lower in
study, infertile: 5.3 µg/L northeast Japan the infertile group than in the control
n = 141, 35 years fertile: 5 µg/L Comparison of two group. Hg levels were higher in the
infertile, 34 years Mean blood Se groups infertile group than in the control group
fertile, 0% Infertile: 189 µg/L (adjusted for age and Se). The authors
fertile: 200 µg/L concluded that Hg and Se exposures
appear to have adverse and protective
effects on female fertility, respectively.
However, the difference in Hg and
Se blood levels between the groups
were small, and the fertile group was
slightly younger than the control group.
Effect
Ai et al., 2019 Tier 2, cross- Mean Hg Men from Taiwan Dose-dependent correlation between
sectional study, blood: 9 µg/L Comparison of two blood Hg and normal sperm count.
n = 84, 37 years, semen: 1 µg/L. groups High predatory fish intake had lower
0% Mean Se percentage of normal morphology in
blood: 205 µg/L sperm.
semen: 82 µg/L No significant difference between
high- and low-quality semen for Se
concentrations.
No effect

Notes: 1 The effects of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”,
“unclear effect” and “no effect”.

6.2.5.1 Weight of evidence for “Reproduction”

In conclusion, when weighing the evidence according to the WCRF criteria, “limited
evidence, no conclusion” is found for the effect of Se on Hg toxicity outcomes on
reproduction. This conclusion was based on:
> The one study (Maeda et al., 2019) claiming to have found an effect based its
claims on very small differences in the levels of Hg concentrations, where the
infertile group had somewhat higher Hg.

222
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

> The other study investigating the hypothesis found no clear effect for the
outcome of sperm count (Ai et al., 2019).
> Both studies investigated very different outcomes and therefore cannot be
compared.
> Non-significant trends in these studies may encourage better or larger studies.

6.2.5.2 Animal studies on “Reproduction”

An animal study that was included from the literature search (Beyrouty et al.,
2006) produced evidence for an alleviation of MeHg toxicity on body weight gain
during lactation in mice by preceding feeding with Se-enriched diet. Reduced
embryotoxicity of MeHg through selenite by concomitant exposure was also
shown in another animal study (Nobunaga et al., 1979). There is some evidence
from animal studies pointing towards an effect of Se on Hg toxicity on reproduction,
but different outcomes (weight gain of offspring, reduced embryotoxicity) from
those in the human studies (female fertility, sperm count) were investigated.

6.2.6 THYROID HORMONES

Gustin et al. (2021) investigated the effects of Se on MeHg toxicity for the outcome
of thyroid hormones. A general summary of the study is given in Table 6.6. This
cohort study included the effects on the thyroid hormones free tri-iodothyronine
(fT3), total T3 and free thyroxine (fT4). An association was found with Hg status,
though no or a very slight effect was found of Se.

TABLE 6.6 STUDIES INCLUDED FOR THE HEALTH OUTCOME “THYROID HORMONES” FOR THE REVIEW
“Se AND MeHg”

STUDIED GROUP,
TIER, TYPE, OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg and Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF Se ON Hg TOXICITY 1
APPROACH
Gustin et al., 2021 Tier 1, cohort study, Median Hg Mothers from Erythrocyte Hg was non-linearly
n = 542, 30 years, Erythrocytes: northern Sweden associated with fT3, TT3 and fT3:fT4
0% 1.5 µg/L Regression analysis ratio. No or very slight effect of Se.
Median Se The authors concluded that Hg can
Plasma: 67 µg/L interfere with thyroid hormones,
but that Se does not affect this
interference.
No effect

Notes: fT3: free triiodothyronine; TT3: total triiodothyronine; fT4: free thyroxine
1
The effects of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”, “unclear
effect” and “no effect”.

223
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

6.2.6.1 Weight of evidence for “Thyroid hormones”

In conclusion, when weighing the evidence according to the WCRF criteria, “limited
evidence, no conclusion” is found of the effect of Se on Hg toxicity outcomes on
thyroid hormones. There was only one study investigating only thyroid hormones,
which found no effect.

6.2.6.2 Animal studies on “Thyroid hormones”

There were no animal studies with the outcomes within the research field of thyroid
hormones included from the literature search.

6.2.7 BIRTH OUTCOMES

Two studies (Wells et al., 2016 and Kobayashi et al., 2019) investigated the effects
of Se on MeHg toxicity on birth outcomes. A general summary of the studies is
given in Table 6.7. Wells et al, a cross-sectional study, found a protective effect of
Se on Hg toxicity on developmental early-life outcomes, while Kobayashi et al., a
cohort study, did not.
In summary, there is evidence from two different studies, one cohort study with
a very high participant number (ranked Tier 1), finding no effect, and one cross-
sectional study with a much lower participant number (ranked Tier 2), finding an
effect. These studies originate from two different population groups (Japan and the
United States) (Table 6.7).

TABLE 6.7 STUDIES INCLUDED FOR THE HEALTH OUTCOME “BIRTH OUTCOMES” FOR THE REVIEW
“Se AND MeHg”

STUDIED GROUP,
TIER, TYPE OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg AND Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF Se ON Hg TOXICITY 1
APPROACH
Wells et al., 2016 Tier 2, cross- Geometric mean the United States The negative association of Hg with
sectional study, cord blood: 1 µg/L (Baltimore) birth weight and ponderal index was
n = 271, 0 years Mean Se mother–child pairs influenced by Se.
cord blood: 70 µg/L Regression analysis Effect
Kobayashi et al., Tier 1, cohort study, Mean/median Hg Japanese children No effect of Hg on birth weight
2019 n = 15 444, 31 blood: 4 µg/L Regression analysis and “small for gestational age” as
years, 52% Mean Se outcomes. Weak correlation with head
Blood: 171 µg/L circumference and blood Hg. Se did not
give any protective effect.
No effect

1
The effect of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”, “unclear
effect” and “no effect”.

224
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

In addition to the two studies that found an effect of Hg on the birth outcomes and
measured the potential effect of Se, one additional study (Chen et al., 2014) was found
in the literature search which included birth outcomes (Appendix 6, Table A6.4).
This study did not measure the effect of Se on Hg toxicity.

6.2.7.1 Weight of evidence on “Birth outcomes”

In conclusion, when weighing the evidence according to WCRF criteria, “limited
evidence, no conclusion” was found regarding the effect of Se on Hg toxicity
outcomes on birth outcomes. This conclusion was based on:
> There is only one study providing evidence of an effect.
> Two independent cohort studies have not yet been conducted.
> There was heterogeneity in the results.

6.2.7.2 Animal studies for “Birth outcomes”

There were no animal studies with the outcomes within the research field of birth
outcomes included from our literature search.

6.2.8 NEURODEVELOPMENT AND COGNITION

Five studies investigated the effects of Se on MeHg toxicity for the outcome of
neurodevelopment and cognition. A general summary of the studies is given in Table 6.8.
All five studies were cohort studies. One cohort study (Tratnik et al., 2017) found an
indirect positive effect, while the other four cohort studies (Steuerwald et al., 2000;
Choi et al., 2008; Golding et al., 2017 and Tatsuta et al., 2017) found no protective
effect of Se for the neurotoxicological effects of Hg.
In conclusion, there is evidence from four cohort studies, two ranked Tier 2 and two
ranked Tier 1, three of which had moderately high participant numbers, originated
from different population groups (Japan, the Faroe Islands and the United Kingdom
of Great Britain and Northern Ireland) reporting that Se had no protective effect
on the neurotoxicological effects of Hg early in life, if cord blood Hg levels were
rather high. The only study finding evidence for a protective effect of Se on Hg
neurotoxicity showed this only for a genetic subgroup and had a participant group
with relatively low levels of Hg in cord blood.
In addition to the five studies that found an effect of Hg on the birth outcomes
and measured the potential effect of Se, five additional studies were found in the
literature search including birth outcomes (Boucher et al., 2014b; Golding et al.,
2016a; Golding et al., 2016b; Oken et al., 2016 and Mao et al., 2019) (Appendix 6
Table A6.4). These studies did not measure the effect of Se on Hg toxicity.

225
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 6.8 STUDIES INCLUDED FOR THE HEALTH OUTCOME “NEURODEVELOPMENT AND COGNITION”
FOR THE REVIEW “Se AND MeHg”

STUDIED GROUP,
TIER, TYPE OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg and Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF SE ON Hg TOXICITY 1
APPROACH
Tratnik et al., 2017 Tier 1, cohort study, Cord blood Children from Hg-related decrease in cognitive score
n = 361, mother– mean Hg Croatia and Slovenia only in children carrying at least one
child pairs Croatia: Hg 3.4 µg/L Regression analysis Apoeε4 allele, and general decrease in
Slovenia: Hg 1.6 fine motor scores. Positive association
µg/L between Se and the language score,
Cord serum mean but not in the subgroup of children
Se 40 µg/L (both carrying the ε4 allele. Unclear if effect
countries) of Se on Hg toxicity or just separate
effects. Weak effect of Se for a genetic
subgroup.
Indirect effect
Choi et al., 2008 Tier 2, cohort study, Geometric mean Hg Children from Faroe No evidence that Se was an important
n = 1 022, 7 years, cord blood 23 µg/L Islands protective factor against MeHg
0% Regression analysis neurotoxicity. Increased Se levels
Geometric mean Se
were not associated with decreased
cord blood 112 µg/L
mercury-related neuropsychological
dysfunctions.
No effect
Golding et al., 2017 Tier 2, cohort study, Median Hg blood UK children Selenium did not influence positive
n = 2 062, 8 years 2 µg/L Regression analysis, effect of fish eating on intelligence
Median Se blood fish eaters and despite correlation of fish eating with
108 µg/L non-fish eaters also Hg.
treated separately No effect
Steuerwald et al., Tier 1, cohort study, Geometric mean Hg Infants from Faroe The authors found a significant
2000 n = 182, 17 days, cord blood 20 µg/L Islands decrease in the neonatal NOS
51% cord serum 3 µg/L Regression analysis connected to methylmercury from
hair 4 µg/g seafood. They also concluded that that
Geometric mean there was no evidence that Se offered
Se cord serum 103 protection against mercury-associated
µg/L decrease in NOS.
No effect
Tatsuta et al., 2017 Tier 1, cohort study, Mean Hg Japanese coastal Prenatal Hg negatively affected
n = 566, 1.5 years, cord blood 15–17 children psychomotor performance as measured
50% µg/L hair 3 µg/g Regression analysis in BSID-II. Good correlation between
Mean Se cord cord blood Hg and hair Hg; small but
plasma 66–67 µg/L significant negative effect of Hg on the
psychomotor part of the BSID-II test.
No effect

Notes: NOS: neurologic optimality score; BSID-II: Bayley Scales of Infant Development second edition
1
The effect of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”, “unclear
effect” and “no effect”.

226
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

6.2.8.1 Weight of evidence for “Neurodevelopment and cognition”

In conclusion, when weighing the evidence according to the WCRF criteria, “limited
evidence, no conclusion” was found regarding the effect of Se on Hg toxicity
outcomes on neurodevelopment and cognition. This conclusion was based on:
> Only cohort studies have been carried out, therefore there is no evidence from
different study types.
> Four different independent cohort studies found no effect on this group of
outcomes.
> There is heterogeneity in the results of the studies.

6.2.8.2 Animal studies on “Neurodevelopment and cognition”

Studies in mice and rats show evidence for an alleviation of MeHg toxicity by Se on
average auditory startle response time and somatosensory sensitivity, grip strength,
clasping reflex, flexion and voluntary wheel running and neuronal degeneration by
preceding feeding with Se-enriched diet (Beyrouty et al., 2006; Heath et al., 2010 and
Sakamoto et al., 2013), on cerebellar damage by concomitant MeHg and Se exposure
by gavage (Tu et al., 2021), on neurotoxicity by concomitant injection with selenite
(Chang, 1983), and on the development of reflexes by subcutaneous injection of
selenite (Satoh et al., 1985). Feeding pregnant mice different concentrations of Se
and MeHg also showed increasingly negative neurobehavioral outcomes in the
offspring in groups exposed to lower Se and higher MeHg (Watanabe et al., 1999).
Another study, where selenite was injected subcutaneously, found that inorganic
selenium was ineffective in preventing most of the MeHg-induced brain biochemical
alterations (Glaser et al., 2010).

6.2.9 VISION FUNCTION

Four studies investigated the effects of Se on MeHg toxicity for the outcome of
vision function. A general summary of the studies is given in Table 6.9. Two cross-
sectional studies found positive effects of Se, one on colour vision in adult people
(Fillion et al., 2013) and one on cataract formation in older people (Lemire et al.,
2010). One cross-sectional study and one cohort study found no effects of Se on
Hg toxicity, one investigated acuity in adult people (Fillion et al., 2011) and one
visual-evoked potential in children (Saint-Amour et al., 2006).
In conclusion, there is evidence from two cross-sectional studies (rated Tier 1) that
specific visual impairments caused by Hg exposure may be alleviated by Se. Since
both studies investigated different outcomes, and two further studies investigating
further different outcomes did not find an effect, the amount of evidence is scarce
for the outcome of visual function.

227
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 6.9 STUDIES INCLUDED FOR THE HEALTH OUTCOME “VISION FUNCTION” FOR THE REVIEW
“Se AND MeHg”

STUDIED GROUP,
TIER, TYPE OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg AND Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF SE ON HG TOXICITY 1
APPROACH
Fillion et al., 2013 Tier 1, cross- Mean/median hair Brazil Tapajos River Increased Se associated with less
sectional study, Hg 14/11.5 µg/g basin population. colour confusion and more near visual
n = 228, 35 years, Mean/median Regression analysis contrast; might counteract colour
50% plasma vision loss associated with hair Hg.
Se 182/145 µg/L Effect
Designed for
Se interference
analysis
Lemire et al., 2010 Tier 1, cross- Median Hg Brazil Tapajos River Weak effect of Se on cataract
sectional study, blood: 44 µg/L basin population. formation, which may be caused by the
n = 211, 50/73 plasma: 6.4 µg/L Regression analysis effect of Se as an antioxidant.
years Median Se Effect
55% blood: 222 µg/L
plasma 133 µg/L
Fillion et al., 2011 Tier 1, cross- Mean/median Hg Brazil Tapajos River Hg is associated with worse acuity in
sectional study, hair: 14/11.5 µg/g basin population. older people, but Se did not change
n = 243, 36 years blood: 52/42 µg/L Regression analysis that. Blood Se was significantly lower
55% plasma: 8/7 µg/L in the group excluded for age-related
Mean/median Se cataracts.
blood: 313/251 µg/L No effect
plasma: 179/141
µg/L
Saint-Amour et al., Tier 1, cohort study, Mean Hg Arctic Nunavik Visual-evoked potentials. Designated
2006 n = 110, 5 years, blood: 6 µg/L; Regression analysis N75 and P100, N150 were negatively
38.5% cord blood: 17 µg/L impacted by Hg, while no significant
Mean Se interaction with Se was found.
blood: 331 µg/L No effect
cord blood: 319 µg/L

Notes: N75: negative deflection at ≈ 75 ms; P100: positive deflection at ≈ 100 ms; N150: negative deflection at ≈ 150 ms.
1
The effects of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”, “unclear
effect” and “no effect”.

6.2.9.1 Weight of evidence for “Vision function”

In conclusion, when weighing the evidence according to the WCRF criteria, “limited
evidence, no conclusion” is found of the effect of Se on Hg toxicity outcomes on
vision outcomes. This conclusion was based on:
> Only one cohort study was undertaken, finding no effect. All other studies
were cross-sectional studies. Thus, there are no results showing effects from
two independent cohort studies.
> The direction of effect is consistent, in that the studies concluding with “effect”
and “no effect” have investigated different visual outcomes.

228
C H A P TE R 6 . RESULTS AND SUMMARIZATION OF THE LITERATURE REVIEW
“THE ROLE OF Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”

6.2.9.2 Animal studies on “Vision function”

There were no animal studies with the outcomes within the research field of vision
function included from the literature search.

6.2.10 MOTOR FUNCTION

One study (Lemire et al., 2011) investigated the effects of Se on MeHg toxicity for
the outcome of motor function. This was a cross-sectional study investigating the
effect of Se on Hg-induced impairments of motor function. A general summary
of the study is given in Table 6.10. The study was rated Tier 1 in the risk-of-bias
assessment. The study showed that high plasma Se was correlated positively with
motor function outcomes, while Hg was a counteracting confounder.

TABLE 6.10 STUDY INCLUDED FOR THE HEALTH OUTCOME “MOTOR FUNCTION” FOR THE REVIEW
“Se AND MeHg”

STUDIED GROUP,
TIER, TYPE OTHER STUDY CONCLUSIONS
AUTHOR, YEAR N, MEAN AGE, Hg and Se CHARACTERISTICS
PERCENT MALE AND STATISTICAL EFFECT OF Se ON Hg TOXICITY 1
APPROACH
Lemire et al., 2011 Tier 1, cross- Mean/median Hg Brazil Tapajos River Positive correlation between high
sectional study, blood: 51/44 µg/L basin population plasma Se and motor function
n = 319, 42 years plasma median: 6 Regression analysis outcomes, more when including Hg as
50% µg/L Mean/median a counteracting confounder.
Se Effect
blood: 288/222 µg/L
plasma median:
133 µg/L

1
The effects of Se on MeHg toxicity were assessed and departed into the categories of “effect”, “indirect effect”, “unclear
effect” and “no effect”.

6.2.10.1 Weight of evidence for “Motor function”

In conclusion, when weighing the evidence according to the WCRF criteria, “limited
evidence, no conclusion” is found regarding the effect of Se on Hg toxicity outcomes
of motor function. There was only one study investigating motor function, which
found an effect.

6.2.10.2 Animal studies on “Motor function”

MeHg was also shown in mice to cause a significant decrease in motor activity,
which was reduced by selenomethionine co-exposure, both if administered by diet
and via injection (Folven et al., 2009).

229
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

6.3 FINAL WEIGHT OF EVIDENCE FOR THE REVIEW “THE ROLE OF

Se WITH REGARD TO THE HEALTH EFFECTS OF MeHg”
A summarization of the final weight of evidence for the different outcomes included
in “The role of Se with regard to the health effects of MeHg” is given in Table 6.11.
The direction of the effect is given according to whether most studies showed
“effect” or “no effect” regarding the effect of Se on MeHg toxicity. If the studies
showed contradictory findings, this was depicted as “unclear”.
As shown in Table 6.11, except for the health outcome “cardiovascular outcomes”,
which was graded “limited, suggestive”, the remaining health outcomes were graded
“limited, no conclusion”.

TABLE 6.11 SUMMARY OF FINAL WEIGHT OF EVIDENCE FOR THE “ROLE OF Se WITH REGARD TO THE
HEALTH EFFECTS OF MeHg”

DIRECTION OF AN EFFECT OF CONCLUSION,

HEALTH OUTCOME Se ON MeHg TOXICITY WEIGHT OF EVIDENCE
Cardiovascular outcomes Effect (protective) Limited, suggestive
Oxidative stress Effect (protective) Limited, no conclusion
Immune system No effect Limited, no conclusion
Reproduction Unclear Limited, no conclusion
Thyroid hormones Unclear Limited, no conclusion
Birth outcomes No effect Limited, no conclusion
Neurodevelopment and cognition No effect Limited, no conclusion
Vision function Effect (protective) Limited, no conclusion
Motor function Effect (protective) Limited, no conclusion

Note: Final weight of evidence is based on the grading from the World Cancer Research Fund grading system (WCRF, 2018
and WCRF, 2018a).

It should be noted that even though, for most of the health outcomes, the evidence
was graded “Limited, no conclusion”, this was mostly related to the low number
of studies published, the variability of the methodology used in the studies, the
investigation of different population groups, and the fact that most of the studies
were not methodologically designed to measure “The role of Se with regards to the
health effects of MeHg”. Thus, although the evidence on this topic is rather low,
the weight of evidence might change if high quality studies designed to measure the
interaction between Se and Hg toxicity are conducted.

230
231
 232
© FAO/Alisa Suwanrumpha
CHAPTER 7
RESULTS AND
SUMMARIZATION OF
“OCCURRENCE DATA
FOR MeHg AND, DIOXINS
AND dl-PCBs”

7.1 LITERATURE SEARCH

Literature searches for the systematic review on the “Occurrence data for MeHg and
dioxins in fisheries and aquaculture products” were performed in Web of Science.
A flow diagram of the results from the literature searches is given in Figure 7.1.
The literature searches in Web of Science resulted in 6 851 records. Of these,
69 duplicate records were identified and removed, and 6 782 records were screened
by title and abstract using the online screening tool Rayyan. A further 4 476 records
were excluded based on inclusion and exclusion criteria during this title and abstract
screening. The remaining 2 306 records were screened based on the full-text articles.
Of these, 1 054 records were excluded based on exclusion criteria: 20 articles because
the language of the article was not English and 1 034 articles because of sampling
before 2011 or sampling year missing. The remaining 1 252 full-text articles were
subjected to quality assessment.

7.2 QUALITY ASSESSMENT OF THE LITERATURE INCLUDED FROM

THE LITERATURE SEARCH
Quality assessment of full-text articles was performed using the quality-assessment
template described in Section 2.5.4 with questions assessing the quality of the article
with respect to food description, component identification, sampling plan, number
of analytical samples, sample handling and analytical method and performance.

233
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

FIGURE 7.1. FLOW DIAGRAM FOR THE REVIEW “OCCURRENCE DATA OR MeHg, AND DIOXINS AND
dl-PCBs IN FISHERIES AND AQUACULTURE PRODUCTS”

IDENTIFICATION OF STUDIES
VIA DATABASES

•
Records identified from database Records removed before

IDENTIFICATION
searching (Web of Science) screening:
(n = 6 851) • Duplicate records removed
• MeHg: (n = 5 884) (n = 69)
• Dioxins: (n = 967)

Records screened by title Records excluded by criteria

and abstract (n = 6 782) (n = 4 476)
SCREENING

Full-text articles assessed Records excluded by criteria

for eligibility (n = 2 306) (n = 1 054)

Quality assessment of full-text Records excluded after

articles (n = 1 252) quality assessment (n = 698)
INCLUDED

Total studies included in final systematic review

(n = 554)

Source: Prepared by authors based on Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D.,
Shamseer, L. et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ, 372: n71.
https://doi.org/10.1136/bmj.n71

Of the 1 252 articles eligible for quality assessment, 698 articles were excluded and
554 articles were included. Of the excluded articles, 383 articles were excluded due to
inadequate food description, 46 due to inadequate component identification, 62 due
to missing sample number, 103 due to inadequate analytical method or performance,
and 104 articles were excluded for other reasons.
Details of the quality assessment for each article are given in Appendix 7, Table A7.3,
which also provides the full reference for each article.

7.3 OVERVIEW OF CONCENTRATION DATA FOR TOTAL Hg, MeHg, DIOXINS

AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS
The data obtained from the literature review and the data from the EFSA database
were treated separately and summarized in separate tables. This was done for
several reasons. Most importantly, there was no way to control which data were

234
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

published both in articles and to EFSA, thus some of the data could be the same.
Moreover, different levels of information were included in the two datasets. That
is, in the literature data, the species or genus name was always included, while the
EFSA data contained more generic names of species or species groups. There was
also a difference with regard to sample origin. The EFSA database is primarily
a European database with data from European laboratories analysing samples of
European seafood or seafood imported from third-party countries, while the data
from the literature search could represent seafood sampled and analysed anywhere
in the world.
In the literature, except for a few articles, data were mostly presented as mean
concentrations, where median values were given (Appendix 7, Table A7.4). In the
following tables, results from the literature are presented as means of mean or
median values. For consistency, the EFSA data are also presented as mean values
in the tables.

7.3.1 OVERVIEW OF CONCENTRATION DATA FROM THE LITERATURE SEARCH

From the 554 articles included, concentration data for total Hg, MeHg or dioxins
and dl-PCBs were extracted, along with necessary metadata, and compiled in an
excel sheet as described in Section 1.1.4. Most of the articles reported data on total
Hg, adding up to a total sample number of 68 998, while only 66 of the articles
had results for MeHg (total n = 7 720). Forty-six articles reported results for either
dioxins, dl-PCBs and/or the sum of dioxins and dl-PCBs, and of these 2 760 samples
were analysed for either dioxins, dl-PCBs or both and/or their total sum.
An overview of concentration data from the literature search for total Hg and
MeHg, in finfish and shellfish, from different regions around the world, is shown in
Table 7.1, Table 7.2, Table 7.3, Table 7.4, Table 7.5 and Table 7.6 (finfish), and in
Table 7.12, Table 7.13, Table 7.14, Table 7.15 and Table 7.16 (shellfish). An overview of
the concentration data from the literature search for dioxins and dl-PCBs is presented
in Table 7.22, Table 7.23, Table 7.24 and Table 7.25 (finfish), and in Table 7.31,
Table 7.32 and Table 7.33 (shellfish).

7.3.2 OVERVIEW OF CONCENTRATION DATA FROM THE DATABASE

OF THE EUROPEAN FOOD SAFETY AUTHORITY
The final dataset on total Hg and MeHg compiled from the EFSA data contained
24 628 samples of finfish and shellfish. Of these, 23 411 samples had results for total
Hg, whereas only 1 217 samples had results for MeHg.
The final dataset on dioxins and dl-PCBs compiled from the EFSA data contained
9 797 samples of finfish and shellfish. Of these, 9 390 samples had results for both
dioxins and dl-PCBs (all 29 congeners), 45 samples had results only for dioxins (17
congeners) and 362 samples had results only for dl-PCBs (12 congeners).

235
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

An overview of concentration data for total Hg and MeHg from the EFSA dataset, in
finfish and shellfish, from different regions around the world, is given in Table 7.7,
Table 7.8, Table 7.9, Table 7.10 and Table 7.11 (finfish), and in Table 7.17, Table 7.18,
Table 7.19, Table 7.20 and Table 7.21 (shellfish). An overview of the concentration
data from the EFSA dataset for dioxins and dl-PCBs is presented in Table 7.26,
Table 7.27, Table 7.28, Table 7.29 and Table 7.30 (finfish) and in Table 7.34,
Table 7.35, Table 7.36, Table 7.37 and Table 7.38 (shellfish).

7.4 MERCURY IN FINFISH AND SHELLFISH FROM DIFFERENT REGIONS

AROUND THE WORLD

7.4.1 MERCURY IN FINFISH FROM THE LITERATURE REVIEW

In the literature, THg results were found for 57 189 analysed samples of finfish,
while the number of MeHg results was much lower, 6 957. Of the analysed finfish,
1 340 and 122 samples for THg and MeHg, respectively, were from farmed fish,
while the rest were from wild stocks (55 621 and 6 685) or unknown (for instance,
sampled at market, with no information about geographic origin). More than
half of the analysed finfish originated from marine waters (n = 33 035 and 4 819,
for THg and MeHg, respectively), while 24 154 and 2 138 finfish samples analysed
for THg and MeHg, respectively, were from inland waters.

7.4.1.1 Farmed finfish from inland waters

Farmed finfish from inland waters had generally low concentrations of Hg (Table 7.1).
Only one species, Argyrosomus regius, meagre, from Europe, had a mean
concentration higher than 0.2 mg/kg wet weight. The rest of the species in this
category had mean concentrations of up to 0.053 mg/kg.

7.4.1.2 Farmed finfish from marine waters

Farmed fish species from the Mediterranean had higher Hg levels than farmed
marine fish from other areas (Table 7.2). The farmed fish from the different areas
did not include the same species. Farmed Thunnus thynnus, Atlantic bluefin tuna
from the Mediterranean, had a mean concentration of 0.6 mg/kg. In area 61, only
one species had a mean concentration of THg and MeHg of more than 0.2 mg/kg
(n = 5). Samples with sample numbers less than 10 were usually excluded from the
table, but in this case, they were kept because both THg and MeHg were analysed
in a number of species with n < 10.

236
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.1 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS (mg/kg WET WEIGHT)
IN MUSCLE TISSUE OF SPECIES OF FARMED FINFISH FROM DIFFERENT INLAND REGIONS
(FAO AREAS)

FAO AREA/SPECIES LATIN NAME N THg N MeHg

01 AFRICA, INLAND WATERS
Chanos chanos 12 0.012
03 SOUTH AMERICA, INLAND WATERS
Piaractus brachypomus 111 0.053
04 ASIA, INLAND WATERS
Misgurnus anguillicaudatus 15 0.050
Monopterus albus 15 0.049
Anguilla japonica 60 0.037
Carassius auratus 13 0.031
Aristichthys nobilis 35 0.027 23 0.002
Carassius carassius 47 0.023
Ctenopharyngodon idella 197 0.022
Pampus argenteus 87 0.015
Cyprinus carpio 158 0.014
Oreochromis mossambicus 73 0.014 1 0.002
Tinca tinca 23 0.010
Pseudosciaena crocea 72 0.009
Oncorhynchus mykiss 40 0.005
05 EUROPE, INLAND WATERS
Argyrosomus regius 16 0.251
Tinca tinca 20 0.048
Sparus aurata 24 0.044
Dicentrarchus labrax 16 0.044
Oncorhynchus mykiss 27 0.009

Note: N is the number of analytical samples, including both individual and composite samples. Cases where N < 10 are
excluded.

237
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.2 TABLE 7.2 MEAN TOTAL Hg (THg) AND METHYLMERCURY (MeHg) LEVELS IN MUSCLE TISSUE
OF SPECIES OF FARMED FINFISH FROM DIFFERENT MARINE REGIONS (FAO AREAS),
IN mg/kg WET WEIGHT

FAO AREA/SPECIES LATIN NAME N THg N MeHg

27 ATLANTIC, NORTHEAST
Salmo salar 14 0.019
37 MEDITERRANEAN AND BLACK SEA
Thunnus thynnus 40 0.600
Dicentrarchus labrax 39 0.322
Salmo trutta 30 0.238
Sparus aurata 37 0.218
61 PACIFIC, NORTHWEST
Epinephelus octofasciatus 5 0.235 5 0.209
Epinephelus coioides 5 0.162 5 0.143
Hypoplectrus indigo 7 0.147 7 0.090
Genyonemus lineatus 5 0.134 5 0.087
Ephippus orbis 6 0.123 6 0.060
Lateolabrax japonicus 5 0.115 5 0.105
Pagrus major 10 0.086 10 0.076
Acanthopagrus latus 5 0.072 5 0.055
Epinephelus awoara 2 0.065 2 0.042
Acanthopagrus schlegelii 5 0.048 5 0.033
Larimichthys crocea 5 0.048 5 0.030
Anguilla japonica 5 0.047 5 0.031
Paralichthys olivaceus 5 0.041 5 0.016
Cynoscion nebulosus 6 0.034 6 0.022
Larimichthys polyactis 8 0.023 8 0.009
Sardinella jussieu 4 0.020 4 0.006
Siganus fuscescens 5 0.013 5 0.004

Note: N is the number of analytical samples, including both individual and composite samples.

7.4.1.3 Wild-caught finfish from inland waters

THg results for wild-caught finfish from inland waters are given per genus and inland
region (FAO area) in Table 7.3 and for MeHg in Table 7.4. The highest mean value,
2.89 mg/kg, was found in Synodontis from Africa. However, the mean value was
particularly high due to an extreme value in Synodontis from one study from the Flag
Boshielo Dam in South Africa (Sara et al., 2017). Three genera from Africa had mean
concentrations above 0.5 mg/kg. In South American inland waters, 16 genera had
THg above 0.5 mg/kg, while in Europe and Asia, no genus had mean concentrations
of THg above 0.5 mg/kg. In North America, only one genus, Petromyzon, had THg
higher than 0.5 mg/kg. With regard to MeHg, two genera from North America and
none in South America, Asia or Europe had mean concentrations above 0.5 mg/kg
(Table 7.4). Higher MeHg concentrations than THg in North America could be due to
a selective analysis of MeHg in fish species or areas expected to have higher Hg levels.

238
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.3 MEAN TOTAL MERCURY (THg) LEVELS IN DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM INLAND REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

01 AFRICA INLAND WATERS Parasilurus 13 0.120
Synodontis 75 (121) 2.89 (0.1601) Erythroculter 112 0.117
Barbus 13 0.755 Leuciscus 79 0.114
Hydrocynus 10 0.510 Sperata 10 0.110
Schilbe 29 0.240 Hemibarbus 31 0.109
Marcusenius 26 0.204 Carassius 259 0.106
Esox 10 0.175 Thymallus 33 0.103
Chrysichthys 109 0.165 Cyprinus 123 0.079
Sander 12 0.151 Liza 35 0.074
Clarias 367 0.138 Puntius 35 0.073
Bagrus 21 0.130 Mystus 25 0.073
Lepomis 35 0.103 Glyptothorax 11 0.070
Ictalurus 10 0.092 Pangasius 67 0.069
Micropterus 26 0.088 Coptodon 40 0.066
Hemichromis 51 0.085 Lethrinus 61 0.063
Labeobarbus 11 0.083 Pseudobagrus 44 0.062
Distichondus 21 0.080 Hypophthalmichthys 230 0.060
Sarotherodon 20 0.072 Spinibarbus 20 0.056
Scardinius 22 0.070 Aristichthys 77 0.053
Citharinus 21 0.060 Lutjanus 21 0.051
Labeo 22 0.058 Coreius 35 0.050
Mugil 203 0.051 Hemiculter 28 0.046
Auchenoglanis 21 0.040 Epinephelus 26 0.046
Lates 205 0.038 Squaliobarbus 20 0.046
Oreochromis 330 0.029 Leiocassis 14 0.045
Cyprinus 11 0.024 Lates 30 0.040
04 ASIA INLAND WATERS Rhinogobio 22 0.035
Barbonymus 16 0.430 Chondrostoma 46 0.030
Chanodichthys 17 0.410 Oncorhynchus 40 0.026
Hemibagrus 20 0.410 Ctenopharyngodon 70 0.019
Puntioplites 17 0.310 Parabramis 12 0.019
Gerres 14 0.300 Culterichthys 18 0.011
Cyclocheilichthys 111 0.275 Oreochromis 15 0.010
Lota 29 0.209 Siganus 21 0.009
Esox 65 0.206 Rhynchocypris 30 0.008
Channa 23 0.194 Hypomesus 13 0.008
Toxotes 23 0.190 05 EUROPE INLAND WATERS
Silurus 76 0.180 Perca 1922 0.333
Leiognathus 24 0.170 Aspius 24 0.312
Brachymystax 35 0.147 Silurus 184 0.276
Gymnocypris 58 0.126 Esox 143 0.241
Pelteobagrus 96 0.123 Sander 151 0.222

239
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.3 MEAN TOTAL MERCURY (THg) LEVELS IN DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM INLAND REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Osmerus 99 0.220 Gambusia (w) 34 0.144
Anguilla 314 0.178 Morone (w+m) 84 0.136
Scardinius 60 0.163 Oncorhynchus (w+m) 398 0.122
Cyprinus 197 0.158 Chirostoma 16 0.113
Carassius 92 0.155 Lepomis (w+m) 312 0.110
Abramis 457 0.135 Perca (w+m) 269 0.105
Stizostedion 20 0.134 Coregonus (w+m) 548 0.105
Alburnus 94 0.129 Eupomotis 10 0.076
Hypophthalmichthys 17 0.129 Alosa 31 0.072
Squalius 276 0.118 Neogobius (w+m) 93 0.067
Rutilus 305 0.110 Semotilus 88 0.061
Leucos 80 0.099 Osmerus (w+m) 80 0.058
Leuciscus 150 0.096 Notomigonus 50 0.050
Alosa 21 0.090 Oreochromis 99 0.050
Salvelinus 136 0.089 Cottus 10 0.048
Salmo 44 0.073 03 SOUTH AMERICA INLAND WATERS
Coregonus 119 0.070 Pseudopimelodus 25 1.40
Eleginus 10 0.063 Hydrolycus/Raphiodon 21 1.20
Chondrostoma 89 0.044 Calophysus 43 0.949
Lota 10 0.035 Rhaphiodon 181 0.868
Oncorhynchus 15 0.021 Acestrorhynchus 73 0.848
02 NORTH AMERICA INLAND WATERS Pellona 173 0.790
Petromyzon (w+m) 116 1.39 Salminus 53 0.783
Sander (w+m) 908 0.425 Pygocentrus 47 0.739
Esox (w+m) 983 0.406 Cichla 464 0.679
Rhinichthys 223 0.400 Ageneiosus 120 0.670
Hiodon 205 0.314 Phractocephalus 42 0.630
Micropterus 652 0.304 Sternopygus 25 0.623
Lota 92 0.293 Hoplias 253 0.583
Pomoxis 100 0.229 Pinirampus 28 0.576
Salvelinus (w+m) 1516 0.214 Plagioscion 119 0.568
Acantharchus (w) 18 0.213 Pseudoplatystoma 87 0.530
Moxostoma 123 0.200 Hypophthalmus 163 0.486
Ictalurus 295 0.192 Sorubim/Hemisorubim 10 0.480
Catostomus (w+m) 154 0.172 Odontesthes 11 0.465
Cyprinus 56 0.168 Anodus 58 0.430
Ameiurus (w+m) 80 0.165 Serrasalmus 215 0.415
Prosopium 30 0.160 Ctenolucius 28 0.385
Aphredoderous (w) 40 0.156 Rhamdia 99 0.383
Notropis 40 0.150 Potamorhina 225 0.331
Porogonias 10 0.150 Caquetaia 160 0.330
Gasterosteus 360 0.147 Trachelyopterus 25 0.320

240
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.3 MEAN TOTAL MERCURY (THg) LEVELS IN DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM INLAND REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Triportheus 92 0.275 Pimelodella 21 0.100
Centropomus 14 0.265 Astyanax 172 0.097
Potamotrygon 14 0.260 Pachyurus 37 0.092
Hoplerythrinus 36 0.255 Prochilodus 508 0.092
Tocantinsia 17 0.240 Mugil 20 0.081
Boulengerella 36 0.230 Bivibranchia 13 0.080
Astronotus 14 0.216 Brycon 73 0.074
Pimelodus 106 0.210 Geophagus 63 0.065
Schizodon 81 0.174 Hypostomus 115 0.064
Hemiodus 68 0.172 Hemiancistrus 14 0.060
Psectrogaster 90 0.156 Oligosarcus 30 0.059
Megalonema/Pimelodus 21 0.150 Cyphocharax 42 0.056
Semaprochilodus 32 0.141 Baryancistrus 10 0.030
Spatuloricaria 16 0.120 Pterygoplichthys 13 0.029
Arius 15 0.120 Myloplus 34 0.023
Hassar 15 0.120 Tometes 15 0.010
Andinoacara 41 0.117 Hypomasticus 10 0.010
Curimata 38 0.109
Leporinus 95 0.109
Mylossoma 261 0.106

Notes: N is the number of analytical samples, including both individual and composite samples. All analysed edible
tissues are included. When whole fish (w) or both whole fish and muscle (w+m) were analysed, this is marked
after the genus name. When nothing is specified, muscle tissue was analysed. Cases where N < 10 are excluded.
1
Excluding fish from one study (N = 63, mean THg 11.1 mg/kg).

241
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.4 MEAN METHYLMERCURY (MeHg) LEVELS IN GENUS OF WILD-CAUGHT FINFISH FROM
DIFFERENT INLAND REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

FAO AREA/GENUS N MeHg FAO AREA/GENUS N MeHg

02 SOUTH AMERICA INLAND WATERS 04 ASIA INLAND WATERS
Rhinichthys 223 0.230 Gymnocypris 58 0.109
Perca 44 0.229 Carassius 48 0.049
Acipenser 35 0.180 Erythroculter 49 0.033
Aphredoderous 40 0.150 Coreius 12 0.026
Esox 49 0.131 Cyprinus 27 0.023
Neogobius 20 0.107 Aristichthys 48 0.023
Catostomus 35 0.095 Hypophthalmichthys 120 0.017
Semotilus 19 0.074 Squaliobarbus 20 0.017
Cyphocharax 12 0.050 Ctenopharyngodon 23 0.005
Salvelinus 20 0.038 05 EUROPE INLAND WATERS
Lepomis 50 0.032 Sander 13 0.399
03 NORTH AMERICA INLAND WATERS Esox 17 0.348
Pseudopimelodus 13 1.06 Aspius 24 0.309
Ageneiosus 50 0.789 Perca 17 0.243
Trachelyopterus 17 0.423 Abramis 39 0.125
Sternopygus 15 0.409 Rutilus 14 0.092
Ctenolucius 20 0.347 Leuciscus 143 0.088
Hoplias 58 0.272 Cyprinus 12 0.083
Caquetaia 57 0.268
Rhamdia 46 0.261
Astyanax 53 0.175
Leporinus 34 0.174
Andinoacara 25 0.117
Pimelodus 16 0.093
Prochilodus 106 0.078
Hypostomus 55 0.064

Notes: N is the number of analytical samples, including both individual and composite samples. All analysed edible
tissues are included. When whole fish (w) or both whole fish and muscle (w+m) were analysed, this is marked after the
genus name. When nothing is specified, muscle tissue was analysed. Cases where N < 10 are excluded.

242
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.4.1.4 Wild-caught finfish from marine waters

Finfish sampled from wild stocks (or from unknown source) in marine waters made
up the largest volume of results published in the literature for both THg and MeHg.
Table 7.5 and Table 7.6 provide results per FAO area and genus.
Almost all FAO areas included several genera with mean concentrations higher
than 0.5 mg/kg, exceptions being 18 Arctic Sea; 58 Indian Ocean, Antarctic and
Southern; 88 Pacific, Antarctic; 61 Pacific, Northwest; and 67 Pacific, Northeast. The
highest mean THg concentration was 4.15 mg/kg, measured in Makaira nigricans
(blue marlin), from Mexico, with only 16 samples analysed. Makaira from the
Caribbean (n = 62) had a lower mean value of 0.91 mg/kg. Several areas included a
number of genera with mean THg concentrations higher than 1 mg/kg. The genera
included a number of sharks (Carcharinus, Rhizopronodon, Galeus, Triakis, Alopias,
Squalus, Scylorhinus, Deania), the infamous predator barracuda (Sphyraena) as well
as the genera Lophius (anglers or monkfishes), Helicolenus (blackbelly rosefish),
Micropterus (bass) and Johnius (croaker). It is also worth noticing that swordfish
(only one species, Xiphias gladius) had relatively high mean concentrations in a
number of different FAO areas, and the same applies to tuna species (Thunnus spp.).
However, tunas generally had lower mean concentrations than swordfish in areas
where both were analysed. As mentioned earlier, fewer samples and genera by far
were analysed for MeHg than for THg (Table 7.6). Only two genera from area 27,
Anguilla (eel) and Brosme (tusk or cusk), and one genus from area 31, Epinephelus
(groupers), had mean MeHg concentrations above 0.5 mg/kg. These species had
higher mean levels of MeHg than of THg, but a lower number of samples were
analysed, and these may have been sampled specifically in areas known for high
Hg levels.

243
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.5 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

04 ASIA INLAND WATERS (CASPIAN SEA) Xiphias 25 0.475
Clupeonella (w) 1050 0.014 Anguilla 111 0.424
18 ARCTIC SEA Thunnus 47 0.338
Gadus 10 0.140 Mullus 17 0.338
Myoxocephalus 56 0.095 Galeorhinus 124 0.315
Stenodus 50 0.081 Esox 21 0.296
Coregonus 125 0.056 Leucoraja 44 0.280
Microgadus 14 0.032 Merluccius 124 0.229
Boreogadus 40 0.030 Platichthys 215 0.221
Platichthys 28 0.026 Perca 205 0.166
Oncorhynchus 22 0.024 Reinhardtius 475 0.142
Ammodytes 13 0.020 Pelecus 60 0.127
Mallotus 11 0.010 Trachurus 79 0.121
Clupea 29 0.005 Solea 41 0.118
21 ATLANTIC, NORTHWEST Anarhichas 93 0.115
Rhizoprionodon 10 1.64 Stizostedion 14 0.111
Thunnus 519 0.650 Scyliorhinus 11 0.108
Pomatomus 80 0.323 Diplodus 10 0.105
Caulolatilus 64 0.300 Limanda 379 0.089
Squalus 159 0.284 Abramis 26 0.085
Cottunculus 10 0.270 Gadus 535 0.083
Catostomus 10 0.143 Capros 11 0.080
Gadus 30 0.134 Rutilus 11 0.077
Coregonus 27 0.122 Scophthalmus 22 0.071
Myoxocephalus 35 0.121 Zoarces 18 0.060
Lopholatilus 484 0.092 Salmo 12 0.056
Gasterosteus (w) 14 0.078 Katsuwonus 16 0.050
Osmerus (w+m) 25 0.074 Micromesistius 85 0.048
Salmo 12 0.073 Pleuronectes 51 0.047
Pleuronectoide 20 0.068 Pollachius 22 0.046
Salvelinus 170 0.066 Clupea 167 0.032
Reinhardtius 10 0.048 Myoxocephalus 58 0.028
Menidia (w) 193 0.040 Neogobius 90 0.025
27 ATLANTIC, NORTHEAST Sprattus (w) 45 0.015
Deania 21 1.80 31 ATLANTIC, WESTERN CENTRAL
Mora 30 0.900 Carcharhinus 46 1.96
Lamna 33 0.840 Micropterus 44 1.35
Molva 150 0.796 Rhizoprionodon 188 1.22
Isurus 48 0.740 Sphyraena 27 1.06
Etmopterus 10 0.700 Makaira 62 0.910
Brosme 433 0.596 Anisotremus 15 0.769
Prionace 40 0.520 Bagre 24 0.730

244
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.5 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Seriola 43 0.575 Sphyraena 13 0.194
Lepomis 33 0.542 Trichiurus 13 0.178
Sciaenops 30 0.508 Brachydeuterus 51 0.150
Genidens 18 0.450 Pseudupeneus 10 0.080
Epinephelus 245 0.399 Lutjanus 16 0.057
Scomberomorus 18 0.394 Chloroscombrus 10 0.026
Caulolatilus 62 0.380 Trachurus 10 0.025
Rhinoptera 25 0.370 Sardinella 51 0.024
Kajikia 45 0.350 Solea 45 0.016
Haemulon 23 0.316 37 MEDITERRANEAN AND BLACK SEA
Acathocybium 23 0.300 Galeus 15 2.06
Centropomus 92 0.296 Scyliorhinus 27 1.84
Cynoscion 12 0.274 Helicolenus 14 1.15
Caranx 32 0.251 Lophius 18 1.14
Squatina 94 0.240 Lepidorhombus 13 0.860
Mycteroperca 326 0.239 Euthynnus 89 0.740
Ariopsis 18 0.239 Xiphias 125 0.694
Istiophorus 38 0.220 Diplodus 153 0.688
Coryphaena 91 0.200 Phycis 24 0.606
Lutjanus 452 0.164 Thunnus 552 0.594
Ocyurus 12 0.162 Trachinus 26 0.524
Elops 58 0.161 Pagellus 215 0.482
Mobula 15 0.158 Zosterisessor 209 0.434
Archosargus 17 0.158 Scorpaena 43 0.420
Thunnus 11 0.150 Microchirus 29 0.397
Lachnolaimus 12 0.132 Sarda 48 0.369
Pterois 366 0.103 Chelidonichthys 35 0.362
Mugil 91 0.076 Squalus 68 0.330
Brevoortia 10 0.063 Raja 94 0.315
Poecilia 11 0.055 Mullus 1 270 0.294
Trinectes 12 0.054 Sparus 258 0.216
Eugerres 53 0.054 Micromesistius 42 0.207
Fundulus 19 0.053 Epinephelus 10 0.185
Eucinostomus 23 0.036 Arnoglossus 30 0.184
Acanthurus 20 0.036 Scomber 292 0.170
Anchoa 13 0.032 Trachurus 375 0.160
Cyprinodon 11 0.030 Boops 40 0.141
Cathorops 35 0.018 Scophthalmus 28 0.137
34 ATLANTIC, EASTERN CENTRAL Katsuwonus 26 0.137
Thunnus 36 0.777 Upeneus 23 0.130
Xiphias 17 0.672 Dicentrarchus 26 0.126
Pseudotolithus 14 0.410 Merluccius 135 0.112

245
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.5 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Sardina 718 0.108 Umbrina 12 0.104
Spicara 13 0.104 Odontesthes 46 0.101
Trisopterus 12 0.101 Atherinella 16 0.097
Etrumeus 30 0.100 Diapterus 11 0.078
Gobius 24 0.096 Orthopristis 62 0.074
Lithognathus 89 0.087 Percophis 43 0.074
Nemipterus 26 0.080 Priacanthus 39 0.061
Saurida 41 0.070 Merluccius 12 0.046
Solea 96 0.067 Sardinella 29 0.033
Sprattus 120 0.059 Brevoortia 10 0.032
Merlangius 362 0.059 Prochilodus 16 0.028
Engraulis 1 204 0.057 Mugil 34 0.028
Belone 50 0.050 Paralonchurus 12 0.026
Mugil 106 0.042 Paralichthys 23 0.024
Sardinella 41 0.040 Aetobatus 12 0.006
Platichthys 10 0.039 47 ATLANTIC, SOUTHEAST
Psetta 67 0.033 Xiphias 17 0.816
Liza 29 0.031 Merluccius 10 0.630
Sepia 23 0.030 Genypterus 10 0.580
Marsupenaeus 39 0.027 Lophius 1 017 0.388
Loligo 35 0.020 Sardinella 10 0.030
Spondylus 30 0.018 48 ATLANTIC, ANTARCTIC
Siganus 10 0.017 Dissostichus 39 0.680
Myctophidae 19 51 INDIAN OCEAN, WESTERN
41 ATLANTIC, SOUTHWEST Johnius 81 1.08
Squalus 32 1.37 Xiphias 54 0.971
Cynoscion 139 1.24 Acanthocybium 12 0.902
Carcharhinus 21 0.719 Sphyrna 12 0.727
Rhizoprionodon 47 0.695 Carcharhinus 89 0.724
Xiphias 23 0.680 Epinephelus 78 0.430
Centropomus 17 0.655 Thunnus 210 0.418
Ramnogaster 51 0.540 Sphyraena 56 0.378
Sphyrna 26 0.485 Katsuwonus 37 0.363
Thunnus 88 0.381 Grammoplites 24 0.286
Genidens 45 0.353 Alepes 20 0.282
Trichiurus 10 0.239 Euthynnus 23 0.271
Hypanus 29 0.178 Cynoglossus 74 0.263
Micropogonias 90 0.173 Platycephalus 16 0.255
Mustelus 83 0.156 Lethrinus 100 0.235
Cathorops 26 0.126 Scomberomorus 140 0.217
Lycengraulis 28 0.106 Lutjanus 39 0.215
Eugerres 45 0.106 Pampus 27 0.202

246
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.5 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Otolithes 34 0.182 61 PACIFIC, NORTHWEST
Parastromateus 10 0.180 Liza 27 0.492
Pomadasys 11 0.170 Sebastes 14 0.452
Psettodes 10 0.151 Prionace 10 0.420
Mugil 21 0.131 Sebastiscus 76 0.290
Coryphaena 30 0.109 Nematalosa 15 0.267
Rhabdosargus 14 0.100 Carcharhinus 17 0.250
Auxis 18 0.092 Gadus 66 0.226
Rhizoprionodon 12 0.089 Pseudolabrus 11 0.209
Rastrelliger 26 0.084 Terapon 34 0.177
Liza 70 0.075 Acanthopagrus 69 0.118
Elagatis 16 0.062 Konosirus 36 0.117
Lobotes 10 0.061 Muraenesox 16 0.113
Carangoides 15 0.050 Pleuronectiformes 11 0.100
Coilia 21 0.045 Lateolabrax 48 0.091
Canthidermis 18 0.028 Epinephelus 166 0.091
Chiloscyllium 84 0.025 Triaenopogon 62 0.089
Uraspis 14 0.018 Plectorhinchus 20 0.086
Kyphosus 32 0.007 Trichiurus 42 0.083
Aluterus 10 0.007 Nemipterus 118 0.075
Gerres 30 Priacanthus 25 0.075
57 INDIAN OCEAN, EASTERN Parargyrops 10 0.074
Xiphias 94 0.612 Paralichthys 26 0.071
Platycephalus 51 0.366 Eupleurogrammus 30 0.070
Thunnus 112 0.300 Sparus 31 0.068
Mobula 15 0.190 Branchiostegus 13 0.067
Sphyraena 16 0.135 Sillago 56 0.064
Mystus 10 0.121 Pholis 24 0.061
Eleutheronema 72 0.077 Callionymus 18 0.059
Daysciaena 72 0.062 Megachasma 27 0.058
Mugil 75 0.041 Thryssa 13 0.055
Etroplus 75 0.040 Inimicus 12 0.053
Megalaspis 10 0.035 Lutjanus 28 0.051
Caranx 17 0.014 Scomberomorus 50 0.049
Sardinella 15 0.013 Plotosus 12 0.048
Scomberomorus 10 0.011 Cynoglossus 129 0.044
Filimanus 13 0.005 PLATYCEPHALUS 69 0.043
Siganus 10 0.002 Synechogobius 19 0.042
58 INDIAN OCEAN, ANTARCTIC Evynnis 18 0.042
Macrourus 15 0.336 Johnius 64 0.040
Dissostichus 57 0.321 Saurida 33 0.036

247
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.5 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Siniperca 11 0.036 Hapalogenys 12
Argyrosomus 25 0.035 67 PACIFIC, NORTHEAST
Pennahia 18 0.034 Raja 20 0.340
Larimichthys 29 0.034 Hemilepidotus 75 0.340
Nibea 12 0.034 Lepidopsetta 60 0.244
Trachurus 30 0.033 Malacoccottus 55 0.224
Trachinotus 140 0.032 Atheresthes 128 0.152
Lagocephalus 11 0.031 Gadus 225 0.135
Cirrhinus 12 0.031 Sebastes 158 0.118
Hexagrammos 16 0.031 Beringraja 20 0.090
Pseudopleuronectes 12 0.028 Pleurogrammus 221 0.045
Morone 11 0.028 71 PACIFIC, WESTERN CENTRAL
Leiognathus 39 0.028 Xiphias 20 0.646
Aristichthys 13 0.026 Acanthopagrus 58 0.335
Setipinna 18 0.024 Toxotes 16 0.260
Pseudosciaena 27 0.023 Decapterus 15 0.250
Siganus 144 0.023 Gerres 10 0.173
Dendrophysa 10 0.023 Lates 17 0.101
Thamnaconus 30 0.022 Megalaspis 30 0.066
Repomucenus 18 0.022 Thunnus 61 0.053
Acanthogobius 12 0.022 Siganus 10 0.030
Carassius 16 0.021 Manta 15 0.009
Gerres 32 0.021 77 PACIFIC, EASTERN CENTRAL
Osteomugil 15 0.019 Makaira 16 4.15
Chaeturichthys 24 0.018 Triakis 10 1.27
Enchelyopus 12 0.016 Kajikia 17 0.880
Pampus 51 0.016 Urolophus 12 0.770
Engraulis 18 0.016 Nezumia 43 0.748
Lates 14 0.015 Dasyatis 43 0.697
Mallotus 12 0.014 Sebastes 11 0.642
Sardinella 26 0.014 Sphyrna 156 0.590
Salmonidae 13 0.014 Thunnus 363 0.566
Acipenser 10 0.014 Istiophorus 67 0.560
Mugil 47 0.012 Alopias 63 0.560
Pneumatophorus 10 0.008 Etelis 75 0.500
Harpadon 62 0.007 Lepophidium 14 0.490
Odontamblyopus 12 0.005 Hydrolagus 140 0.490
Pangasius 13 0.005 Coelorinchus 26 0.487
Miichthys 11 0.005 Sebastolobus 19 0.466
Collichthys 21 0.004 Aprion 60 0.450
Ctenopharyngodon 14 0.004 Seriola 11 0.437
Decapterus 13 Cephalurus 12 0.434

248
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.5 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT FINFISH
FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N THg FAO AREA/GENUS N THg

Prionace 30 0.390 Allocyttus 20 0.100
Merluccius 241 0.334 Thyrsites 10 0.060
Symphurus 20 0.305 87 PACIFIC, SOUTHEAST
Carcharhinus 30 0.280 Carcharhinus 27 1.53
Sciades 24 0.266 Sphyrna 19 1.28
Platyrhinoidis 11 0.263 Alopias 24 1.07
Haemulopsis 39 0.234 Prionace 132 0.985
Zapteryx 88 0.226 Coryphaena 46 0.847
Hoplopagrus 12 0.218 Thunnus 68 0.801
Caranx 25 0.217 Isurus 22 0.608
Sphoeroides 15 0.203 Xiphias 42 0.551
Micropogonias 51 0.159 Lampris 15 0.482
Pristipomoides 60 0.150 Mustelus 11 0.338
Gerres 15 0.145 Brotula 10 0.276
Diapterus 110 0.135 Paralichthys 11 0.196
Pseudupeneus 32 0.110 Cynoscion 14 0.195
Coryphaena 79 0.103 Cilus 11 0.112
Pseudobatos 97 0.082 Macruronus 12 0.105
Ancylopsetta 10 0.080 Sarda 15 0.097
Pomadasys 57 0.079 Mugil 10 0.009
Scomberomorus 22 0.072 88 PACIFIC, ANTARCTIC
Mustelus 163 0.068 Macrourus 15 0.365
Larimus 11 0.060 Dissostichus 89 0.210
Myliobatis 83 0.058 UNKNOWN FAO AREA
Achirus 36 0.058 Xiphias 90 0.869
Caulolatilus 22 0.053 Thunnus 17 0.420
Lutjanus 36 0.052 Pagrus 20 0.310
Mugil 50 0.047 Lates 12 0.300
Hemicaranx 10 0.040 Lutjanus 35 0.196
Trachinotus 30 0.030 Macruronus 12 0.160
Pronotogrammus 10 0.024 Platycephalus 22 0.160
Rhincodon 72 0.022 Galeorhinus 10 0.130
Balistes 19 0.018 Thunnus 7 0.101
81 PACIFIC, SOUTHWEST Scomberomorus 36 0.088
Genypterus 120 0.530 Monacanthidae 20 0.070
Hoplostethus 101 0.510 Sillago 10 0.060
Rexea 10 0.240 Scomberomorus 11 0.040
Pseudocyttus 20 0.150 Merluccius 3 0.023

Notes: N is the number of analytical samples, including both individual and composite samples. THg: total Hg.
When whole fish (w) or both whole fish and muscle (w+m) were analysed, this is marked after the genus name.
When nothing is specified, muscle tissue was analysed.

249
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.6 MEAN METHYLMERCURY (MeHg) LEVELS IN WILD-CAUGHT FINFISH FROM DIFFERENT MARINE
REGIONS (FAO AREAS), BY GENUS, IN mg/kg WET WEIGHT

FAO AREA/GENUS N MeHg FAO AREA/GENUS N MeHg

18 ARCTIC SEA Macrourus 15 0.145
Gadus 10 0.140 Dissostichus 50 0.106
Myoxocephalus 20 0.125 61 PACIFIC, NORTHWEST
Stenodus 38 0.075 Carcharhinus 17 0.164
Coregonus 41 0.061 Acanthopagrus 48 0.107
Boreogadus 40 0.030 Terapon 34 0.102
Oncorhynchus 22 0.023 Plectorhinchus 10 0.100
Platichthys 22 0.021 Muraenesox 12 0.086
Ammodytes 13 0.020 Nemipterus 96 0.075
Mallotus 11 0.010 Sillago 45 0.067
Clupea 29 0.003 Lateolabrax 14 0.060
21 ATLANTIC, NORTHWEST Lutjanus 15 0.060
Squalus 102 0.378 Argyrosomus 21 0.052
Cottunculus 10 0.260 Hapalogenys 12 0.051
Fundulus 290 0.037 Inimicus 12 0.050
27 ATLANTIC, NORTHEAST Parargyrops 10 0.047
Anguilla 29 0.729 Evynnis 18 0.046
Brosme 137 0.616 Johnius 63 0.046
Stizostedion 14 0.095 Decapterus 13 0.045
Rutilus 10 0.062 Sebastiscus 16 0.044
Gadus 10 0.054 Konosirus 36 0.044
Abramis 10 0.030 Sparus 24 0.043
Clupea 10 0.015 Callionymus 12 0.043
Sprattus 10 0.006 Epinephelus 133 0.042
Trachurus 40 Pholis 24 0.042
31 ATLANTIC, WESTERN CENTRAL Trichiurus 19 0.042
Epinephelus 168 0.755 Paralichthys 16 0.042
37 MEDITERRANEAN AND BLACK SEA Triaenopogon 62 0.038
Xiphias 45 0.495 Cynoglossus 99 0.036
Sardina 44 0.033 Thryssa 13 0.036
Myctophidae 19 0.019 Leiognathus 32 0.035
41 ATLANTIC, SOUTHWEST Platycephalus 62 0.035
Hypanus 11 0.019 Scomberomorus 37 0.035
47 Atlantic, Southeast Nibea 12 0.033
Lophius 650 0.130 Synechogobius 18 0.032
51 INDIAN OCEAN, WESTERN Pennahia 18 0.030
Lethrinus 45 0.320 Saurida 26 0.029
Gerres 30 0.040 Lagocephalus 11 0.029
57 INDIAN OCEAN, EASTERN Hexagrammos 16 0.028
Xiphias 20 0.013 Pseudopleuronectes 12 0.026
Caranx 11 0.005 Larimichthys 24 0.024
Filimanus 13 0.004 Trachinotus 18 0.023

250
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.6 MEAN METHYLMERCURY (MeHg) LEVELS IN WILD-CAUGHT FINFISH FROM DIFFERENT MARINE
REGIONS (FAO AREAS), BY GENUS, IN mg/kg WET WEIGHT (cont.)

FAO AREA/GENUS N MeHg FAO AREA/GENUS N MeHg

Setipinna 18 0.023 81 PACIFIC, SOUTHWEST
Repomucenus 18 0.021 Genypterus 120 0.460
Eupleurogrammus 24 0.019 Hoplostethus 101 0.430
Dendrophysa 10 0.019 Rexea 10 0.220
Pampus 38 0.019 Pseudocyttus 20 0.140
Liza 14 0.019 Allocyttus 20 0.100
Osteomugil 14 0.018 Thyrsites 10 0.060
Thamnaconus 26 0.017 88 PACIFIC, ANTARCTIC
Chaeturichthys 24 0.017 Macrourus 15 0.184
Nematalosa 13 0.016 Dissostichus 52 0.035
Pseudosciaena 14 0.015 57 INDIAN OCEAN, E + 81 PACIFIC, SW
Harpadon 18 0.014 Thunnus 17 0.410
Enchelyopus 12 0.014 Lates 12 0.260
Siganus 129 0.014 Pagrus 20 0.250
Engraulis 18 0.011 Platycephalus 22 0.150
Sardinella 25 0.011 Galeorhinus 10 0.120
Mugil 22 0.010 Macruronus 12 0.110
Monacanthidae 20 0.060
Sillago 10 0.050
Scomberomorus 11 0.030

Note: N is the number of analytical samples, including both individual and composite samples. Cases where N < 10 are
excluded.

7.4.2 MERCURY IN FINFISH FROM DATA FROM THE EUROPEAN FOOD SAFETY
AUTHORITY
In the data from EFSA, THg and MeHg values were reported using codes for species
or species groups, rather than the less ambiguous Latin names (which are used in
the literature data). Because of this, some species groups may include a wide range
of species which may have very different Hg levels. Furthermore, some species may
be present both with their specific species names and as part of a larger group of
species. For instance, the species named Atlantic salmon could be included also in
the species group “salmons”. The EFSA data includes samples analysed in Europe,
but includes both samples of fish of European origin and fish imported to Europe
from other countries. In the EFSA data for finfish, only data for muscle tissue was
included.
In the dataset for Hg in finfish reported to EFSA and included here, there are 17 799
results for THg and 1 006 results for MeHg. The majority of the samples reported
for THg (n = 15 602), were from marine waters, while 1 755 were from inland waters.
For 412 samples of finfish, it was not clear whether the samples were of marine or
inland origin, and these were grouped together with the marine fish in the tables.

251
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Most of the fish samples (n = 11 038) originated from wild stocks, whereas 1 276
samples were from aquaculture production. For 5 485 samples, it was not specified
whether the samples originated from wild stocks or aquaculture, and these were
included with the wild-caught fish in the results tables.

7.4.2.1 Farmed and wild finfish from inland waters

The samples were classified as farmed, wild-caught or “unspecified”, and the latter
constituted the largest group. In Table 7.7, results are given for species or species
groups of freshwater finfish, independent of area or whether produced by farming,
wild capture or unknown. Barbs had the highest mean THg concentration with 0.621
mg/kg, followed by pike, with a mean of 0.424 mg/kg. The analysed barbs were
all from wild stocks or unspecified in Europe (Table 7.8), whereas pike included
both wild, farmed and unspecified in Europe. The unspecified pike samples had the
highest mean THg concentration, analysed in very few samples (mean 1.12 mg/kg,
n = 7). In general, fewer numbers by far were reported for MeHg than for THg,
and mean concentrations were mostly lower for MeHg than for THg (Table 7.7).

TABLE 7.7 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN MUSCLE TISSUE
OF SPECIES OR SPECIES GROUPS OF FARMED AND WILD-CAUGHT FINFISH FROM INLAND
WATERS (ALL REGIONS), IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP N THg N MeHg

Barbs 32 0.621
Pike 44 0.424 1 0.078
Freshwater bream, Europe 172 0.198
Eel, European 17 0.159
Roaches 42 0.141
Perch 154 0.138 8 0.101
Pike-perch 23 0.130
River eels 125 0.114 5 0.065
White crappie 13 0.101
Carps, barbels and other cyprinids 555 0.096 105 0.021
Pangas catfishes 275 0.090 29 0.023
Catfishes (freshwater) 48 0.078 1 0.045
Salmons, trouts, smelts 170 0.028 25 0.021
African catfish 13 0.026
Char 13 0.023 10 0.018
Tilapias and similar 42 (41 )1
0.103 (0.006 )
1
6 0.020

Note: N is the number of analytical samples and may include both individual and composite samples. For THg, cases
where N < 10 are excluded.
1
Excluding one sample with THg-concentration of 4.1 mg/kg ww.

252
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.8 MEAN TOTAL MERCURY (THg) LEVELS IN MUSCLE TISSUE OF SPECIES OR SPECIES GROUPS
OF WILD-CAUGHT, FARMED OR UNSPECIFIED FINFISH FROM DIFFERENT INLAND REGIONS
(FAO AREAS), IN MG/KG WET WEIGHT

WILD FARMED UNSPECIFIED TOTAL SUM

SPECIES/SPECIES GROUP
N THg N THg N THg N THg
01 AFRICA INLAND WATERS
Freshwater bream, Europe 15 0.502 15 0.502
Perch 55 0.107 55 0.107
04 ASIA INLAND WATERS
Pangas catfishes 1 0.010 5 0.022 190 0.118 196 0.115
River eels 10 0.061 10 0.061
Perch 5 0.029 10 0.044 15 0.039
Tilapias and similar 33 0.004 33 0.004
05 EUROPE INLAND WATERS
Barbs 15 0.884 15 0.423 30 0.654
Pikes 34 0.282 3 0.402 7 1.12 44 0.424
Perch 25 0.157 2 0.073 47 0.201 74 0.183
Eel, European 16 0.158 1 0.164 17 0.159
Freshwater bream, Europe 71 0.197 7 0.089 67 0.121 145 0.157
Pike-perch 4 0.252 1 0.174 12 0.103 17 0.142
Roaches 13 0.136 1 0.085 28 0.146 42 0.141
River eels 113 0.117 4 0.066 117 0.116
White crappie 1 0.125 11 0.099 12 0.101
Carps, barbels and other cyprinids 112 0.173 306 0.047 136 0.142 554 0.096
Catfishes (freshwater) 7 0.361 4 0.018 36 0.032 47 0.080
Salmons, trouts, smelts 12 0.043 64 0.023 93 0.029 169 0.027
Char 13 0.023 13 0.023
Pangas catfishes 21 0.005 21 0.005
SPAIN, UNSPECIFIED
Freshwater bream 3 0.106 8 0.223 11 0.191
Perch 12 0.130 12 0.130
Pangas catfishes 26 0.039 26 0.039
UNKNOWN AREA
Pangas catfishes 28 0.027 0.027

Note: N is the number of analytical samples and may include both individual and composite samples. Cases where
N < 10 are excluded.

253
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

7.4.2.2 Farmed finfish from marine waters

The data on Hg in farmed marine finfish reported to EFSA included much fewer
samples than the data on marine finfish from wild stocks. The results for THg and
MeHg are given in Table 7.9. All species or species groups had low concentrations of
THg and MeHg (well below 0.2 mg/kg). Sea bream from area 27 Northeast Atlantic
had the highest mean THg and MeHg levels, while Atlantic salmon had the lowest.

TABLE 7.9 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN MUSCLE TISSUE OF
SPECIES OR SPECIES GROUPS OF FARMED FINFISH FROM DIFFERENT MARINE
(OR UNSPECIFIED) REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP N THg N MeHg

27 ATLANTIC, NORTHEAST
Sea bream 17 0.128 9 0.127
Sea bass 10 0.103 1 0.050
Salmons 23 0.044
Trouts 49 0.027 3 0.021
Rainbow trout 53 0.024 6 0.026
Atlantic salmon 528 0.019 96 0.020
37 MEDITERRANEAN AND BLACK SEA
Sea bass 34 0.081
Gilthead seabream 14 0.068
Trouts 20 0.047
FRANCE, UNSPECIFIED
Rainbow trout 20 0.048
Salmons 35 0.038

Notes: N is the number of analytical samples and may include both individual and composite samples. Cases where N <
10 for THg are excluded.

7.4.2.3 Wild-caught finfish from marine waters

The highest mean THg concentration in wild-caught finfish was reported in sharks
from Spain (FAO area not specified), at 1.74 mg/kg (Table 7.10). In many areas,
however, swordfish was the species or species group with the highest mean THg level,
varying between 0.574 mg/kg in Morocco (FAO area unspecified) and 1.55 mg/kg
in area 87 Southeast Pacific. Other species or species groups with mean THg
concentrations above 0.5 mg/kg included sharks from Northeast Atlantic (FAO area 27);
anglerfish, monkfish and stargazers (FAO area 37); European conger from the
Mediterranean and Black Sea (also FAO area 37); butterfish from Western Central
Pacific; yellowfin tuna from Eastern Central and Southeast Pacific (FAO areas 77 and 87);
and blue shark from Spain (FAO area unspecified). Tunas were classified as “tuna” or
“yellowfin tuna”, and for several areas results for both groups were reported. Mean THg
concentrations in the diverse species group “tuna” varied between 0.14 and 0.81 mg/kg,
perhaps because different tuna species were analysed. However, yellowfin tuna, a
large and globally migrating tuna species, also showed large variations between FAO

254
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

areas, from 0.21 mg/kg in Western-Central Pacific (FAO area 71, n = 22) to 0.66 mg/kg
in Southeast Pacific (FAO area 87, n = 14). The number of samples analysed in each
of these areas was relatively low and the geographic differences should be regarded
cautiously.
Results for both THg and MeHg in species where both analytes were included
are shown in Table 7.11 (with all sampling areas pooled together). Mean MeHg
concentrations exceeding 0.5 mg/kg were reported for swordfish, bonito, European
conger, northern bluefin tuna and anglerfish, monkfish and stargazers. For
swordfish, the mean MeHg concentration was lower than for THg, while for the
rest of the species or species groups, MeHg values were mostly higher than THg
values. Because many more samples were analysed for THg than for MeHg, the THg
results are likely more representative than the MeHg results of the actual MeHg
concentrations in marine fish.
One species group, river eels, included in Table 7.10, appear to be freshwater fish,
but in the EFSA data these samples were reported as coming from the Mediterranean
Sea and they are therefore included here.

255
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.10 MEAN TOTAL MERCURY (THg) LEVELS IN MUSCLE TISSUE OF SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE (OR UNSPECIFIED) REGIONS
(FAO AREAS), IN mg/kg WET WEIGHT

SPECIES/ SPECIES/
N THg N THg
SPECIES GROUP SPECIES GROUP
27 ATLANTIC, NORTHEAST Plaice 269 0.053
Swordfish 120 1.02 Herring, Atlantic 105 0.048
Sharks 38 0.835 Rainbow trout 36 0.040
Blue ling 50 0.448 European sardine 27 0.039
Bluefish 44 0.448 Atlantic salmon 83 0.034
Tusk 1461 0.339 Trouts 90 0.033
Albacore 23 0.315 Salmons 319 0.032
Tuna 199 0.303 Salmons, trouts, smelts 153 0.026
Anglerfish, monkfish and 345 0.256 Sprat 61 0.020
stargazers
31 ATLANTIC, WESTERN CENTRAL
Sea bass 48 0.234
Tuna, yellowfin 13 0.404
Halibut, Atlantic 797 0.206
34 ATLANTIC, EASTERN CENTRAL
Common ling 827 0.185
Swordfish 24 1.06
Hakes 330 0.138
Tuna, yellowfin 32 0.495
Rays 29 0.131
Tuna 67 0.404
Whiting 40 0.130
Groupers 36 0.082
Eel, European 23 0.126
37 MEDITERRANEAN AND BLACK SEA
Halibut 11 0.121
Tuna 59 0.807
Pollack 330 0.113
Anglerfish, monkfish 24 0.616
Ocean perch 57 0.109 and stargazers
Coalfish 43 0.105 Conger, European 21 0.563
Flounders 33 0.104 Sole 19 0.223
River eels 28 0.103 Hakes 35 0.210
Grenadiers 22 0.101 Scorpion fishes 23 0.193
Dab or common dab 25 0.098 Mullets 20 0.114
Mullets 11 0.098 Anchovies 25 0.092
Beaked redfish 524 0.093 Sea bream 14 0.071
Pangas catfishes 85 0.090 Gilthead seabream 28 0.067
Cod, Atlantic 779 0.090 Sea bass 109 0.065
Golden redfish 223 0.089 Sardines and 29 0.060
sardine-type fishes
Perch 19 0.085
Trouts 33 0.015
Sole 54 0.085
Salmons, trouts, smelts 35 0.013
Mackerel, Atlantic 43 0.076
41 ATLANTIC, SOUTHWEST
Cods, hakes, haddocks 488 0.074
Swordfish 33 0.676
haddock 235 0.071
Tuna 10 0.494
Mackerel 402 0.066
Hakes 77 0.052
Sardines and 90 0.062
sardine-type fishes 47 ATLANTIC, SOUTHEAST
Herrings 249 0.060 Swordfish 57 1.01
Plaice, European 458 0.057 Hakes 164 0.180
Anchovies 66 0.056 Cods, hakes, haddocks 15 0.056

256
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.10 MEAN TOTAL MERCURY (THg) LEVELS IN MUSCLE TISSUE OF SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE (OR UNSPECIFIED) REGIONS
(FAO AREAS), IN mg/kg WET WEIGHT (cont.)

SPECIES/ SPECIES/
N THg N THg
SPECIES GROUP SPECIES GROUP
51 INDIAN OCEAN, WESTERN Hakes 13 0.194
Swordfish 57 1.34 Sea bass 22 0.135
Tuna 82 0.363 Dolphinfishes 13 0.107
Tuna, yellowfin 42 0.261 Mackerel 23 0.070
57 INDIAN OCEAN, EASTERN Cods, hakes, haddocks 23 0.065
Swordfish 46 0.642 Trouts 24 0.059
Tuna 37 0.427 Sardines and 12 0.042
sardine-type fishes
Tuna, yellowfin 17 0.232
GREENLAND, UNSPECIFIED
61 PACIFIC, NORTHWEST
Halibut, Greenland 16 0.045
Swordfish 11 1.11
INDIA, UNSPECIFIED
Hakes 16 0.097
Swordfish 11 0.591
Grenadiers 12 0.088
Tuna 11 0.210
Anglerfish, monkfish and 15 0.079
stargazers INDIAN OCEAN, UNSPECIFIED
Cods, hakes, haddocks 82 0.037 Swordfish 47 1.13

Alaska pollock 17 0.027 Tuna 102 0.361

Salmons, trouts, smelts 31 0.026 MOROCCO, UNSPECIFIED

Salmons 66 0.025 Swordfish 14 0.574

67 PACIFIC, NORTHEAST Anchovies 11 0.044

Hakes 37 0.344 Sardines and 24 0.032

sardine-type fishes
Salmons 26 0.031
PACIFIC OCEAN, UNSPECIFIED
Cods, hakes, haddocks 25 0.020
Swordfish 46 0.757
71 PACIFIC, WESTERN CENTRAL Tuna 21 0.347
Swordfish 10 0.925
Cods, hakes, haddocks 17 0.026
Butterfish 11 0.743
Salmons 11 0.021
Tuna, yellowfin 22 0.211
RUSSIA, UNSPECIFIED
Tuna 204 0.140
Grenadiers 12 0.044
77 PACIFIC, EASTERN CENTRAL SPAIN, UNSPECIFIED
Tuna, yellowfin 29 0.635
Sharks 40 1.74
87 PACIFIC, SOUTHEAST Swordfish 238 1.08
Swordfish 108 1.55
Blue shark 23 1.05
Tuna, yellowfin 14 0.663
Bonito 37 0.480
Tuna 48 0.314
Tuna 256 0.423
Hakes 20 0.096
Tuna, yellowfin 20 0.410
Trouts 10 0.064
Albacore 18 0.273
Salmons, trouts, smelts 13 0.017
Blue whitings 15 0.138
Salmons 60 0.013
Perch 10 0.133
Atlantic salmon 11 0.010
Mackerel 70 0.117
FRANCE, UNSPECIFIED Hakes 104 0.110
Swordfish 14 1.06
Sea bass 21 0.105
Tuna 12 0.362

257
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.10 MEAN TOTAL MERCURY (THg) LEVELS IN MUSCLE TISSUE OF SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE (OR UNSPECIFIED) REGIONS
(FAO AREAS), IN mg/kg WET WEIGHT (cont.)

SPECIES/ SPECIES/
N THg N THg
SPECIES GROUP SPECIES GROUP
Horse mackerels 20 0.092 VIET NAM, UNSPECIFIED
Anglerfish, monkfish and 50 0.091 Swordfish 18 1.10
stargazers
Mackerel 27 0.255
Anchovies 120 0.073
Tuna 81 0.172
Cods, hakes, haddocks 51 0.071
Wolffishes 11 0.004
Sole 13 0.068
UNKNOWN AREA
Sardines and 35 0.060
sardine-type fishes Tuna 15 0.277
Salmons, trouts, smelts 40 0.056 Hakes 25 0.055
European sardine 13 0.055 Mackerel 10 0.053
THE UNITED STATES, UNSPECIFIED Cods, hakes, haddocks 31 0.040
Cods, hakes, haddocks 15 0.030 Herrings 15 0.037
Salmons, trouts, smelts 10 0.028 Char 39 0.033
Salmons 13 0.025 Trouts 31 0.024
Salmons, trouts, smelts 79 0.022

Note: N is the number of analytical samples and may include both individual and composite samples. Cases where N <
10 are excluded.

TABLE 7.11 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (K) LEVELS IN MUSCLE TISSUE OF
SPECIES OR SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM MARINE
OR UNSPECIFIED AREAS AROUND THE WORLD (ALL REGIONS), IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP N THg N MeHg

Swordfish 899 1.06 195 0.796
Bonito 43 0.424 15 0.691
Conger, European 21 0.563 11 0.682
Northern bluefin tuna 17 0.696 10 0.567
Anglerfish, monkfish and stargazers 448 0.250 20 0.548
Tuna, yellowfin 228 0.390 17 0.443
Tuna 1 247 0.340 137 0.363
Albacore 75 0.324 24 0.332
Sole 102 0.107 11 0.243
Hakes 853 0.144 44 0.159
Mackerel 579 0.091 20 0.119
Gilthead seabream 53 0.069 15 0.071
Plaice 306 0.054 15 0.064
Cods, hakes, haddocks 800 0.063 84 0.045

Notes: Results are shown only for species or species groups where both THg and MeHg were analysed. Cases where N <
10 are excluded.

258
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.4.3 MERCURY IN FINFISH, SUMMARY

Data on THg and MeHg from both the literature search and from the EFSA database
showed that, in general, farmed finfish had lower Hg concentrations than wild
finfish and farmed freshwater finfish had lower Hg levels than farmed marine finfish.
Fish captured from wild stocks in marine waters had the highest concentrations,
although the differences between marine and freshwater fish were not very large.
The EFSA data exhibited a larger difference between average THg concentrations
in marine and freshwater fish than what was found in the literature, with overall
mean values for marine and freshwater fish in the EFSA data of 0.27 and 0.19 mg/kg,
respectively, and mean values in the literature of 0.26 and 0.23 mg/kg, respectively.
While the mean values for marine fish were remarkably similar in the two datasets,
the results for fish from inland waters in the two datasets were quite different.
Freshwater fish from the EFSA database included very few species, mostly from
European inland waters, and most of them had low Hg levels. On the other hand,
data from the literature included data on fish from inland waters of most continents,
including a wide range of fish species with relatively high Hg levels, particularly
from South America.
Within the category marine finfish from wild capture, certain species or species
groups had particularly high Hg levels according to both the literature data and the
data from EFSA. These included swordfish (Xiphias gladius) and many different
sharks (Table 7.5 and Table 7.10). Also, large tuna species in the genus Thunnus
from literature and tuna or yellowfin tuna (Thunnus albacares) from the EFSA
database, were among the groups with relatively high Hg levels throughout different
regions of the world’s oceans. However, large variations were observed within the
groups. In addition, there were some genera or species groups more specific to the
different areas that had relatively high mean Hg concentrations. It appears from a
superficial look at the data and without statistical analyses, that the differences in
Hg levels among species or species groups were more important than differences
between geographical areas.

7.4.4 MERCURY IN SHELLFISH FROM THE LITERATURE REVIEW

In the literature, THg results were found for 11 809 analysed samples of shellfish,
while MeHg results were found for only 763 analysed samples. Of the analysed
shellfish, 1 752 and 7 samples for THg and MeHg, respectively, were from
aquaculture, while the rest were from wild stocks (10 057 and 756, respectively) or
unknown (including, for example, sampled at market with no information about
origin). The majority of the analysed shellfish samples originated from marine waters
(n = 9 850 and 689, respectively), while 1 959 and 74 shellfish samples analysed for
THg and MeHg, respectively, were from inland waters.

259
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

7.4.4.1 Farmed shellfish from inland and marine waters

In the literature, results for Hg in farmed shellfish from inland waters were limited to
three species, all from Asia (Table 7.12). These all had very low THg concentrations, with
Chinese mitten crab (Eirocheir sinensis), having the highest mean level, at 0.05 mg/kg.
Methylmercury was only reported from seven samples of whiteleg shrimp
(Litopenaeus vannamei), and mean values of both THg and MeHg in this species
were very low.
Farmed shellfish from marine regions included nine different species, distributed
between three different FAO areas (Table 7.13). These also had very low THg
levels, with the great Mediterranean scallop (Pecten jacobaeus) having the highest
mean concentration, at 0.12 mg/kg (n = 10). No MeHg results were reported in the
literature for farmed shellfish from marine areas.

TABLE 7.12 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN SPECIES OF FARMED
SHELLFISH FROM INLAND WATERS, IN mg/kg WET WEIGHT

FAO AREA/SPECIES LATIN NAME N THg N MeHg

04 ASIA INLAND WATERS
Eriocheir sinensis 36 0.050
Meretrix lusoria 1 021 0.003
Litopenaeus vannamei 7 0.002 7 0.001

Notes: Only FAO area 04 Asia inland included shellfish from inland aquaculture. N is the number of analytical samples,
including both individual and composite samples.

TABLE 7.13 MEAN TOTAL MERCURY (THg) LEVELS IN SPECIES OF FARMED SHELLFISH FROM DIFFERENT
MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

FAO-AREA/SPECIES LATIN NAME N THg

37 MEDITERRANEAN AND BLACK SEA
Pecten jacobaeus 10 0.120
Ostrea edulis 96 0.054
Mytilus galloprovincialis 261 0.052
Venus verrucosa 12 0.020
47 ATLANTIC, SOUTHEAST
Mytilus galloprovincialis 30 0.004
Choromytilus meridionalis 30 0.003
61 PACIFIC, NORTHWEST
Ostrea plicatula 60 0.016
Tegillarca granosa 60 0.008
Sinonovacula constricta 60 0.008
Ruditapes philippinarum 69 0.007

Notes: No farmed shellfish from marine waters were analysed for methylmercury. N is the number of analytical samples,
including both individual and composite samples.

260
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.4.4.2 Wild-caught shellfish from inland waters

The results for THg and MeHg in shellfish species sampled from wild stocks in
different FAO areas are shown in Table 7.14. The table also includes samples whose
origin (aquaculture or wild stocks) is not known. The THg concentrations were low
in most species, except two bivalve species from Africa, Aspatharia senegalensis and
Etheria elliptica, which had mean Hg levels above 0.5 mg/kg, but where the numbers
of analysed samples were very low (n = 5 and n = 2, respectively). Apart from these
two, red swamp crayfish (Procambarus clarkii), from Europe was the only shellfish
species with a mean THg concentration higher than 0.1 mg/kg.

7.4.4.3 Wild-caught shellfish from marine waters

Most shellfish species from marine waters had low Hg concentrations, with a
few exceptions (Table 7.15). The mud crab genus Scylla had a high mean THg
concentration due to one sample from a closed saltwater pond in China, at the site
of a former chloralkali factory. When this sample was excluded, the mean THg
concentration in Scylla was very low. Apart from this, Norway lobster (Nephrops
norvegicus) and blue and red shrimp (Aristeus antennatus) had mean concentrations
close to 0.5 mg/kg. For Norway lobster the samples from Mediterranean and Black
Sea had a considerably higher mean concentration (0.537 mg/kg) than those from the
Northeast Atlantic (0.2 mg/kg) (Table 7.16). The shrimp genus Penaeus had higher
THg concentrations in Mediterranean and Black Sea than in Eastern Central Atlantic
(FAO area 34) and Western Indian Ocean (FAO area 51). The genus Homarus had a
higher mean concentration in Northeast Atlantic (FAO area 27) than in Northwest
Atlantic (FAO area 21) and in Northwest Pacific (FAO area 61). The samples from
Northeast Atlantic were European lobsters (Homarus gammarus), while the others
were American lobster (H. americanus).
The mean MeHg concentration in Norway lobster was higher than the THg
concentration, but only three samples were analysed for MeHg (Table 7.15). Genera
with mean THg above 0.2 mg/kg included lobsters, shrimps, cephalopods and
gastropods, as well as the Atlantic bay scallop (Argopecten irradians).

261
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.14 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN TISSUES OF
DIFFERENT SPECIES OF WILD-CAUGHT (OR UNKNOWN) SHELLFISH FROM DIFFERENT INLAND
REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

SPECIES N THg N MeHg

01 AFRICA, INLAND WATERS
Aspatharia senegalensis 5 0.790
Etheria elliptica 2 0.600
Macrobrachium rosenbergii 50 0.010
02 NORTH AMERICA, INLAND WATERS
Cambarus carinirostris 20 0.050
Corbicula fluminea 5 0.018
Dreissena bugensis 2 0.005
03 SOUTH AMERICA INLAND WATERS
Macrobrachium amazonicum 349 0.018
05 EUROPE INLAND WATERS
Procambarus clarkii 105 0.113
Mytilus galloprovincialis 4 0.026 4 0.010
04 ASIA INLAND WATERS
Eriocheir sinensis 3 0.081
Portunus pelagicus 5 0.075
Scylla olivacea 12 0.072
Crassostrea sp. 23 0.071
Meretrix lusoria 27 0.065
Anodonta woodiana 1
8 0.052
Thalamita crenata 6 0.048
Polymesoda expansa 21 0.021
Sinanodonta woodiana 9 0.020
Chlamys farreri 1 5 0.017
Pilsbryoconcha compressa 3 0.014
Perna viridis 1
5 0.013
Argopecten irradians 1 5 0.012
Paphia undulata 1
5 0.010
Ruditapes philippinarum 1 5 0.010
Crassostrea ariakensis 1
5 0.008
Babylonia areolata 1
5 0.007
Sinonovacula constricta 1 5 0.007
Corbicula leana 1
52 0.005
Haliotis discus 1 5 0.004
Cipangopaludina japonica 1
71 0.002
Penaeus monodon 1 0.0002 1 0.0002
Prawn, unknown 1 0.0021
Bellamya javanica 61 0.260

Notes: N is the number of analytical samples, including both individual and composite samples. All analysed edible
tissues are included.
1
Unknown whether farmed or wild.

262
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.15 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN TISSUE OF
DIFFERENT GENERA OF WILD-CAUGHT SHELLFISH FROM MARINE WATERS (ALL REGIONS),
IN mg/kg WET WEIGHT

GENUS N THg N MeHg

Scylla 71 (701) 0.93 (0.0481) 2 0.001
Nephrops 350 0.495 3 0.664
Aristeus 16 0.465
Eledone 12 0.257
Busycon 38 0.250
Parapenaeus 753 0.249
Hexaplex 20 0.245
Argopecten 34 0.233
Austramacoma 24 0.195
Squilla 78 0.182
Palaemon 38 0.180 32 0.185
Cancer 20 0.170
Dosinia 15 0.169
Maja 22 0.164
Penaeus 465 0.161
Homarus 197 0.160
Chlamys 24 0.146 8 0.139
Eriphia 634 0.144
Haliotis 81 0.128
Sepia 148 0.118
Pandalus 19 0.113 10 0.200
Callinectes 235 0.105
Chionoecetes 227 0.101 39 0.020
Thysanoteuthis 24 0.101
Concha 47 0.095
Panulirus 124 0.083
Nototodarus 30 0.080
Litopenaeus 50 0.079
Achelous 32 0.076
Pinna 15 0.074 1.16
Lithophaga 58 0.073
Megapitaria 89 0.070
Crassostrea 183 0.062 1 0.028
Octopus 302 0.062 14 0.036
Mytilus 605 0.054 15 0.015
Solen 54 0.053 12 0.011
Metapenaeus 197 0.052
"Sea prawns" 2 0.003 2 0.003
Macrobrachium 24 0.048
Anadara 232 0.044 42 0.009
Rapana 41 0.043 14 0.034

263
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.15 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN TISSUE OF
DIFFERENT GENERA OF WILD-CAUGHT SHELLFISH FROM MARINE WATERS (ALL REGIONS),
IN mg/kg WET WEIGHT (cont.)

GENUS N THg N MeHg

Sinonovacula 63 0.041 6 0.002
Todarodes 113 0.041
Donax 24 0.041
Charybdis 160 0.040 9 0.031
Pachygrapsus 13 0.040
Doryteuthis 23 0.039
Berryteuthis 97 0.038
Diplodonta 12 0.037
Loligo 105 0.033 20 0.018
Callista 98 0.032
Paralithodes 43 0.031
Portunus 221 0.030 17 0.014
Senilia 30 0.030
Osilinus 10 0.029
Neptunea 51 0.029 24 0.003
Ruditapes 786 0.029 79 0.012
Fenneropenaeus 256 0.028 30 0.005
Patella 35 0.028
Paracentrotus 136 0.025
Farfantepenaeus 51 0.024
Oratosquilla 32 0.021 24 0.021
Enteroctopus 16 0.021
Paroctopus 99 0.020
Cerithium 27 0.017
Holothuria 42 0.016
Mactra 12 0.016 12 0.012
Meretrix 117 0.015 42 0.007
Cyclina 12 0.014 12 0.010
Perna 103 0.014
Cassidula 37 0.013
Thais 12 0.012 6 0.019
Dosidicus 18 0.010
Acetes 78 0.010
Paphia 25 0.010
Alpheus 43 0.009 36 0.006
Crangon 44 0.009 38 0.005
Ostrea 28 0.006 24 0.002
Trachypenaeus 43 0.006 36 0.006
Chamelea 14 0.005
Nordotis 12 0.004
Monetaria 12 0.004 12 0.002

264
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.15 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN TISSUE OF
DIFFERENT GENERA OF WILD-CAUGHT SHELLFISH FROM MARINE WATERS (ALL REGIONS),
IN mg/kg WET WEIGHT (cont.)

GENUS N THg N MeHg

Harpiosquilla 36 0.00005
"Pacific octopus" 21 0.095

Note: 1 1 sample removed: THg 3.6 mg/kg, n = 1.

TABLE 7.16 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT
SHELLFISH FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT

GENUS N THg GENUS N THg

21 ATLANTIC, NORTHWEST Sepia 42 0.153
Homarus 105 0.162 Chlamys 16 0.146
Pandalus 19 0.113 Eriphia 634 0.144
27 ATLANTIC, NORTHEAST Octopus 79 0.134
Homarus 26 0.273 Crassostrea 73 0.131
Nephrops 10 0.200 Solen 42 0.091
Cancer 20 0.170 Mytilus 239 0.086
Lithophaga 46 0.126 Ruditapes 19 0.040
Ruditapes 484 0.031 Portunus 39 0.080
Mytilus 34 0.027 Anadara 15 0.051
Patella 15 0.012 Donax 24 0.041
31 ATLANTIC, WESTERN CENTRAL Patella 20 0.036
Callinectes 15 0.125 Pachygrapsus 13 0.040
Thysanoteuthis 24 0.101 Pinna 15 0.074
Crassostrea 17 0.009 Rapana 18 0.034
Litopenaeus 31 0.033 Callista 98 0.032
34 ATLANTIC, EASTERN CENTRAL Paracentrotus 130 0.031
Austramacoma Osilinus 10 0.029
24 0.195 Lithophaga 12 0.020
Penaeus 48 0.092 Holothuria 42 0.016
Diplodonta 12 0.037 Cerithium 23 0.015
Senilia 30 0.030 Haliotis 60 0.012
Crassostrea 62 0.025 Chamelea 14 0.005
Perna 36 0.011 41 ATLANTIC, SOUTHWEST
37 MEDITERRANEAN AND BLACK SEA Callinectes 210 0.094
Nephrops 340 0.537 Achelous 32 0.076
Aristeus 16 0.465 Doryteuthis 17 0.056
Parapenaeus 748 0.277 Farfantepenaeus 45 0.027
Eledone 12 0.257 Macrobrachium 20 0.025
Hexaplex 20 0.245 Perna 30 0.024
Penaeus 98 0.223 47 ATLANTIC, SOUTHEAST
Squilla 78 0.182 Loligo 35 0.029
Maja 22 0.164

265
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.16 MEAN TOTAL MERCURY (THg) LEVELS IN TISSUE OF DIFFERENT GENERA OF WILD-CAUGHT
SHELLFISH FROM DIFFERENT MARINE REGIONS (FAO AREAS), IN mg/kg WET WEIGHT (cont.)

GENUS N THg GENUS N THg

51 INDIAN OCEAN, WESTERN Concha 47 0.095
Metapenaeus 104 0.131 Sinonovacula 45 0.059
Fenneropenaeus 60 0.116 Rapana 23 0.047
Penaeus 300 0.085 Charybdis 153 0.047
Panulirus 120 0.076 Homarus 66 0.045
57 INDIAN OCEAN, EASTERN Todarodes 113 0.041
Mytilus 298 0.047 Ruditapes1 18 0.016
Portunus 13 0.014 Acetes1
78 0.01
Cassidula 37 0.013 Meretrix 1
36 0.007
61 PACIFIC, NORTHWEST Metapenaeus1 86 0.007
Busycon 38 0.250 Sinonovacula 1
18 0.006
Haliotis 21 0.243 Paphia1 18 0.005
Argopecten 34 0.233 Fenneropenaeus 1
161 0.003
Chionoecetes 227 0.101

Note: 1 Unknown whether farmed or wild.

7.4.5 MERCURY IN SHELLFISH FROM DATA FROM THE EUROPEAN FOOD SAFETY
AUTHORITY
In the data from EFSA, THg and MeHg values were reported using codes for species
or species groups, rather than the less ambiguous Latin names (which are used in
the literature data). Because of this, some species groups may include a wide range
of species which may have very different Hg levels. Also, some species may be
present both with its specific species name and as part of a larger group of species.
For instance, blue mussel could be included also in the species group “mussels”.
The EFSA data includes samples analysed in Europe, but includes both samples of
fish of European origin and fish imported to Europe from other countries.
Shellfish data reported to EFSA included 5 610 results for THg and 211 results for
MeHg. Of the THg results, as many as 4 124 were reported without information
about whether the samples were from shellfish farming or wild stocks. It was
assumed that they were most likely from wild stocks, but they may also have been
farmed. Of the shellfish for which the origin was given, 495 samples analysed for
THg were reported as farmed and 891 were reported as being from wild stocks.
The majority of the reported samples (THg n = 3 630) were from marine waters,
while only 174 samples were from inland areas. A large portion of the samples
(THg n = 1 806) were reported without indications of the origin, and these were
included together with the results for marine shellfish. Analysed tissue for shellfish
was not specified in the dataset, but it is assumed that edible tissue was used, either
muscle tissue (for instance, for crustaceans) or the whole soft part of the animal (for
instance, for bivalves).

266
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.4.5.1 Farmed shellfish from inland and marine waters

The dataset received from EFSA’s database included only four samples of shellfish
from inland aquaculture (Table 7.17). All were freshwater shrimps or prawns with
very low mean THg concentrations. No results were reported for MeHg.

TABLE 7.17 MEAN TOTAL MERCURY (THg) LEVELS IN SPECIES OR SPECIES GROUPS OF FARMED
SHELLFISH FROM INLAND AND MARINE WATERS, IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP N THg

03 SOUTH AMERICA INLAND WATERS
Freshwater shrimps or prawns 3 0.022
05 EUROPE INLAND WATERS
Freshwater shrimps or prawns 1 0.032
27 ATLANTIC, NORTHEAST
Oysters 80 (71) 0.025
Mussels 103 (98) 0.022
Oyster, European 28 0.013
Blue mussel 297 0.013
Scallop, great 34 0.012
Scallops, pectens 23 0.012

Note: N is the number of analytical samples and may include both individual and composite samples.

7.4.5.2 Wild-caught shellfish from inland waters

Overall mean concentrations of THg and MeHg in shellfish species groups sampled
from wild stocks (or unspecified origin) in inland waters, included 127 samples of
shrimps and prawns, 37 samples of freshwater crayfish, 5 oyster samples and 1 clam
sample (Table 7.18). All these groups had low concentrations of THg. One sample of
the group “shrimps and prawns” was analysed for MeHg. This sample had relatively
high MeHg, with 0.349 mg/kg. Of the different areas, freshwater shrimps or prawns
from Spain (FAO area not specified), had the highest mean concentration of THg,
with 0.115 mg/kg, except a single sample of crayfish from FAO area 06 Oceania
with a THg level of 0.120 mg/kg (Table 7.19).

TABLE 7.18 MEAN TOTAL MERCURY (THg) LEVELS IN SPECIES OR SPECIES GROUPS OF WILD-CAUGHT
(OR UNSPECIFIED) SHELLFISH FROM INLAND WATERS (ALL REGIONS), IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP THg N THg MeHg N MeHg

Clams 1 0.100
Shrimps and prawns1 127 0.072 1 0.349
Freshwater crayfishes 37 0.052
Oysters 5 0.031

Notes: N is the number of analytical samples and may include both individual and composite samples.
1
Includes “Freshwater shrimps and prawns” (n = 118) and “Shrimps and prawns” (n = 9).

267
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.19 MEAN TOTAL MERCURY (THg) LEVELS IN SPECIES OR SPECIES GROUPS OF WILD-CAUGHT
(OR UNSPECIFIED) SHELLFISH FROM DIFFERENT INLAND REGIONS (FAO AREAS),
IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP N THg

01 AFRICA INLAND WATERS
Freshwater shrimps or prawns 5 0.070
02 NORTH AMERICA INLAND WATERS
Freshwater shrimps or prawns 2 0.030
03 SOUTH AMERICA INLAND WATERS
Freshwater shrimps or prawns 7 0.038
04 ASIA INLAND WATERS
Freshwater shrimps or prawns 27 0.048
Freshwater crayfishes 3 0.043
Shrimps and prawns 9 0.009
05 EUROPE INLAND WATERS
Clams 1 0.100
Freshwater crayfishes 33 0.051
Freshwater shrimps or prawns 26 0.070
Oysters 5 0.031
06 OCEANIA INLAND WATERS
Freshwater crayfishes 1 0.120
Freshwater shrimps or prawns 1 0.010
CANADA, UNSPECIFIED
Freshwater shrimps or prawns 2 0.054
GREENLAND, UNSPECIFIED
Freshwater shrimps or prawns 1 0.050
SPAIN, UNSPECIFIED
Freshwater shrimps or prawns 48 0.115

Note: N is the number of analytical samples and may include both individual and composite samples.

268
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.4.5.3 Wild-caught shellfish from marine waters

The data received from EFSA contained THg in 4 529 samples of marine shellfish
originating from wild stocks, belonging to 39 different species or species groups,
where 32 had a sample number of 10 or greater (Table 7.20). MeHg was analysed in
209 samples of 19 species or groups. All groups had mean THg concentrations below
0.2 mg/kg. Tissue samples include muscle (most crustaceans), whole body or soft or
edible parts (most bivalves). Edible crab, common octopus and Norway lobster had
the highest mean THg concentrations, with mean concentrations around 0.14 mg/
kg. When comparing species or species groups in different areas, common octopus
from Spain (FAO area not specified) had the highest mean THg concentration, with
0.26 mg/kg (Table 7.21), followed by shrimps and prawns from the Mediterranean
and Black Sea area, with 0.20 mg/kg. Shrimps and prawns from the Mediterranean
and Black Sea area had higher mean THg than shrimps and prawns from all other
areas where this group was included.
Results for MeHg included much fewer samples than the results for THg, and mean
concentrations of MeHg where higher than THg for Norway lobster, common
shrimp and shrimps and prawns, but lower for most species (Table 7.20).

269
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.20 MEAN TOTAL MERCURY (THg) AND METHYLMERCURY (MeHg) LEVELS IN SPECIES OR SPECIES
GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) SHELLFISH FROM MARINE WATERS
(ALL REGIONS), IN mg/kg WET WEIGHT

SPECIES/SPECIES GROUP N THg N MeHg

Edible crab 100 0.141
Octopus, common 68 0.137 4 0.121
Lobster, Norway 169 0.137 5 0.160
Crabs, sea-spiders 281 0.123
Lobsters 33 0.102
Freshwater shrimps or prawns 36 0.096 1 0.020
Spiny and rock lobsters 12 0.081
Cuttlefish, common 32 0.081 8 0.034
Freshwater crayfishes 36 0.071
Crustaceans 20 0.061
Cuttlefishes 157 0.055 15 0.016
Clams, cockles, arkshells 21 0.050
Squids 238 0.049 15 0.036
Razor clam 22 0.048 13 0.010
Shrimps and prawns 628 0.045 29 0.138
Octopus, curled 106 0.043
Cockles 153 0.039
Water snails, conches and whelks 23 0.039
Clams 314 0.038 22 0.011
Shrimps, common 312 0.036 20 0.083
Squids, cuttlefishes, octopuses 18 0.036
Oysters 500 0.035
Scallops, pectens 257 0.030
Mussels 950 0.030 71 0.012
Octopuses 37 0.026
Scallop, queen 93 0.025
Prawn, northern 84 0.024
Blue mussel 44 0.022
White shrimp 11 0.021 2 0.015
Metapenaeus shrimps 27 0.012
Oyster, European 20 0.011
Scallop, great 20 0.011 4 0.020

Notes: N is the number of analytical samples and may include both individual and composite samples. Cases where N <
10 are excluded.

270
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.21 MEAN TOTAL MERCURY (THg) LEVELS IN SPECIES OR SPECIES GROUPS OF WILD-CAUGHT
(OR UNSPECIFIED) SHELLFISH FROM DIFFERENT MARINE (OR UNSPECIFIED) REGIONS (FAO AREAS),
IN mg/kg WET WEIGHT

SPECIES/ SPECIES/
N THg N THg
SPECIES GROUP SPECIES GROUP
21 ATLANTIC, NORTHWEST Scallops, pectens 12 0.023
Scallops, pectens 17 0.038 51 INDIAN OCEAN, WESTERN
Oysters 23 0.017 Squids 13 0.056
27 ATLANTIC, NORTHEAST 57 INDIAN OCEAN, EASTERN
Edible crab 79 0.143 Shrimps and prawns 53 0.013
Crabs, sea-spiders 187 0.136 Shrimps, common 29 0.012
Lobsters 13 0.133 61 PACIFIC, NORTHWEST
Lobster, Norway 156 0.127 Crabs, sea-spiders 19 0.064
Freshwater crayfishes 35 0.073 Squids 40 0.035
Crustaceans 20 0.061 Shrimps and prawns 16 0.023
Octopus, curled 17 0.058 71 PACIFIC WESTERN CENTRAL
Shrimps, common 91 0.053 Squids 10 0.029
Squids 38 0.051 Shrimps and prawns 34 0.014
Cuttlefishes 16 0.051 Shrimps, common 24 0.010
Water snails, conches and 16 0.042 81 PACIFIC, SOUTHWEST
whelks
Mussels 25 0.033
Octopus, common 18 0.041
87 PACIFIC, SOUTHEAST
Oysters 181 0.039
Squids 16 0.044
Scallops, pectens 156 0.038
Mussels 78 0.025
Cockles 99 0.037
Scallop, queen 18 0.023
Shrimps and prawns 81 0.036
Scallops, pectens 15 0.023
Clams 118 0.033
Shrimps and prawns 39 0.022
Mussels 515 0.028
CANADA, UNSPECIFIED
Scallop, queen 52 0.028
Lobsters 12 0.074
Blue mussel 38 0.023
France, unspecified
Oyster, European 20 0.011
Edible crab 10 0.150
Scallop, great 20 0.011
Crabs, sea-spiders 10 0.122
31 ATLANTIC, WESTERN CENTRAL
Clams 11 0.046
Shrimps and prawns 23 0.046
Cockles 35 0.038
34 ATLANTIC, EASTERN CENTRAL
Oysters 255 0.032
Cuttlefishes 17 0.045
Mussels 97 0.029
37 MEDITERRANEAN AND BLACK SEA
Scallop, queen 11 0.021
Shrimps and prawns 21 0.201
Shrimps and prawns 15 0.016
Freshwater shrimps or 18 0.152
prawns Shrimps, common 12 0.008

Octopus, curled 12 0.095 Greenland, unspecified

Mussels 19 0.038 Crabs, sea-spiders 15 0.046

Oysters 26 0.038 Prawn, northern 58 0.027

41 ATLANTIC, SOUTHWEST Scallops, pectens 18 0.016

Squids 12 0.063
Shrimps and prawns 14 0.030

271
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.21 MEAN TOTAL MERCURY (THg) LEVELS IN SPECIES OR SPECIES GROUPS OF WILD-CAUGHT
(OR UNSPECIFIED) SHELLFISH FROM DIFFERENT MARINE (OR UNSPECIFIED) REGIONS (FAO AREAS),
IN mg/kg WET WEIGHT (cont.)

SPECIES/ SPECIES/
N THg N THg
SPECIES GROUP SPECIES GROUP
INDIA, UNSPECIFIED Squids 45 0.066
Squids 25 0.034 Clams, cockles, arkshells 15 0.058
Shrimps and prawns 43 0.020 Clams 118 0.051
Cuttlefishes 29 0.019 Razor clam 11 0.047
Octopuses 13 0.014 Mussels 186 0.038
Shrimps, common 16 0.011 Octopus, curled 44 0.034
INDIAN OCEAN, UNSPECIFIED Cockles 11 0.027
Shrimps and prawns 23 0.016 VIET NAM, UNSPECIFIED
SPAIN, UNSPECIFIED Clams 34 0.015
Octopus, common 27 0.258 Prawn, northern 10 0.014
Shrimps and prawns 78 0.160 Metapenaeus shrimps 12 0.012
Crabs, sea-spiders 16 0.142 Shrimps and prawns 134 0.011
Shrimps, common 30 0.080 Shrimps, common 66 0.010
Cuttlefishes 65 0.074 Mussels 10 0.010
Cuttlefish, common 12 0.067

Notes: N is the number of analytical samples and may include both individual and composite samples. Cases where N <
10 are excluded.

7.4.6 MERCURY IN SHELLFISH, SUMMARY

The data from both the literature and the EFSA database on Hg in farmed
shellfish and in shellfish from inland waters was very limited and included few
species or species groups. Farmed shellfish from inland waters all had very low
Hg concentrations. In general, wild-caught shellfish from inland waters also had
low concentrations. Two species with mean THg levels exceeding 0.5 mg/kg were
represented with results from only two and five samples, which does not provide
sufficiently conclusive evidence. Apart from these, only one species from the
literature, red swamp crayfish (Procambrus clarkii) from Europe, and one group
in the EFSA data, freshwater shrimps and prawns from Spain, had mean THg
concentrations higher than 0.1 mg/kg.
According to data from both the literature and EFSA, farmed marine shellfish
(all bivalves) also had very low Hg levels. Marine shellfish captured from wild stocks
was the most diverse category of shellfish with the largest volume of Hg data and
the largest number of species or species groups, both from the literature and from
EFSA. Consequently, the range of Hg concentrations was also wider. However,
Hg levels were generally low also in wild-caught marine shellfish, with only the
Norway lobster from the Mediterranean and Black Sea area having a mean THg
concentration slightly above 0.5 mg/kg.

272
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

Both the EFSA data and the data from the literature showed that, in general, in the
different areas, the highest Hg levels were found in decapod crustaceans such as
lobsters, crabs and shrimps and prawns, as well as in octopuses. In the literature data,
some gastropod genera (abalone and whelk) from the Northwest Pacific and one
gastropod from the Mediterranean (Hexaplex, murex) also had among the highest
mean THg concentrations in those areas. Bivalves in general had very low THg
concentrations.
It is difficult to compare geographical areas, since different species are represented in
the different areas and very different amounts of data have been reported from the
different areas. In the literature data, the Mediterranean and Black Sea area (FAO
area 37) had the highest number of genera with Hg concentrations higher than
0.2 mg/kg, followed by Northwest Pacific (FAO area 61) and Northeast Atlantic
(FAO area 27). In the EFSA data, this geographical trend was not as clear, having
only one species group with THg > 0.2 mg/kg; but here the Hg concentrations
in general were lower. At least part of the geographical differences may be due to
different species analysed in the different areas. However, for some genera or species
groups, including Nephrops, Penaeus and the group shrimps and prawns, mean THg
concentrations in samples from the Mediterranean and Black Sea area were higher
than in those from other parts of the world.

7.5 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS

IN FINFISH AND SHELLFISH FROM DIFFERENT REGIONS AROUND
THE WORLD

7.5.1 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS

IN FINFISH FROM THE LITERATURE REVIEW
In the literature, results for dioxins and dl-PCBs were found for 2 408 analysed
samples of finfish. Of these, 805 samples were from farmed fish, while the rest were
from wild stocks (1 139 samples) or unknown (464 samples were sampled at market
with no information about their origin). Most of the analysed finfish originated from
marine waters (1 915 samples), while 461 finfish samples were from inland waters
or unknown (32 finfish samples were sampled at market with no information about
their origin).

7.5.1.1 Farmed finfish from inland and marine waters

Farmed finfish from inland and marine waters had generally low concentrations of
dioxins and dl-PCBs (Table 7.22), and only one species, milk fish (Chanos chanos),
from Asia inland waters, had a high mean value of sum dioxins of 4.36 ng TEQ/kg.
The rest of the species in this category had mean concentrations of both sum dioxins
and sum dioxins + dl-PCBs of 0.513 ng TEQ/kg or lower.

273
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.22 MEAN CONCENTRATIONS OF SUM DIOXINS, SUM DIOXIN-LIKE POLYCHLORINATED BIPHENYLS
(dl-PCBs) AND SUM DIOXINS AND dl-PCBs IN FARMED FINFISH FROM DIFFERENT INLAND
AND MARINE REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

DIOXINS+
FAO AREA/SPECIES LATIN NAME N DIOXINS Dl-PCBs dl-PCBs
04 ASIA INLAND WATERS
Chanoschanos 4 4.36
Oreochromis mossambicus 46 0.354
Elops machnata 3 0.237
27 ATLANTIC, NORTHEAST
Salmo salar 738 0.273 0.247 0.513
37 MEDITERRANEAN AND BLACK SEA
Salmo salar 2 0.117 0.390 0.507

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N =1 are
excluded.

7.5.1.2 Wild-caught finfish from inland waters

For wild-caught finfish from inland waters, the literature review showed that the
concentrations of dioxins and dl-PCBs were relatively low for most of the species
investigated (Table 7.23). The highest mean values of dioxins+dl-PCBs were found
in muscle of grass carp (Ctenopharyngodon idella) and goldfish (Carassius aurata)
from Asia. The concentrations in these species, 38–49 ng TEQ/kg wet weight,
were far above 6.5 ng TEQ/kg (maximum level in the European Union), but the
results were based on very few samples and should be interpreted with caution.
High levels of dioxins+dl-PCBs were also found in whole fish of lake trout (Salvelinus
namaycush) and whole fish of walleye (Sander vitreus), from North America
(12.6–29.5 ng TEQ/kg wet weight). In general, whole fish have higher levels than
muscle tissue from the same species, and muscle of lake trout had a much lower mean
concentration of dioxins than whole fish of this species. The difference between
whole fish and muscle tissue could also be seen for chub (Squalius cephalus), from
Europe, where whole fish had concentrations of dioxins+dl-PCBs more than two
times higher than muscle tissue (Table 7.23). Apart from the species mentioned
above, no other species had mean concentrations of dioxins and dl-PCBs above 6.5
ng TEQ/kg, and only one other species, wels catfish (Silurus glanis), from Europe,
had mean concentrations of dioxins and dl-PCBs in muscle tissue above 3.0 ng
TEQ/kg (Table 7.23).

7.5.1.3 Wild-caught finfish from marine waters

Finfish sampled from wild stocks (or unknown) in marine waters made up the
largest volume of results published in the literature for dioxins and dl-PCBs.
Table 7.24 provides results for muscle tissue (per FAO area and genus). The highest
mean values for dioxins in muscle tissue were found in five genera from FAO area
61 Pacific, Northwest – Elops, Liza, Chanos, Nematolosa and Oreochromis, with
concentrations between 5.35 and 330 ng TEQ/kg. All these results came from the

274
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.23 MEAN CONCENTRATIONS OF SUM DIOXINS, SUM DIOXIN-LIKE POLYCHLORINATED BIPHENYLS
(dl-PCBs) AND SUM DIOXINS AND dl-PCBs IN SPECIES OF WILD FINFISH FROM DIFFERENT
INLAND REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT/kg WET WEIGHT)

DIOXINS+
FAO AREA/SPECIES LATIN NAME N DIOXINS Dl-PCBs dl-PCBs
01 AFRICA INLAND WATERS
Lates niloticus 31 0.030 0.063
Oreochromis niloticus 31 0.027 0.030
02 NORTH AMERICA INLAND WATERS
Salvelinus namaycush (muscle) 12 2.02
Salvelinus namaycush (whole) 71 4.93 23.6 29.5
Sander vitreus (whole) 4 2.77 9.86 12.6
04 ASIA INLAND WATERS
Carassius auratus 5 49
Ctenopharyngodon idella 5 38
Cyprinus carpio 16 0.31 0.37 0.68
Silurus asotus 16 0.090 0.13 0.22
Carassius carassius 13 0.020 0.060 0.09
05 EUROPE INLAND WATERS
Squalius cephalus (whole) 4 0.698 5.02 5.74
Silurus glanis 31 0.17 2.78 3.1
Abramis brama 26 0.89 1.19 2.4
Squalius cephalus (muscle) 10 0.37 2.12 2.3
Abramis bjoerkena 3 0.85 1.22 2.1
Oncorhynchus mykiss (whole) 4 0.230 1.53 1.74
Coregonus renke 2 0.26 0.69 0.940
Rutilus rutilus 32 0.20 0.56 0.763
Perca fluviatilis 45 0.13 0.36 0.495
Barbus barbus (whole) 2 0.125 0.28 0.41
Esox lucius 17 0.14 0.20 0.340
Stizostedion lucioperca 9 0.13 0.18 0.305
Anguilla anguilla 16 0.013

Notes: N is the number of analytical samples, including both individual and composite samples. The number of samples
analysed for dioxins, dl-PCBs and dioxins+dl-PCBs may be different for some species. Cases where N = 1 are excluded.
Where whole fish was analysed, this is indicated; otherwise, muscle tissue was analysed.

same study (Liao et al., 2016) with samples collected from a closed saltwater pond
in China at the site of a former chloralkali factory, which likely explains the high
values of dioxins. Apart from these genera, the remaining species in this category
had mean values of dioxins below 2.0 ng TEQ/kg and dioxins+dl-PCBs below 3.8
ng TEQ/kg (Table 7.24).
Results for whole body tissue of wild-caught finfish from marine waters published
in the literature are given in Table 7.25. Results for whole fish were available for
four different species from FAO area 27 (Atlantic, Northeast), with mean values of
dioxins and dl-PCBs between 1.6 and 2.0 ng TEQ/kg.

275
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.24 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF WILD FINFISH FROM
DIFFERENT MARINE REGIONS (FAO AREAS), BY GENUS (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT)

DIOXINS+
GENUS OR OTHER DESCRIPTION N DIOXINS Dl-PCBs dl-PCBs
27 ATLANTIC, NORTHEAST
Sprattus 30 1.79 1.95 3.74
Salmo 141 1.13 2.00 3.13
«Sea bass» 25 0.440 2.06 2.50
Clupea 45 0.990 0.920 1.91
Platichthys 23 0.870 0.940 1.82
Scomber 41 0.430 0.970 1.40
«Grey mullet» 26 0.140 0.530 0.670
Scophthalmus 16 0.170 0.500 0.670
«Sharks» 14 0.120 0.210 0.320
Gadus 40 0.105 0.151 0.257
Sebastes 2 0.157 0.388
Zoarces 15 0.675
37 MEDITERRANEAN AND BLACK SEA
Sarda 3 0.458 3.23 3.69
Thunnus 26 1.90 0.700 2.60
«Sardine» 3 0.270 2.13 2.39
Scomber 5 0.366 1.81 2.17
Mullus 2 0.231 1.90 2.13
Boops 3 0.352 1.05 1.40
Xiphias 52 0.179 1.09 1.27
«Anchovy» 2 0.091 1.08 1.17
«Tuna» 2 0.042 0.58 0.622
Trachurus 6 0.332 0.27 0.602
Gadus 39 0.065 0.495 0.561
Merluccius 5 0.217 0.134 0.351
Solea 2 0.057 0.170 0.227
27 ATLANTIC NORTHEAST OR 37 MEDITERRANEAN AND BLACK SEA
Thunnus 2 0.080 1.24 1.32
Solea 2 0.130 0.260 0.390
Sebastes 2 0.140 0.190 0.330
Mullus 2 0.100 0.170 0.270
Merluccius 2 0.040 0.140 0.180
41 ATLANTIC, SOUTHWEST
Micropogonias 14 0.110
48 ATLANTIC, ANTARCTIC
Trematomus 10 0.360
Chionodraco 11 0.089
61 PACIFIC, NORTHWEST
Pagrus 13 0.600 0.880 1.47

276
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.24 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF WILD FINFISH FROM
DIFFERENT MARINE REGIONS (FAO AREAS), BY GENUS (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT) (cont.)

DIOXINS+
GENUS OR OTHER DESCRIPTION N DIOXINS Dl-PCBs dl-PCBs
Okamejei 12 0.470 0.730 1.20
Scomber 15 0.410 0.700 1.12
Hippoglossus 5 0.140 0.830 0.970
Selachimorpha 15 0.210 0.720 0.930
Arctoscopus 11 0.260 0.610 0.870
Larimichthys 29 0.256 0.888 0.784
Theragra 33 0.135 0.420 0.555
Astroconger 15 0.122 0.100 0.444
Pleurogrammus 15 0.060 0.330 0.390
Cololabis 16 0.114 0.271 0.385
Pleuronectes 17 0.140 0.230 0.370
Mugil 11 0.148 0.200 0.369
Ostracion 14 0.060 0.300 0.360
Trichiurus 17 0.136 0.314 0.354
Clupea 17 0.210 0.095 0.305
Stephanolepis 16 0.060 0.240 0.300
Lophiomus 12 0.110 0.180 0.290
Scomberomorus 7 0.107 0.168 0.285
Paralichthys 2 0.152 0.200 0.276
Sebastes 17 0.040 0.210 0.250
Pampus 11 0.073 0.040 0.220
Miichthys 11 0.055 0.210 0.206
Gadus 15 0.063 0.092 0.155
Engraulis 9 0.046 0.010 0.150
Lateolabrax 14 0.074 0.192 0.128
Konosirus 10 0.030 0.080 0.110
Pleuronichthys 17 0.030 0.050 0.080
Cynoglossus 8 0.040 0.030 0.060
Thunnus 12 0.040 0.020 0.050
Misgurnus 16 0.030 0.020 0.040
Elops 7 330
Liza 2 58.8
Chanos 3 51
Nematalosa 4 27
Oreochromis 2 5.35
«Hairtail» 5 0.946
«Flatfish» 5 1.28
«Spanish mackerel» 10 1.65

Notes: N is the number of analytical samples, including both individual and composite samples. The number of samples
analysed for dioxins, dl-PCBs and dioxins+dl-PCBs may be different for some species. Cases where N = 1 are excluded.

277
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.25 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs) AND
THE SUM OF DIOXINS AND dl-PCBs IN WHOLE BODY TISSUE OF SPECIES OF WILD FINFISH FROM
DIFFERENT MARINE REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

FAO AREA/ DIOXINS+

N DIOXINS Dl-PCBs
SPECIES LATIN NAME dl-PCBs
27 ATLANTIC, NORTHEAST
Maurolicus muelleri 4 1.10 0.97 2.00
Sprattus sprattus 25 0.910 1.09 2.00
«Sardine» 16 0.400 1.57 1.97
Benthosema glaciale 5 0.770 0.84 1.60

Note: N is the number of analytical samples, including both individual and composite samples.

7.5.2 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS FROM DATA

FROM THE EUROPEAN FOOD SAFETY AUTHORITY
In the data from EFSA, values for dioxins and dl-PCBs were reported using codes for
species or species groups, rather than the less ambiguous Latin names (which are used
in the literature data). Because of this, some species groups may include a wide range
of species which may have very different levels of dioxins and dl-PCBs. Also, some
species may be present both with their specific species names and in terms of a larger
group of species. For instance, the species named Atlantic salmon could be included also
in the species groups “salmons” or “salmons, trouts, smelts”. The EFSA data includes
samples analysed in Europe, but includes both samples of fish of European origin and
fish imported to Europe from other countries. Analysed tissue for finfish was muscle
tissue (“fish meat”), and no data for dioxins and dl-PCBs in whole fish were available
from the EFSA data.
The final dataset on dioxins and dl-PCBs compiled from the EFSA data contained 7
668 analysed samples of finfish. Of these, 7 438 samples had results for both dioxins
and dl-PCBs (all 29 congeners), 42 samples had results only for dioxins (17 congeners)
and 188 samples had results only for dl-PCBs (12 congeners). Of the analysed finfish, 1
142 samples were from farmed shellfish, 3 925 samples were from wild stocks, and the
remaining 2 601 samples were reported without indications as to whether the samples
were from farmed or wild-caught finfish. These unspecified samples were assumed to be
from wild stocks and combined with the wild-caught finfish, even if they may have been
farmed. Most of the analysed shellfish originated from marine waters (6 524 samples),
while 1 033 samples were reported to be from inland waters. For the remaining 111
samples, this information was not given, and these unspecified samples were included
together with the results for marine finfish.

7.5.2.1 Farmed finfish from inland waters

Farmed finfish from inland waters had generally low concentrations of dioxins and dl-
PCBs (Table 7.26), but the number of samples in this category was limited. The highest
mean value of dioxins and dl-PCB, 3.36 ng TEQ/kg, was found in European eel (only
two samples), while the rest of the species and species groups in this category had mean
concentrations of both sum dioxins and sum dioxins+dl-PCBs below 0.6 ng TEQ/kg.

278
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.26 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS

(dl-PCBs) AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT
SPECIES OR SPECIES GROUPS OF FARMED FINFISH FROM INLAND WATERS (ALL REGIONS)
(ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Eel, European 2 0.352 2 3.01 2 3.36
Freshwater bream, Europe 11 0.271 11 0.292 11 0.563
Carp, common 33 0.163 33 0.345 33 0.508
Carps 16 0.133 16 0.330 16 0.463
Salmons, trouts, smelts 2 0.116 2 0.325 2 0.441
Whitefishes or coregonus 3 0.110 4 0.407 2 0.404
Salmons 2 0.087 2 0.247 2 0.334
Trouts 9 0.039 9 0.140 9 0.179
Chum salmon 2 0.049 2 0.065 2 0.114
Tilapias and similar 7 0.035 7 0.008 7 0.044
Pangas catfishes 6 0.007 6 0.007 6 0.014

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

7.5.2.2 Farmed finfish from marine waters

Farmed finfish from marine (and unspecified) waters had low mean concentrations for
dioxins and dl-PCBs for most species and species groups (Table 7.27). The highest mean
values of dioxins and dl-PCBs were found in the species group “whitefishes or coregonus”
with sum dioxins of 8.55 ng TEQ/kg and sum dioxins and dl-PCBs of 12.2 ng TEQ/kg.
The results for this species group were, however, based on only four samples and should
be interpreted with caution. The sample number was limited for most species/species
groups in this category, except for Atlantic salmon, rainbow trout and trouts. Apart from
the whitefishes or coregonus group, all species or species groups in this category had mean
values of dioxins and dl-PCBs of 2.05 ng TEQ/kg or lower (Table 7.27).

7.5.2.3 Wild-caught finfish from inland waters

For wild-caught (or unspecified) finfish from inland waters, the EFSA data showed that
the concentrations of dioxins and dl-PCBs were well below 6.5 ng TEQ/kg (maximum
level in the European Union) for most species and species groups (Table 7.28).
Only three species and species groups (river eels, European eel and barbs from FAO
area 05 Europe), had higher mean values of dioxins and dl-PCBs, between 8.24 and
9.12 ng TEQ/kg. Dl-PCBs contributed most to the total sum for these three species
and species groups, and the mean values for sum dioxins in these species groups
were not particularly high (0.803-1.15 ng TEQ/kg). All other species and species
groups had mean concentrations of sum dioxins and dl-PCBs below 2.0 ng TEQ/kg.
Most of the samples originated from FAO area 05 Europe, and most of the species
and species groups were different between the different FAO areas. Two species and
species groups, perch and European perch, were reported from both FAO area 05
Europe and FAO area 01 Africa, and for these two species and species groups, the
concentrations of dioxins and dl-PCBs were higher in Europe than in Africa.

279
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.27 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF FARMED FINFISH FROM MARINE WATERS (ALL REGIONS) (ng TOXIC
EQUIVALENT QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Whitefishes or coregonus 4 8.55 4 3.64 4 12.19
Brown trout 4 1.09 4 0.960 4 2.05
Herrings 2 0.891 2 1.05 2 1.94
River eels 5 0.317 5 1.06 5 1.37
Salmons, trouts, smelts 15 0.472 18 1.08 13 1.37
Halibut, Atlantic 10 0.313 10 0.666 10 0.980
Arctic char 8 0.206 8 0.557 8 0.764
Char 2 0.171 2 0.442 2 0.613
Atlantic salmon 675 0.247 675 0.315 675 0.563
Gilthead seabream 2 0.112 2 0.431 2 0.543
Turbot 7 0.153 7 0.340 7 0.493
Trouts 95 0.117 94 0.319 94 0.436
Rainbow trout 110 0.133 110 0.215 110 0.348
Sturgeon 13 0.161 13 0.179 13 0.340
Sea bass 30 0.047 30 0.279 30 0.326
Sea bream 39 0.062 38 0.221 37 0.280
Salmons 4 0.054 4 0.125 4 0.179
Cod 2 0.030 2 0.010 2 0.040

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

7.5.2.4 Wild-caught finfish from marine waters

Finfish sampled from wild stocks (or unknown) in marine waters made up the largest
volume of results reported in the EFSA dataset for dioxins and dl-PCBs, and results for
a wide range of species/species groups from several different geographical areas were
reported (Table 7.29 and Table 7.30). When the results were summarized independent
of geographic origin (Table 7.29), mean values for all species/species groups were
below 3.5 ng TEQ/kg for dioxins and below 6.5 ng TEQ/kg for dioxins and dl-PCBs
(maximum levels in European Union). Nevertheless, several species/species groups,
such as whitefishes or coregonus, smelt, shads, European eel and river eels had quite
high mean concentrations of sum dioxins and sum dioxins and dl-PCBs, close to these
levels. When the results were summarized for species/species groups within different
areas (Table 7.30), the highest mean values of dioxins and dl-PCBs were found for tuna
from FAO area 37 (Mediterranean and Black Sea), at 26.1 ng TEQ/kg, and shads from
FAO area (27 Atlantic Northeast), 7.49 ng TEQ/kg. These results were, however, based
on very few samples and should therefore be interpreted with caution. Tuna from several
other areas around the world had much lower concentrations of dioxins and dl-PCBs.
No other species/species groups had concentrations of dioxins and dl-PCBs above 6.5 ng
TEQ/kg, but several species/species groups from FAO area 27 (Atlantic, Northeast) had

280
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

relatively high mean values of dioxins and dl-PCBs, up to 6.19 ng TEQ/kg (Table 7.30).
TABLE 7.28 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)
AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT INLAND
REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
01 AFRICA INLAND WATERS
Perch 16 0.016 17 0.024 16 0.041
Nile perch 5 0.018 5 0.013 5 0.031
Perch, European 2 0.015 2 0.008 2 0.024
04 ASIA INLAND WATERS
Pangas catfishes 44 0.056 45 0.013 44 0.069
Tilapias and similar 2 0.024 2 0.007 2 0.032
05 EUROPE INLAND WATERS
River eels 245 0.804 248 8.24 245 9.12
Eel, European 60 0.803 60 7.93 60 8.74
Barbs 20 1.15 21 6.97 20 8.24
Brook trout 2 0.146 2 1.84 2 1.98
Freshwater bream - Europe 104 0.405 104 1.32 102 1.74
Whitefishes or coregonus 54 0.407 61 0.983 54 1.42
Salmons 5 0.294 5 0.995 5 1.29
Northern pike 45 0.158 45 0.686 45 0.844
Perch, European 15 0.129 15 0.673 15 0.801
Roaches 23 0.149 23 0.611 23 0.760
Perch 35 0.156 37 0.500 35 0.681
Pike 4 0.144 4 0.416 4 0.560
Pike-perch 13 0.149 13 0.314 13 0.463
Carp, common 56 0.105 57 0.347 56 0.457
Carps 43 0.206 42 0.239 41 0.456
River lamprey 3 0.068 3 0.373 3 0.441
Salmons, trouts, smelts 2 0.109 2 0.297 2 0.406
Brown trout 3 0.060 3 0.242 3 0.303
Trouts 37 0.047 37 0.217 37 0.264
Catfishes (freshwater) 5 0.057 5 0.160 5 0.217
Rainbow trout 18 0.042 18 0.170 18 0.212
Char 21 0.060 22 0.215 21 0.193
African catfish 3 0.014 3 0.039 3 0.053

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

281
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.29 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD (OR UNSPECIFIED) FINFISH FROM MARINE OR UNSPECIFIED
WATERS (ALL REGIONS ) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Whitefishes or coregonus 56 3.02 56 2.97 56 5.98
Smelt 3 2.67 3 2.61 3 5.27
Shads 5 0.865 5 3.92 5 4.78
Eel, European 39 0.715 39 3.84 39 4.56
River eels 57 1.24 58 2.92 57 4.19
Herring, Baltic 68 2.53 68 1.31 68 3.84
Sprat 97 1.45 98 1.74 97 3.18
Salmons, trouts, smelts 79 0.810 65 1.70 61 2.77
Atlantic salmon 202 1.03 202 1.68 202 2.71
Conger 2 0.326 2 2.19 2 2.52
Mackerel, Atlantic 110 0.515 110 1.72 110 2.24
Herrings 506 1.16 510 1.07 505 2.24
Herring, Atlantic 128 1.22 128 0.901 128 2.12
Mackerel, chub 2 0.164 2 1.75 2 1.91
Mackerel 309 0.621 328 1.30 309 1.88
Brown trout 4 0.768 4 0.953 4 1.72
Flounders 22 0.763 22 0.934 22 1.70
Garfish 3 0.337 3 1.30 3 1.64
European sardine 91 0.308 91 1.22 91 1.53
Halibut, Greenland 90 0.607 90 0.885 90 1.49
Trouts 225 0.603 228 0.785 223 1.41
Halibut, Atlantic 466 0.408 466 0.999 466 1.41
Sardines and sardine-type fishes 71 0.246 81 1.17 71 1.37
Herring, Pacific 3 0.622 3 0.604 3 1.23
Atlantic mackerel 30 0.357 30 0.820 30 1.18
Brook trout 6 0.087 6 1.079 6 1.17
Salmons 321 0.384 333 0.744 321 1.14
Mullets 8 0.209 10 0.692 8 1.07
Sea bream 4 0.223 4 0.735 4 0.958
Turbot 14 0.295 14 0.627 14 0.922
Herrings, sardines, anchovies 3 0.179 4 0.946 3 0.860
Sturgeon 6 0.372 6 0.430 6 0.802
Char 6 0.208 6 0.564 6 0.772
Tuna 207 0.063 214 0.722 197 0.740
Sea bass 160 0.113 165 0.571 160 0.685
Beaked redfish 525 0.199 525 0.416 525 0.615
Golden redfish 227 0.208 227 0.395 227 0.603
Halibut 4 0.172 4 0.395 4 0.567
Plaice 101 0.221 149 0.277 101 0.501
Brill 5 0.168 5 0.307 5 0.475

282
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.29 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD (OR UNSPECIFIED) FINFISH FROM MARINE OR UNSPECIFIED
WATERS (ALL REGIONS ) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT) (cont.)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Swordfish 35 0.066 35 0.387 35 0.453
Plaice, European 74 0.171 74 0.267 74 0.439
Anchovies 4 0.103 4 0.322 4 0.426
Gilthead seabream 7 0.103 7 0.315 7 0.418
Cod 385 0.105 404 0.236 385 0.349
Pangas catfishes 22 0.152 22 0.180 22 0.332
Rainbow trout 98 0.091 98 0.207 98 0.297
Hakes 79 0.073 84 0.221 79 0.293
Sole 39 0.118 41 0.157 39 0.278
Ocean perch 25 0.100 25 0.176 25 0.276
Horse mackerels 8 0.104 8 0.141 8 0.245
Whiting 15 0.109 15 0.135 15 0.244
Pink salmon 5 0.064 5 0.174 5 0.238
Wolffishes 4 0.129 4 0.102 4 0.231
Dab or common dab 7 0.092 7 0.125 7 0.217
Dolphinfishes 4 0.046 4 0.166 4 0.212
Tuna and bonito (generic) 2 0.089 2 0.115 2 0.204
Sharks 12 0.058 12 0.094 12 0.152
Rays 16 0.069 17 0.080 16 0.151
Haddock 95 0.068 96 0.054 95 0.122
Rat fish 2 0.077 2 0.040 2 0.117
Anglerfish, monkfish and stargazers 56 0.051 56 0.053 56 0.104
Pollack, pollock 2 0.058 2 0.035 2 0.093
Coalfish 50 0.026 51 0.053 50 0.078
Pollack 36 0.058 36 0.016 36 0.075
Cod, Atlantic 44 0.017 44 0.048 44 0.065
Tuna, bigeye 6 0.014 6 0.040 6 0.055
Bonito, Eastern Pacific 2 0.019 2 0.035 2 0.054
Tuna, yellowfin 2 0.015 2 0.025 2 0.039
Snappers 10 0.013 10 0.020 10 0.032
Freshwater bream - Europe 2 0.005 2 0.016 2 0.021
Pacific salmon (generic) 10 0.003 9 0.007 9 0.011

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

283
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.30 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE
(OR UNSPECIFIED) REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
18 ARCTIC SEA
Halibut, Greenland 4 0.381 4 0.686 4 1.07
Atlantic salmon 2 0.061 2 0.163 2 0.225
21 ATLANTIC, NORTHWEST
Mackerel, Atlantic 3 0.700 3 2.11 3 2.81
European sardine 3 0.464 3 2.06 3 2.53
Halibut, Atlantic 19 0.308 19 0.562 19 0.870
Mackerel 2 0.170 2 0.648 2 0.818
Halibut, Greenland 9 0.259 9 0.377 9 0.636
Atlantic salmon 4 0.098 4 0.202 4 0.300
Sea bass 2 0.030 2 0.236 2 0.266
Coalfish 2 0.028 2 0.106 2 0.135
Rays 4 0.054 4 0.063 4 0.117
Cod, Atlantic 2 0.014 2 0.101 2 0.115
Cod 8 0.021 8 0.040 8 0.061
Haddock 4 0.016 4 0.042 4 0.058
27 ATLANTIC, NORTHEAST
Shads 2 1.26 2 6.22 2 7.49
Whitefishes or coregonus 54 3.12 54 3.07 54 6.19
Eel, European 29 0.910 29 4.81 29 5.72
Smelt 3 2.67 3 2.60 3 5.27
River eels 55 1.28 56 3.01 55 4.32
Herring, Baltic 68 2.53 68 1.31 68 3.84
Sprat 97 1.45 98 1.73 97 3.18
Salmons, trouts, smelts 70 0.894 57 1.90 53 3.12
Atlantic salmon 178 1.16 178 1.88 178 3.04
Herrings 499 1.17 500 1.08 498 2.25
Mackerel, Atlantic 86 0.511 86 1.64 86 2.15
Herring, Atlantic 127 1.22 127 0.897 127 2.12
Mullets 3 0.374 3 1.69 3 2.07
Halibut, Greenland 41 0.890 41 1.17 41 2.06
Mackerel 279 0.661 283 1.30 279 1.98
Brown trout 4 0.768 4 0.953 4 1.72
Flounders 22 0.763 22 0.934 22 1.70
Sea bass 32 0.255 33 1.40 32 1.69
Trouts 181 0.735 181 0.952 181 1.69
Garfish 3 0.337 3 1.30 3 1.64
European sardine 63 0.336 63 1.14 63 1.48
Halibut, Atlantic 428 0.411 428 1.03 428 1.44

284
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.30 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE
(OR UNSPECIFIED) REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT) (cont.)
Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Sardines and sardine-type fishes 44 0.278 44 1.05 44 1.33
Salmons 280 0.423 291 0.811 280 1.25
Herring, pacific 3 0.622 3 0.604 3 1.23
Atlantic mackerel 30 0.357 30 0.820 30 1.18
Brook trout 6 0.087 6 1.08 6 1.17
Sea bream 4 0.223 4 0.735 4 0.958
Turbot 14 0.295 14 0.627 14 0.922
Herrings, sardines, anchovies 3 0.179 3 0.681 3 0.860
Sturgeon 6 0.372 6 0.430 6 0.802
Char 6 0.208 6 0.564 6 0.772
Beaked redfish 525 0.199 525 0.416 525 0.615
Golden redfish 227 0.208 227 0.395 227 0.603
Plaice 82 0.264 127 0.309 82 0.601
Swordfish 3 0.104 3 0.418 3 0.522
Brill 5 0.168 5 0.307 5 0.475
Plaice, European 68 0.184 68 0.288 68 0.472
Cod 348 0.114 358 0.258 348 0.377
Sole 29 0.149 29 0.202 29 0.351
Hakes 54 0.082 54 0.268 54 0.350
Pangas catfishes 21 0.159 21 0.189 21 0.348
Ocean perch 21 0.114 21 0.203 21 0.316
Rainbow trout 92 0.093 92 0.209 92 0.301
Horse mackerels 7 0.104 7 0.147 7 0.251
Anchovies 2 0.078 2 0.161 2 0.239
Wolffishes 4 0.129 4 0.102 4 0.231
Dab or common dab 7 0.092 7 0.125 7 0.217
Whiting 14 0.114 14 0.101 14 0.215
Rays 9 0.086 9 0.089 9 0.175
Sharks 11 0.062 11 0.099 11 0.161
Haddock 76 0.080 77 0.058 76 0.139
Rat fish 2 0.077 2 0.040 2 0.117
Anglerfish, monkfish and stargazers 54 0.053 54 0.054 54 0.106
Tuna 43 0.032 37 0.053 35 0.093
Pollack, pollock 2 0.058 2 0.035 2 0.093
Coalfish 40 0.024 40 0.055 40 0.080
Pollack 36 0.058 36 0.016 36 0.075
Cod, Atlantic 39 0.016 39 0.047 39 0.064
31 ATLANTIC, WESTERN CENTRAL
Snappers 7 0.012 7 0.021 7 0.032

285
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.30 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE
(OR UNSPECIFIED) REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT) (cont.)
Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Eel, European 5 0.007 5 0.006 5 0.013
34 ATLANTIC, EASTERN CENTRAL
Tuna 4 0.016 4 0.129 4 0.146
Sardines and sardine-type fishes 4 0.060 4 0.060 4 0.120
Sole 4 0.009 4 0.023 4 0.032
37 MEDITERRANEAN AND BLACK SEA
Tuna 5 1.21 9 15.5 5 26.1
Eel, European 4 0.328 4 2.34 4 2.67
Mackerel, Atlantic 5 0.359 5 2.23 5 2.58
Sardines and sardine-type fishes 15 0.217 16 1.75 15 2.02
Mackerel, chub 2 0.164 2 1.75 2 1.91
European sardine 15 0.209 15 1.56 15 1.77
Salmons, trouts, smelts 4 0.256 3 0.631 3 0.870
Mackerel 2 0.241 2 0.570 2 0.811
Salmons 5 0.205 5 0.521 5 0.725
Mullets 3 0.132 3 0.398 3 0.530
Swordfish 3 0.080 3 0.439 3 0.518
Hakes 4 0.100 4 0.358 4 0.458
Gilthead seabream 6 0.111 6 0.329 6 0.440
Sea bass 96 0.078 97 0.353 96 0.432
Rainbow trout 5 0.066 5 0.189 5 0.255
Trouts 13 0.087 12 0.129 11 0.204
Cod 2 0.087 2 0.086 2 0.173
Sole 3 0.040 3 0.051 3 0.091
47 ATLANTIC, SOUTHEAST
Hakes 2 0.010 2 0.003 2 0.013
48 ATLANTIC, ANTARCTIC
Halibut, Greenland 2 0.837 2 1.08 2 1.92
51 INDIAN OCEAN, WESTERN
Swordfish 17 0.061 17 0.215 17 0.275
Tuna, bigeye 6 0.014 6 0.040 6 0.055
Tuna 18 0.015 20 0.038 17 0.049
57 INDIAN OCEAN, EASTERN
Swordfish 4 0.023 4 0.423 4 0.445
Tuna 16 0.052 16 0.082 16 0.134
61 PACIFIC, NORTHWEST
Mackerel 7 0.257 7 0.618 7 0.875
Halibut, Atlantic 2 0.167 2 0.321 2 0.489
Pink salmon 3 0.095 3 0.279 3 0.374

286
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.30 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE
(OR UNSPECIFIED) REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT) (cont.)
Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Atlantic salmon 2 0.064 2 0.125 2 0.190
Salmons 10 0.051 10 0.100 10 0.151
Coalfish 4 0.046 4 0.019 4 0.066
Salmons, trouts, smelts 2 0.025 2 0.032 2 0.057
Hakes 6 0.019 6 0.031 6 0.050
Cod 10 0.014 10 0.009 10 0.022
Pacific salmon (generic) 5 0.003 5 0.005 5 0.009
67 PACIFIC, NORTHEAST
Ocean perch 3 0.029 3 0.041 3 0.069
Plaice, European 5 0.032 5 0.030 5 0.062
Plaice 14 0.026 15 0.031 14 0.058
Salmons 2 0.012 2 0.024 2 0.036
71 PACIFIC, WESTERN CENTRAL
Tuna 52 0.039 55 0.019 52 0.058
87 PACIFIC, SOUTHEAST
Mackerel 2 0.152 2 0.395 2 0.547
Salmons 9 0.109 9 0.348 9 0.456
Atlantic salmon 4 0.050 4 0.148 4 0.198
Tuna 3 0.076 3 0.041 3 0.117
Hakes 2 0.013 2 0.004 2 0.017
CANADA, UNSPECIFIED
Halibut, Greenland 4 0.283 4 0.440 4 0.723
Herrings 3 0.268 3 0.345 3 0.613
FRANCE, UNSPECIFIED
Halibut, Atlantic 6 0.992 6 1.55 6 2.54
Mackerel, Atlantic 15 0.499 15 1.76 15 2.26
Mackerel 6 0.558 19 2.41 6 2.25
Halibut, Greenland 8 0.648 8 1.25 8 1.90
Herrings 2 0.562 5 0.62 2 1.70
Shads 2 0.583 2 1.00 2 1.58
Sardines and sardine-type fishes 3 0.303 12 1.45 3 1.55
European sardine 6 0.239 6 1.01 6 1.25
Salmons 6 0.246 7 0.512 6 0.770
Sea bass 27 0.079 30 0.412 27 0.452
Atlantic salmon 4 0.087 4 0.221 4 0.308
Hakes 5 0.028 10 0.211 5 0.216
Trouts 27 0.039 31 0.145 27 0.185
Cod 12 0.028 21 0.110 12 0.149
Sole 2 0.034 4 0.084 2 0.107

287
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.30 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN MUSCLE TISSUE OF DIFFERENT SPECIES OR
SPECIES GROUPS OF WILD-CAUGHT (OR UNSPECIFIED) FINFISH FROM DIFFERENT MARINE
(OR UNSPECIFIED) REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg
WET WEIGHT) (cont.)
Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Coalfish 2 0.017 3 0.046 2 0.060
Haddock 12 0.019 12 0.036 12 0.055
Tuna 3 0.016 5 1.19 3 0.031
GREENLAND, UNSPECIFIED
Halibut, Greenland 20 0.238 20 0.431 20 0.669
Halibut 2 0.162 2 0.317 2 0.479
Halibut, Atlantic 10 0.192 10 0.267 10 0.459
INDIAN OCEAN, UNSPECIFIED
Swordfish 6 0.044 6 0.373 6 0.417
Tuna 32 0.022 32 0.046 31 0.071
Freshwater bream - Europe 2 0.005 2 0.016 2 0.021
PACIFIC OCEAN, UNSPECIFIED
Tuna 11 0.017 11 0.055 11 0.072
Cod, Atlantic 2 0.047 2 0.019 2 0.066
Cod 2 0.010 2 0.005 2 0.015
Pacific salmon (generic) 4 0.0003 3 0.0001 3 0.0004
SPAIN, UNSPECIFIED
European sardine 2 0.292 2 1.23 2 1.53
Sea bass 3 0.061 3 0.359 3 0.420
Salmons 2 0.131 2 0.206 2 0.338
Trouts 3 0.087 3 0.154 3 0.242
Tuna 4 0.018 5 0.027 4 0.049
THE UNITED STATES, UNSPECIFIED
Salmons, trouts, smelts 2 0.124 2 0.177 2 0.301
Salmons 2 0.120 2 0.163 2 0.283
Atlantic salmon 6 0.086 6 0.131 6 0.217
VIET NAM, UNSPECIFIED
Tuna and bonito (generic) 2 0.089 2 0.115 2 0.204
Tuna 14 0.077 14 0.068 14 0.145
Mackerel 5 0.024 5 0.016 5 0.040

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

288
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.5.3 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS IN FINFISH,

SUMMARY
Data on dioxins and dl-PCBs from both the literature search and the EFSA
database showed that there was a large variation in concentrations of dioxins and
dl-PCBs depending on species/species groups. On average, farmed finfish had
lower concentrations of dioxins and dl-PCBs than wild finfish, with little difference
between farmed freshwater and farmed marine fish. Fish captured from wild stocks
in inland waters had higher concentrations of dioxins and dl-PCBs than wild-caught
finfish from marine waters.
Within the category of wild-caught finfish from inland waters, two species from the
literature data, grass carp (Ctenopharyngodon idella) and goldfish (Carassius aurata),
from Asia, and three species/species groups from the EFSA data, river eels, European
eel and barbs from Europe, stood out with particularly high concentrations of
dioxins and dl-PCBs in muscle (8.24–49 ng TEQ/kg). For European eel from FAO
area 05 Europe there was, however, a large difference between the results from
the literature data and the EFSA data. European eel (Anguilla anguilla) from the
literature data (n = 16) had a very low mean value of dioxins and dl-PCBs of 0.013
ng TEQ/kg, compared with 8.74 ng TEQ/kg for European eel from the EFSA data
(n = 60) (Table 7.23 and Table 7.28).
Wild-caught finfish from marine waters made up the largest volume of results in
both the literature data and the EFSA data. Within this category, two species groups
from the EFSA data, shads from FAO area 27 (Atlantic Northeast) and tuna from
the FAO area 37 (Mediterranean and Black Sea) stood out with particularly high
concentrations of dioxins and dl-PCBs in muscle (7.49–26.1 ng TEQ/kg) (Table 7.30).
Also, in the literature data the large tuna species in the genus Thunnus was among
the genera with highest concentrations in the area Mediterranean and Black
Sea, but the concentration in the literature data was nevertheless much lower
(2.6 ng TEQ/kg) (Table 7.24). In other marine areas, tuna and bigeye tuna from the
EFSA data and Thunnus from the literature data had much lower levels of dioxins
and dl-PCBs. Apart from shads and tuna, no other finfish from marine areas had
concentrations of dioxins and dl-PCBs above 6.5 ng TEQ/kg (maximum level in
European Union), but several species/species groups had high concentrations close
to this level, including wild-caught whitefishes or coregonus, European eel, smelt
and river eels from FAO area 27 (Table 7.30). Mackerel from the EFSA data and
the genus Scomber from the literature data had relatively high concentrations of
dioxins and dl-PCBs throughout different regions of the world’s oceans, but, in
general, there was a wide variation with regard to which species/species groups had
the highest levels in the different marine areas. Comparison of geographical areas
is difficult, since different species are represented in the different areas, and very
different amounts of data have been reported from the different areas. It appears
from the data (without any statistical analyses), that the differences in concentrations
of dioxins and dl-PCBs among species or species groups were more important than
differences between geographical areas.

289
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

7.5.4 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS IN SHELLFISH

FROM THE LITERATURE REVIEW
In the literature, results for dioxins and dl-PCBs were found for 352 analysed
samples of shellfish. Of the analysed shellfish, 212 samples were from farmed
shellfish, while the rest were from wild stocks (140 samples). Most of the analysed
shellfish originated from marine waters (328 samples), and only 24 shellfish samples
originated from inland waters.

7.5.4.1 Farmed shellfish from inland and marine waters

No data was found in the literature for dioxins and dl-PCBs in farmed shellfish
from inland waters.
Published data on farmed shellfish from marine waters was limited to 210 samples
of Mediterranean mussel (Mytilus galloprovincialis) from the coast of Europe (Table
7.31). The mean value of dioxins and dioxins and dl-PCBs in this species was higher
in FAO area 37, Mediterranean and Black Sea, than in FAO area 27, Atlantic,
Northeast, but in both areas the mean values were well below the maximum levels
for both dioxins and dioxins and dl-PCBs.

TABLE 7.31 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN TISSUE OF FARMED SHELLFISH FROM DIFFERENT
MARINE REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

FAO AREA/ DIOXINS+

N DIOXINS Dl-PCBs
SPECIES LATIN NAME dl-PCBs
27 ATLANTIC, NORTHEAST
Mytilus galloprovincialis 8 0.135 0.573 0.712
37 MEDITERRANEAN AND BLACK SEA
Mytilus galloprovincialis 202 0.465 1.78 2.25

Note: N is the number of analytical samples, including both individual and composite samples.

7.5.4.2 Wild-caught shellfish from inland waters

For wild-caught shellfish from inland waters, data from the literature review was
limited to only three species. The highest concentration of dioxins and dl-PCBs was
found in the wild-caught snail Bellamya purificata from FAO area 04, Asia, with a
very high concentration of sum dioxins and dl-PCBs of 38.5 ng TEQ/kg wet weight
(Table 7.32). The result for this species was, however, based on only three samples
and should therefore be interpreted with caution. Two other wild-caught species
or species groups of shellfish from inland waters had concentrations of dioxins and
dl-PCBs of 1.06 ng TEQ/kg wet weight, or lower (Table 7.32).

290
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.32 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN TISSUE OF WILD SHELLFISH FROM DIFFERENT
INLAND REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

SHELLFISH Dioxins+
FAO AREA/SPECIES LATIN NAME N DIOXINS Dl-PCBs
GROUP dl-PCBs
04 ASIA INLAND WATERS
Bellamya purificata Snail 3 38.5
Ampullaria gigas apix Snail 4 1.06
05 EUROPE INLAND WATERS
Eriocheir sinensis Crab 16 0.375 0.320 0.680

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

7.5.4.3 Wild-caught shellfish from marine waters

Wild-caught shellfish from marine waters made up the largest volume of data for
dioxins and dl-PCBs in shellfish from the literature review. Results were found
for several species and species groups, but the number of samples was low for
many of the species/species groups (Table 7.33). The concentrations were low in
all reported species/species groups from all areas, with mean values of sum dioxins
and dl-PCBs of 1.60 ng TEQ/kg, or lower. Wild-caught Mediterranean mussels
(Mytilus galloprovincialis) were reported for both FAO area 37, Mediterranean and
Black Sea, and for FAO area 27, Atlantic, Northeast. The results showed that this
species had lower concentrations of dioxins and dl-PCBs in FAO area 37 than in
FAO area 27 (Table 7.33), contrary to the results for farmed Mediterranean mussel
from the same areas (Table 7.31). Farmed Mediterranean mussel from FAO area
37 had the highest concentrations overall, with mean values about ten times higher
than in wild-caught Mediterranean mussel from FAO area 37 and about three times
higher than in both farmed and wild-caught Mediterranean mussel from FAO area
27 (Table 7.31 and Table 7.33).

7.5.5 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS IN SHELLFISH

FROM DATA FROM THE EUROPEAN FOOD SAFETY AUTHORITY
In the data from EFSA, values for dioxins and dl-PCBs in shellfish (as in finfish) were
reported using codes for species or species groups, rather than their less ambiguous
Latin names (which are used in the literature data). Because of this, some species
groups may include a wide range of species that may have very different levels of
dioxins and dl-PCBs. Also, some species may be present both with their specific
species names and in terms of a larger group of species. For instance, the species named
blue mussel could be included also in the species groups “mussels” or “molluscs”. The
EFSA data includes samples analysed in Europe, but includes both samples of fish of
European origin and fish imported to Europe from other countries.
The final dataset on dioxins and dl-PCBs compiled from the EFSA data contained
2 129 analysed samples of shellfish. Of these, 1 952 samples had results for both dioxins
and dl-PCBs (all 29 congeners), 3 samples had results only for dioxins (17 congeners)

291
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.33 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN TISSUE OF WILD SHELLFISH FROM DIFFERENT
MARINE REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

SHELLFISH Dioxins+
FAO AREA/SPECIES LATIN NAME N DIOXINS Dl-PCBs
GROUP dl-PCBs
27 ATLANTIC, NORTHEAST
Eusergestes arcticus Shrimp 4 0.830 0.720 1.60
Mytilus galloprovincialis Bivalve 12 0.209 0.504 0.708
Pasiphaea spp. Shrimp 3 0.370 0.280 0.660
Meganyctiphanes norvegica Krill 3 0.290 0.260 0.540
Ostrea edulis Bivalve 2 0.220 0.670
37 MEDITERRANEAN AND BLACK SEA
Illex coindetii Squid 3 0.774 0.405 0.921
Mytilus galloprovincialis Bivalve 33 0.0835 0.240 0.220
«Clam» Bivalve 2 0.047 0.12 0.167
«Squid» Squid 2 0.025 0.13 0.155
Chamelea gallina Bivalve 14 0.036 0.154
«Shrimp» Shrimp 2 0.051 0.09 0.141
Cuttlefish Cuttlefish 2 0.025 0.04 0.065
Hexaplex trunculus Snail 9 2.72
27 ATLANTIC, NORTHEAST OR 37 MEDITERRANEAN AND BLACK SEA
Mytilus galloprovincialis Bivalve 2 0.320 0.120 0.440
Chamelea gallina Bivalve 2 0.280 0.060 0.340
Argopecten irradians or Aequipecten Bivalve 2 0.120 0.070 0.190
opercularis
Aristeus antennatus Shrimp 2 0.080 0.010 0.090
Squilla mantis Shrimp 2 0.080 0.010 0.090
Nephrops norvegicus Lobster 2 0.070 0.010 0.080
Loligo vulgaris Squid 2 0.020 0.020 0.040
Octopus vulgaris Octopus 2 0.040 0.001 0.040
Sepia officinalis Cuttlefish 2 0.010 0.001 0.010

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

and 174 samples had results only for dl-PCBs (12 congeners). Of the analysed
shellfish, 218 samples were from farmed shellfish and 223 samples were from wild
stocks. The remaining 1 688 samples were reported without indication as to whether
the samples were from shellfish farming or from wild stocks. These unspecified
samples were assumed to be from wild stocks and were combined with the wild-
caught shellfish, even if they may have been farmed. Most of the analysed shellfish
originated from marine waters (1 459 samples), with only 75 samples being reported
to be from inland waters. There was also a large portion of the samples (n = 595)
for which this information was not given, and these were included together with
the results for marine shellfish. Analysed tissue for shellfish was not specified in the
dataset, but it is assumed that edible tissue was used, either muscle tissue (such as
for crustaceans) or the whole soft part of the animal (as in the case of for bivalves).

292
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

7.5.5.1 Farmed shellfish from inland and marine waters

The dataset obtained from the EFSA database included no data on dioxins and dl-
PCBs in farmed shellfish from inland waters, except for a single sample of mussels
from European inland waters.
For farmed shellfish from marine waters, data was available for several species/
species groups, and the results showed that the mean values of dioxins and dl-PCBs
were below 0.80 ng TEQ/kg in all species and species groups reported (Table 7.34).
The highest mean values were found in the species/species groups oysters, mussels
and European oyster, with values between 0.601 and 0.792 ng TEQ/kg. All other
species/species groups had mean values below 0.25 ng TEQ/kg.

TABLE 7.34 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-

PCBs) AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS
OF FARMED SHELLFISH FROM MARINE WATERS (ALL REGIONS) (ng TOXIC EQUIVALENT
QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Oysters 20 0.430 20 0.362 20 0.792
Mussels 22 0.218 22 0.407 22 0.625
Oyster, European 13 0.478 13 0.123 13 0.601
Blue mussel 117 0.189 117 0.053 117 0.242
Lobster, European 2 0.095 2 0.098 2 0.193
Molluscs 2 0.120 2 0.054 2 0.174
Clams 3 0.052 3 0.065 3 0.117
Crabs, sea-spiders 2 0.041 2 0.067 2 0.108
Scallop, great 17 0.080 17 0.006 17 0.087
Water snails, conches and whelks 3 0.072 3 0.006 3 0.077
Scallops, pectens 9 0.068 9 0.005 9 0.073
Shrimps and prawns 6 0.017 6 0.010 6 0.027

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

7.5.5.2 Wild-caught shellfish from inland waters

For wild-caught shellfish from inland waters, the data received from the EFSA
database showed low concentrations of dioxins and dl-PCBs for all species groups
(Table 7.35). The concentrations of dioxins and dl-PCBs in freshwater shrimps and
prawns and freshwater crayfishes varied between the different FAO areas, with
the highest mean values in FAO area 05 (Europe), and much lower mean values
in the FAO areas 04 (Asia) and 03 (South America) (Table 7.36). The results for
the two latter FAO areas were, however, based on very few samples and should be
interpreted with caution.

293
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.35 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS OF WILD
(OR UNSPECIFIED) SHELLFISH FROM INLAND WATERS (ALL REGIONS) (ng TOXIC EQUIVALENT
QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Freshwater shrimps or prawns 29 0.291 31 0.382 29 0.698
Mussels 2 0.124 2 0.331 2 0.454
Freshwater crayfishes 27 0.228 29 0.191 27 0.420
Oysters 5 0.116 5 0.154 5 0.270
Shrimps and prawns 4 0.096 4 0.022 4 0.117
Metapenaeus shrimps 3 0.006 3 0.004 3 0.010

Notes: N is the number of analytical samples and may include both individual and composite samples. Cases where N = 1 are
excluded.

TABLE 7.36 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS OF WILD-
CAUGHT (OR UNSPECIFIED) SHELLFISH FROM DIFFERENT INLAND REGIONS (FAO AREAS) (NG
TOXIC EQUIVALENT QUOTIENT/KG WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
03 SOUTH-AMERICA INLAND WATERS
Freshwater shrimps or prawns 2 0.014 2 0.009 2 0.023
04 ASIA INLAND WATERS
Freshwater crayfishes 2 0.051 2 0.043 2 0.094
Shrimps and prawns 2 0.014 2 0.004 2 0.018
Freshwater shrimps or prawns 2 0.011 3 0.012 2 0.018
Metapenaeus shrimps 3 0.006 3 0.004 3 0.010
05 EUROPE INLAND WATERS
Freshwater shrimps or prawns 25 0.336 25 0.471 25 0.807
Mussels 2 0.124 2 0.331 2 0.454
Freshwater crayfishes 25 0.242 27 0.202 25 0.447
Oysters 5 0.116 5 0.154 5 0.270

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

7.5.5.3 Wild-caught shellfish from marine waters

Wild-caught (or unspecified) shellfish from marine waters made up the largest
volume of shellfish results for dioxins and dl-PCBs reported in the EFSA dataset.
Results for several species/species groups from a wide range of geographical areas
were reported (Table 7.37 and Table 7.38). The highest concentrations of dioxins
and dl-PCBs were found in squids from FAO area 27 (Atlantic, Northeast) with a
mean value for sum dioxins of 5.15 ng TEQ/kg, and a mean value for sum dioxins
and dl-PCBs of 5.20 ng TEQ/kg (Table 7.38). These results were, however, based
on only six samples and should therefore be interpreted with caution. The mean
value of dioxins and dl-PCBs in wild-caught squids from other marine regions
(FAO areas 61 and 87, and Spain, unspecified area), showed much lower levels of

294
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

dioxins and dl-PCBs in squids (Table 7.38). All other wild-caught shellfish species/
species groups from all the different marine areas had mean concentrations of dioxins
and dl-PCBs of 1.72 ng TEQ/kg or lower (Table 7.38).

TABLE 7.37 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) SHELLFISH FROM MARINE OR UNSPECIFIED WATERS (ALL
REGIONS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Squids 24 1.31 27 0.064 24 1.36
Oysters 356 0.382 444 0.456 356 0.850
Crabs, sea-spiders 242 0.405 253 0.423 241 0.836
Mussels 406 0.191 426 0.354 404 0.548
Water snails, conches and whelks 22 0.124 22 0.221 22 0.346
Cuttlefish, common 7 0.112 7 0.173 7 0.285
Lobsters 14 0.104 14 0.161 14 0.265
Oyster, Pacific cupped 2 0.150 2 0.099 2 0.249
Blue mussel 17 0.136 17 0.064 17 0.200
Lobster, European 3 0.136 3 0.062 3 0.198
Lobster, Norway 63 0.125 65 0.071 63 0.196
Edible crab 17 0.104 17 0.058 17 0.162
Shrimps, common 17 0.079 17 0.071 17 0.149
Shrimps and prawns 49 0.065 65 0.059 49 0.140
Cockles 11 0.062 13 0.074 11 0.140
Scallops, pectens 274 0.064 277 0.042 274 0.105
Scallop, great 6 0.079 6 0.021 6 0.100
Clams 20 0.048 26 0.036 20 0.085
Clams, cockles, arkshells 3 0.037 3 0.024 3 0.061
Cuttlefishes 7 0.036 8 0.021 7 0.058
Octopus, curled 6 0.028 6 0.028 6 0.057
Scallop, queen 64 0.034 78 0.025 64 0.056
Squids, cuttlefishes, octopuses 17 0.021 17 0.028 17 0.049
Octopus, common 2 0.018 2 0.028 2 0.046
Prawn, northern 8 0.035 8 0.006 8 0.041
Spiny and rock lobsters 8 0.011 9 0.008 8 0.020

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

295
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.38 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) SHELLFISH FROM DIFFERENT MARINE (OR UNSPECIFIED)
REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
21 ATLANTIC, NORTHWEST
Mussels 7 0.210 7 0.411 7 0.621
Oysters 6 0.209 6 0.249 6 0.457
Lobsters 3 0.111 3 0.117 3 0.229
Crabs, sea-spiders 9 0.066 9 0.041 9 0.107
Edible crab 2 0.062 2 0.034 2 0.096
Scallops, pectens 33 0.031 33 0.018 33 0.050
27 ATLANTIC, NORTHEAST
Squids 6 5.15 6 0.050 6 5.20
Shrimps and prawns 5 0.454 6 0.548 5 1.11
Crabs, sea-spiders 204 0.452 212 0.477 203 0.936
Oysters 142 0.406 144 0.453 142 0.863
Mussels 146 0.278 148 0.462 145 0.746
Cuttlefish, common 4 0.179 4 0.296 4 0.475
Shrimps, common 4 0.196 4 0.276 4 0.473
Water snails, conches and whelks 20 0.119 20 0.199 20 0.318
Oyster, pacific cupped 2 0.150 2 0.099 2 0.249
Lobsters 6 0.109 6 0.108 6 0.218
Blue mussel 17 0.136 17 0.064 17 0.200
Lobster, Norway 54 0.132 54 0.068 54 0.200
Lobster, European 3 0.136 3 0.062 3 0.198
Edible crab 10 0.100 10 0.056 10 0.156
Scallops, pectens 184 0.071 184 0.046 184 0.117
Scallop, great 6 0.079 6 0.021 6 0.100
Squids, cuttlefishes, octopuses 3 0.033 3 0.062 3 0.095
Cockles 4 0.043 4 0.039 4 0.082
Cuttlefishes 5 0.045 5 0.030 5 0.075
Scallop, queen 32 0.046 38 0.023 32 0.071
Clams 6 0.032 6 0.021 6 0.053
31 ATLANTIC, WESTERN CENTRAL
Crabs, sea-spiders 2 0.500 2 0.740 2 1.24
Oysters 10 0.380 10 0.499 10 0.879
Mussels 4 0.264 4 0.474 4 0.739
Shrimps and prawns 5 0.019 6 0.010 5 0.030
Spiny and rock lobsters 3 0.009 4 0.005 3 0.015
34 ATLANTIC, EASTERN CENTRAL
Oysters 2 0.304 2 0.362 2 0.667
Octopus, curled 2 0.019 2 0.026 2 0.044
37 MEDITERRANEAN AND BLACK SEA
Oysters 28 0.278 29 0.447 28 0.729

296
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

TABLE 7.38 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) SHELLFISH FROM DIFFERENT MARINE (OR UNSPECIFIED)
REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT) (cont.)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Mussels 49 0.138 49 0.389 48 0.529
Scallops, pectens 3 0.057 3 0.025 3 0.081
41 ATLANTIC, SOUTHWEST
Oysters 2 0.661 2 1.06 2 1.72
Shrimps and prawns 8 0.008 8 0.003 8 0.011
Scallop, queen 5 0.007 6 0.002 5 0.010
Scallops, pectens 5 0.006 6 0.002 5 0.008
51 INDIAN OCEAN, WESTERN
Shrimps and prawns 5 0.035 5 0.008 5 0.043
57 Indian Ocean, Eastern
Shrimps and prawns 3 0.040 3 0.006 3 0.046
Shrimps, common 4 0.021 4 0.004 4 0.025
61 PACIFIC, NORTHWEST
Scallop, queen 2 0.037 2 0.036 2 0.073
Squids 4 0.013 5 0.014 4 0.030
Scallops, pectens 4 0.013 4 0.014 4 0.026
77 Pacific, Eastern Central
Scallops, pectens 2 0.010 2 0.006 2 0.015
87 PACIFIC, SOUTHEAST
Scallop, queen 9 0.017 12 0.028 9 0.053
Mussels 32 0.017 36 0.015 32 0.034
Squids, cuttlefishes, octopuses 3 0.014 3 0.004 3 0.018
Scallops, pectens 19 0.011 19 0.006 19 0.017
Shrimps and prawns 4 0.009 9 0.009 4 0.015
Squids 2 0.007 2 0.005 2 0.012
CANADA, UNSPECIFIED
Shrimps, common 2 0.020 2 0.012 2 0.032
Scallop, queen 2 0.008 2 0.008 2 0.016
Scallops, pectens 2 0.009 2 0.005 2 0.015
FRANCE, UNSPECIFIED
Oysters 164 0.386 249 0.461 164 0.870
Mussels 104 0.216 114 0.429 104 0.637
Crabs, sea-spiders 24 0.161 27 0.152 24 0.318
Scallops, pectens 15 0.158 16 0.126 15 0.270
Lobsters 4 0.046 4 0.200 4 0.246
Lobster, Norway 4 0.089 6 0.119 4 0.242
Cockles 6 0.071 8 0.097 6 0.183
Squids, cuttlefishes, octopuses 2 0.055 2 0.101 2 0.156
Clams 8 0.036 12 0.045 8 0.086
Scallop, queen 9 0.034 13 0.050 9 0.052

297
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE 7.38 MEAN CONCENTRATIONS OF DIOXINS, DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (dl-PCBs)

AND THE SUM OF DIOXINS AND dl-PCBs IN VARIOUS SPECIES OR SPECIES GROUPS OF
WILD-CAUGHT (OR UNSPECIFIED) SHELLFISH FROM DIFFERENT MARINE (OR UNSPECIFIED)
REGIONS (FAO AREAS) (ng TOXIC EQUIVALENT QUOTIENT/kg WET WEIGHT) (cont.)

Dioxins+
SPECIES/SPECIES GROUP N DIOXINS N Dl-PCBs N dl-PCBs
Shrimps and prawns 2 0.012 2 0.020 2 0.032
Squids 2 0.267
INDIAN OCEAN, UNSPECIFIED
Shrimps and prawns 4 0.031 8 0.019 4 0.053
Octopus, curled 2 0.027 2 0.008 2 0.035
Squids, cuttlefishes, octopuses 3 0.015 3 0.012 3 0.027
SPAIN, UNSPECIFIED
Mussels 60 0.077 64 0.137 60 0.211
Clams 6 0.080 6 0.037 6 0.117
Squids 2 0.024 2 0.012 2 0.037
Squids, cuttlefishes, octopuses 4 0.009 4 0.006 4 0.015
the United States, unspecified
Scallop, queen 2 0.013 2 0.007 2 0.020
Unspecified
Crabs, sea-spiders 2 0.069 2 0.008 2 0.078
VIET NAM, UNSPECIFIED
Edible crab 2 0.106 2 0.057 2 0.163
Prawn, northern 7 0.036 7 0.006 7 0.042
Shrimps and prawns 10 0.026 12 0.007 10 0.033

Notes: N is the number of analytical samples, including both individual and composite samples. Cases where N = 1 are excluded.

7.5.6 DIOXINS AND DIOXIN-LIKE POLYCHLORINATED BIPHENYLS IN SHELLFISH,

SUMMARY
The data for dioxins and dl-PCBs were much more limited for shellfish than for
finfish, with only 352 analysed samples of shellfish from the literature review and 2
129 analysed samples of shellfish from the EFSA data. The overall results for these
data showed that, in general, the concentrations of dioxins and dl-PCBs were lower
in shellfish than in finfish. Farmed shellfish had, on average, lower concentrations
than wild shellfish, and the highest concentrations of dioxins and dl-PCBs in
shellfish were found among shellfish from wild stocks. Most of the shellfish data
were from marine waters, with only a very limited number of shellfish samples from
inland waters (24 samples from the literature data and 75 samples from the EFSA
data, all wild shellfish).
For wild-caught shellfish from marine waters, the highest concentrations of dioxins
and dl-PCBs were found in the species group “squids” from FAO area 27 (Atlantic,
Northeast), with a mean value of sum dioxins exceeding 6.5 ng TEQ/kg (maximum
level in the European Union). These results were based on only six samples and

298
C H A P TE R 7 . RESULTS AND SUMMARIZATION OF
“OCCURRENCE DATA FOR MeHg AND, DIOXINS AND dl-PCBs”

should therefore be interpreted with caution. The mean value of wild-caught squids
from other marine regions, showed much lower levels. No other shellfish species/
species groups had concentrations of dioxins or dioxins and dl-PCBs higher than
6.5 ng TEQ/kg. With very few exceptions, most species/species groups (wild or
farmed) from marine areas had concentrations of dioxins and dl-PCBs well below
1.0 ng TEQ/kg wet weight.
For wild-caught shellfish from inland waters, the highest mean concentration of
dioxins and dl-PCBs were found in the snail Bellamya purificata from China, with
a concentration of sum dioxins and dl-PCBs of 38.5 ng TEQ/kg wet weight. The
results for this species were based on only three samples and should therefore be
regarded with caution. All other wild-caught species/species groups of shellfish from
inland waters had concentrations of dioxins and dl-PCBs of 1.06 ng TEQ/kg wet
weight or lower, and no data were available for farmed shellfish from inland waters.
Comparison of geographical areas is difficult, since different species are represented
in the different areas and because very different amounts of data were reported from
the different areas. Most of the shellfish samples analysed for dioxins and dl-PCBs
were from European waters, mainly from FAO areas 27 (Atlantic, Northeast) and
37 (Mediterranean and Black Sea) or from the unspecified marine areas of France
and Spain (likely to be either FAO area 27 or FAO area 37). Mean concentrations of
dioxins and dl-PCBs in bivalves from wild stocks, including Mytilus galloprovincialis,
“oysters”, “mussels” and “scallops, pectens” were generally higher in FAO area 27
than in FAO area 37. In contrast, for farmed Mytilus galloprovincialis, the mean
concentration of dioxins and dl-PCBs was considerably higher in area 37 than in area
27. A limited number of shellfish results were available from several other marine
areas of the world, with some species groups represented in several different marine
areas. The species groups “shrimps and prawns” and “scallops, pectens” were among
the species groups with very low concentrations of dioxins and dl-PCBs – below
0.10 ng TEQ/kg in all the marine areas where they were represented, except in area
27, where the concentrations were somewhat higher. In general, it appears from
the data (without any statistical analyses), that the differences in concentrations of
dioxins and dl-PCBs among species or species groups were more important than
differences between geographical areas.

299
 300
© FAO/Lalo de Almeida
REFERENCES

Acosta, S., Johansson, A. & Drake, I. 2021. Diet and lifestyle factors and risk of
atherosclerotic cardiovascular disease—A prospective cohort study. Nutrients, 13(11): 3822.
Aglago, E.K., Huybrechts, I., Murphy, N., Casagrande, C., Nicolas, G., Pischon, T.,
Fedirko, V. et al. 2020. Consumption of Fish and Long-chain n-3 Polyunsaturated Fatty
Acids Is Associated With Reduced Risk of Colorectal Cancer in a Large European Cohort.
Clin Gastroenterol Hepatol, 18(3): 654-666.e6. https://doi.org/https://doi.org/10.1016/j.
cgh.2019.06.031
Ahlqwist, M., Bengtsson, C., Lapidus, L., Bergdahl, I.A. & Schütz, A. 1999. Serum
mercury concentration in relation to survival, symptoms, and diseases: results from the
prospective population study of women in Gothenburg, Sweden. Acta Odontol Scand, 57(3):
168-174.
Ai, C.E., Li, C.J., Tsou, M.C., Chen, J.L., Hsi, H.C. & Chien, L.C. 2019. Blood and seminal
plasma mercury levels and predatory fish intake in relation to low semen quality. Environ
Sci Pollut Res Int, 26(19): 19425-19433. https://doi.org/10.1007/s11356-019-04592-6
Al-Ghannami, S.S., Al-Adawi, S., Ghebremeskel, K., Hussein, I.S., Min, Y., Jeyaseelan, L.,
Al-Shammakhi, S.M., Mabry, R.M. & Al-Oufi, H.S. 2019. Randomized open-label trial of
docosahexaenoic acid-enriched fish oil and fish meal on cognitive and behavioral functioning
in Omani children. Nutrition, 57: 167-172. https://doi.org/10.1016/j.nut.2018.04.008
Al-Saleh, I., Moncari, L., Jomaa, A., Elkhatib, R., Al-Rouqi, R., Eltabache, C., Al-Rajudi, T.,
Alnuwaysir, H., Nester, M. & Aldhalaan, H. 2020. Effects of early and recent mercury and lead
exposure on the neurodevelopment of children with elevated mercury and/or developmental
delays during lactation: A follow-up study. Int J Hyg Environ Health, 230: 113629.
Al-Saleh, I., Nester, M., Abduljabbar, M., Al-Rouqi, R., Eltabache, C., Al-Rajudi, T.
& Elkhatib, R. 2016. Mercury (Hg) exposure and its effects on Saudi breastfed infant's
neurodevelopment. Int J Hyg Environ Health, 219(1): 129-141.
Alves, M.F.A., Fraiji, N.A., Barbosa, A.C., De Lima, D.S., Souza, J.R., Dórea, J.G. &
Cordeiro, G.W. 2006. Fish consumption, mercury exposure and serum antinuclear antibody
in Amazonians. Int J Environ Health Res, 16(4): 255-262.
Antunes dos Santos, A., Appel Hort, M., Culbreth, M., López-Granero, C., Farina,
M., Rocha, J.B.T. & Aschner, M. 2016. Methylmercury and brain development: A review
of recent literature. J. Trace Elem. Med. Biol., 38(Supplement C): 99-107. https://doi.org/
https://doi.org/10.1016/j.jtemb.2016.03.001
Aromataris, E., Fernandez, R., Godfrey, C.M., Holly, C., Khalil, H. & Tungpunkom, P.
2015. Summarizing systematic reviews: methodological development, conduct and reporting
of an umbrella review approach. JBI Evid Implement, 13(3): 132-140.
ATSDR (Agency for Toxic Substances and Disease Registry). 2022. Toxilogical profile
for Mercury, Atlanta, Georgia: U.S. Department of Health and Human Services, Public
Health Service. Retrived from https://www.atsdr.cdc.gov/toxprofiles/tp46.pdf [Cited 24th
October 2022].

301
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Ayotte, P., Carrier, A., Ouellet, N., Boiteau, V., Abdous, B., Sidi, E.A., Château-
Degat, M.L. & Dewailly, É. 2011. Relation between methylmercury exposure and plasma
paraoxonase activity in inuit adults from Nunavik. Environ Health Perspect, 119(8): 1077-83.
https://doi.org/10.1289/ehp.1003296
Baba, T., Ito, S., Yuasa, M., Yoshioka, E., Miyashita, C., Araki, A., Sasaki, S., Kobayashi,
S., Kajiwara, J. & Hori, T. 2018. Association of prenatal exposure to PCDD/Fs and PCBs
with maternal and infant thyroid hormones: The Hokkaido Study on Environment and
Children's Health. Sci Total Environ, 615: 1239-1246.
Baccarelli, A., Giacomini, S.M., Corbetta, C., Landi, M.T., Bonzini, M., Consonni, D.,
Grillo, P., Patterson Jr, D.G., Pesatori, A.C. & Bertazzi, P.A. 2008. Neonatal thyroid
function in Seveso 25 years after maternal exposure to dioxin. PLoS Med, 5(7): e161.
Bakre, A.T., Chen, R., Khutan, R., Wei, L., Smith, T., Qin, G., Danat, I.M., Zhou,
W., Schofield, P. & Clifford, A. 2018. Association between fish consumption and risk of
dementia: a new study from China and a systematic literature review and meta-analysis.
Public Health Nutr., 21(10): 1921-1932.
Ballester, F., Iñiguez, C., Murcia, M., Guxens, M., Basterretxea, M., Rebagliato, M.,
Vioque, J., Lertxundi, A., Fernandez-Somoano, A. & Tardon, A. 2018. Prenatal exposure
to mercury and longitudinally assessed fetal growth: relation and effect modifiers. Environ
Res, 160: 97-106.
Balthrop, J.E. & Braddon, S.A. 1985. Effects of selenium and methylmercury upon
glutathione and glutathione-s-transferase in mice. Arch Environ Contam Toxicol, 14(2):
197-202. https://doi.org/10.1007/bf01055612
Barbone, F., Valent, F., Pisa, F., Daris, F., Fajon, V., Gibicar, D., Logar, M. & Horvat,
M. 2020. Prenatal low-level methyl mercury exposure and child development in an Italian
coastal area. Neurotoxicology, 81: 376-381.
Bautista, L.E., Stein, J.H., Morgan, B.J., Stanton, N., Young, T. & Nieto, F.J. 2009.
Association of blood and hair mercury with blood pressure and vascular reactivity. WMJ:
official publication of the State Medical Society of Wisconsin, 108(5): 250.
Bechthold, A., Boeing, H., Schwedhelm, C., Hoffmann, G., Knüppel, S., Iqbal, K., De
Henauw, S., Michels, N., Devleesschauwer, B. & Schlesinger, S. 2019. Food groups and risk
of coronary heart disease, stroke and heart failure: a systematic review and dose-response
meta-analysis of prospective studies. Crit Rev Food Sci Nutr, 59(7): 1071-1090.
Bélanger, M.C., Mirault, M.E., Dewailly, E., Berthiaume, L. & Julien, P. 2008a.
Environmental contaminants and redox status of coenzyme Q10 and vitamin E in Inuit
from Nunavik. Metabolism, 57(7): 927-33. https://doi.org/10.1016/j.metabol.2008.02.007
Bélanger, M.C., Mirault, M.E., Dewailly, E., Plante, M., Berthiaume, L., Noël, M. &
Julien, P. 2008b. Seasonal mercury exposure and oxidant-antioxidant status of James Bay
sport fishermen. Metabolism, 57(5): 630-6. https://doi.org/10.1016/j.metabol.2007.12.006
Bélanger, M.C., Dewailly, E., Berthiaume, L., Noël, M., Bergeron, J., Mirault, M.E. &
Julien, P. 2006. Dietary contaminants and oxidative stress in Inuit of Nunavik. Metabolism,
55(8): 989-95. https://doi.org/10.1016/j.metabol.2006.03.007
Belles-Isles, M., Ayotte, P., Dewailly, E., Weber, J.-P. & Roy, R. 2002. Cord blood
lymphocyte functions in newborns from a remote maritime population exposed to
organochlorines and methylmercury. J Toxicol Env Heal A, 65(2): 165-182.

302
REFERENCES

Bergdahl, I.A., Ahlqwist, M., Barregard, L., Björkelund, C., Blomstrand, A., Skerfving,
S., Sundh, V., Wennberg, M. & Lissner, L. 2013. Mercury in serum predicts low risk of
death and myocardial infarction in Gothenburg women. Int Arch Occup Environ Health,
86: 71-77.
Berk, M., Williams, L.J., Andreazza, A.C., Pasco, J.A., Dodd, S., Jacka, F.N., Moylan, S.,
Reiner, E.J. & Magalhaes, P.V. 2014. Pop, heavy metal and the blues: secondary analysis of
persistent organic pollutants (POP), heavy metals and depressive symptoms in the NHANES
National Epidemiological Survey. BMJ Open, 4(7): e005142.
Berlin M, Zalups, R.K. & Fowler, B.A. Chapter 33 Mercury. In: Nordberg G.F., Fowler
B.A., Nordberg M., Friberg L.T., eds. Handbook on the Toxicology of Metals. Amsterdam:
Elsevier Science.
Beulen, Y., Martinez-Gonzalez, M.A., van de Rest, O., Salas-Salvado, J., Sorli, J.V.,
Gomez-Gracia, E., Fiol, M. et al. 2018. Quality of Dietary Fat Intake and Body Weight
and Obesity in a Mediterranean Population: Secondary Analyses within the PREDIMED
Trial. Nutrients, 10(12). https://doi.org/10.3390/nu10122011
Beyrouty, P. & Chan, H.M. 2006. Co-consumption of selenium and vitamin E altered the
reproductive and developmental toxicity of methylmercury in rats. Neurotoxicol Teratol,
28(1): 49-58. https://doi.org/10.1016/j.ntt.2005.11.002
Bjerregaard, P. & Christensen, A. 2012. Selenium reduces the retention of methyl mercury
in the brown shrimp Crangon crangon. Environ Sci Tech, 46(11): 6324-9. https://doi.
org/10.1021/es300549y
Boucher, O., Muckle, G., Ayotte, P., Dewailly, E., Jacobson, S.W. & Jacobson, J.L. 2016.
Altered fine motor function at school age in Inuit children exposed to PCBs, methylmercury,
and lead. Environ Int, 95: 144-151.
Boucher, O., Muckle, G., Jacobson, J.L., Carter, R.C., Kaplan-Estrin, M., Ayotte, P.,
Dewailly, É. & Jacobson, S.W. 2014a. Domain-specific effects of prenatal exposure to PCBs,
mercury, and lead on infant cognition: results from the Environmental Contaminants and
Child Development Study in Nunavik. Environ Health Perspect, 122(3): 310-6. https://
doi.org/10.1289/ehp.1206323
Boucher, O., Muckle, G., Jacobson, J.L., Carter, R.C., Kaplan-Estrin, M., Ayotte, P.,
Dewailly, E. & Jacobson, S.W. 2014b. Domain-specific effects of prenatal exposure to PCBs,
mercury, and lead on infant cognition: results from the Environmental Contaminants and
Child Development Study in Nunavik. Environ Health Perspect, 122(3): 310-316.
Boucher, O., Bastien, C.H., Saint-Amour, D., Dewailly, E., Ayotte, P., Jacobson, J.L.,
Jacobson, S.W. & Muckle, G. 2010. Prenatal exposure to methylmercury and PCBs affects
distinct stages of information processing: an event-related potential study with Inuit children.
Neurotoxicology, 31(4): 373-384.
Bradbury, K.E., Murphy, N. & Key, T.J. 2020. Diet and colorectal cancer in UK Biobank:
a prospective study. Int J Epidemiol, 49(1): 246-258. https://doi.org/10.1093/ije/dyz064
Bradley, M.A., Barst, B.D. & Basu, N. 2017. A Review of Mercury Bioavailability in Humans
and Fish. Int J Environ Res Public Health, 14(2). https://doi.org/10.3390/ijerph14020169
Byrd, K.A., Thilsted, S.H. & Fiorella, K.J. 2021. Fish nutrient composition: a review of
global data from poorly assessed inland and marine species. Public Health Nutr, 24(3):
476-486. https://doi.org/10.1017/S1368980020003857
Cace, I.B., Milardovic, A., Prpic, I., Krajina, R., Petrovic, O., Vukelic, P., Spiric, Z., Horvat,
M., Mazej, D. & Snoj, J. 2011. Relationship between the prenatal exposure to low-level of
mercury and the size of a newborn's cerebellum. Med Hypotheses, 76(4): 514-516.

303
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Cai, H., Sobue, T., Kitamura, T., Ishihara, J., Sawada, N., Iwasaki, M., Shimazu, T. &
Tsugane, S. 2020. Association between meat and saturated fatty acid intake and lung cancer
risk: The Japan Public Health Center-based prospective study. Int J Cancer, 147(11): 3019-
3028. https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/ijc.33112?download=true
Cao, Y., Chen, A., Jones, R.L., Radcliffe, J., Caldwell, K.L., Dietrich, K.N. & Rogan, W.J.
2010. Does background postnatal methyl mercury exposure in toddlers affect cognition and
behavior? Neurotoxicology, 31(1): 1-9.
Carrasco, P., Estarlich, M., Iñiguez, C., Ferrero, A., Murcia, M., Esplugues, A., Vioque, J.,
Santa Marina, L., Zabaleta, C. & Iriarte, G. 2021. Pre and postnatal exposure to mercury
and respiratory health in preschool children from the Spanish INMA Birth Cohort Study.
Sci Total Environ, 782: 146654.
Castriotta, L., Rosolen, V., Biggeri, A., Ronfani, L., Catelan, D., Mariuz, M., Bin, M.,
Brumatti, L.V., Horvat, M. & Barbone, F. 2020. The role of mercury, selenium and the
Se-Hg antagonism on cognitive neurodevelopment: A 40-month follow-up of the Italian
mother-child PHIME cohort. Int J Hyg Environ Health, 230: 113604.
Chan, P.H.Y., Kwok, K.M., Chan, M.H.M., Li, A.M., Chan, I.H.S., Fok, T.F. & Lam, H.S.
2021. Prenatal methylmercury exposure is associated with decrease heart rate variability in
children. Environ Res, 200: 111744.
Chang, L.W. 1983. Protective effects of selenium against methylmercury neurotoxicity: a
morphological and biochemical study. Experimental Pathology, 23(3): 143-156. https://doi.
org/10.1016/s0232-1513(83)80052-8
Chapman, L. & Chan, H.M. 2000. The influence of nutrition on methyl mercury
intoxication. Environ Health Perspect, 108 Suppl 1: 29-56. https://www.ncbi.nlm.nih.gov/
pmc/articles/PMC1637774/pdf/envhper00310-0034.pdf
Chen, C., Xun, P., McClure, L.A., Brockman, J., MacDonald, L., Cushman, M., Cai, J.,
Kamendulis, L., Mackey, J. & He, K. 2018. Serum mercury concentration and the risk of
ischemic stroke: the reasons for geographic and racial differences in stroke trace element
study. Environ Int, 117: 125-131. https://doi.org/10.1016/j.envint.2018.05.001
Chen, Z., Franco, O.H., Lamballais, S., Ikram, M.A., Schoufour, J.D., Muka, T. &
Voortman, T. 2020. Associations of specific dietary protein with longitudinal insulin
resistance, prediabetes and type 2 diabetes: The Rotterdam Study. Clin Nutr, 39(1): 242-
249. https://doi.org/10.1016/j.clnu.2019.01.021
Chen, Z., Myers, R., Wei, T.Y., Bind, E., Kassim, P., Wang, G.Y., Ji, Y.L. et al. 2014.
Placental transfer and concentrations of cadmium, mercury, lead, and selenium in mothers,
newborns, and young children. J Expo Sci Environ Epidemiol, 24(5): 537-544. https://doi.
org/10.1038/jes.2014.26
Cho, G.J., Park, H.T., Shin, J.H., Hur, J.Y., Kim, S.H., Lee, K.W. & Kim, T. 2012. The
relationship between blood mercury level and osteoporosis in postmenopausal women.
Menopause, 19(5): 576-581.
Choi, A.L., Mogensen, U.B., Bjerve, K.S., Debes, F., Weihe, P., Grandjean, P. & Budtz-
Jørgensen, E. 2014. Negative confounding by essential fatty acids in methylmercury
neurotoxicity associations. Neurotoxicol Teratol, 42: 85-92.
Choi, A.L., Weihe, P., Budtz-Jørgensen, E., Jørgensen, P.J., Salonen, J.T., Tuomainen, T.-
P., Murata, K., Nielsen, H.P., Petersen, M.S. & Askham, J. 2009. Methylmercury exposure
and adverse cardiovascular effects in Faroese whaling men. Environ Health Perspect, 117(3):
367-372.

304
REFERENCES

Choi, A.L., Budtz-Jørgensen, E., Jørgensen, P.J., Steuerwald, U., Debes, F., Weihe, P. &
Grandjean, P. 2008. Selenium as a potential protective factor against mercury developmental
neurotoxicity. Environ Res, 107(1): 45-52. https://doi.org/10.1016/j.envres.2007.07.006
Chowdhury, R., Ramond, A., O'Keeffe, L.M., Shahzad, S., Kunutsor, S.K., Muka, T.,
Gregson, J., Willeit, P., Warnakula, S. & Khan, H. 2018. Environmental toxic metal
contaminants and risk of cardiovascular disease: systematic review and meta-analysis. BMJ,
362.
Chowdhury, R., Stevens, S., Gorman, D., Pan, A., Warnakula, S., Chowdhury, S., Ward, H.
et al. 2012. Association between fish consumption, long chain omega 3 fatty acids, and risk
of cerebrovascular disease: systematic review and meta-analysis. BMJ, 345: e6698. https://
doi.org/10.1136/BMJ.e6698
Choy, C.M., Yeung, Q.S., Briton-Jones, C.M., Cheung, C.-K., Lam, C.W. & Haines, C.J.
2002. Relationship between semen parameters and mercury concentrations in blood and in
seminal fluid from subfertile males in Hong Kong. Fertil Steril, 78(2): 426-428.
Clarke, J. 2011. What is a systematic review? Evidence-Based Nursing, 14(3): 64-64.
Cochrane Library. 2023. About Cochrane Reviews. [Cited 20 September 2023]. https://
www.cochranelibrary.com/about/about-cochrane-reviews.
Crump, K.S., Kjellstrom, T., Shipp, A.M., Silvers, A. & Stewart, A. 1998. Influence of
prenatal mercury exposure upon scholastic and psychological test performance: benchmark
analysis of a New Zealand cohort. Risk Anal, 18(6): 701-13. https://doi.org/10.1023/
B:RIAN.0000005917.52151.e6
Cuvin-Aralar, M.L.A. & Furness, R.W. 1991. Mercury and selenium interaction: a review.
Ecotoxicol Environ Saf, 21(3): 348-364.
Dack, K., Fell, M., Taylor, C.M., Havdahl, A. & Lewis, S.J. 2021. Mercury and prenatal
growth: a systematic review. Int J Environ Res Public Health, 18(13): 7140.
Davidson, P.W., Myers, G.J., Cox, C., Axtell, C., Shamlaye, C., Sloane-Reeves, J.,
Cernichiari, E. et al. 1998. Effects of prenatal and postnatal methylmercury exposure from
fish consumption on neurodevelopment: outcomes at 66 months of age in the Seychelles
Child Development Study. JAMA., 280(8): 701-7. https://doi.org/10.1001/JAMA.280.8.701
De Craemer, S., Croes, K., Van Larebeke, N., De Henauw, S., Schoeters, G., Govarts,
E., Loots, I., Nawrot, T., Nelen, V. & Den Hond, E. 2017. Metals, hormones and sexual
maturation in Flemish adolescents in three cross-sectional studies (2002–2015). Environ
Int, 102: 190-199.
Debes, F., Weihe, P. & Grandjean, P. 2016. Cognitive deficits at age 22 years associated with
prenatal exposure to methylmercury. Cortex, 74: 358-369.
DeVito, M., Bokkers, B., van Duursen, M. B., van Ede, K., Feeley, M., Gáspár, E.A.F.,
Haws, L. et al. 2024. The 2022 World Health Organization reevaluation of human and
mammalian toxic equivalency factors for polychlorinated dioxins, dibenzofurans and
biphenyls. Regul Toxicol Pharmacol, 146:105525.
Di Giuseppe, D., Crippa, A., Orsini, N. & Wolk, A. 2014. Fish consumption and risk of
rheumatoid arthritis: a dose-response meta-analysis. Arthritis Res Ther, 16: 1-7.
Dianatinasab, M., Wesselius, A., de Loeij, T., Salehi-Abargouei, A., Yu, E.Y.W., Fararouei,
M., Brinkman, M. et al. 2021. The association between meat and fish consumption and
bladder cancer risk: a pooled analysis of 11 cohort studies. Eur J Epidemiol, 36(8): 781-792.
https://doi.org/10.1007/s10654-021-00762-4

305
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Dinse, G.E., Jusko, T.A., Whitt, I.Z., Co, C.A., Parks, C.G., Satoh, M., Chan, E.K., Rose,
K.M., Walker, N.J. & Birnbaum, L.S. 2016. Associations between selected xenobiotics and
antinuclear antibodies in the national health and nutrition examination survey, 1999–2004.
Environ Health Perspect, 124(4): 426-436.
Dórea, J.G., de Souza, J.R., Rodrigues, P., Ferrari, Í. & Barbosa, A.C. 2005. Hair mercury
(signature of fish consumption) and cardiovascular risk in Munduruku and Kayabi Indians
of Amazonia. Environ Res, 97(2): 209-219.
Drouillet-Pinard, P., Huel, G., Slama, R., Forhan, A., Sahuquillo, J., Goua, V.,
Thiébaugeorges, O., Foliguet, B., Magnin, G. & Kaminski, M. 2010. Prenatal mercury
contamination: relationship with maternal seafood consumption during pregnancy and fetal
growth in the “EDEN mother–child” cohort. Br J Nutr, 104(8): 1096-1100.
EFSA (European Food Safety Authority). 2023. Panel on Nutrition, Novel Foods and
Food Allergens (NDA). Scientific opinion on the tolerable upper intake level for selenium.
EFSA Journal, 21(1): 7704. https://doi.org/10.2903/j.efsa.2023.7704
EFSA. 2022. Dioxins and dioxin-like PCBs. https://www.EFSA.europa.eu/en/topics/topic/
dioxins-and-pcbs
EFSA. 2014a. Dietetic Products, Nutrition, Allergies. Scientific Opinion on health benefits of
seafood (fish and shellfish) consumption in relation to health risks associated with exposure
to methylmercury. EFSA Journal, 12(7): 3761.
EFSA. 2014b. Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific
opinion on dietary reference values for selenium. EFSA Journal, 12(10): 3846. https://doi.
org/10.2903/j.EFSA.2014.3846
EFSA. 2012. Scientific Opinion on the risk for public health related to the presence of
mercury and methylmercury in food. EFSA Journal, 10(12): 2985. https://doi.org/10.2903/j.
EFSA.2012.2985
EFSA. 2011. Panel on Contaminants in the Food Chain. Scientific Opinion on the risk to
public health related to the presence of high levels of dioxins and dioxin‐like PCBs in liver
from sheep and deer. EFSA Journal, 9(7): 2297.
EFSA. 2018. Panel on Contaminants in the Food Chain (CONTAM). Risk for animal and
human health related to the presence of dioxins and dioxin‐like PCBs in feed and food.
EFSA Journal, 16(11): e05333. https://doi.org/https://doi.org/10.2903/j.EFSA.2018.5333
Emanuele, E. & Meliker, J. 2017. Seafood intake, polyunsaturated fatty acids, blood
mercury, and serum C-reactive protein in US National Health and Nutrition Examination
Survey (2005-2006). Int J Environ Health Res, 27(2): 136-143. https://doi.org/10.1080/09
603123.2017.1292495
Emeny, R.T., Korrick, S.A., Li, Z., Nadeau, K., Madan, J., Jackson, B., Baker, E. & Karagas,
M.R. 2019. Prenatal exposure to mercury in relation to infant infections and respiratory
symptoms in the New Hampshire Birth Cohort Study. Environ Res, 171: 523-529.
Engström, K.S., Wennberg, M., Strömberg, U., Bergdahl, I.A., Hallmans, G., Jansson,
J.H., Lundh, T. et al. 2011. Evaluation of the impact of genetic polymorphisms in glutathione-
related genes on the association between methylmercury or n-3 polyunsaturated long chain
fatty acids and risk of myocardial infarction: a case-control study. Environ Health, 10: 33.
https://doi.org/10.1186/1476-069x-10-33
Etemadi, A., Abnet, C.C., Graubard, B.I., Beane-Freeman, L., Freedman, N.D., Liao, L.,
Dawsey, S.M. & Sinha, R. 2018. Anatomical subsite can modify the association between
meat and meat compounds and risk of colorectal adenocarcinoma: Findings from three large
US cohorts. Int J Cancer, 143(9): 2261-2270. https://doi.org/10.1002%2Fijc.31612

306
REFERENCES

FAO (Food and Agriculture Organization of the United Nations). 2022. The State of
World Fisheries and Aquaculture 2022. Towards Blue Transformation. Rome. https://doi.
org/10.4060/cc0461en
FAO. 2020. The State of World Fisheries and Aquaculture 2020. Sustainability in action.
Rome. . https://doi.org/10.4060/ca9229en
FAO/WHO (World Health Organization). 2010. Report of the Joint FAO/WHO Expert
Consultation on the Risks and Benefits of Fish Consumption. Rome, 25-29 January 2010.
http://www.fao.org/docrep/014/ba0136e/ba0136e00.pdf
FAO/WHO. 2004a. Safety evaluation of certain food additives and contaminants.
Methylmercury. WHO Food Additives Series, 52: 565-623. https://inchem.org/documents/
jecfa/jecmono/v52je23.htm
FAO/WHO. 2004b. Safety evaluation of certain food additives and contaminants.
Methylmercury. WHO Food Additives Series, 52: 565-623.
Fillion, M., Lemire, M., Philibert, A., Frenette, B., Weiler, H.A., Deguire, J.R., Guimaraes,
J.R.D., Larribe, F., Barbosa, F. & Mergler, D. 2011. Visual acuity in fish consumers of the
Brazilian Amazon: risks and benefits from local diet. Public Health Nutr, 14(12): 2236-2244.
Fillion, M., Lemire, M., Philibert, A., Frenette, B., Weiler, H.A., Deguire, J.R., Guimarães,
J.R., Larribe, F., Barbosa, F., Jr. & Mergler, D. 2013. Toxic risks and nutritional benefits of
traditional diet on near visual contrast sensitivity and color vision in the Brazilian Amazon.
Neurotoxicology, 37: 173-81. https://doi.org/10.1016/j.neuro.2013.04.010
Fillion, M., Mergler, D., Passos, C.J.S., Larribe, F., Lemire, M. & Guimarães, J.R.D. 2006.
A preliminary study of mercury exposure and blood pressure in the Brazilian Amazon.
Environ Health, 5(1): 1-9.
Folven, K.I., Glover, C.N., Malde, M.K. & Lundebye, A.K. 2009. Does selenium modify
neurobehavioural impacts of developmental methylmercury exposure in mice? Environ
Toxicol Pharmacol, 28(1): 111-9. https://doi.org/10.1016/j.etap.2009.03.007
Frost, L. & Vestergaard, P. 2005. n− 3 Fatty acids consumed from fish and risk of atrial
fibrillation or flutter: the Danish Diet, Cancer, and Health Study. The American journal of
clinical nutrition, 81(1): 50-54.
Gallego-Vinas, G., Ballester, F. & Llop, S. 2019. Chronic mercury exposure and blood
pressure in children and adolescents: a systematic review. Environ Sci Pollut Res, 26: 2238-
2252.
Gammelmark, A., Nielsen, M.S., Bork, C.S., Lundbye-Christensen, S., Tjønneland, A.,
Overvad, K. & Schmidt, E.B. 2016. Association of fish consumption and dietary intake
of marine n-3 PUFA with myocardial infarction in a prospective Danish cohort study. Br
J Nutr, 116(1): 167-177.
Gao, Y., Ma, Y., Yu, M., Li, G., Chen, Y., Li, X., Chen, X., Xie, Y. & Wang, X. 2022. Poultry
and Fish Intake and Pancreatic Cancer Risk: A Systematic Review and Meta-Analysis. Nutr
Cancer, 74(1): 55-67. https://doi.org/10.1080/01635581.2020.1869276
García-Esquinas, E., Pérez-Gómez, B., Fernández-Navarro, P., Fernández, M.A., De Paz,
C., Pérez-Meixeira, A.M., Gil, E., Iriso, A., Sanz, J.C. & Astray, J. 2013. Lead, mercury and
cadmium in umbilical cord blood and its association with parental epidemiological variables
and birth factors. BMC Public Health, 13: 1-11.

307
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Geelen, A., Schouten, J.M., Kamphuis, C., Stam, B.E., Burema, J., Renkema, J.M.S.,
Bakker, E.J., van't Veer, P. & Kampman, E. 2007. Fish consumption, n-3 fatty acids, and
colorectal cancer: A meta-analysis of prospective cohort studies. Am J Epidemiol, 166(10):
1116-1125. https://doi.org/10.1093/aje/kwm197
Glaser, V., Nazari, E.M., Müller, Y.M., Feksa, L., Wannmacher, C.M., Rocha, J.B., de
Bem, A.F., Farina, M. & Latini, A. 2010. Effects of inorganic selenium administration in
methylmercury-induced neurotoxicity in mouse cerebral cortex. Int J Dev Neurosci, 28(7):
631-7. https://doi.org/10.1016/j.ijdevneu.2010.07.225
Golden, C.D., Koehn, J.Z., Shepon, A., Passarelli, S., Free, C.M., Viana, D.F., Matthey,
H. et al. 2021. Aquatic foods to nourish nations. Nature, 598(7880): 315-320. https://doi.
org/10.1038/s41586-021-03917-1
Golding, J., Hibbeln, J.R., Gregory, S.M., Iles-Caven, Y., Emond, A. & Taylor, C.M. 2017.
Maternal prenatal blood mercury is not adversely associated with offspring IQ at 8 years
provided the mother eats fish: A British prebirth cohort study. Int J Hyg Environ Health,
220(7): 1161-1167. https://doi.org/10.1016/j.ijheh.2017.07.004
Golding, J., Gregory, S., Iles-Caven, Y., Hibbeln, J., Emond, A. & Taylor, C.M. 2016a.
Associations between prenatal mercury exposure and early child development in the
ALSPAC study. Neurotoxicology, 53: 215-222. https://doi.org/10.1016/j.neuro.2016.02.006
Golding, J., Gregory, S., Emond, A., Iles-Caven, Y., Hibbeln, J. & Taylor, C.M. 2016b.
Prenatal mercury exposure and offspring behaviour in childhood and adolescence.
Neurotoxicology, 57: 87-94. https://doi.org/10.1016/j.neuro.2016.09.003
Grandjean, P., Poulsen, L.K., Heilmann, C., Steuerwald, U. & Weihe, P. 2010. Allergy and
sensitization during childhood associated with prenatal and lactational exposure to marine
pollutants. Environ Health Perspect, 118(10): 1429-1433.
Grandjean, P., Weihe, P., White, R.F., Debes, F., Araki, S., Yokoyama, K., Murata, K.,
Sorensen, N., Dahl, R. & Jorgensen, P.J. 1997. Cognitive deficit in 7-year-old children
with prenatal exposure to methylmercury. Neurotoxicol Teratol, 19(6): 417-28. https://doi.
org/10.1016/S0892-0362(97)00097-4
Gregory, S., Iles-Caven, Y., Hibbeln, J.R., Taylor, C.M. & Golding, J. 2016. Are
prenatal mercury levels associated with subsequent blood pressure in childhood and
adolescence? The Avon prebirth cohort study. BMJ Open, 6(10). https://doi.org/10.1136/
BMJopen-2016-012425
Gribble, M.O., Cheng, A., Berger, R.D., Rosman, L. & Guallar, E. 2015. Mercury exposure
and heart rate variability: a systematic review. Current Environ Health Rep, 2: 304-314.
Grosso, G., Micek, A., Marventano, S., Castellano, S., Mistretta, A., Pajak, A. & Galvano,
F. 2016. Dietary n-3 PUFA, fish consumption and depression: A systematic review and
meta-analysis of observational studies. J Affect Disord, 205: 269-281.
Grotto, D., Barcelos, G.R.M., Valentini, J., Antunes, L.M.G., Angeli, J.P.F., Garcia, S.C.
& Barbosa, F. 2009. Low levels of methylmercury induce DNA damage in rats: protective
effects of selenium. Arch Toxicol, 83(3): 249-254. https://doi.org/10.1007/s00204-008-0353-3
Grotto, D., Barcelos, G.R.M., Valentini, J., Garcia, S.C. & Barbosa, F. 2008. Selenium
minimize DNA damage in rats exposed to low dose of methylmercury. Toxicol Lett, 180:
S183-S184. https://doi.org/10.1016/j.toxlet.2008.06.174
Guallar, E., Sanz-Gallardo, M.I., Veer, P.v.t., Bode, P., Aro, A., Gómez-Aracena, J., Kark,
J.D., Riemersma, R.A., Martín-Moreno, J.M. & Kok, F.J. 2002. Mercury, fish oils, and the
risk of myocardial infarction. N Engl J Med, 347(22): 1747-1754.

308
REFERENCES

Guangliang, L., Yong, C. & Nelson, O.D. 2012. Environmental Chemistry and Toxicology
of Mercury. Hoboken, New Jersey, John Wiley & Sons Inc.
Gupta, A., Ketchum, N., Roehrborn, C.G., Schecter, A., Aragaki, C.C. & Michalek, J.E.
2006. Serum dioxin, testosterone, and subsequent risk of benign prostatic hyperplasia: a
prospective cohort study of Air Force veterans. Environ Health Perspect, 114(11): 1649-1654.
Gustin, K., Barman, M., Skröder, H., Jacobsson, B., Sandin, A., Sandberg, A.S., Wold,
A.E., Vahter, M. & Kippler, M. 2021. Thyroid hormones in relation to toxic metal exposure
in pregnancy, and potential interactions with iodine and selenium. Environ Int, 157: 106869.
https://doi.org/10.1016/j.envint.2021.106869
Hallgren, C., Hallmans, G., Jansson, J.-H., Marklund, S., Huhtasaari, F., Schütz, A.,
Strömberg, U., Vessby, B. & Skerfving, S. 2001. Markers of high fish intake are associated
with decreased risk of a first myocardial infarction. Br J Nutr, 86(3): 397-404.
Hansen-Krone, I.J., Enga, K.F., Südduth-Klinger, J.M., Mathiesen, E.B., Njølstad, I.,
Wilsgaard, T., Watkins, S., Brækkan, S.K. & Hansen, J.-B. 2014. High fish plus fish oil
intake is associated with slightly reduced risk of venous thromboembolism: the Tromsø
Study. The Journal of nutrition, 144(6): 861-867.
He, K., Song, Y., Daviglus, M.L., Liu, K., Van Horn, L., Dyer, A.R. & Greenland, P. 2004.
Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-
analysis of cohort studies. Circulation, 109(22): 2705-2711.
Heath, J.C., Banna, K.M., Reed, M.N., Pesek, E.F., Cole, N., Li, J. & Newland, M.C. 2010.
Dietary selenium protects against selected signs of aging and methylmercury exposure.
Neurotoxicology, 31(2): 169-179. https://doi.org/10.1016/j.neuro.2010.01.003
Heilmann, C., Budtz-Jørgensen, E., Nielsen, F., Heinzow, B., Weihe, P. & Grandjean,
P. 2010. Serum concentrations of antibodies against vaccine toxoids in children exposed
perinatally to immunotoxicants. Environ Health Perspect, 118(10): 1434-1438.
Hermans, K., van den Brandt, P.A., Loef, C., Jansen, R.L.H. & Schouten, L.J. 2021. Meat
consumption and cancer of unknown primary (CUP) risk: results from The Netherlands
cohort study on diet and cancer. Eur J Nutr, 60(8): 4579-4593. https://doi.org/10.1007%2
Fs00394-021-02600-5
Hibbeln, J.R., Spiller, P., Brenna, J.T., Golding, J., Holub, B.J., Harris, W.S., Kris-
Etherton, P. et al. 2019. Relationships between seafood consumption during pregnancy
and childhood and neurocognitive development: Two systematic reviews. Prostaglandins
Leukot Essent Fatty Acids, 151: 14-36. https://doi.org/10.1016/j.plefa.2019.10.002
Hirota, T., Kusu, T. & Hirota, K. 2005. Improvement of nutrition stimulates bone mineral
gain in Japanese school children and adolescents. Osteoporosis Int, 16: 1057-1064.
Hu, X.F., Eccles, K.M. & Chan, H.M. 2017. High selenium exposure lowers the odds ratios
for hypertension, stroke, and myocardial infarction associated with mercury exposure among
Inuit in Canada. Environ Int, 102: 200-206. https://doi.org/10.1016/j.envint.2017.03.002
Hu, X.F., Lowe, M. & Chan, H.M. 2021. Mercury exposure, cardiovascular disease, and
mortality: A systematic review and dose-response meta-analysis. Environ Res, 193: 110538.
Hu, X.F., Singh, K. & Chan, H.M. 2018. Mercury exposure, blood pressure, and
hypertension: A systematic review and dose–response meta-analysis. Environ Health
Perspect, 126(07): 076002.
Hui, L.L., Chan, M.H.M., Lam, H.S., Chan, P.H.Y., Kwok, K.M., Chan, I.H.S., Li, A.M.
& Fok, T.F. 2016. Impact of fetal and childhood mercury exposure on immune status in
children. Environ Res, 144: 66-72. https://doi.org/10.1016/j.envres.2015.11.005

309
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Ierodiakonou, D., Garcia-Larsen, V., Logan, A., Groome, A., Cunha, S., Chivinge, J.,
Robinson, Z. et al. 2016. Timing of Allergenic Food Introduction to the Infant Diet and
Risk of Allergic or Autoimmune Disease: A Systematic Review and Meta-analysis. JAMA,
316(11): 1181-1192. https://doi.org/10.1001/JAMA.2016.12623
Iso, H., Kobayashi, M., Ishihara, J., Sasaki, S., Okada, K., Kita, Y., Kokubo, Y. & Tsugane,
S. 2006. Intake of fish and n3 fatty acids and risk of coronary heart disease among Japanese:
the Japan Public Health Center-Based (JPHC) Study Cohort I. Circulation, 113(2): 195-202.
https://doi.org/10.1007%2Fs00394-021-02600-5
Jackson, L.W., Zullo, M.D. & Goldberg, J. 2008. The association between heavy metals,
endometriosis and uterine myomas among premenopausal women: National Health and
Nutrition Examination Survey 1999–2002. Hum Reprod, 23(3): 679-687.
Jacobson, J.L., Muckle, G., Ayotte, P., Dewailly, É. & Jacobson, S.W. 2015. Relation of
prenatal methylmercury exposure from environmental sources to childhood IQ. Environ
Health Perspect, 123(8): 827-833.
Jafari, T., Rostampour, N., Fallah, A.A. & Hesami, A. 2017. The association between
mercury levels and autism spectrum disorders: a systematic review and meta-analysis.
J Trace Elem Med Biol, 44: 289-297.
Jayedi, A. & Shab-Bidar, S. 2020. Fish Consumption and the Risk of Chronic Disease: An
Umbrella Review of Meta-Analyses of Prospective Cohort Studies. Adv Nutr, 11(5): 1123-
1133. https://doi.org/10.1093/advances/nmaa029
Jayedi, A., Zargar, M.S. & Shab-Bidar, S. 2019. Fish consumption and risk of myocardial
infarction: a systematic review and dose-response meta-analysis suggests a regional
difference. Nutr Res, 62: 1-12.
JECFA (Joint FAO/WHO Expert Committee on Food Additives). 2002. Evaluation
of Certain Food Additives and Contaminants: Fifty-seventh Report of the Joint FAO/
WHO Expert Committee on Food Additives. Rome. https://www.who.int/publications/i/
item/9241209097
Jin, X.L., Hidiroglou, N., Lok, E., Taylor, M., Kapal, K., Ross, N., Sarafin, K. et al. 2012.
Dietary Selenium (Se) and Vitamin E (V-E) Supplementation Modulated Methylmercury-
Mediated Changes in Markers of Cardiovascular Diseases in Rats. Cardiovasc Toxicol, 12(1):
10-24. https://doi.org/10.1007/s12012-011-9134-y
Joshi, D., Mittal, D.K., Shukla, S., Srivastav, A.K. & Srivastav, S.K. 2014. Methylmercury
toxicity: amelioration by selenium and water-soluble chelators as N-acetyl cysteine and
dithiothreitol. Cell Biochem Funct, 32(4): 351-360. https://doi.org/10.1002/cbf.3023
Kanan, S. & Samara, F. 2018. Dioxins and furans: A review from chemical and environmental
perspectives. Tren Environ Anal Chem, 17: 1-13.
Karimi, R., Vacchi-Suzzi, C. & Meliker, J.R. 2016. Mercury exposure and a shift toward
oxidative stress in avid seafood consumers. Environ Res, 146: 100-107. https://doi.
org/10.1016/j.envres.2015.12.023
Karita, K., Iwata, T., Maeda, E., Sakamoto, M. & Murata, K. 2018. Assessment of cardiac
autonomic function in relation to methylmercury neurotoxicity. Toxics, 6(3): 38.
Kazemi, A., Barati-Boldaji, R., Soltani, S., Mohammadipoor, N., Esmaeilinezhad, Z.,
Clark, C.C.T., Babajafari, S. & Akbarzadeh, M. 2021. Intake of Various Food Groups
and Risk of Breast Cancer: A Systematic Review and Dose-Response Meta-Analysis of
Prospective Studies. Adv Nutr, 12(3): 809-849. https://doi.org/10.1093/advances/nmaa147

310
REFERENCES

Kerger, B.D., Leung, H.-W., Scott, P.K. & Paustenbach, D.J. 2007. Refinements on the age-
dependent half-life model for estimating child body burdens of polychlorodibenzodioxins
and dibenzofurans. Chemosphere, 67(9): S272-S278.
Kim, Y., Ha, E.-H., Park, H., Ha, M., Kim, Y., Hong, Y.-C., Lee, E.J., Kim, H., Chang, N.
& Kim, B.-N. 2018. Prenatal mercury exposure, fish intake and neurocognitive development
during first three years of life: Prospective cohort Mothers and Children's environmental
health (MOCEH) study. Sci Total Environ, 615: 1192-1198.
Kim, Y.H., Shim, J.Y., Seo, M.S., Yim, H.J. & Cho, M.R. 2016. Relationship between blood
mercury concentration and bone mineral density in Korean men in the 2008–2010 Korean
national health and nutrition examination survey. Korean Journal of Family Medicine, 37(5):
273.
Kindgren, E., Guerrero-Bosagna, C. & Ludvigsson, J. 2019. Heavy metals in fish and
its association with autoimmunity in the development of juvenile idiopathic arthritis: a
prospective birth cohort study. Pediatr Rheumatol, 17: 1-9.
Kobayashi, S., Kishi, R., Saijo, Y., Ito, Y., Oba, K., Araki, A., Miyashita, C. et al. 2019.
Association of blood mercury levels during pregnancy with infant birth size by blood
selenium levels in the Japan Environment and Children's Study: A prospective birth cohort.
Environ Int, 125: 418-429. https://doi.org/10.1016/j.envint.2019.01.051
Kobayashi, S., Sata, F., Miyashita, C., Sasaki, S., Ban, S., Araki, A., Goudarzi, H.,
Kajiwara, J., Todaka, T. & Kishi, R. 2017. Dioxin-metabolizing genes in relation to effects
of prenatal dioxin levels and reduced birth size: the Hokkaido study. Reprod Toxicol, 67:
111-116.
Kosti, R.I., Kasdagli, M.I., Kyrozis, A., Orsini, N., Lagiou, P., Taiganidou, F. & Naska,
A. 2022. Fish intake, n-3 fatty acid body status, and risk of cognitive decline: a systematic
review and a dose–response meta-analysis of observational and experimental studies. Nutr
Rev, 80(6): 1445-1458.
Koual, M., Cano-Sancho, G., Bats, A.-S., Tomkiewicz, C., Kaddouch-Amar, Y., Douay-
Hauser, N., Ngo, C., Bonsang, H., Deloménie, M. & Lecuru, F. 2019. Associations between
persistent organic pollutants and risk of breast cancer metastasis. Environ Int, 132: 105028.
Kuras, R., Kozlowska, L., Reszka, E., Wieczorek, E., Jablonska, E., Gromadzinska, J.,
Stanislawska, M., Janasik, B. & Wasowicz, W. 2019. Environmental mercury exposure and
selenium-associated biomarkers of antioxidant status at molecular and biochemical level. A
short-term intervention study. Food Chem Toxicol, 130: 187-198. https://doi.org/10.1016/j.
fct.2019.04.056
Kuras, R., Reszka, E., Wieczorek, E., Jablonska, E., Gromadzinska, J., Malachowska, B.,
Kozlowska, L., Stanislawska, M., Janasik, B. & Wasowicz, W. 2018. Biomarkers of selenium
status and antioxidant effect in workers occupationally exposed to mercury. J Trace Elem
Med Biol, 49: 43-50. https://doi.org/10.1016/j.jtemb.2018.04.032
Lajous, M., Willett, W.C., Robins, J., Young, J.G., Rimm, E., Mozaffarian, D. & Hernán,
M.A. 2013. Changes in fish consumption in midlife and the risk of coronary heart disease
in men and women. Am J Epidemiol, 178(3): 382-391.
Lasota, A.N., Grønholdt, M.-L.M., Bork, C.S., Lundbye-Christensen, S., Schmidt, E.B. &
Overvad, K. 2019. Substitution of poultry and red meat with fish and the risk of peripheral
arterial disease: a Danish cohort study. Eur J Nutr, 58: 2731-2739.

311
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Lederman, S.A., Jones, R.L., Caldwell, K.L., Rauh, V., Sheets, S.E., Tang, D., Viswanathan,
S., Becker, M., Stein, J.L. & Wang, R.Y. 2008. Relation between cord blood mercury levels
and early child development in a World Trade Center cohort. Environ Health Perspect,
116(8): 1085-1091.
Lee, B.-E., Hong, Y.-C., Park, H., Ha, M., Koo, B.S., Chang, N., Roh, Y.-M., Kim, B.-N.,
Kim, Y.-J. & Kim, B.-M. 2010. Interaction between GSTM1/GSTT1 polymorphism and
blood mercury on birth weight. Environ Health Perspect, 118(3): 437-443.
Lemire, M., Fillion, M., Frenette, B., Passos, C.J., Guimarães, J.R., Barbosa, F., Jr. &
Mergler, D. 2011. Selenium from dietary sources and motor functions in the Brazilian
Amazon. Neurotoxicology, 32(6): 944-53. https://doi.org/10.1016/j.neuro.2011.04.005
Lemire, M., Fillion, M., Frenette, B., Mayer, A., Philibert, A., Passos, C.J.S., Guimarães,
J.R.D., Barbosa Jr, F. & Mergler, D. 2010. Selenium and mercury in the Brazilian Amazon:
opposing influences on age-related cataracts. Environ Health Perspect, 118(11): 1584-1589.
https://doi.org/10.1289/ehp.0901284
Li, F.-R., Chen, G.-C., Qin, J. & Wu, X. 2017. Dietary fish and long-chain n-3
polyunsaturated fatty acids intake and risk of atrial fibrillation: a meta-analysis. Nutrients,
9(9): 955.
Li, F., Liu, X. & Zhang, D. 2016. Fish consumption and risk of depression: a meta-analysis.
J Epidemiol Community Health, 70(3): 299-304.
Li, X., Yin, D.Q., Li, J. & Wang, R. 2014a. Protective Effects of Selenium on Mercury
Induced Immunotoxic Effects in Mice by Way of Concurrent Drinking Water Exposure.
Arch Environ Contam Toxicol, 67(1): 104-114. https://doi.org/10.1007/s00244-014-0001-2
Li, X., Yin, D.Q., Yin, J.Y., Chen, Q.Q. & Wang, R. 2014b. Dietary selenium protect
against redox-mediated immune suppression induced by methylmercury exposure. Food
Chem Toxicol, 72: 169-177. https://doi.org/10.1016/j.fct.2014.07.023
Li, Z.-M., Albrecht, M., Fromme, H., Schramm, K.-W. & De Angelis, M. 2019. Persistent
organic pollutants in human breast milk and associations with maternal thyroid hormone
homeostasis. Environ Sci Tech, 54(2): 1111-1119.
Li, Z.-M., Hernandez-Moreno, D., Main, K.M., Skakkebæk, N.E., Kiviranta, H., Toppari,
J., Feldt-Rasmussen, U., Shen, H., Schramm, K.-W. & De Angelis, M. 2018. Association
of in utero persistent organic pollutant exposure with placental thyroid hormones.
Endocrinology, 159(10): 3473-3481.
Liao, P.-Y., Liu, C.-W. & Liu, W.-Y. 2016. Bioaccumulation of mercury and polychlorinated
dibenzo-p-dioxins and dibenzofurans in salty water organisms. Environ Monit Assess, 188:
1-15.
Lim, J.-e., Nam, C., Yang, J., Rha, K.H., Lim, K.-M. & Jee, S.H. 2017. Serum persistent
organic pollutants (POPs) and prostate cancer risk: A case-cohort study. Int J Hyg Environ
Health, 220(5): 849-856.
Lim, S., Chung, H.-U. & Paek, D. 2010. Low dose mercury and heart rate variability
among community residents nearby to an industrial complex in Korea. Neurotoxicology,
31(1): 10-16.
Liu, X., Zhang, L., Li, J., Wang, J., Meng, G., Chi, M., Zhao, Y. & Wu, Y. 2019. Relative
effect potency estimates for dioxin-like compounds in pregnant women with gestational
diabetes mellitus and blood glucose outcomes based on a nested case-control study. Environ
Sci Tech, 53(13): 7792-7802.

312
REFERENCES

Llop, S., Murcia, M., Amorós, R., Julvez, J., Santa-Marina, L., Soler-Blasco, R.,
Rebagliato, M., Iñiguez, C., Aguinagalde, X. & Iriarte, G. 2020. Postnatal exposure to
mercury and neuropsychological development among preschooler children. Eur J Epidemiol,
35: 259-271.
Lucas, M., Dewailly, É., Muckle, G., Ayotte, P., Bruneau, S., Gingras, S., Rhainds, M. &
Holub, B.J. 2004. Gestational age and birth weight in relation to n− 3 fatty acids among
Inuit (Canada). Lipids, 39(7): 617-626.
Lucey, A.J., Paschos, G.K., Cashman, K.D., Martínéz, J.A., Thorsdottir, I. & Kiely, M.
2008. Influence of moderate energy restriction and seafood consumption on bone turnover
in overweight young adults. Am J Clin Nutr, 87(4): 1045-1052.
Lutsey, P.L., Steffen, L.M., Virnig, B.A. & Folsom, A.R. 2009. Diet and incident venous
thromboembolism: the Iowa Women's Health Study. Am Heart J, 157(6): 1081-1087.
Ma, Y., Yang, W., Li, T., Liu, Y., Simon, T.G., Sui, J., Wu, K., Giovannucci, E.L., Chan,
A.T. & Zhang, X. 2019. Meat intake and risk of hepatocellular carcinoma in two large US
prospective cohorts of women and men. Int J Epidemiol, 48(6): 1863-1871. https://doi.
org/10.1093/ije/dyz146
Maeda, E., Murata, K., Kumazawa, Y., Sato, W., Shirasawa, H., Iwasawa, T., Izumo, K.,
Tatsuta, N., Sakamoto, M. & Terada, Y. 2019. Associations of environmental exposures
to methylmercury and selenium with female infertility: A case-control study. Environ Res,
168: 357-363. https://doi.org/10.1016/j.envres.2018.10.007
Makiuchi, T., Sobue, T., Kitamura, T., Ishihara, J., Sawada, N., Iwasaki, M., Yamaji, T.,
Shimazu, T. & Tsugane, S. 2020. Relationship between Meat/Fish Consumption and Biliary
Tract Cancer: The Japan Public Health Center-Based Prospective Study. Cancer Epidemiol
Biomarkers Prev, 29(1): 95-102. https://doi.org/10.1158/1055-9965.epi-19-0514
Malmir, H., Larijani, B. & Esmaillzadeh, A. 2021. Fish consumption during pregnancy
and risk of allergic diseases in the offspring: A systematic review and meta-analysis. Crit
Rev Food Sci Nutr, 62(27): 7449-7459.
Mao, X.X., Chen, C., Xun, P.C., Daviglus, M., Steffen, L.M., Jacobs, D.R., Van Horn, L.,
Sidney, S., Zhu, N. & He, K. 2019. Effects of seafood consumption and toenail mercury and
selenium levels on cognitive function among American adults: 25 y of follow up. Nutrition,
61: 77-83. https://doi.org/10.1016/j.nut.2018.11.002
Marques, R.C., Dórea, J.G., McManus, C., Leao, R.S., Brandao, K.G., Marques, R.C.,
Vieira, I.H.I., Guimaraes, J.-R.D. & Malm, O. 2011. Hydroelectric reservoir inundation
(Rio Madeira Basin, Amazon) and changes in traditional lifestyle: impact on growth and
neurodevelopment of pre-school children. Public Health Nutr, 14(4): 661-669.
Marshall, W.A. & Tanner, J.M. 1970. Variations in the pattern of pubertal changes in boys.
Arch Dis Child, 45(239): 13-23.
Marumoto, M., Sakamoto, M., Nakamura, M., Marumoto, K. & Tsuruta, S. 2022. Organ-
specific accumulation of selenium and mercury in Indo-Pacific bottlenose dolphins (Tursiops
aduncus). Acta Vet Scand, 64(1): 1-7.
Matheson, E.M., Mainous III, A.G., Hill, E.G. & Carnemolla, M.A. 2009. Shellfish
consumption and risk of coronary heart disease. J Am Diet Assoc, 109(8): 1422-1426.
Matison, A.P., Mather, K.A., Flood, V.M. & Reppermund, S. 2021. Associations between
nutrition and the incidence of depression in middle-aged and older adults: A systematic
review and meta-analysis of prospective observational population-based studies. Ageing
Res Rev, 70: 101403.

313
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Mellingen, R.M., Myrmel, L.S., Rasinger, J.D., Lie, K.K., Bernhard, A., Madsen, L.
& Nøstbakken, O.J. 2022. Dietary selenomethionine reduce mercury tissue levels and
modulate methylmercury induced proteomic and transcriptomic alterations in hippocampi
of adolescent BALB/c mice. Int J Mol Sci, 23(20): 12242.
Mente, A., de Koning, L., Shannon, H.S. & Anand, S.S. 2009. A systematic review of the
evidence supporting a causal link between dietary factors and coronary heart disease. Arch
Intern Med, 169(7): 659-669.
Mesirow, M.S.C., Cecil, C., Maughan, B. & Barker, E.D. 2017. Associations between
Prenatal and Early Childhood Fish and Processed Food Intake, Conduct Problems, and
Co-Occurring Difficulties. J Abnorm Child Psych, 45(5): 1039-1049. https://doi.org/10.1007/
s10802-016-0224-y
Micha, R., Mannar, V., Afshin, A., Allemandi, L., Baker, P., Battersby, J., Bhutta, Z.,
Chen, K., Corvalan, C. & Di Cesare, M. 2020. Global nutrition report: action on equity to
end malnutrition. Bristol, UK, Development Initiatives Poverty Research.
Michalek, J.E., Akhtar, F.Z., Longnecker, M.P. & Burton, J.E. 2001. Relation of serum 2,
3, 7, 8-tetrachlorodibenzo-p-dioxin (TCDD) level to hematological examination results in
veterans of Operation Ranch Hand. Arch Environ Health, 56(5): 396-405.
Mínguez-Alarcón, L., Sergeyev, O., Burns, J.S., Williams, P.L., Lee, M.M., Korrick, S.A.,
Smigulina, L., Revich, B. & Hauser, R. 2017. A longitudinal study of peripubertal serum
organochlorine concentrations and semen parameters in young men: the Russian children's
study. Environ Health Perspect, 125(3): 460-466.
Miyake, Y., Tanaka, K., Yasutake, A., Sasaki, S. & Hirota, Y. 2011. Lack of association
of mercury with risk of wheeze and eczema in Japanese children: the Osaka Maternal and
Child Health Study. Environ Res, 111(8): 1180-1184.
Miyashita, C., Sasaki, S., Ikeno, T., Araki, A., Ito, S., Kajiwara, J., Todaka, T., Hachiya, N.,
Yasutake, A. & Murata, K. 2015. Effects of in utero exposure to polychlorinated biphenyls,
methylmercury, and polyunsaturated fatty acids on birth size. Sci Total Environ, 533: 256-
265.
Miyashita, C., Sasaki, S., Saijo, Y., Washino, N., Okada, E., Kobayashi, S., Konishi, K.,
Kajiwara, J., Todaka, T. & Kishi, R. 2011. Effects of prenatal exposure to dioxin-like
compounds on allergies and infections during infancy. Environ Res, 111(4): 551-558.
Mocarelli, P., Gerthoux, P.M., Needham, L.L., Patterson Jr, D.G., Limonta, G., Falbo, R.,
Signorini, S., Bertona, M., Crespi, C. & Sarto, C. 2011. Perinatal exposure to low doses
of dioxin can permanently impair human semen quality. Environ Health Perspect, 119(5):
713-718.
Mocarelli, P., Gerthoux, P.M., Patterson Jr, D.G., Milani, S., Limonta, G., Bertona, M.,
Signorini, S., Tramacere, P., Colombo, L. & Crespi, C. 2008. Dioxin exposure, from infancy
through puberty, produces endocrine disruption and affects human semen quality. Environ
Health Perspect, 116(1): 70-77.
Mocevic, E., Specht, I.O., Marott, J.L., Giwercman, A., Jönsson, B.A., Toft, G., Lundh,
T. & Bonde, J.P. 2013. Environmental mercury exposure, semen quality and reproductive
hormones in Greenlandic Inuit and European men: a cross-sectional study. Asian J Androl,
15(1): 97.
Monastero, R.N., Karimi, R., Nyland, J.F., Harrington, J., Levine, K. & Meliker, J.R.
2017. Mercury exposure, serum antinuclear antibodies, and serum cytokine levels in the
Long Island Study of Seafood Consumption: A cross-sectional study in NY, USA. Environ
Res, 156: 334-340. https://doi.org/10.1016/j.envres.2017.03.037

314
REFERENCES

Moniruzzaman, M., Lee, S., Park, Y., Min, T. & Bai, S.C. 2021. Evaluation of dietary
selenium, vitamin C and E as the multi-antioxidants on the methylmercury intoxicated
mice based on mercury bioaccumulation, antioxidant enzyme activity, lipid peroxidation
and mitochondrial oxidative stress. Chemosphere, 273: 129673. https://doi.org/10.1016/j.
Chemosphere.2021.129673
Mozaffarian, D., Shi, P., Morris, J.S., Grandjean, P., Siscovick, D.S., Spiegelman, D.,
Willett, W.C., Rimm, E.B., Curhan, G.C. & Forman, J.P. 2012. Mercury exposure and
risk of hypertension in US men and women in 2 prospective cohorts. Hypertension, 60(3):
645-52. https://doi.org/10.1161/hypertensionaha.112.196154
Mozaffarian, D., Shi, P.L., Morris, J.S., Spiegelman, D., Grandjean, P., Siscovick, D.S.,
Willett, W.C. & Rimm, E.B. 2011. Mercury exposure and risk of cardiovascular disease in
two US cohorts. N Engl J Med, 364(12): 1116-1125. https://doi.org/10.1056/NEJMoa1006876
Mozaffarian, D. & Rimm, E.B. 2006. Fish intake, contaminants, and human health:
evaluating the risks and the benefits. JAMA, 296(15): 1885-1899.
Muley, A., Muley, P. & Shah, M. 2014. ALA, fatty fish or marine n-3 fatty acids for
preventing DM?: a systematic review and meta-analysis. Curr Diabetes Rev, 10(3): 158-65.
https://doi.org/10.2174/1573399810666140515113137
Myers, G.J., Davidson, P.W., Shamlaye, C., Cox, C., Kost, J., Beck, C., Huang, L.-S. &
Weiss, B. 2020. The Seychelles Child Development Study of methyl mercury from fish
consumption: analysis of subscales from the Child Behaviour Checklist at age 107 months
in the main cohort. Neurotoxicology, 81: 331-338.
Myers, G.J., Davidson, P.W., Cox, C., Shamlaye, C.F., Palumbo, D., Cernichiari, E., Sloane-
Reeves, J. et al. 2003. Prenatal methylmercury exposure from ocean fish consumption in the
Seychelles child development study. Lancet, 361(9370): 1686-92. https://doi.org/10.1016/
s0140-6736(03)13371-5
Nagayama, J., Tsuji, H., Iida, T., Nakagawa, R., Matsueda, T., Hirakawa, H., Yanagawa,
T., Fukushige, J.i. & Watanabe, T. 2007. Immunologic effects of perinatal exposure to
dioxins, PCBs and organochlorine pesticides in Japanese infants. Chemosphere, 67(9):
S393-S398.
Nakamura, M., Hachiya, N., Murata, KY., Nakanishi, I., Kondo, T., Yasutake, A.,
Miyamoto, K., Ser, P.H., Omi, S. & Furusawa, H. 2014. Methylmercury exposure and
neurological outcomes in Taiji residents accustomed to consuming whale meat. Environ
Int, 68: 25-32.
Namazi, N., Brett, N.R., Bellissimo, N., Larijani, B., Heshmati, J. & Azadbakht, L. 2019.
The association between types of seafood intake and the risk of type 2 diabetes: a systematic
review and meta-analysis of prospective cohort studies. Health Promot Perspect, 9(3): 164.
National Research Council. 2000. Toxicological Effects of Methylmercury. Washington,
DC, The National Academies Press. https://doi.org/10.17226/9899
Netting, M.J., Middleton, P.F. & Makrides, M. 2014. Does maternal diet during pregnancy
and lactation affect outcomes in offspring? A systematic review of food-based approaches.
Nutrition, 30(11-12): 1225-1241.
Ng, S., Lin, C.C., Hwang, Y.H., Hsieh, W.S., Liao, H.F. & Chen, P.C. 2013. Mercury,
APOE, and children's neurodevelopment. Neurotoxicology, 37: 85-92.
Nielsen, A.B.S., Davidsen, M. & Bjerregaard, P. 2012. The association between blood
pressure and whole blood methylmercury in a cross-sectional study among Inuit in
Greenland. Environ Health, 11(1): 1-10.

315
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Nišević, J.R., Prpić, I., Kolić, I., Baždarić, K., Tratnik, J.S., Prpić, I.Š., Mazej, D., Špirić,
Z., Barbone, F. & Horvat, M. 2019. Combined prenatal exposure to mercury and LCPUFA
on newborn's brain measures and neurodevelopment at the age of 18 months. Environ Res,
178: 108682.
Nobunaga, T., Satoh, H. & Suzuki, T. 1979. Effects of sodium selenite on methylmercury
embryotoxicity and teratogenicity in mice. Toxicol Appl Pharmacol, 47(1): 79-88. https://
doi.org/10.1016/0041-008x(79)90074-7
Nyland, J.F., Wang, S.B., Shirley, D.L., Santos, E.O., Ventura, A.M., de Souza, J.M. &
Silbergeld, E.K. 2011a. Fetal and maternal immune responses to methylmercury exposure:
a cross-sectional study. Environ Res, 111(4): 584-589.
Nyland, J.F., Fillion, M., Barbosa, F., Jr., Shirley, D.L., Chine, C., Lemire, M., Mergler, D.
& Silbergeld, E.K. 2011b. Biomarkers of methylmercury exposure immunotoxicity among
fish consumers in Amazonian Brazil. Environ Health Perspect, 119(12): 1733-8. https://doi.
org/10.1289/ehp.1103741
OHAT (Office of Health Assessment and Translation). 2015. OHAT risk of bias rating
tool for human and animal studies. Office of Health Assessment and Translation (OHAT).
Division of the National Toxicology Program. National Institute of Environmental Health
Sciences. . https://ntp.niehs.nih.gov/ntp/ohat/pubs/riskofbiastool_508.pdf
OHAT. 2019. Handbook for Conducting a Literature-Based Health Assessment Using
OHAT Approach for Systematic Review and Evidence Integration. Office of Health
Assessment and Translation (OHAT). Division of the National Toxicology Program.
National Institute of Environmental Health Sciences. https://ntp.niehs.nih.gov/ntp/ohat/
pubs/handbookmarch2019_508.pdf
Ohi, G., Nishigaki, S., Seki, H., Tamura, Y., Maki, T., Minowa, K., Shimamura, Y.,
Mizoguchi, I., Inaba, Y. & Takizawa, Y. 1980. The protective potency of marine animal
meat against the neurotoxicity of methylmercury: its relationship with the organ distribution
of mercury and selenium in the rat. Food Cosmet Toxicol, 18(2): 139-145.
Oken, E., Rifas-Shiman, S.L., Amarasiriwardena, C., Jayawardene, I., Bellinger, D.C.,
Hibbeln, J.R., Wright, R.O. & Gillman, M.W. 2016. Maternal prenatal fish consumption
and cognition in mid childhood: Mercury, fatty acids, and selenium. Neurotoxicol Teratol,
57: 71-78. https://doi.org/10.1016/j.ntt.2016.07.001
Oken, E., Kleinman, K.P., Olsen, S.F., Rich-Edwards, J.W. & Gillman, M.W. 2004.
Associations of seafood and elongated n-3 fatty acid intake with fetal growth and length of
gestation: results from a US pregnancy cohort. Am J Epidemiol, 160(8): 774-83. https://doi.
org/10.1093/aje/kwh282
Olsén, L., Lind, P.M. & Lind, L. 2012. Gender differences for associations between
circulating levels of metals and coronary risk in the elderly. Int J Hyg Environ Health,
215(3): 411-417.
Orct, T., Lazarus, M., Ljubojevic, M., Sekovanic, A., Sabolic, I. & Blanusa, M. 2015.
Metallothionein, essential elements and lipid peroxidation in mercury-exposed suckling
rats pretreated with selenium. Biometals, 28(4): 701-712. https://doi.org/10.1007/s10534-
015-9859-3
Orenstein, S.T., Thurston, S.W., Bellinger, D.C., Schwartz, J.D., Amarasiriwardena, C.J.,
Altshul, L.M. & Korrick, S.A. 2014. Prenatal organochlorine and methylmercury exposure
and memory and learning in school-age children in communities near the New Bedford
Harbor Superfund site, Massachusetts. Environ Health Perspect, 122(11): 1253-1259.

316
REFERENCES

Oseredczuk, M., Salvini, S., Roe, M. & Moller, A. 2009. EuroFIR Workpackage 1.3., Task
group 4. Guidelines for Quality Index Attribution to original data from scientific literature
or reports for EuroFIR data interchange. [Cited 26 October 2022]. [https://www.eurofir.
org/wp-admin/wp-content/uploads/Deliverables/EuroFIR_Quality_Index_Guidelines.pdf].
Outzen, M., Tjønneland, A., Christensen, J. & Olsen, A. 2018. Fish consumption and
prostate cancer risk and mortality in a Danish cohort study. Eur J Cancer Prev, 27(4): 355-
360. https://doi.org/10.1097/CEJ.0000000000000330
Ouzzani, M., Hammady, H., Fedorowicz, Z. & Elmagarmid, A. 2016. Rayyan—a web
and mobile app for systematic reviews. Syst Rev, 5(1): 1-10.
Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D.,
Shamseer, L. et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting
systematic reviews. BMJ, 372: n71. https://doi.org/10.1136/BMJ.n71
Pan, Z., Guo, Y., Xiang, H., Hui, Y., Ju, H., Xu, S. & Li, L. 2020. Effects of lead, mercury,
and cadmium co-exposure on children's pulmonary function. Biol Trace Elem Res, 194(1):
115-120.
Papadopoulou, E., Botton, J., Caspersen, I.H., Alexander, J., Eggesbø, M., Haugen, M.,
Iszatt, N., Jacobsson, B., Knutsen, H.K. & Meltzer, H.M. 2021. Maternal seafood intake
during pregnancy, prenatal mercury exposure and child body mass index trajectories up to
8 years. Int J Epidemiol, 50(4): 1134-1146.
Park, H. & Kim, K. 2011. Association of blood mercury concentrations with atopic
dermatitis in adults: a population-based study in Korea. Environ Res, 111(4): 573-578.
Park, K. & Seo, E. 2017. Toenail mercury and dyslipidemia: Interaction with selenium. J
Trace Elem Med Biol, 39: 43-49. https://doi.org/10.1016/j.jtemb.2016.07.005
Park, K. & Seo, E. 2016. Association between Toenail Mercury and Metabolic Syndrome Is
Modified by Selenium. Nutrients, 8(7). https://doi.org/10.3390/nu8070424
Park, S.B., Choi, S.W. & Nam, A.Y. 2009. Hair tissue mineral analysis and metabolic
syndrome. Biol Trace Elem Res, 130: 218-228.
Park, S.K., Lee, S., Basu, N. & Franzblau, A. 2013. Associations of blood and urinary
mercury with hypertension in U.S. adults: the NHANES 2003-2006. Environ Res, 123:
25-32. https://doi.org/10.1016/j.envres.2013.02.003
Pastorino, S., Bishop, T., Sharp, S.J., Pearce, M., Akbaraly, T., Barbieri, N.B., Bes-
Rastrollo, M., Beulens, J.W., Chen, Z. & Du, H. 2021. Heterogeneity of associations
between total and types of fish intake and the incidence of type 2 diabetes: federated meta-
analysis of 28 prospective studies including 956,122 participants. Nutrients, 13(4): 1223.
Pattison, D.J., Harrison, R.A. & Symmons, D.P. 2004. The role of diet in susceptibility to
rheumatoid arthritis: a systematic review. J Rheumatol, 31(7): 1310-1319.
Paul, R., Moltó, J., Ortuño, N., Romero, A., Bezos, C., Aizpurua, J. & Gómez-Torres,
M.J. 2017. Relationship between serum dioxin-like polychlorinated biphenyls and post-
testicular maturation in human sperm. Reprod Toxicol, 73: 312-321.
Pedersen, E.B., Jørgensen, M.E., Pedersen, M.B., Siggaard, C., Sørensen, T.B., Mulvad,
G., Hansen, J.C., Asmund, G. & Skjoldborg, H. 2005. Relationship between mercury in
blood and 24-h ambulatory blood pressure in Greenlanders and Danes. Amer J Hypertens,
18(5): 612-618.

317
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Pertiwi, K., Küpers, L.K., de Goede, J., Zock, P.L., Kromhout, D. & Geleijnse, J.M. 2021.
Dietary and Circulating Long-Chain Omega-3 Polyunsaturated Fatty Acids and Mortality
Risk After Myocardial Infarction: A Long-Term Follow-Up of the Alpha Omega Cohort.
J Am Heart Assoc: e022617.
Petermann-Rocha, F., Parra-Soto, S., Gray, S., Anderson, J., Welsh, P., Gill, J., Sattar,
N., Ho, F.K., Celis-Morales, C. & Pell, J.P. 2021. Vegetarians, fish, poultry, and meat-
eaters: who has higher risk of cardiovascular disease incidence and mortality? A prospective
study from UK Biobank. Eur Heart J, 42(12): 1136-1143. https://doi.org/10.1097/
CEJ.0000000000000330
Pham, N.T., Nishijo, M., Nghiem, T.T.G., Pham, T.T., Tran, N.N., Vu, T.H., Tran, H.A.,
Phan, H.A.V., Do, Q. & Takiguchi, T. 2021. Effects of perinatal dioxin exposure on neonatal
electroencephalography (EEG) activity of the quiet sleep stage in the most contaminated
area from Agent Orange in Vietnam. Int J Hyg Environ Health, 232: 113661.
Pollack, A.Z., Schisterman, E.F., Goldman, L.R., Mumford, S.L., Albert, P.S., Jones, R.L.
& Wactawski-Wende, J. 2011. Cadmium, lead, and mercury in relation to reproductive
hormones and anovulation in premenopausal women. Environ Health Perspect, 119(8):
1156-1161.
Puty, B., Leão, L.K.R., Crespo-Lopez, M.E., Carvalho, A.P.C.P.S., Fagundes, N.C.F., Maia,
L.C. & Lima, R.R. 2019. Association between methylmercury environmental exposure and
neurological disorders: a systematic review. J Trace Elem Med Biol, 52: 100-110.
Ramon, R., Ballester, F., Aguinagalde, X., Amurrio, A., Vioque, J., Lacasana, M.,
Rebagliato, M., Murcia, M. & Iniguez, C. 2009. Fish consumption during pregnancy,
prenatal mercury exposure, and anthropometric measures at birth in a prospective mother-
infant cohort study in Spain. Am J Clin Nutr, 90(4): 1047-1055.
Raymond, L.J. & Ralston, N.V. 2004. Mercury: selenium interactions and health
implications. Seychelles Medical and Dental Journal, 7(1): 72-77.
Rhee, J., Vance, T., Lim, R., Christiani, D., Qureshi, A. & Cho, E. 2020. Association of
blood mercury levels with nonmelanoma skin cancer in the USA using National Health and
Nutrition Examination Survey data (2003–2016). Br J Dermatol, 183(3): 480-487.
Rignell-Hydbom, A., Axmon, A., Lundh, T., Jönsson, B.A., Tiido, T. & Spano, M. 2007.
Dietary exposure to methyl mercury and PCB and the associations with semen parameters
among Swedish fishermen. Environ Health, 6: 1-10.
Rocha, A.V., Cardoso, B.R., Zavarize, B., Almondes, K., Bordon, I., Hare, D.J., Favaro,
D.I.T. & Cozzolino, S.M.F. 2016. GPX1 Pro198Leu polymorphism and GSTM1 deletion
do not affect selenium and mercury status in mildly exposed Amazonian women in an urban
population. Sci Total Environ, 571: 801-808. https://doi.org/10.1016/j.scitotenv.2016.07.054
Rooney, A.A., Boyles, A.L., Wolfe, M.S., Bucher, J.R. & Thayer, K.A. 2014. Systematic
review and evidence integration for literature-based environmental health science
assessments. Environmental Health Perspect, 122(7): 711-718.
Rossa-Roccor, V. & Karim, M.E. 2021. Are US adults with low-exposure to methylmercury
at increased risk for depression? A study based on 2011–2016 National Health and Nutrition
Examination Surveys (NHANES). Int Arch Occup Environ Health, 94: 419-431.
Rothenberg, S.E., Korrick, S.A., Liu, J., Nong, Y., Nong, H., Hong, C., Trinh, E.P.,
Jiang, X., Biasini, F.J. & Ouyang, F. 2021. Maternal methylmercury exposure through rice
ingestion and child neurodevelopment in the first three years: a prospective cohort study in
rural China. Environ Health, 20(1): 50.

318
REFERENCES

Rothenberg, S.E., Yu, X., Liu, J., Biasini, F.J., Hong, C., Jiang, X., Nong, Y., Cheng, Y. &
Korrick, S.A. 2016. Maternal methylmercury exposure through rice ingestion and offspring
neurodevelopment: a prospective cohort study. Int J Hyg Environ Health, 219(8): 832-842.
Roy, C., Tremblay, P.-Y. & Ayotte, P. 2017. Is mercury exposure causing diabetes, metabolic
syndrome and insulin resistance? A systematic review of the literature. Environ Res, 156:
747-760.
Sadeghi, O., Djafarian, K., Ghorabi, S., Khodadost, M., Nasiri, M. & Shab-Bidar, S. 2019.
Dietary intake of fish, n-3 polyunsaturated fatty acids and risk of hip fracture: A systematic
review and meta-analysis on observational studies. Crit Rev Food Sci Nutr, 59(8): 1320-1333.
https://doi.org/10.1080/10408398.2017.1405908
Saghazadeh, A. & Rezaei, N. 2017. Systematic review and meta-analysis links autism and
toxic metals and highlights the impact of country development status: Higher blood and
erythrocyte levels for mercury and lead, and higher hair antimony, cadmium, lead, and
mercury. Prog Neuropsychopharmacology Biol Psychiatry, 79: 340-368.
Saint-Amour, D., Roy, M.S., Bastien, C., Ayotte, P., Dewailly, E., Després, C., Gingras, S. &
Muckle, G. 2006. Alterations of visual evoked potentials in preschool Inuit children exposed
to methylmercury and polychlorinated biphenyls from a marine diet. Neurotoxicology, 27(4):
567-78. https://doi.org/10.1016/j.neuro.2006.02.008
Sakamoto, M., Yasutake, A., Kakita, A., Ryufuku, M., Chan, H.M., Yamamoto, M.,
Oumi, S., Kobayashi, S. & Watanabe, C. 2013. Selenomethionine protects against neuronal
degeneration by methylmercury in the developing rat cerebrum. Environ Sci Technol, 47(6):
2862-8. https://doi.org/10.1021/es304226h
Sanders, A.P., Mazzella, M.J., Malin, A.J., Hair, G.M., Busgang, S.A., Saland, J.M. &
Curtin, P. 2019. Combined exposure to lead, cadmium, mercury, and arsenic and kidney
health in adolescents age 12–19 in NHANES 2009–2014. Environ Int, 131: 104993.
Sara, J., Marr, S., Smit, W., Erasmus, L. & Luus-Powell, W. 2017. Human health risks of
metals and metalloids in muscle tissue of Synodontis zambezensis Peters, 1852 from Flag
Boshielo Dam, South Africa. Afr J Aquat Sci, 42(3): 287-291.
Sarihi, S., Niknam, M., Mahjour, S., Hosseini-Bensenjan, M., Moazzen, F., Soltanabadi,
S. & Akbari, H. 2021. Toxic heavy metal concentrations in multiple sclerosis patients: A
systematic review and meta-analysis. EXCLI J, 20: 1571.
Satoh, H., Yasuda, N. & Shimai, S. 1985. Development of reflexes in neonatal mice
prenatally exposed to methylmercury and selenite. Toxicol Lett, 25(2): 199-203. https://doi.
org/10.1016/0378-4274(85)90082-7
Schwingshackl, L., Hoffmann, G., Lampousi, A.-M., Knüppel, S., Iqbal, K., Schwedhelm,
C., Bechthold, A., Schlesinger, S. & Boeing, H. 2017. Food groups and risk of type 2
diabetes mellitus: a systematic review and meta-analysis of prospective studies. Eur J
Epidemiol, 32: 363-375.
Shea, B.J., Reeves, B.C., Wells, G., Thuku, M., Hamel, C., Moran, J., Moher, D. et al.
2017. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised
or non-randomised studies of healthcare interventions, or both. BMJ, 358: j4008. https://
doi.org/10.1136/BMJ.j4008
Sheehan M.C., Burke T.A., Navas-Acien A., Breysse P.N., McGready J. & Fox M.A. 2014.
Global methylmercury exposure from seafood consumption and risk of developmental
neurotoxicity: a systematic review. Bull World Health Organ, 92(4): 254-269. https://doi.
org/10.2471/blt.12.116152

319
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Sirohi, D., Al Ramadhani, R. & Knibbs, L.D. 2021. Environmental exposures to endocrine
disrupting chemicals (EDCs) and their role in endometriosis: A systematic literature review.
Rev Environ Health, 36(1): 101-115.
Smith, J.D., Hou, T., Ludwig, D.S., Rimm, E.B., Willett, W., Hu, F.B. & Mozaffarian,
D. 2015. Changes in intake of protein foods, carbohydrate amount and quality, and long-
term weight change: results from 3 prospective cohorts. Am J Clin Nutr, 101(6): 1216-1224.
https://doi.org/10.3945/ajcn.114.100867
Steenland, K., Piacitelli, L., Deddens, J., Fingerhut, M. & Chang, L.I. 1999. Cancer, heart
disease, and diabetes in workers exposed to 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin. J Natl
Cancer Inst, 91(9): 779-786.
Steuerwald, U., Weihe, P., Jorgensen, P.J., Bjerve, K., Brock, J., Heinzow, B., Budtz-
Jorgensen, E. & Grandjean, P. 2000. Maternal seafood diet, methylmercury exposure,
and neonatal neurologic function. J Pediatr, 136(5): 599-605. https://doi.org/10.1067/
mpd.2000.102774
Stevens, G.A., Beal, T., Mbuya, M.N., Luo, H., Neufeld, L.M., Addo, O.Y., Adu-
Afarwuah, S., Alayón, S., Bhutta, Z. & Brown, K.H. 2022. Micronutrient deficiencies
among preschool-aged children and women of reproductive age worldwide: a pooled analysis
of individual-level data from population-representative surveys. The Lancet Glob Health,
10(11): e1590-e1599.
Stillings, B.R., Lagally, H., Bauersfeld, P. & Soares, J. 1974. Effect of cystine, selenium,
and fish protein on the toxicity and metabolism of methylmercury in rats. Toxicol Appl
Pharmacol, 30(2): 243-254.
Strain, J., Yeates, A.J., van Wijngaarden, E., Thurston, S.W., Mulhern, M.S., McSorley,
E.M., Watson, G.E., Love, T.M., Smith, T.H. & Yost, K. 2015. Prenatal exposure to methyl
mercury from fish consumption and polyunsaturated fatty acids: associations with child
development at 20 mo of age in an observational study in the Republic of Seychelles. Am J
Clin Nutr, 101(3): 530-537.
Strain, J.J., Davidson, P.W., Thurston, S.W., Harrington, D., Mulhern, M.S., McAfee, A.J.,
van Wijngaarden, E. et al. 2012. Maternal PUFA status but not prenatal methylmercury
exposure is associated with children's language functions at age five years in the Seychelles.
J Nutr, 142(11): 1943-9. https://doi.org/10.3945/jn.112.163493
Stratakis, N., Conti, D.V., Borras, E., Sabido, E., Roumeliotaki, T., Papadopoulou, E.,
Agier, L., Basagana, X., Bustamante, M. & Casas, M. 2020. Association of fish consumption
and mercury exposure during pregnancy with metabolic health and inflammatory biomarkers
in children. JAMA network open, 3(3): e201007-e201007.
Streppel, M.T., Ocke, M.C., Boshuizen, H.C., Kok, F.J. & Kromhout, D. 2008. Long-
term fish consumption and n-3 fatty acid intake in relation to (sudden) coronary heart
disease death: the Zutphen study. Eur Heart J, 29(16): 2024-2030. https://doi.org/10.1093/
eurheartj/ehn294
Sun, Y., Liu, B., Rong, S., Zhang, J., Du, Y., Xu, G., Snetselaar, L.G., Wallace, R.B.,
Lehmler, H.J. & Bao, W. 2021. Association of Seafood Consumption and Mercury Exposure
With Cardiovascular and All-Cause Mortality Among US Adults. JAMA Netw Open, 4(11):
e2136367. https://doi.org/10.1001/JAMAnetworkopen.2021.36367
Szymanski, K.M., Wheeler, D.C. & Mucci, L.A. 2010. Fish consumption and prostate cancer
risk: a review and meta-analysis. Am J Clin Nutr, 92(5): 1223-1233.

320
REFERENCES

Saavedra, S., Fernández-Recamales, Á., Sayago, A., Cervera-Barajas, A., González-

Domínguez, R. & Gonzalez-Sanz, J.D. 2022. Impact of dietary mercury intake during
pregnancy on the health of neonates and children: a systematic review. Nutr Rev, 80(2):
317-328.
Tajik, B., Kurl, S., Tuomainen, T.P., Savonen, K. & Virtanen, J.K. 2018. Associations of the
serum long-chain n-3 PUFA and hair mercury with resting heart rate, peak heart rate during
exercise and heart rate recovery after exercise in middle-aged men. Br J Nutr, 119(1): 66-73.
Tajik, B., Kurl, S., Tuomainen, T.P. & Virtanen, J.K. 2016. The association of serum long-
chain n-3 PUFA and hair mercury with exercise cardiac power in men. Br J Nutr, 116(3):
487-495.
Tang, M., Xu, C., Lin, N., Liu, K., Zhang, Y., Yu, X. & Liu, W. 2016. Lead, mercury,
and cadmium in umbilical cord serum and birth outcomes in Chinese fish consumers.
Chemosphere, 148: 270-275.
Tatsuta, N., Nakai, K., Sakamoto, M., Murata, K. & Satoh, H. 2018. Methylmercury
exposure and developmental outcomes in Tohoku study of child development at 18 months
of age. Toxics, 6(3): 49.
Tatsuta, N., Murata, K., Iwai-Shimada, M., Yaginuma-Sakurai, K., Satoh, H. & Nakai,
K. 2017. Psychomotor Ability in Children Prenatally Exposed to Methylmercury: The
18-Month Follow-Up of Tohoku Study of Child Development. Tohoku J Exp Med, 242(1):
1-8. https://doi.org/10.1620/tjem.242.1
Thacher, T.D., Bommersbach, T.J., Pettifor, J.M., Isichei, C.O. & Fischer, P.R. 2015.
Comparison of limestone and ground fish for treatment of nutritional rickets in children in
Nigeria. J Pediatr, 167(1): 148-154. e1.
Timmerman, R. & Omaye, S.T. 2021. Selenium's Utility in Mercury Toxicity: A Mini-
Review. Food and Nutrition Sciences, 12(02): 124.
Tong, T.Y., Appleby, P.N., Armstrong, M.E., Fensom, G.K., Knuppel, A., Papier, K., Perez-
Cornago, A., Travis, R.C. & Key, T.J. 2020. Vegetarian and vegan diets and risks of total
and site-specific fractures: results from the prospective EPIC-Oxford study. BMC medicine,
18(1): 1-15.
Tong, T.Y., Appleby, P.N., Bradbury, K.E., Perez-Cornago, A., Travis, R.C., Clarke, R.
& Key, T.J. 2019. Risks of ischaemic heart disease and stroke in meat eaters, fish eaters, and
vegetarians over 18 years of follow-up: results from the prospective EPIC-Oxford study.
BMJ, 366.
Tratnik, J.S., Falnoga, I., Trdin, A., Mazej, D., Fajon, V., Miklavcic, A., Kobal, A.B. et
al. 2017. Prenatal mercury exposure, neurodevelopment and apolipoprotein E genetic
polymorphism. Environ Res, 152: 375-385. https://doi.org/10.1016/j.envres.2016.08.035
Tsai, T.L., Kuo, C.C., Pan, W.H., Wu, T.N., Lin, P. & Wang, S.L. 2019. Type 2 diabetes
occurrence and mercury exposure–From the National Nutrition and Health Survey in
Taiwan. Environ Int, 126: 260-267.
Tu, R., Zhang, C., Feng, L., Wang, H., Wang, W. & Li, P. 2021. Impact of selenium on
cerebellar injury and mRNA expression in offspring of rat exposed to methylmercury.
Ecotoxicol Environ Saf, 223: 112584. https://doi.org/10.1016/j.ecoenv.2021.112584
Tørris, C., Molin, M. & Småstuen, M.C. 2017. Lean Fish Consumption Is Associated with
Beneficial Changes in the Metabolic Syndrome Components: A 13-Year Follow-Up Study
from the Norwegian Tromsø Study. Nutrients, 9(3).

321
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Valera, B., Dewailly, É. & Poirier, P. 2013. Association between methylmercury and
cardiovascular risk factors in a native population of Quebec (Canada): a retrospective
evaluation. Environ Res, 120: 102-108.
Valera, B., Dewailly, É., Poirier, P., Counil, E. & Suhas, E. 2011a. Influence of mercury
exposure on blood pressure, resting heart rate and heart rate variability in French Polynesians:
a cross-sectional study. Environ Health, 10: 1-10.
Valera, B., Dewailly, E. & Poirier, P. 2011b. Impact of mercury exposure on blood pressure
and cardiac autonomic activity among Cree adults (James Bay, Quebec, Canada). Environ
Res, 111(8): 1265-1270.
Valera, B., Dewailly, E. & Poirier, P. 2009. Environmental mercury exposure and blood
pressure among Nunavik Inuit adults. Hypertension, 54(5): 981-986.
Valera, B., Dewailly, E. & Poirier, P. 2008. Cardiac autonomic activity and blood pressure
among Nunavik Inuit adults exposed to environmental mercury: a cross-sectional study.
Environ Health, 7: 1-11.
Van den Berg, M., Birnbaum, L.S., Denison, M., De Vito, M., Farland, W., Feeley, M.,
Fiedler, H., Hakansson, H., Hanberg, A. & Haws, L. 2006. The 2005 World Health
Organization reevaluation of human and mammalian toxic equivalency factors for dioxins
and dioxin-like compounds. Toxicol Sci, 93(2): 223-241.
Van Den Heuvel, R.L., Koppen, G., Staessen, J.A., Hond, E.D., Verheyen, G., Nawrot,
T.S., Roels, H.A., Vlietinck, R. & Schoeters, G.E. 2002. Immunologic biomarkers in relation
to exposure markers of PCBs and dioxins in Flemish adolescents (Belgium). Environ Health
Perspect, 110(6): 595-600.
Van Wijngaarden, E., Thurston, S.W., Myers, G.J., Strain, J.J., Weiss, B., Zarcone, T.,
Watson, G.E. et al. 2013a. Prenatal methyl mercury exposure in relation to neurodevelopment
and behavior at 19 years of age in the Seychelles Child Development Study. Neurotoxicol
Teratol, 39: 19-25. https://doi.org/10.1016/j.ntt.2013.06.003
Van Wijngaarden, E., Thurston, S., Myers, G., Strain, J., Weiss, B., Zarcone, T., Watson,
G., Zareba, G., McSorley, E. & Mulhern, M.S. 2013b. Prenatal methyl mercury exposure
in relation to neurodevelopment and behavior at 19 years of age in the Seychelles Child
Development Study. Neurotoxicol Teratol, 39: 19-25.
Vedrine, M.L., Hanlon, J., Bevan, R., Floyd, P., Brown, T. & Matthies, F. 2018. Extensive
literature search, selection for relevance and data extraction of studies related to the toxicity
of PCDD/Fs and DL‐PCB s in humans. EFSA Supporting Publications, 15(11): 1136E.
Venter, C., Agostoni, C., Arshad, S.H., Ben-Abdallah, M., Du Toit, G., Fleischer,
D.M., Greenhawt, M. et al. 2020. Dietary factors during pregnancy and atopic outcomes
in childhood: A systematic review from the European Academy of Allergy and Clinical
Immunology. Pediatr Allergy Immunol, 31(8): 889-912. https://doi.org/10.1111/pai.13303
Venø, S.K., Bork, C.S., Jakobsen, M.U., Lundbye-Christensen, S., Bach, F.W., McLennan,
P.L., Tjønneland, A., Schmidt, E.B. & Overvad, K. 2018. Substitution of fish for red meat
or poultry and risk of ischemic stroke. Nutrients, 10(11): 1648.
Virtanen, J.K., Voutilainen, S., Rissanen, T.H., Mursu, J., Tuomainen, T.P., Korhonen,
M.J., Valkonen, V.P., Seppänen, K., Laukkanen, J.A. & Salonen, J.T. 2005. Mercury, fish
oils, and risk of acute coronary events and cardiovascular disease, coronary heart disease, and
all-cause mortality in men in eastern Finland. Arterioscler Thromb Vasc Biol, 25(1): 228-233.

322
REFERENCES

VKM (Norwegian Scientific Committee for Food and Environment). 2022. Benefit and
risk assessment of fish in the Norwegian diet. Scientific Opinion of the Steering Committee
of the Norwegian Scientific Committee for Food and Environment. https://vkm.no/
risikovurderinger/allevurderinger/fiskinorskkostholdnytteogrisikovurdering.4.413ea9241
6707dc43759fba3.html
VKM. 2020. Protocol and description of literature search for the risk-benefit assessment of fish
in the Norwegian diet. From the Steering Committee of the Norwegian Scientific Committee
for Food and Environment. ISBN: 978-82-8259-335-9. Oslo, Norway. https://vkm.no/do
wnload/18.1f82914172cd85756e85e21/1593150372748/Protocol%20for%20the%20risk-
benefit%20assessment%20of%20fish%20in%20the%20Norwegian%20diet%20final.pdf
VKM. 2014. Benefit-risk assessment of fish and fish products in the Norwegian diet - an
update. Scientific Opinion of the Scientific Steering Committee. VKM Report 15 [293 pp].
https://vkm.no/download/18.2994e95b15cc54507161ea1a/1498222018046/0a646edc5e.pdf
Vupputuri, S., Longnecker, M.P., Daniels, J.L., Guo, X. & Sandler, D.P. 2005. Blood
mercury level and blood pressure among US women: results from the National Health and
Nutrition Examination Survey 1999–2000. Environ Res, 97(2): 195-200.
Walda, I.C., Tabak, C., Smit, H.A., Rasanen, L., Fidanza, F., Menotti, A., Nissinen, A.,
Feskens, E.J.M. & Kromhout, D. 2002. Diet and 20-year chronic obstructive pulmonary
disease mortality in middle-aged men from three European countries. Eur J Clin Nutr, 56(7):
638-643. https://doi.org/10.1038/sj.ejcn.1601370
Wallin, A., Di Giuseppe, D., Orsini, N., Patel, P.S., Forouhi, N.G. & Wolk, A. 2012. Fish
consumption, dietary long-chain n-3 fatty acids, and risk of type 2 diabetes: systematic
review and meta-analysis of prospective studies. Diabetes Care, 35(4): 918-929.
Wang, Y., Chen, A., Dietrich, K.N., Radcliffe, J., Caldwell, K.L. & Rogan, W.J. 2014.
Postnatal exposure to methyl mercury and neuropsychological development in 7-year-old
urban inner-city children exposed to lead in the United States. Child Neuropsychol, 20(5):
527-538.
Warner, M., Rauch, S., Ames, J., Mocarelli, P., Brambilla, P., Signorini, S. & Eskenazi, B.
2020a. Prenatal dioxin exposure and thyroid hormone levels in the Seveso second generation
study. Environ Res, 183: 109280.
Warner, M., Rauch, S., Brambilla, P., Signorini, S., Mocarelli, P. & Eskenazi, B. 2020b.
Prenatal dioxin exposure and glucose metabolism in the Seveso Second Generation study.
Environ Int, 134: 105286.
Warner, M., Rauch, S., Ames, J., Mocarelli, P., Brambilla, P., Signorini, S. & Eskenazi, B.
2019. In utero dioxin exposure and cardiometabolic risk in the Seveso Second Generation
Study. Int J Obes, 43(11): 2233-2243.
Watanabe, C., Yin, K., Kasanuma, Y. & Satoh, H. 1999. In utero exposure to methylmercury
and Se deficiency converge on the neurobehavioral outcome in mice. Neurotoxicol Teratol,
21(1): 83-8. https://doi.org/10.1016/s0892-0362(98)00036-1
WCRF (World Cancer Research Fund)/American Institute for Cancer Research. 2018.
Diet, Nutrition, Physical Activity and Cancer: a Global Perspective. Continuous Update
Project Expert Report 2018. http://www.dietandcancerreport.org
WCRF. 2018a. Judging the evidence. Judging the evidence. Continuous Update Project Expert
Report 2018. Available at dietandcancerreport.org and https://www.wcrf.org/wp-content/
uploads/2021/02/judging-the-evidence.pdf.

323
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Weisglas-Kuperus, N., Patandin, S., Berbers, G., Sas, T., Mulder, P., Sauer, P. & Hooijkaas,
H. 2000. Immunologic effects of background exposure to polychlorinated biphenyls and
dioxins in Dutch preschool children. Environ Health Perspect, 108(12): 1203-1207.
Wells, E.M., Herbstman, J.B., Lin, Y.H., Jarrett, J., Verdon, C.P., Ward, C., Caldwell,
K.L. et al. 2016. Cord Blood Methylmercury and Fetal Growth Outcomes in Baltimore
Newborns: Potential Confounding and Effect Modification by Omega-3 Fatty Acids,
Selenium, and Sex. Environ Health Perspect, 124(3): 373-9. https://doi.org/10.1289/
ehp.1408596
Wennberg, M., Bergdahl, I.A., Hallmans, G., Norberg, M., Lundh, T., Skerfving, S.,
Strömberg, U., Vessby, B. & Jansson, J.H. 2011. Fish consumption and myocardial
infarction: a second prospective biomarker study from northern Sweden. Am J Clin Nutr,
93(1): 27-36. https://doi.org/10.3945/ajcn.2010.29408
Wennberg, M., Bergdahl, I.A., Stegmayr, B., Hallmans, G., Lundh, T., Skerfving, S.,
Strömberg, U., Vessby, B. & Jansson, J.H. 2007. Fish intake, mercury, long-chain n-3
polyunsaturated fatty acids and risk of stroke in northern Sweden. Br J Nutr, 98(5): 1038-
1045.
WHO. 2022. WHO expert consultation on updating the 2005 toxic equivalency factors for
dioxin like compounds, including some polychlorinated biphenyls. https://www.who.int/
news/item/15-11-2022-who-expert-consultation-on-updating-the-2005-toxic-equivalency-
factors-for-dioxin-like-compounds-including-some-polychlorinated-biphenyls
WHO. 2016. Dioxins and their effects on human health. [Cited 27 October 2022]. https://
www.who.int/news-room/fact-sheets/detail/dioxins-and-their-effects-on-human-health
Wu, J.H., Micha, R., Imamura, F., Pan, A., Biggs, M.L., Ajaz, O., Djousse, L., Hu, F.B.
& Mozaffarian, D. 2012. Omega-3 fatty acids and incident type 2 diabetes: a systematic
review and meta-analysis. Br J Nutr, 107(S2): S214-S227.
Xu, P., Liu, A., Li, F., Tinkov, A.A., Liu, L. & Zhou, J.C. 2021. Associations between
metabolic syndrome and four heavy metals: a systematic review and meta-analysis. Environ
Pollut, 273: 116480.
Xue, F., Holzman, C., Rahbar, M.H., Trosko, K. & Fischer, L. 2007. Maternal fish
consumption, mercury levels, and risk of preterm delivery. Environ Health Perspect, 115(1):
42-47.
Xun, P. & He, K. 2012. Fish consumption and incidence of diabetes: meta-analysis of data
from 438,000 individuals in 12 independent prospective cohorts with an average 11-year
follow-up. Diabetes Care, 35(4): 930-938.
Yaginuma-Sakurai, K., Murata, K., Shimada, M., Nakai, K., Kurokawa, N., Kameo, S. &
Satoh, H. 2010. Intervention study on cardiac autonomic nervous effects of methylmercury
from seafood. Neurotoxicol Teratol, 32(2): 240-245.
Yamagishi, K., Iso, H., Shimazu, T., Tamakoshi, A., Sawada, N., Matsuo, K., Ito, H.
et al. 2019. Fish intake and risk of mortality due to aortic dissection and aneurysm: A
pooled analysis of the Japan cohort consortium. Clin Nutr, 38(4): 1678-1683. https://doi.
org/10.1016/j.clnu.2018.08.007
Yang, Y., Kim, Y. & Je, Y. 2018. Fish consumption and risk of depression: Epidemiological
evidence from prospective studies. Asia‐Pacific Psychiatry, 10(4): e12335.
Yorifuji, T., Murata, K., Bjerve, K.S., Choi, A.L., Weihe, P. & Grandjean, P. 2013. Visual
evoked potentials in children prenatally exposed to methylmercury. Neurotoxicology,
37: 15-18.

324
REFERENCES

Yoshimasu, K., Kiyohara, C., Takemura, S. & Nakai, K. 2014. A meta-analysis of the
evidence on the impact of prenatal and early infancy exposures to mercury on autism and
attention deficit/hyperactivity disorder in the childhood. Neurotoxicology, 44: 121-131.
Yoshizawa, K., Rimm, E.B., Morris, J.S., Spate, V.L., Hsieh, C., Spiegelman, D., Stampfer,
M.J. & Willett, W.C. 2002. Mercury and the risk of coronary heart disease in men. N Engl
J Med, 347(22): 1755-1760.
Young, E.C., Davidson, P.W., Wilding, G., Myers, G.J., Shamlaye, C., Cox, C., de Broeck,
J., Bennett, C.M. & Reeves, J.S. 2020. Association between prenatal dietary methyl mercury
exposure and developmental outcomes on acquisition of articulatory-phonologic skills in
children in the Republic of Seychelles. Neurotoxicology, 81: 353-357.
Zamani, S.A., McClain, K.M., Graubard, B.I., Liao, L.M., Abnet, C.C., Cook, M.B. &
Petrick, J.L. 2020. Dietary Polyunsaturated Fat Intake in Relation to Head and Neck,
Esophageal, and Gastric Cancer Incidence in the National Institutes of Health-AARP Diet
and Health Study. Am J Epidemiol, 189(10): 1096-1113. https://doi.org/10.1093/aje/kwaa024
Zeng, L.F., Cao, Y., Liang, W.X., Bao, W.H., Pan, J.K., Wang, Q., Liu, J., Liang, H.D., Xie,
H. & Chai, Y.T. 2017. An exploration of the role of a fish-oriented diet in cognitive decline:
a systematic review of the literature. Oncotarget, 8(24): 39877.
Zhang, B., Xiong, K., Cai, J. & Ma, A. 2020. Fish consumption and coronary heart disease:
a meta-analysis. Nutrients, 12(8): 2278.
Zhang, G.Q., Liu, B., Li, J., Luo, C.Q., Zhang, Q., Chen, J.L., Sinha, A. & Li, Z.Y. 2017.
Fish intake during pregnancy or infancy and allergic outcomes in children: A systematic
review and meta‐analysis. Pediatr Allergy Immunol, 28(2): 152-161.
Zhang, H., Zeng, Y., Yang, H., Hu, Y., Hu, Y., Chen, W., Ying, Z., Sun, Y., Qu, Y. & Li,
Q. 2021a. Familial factors, diet, and risk of cardiovascular disease: a cohort analysis of the
UK Biobank. Am J Clin Nutr, 114(5): 1837-1846.
Zhang, J., Li, X., Shen, L., Khan, N.U., Zhang, X., Chen, L., Zhao, H. & Luo, P. 2021b.
Trace elements in children with autism spectrum disorder: a meta-analysis based on case-
control studies. J Trace Elem Med Biol, 67: 126782.
Zhang, M., Picard-Deland, E. & Marette, A. 2013. Fish and marine omega-3 polyunsatured
fatty acid consumption and incidence of type 2 diabetes: a systematic review and meta-
analysis. Int J Endocrinol, 2013: 501015. https://doi.org/10.1155/2013/501015
Zhang, Y., Chen, J., Qiu, J., Li, Y., Wang, J. & Jiao, J. 2015. Intakes of fish and polyunsaturated
fatty acids and mild-to-severe cognitive impairment risks: a dose-response meta-analysis of
21 cohort studies. Am J Clin Nutr, 103(2): 330-340.
Zhao, R., Gao, Q., Xiong, T., Zhou, J., Wang, S., Zhang, Z., Du, Y. et al. 2022. Moderate
Freshwater Fish Intake, but Not n-3 Polyunsaturated Fatty Acids, Is Associated with a
Reduced Risk of Small for Gestational Age in a Prospective Cohort of Chinese Pregnant
Women. J Acad Nutr Diet, 122(4): 722-730 e12. https://doi.org/10.1016/j.jand.2021.10.016
Zhao, R., Gao, Q., Wang, S., Yang, X. & Hao, L. 2021. The effect of maternal seafood
consumption on perinatal outcomes: a systematic review and dose-response meta-analysis.
Crit Rev Food Sci Nutr, 61(21): 3504-3517.
Zhao, W., Tang, H., Yang, X., Luo, X., Wang, X., Shao, C. & He, J. 2019. Fish consumption
and stroke risk: a meta-analysis of prospective cohort studies. J Stroke Cerebrovasc Dis,
28(3): 604-611.

325
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Zheng, J.S., Huang, T., Yang, J., Fu, Y.Q. & Li, D. 2012. Marine N-3 polyunsaturated
fatty acids are inversely associated with risk of type 2 diabetes in Asians: a systematic review
and meta-analysis. PLoS One, 7(9): e44525. https://doi.org/10.1371/journal.pone.0044525
Zhong, V.W., Allen, N.B., Greenland, P., Carnethon, M.R., Ning, H.Y., Wilkins, J.T.,
Lloyd-Jones, D.M. & Van Horn, L. 2021. Protein foods from animal sources, incident
cardiovascular disease and all-cause mortality: a substitution analysis. Int J Epidemiol, 50(1):
223-233. https://doi.org/10.1093/ije/dyaa205
Zhou, Y., Tian, C. & Jia, C. 2012. Association of fish and n-3 fatty acid intake with the risk
of type 2 diabetes: a meta-analysis of prospective studies. Br J Nutr, 108(3): 408-17. https://
doi.org/10.1017/s0007114512002036
Zidane, M., Ren, Y., Xhaard, C., Leufroy, A., Côte, S., Dewailly, E., Noël, L., Guérin, T.,
Bouisset, P. & Bernagout, S. 2019. Non-essential trace elements dietary exposure in French
Polynesia: intake assessment, nail bio monitoring and thyroid cancer risk. Asian Pac J Cancer
Prev, 20(2): 355. https://doi.org/10.31557%2FAPJCP.2019.20.2.355

References for all included articles for the “Occurrence data for MeHg, dioxins and dl-
PCBs” are given in Appendix 7, Table A7.3.

326
327
 328
© FAO/Giulio Napolitano
APPENDIX 1
TERMS OF REFERENCE
PROVIDED BY FAO/WHO

DEFINITION OF OUTCOME AND OUTPUTS

OUTCOME
Scientific evidence provided about risks and benefits of fish consumption for an
update of the Report of the Joint FAO/WHO Expert Consultation on the Risks
and Benefits of Fish Consumption.
Outputs
1. Background document including a literature review.
2. Draft update of the Report of the Joint FAO/WHO Expert Consultation on
the Risks and Benefits of Fish Consumption.

DESCRIPTION OF SERVICES
The Service Provider will:
1. Develop an outline for the background document (including the literature
review) to be discussed with FAO.
2. Carry out a systematic literature review that will include:
c. Evidence of health benefits from fish consumption other than from long-
chain n-3 polyunsaturated fatty acids (LCn3PUFAs).
d. New data (published in the last 10 years) on toxic effects of dioxins
(defined here to include polychlorinated dibenzo-p-dioxins [PCDDs],
polychlorinated dibenzofurans [PCDFs] and dioxin-like polychlorinated
biphenyls [dl-PCBs]) for all population groups.
e. New data (published in the last 10 years) on toxic effects of methylmercury
(MeHg) for all population groups.

329
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

f. The role of selenium with regard to the health effects of MeHg.

g. Occurrence data for MeHg and dioxins in fisheries and aquaculture products
published in the last 10 years.
Develop a background document including the above-mentioned literature review
and other relevant information that could add value to the background document.

330
APPENDICES

APPENDIX 2
QUALITY ASSESSMENT TOOLS
(RISK OF BIAS)

TABLE A2.1 QUESTIONS INCLUDED IN AMSTAR 2 – A CRITICAL APPRAISAL TOOL FOR SYSTEMATIC REVIEWS
1. Did the research questions and inclusion criteria for the review include the components of PICO?
For Yes: Optional (recommended) † Yes
† Population † Timeframe for follow-up † No
† Intrvention
† Comparator group
† Outcome

2. Did the report of the review contain an explicit statement that the review methods were established prior to the conduct of the review and did the report justify any
significant deviations from the protocol?
For Partial Yes: For Yes: † Yes
The authors state that they had a written protocol or guide that As for partial yes, plus the protocol should be registered and should † Partial Yes
included ALL the following: also have specified: † No
† review question(s) † a meta-analysis/synthesis plan, if appropriate, and
† a search strategy † a plan for investigating causes of heterogeneity
† inclusion/exclusion criteria † justification for any deviations from the protocol
† a risk of bias assessment
3. Did the review authors explain their selection of the study designs for inclusion in the review?
For Yes, the review should satisfy ONE of the following: † Yes
† Explanation for including only RCTs † No
† OR Explanation for including only NRSI
† OR Explanation for including both RCTs and NRSI
4. Did the review authors use a comprehensive literature search strategy?
For Partial Yes (all the following): For Yes, should also have (all the following): † Yes
† searched at least 2 databases (relevant to research question) † searched the reference lists / bibliographies of included studies † Partial Yes
† provided key word and/or search strategy † searched trial/study registries † No
† justified publication restrictions (e.g. language) † included/consulted content experts in the field
† where relevant, searched for grey literature
† conducted search within 24 months of completion of the review
5. Did the review authors perform study selection in duplicate?
For Yes, either ONE of the following: † Yes
† at least two reviewers independently agreed on selection of eligible studies and achieved consensus on which studies to include † No
† OR two reviewers selected a sample of eligible studies and achieved good agreement (at least 80 percent), with the remainder selected by
one reviewer.
6. Did the review authors perform data extraction in duplicate?
For Yes, either ONE of the following: † Yes
† at least two reviewers achieved consensus on which data to extract from included studies † No
† OR two reviewers extracted data from a sample of eligible studies and achieved good agreement (at least 80 percent), with the remainder
extracted by one reviewer.
7. Did the review authors provide a list of excluded studies and justify the exclusions?
For Partial Yes: For Yes, must also have: † Yes
† provided a list of all potentially relevant studies that were read † Justified the exclusion from the review of each potentially relevant † Partial Yes
† in full-text form but excluded from the review † No

331
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A2.1 QUESTIONS INCLUDED IN AMSTAR 2 – A CRITICAL APPRAISAL TOOL FOR SYSTEMATIC REVIEWS (cont.)
8. Did the review authors describe the included studies in adequate detail?
For Partial Yes (ALL the following): For Yes, should also have ALL the following: † Yes
† described populations † described population in detail † Partial Yes
† described interventions † described intervention in detail (including doses where relevant) † No
† described comparators † described comparator in detail (including doses where relevant)
† described outcomes † described study’s setting
† described research designs † timeframe for follow-up
9. Did the review authors use a satisfactory technique for assessing the risk of bias (RoB) in individual studies that were included in the review?
RCTs For Yes, must also have assessed RoB from: † Yes
For Partial Yes, must have assessed RoB from † allocation sequence that was not truly random, and † Partial Yes
† unconcealed allocation, and † selection of the reported result from among multiple † No
† lack of blinding of patients and assessors when assessing measurements or analyses of a specified outcome † Includes only NRSI
outcomes (unnecessary for objective outcomes such as all-cause
mortality)
NRSI For Yes, must also have assessed RoB: † Yes
For Partial Yes, must have assessed RoB: † methods used to ascertain exposures and outcomes, and † Partial Yes
† from confounding, and † selection of the reported result from among multiple † No
† from selection bias measurements or analyses of a specified outcome † Includes only RCTs
10. Did the review authors report on the sources of funding for the studies included in the review?
For Yes † Yes
† Must have reported on the sources of funding for individual studies included in the review. Note: Reporting that the reviewers looked for this † No
information but it was not reported by study authors also qualifies
11. If meta-analysis was performed did the review authors use appropriate methods for statistical combination of results?
RCTs † Yes
For Yes: † No
† The authors justified combining the data in a meta-analysis † No meta-analysis conducted
† AND they used an appropriate weighted technique to combine study results and adjusted for heterogeneity if present.
† AND investigated the causes of any heterogeneity
For NRSI † Yes
For Yes: † No
† The authors justified combining the data in a meta-analysis † No meta-analysis conducted
† AND they used an appropriate weighted technique to combine study results, adjusting for heterogeneity if present
† AND they statistically combined effect estimates from NRSI that were adjusted for confounding, rather than combining raw data, or justified
combining raw data when adjusted effect estimates were not available
† AND they reported separate summary estimates for RCTs and NRSI separately when both were included in the review
12. If meta-analysis was performed, did the review authors assess the potential impact of RoB in individual studies on the results of the meta-analysis or other evidence
synthesis?
For Yes: † Yes
† included only low risk of bias RCTs † No
† OR, if the pooled estimate was based on RCTs and/or NRSI at variable RoB, the authors performed analyses to investigate possible impact † No meta-analysis conducted
of RoB on summary estimates of effect.
13. Did the review authors account for RoB in individual studies when interpreting/ discussing the results of the review?
For Yes: † Yes
† included only low risk of bias RCTs † No
† OR, if RCTs with moderate or high RoB, or NRSI were included the review provided a discussion of the likely impact of RoB on the results
14. Did the review authors provide a satisfactory explanation for, and discussion of, any heterogeneity observed in the results of the review?
For Yes: † Yes
† There was no significant heterogeneity in the results † No
† OR if heterogeneity was present the authors performed an investigation of sources of any heterogeneity in the results and discussed the
impact of this on the results of the review
15. If they performed quantitative synthesis did the review authors carry out an adequate investigation of publication bias (small study bias) and discuss its likely impact on
the results of the review?
For Yes: † Yes
† performed graphical or statistical tests for publication bias and discussed the likelihood and magnitude of impact of publication bias † No
† No meta-analysis conducted

Source: Shea B.J. Reeves B.C. Wells G. Thuku M. Hamel C. Moran J. Moher D. et al. 2017. AMSTAR 2: a critical appraisal tool for systematic
reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ.
https://doi.org/10.1136/bmj.j4008

332
APPENDICES

TABLE A2.2 QUESTIONS INCLUDED IN THE RISK-OF-BIAS TOOL FOR THE QUALITY ASSESSMENT OF RANDOMIZED CONTROLLED
TRIALS AND TRIALS IN THE REVIEW “EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”
Can't If "no" on this question, the study
Quality assessment of RCTs and trials Yes No NA Requires yes for level
tell is graded C and further excluded
Questions A B C
1. General questions and study design
1a) Research question/hypothesis clearly formulated? ✖ ✖ ✖
1b) Was the study design suited to test the research question? ✖ ✖ ✖
1c) Was the duration of the study suited to test the research hypothesis? ✖
2. Participation and compliance
2a) Population (target group) well described? ✖ ✖ ✖
2b) Sample (possible participants) recruited in an acceptable way? ✖
2c) Criteria for inclusion/exclusion clearly formulated and acceptable? ✖
2d) Actual participants comparable with the relevant (target) population? ✖
2e) Method of randomization allocation stated and appropriate? ✖
2f) Was there an account for the comparability of intervention and control
groups with regard to relevant/possible factors that might affect ✖
outcome?
2g) Compliance reported in an acceptable way, and compliance acceptable? ✖
2h) Drop-out rate within an acceptable range? 6mo<30%, 12mo<40%,
24mo<50% ✖
2i) The drop-outs did not differ between the groups? ✖
3. Dietary interventions and assessment
3a) Intervention diets clearly defined and characterized (fish intake)? ✖ ✖ ✖
3b) Method used for dietary assessment valid/adequately validated? ✖
3c) Intervention diets consist of normal foods/relevance to research question? ✖
3d) Measurement errors in dietary reporting considered? ✖
3e) Energy adjustment adequately done? ✖
4. Anthropometry
4a) Assessment details clearly reported, and assessment adequately
performed? ✖
5. Outcome, results, and analyses
5a) Acceptable and clear definition of the outcome/endpoint? ✖ ✖ ✖
5b) Results analysed blind? ✖
5c) Attempts in the analysis phase made to adjust for imbalances between
treatment arms with regard to important determinants for the outcome ✖
(e.g. through multivariate modelling)?
5d) Possible use of medication/supplements taken into account? ✖
6. Statistical power
6a) Sample size and power calculation reported/considered (relevant for main
outcome variable)? ✖
Summary of the study quality (A, B or C):
Reasons for grade C and other comments:

333
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A2.3 QUESTIONS INCLUDED IN THE RISK-OF-BIAS TOOL FOR THE QUALITY ASSESSMENT OF PROSPECTIVE COHORT
STUDIES IN THE REVIEW “EVIDENCE OF HEALTH BENEFITS FROM FISH CONSUMPTION”
Can't If "no" on this question, the study
Quality assessment of prospective cohort studies Yes No NA Requires yes for level
tell is graded C and further excluded
Questions A B C
1 General questions and study design
1a) Research question clearly formulated? ✖ ✖ ✖
1b) Endpoint/outcome clearly formulated? ✖ ✖ ✖
1c) Was the study design suited to test the research hypothesis? ✖ ✖ ✖
2 Sampling (ascertainment of cases and non-cases)
2a) Source population/study base well defined? ✖ ✖ ✖
2b) Response rate reported and acceptable? ✖ ✖ ✖
2c) Time period of baseline examinations clearly identified? ✖
2d) Endpoint clearly ascertained and assessed in a valid way? ✖ ✖ ✖
2e) Follow-up period clearly identified? ✖
3 Dietary exposure
3a) Fish or seafood intake according to inclusion criteria (individual intake
and at least frequency)? ✖ ✖ ✖
3b) Was the dietary assessment method validated? ✖
3c) Measurement errors in dietary reporting considered or mentioned? ✖
3d) Energy adjustment adequately done? ✖
3e) Repeated assessment of diet during follow up, and data considered? ✖
4 Anthropometry
4a) Assessment details clearly reported, and assessment adequately
performed? ✖
5 Confounding
5a) Were important confounders identified/ascertained and considered by
authors? ✖ ✖ ✖
6 Statistical power
6a) Was the study power considered and power calculations and sample size
reported? ✖
6b) In view of multiple tests, were by chance findings considered? ✖ ✖
6c) Sufficient size of study population and no. of outcomes/cases? ✖
7 Statistical analysis
7a) Appropriately handled? ✖
7b) Relevant confounders adequately handled (e.g., restriction, stratified
analyses, multivariate modelling, interaction tested)? ✖ ✖ ✖
7c) Ascertainment/detection bias considered (e.g. cases detected due to
screening)? ✖
7d) Cases detected early during the follow-up period removed? ✖
Summary of the study quality (A, B or C):
Reasons for grade C and other comments:

334
APPENDICES

TABLE A2.4 QUESTIONS INCLUDED IN THE OHAT RISK OF BIAS RATING TOOL FOR HUMAN AND ANIMAL STUDIES OF THE OFFICE
OF HEALTH ASSESSMENT AND TRANSLATION
Q Bias domain Questions Applies to study designs
Q1 Selection bias Was administered dose or exposure level adequately randomized? EA, HCT
Q2 Selection bias Was allocation to study groups adequately concealed? EA, HCT
Q3 Selection bias Did selection of study participants result in appropriate comparison groups? Co, CaCo, CrSe
Q4 (Key)* Confounding Did the study design or analysis account for important confounding and Co, CaCo, CrSe, CaS
modifying variables?
Q5 Performance bias Were experimental conditions identical across study groups? EA
Q6 Performance bias Were the research personnel and human subjects blinded to the study group EA, HCT
during the study?
Q7 Attrition/exclusion bias Were outcome data complete without attrition or exclusion from analysis? EA, HCT, Co, CaCo CrSe
Q8 (Key)* Detection bias Can we be confident in the exposure characterization? EA, HCT, Co, CaCo CrSe, CaS
Q9 (Key)* Detection bias Can we be confident in the outcome assessment? EA, HCT, Co, CaCo CrSe, CaS
Q10 Selective reporting bias Were all measured outcomes reported? EA, HCT, Co, CaCo CrSe, CaS
Q11 Other sources of bias Were there no other potential threats to internal validity (e.g. statistical EA, HCT, Co, CaCo CrSe, CaS
methods were appropriate, and researchers adhered to the study protocol)?

Notes: *Questions are considered key questions when assessing the risk of bias for human observational studies (Co, CaCo, CrSe).
Abbreviations: EA: experimental animal; HCT: human controlled trials; Co: cohort; CaCo: case-control; CrSe: cross-sectional; CaS: case series
Source: OHAT (Office of Health Assessment and Translation). 2015. OHAT Risk of Bias Tool for human and Animal Studies. Division of the National
Toxicology Program. National Institute of Environmental Health Sciences.

335
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

APPENDIX 3
EVIDENCE OF HEALTH
BENEFITS FROM FISH
CONSUMPTION

LITERATURE SEARCH STRATEGY

TABLE A3.1 LITERATURE SEARCH STRATEGY FOR THE SYSTEMATIC REVIEW “EVIDENCE OF HEALTH BENEFITS
FROM FISH CONSUMPTION”
Databases: Web of Science, PubMed, Cochrane
Date of literature searches: Week 47, 2021
# Search group Literature search string
#1 Fish (Fish* or finfish* or crayfish* or crawfish* or cuttlefish* or inkfish* or milkfish* or catfish* or Pisces or
seafood* or “sea food*” or sea-food* or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or
anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin* or mackerel* or pilchard* or crab*
or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod
or “Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad*
or Crassostrea or tagelus or clams or cockle* or urchin* or echinoderm* or echinoid* or “sea cucumber*” or
holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark* or skate* or crustacea*
or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* or whelk* or gastropod* or abalone*
or snail* or limpet* or conch* or periwinkle* or nautilus*)
#2 Study design ("systematic review*" or "systematic report*" or "cochrane review*" or "umbrella review*" or meta-analysis
or "meta analysis" or “meta analyses” or RCT* or "randomized controlled" or prospective cohort or prospective
observational or cohort or longitudinal or controlled trial*)
#3 Dietary (eat* or ate or intake* or consumption or consume* or consuming or ingestion or serving* or meal* or diet* or
consumption dine or dinner* or lunch* or breakfast* or snack*)
Search string Hits Web of Science Hits PubMed Hits Cochrane Total hits
#4 #1 #2 #3 11 719 13 583 46 25 348
# Search group Literature search string
#5 Overweight and (obesity or obesities or obese or adiposity or fatness or overweight or “over weight” or (excess* adj (fat or
obesity weight)) or "BMI" or (body adj (height or size or weight or mass)) or (abdominal adj (diameter index or height))
or "height weight ratio*" or "waist circumference*" or "waist height ratio*" or "waist to height ratio*" or (weight
adj (change* or gain* or loss)))
#6 Cardiovascular ((Cardiovascular or heart or cardiac or myocardial or myo cardial or cerebrovascular or vascular or
diseases and coronary or cerebral or peripheral or endothelial) adj (disease* or disorder* or failure or event* or health or
outcomes effect* or accident* or calcification* or “risk factor*” or riskfactor* or syndrom* or revascularization* or
revascularisation* or arter* or function* or dysfunction* or attack* or arrest or apoplex* or insufficienc*
or injur* or insult* or scleros* or stenos* or restenos*)) or cardioprotect* or "cardio protect*" or "high
cardiovascular risk*" or CVD or infarct* or reinfarction* or aneurysm* or angina or artherosclero* or "arthero
sclero*" or arteriosclero* or "arterio sclero*" or ischemi* or ischaemi* or nonischemi* nonischaemi* or
"non ischemi*" or “non ischaemi*” or thrombos* or thrombolism* or tachycardia* or tachyarrhythmia* or
arrhythmia* or ((ventricular or arterial) adj (fibrillation* or compliance* or stiffness*)) or "sudden cardiac
death*" or stroke* or TIA or (brain adj (hemorrhage* or haemorrhage* or accident* or attack* or infarct* or
insult*))

336
APPENDICES

TABLE A3.1 LITERATURE SEARCH STRATEGY FOR THE SYSTEMATIC REVIEW “EVIDENCE OF HEALTH BENEFITS
FROM FISH CONSUMPTION” (cont.)
Databases: Web of Science, PubMed, Cochrane
Date of literature searches: Week 47, 2021
#7 Type 2 diabetes (diabetes or sugar sickness or hypoglycemia or hypo glycemia or hyperglycemia or hyper glycemia or insulin
resistance)
#8 Birth and growth (growth or ((premature or “pre term” or preterm) adj birth*) or SGA or ((birth or gestational or neonatal or neo
outcomes natal or newborn or "new born" or foetal or fetal or foetus or fetus or baby or babies or infan*) adj (weight or
size*)) or ((pregnancy or birth or obstetric) adj outcome*))
#9 Allergy (("allerg*" or "hypersensitivit*" or "hyper sensitivit*" or "sensiti#ation*" or "atopic?" or "atopy" or "atopies"))
#10 Neurodevelopment (((Child* or infant* or fetal or foetal or prenatal or "pre natal" or postnatal or "post natal" or human or
and cognitive "antepartum period*" or "ante partum period*") adj3 development*) or inhibition or (brain adj2 (damage* or
diseases injur* or development* or disorder*)) or psychomotor* or "psycho motor*" or motor or sensorimotor or "sensori
motor" or sensorymotor or "sensory motor" or cognition or "cognitive function*" or "Mental health" or “Disorder*
of higher cerebral function*" or (psychological adj ("well being" or wellbeing)) or ((neurocognit* or "neuro
cognit*" or neurological or "nervous system" or nervoussystem or cognitive or development* or mental) adj2
(dysfunction* or function* or decline* or deterioration* or Defici* or illness* or retardation* or disturbance*
or impairment* or disorder* or impact* or disabilit* or deviation* or development*)) or neurodevelopment* or
"neuro development*" or autis* or Asperger* or kanner* or ASD or "attention deficit" or hyperactiv* or ADDH or
ADHD or AD/HD or ADD or "minimal brain dysfunction" or impulsiveness or dyslexia or dyslexic* or dyscalculia or
dyscalculic* or attention or learning or reading or mathematic* or math* or (aptitude adj1 test*) or ((Education
or Educational or academic or school) adj1 (Status or attainment* or achievement* or performance* or
underachievement* or "under achievement*" or score* or success* or failure*)) or "executive function*" or
"information processing" or "school readiness" or "school ready" or Emotion* or socioemotional or "social
emotional" or socioemotional or "socio emotional" or (social adj (development* or behaviour* or adjustment*))
or (intellectual adj2 (development* or deficien* or disorder* or retardation* or disabilit* or disturbance*
or impairment*)) or Communication or language* or literacy or literacies or IQ or intelligence or “Speech
disorder*" or mutism* or aphasia or stutter* or dysphasia or alexia or anxiet* or depression* or depressive or
"mood disorder*” or schizophrenia or schizophrenic or aggression or behaviour* or affect or anger or bipolar or
Temperament* or personalit* or amnesia or dementia or Alzheimer* or Parkinson* or huntington* or (memory
adj3 (disorder* or impairment* or disturbance* or deficianc* or disabilit* or "short term" or shortterm or “long
term" or longterm or "verbal recognition*")))
#11 Dental health ((("dental" or "tooth" or "teeth" or "enamel") and ("enamel" or "discolo?ration?" or "malformation?" or
"opacit*")) or "hypo?minerali#ation" or ("developmental" and ("dental" or "teeth" or "tooth" or "enamel") and
"defect?"))
#12 Immunology (("immunolog*" or "infection resistance" or immunity or autoimmunity or "auto immunity" or immunodeficienc*
or "immuno deficienc*" or (immun* adj (system or status or defense* or defence* or deficienc*)) or "vaccination
response*" or ((upper or lower) adj "respiratory tract infection*") or "respiratory Sound*" or wheez* or asthma*
or psoriasis or eczema* or dermatiti* or rheumatoid arthritis or (((Sjogren* or sicca) adj syndrome*) or
syndrome*) or Antinuclear antibod* or "Multiple scleros*" or "Systemic lupus erythematosus" or ((Scleroderma
or scleros*) adj (localized or systemic))))
#13 Cancer (cyst* or neoplasm* or neurofibroma* or tumor* or tumour* or cancer* or malign*)
#14 Mortality (mortalit* or "death rate*" or deathrate* or death*)
#15 Bone health ((Osteoporos* or Osteopenia or Rickets or Osteomalacia or "vitamin D deficienc*" or (bone adj2 (diseas* or
density or (low adj fractur*) or fragil* or broken or demineralization* or demineralisation* decalcification*)) or
"Accidental Fall*" or ((Slip* or trip*) adj2 fall*)))
# Search group Search string Hits Web of Science Hits PubMed Hits Cochrane Total hits
#16 Overweight and #4 #5 3 498 1 584 13 5 095
obesity
#17 Cardiovascular #4 #6 2 425 1 282 10 3 717
diseases and
outcomes
#18 Diabetes #4 #7 848 1 370 5 2 223
#19 Birth and growth #4 #8 4 236 1 731 6 5 973
outcomes
#20 Allergy #4 #9 345 530 3 878
#21 Neurodevelopment #4 #10 4 033 4 533 46 8 612
and cognitive
diseases
#22 Dental health #4 #11 5 13 1 19
#23 Immunology #4 #12 2 037 2 220 1 4 258
#24 Cancer #4 #13 1 554 2 273 4 3 831
#25 Mortality #4 #14 1 825 1 893 21 3 739
#26 Bone health #4 #15 304 387 2 693
Number of total hits in all databases 39 092

337
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

RECORDS EXCLUDED DURING FULL-TEXT SCREENING

ALLERGY AND IMMUNOLOGY
TABLE A3.2 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “ALLERGY AND IMMUNOLOGY”,
BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 8) Reason for exclusion
Shoormasti, R. S. Sabetkish, N. Kazemnejad, A. Vahabi, N. Fazlollahi, M. R. & Pourpak, Z. 2019. Are the most common food Excluded based on inclusion and exclusion criteria. Not
allergens in an Iranian atopic population compatible with worldwide reports? A systemic review and meta-analysis with measured health outcome.
molecular classification of frequent allergens. Allergologia et Immunopathologia, 47(6):604-618.
Best, K. P. Gold, M. Kennedy, D. Martin, J. & Makrides, M. 2016. Omega-3 long-chain PUFA intake during pregnancy Excluded as the review has already been assessed in
and allergic disease outcomes in the offspring: a systematic review and meta-analysis of observational studies and VKM 2022.
randomized controlled trials. The American Journal of Clinical Nutrition, 103(1):128-143.
Malmir, H. Larijani, B. & Esmaillzadeh, A. 2022. Fish consumption during pregnancy and risk of allergic diseases in the Excluded as the review has already been assessed in
offspring: A systematic review and meta-analysis. Critical Reviews in Food Science and Nutrition, 62(27):7449-7459. VKM 2022.
Miles, E. A. & Calder, P. C. 2012. Influence of marine n-3 polyunsaturated fatty acids on immune function and Excluded based on inclusion and exclusion criteria: focus
a systematic review of their effects on clinical outcomes in rheumatoid arthritis. British Journal of Nutrition, on n-3 PUFAs and not fish consumption.
107(S2):S171-S184.
Papamichael, M. M. Shrestha, S. K. Itsiopoulos, C. & Erbas, B. 2018. The role of fish intake on asthma in children: A meta‐ Excluded as the review has already been assessed in
analysis of observational studies. Pediatric Allergy and Immunology, 29(4):350-360. VKM 2022.
Rezaeizadeh, H. Mohammadpour, Z. Bitarafan, S. Harirchian, M. H. Ghadimi, M. & Homayon, I. A. 2022. Dietary fish Excluded as the review has already been assessed in
intake and the risk of multiple sclerosis: a systematic review and meta-analysis of observational studies. Nutritional VKM 2022.
Neuroscience, 25(4):681-689.
Yang, H. Xun, P. & He, K. 2013. Fish and fish oil intake in relation to risk of asthma: a systematic review and meta- Excluded as the review has already been assessed in
analysis. PloS One, 8(11):e80048. VKM 2022.
Zhang, G. Q. Liu, B. Li, J. Luo, C. Q. Zhang, Q. Chen, J. L. Li, Z. Y. et al. 2017. Fish intake during pregnancy or infancy and Excluded as the review has already been assessed in
allergic outcomes in children: A systematic review and meta‐analysis. Pediatric Allergy and Immunology, 28(2):152-161. VKM 2022.

338
APPENDICES

TABLE A3.3 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “ALLERGY AND IMMUNOLOGY”
Study (n = 120) Reason for exclusion
Abdollahpour, I., Sormani, M.P., Nedjat, S., Mansournia, M.A. & van der Mei, I. 2021. The role of nutritional factors during Excluded based on inclusion and exclusion criteria:
adolescence in multiple sclerosis onset: a population-based incident case-control study. Nutritional Neuroscience, population-based incident case–control study
24(7):500-507.
Andersen, V., Olsen, A., Carbonnel, F., Tjonneland, A. & Vogel, U. 2012. Diet and risk of inflammatory bowel disease. Excluded based on inclusion and exclusion criteria:
Digestive and Liver Disease, 44(3): 185-194. retrospective case–control
Andrusaityte, S., Grazuleviciene, R. & Petraviciene, I. 2017. Effect of diet and maternal education on allergies among Excluded based on inclusion and exclusion criteria:
preschool children: A case-control study. Environ Res, 159: 374-380. nested case-control study
Atkins, F.M., Steinberg, S.S. & Metcalfe, D.D. 1985. Evaluation of immediate adverse reactions to foods in adult patients. Excluded based on inclusion and exclusion criteria:
I. Correlation of demographic, laboratory, and prick skin test data with response to controlled oral food challenge. J Allergy people with immediate adverse reaction to food
Clin Immunol, 75(3): 348-55
Barman, M., Rabe, H., Hesselmar, B., Johansen, S., Sandberg, A.S. & Wold, A.E. 2020. Cord Blood Levels of EPA, a Marker Excluded based on inclusion and exclusion criteria:
of Fish Intake, Correlate with Infants' T- and B-Lymphocyte Phenotypes and Risk for Allergic Disease. Nutrients, 12(10). inflammatory markers and not specified health outcome
Barman, M., Jonsson, K., Sandin, A., Wold, A.E. & Sandberg, A.S. 2014. Serum fatty acid profile does not reflect seafood Excluded based on inclusion and exclusion criteria:
intake in adolescents with atopic eczema. Acta Paediatr, 103(9): 968-76. cross-sectional study
Benvenga, S., Vigo, M.T., Metro, D., Granese, R., Vita, R. & Le Donne, M. 2016. Type of fish consumed and thyroid Excluded based on inclusion and exclusion criteria:
autoimmunity in pregnancy and postpartum. Endocrine, 52(1): 120-9. autoimmunity markers and not measured specified
health outcome
Beyer, K. & Niggemann, B. 2017. Immunoglobulin E-mediated food allergies in childhood. Monatsschrift Kinderheilkunde, Excluded based on inclusion and exclusion criteria:
165(2): 108-116. non-English language
Chandra, R.K., Puri, S. & Hamed, A. 1989. Influence of maternal diet during lactation and use of formula feeds on Excluded based on inclusion and exclusion criteria:
development of atopic eczema in high risk infants. BMJ, 299(6693): 228-30. retracted paper
Chandra, R.K. 2002. Breast feeding, hydrolysate formulas and delayed introduction of selected foods in the prevention of Excluded based on inclusion and exclusion criteria:
food hypersensitivity and allergic disease. Nutrition Research, 22(1-2): 125-135. narrative review
Chiang, W.C., Kidon, M.I., Liew, W.K., Goh, A., Tang, J.P. & Chay, O.M. 2007. The changing face of food hypersensitivity in Excluded based on inclusion and exclusion criteria:
an Asian community. Clin Exp Allergy, 37(7): 1055-61. patient population
Collier, P.M., Ursell, A., Zaremba, K., Payne, C.M., Staughton, R.C. & Sanders, T. 1993. Effect of regular consumption of oily Excluded based on inclusion and exclusion criteria:
fish compared with white fish on chronic plaque psoriasis. Eur J Clin Nutr, 47(4): 251-4. patient population
Connett, G.J., Gerez, I., Cabrera-Morales, E.A., Yuenyongviwat, A., Ngamphaiboon, J., Chatchatee, P., Sangsupawanich, P. Excluded based on inclusion and exclusion criteria:
et al. 2012. A population-based study of fish allergy in the Philippines, Singapore and Thailand. Int Arch Allergy Immunol, cross-sectional study
159(4): 384-90.
de Mello, V.D., Dahlman, I., Lankinen, M., Kurl, S., Pitkänen, L., Laaksonen, D.E., Schwab, U.S. & Erkkilä, A.T. 2019. The Excluded based on inclusion and exclusion criteria:
effect of different sources of fish and camelina sativa oil on immune cell and adipose tissue mRNA expression in subjects patient population
with abnormal fasting glucose metabolism: a randomized controlled trial. Nutr Diabetes, 9(1): 1
Devereux, G. 2008. Maternal diet during pregnancy: an emerging risk factor for childhood asthma. Expert Rev Clin Excluded based on inclusion and exclusion criteria:
Immunol, 4(6): 663-8. narrative review
D'Hooghe M, B., Haentjens, P., Nagels, G. & De Keyser, J. 2012. Alcohol, coffee, fish, smoking and disease progression in Excluded based on inclusion and exclusion criteria:
multiple sclerosis. Eur J Neurol, 19(4): 616-24. cross-sectional study
Douros, K., Tsabouri, S., Feketea, G., Grammeniatis, V., Koliofoti, E.G., Papadopoulos, M., Sardeli, O., Triga, M. & Priftis, Excluded based on inclusion and exclusion criteria:
K.N. 2019. Retrospective study identified fish and milk as the main culprits in cases of food protein-induced enterocolitis retrospective study
syndrome. Acta Paediatr, 108(10): 1901-1904
Eigenmann, P.A., Sicherer, S.H., Borkowski, T.A., Cohen, B.A. & Sampson, H.A. 1998. Prevalence of IgE-mediated food Excluded based on inclusion and exclusion criteria:
allergy among children with atopic dermatitis. Pediatrics, 101(3): E8. patient population
Ellul-Micallef, R. 1983. Effect of oral sodium cromoglycate and ketotifen in fish-induced bronchial asthma. Thorax, 38(7): Excluded based on inclusion and exclusion criteria:
527-30. patient population
Ferraro, V., Zanconato, S. & Carraro, S. 2019. Timing of Food Introduction and the Risk of Food Allergy. Nutrients, 11(5) Excluded based on inclusion and exclusion criteria:
narrative review
García-Rodríguez, C.E., Olza, J., Aguilera, C.M., Mesa, M.D., Miles, E.A., Noakes, P.S., Vlachava, M. et al. 2012. Plasma Excluded based on inclusion and exclusion criteria:
inflammatory and vascular homeostasis biomarkers increase during human pregnancy but are not affected by oily fish inflammatory markers and not measured health outcome
intake. J Nutr, 142(7): 1191-6.
Hageman, J.H., Hooyenga, P., Diersen-Schade, D.A., Scalabrin, D.M., Wichers, H.J. & Birch, E.E. 2012. The impact of dietary Excluded based on inclusion and exclusion criteria:
long-chain polyunsaturated fatty acids on respiratory illness in infants and children. Curr Allergy Asthma Rep, 12(6): narrative review
564-73.
Hansen, T.K. & Bindslev-Jensen, C. 1992. Codfish allergy in adults. Identification and diagnosis. Allergy, 47(6): 610-7 Excluded based on inclusion and exclusion criteria:
patient population
Hanson, C., Sayles, H., Rutten, E., Wouters, E.F.M., MacNee, W., Calverley, P., Meza, J.L. & Rennard, S. 2014. The Excluded based on inclusion and exclusion criteria: not
Association Between Dietary Intake and Phenotypical Characteristics of COPD in the ECLIPSE Cohort. Chronic Obstr Pulm included health outcome
Dis, 1(1): 115-124.
Hattevig, G., Kjellman, B., Sigurs, N., Grodzinsky, E., Hed, J. & Björkstén, B. 1990. The effect of maternal avoidance of Excluded based on inclusion and exclusion criteria:
eggs, cow's milk, and fish during lactation on the development of IgE, IgG, and IgA antibodies in infants. J Allergy Clin wrong health outcome (only IgE antibodies)
Immunol, 85(1 Pt 1): 108-15..

339
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.3 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “ALLERGY AND IMMUNOLOGY” (cont.)
Study (n = 120) Reason for exclusion
Helbling, A., Haydel, R., Jr., McCants, M.L., Musmand, J.J., El-Dahr, J. & Lehrer, S.B. 1999. Fish allergy: is cross-reactivity Excluded based on inclusion and exclusion criteria:
among fish species relevant? Double-blind placebo-controlled food challenge studies of fish allergic adults. Ann Allergy patients with fish allergy
Asthma Immunol, 83(6 Pt 1): 517-23
Hong, S.J., Lee, M.S., Lee, S.Y., Ahn, K.M., Oh, J.W., Kim, K.E., Lee, J.S. & Lee, H.B. 2006. High body mass index and dietary Excluded based on inclusion and exclusion criteria:
pattern are associated with childhood asthma. Pediatr Pulmonol, 41(12): 1118-24. cross-sectional study
Hooper, R., Heinrich, J., Omenaas, E., Sausenthaler, S., Garcia-Larsen, V., Bakolis, I. & Burney, P. 2010. Dietary patterns Excluded based on inclusion and exclusion criteria:
and risk of asthma: results from three countries in European Community Respiratory Health Survey-H. British Journal of cross-sectional study
Nutrition, 103(9): 1354-1365.
Hua, M.C., Yao, T.C., Chen, C.C., Tsai, M.H., Liao, S.L., Lai, S.H., Chiu, C.Y., Su, K.W., Yeh, K.W. & Huang, J.L. 2017. Excluded based on inclusion and exclusion criteria:
Introduction of various allergenic foods during infancy reduces risk of IgE sensitization at 12 months of age: a birth wrong health outcome (only IgE sensitization)
cohort study. Pediatr Res, 82(5): 733-740.
Infante, S., Marco-Martín, G., Sánchez-Domínguez, M., Rodríguez-Fernández, A., Fuentes-Aparicio, V., Alvarez-Perea, A., Excluded based on inclusion and exclusion criteria:
Cabrera-Freitag, P., Morales-Cabeza, C., Zubeldia, J.M. & Zapatero, L. 2018. Food protein-induced enterocolitis syndrome not included health outcome (food protein-induced
by fish: Not necessarily a restricted diet. Allergy, 73(3): 728-732.. enterocolitis syndrome)
Infante, S., Pérez-Pallisé, E., Skrabski, F., Cabrera-Freitag, P., Morales-Cabeza, C., Fuentes-Aparicio, V., Alvarez-Perea, Excluded based on inclusion and exclusion criteria:
A. & Zubeldia, J.M. 2021. Poor prognosis of food protein-induced enterocolitis syndrome to fish. Pediatr Allergy Immunol, not included health outcome (food protein-induced
32(3): 560-565 enterocolitis syndrome)
Jelinek, G.A., Hadgkiss, E.J., Weiland, T.J., Pereira, N.G., Marck, C.H. & van der Meer, D.M. 2013. Association of fish Excluded based on inclusion and exclusion criteria:
consumption and Ω 3 supplementation with quality of life, disability and disease activity in an international cohort of omega-3 supplementation only
people with multiple sclerosis. Int J Neurosci, 123(11): 792-800.
Jonsson, K., Barman, M., Brekke, H.K., Hesselmar, B., Johansen, S., Sandberg, A.S. & Wold, A.E. 2017. Late introduction of Excluded based on inclusion and exclusion criteria:
fish and eggs is associated with increased risk of allergy development - results from the FARMFLORA birth cohort. Food & retrospective study
Nutrition Research, 61.
Jonsson, K., Barman, M., Moberg, S., Sjöberg, A., Brekke, H.K., Hesselmar, B., Sandberg, A.S. & Wold, A.E. 2016. Serum Excluded based on inclusion and exclusion criteria: not
fatty acids in infants, reflecting family fish consumption, were inversely associated with allergy development but not measured fish consumption (only fatty acids)
related to farm residence. Acta Paediatr, 105(12): 1462-1471.
Jonsson, K., Green, M., Barman, M., Sjöberg, A., Brekke, H.K., Wold, A.E. & Sandberg, A.S. 2016. Diet in 1-year-old farm Excluded based on inclusion and exclusion criteria:
and control children and allergy development: results from the FARMFLORA birth cohort. Food Nutr Res, 60: 32721. retrospective study
Klingberg, S., Brekke, H.K. & Ludvigsson, J. 2019. Introduction of fish and other foods during infancy and risk of asthma Excluded based on inclusion and exclusion criteria:
in the All Babies In Southeast Sweden cohort study. European Journal of Pediatrics, 178(3): 395-402.. covered by birth and growth
Knope, K., Sloan-Gardner, T.S. & Stafford, R.J. 2014. Histamine fish poisoning in Australia, 2001 to 2013. Commun Dis Excluded based on inclusion and exclusion criteria:
Intell Q Rep, 38(4): E285-93.. patient population
Kusunoki, T., Takeuchi, J., Morimoto, T., Sakuma, M., Yasumi, T., Nishikomori, R., Higashi, A. & Heike, T. 2017. Fruit intake Excluded based on inclusion and exclusion criteria:
reduces the onset of respiratory allergic symptoms in schoolchildren. Pediatr Allergy Immunol, 28(8): 793-800. questionnaire
Lavon, O., Lurie, Y. & Bentur, Y. 2008. Scombroid fish poisoning in Israel, 2005-2007. Isr Med Assoc J, 10(11): 789-92. Excluded based on inclusion and exclusion criteria: fish
poisoning
Lee, H.L., Tang, M.M., Bakhtiar, M.F., Mohamad Yadzir, Z.H. & Johar, A. 2021. Sensitization to Local Seafood Allergens in Excluded based on inclusion and exclusion criteria:
Adult Patients with Atopic Dermatitis in Malaysia. Int Arch Allergy Immunol, 182(2): 153-157 patients with atopic dermatitis
Lindqvist, H.M., Gjertsson, I., Eneljung, T. & Winkvist, A. 2018. Influence of Blue Mussel (Mytilus edulis) Intake on Disease Excluded based on inclusion and exclusion criteria:
Activity in Female Patients with Rheumatoid Arthritis: The MIRA Randomized Cross-Over Dietary Intervention. Nutrients, patients with rheumatoid arthritis
10(4).
Losol, P., Rezwan, F.I., Patil, V.K., Venter, C., Ewart, S., Zhang, H., Arshad, S.H., Karmaus, W. & Holloway, J.W. 2019. Effect Excluded based on inclusion and exclusion criteria:
of gestational oily fish intake on the risk of allergy in children may be influenced by FADS1/2, ELOVL5 expression and DNA children
methylation. Genes Nutr, 14: 20
Ludman, S., Harmon, M., Whiting, D. & du Toit, G. 2014. Clinical presentation and referral characteristics of food protein- Excluded based on inclusion and exclusion criteria: food
induced enterocolitis syndrome in the United Kingdom. Ann Allergy Asthma Immunol, 113(3): 290-4. protein-induced enterocolitis syndrome
Machowicz, A., Hall, I., de Pablo, P., Rauz, S., Richards, A., Higham, J., Poveda-Gallego, A. et al. 2020. Mediterranean diet Excluded based on inclusion and exclusion criteria:
and risk of Sjogren's syndrome. Clinical and Experimental Rheumatology, 38(4): S216-S221 patients with Sjögren’s syndrome
Merchant, A.T., Curhan, G.C., Rimm, E.B., Willett, W.C. & Fawzi, W.W. 2005. Intake of n-6 and n-3 fatty acids and fish and Excluded based on inclusion and exclusion criteria:
risk of community-acquired pneumonia in US men. Am J Clin Nutr, 82(3): 668-74. pneumonia is not allergy, mostly caused by
microorganisms
Miles, E.A. & Calder, P.C. 2017. Can Early Omega-3 Fatty Acid Exposure Reduce Risk of Childhood Allergic Disease? Excluded based on inclusion and exclusion criteria: only
Nutrients, 9(7) omega-3
Milewska-Wróbel, D. & Lis-Święty, A. 2020. Does maternal diet during pregnancy influence clinical and laboratory Excluded based on inclusion and exclusion criteria:
characteristics of infantile-onset atopic dermatitis? Eur Ann Allergy Clin Immunol, 52(6): 277-279. exposure during maternal diet during pregnancy
Minamino, H., Katsushima, M., Torii, M., Hashimoto, M., Fujita, Y., Ikeda, K., Yamamoto, W. et al. 2021. Habitual fish Excluded based on inclusion and exclusion criteria:
intake negatively correlates with prevalence of frailty among patients with rheumatoid arthritis. Scientific Reports, 11(1). cross-sectional study
Miyake, Y., Sasaki, S., Tanaka, K., Ohya, Y., Matsunaga, I., Yoshida, T., Hirota, Y. & Oda, H. 2008. Relationship between Excluded based on inclusion and exclusion criteria:
dietary fat and fish intake and the prevalence of atopic eczema in pregnant Japanese females: baseline data from the cross-sectional study
Osaka Maternal and Child Health Study. Asia Pac J Clin Nutr, 17(4): 612-9.

340
APPENDICES

TABLE A3.3 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “ALLERGY AND IMMUNOLOGY” (cont.)
Study (n = 120) Reason for exclusion
Mori, T.A. & Beilin, L.J. 2004. Omega-3 fatty acids and inflammation. Curr Atheroscler Rep, 6(6): 461-7. Excluded based on inclusion and exclusion criteria: only
omega-3
Muche-Borowski, C., Kopp, M., Reese, I., Sitter, H., Werfel, T. & Schäfer, T. 2009. Allergy prevention. Dtsch Arztebl Int, Excluded based on inclusion and exclusion criteria:
106(39): 625-31. review article
Murakami, I., Murakami, K., Hashimoto, M., Tanaka, M., Ito, H., Fujii, T., Torii, M. et al. 2020. Intake frequency of Excluded based on inclusion and exclusion criteria:
vegetables or seafoods negatively correlates with disease activity of rheumatoid arthritis. Plos One, 15(2). patient population
Nagel, G., Weinmayr, G., Kleiner, A., Garcia-Marcos, L., Strachan, D.P. & Grp, I.P.T.S. 2010. Effect of diet on asthma and Excluded based on inclusion and exclusion criteria:
allergic sensitisation in the International Study on Allergies and Asthma in Childhood (ISAAC) Phase Two. Thorax, 65(6): cross-sectional study
516-522.
Norback, D., Zhao, Z.H., Wang, Z.H., Wieslander, G., Mi, Y.H. & Zhang, Z. 2007. Asthma, eczema, and reports on pollen and Excluded based on inclusion and exclusion criteria: not
cat allergy among pupils in Shanxi province, China. International Archives of Occupational and Environmental Health, related to fish consumption
80(3): 207-216.
Papadopol, A. 2018. A retrospective cohort analysis investigating the effects of salmon consumption during pregnancy on Excluded based on inclusion and exclusion criteria:
signs of allergy in children at 30 months of age. Clinical and Experimental Allergy, 48(11):1565-1565. covered by birth and growth
Papadopoulou, A., Lagousi, T., Hatzopoulou, E., Korovessi, P., Kostaridou, S. & Mermiri, D.Z. 2021. Atypical food protein- Excluded based on inclusion and exclusion criteria: not
induced enterocolitis syndrome in children: Is IgE sensitisation an issue longitudinally? Allergol Immunopathol (Madr), related to fish consumption
49(3): 73-82.
Resano, A., Crespo, E., Fernández Benítez, M., Sanz, M.L. & Oehling, A. 1998. Atopic dermatitis and food allergy. J Investig Excluded based on inclusion and exclusion criteria:
Allergol Clin Immunol, 8(5): 271-6. cross-sectional study
Rodriguez-Rodriguez, E., Perea, J.M., Jimenez, A.I., Rodriguez-Rodriguez, P., Lopez-Sobaler, A.M. & Ortega, R.M. 2010. Fat Excluded based on inclusion and exclusion criteria:
intake and asthma in Spanish schoolchildren. European Journal of Clinical Nutrition, 64(10): 1065-1071 cross-sectional study
Sampson, H.A. 2001. Utility of food-specific IgE concentrations in predicting symptomatic food allergy. J Allergy Clin Excluded based on inclusion and exclusion criteria:
Immunol, 107(5): 891-6. modelling
Shams, K., Grindlay, D.J.C. & Williams, H.C. 2011. What's new in atopic eczema? An analysis of systematic reviews Excluded based on inclusion and exclusion criteria:
published in 2009-2010. Clinical and Experimental Dermatology, 36(6): 573-578. review of systematic review studies
Sinitkul, R., Manuyakorn, W., Kamchaisatian, W., Vilaiyuk, S., Benjaponpitak, S., Lertudompholwanit, C. & Excluded based on inclusion and exclusion criteria:
Treepongkaruna, S. 2018. De novo food allergy in pediatric liver transplantation recipients. Asian Pac J Allergy Immunol, retrospective cohort study
36(3): 166-174.
Sopo, S.M., Giorgio, V., Dello Iacono, I., Novembre, E., Mori, F. & Onesimo, R. 2012. A multicentre retrospective study of 66 Excluded based on inclusion and exclusion criteria:
Italian children with food protein-induced enterocolitis syndrome: different management for different phenotypes. Clin retrospective study
Exp Allergy, 42(8): 1257-65.
Sparks, J.A., Iversen, M.D., Kroouze, R.M., Mahmoud, T.G., Triedman, N.A., Kalia, S.S., Atkinson, M.L. et al. 2014. Excluded based on inclusion and exclusion criteria:
Personalized Risk Estimator for Rheumatoid Arthritis (PRE-RA) Family Study: Rationale and design for a randomized protocol
controlled trial evaluating rheumatoid arthritis risk education to first-degree relatives. Contemporary Clinical Trials,
39(1): 145-157.
Stratakis, N., Conti, D.V., Borras, E., Sabido, E., Roumeliotaki, T., Papadopoulou, E., Agier, L. et al. 2020. Association Excluded based on inclusion and exclusion criteria:
of Fish Consumption and Mercury Exposure During Pregnancy With Metabolic Health and Inflammatory Biomarkers in mercury and inflammatory markers; exclusion criteria
Children. Jama Network Open, 3(3)
Syrjälä, E., Nevalainen, J., Peltonen, J., Takkinen, H.M., Hakola, L., Åkerlund, M., Veijola, R. et al. 2019. A Joint Modeling Excluded based on inclusion and exclusion criteria:
Approach for Childhood Meat, Fish and Egg Consumption and the Risk of Advanced Islet Autoimmunity. Sci Rep, 9(1): modelling
7760.
Saadeh, D., Salameh, P., Caillaud, D., Charpin, D., De Blay, F., Kopferschmitt, C., ... & Raherison, C. 2015. Prevalence and Excluded based on inclusion and exclusion criteria:
association of asthma and allergic sensitization with dietary factors in schoolchildren: data from the french six cities cross-sectional study
study. BMC Public Health, 15, 1-11.
Tamay, Z., Akcay, A., Ergin, A. & Guler, N. 2013. Effects of dietary habits and risk factors on allergic rhinitis prevalence Excluded based on inclusion and exclusion criteria:
among Turkish adolescents. Int J Pediatr Otorhinolaryngol, 77(9): 1416-23. cross-sectional study
Tani, S., Matsuo, R., Atsumi, W., Kawauchi, K., Ashida, T., Yagi, T., Imatake, K. et al. 2021. Higher Frequency of Fish Intake Excluded based on inclusion and exclusion criteria:
May Be Associated with a Lower Neutrophil/Lymphocyte Ratio: Anti-Atherosclerotic Effects of Fish Consumption. Annals of cross-sectional study
Nutrition and Metabolism, 77(3): 146-153
Tedeschi, S.K., Bathon, J.M., Giles, J.T., Lin, T.C., Yoshida, K. & Solomon, D.H. 2018. Relationship Between Fish Excluded based on inclusion and exclusion criteria:
Consumption and Disease Activity in Rheumatoid Arthritis. Arthritis Care & Research, 70(3): 327-332 cross-sectional analysis
Thong, B.Y., Cheng, Y.K., Leong, K.P., Tang, C.Y. & Chng, H.H. 2007. Immediate food hypersensitivity among adults Excluded based on inclusion and exclusion criteria:
attending a clinical immunology/allergy centre in Singapore. Singapore Med J, 48(3): 236-40. retrospective review
Turner, P., Ng, I., Kemp, A. & Campbell, D. 2011. Seafood allergy in children: a descriptive study. Ann Allergy Asthma Excluded based on inclusion and exclusion criteria:
Immunol, 106(6): 494-501. retrospective review
Untersmayr, E., Vestergaard, H., Malling, H.J., Jensen, L.B., Platzer, M.H., Boltz-Nitulescu, G., Scheiner, O., Skov, P.S., Excluded based on inclusion and exclusion criteria:
Jensen-Jarolim, E. & Poulsen, L.K. 2007. Incomplete digestion of codfish represents a risk factor for anaphylaxis in absorption kinetics of fish protein was tested in patients
patients with allergy. J Allergy Clin Immunol, 119(3): 711-7. with fish allergy
Urwin, H.J., Miles, E.A., Noakes, P.S., Kremmyda, L.S., Vlachava, M., Diaper, N.D., Godfrey, K.M., Calder, P.C., Vulevic, J. & Excluded based on inclusion and exclusion criteria:
Yaqoob, P. 2014. Effect of salmon consumption during pregnancy on maternal and infant faecal microbiota, secretory IgA cross-sectional study
and calprotectin. Br J Nutr, 111(5): 773-84.

341
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.3 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “ALLERGY AND IMMUNOLOGY” (cont.)
Study (n = 120) Reason for exclusion
Wang, J.J., Rochtchina, E., Smith, W., Klein, R., Klein, B.E., Joshi, T., Sivakumaran, T.A., Iyengar, S. & Mitchell, P. 2009. Excluded based on inclusion and exclusion criteria: age-
Combined effects of complement factor H genotypes, fish consumption, and inflammatory markers on long-term risk for related macular degeneration
age-related macular degeneration in a cohort. Am J Epidemiol, 169(5): 633-41.
Xepapadaki, P., Kitsioulis, N.A., Manousakis, E., Manolaraki, I., Douladiris, N. & Papadopoulos, N.G. 2019. Remission Excluded based on inclusion and exclusion criteria:
Patterns of Food Protein-Induced Enterocolitis Syndrome in a Greek Pediatric Population. Int Arch Allergy Immunol, 180(2): retrospective study
113-119.
Yadana, S., Talegawkar, S.A., Mathad, J.S., Alexander, M., Rajagopalan, K., Kumar, P., Naik, S. et al. 2020. Association of Excluded based on inclusion and exclusion criteria:
Vegetable and Animal Flesh Intake with Inflammation in Pregnant Women from India. Nutrients, 12(12). dietary patterns study and markers of immunity
Ye, S.Q., Mo, X.M., Liu, J.F., Yan, F.G. & Chen, D.C. 2019. Factors Influencing Atopic Dermatitis Incidence in Offspring. Excluded based on inclusion and exclusion criteria:
Iranian Journal of Allergy Asthma and Immunology, 18(4): 347-357. review paper
Zhan, T., Ali, A., Choi, J.G., Lee, M., Leung, J., Dellon, E.S., Garber, J.J. & Hur, C. 2018. Model to Determine the Optimal Excluded based on inclusion and exclusion criteria:
Dietary Elimination Strategy for Treatment of Eosinophilic Esophagitis. Clin Gastroenterol Hepatol, 16(11): 1730-1737.e2. modelling study
Calvani, M., Alessandri, C., Sopo, S.M., Panetta, V., Pingitore, G., Tripodi, S., Zappalà, D. & Zicari, A.M. 2006. Consumption Excluded for further assessment as the primary study
of fish, butter and margarine during pregnancy and development of allergic sensitizations in the offspring: role of had already been assessed in one of the included
maternal atopy. Pediatr Allergy Immunol, 17(2): 94-102. systematic reviews
Alm, B., Goksor, E., Thengilsdottir, H., Pettersson, R., Mollborg, P., Norvenius, G., Erdes, L., Aberg, N. & Wennergren, G. Excluded for further assessment as the primary study
2011. Early protective and risk factors for allergic rhinitis at age 41/2 yr. Pediatric Allergy and Immunology, 22(4): 398- had already been assessed in a systematic review that
404 was included in VKM 2022
Benito-Garcia, E., Feskanich, D., Hu, F.B., Mandl, L.A. & Karlson, E.W. 2007. Protein, iron, and meat consumption and risk Excluded for further assessment as the primary study
for rheumatoid arthritis: a prospective cohort study. Arthritis Research & Therapy, 9(1). had already been assessed in VKM 2022
Di Giuseppe, D., Wallin, A., Bottai, M., Askling, J. & Wolk, A. 2014. Long-term intake of dietary long-chain n-3 Excluded for further assessment, as the primary study
polyunsaturated fatty acids and risk of rheumatoid arthritis: a prospective cohort study of women. Annals of the had already been assessed in VKM 2022
Rheumatic Diseases, 73(11): 1949-1953.
Dotterud, C.K., Storrø, O., Simpson, M.R., Johnsen, R. & Øien, T. 2013. The impact of pre- and postnatal exposures on Excluded for further assessment, as the primary study
allergy related diseases in childhood: a controlled multicentre intervention study in primary health care. BMC Public had already been assessed in VKM 2022
Health, 13: 123.
Hamazaki, K., Tsuchida, A., Takamori, A., Tanaka, T., Ito, M., Inadera, H., Kawamoto, T. et al. 2019. Dietary intake of Excluded for further assessment, as the primary study
fish and omega-3 polyunsaturated fatty acids and physician-diagnosed allergy in Japanese population: The Japan had already been assessed in VKM 2022
Environment and Children's Study. Nutrition, 61: 194-201.
Jedrychowski, W., Flak, E., Mroz, E., Pac, A., Jacek, R., Sochacka-Tatara, E., Spengler, J., Rauh, V. & Perera, F. 2008. Excluded for further assessment, as the primary study
Modulating effects of maternal fish consumption on the occurrence of respiratory symptoms in early infancy attributed to had already been assessed in one of the included
prenatal exposure to fine particles. Annals of Nutrition and Metabolism, 52(1): 8-16. systematic reviews
Jedrychowski, W., Perera, F., Maugeri, U., Mrozek-Budzyn, D., Miller, R.L., Flak, E., Mroz, E., Jacek, R. & Spengler, J.D. 2011. Excluded for further assessment, as the primary study
Effects of Prenatal and Perinatal Exposure to Fine Air Pollutants and Maternal Fish Consumption on the Occurrence of had already been assessed in one of the included
Infantile Eczema. International Archives of Allergy and Immunology, 155(3): 275-281 systematic reviews
Kampman, M.T., Wilsgaard, T. & Mellgren, S.I. 2007. Outdoor activities and diet in childhood and adolescence relate to MS Excluded for further assessment, as the primary study
risk above the Arctic Circle. J Neurol, 254(4): 471-7. had already been assessed in VKM 2022
Khan, F., Orson, F., Ogawa, Y., Parker, C. & Davis, C.M. 2011. Adult seafood allergy in the Texas Medical Center: A 13-year Excluded for further assessment, as the primary study
experience. Allergy Rhinol (Providence), 2(2): e71-7. had already been assessed in VKM 2022
Kiefte-de Jong, J.C., de Vries, J.H., Franco, O.H., Jaddoe, V.W., Hofman, A., Raat, H., de Jongste, J.C. & Moll, H.A. 2012. Fish Excluded for further assessment, as the primary study
consumption in infancy and asthma-like symptoms at preschool age. Pediatrics, 130(6): 1060-8. had already been assessed in VKM 2022
Kull, I., Bergström, A., Lilja, G., Pershagen, G. & Wickman, M. 2006. Fish consumption during the first year of life and Excluded for further assessment, as the primary study
development of allergic diseases during childhood. Allergy, 61(8): 1009-15. had already been assessed in one of the included
systematic reviews
Leermakers, E.T., Sonnenschein-van der Voort, A.M., Heppe, D.H., de Jongste, J.C., Moll, H.A., Franco, O.H., Hofman, Excluded for further assessment, as the primary study
A., Jaddoe, V.W. & Duijts, L. 2013. Maternal fish consumption during pregnancy and risks of wheezing and eczema in had already been assessed in a systematic review that
childhood: the Generation R Study. Eur J Clin Nutr, 67(4): 353-9. was included in VKM 2022
Li, J., Xun, P., Zamora, D., Sood, A., Liu, K., Daviglus, M., Iribarren, C., Jacobs, D., Jr., Shikany, J.M. & He, K. 2013. Intakes Excluded for further assessment, as the primary study
of long-chain omega-3 (n-3) PUFAs and fish in relation to incidence of asthma among American young adults: the CARDIA had already been assessed in VKM 2022
study. Am J Clin Nutr, 97(1): 173-8.
Loo, E.X.L., Ong, L., Goh, A., Chia, A.R., Teoh, O.H., Colega, M.T., Chan, Y.H. et al. 2017. Effect of Maternal Dietary Patterns Excluded for further assessment, as the primary study
during Pregnancy on Self-Reported Allergic Diseases in the First 3 Years of Life: Results from the GUSTO Study. Int Arch had already been assessed in one of the included
Allergy Immunol, 173(2): 105-113. systematic reviews
Lumia, M., Luukkainen, P., Tapanainen, H., Kaila, M., Erkkola, M., Uusitalo, L., Niinistö, S. et al. 2011. Dietary fatty acid Excluded for further assessment, as the primary study
composition during pregnancy and the risk of asthma in the offspring. Pediatr Allergy Immunol, 22(8): 827-35 had already been assessed in one of the included
systematic reviews
Lumia, M., Takkinen, H.M., Luukkainen, P., Kaila, M., Lehtinen-Jacks, S., Nwaru, B.I., Tuokkola, J. et al. 2015. Food Excluded for further assessment, as the primary study
consumption and risk of childhood asthma. Pediatr Allergy Immunol, 26(8): 789-96. had already been assessed in VKM 2022
Lundgren, S.N., Madan, J.C., Emond, J.A., Morrison, H.G., Christensen, B.C., Karagas, M.R. & Hoen, A.G. 2018. Maternal Excluded for further assessment, as the primary study
diet during pregnancy is related with the infant stool microbiome in a delivery mode-dependent manner. Microbiome, 6 had already been assessed in VKM 2022
Magnusson, J., Kull, I., Westman, M., Håkansson, N., Wolk, A., Melén, E., Wickman, M. & Bergström, A. 2015. Fish and Excluded for further assessment, as the primary study
polyunsaturated fat intake and development of allergic and nonallergic rhinitis. J Allergy Clin Immunol, 136(5): 1247-53.e1-2. had already been assessed in VKM 2022

342
APPENDICES

TABLE A3.3 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “ALLERGY AND IMMUNOLOGY” (cont.)
Study (n = 120) Reason for exclusion
Magnusson, J., Kull, I., Rosenlund, H., Håkansson, N., Wolk, A., Melén, E., Wickman, M. & Bergström, A. 2013. Fish Excluded for further assessment, as the primary study
consumption in infancy and development of allergic disease up to age 12 y. Am J Clin Nutr, 97(6): 1324-30. had already been assessed in VKM 2022
Maslova, E., Strøm, M., Oken, E., Campos, H., Lange, C., Gold, D. & Olsen, S.F. 2013. Fish intake during pregnancy and the Excluded for further assessment, as the primary study
risk of child asthma and allergic rhinitis - longitudinal evidence from the Danish National Birth Cohort. Br J Nutr, 110(7): had already been assessed in one of the included
1313-25. systematic reviews
Miyake, Y., Tanaka, K., Okubo, H., Sasaki, S. & Arakawa, M. 2012. Dietary meat and fat intake and prevalence of Excluded for further assessment, as the primary study
rhinoconjunctivitis in pregnant Japanese women: baseline data from the Kyushu Okinawa Maternal and Child Health had already been assessed in VKM 2022
Study. Nutr J, 11: 19.
Miyake, Y., Sasaki, S., Tanaka, K., Ohya, Y., Miyamoto, S., Matsunaga, I., Yoshida, T., Hirota, Y. & Oda, H. 2007. Fish and Excluded for further assessment, as the primary study
fat intake and prevalence of allergic rhinitis in Japanese females: the Osaka Maternal and Child Health Study. J Am Coll had already been assessed in VKM 2022
Nutr, 26(3): 279-87.
Miyake, Y., Tanaka, K., Okubo, H., Sasaki, S. & Arakawa, M. 2013. Maternal fat intake during pregnancy and wheeze and Excluded for further assessment, as the primary study
eczema in Japanese infants: the Kyushu Okinawa Maternal and Child Health Study. Annals of Epidemiology, 23(11): had already been assessed in one of the included
674-680. systematic reviews
Nafstad, P., Nystad, W., Magnus, P. & Jaakkola, J.J.K. 2003. Asthma and allergic rhinitis at 4 years of age in relation to fish Excluded for further assessment, as the primary study
consumption in infancy. Journal of Asthma, 40(4): 343-348 had already been assessed in one of the included
systematic reviews
Noakes, P.S., Vlachava, M., Kremmyda, L.S., Diaper, N.D., Miles, E.A., Erlewyn-Lajeunesse, M., Williams, A.P., Godfrey, K.M. Excluded for further assessment, as the primary study
& Calder, P.C. 2012. Increased intake of oily fish in pregnancy: effects on neonatal immune responses and on clinical had already been assessed in one of the included
outcomes in infants at 6 mo. Am J Clin Nutr, 95(2): 395-404. systematic reviews
Nwaru, B.I., Takkinen, H.M., Niemelä, O., Kaila, M., Erkkola, M., Ahonen, S., Tuomi, H. et al. 2013. Introduction of Excluded for further assessment, as the primary study
complementary foods in infancy and atopic sensitization at the age of 5 years: timing and food diversity in a Finnish birth had already been assessed in VKM 2022
cohort. Allergy, 68(4): 507-16.
Oien, T., Storrø, O. & Johnsen, R. 2010. Do early intake of fish and fish oil protect against eczema and doctor-diagnosed Excluded for further assessment, as the primary study
asthma at 2 years of age? A cohort study. J Epidemiol Community Health, 64(2): 124-9. had already been assessed in one of the included
systematic reviews
Ozawa, N., Shimojo, N., Suzuki, Y., Ochiai, S., Nakano, T., Morita, Y., Inoue, Y., Arima, T., Suzuki, S. & Kohno, Y. 2014. Excluded for further assessment, as the primary study
Maternal intake of Natto, a Japan's traditional fermented soybean food, during pregnancy and the risk of eczema in had already been assessed in one of the included
Japanese babies. Allergol Int, 63(2): 261-6. systematic reviews
Pedersen, M., Stripp, C., Klarlund, M., Olsen, S.F., Tjonneland, A.M. & Frisch, M. 2005. Diet and risk of rheumatoid arthritis Excluded for further assessment, as the primary study
in a prospective cohort. Journal of Rheumatology, 32(7): 1249-1252 had already been assessed in VKM 2022
Pele, F., Bajeux, E., Gendron, H., Monfort, C., Rouget, F., Multigner, L., Viel, J.F. & Cordier, S. 2013. Maternal fish and Excluded for further assessment, as the primary study
shellfish consumption and wheeze, eczema and food allergy at age two: a prospective cohort study in Brittany, France. had already been assessed in VKM 2022
Environmental Health, 12..
Romieu, I., Torrent, M., Garcia-Esteban, R., Ferrer, C., Ribas-Fito, N., Anto, J.M. & Sunyer, J. 2007. Maternal fish intake Excluded for further assessment, as the primary study
during pregnancy and atopy and asthma in infancy. Clinical and Experimental Allergy, 37(4): 518-525. had already been assessed in one of the included
systematic reviews
Saito, K., Yokoyama, T., Miyake, Y., Sasaki, S., Tanaka, K., Ohya, Y. & Hirota, Y. 2010. Maternal meat and fat consumption Excluded for further assessment, as the primary study
during pregnancy and suspected atopic eczema in Japanese infants aged 3-4 months: The Osaka Maternal and Child had already been assessed in one of the included
Health Study. Pediatric Allergy and Immunology, 21(1): 38-46. systematic reviews
Sausenthaler, S., Koletzko, S., Schaaf, B., Lehmann, I., Borte, M., Herbarth, O., von Berg, A., Wichmann, H.E. & Heinrich, J. Excluded for further assessment, as the primary study
2007. Maternal diet during pregnancy in relation to eczema and allergic sensitization in the offspring at 2 y of age. Am J had already been assessed in one of the included
Clin Nutr, 85(2): 530-7 systematic reviews
Sparks, J.A., O'Reilly, E.J., Barbhaiya, M., Tedeschi, S.K., Malspeis, S., Lu, B., Willett, W.C., Costenbader, K.H. & Karlson, Excluded for further assessment, as the primary study
E.W. 2019. Association of fish intake and smoking with risk of rheumatoid arthritis and age of onset: a prospective cohort had already been assessed in VKM 2022
study. BMC Musculoskeletal Disorders, 20:1-13.
Stratakis, N., Roumeliotaki, T., Oken, E., Ballester, F., Barros, H., Basterrechea, M., Cordier, S. et al. 2017. Fish and Excluded for further assessment, as the primary study
seafood consumption during pregnancy and the risk of asthma and allergic rhinitis in childhood: a pooled analysis of 18 had already been assessed in one of the included
European and US birth cohorts. Int J Epidemiol, 46(5): 1465-1477. systematic reviews
Talaei, M., Sdona, E., Calder, P.C., Jones, L.R., Emmett, P.M., Granell, R., Bergstrom, A., Melen, E. & Shaheen, S.O. 2021. Excluded for further assessment, as the primary study
Intake of n-3 polyunsaturated fatty acids in childhood, FADS genotype and incident asthma. European Respiratory had already been assessed in VKM 2022
Journal, 58(3)
Willers, S.M., Wijga, A.H., Brunekreef, B., Scholtens, S., Postma, D.S., Kerkhof, M., de Jongste, J.C. & Smit, H.A. 2011. Excluded for further assessment, as the primary study
Childhood diet and asthma and atopy at 8 years of age: the PIAMA birth cohort study. Eur Respir J, 37(5): 1060-7. had already been assessed in VKM 2022
Willers, S.M., Devereux, G., Craig, L.C., McNeill, G., Wijga, A.H., Abou El-Magd, W., Turner, S.W., Helms, P.J. & Seaton, A. Excluded for further assessment, as the primary study
2007. Maternal food consumption during pregnancy and asthma, respiratory and atopic symptoms in 5-year-old children. had already been assessed in one of the included
Thorax, 62(9): 773-9. systematic reviews
Willers, S.M., Wijga, A.H., Brunekreef, B., Kerkhof, M., Gerritsen, J., Hoekstra, M.O., de Jongste, J.C. & Smit, H.A. 2008. Excluded for further assessment, as the primary study
Maternal food consumption during pregnancy and the longitudinal development of childhood asthma. Am J Respir Crit had already been assessed in one of the included
Care Med, 178(2): 124-31. systematic reviews
Zeiger, R.S., Heller, S., Mellon, M.H., Forsythe, A.B., O'Connor, R.D., Hamburger, R.N. & Schatz, M. 1989. Effect of combined Excluded for further assessment, as the primary study
maternal and infant food-allergen avoidance on development of atopy in early infancy: a randomized study. J Allergy Clin had already been assessed in one of the included
Immunol, 84(1): 72-89. systematic reviews

343
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

BIRTH AND GROWTH OUTCOMES

TABLE A3.4 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “BIRTH AND GROWTH OUTCOMES”
Study (n = 1) Reason for exclusion
Hibbeln, J. R. Spiller, P. Brenna, J. T. Golding, J. Holub, B. J. Harris, W. S. ... & Carlson, S. E. (2019). Relationships between Excluded based on inclusion and exclusion criteria:
seafood consumption during pregnancy and childhood and neurocognitive development: Two systematic reviews. Wrong outcome (review included in neurocognitive
Prostaglandins, Leukotrienes and Essential Fatty Acids, 151, 14-36. outcomes)

TABLE A3.5 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “BIRTH AND GROWTH”
Study (n = 31) Reason for exclusion
Andersen, R. Biltoft-Jensen, A. Christensen, T. Andersen, E.W., Ege, M. Thorsen, A.V., Tetens, I. et al. 2014. Dietary effects Excluded based on inclusion and exclusion criteria:
of introducing school meals based on the New Nordic Diet–a randomised controlled trial in Danish children. The OPUS Nordic diet meal study, data on nutrient intake and fish
School Meal Study. British Journal of Nutrition, 111(11):1967-1976. but no relevant outcome
Bryant, J. Hanson, M. Peebles, C. Davies, L. Inskip, H. Robinson, S. Godfrey, K.M. et al. 2015. Higher oily fish consumption Excluded based on inclusion and exclusion criteria: oily
in late pregnancy is associated with reduced aortic stiffness in the child at age 9 years. Circulation research, fish consumption and aortic stiffness; wrong outcome
116(7):1202-1205.
Buck, G.M. Tee, G.P. Fitzgerald, E. F. Vena, J. E. Weiner, J.M. Swanson, M. & Msall, M.E. 2003. Maternal fish consumption Excluded based on inclusion and exclusion criteria:
and infant birth size and gestation: New York State Angler Cohort Study. Environmental Health, 2(1):1-9. retrospective fish consumption linked to cross-sectional
infant data
Butler, L.J. Janulewicz, P.A. Carwile, J.L. White, R.F. Winter, M.R. & Aschengrau, A. 2017. Childhood and adolescent fish Excluded based on inclusion and exclusion criteria:
consumption and adult neuropsychological performance: An analysis from the Cape Cod Health Study. Neurotoxicology retrospective cohort
and teratology, 61:47-57.
Daniels, J.L. Longnecker, M.P. Rowland, A.S. Golding, J. & ALSPAC Study Team – University of Bristol Institute of Child Excluded based on inclusion and exclusion criteria:
Health. 2004. Fish intake during pregnancy and early cognitive development of offspring. Epidemiology, 394-402. mercury and wrong outcome (cognitive development)
Emmett, P. M. Jones, L. R. & Golding, J. 2015. Pregnancy diet and associated outcomes in the Avon Longitudinal Study of Excluded based on inclusion and exclusion criteria:
Parents and Children. Nutrition reviews, 73(suppl_3):154-174. supplement article, narrative review
Fereidooni, B. & Jenabi, E. 2014. The use of omega 3 on pregnancy outcomes: a single-center study. J Pak Med Assoc, Excluded based on inclusion and exclusion criteria: n-3
64(12):1363-5. use and effects on pregnancy outcome
Larsen, S.C. Ängquist, L. Laurin, C. Morgen, C.S. Jakobsen, M.U. Paternoster, L. Nohr, E.A. et al. 2016. Association between Excluded based on inclusion and exclusion criteria: no
maternal fish consumption and gestational weight gain: influence of molecular genetic predisposition to obesity. PloS data on children (maternal fish intake and weight gain
One, 11(3):e0150105. in pregnancy)
Maslova, E. Hansen, S. Strøm, M. Halldorsson, T.I. Grunnet, L.G. Vaag, A.A. & Olsen, S.F. 2018. Fish Intake in Pregnancy Excluded based on inclusion and exclusion criteria:
and Offspring Metabolic Parameters at Age 9–16 – Does Gestational Diabetes Modify the Risk? Nutrients, 10(10):1534. case-control study
Rylander, L. Strömberg, U. & Hagmar, L. 1995. Decreased birthweight among infants born to women with a high dietary Excluded based on inclusion and exclusion criteria:
intake of fish contaminated with persistent organochlorine compounds. Scandinavian journal of work, environment & retrospective cohort study
health, 368-375.
Rylander, L. Strömberg, U. & Hagmar, L. 2000. Lowered birth weight among infants born to women with a high intake of Excluded based on inclusion and exclusion criteria:
fish contaminated with persistent organochlorine compounds. Chemosphere, 40(9-11):1255-1262. retrospective cohort study
Amezcua-Prieto, C. Martínez-Galiano, J. M. Salcedo-Bellido, I. Olmedo-Requena, R. Bueno-Cavanillas, A. & Delgado- Excluded for further assessment, as the primary study
Rodríguez, M. 2018. Maternal seafood intake and the risk of small for gestational age newborns: a case–control study in had already been assessed in VKM 2022
Spanish women. BMJ Open, 8(8):e020424.
Brantsæter, A.L. Birgisdottir, B.E. Meltzer, H.M. Kvalem, H.E. Alexander, J. Magnus, P. & Haugen, M. 2012. Maternal Excluded for further assessment, as the primary study
seafood consumption and infant birth weight, length and head circumference in the Norwegian Mother and Child Cohort had already been assessed in VKM 2022
Study. British Journal of Nutrition, 107(3):436-444.
Brantsæter, A.L. Englund-Ögge, L. Haugen, M. Birgisdottir, B.E. Knutsen, H.K. Sengpiel, V. Meltzer, H.M. et al. 2017. Excluded for further assessment, as the primary study
Maternal intake of seafood and supplementary long chain n-3 poly-unsaturated fatty acids and preterm delivery. BMC had already been assessed in VKM 2022
pregnancy and childbirth, 17(1):1-15.
Drouillet, P. Kaminski, M. De Lauzon‐Guillain, B. Forhan, A. Ducimetière, P. Schweitzer, M. Charles, M.A. 2009. Association Excluded for further assessment, as the primary study
between maternal seafood consumption before pregnancy and fetal growth: evidence for an association in overweight had already been assessed in VKM 2022
women. The EDEN mother‐child cohort. Paediatric and perinatal epidemiology, 23(1):76-86.
Guldner, L. Monfort, C. Rouget, F. Garlantezec, R. & Cordier, S. 2007. Maternal fish and shellfish intake and pregnancy Excluded for further assessment, as the primary study
outcomes: a prospective cohort study in Brittany, France. Environmental Health, 6, 1-8. had already been assessed in VKM 2022
Halldorsson, T.I. Meltzer, H.M. Thorsdottir, I. Knudsen, V. & Olsen, S.F. 2007. Is high consumption of fatty fish during Excluded for further assessment, as the primary study
pregnancy a risk factor for fetal growth retardation? A study of 44,824 Danish pregnant women. American journal of had already been assessed in VKM 2022
epidemiology, 166(6):687-696.
Heppe, D.H. Steegers, E.A. Timmermans, S. den Breeijen, H. Tiemeier, H. Hofman, A. & Jaddoe, V. W. 2011. Maternal fish Excluded for further assessment, as the primary study
consumption, fetal growth and the risks of neonatal complications: the Generation R Study. British journal of nutrition, had already been assessed in VKM 2022
105(6):938-949.
Guldner, L. Monfort, C. Rouget, F. Garlantezec, R. & Cordier, S. 2007. Maternal fish and shellfish intake and pregnancy Excluded for further assessment, as the primary study
outcomes: a prospective cohort study in Brittany, France. Environmental Health, 6, 1-8. had already been assessed in VKM 2022

344
APPENDICES

TABLE A3.5 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “BIRTH AND GROWTH” (cont.)
Klebanoff, M.A. Harper, M. Lai, Y. Thorp Jr, J. Sorokin, Y. Varner, M.W. Anderson, G.D. 2011. Fish consumption, erythrocyte Excluded for further assessment, as the primary study
fatty acids, and preterm birth. Obstetrics and gynecology, 117(5):1071. had already been assessed in VKM 2022
Leventakou, V. Roumeliotaki, T. Martinez, D. Barros, H. Brantsaeter, A.L. Casas, M. Chatzi, L. et al. 2014. Fish intake Excluded for further assessment, as the primary study
during pregnancy, fetal growth, and gestational length in 19 European birth cohort studies. The American journal of had already been assessed in VKM 2022
clinical nutrition, 99(3):506-516.
Mendez, M.A. Plana, E. Guxens, M. Morillo, C.M.F. Albareda, R.M. Garcia-Esteban, R. Sunyer, J. et al. 2010 Seafood Excluded for further assessment, as the primary study
consumption in pregnancy and infant size at birth: results from a prospective Spanish cohort. Journal of Epidemiology & had already been assessed in VKM 2022
Community Health, 64(3):216-222.
Mohanty, A.F. Thompson, M.L. Burbacher, T.M. Siscovick, D.S. Williams, M.A. & Enquobahrie, D.A. 2015. Periconceptional Excluded for further assessment, as the primary study
seafood intake and fetal growth. Paediatric and perinatal epidemiology, 29(5):376-387. had already been assessed in VKM 2022
Muthayya, S. Dwarkanath, P. Thomas, T. Ramprakash, S. Mehra, R. Mhaskar, A. Kurpad, A. et al. 2009. The effect of fish Excluded for further assessment, as the primary study
and ω-3 LCPUFA intake on low birth weight in Indian pregnant women. European journal of clinical nutrition, 63(3):340- had already been assessed in VKM 2022
346.
Nykjaer, C. Higgs, C. Greenwood, D.C. Simpson, N.A. Cade, J.E. & Alwan, N.A. 2019. Maternal fatty fish intake prior to and Excluded for further assessment, as the primary study
during pregnancy and risks of adverse birth outcomes: findings from a British Cohort. Nutrients, 11(3):643. had already been assessed in VKM 2022
Olsen, S. F. 2002. Low consumption of seafood in early pregnancy as a risk factor for preterm delivery: prospective cohort Excluded for further assessment, as the primary study
study. Bmj, 324(7335):447. had already been assessed in VKM 2022
Ramon, R. Ballester, F. Aguinagalde, X. Amurrio, A. Vioque, J. Lacasana, M. Iniguez, C. et al. 2009. Fish consumption Excluded for further assessment, as the primary study
during pregnancy, prenatal mercury exposure, and anthropometric measures at birth in a prospective mother-infant had already been assessed in VKM 2022
cohort study in Spain. The American journal of clinical nutrition, 90(4):1047-1055.
Rogers, I. Emmett, P. Ness, A. & Golding, J. 2004. Maternal fish intake in late pregnancy and the frequency of low birth Excluded for further assessment, as the primary study
weight and intrauterine growth retardation in a cohort of British infants. Journal of Epidemiology & Community Health, had already been assessed in VKM 2022
58(6):486-492.
Smid, M.C. Stuebe, A.M. Manuck, T.A. & Sen, S. 2019. Maternal obesity, fish intake, and recurrent spontaneous preterm Excluded for further assessment, as the primary study
birth. The Journal of Maternal-Fetal & Neonatal Medicine, 32(15):2486-2492. had already been assessed in VKM 2022
Stratakis, N. Roumeliotaki, T. Oken, E. Barros, H. Basterrechea, M. Charles, M.A. Chatzi, L. et al. 2016. Fish intake in Excluded for further assessment, as the primary study
pregnancy and child growth: a pooled analysis of 15 European and US birth cohorts. JAMA pediatrics, 170(4):381-390. had already been assessed in VKM 2022
van Den Berg, S.W. Wijga, A.H. van Rossem, L. Gehring, U. Koppelman, G.H. Smit, H.A. & Boer, J.M. 2016. Maternal fish Excluded for further assessment, as the primary study
consumption during pregnancy and BMI in children from birth up to age 14 years: the PIAMA cohort study. European had already been assessed in VKM 2022
journal of nutrition, 55, 799-808.

BONE HEALTH
TABLE A3.6 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “BONE HEALTH”
Study (n = 7) Reason for exclusion
Abdelhamid, A. Hooper, L. Sivakaran, R. Hayhoe, R. P. & Welch, A. 2019. The relationship between omega-3, omega-6 and Excluded based on inclusion and exclusion criteria:
total polyunsaturated fat and musculoskeletal health and functional status in adults: a systematic review and meta- Studies included in the review investigated dietary
analysis of RCTs. Calcified tissue international, 105(4):353-372. supplements (two studies included nuts), none with fish.
All studies used dietary supplements.
Salari, P. Rezaie, A. Larijani, B. & Abdollahi, M. 2008. A systematic review of the impact of n-3 fatty acids in bone Excluded based on inclusion and exclusion criteria:
health and osteoporosis. Medical science monitor: International medical journal of experimental and clinical research, Studies included in the review only investigated intake
14(3):RA37-44. of fats, no mention of fish.
Sadeghi, O. Djafarian, K. Ghorabi, S. Khodadost, M. Nasiri, M. & Shab-Bidar, S. 2019. Dietary intake of fish, n-3 Excluded, as the review had already been assessed in
polyunsaturated fatty acids and risk of hip fracture: A systematic review and meta-analysis on observational studies. VKM 2022.
Critical reviews in food science and nutrition, 59(8):1320-1333.
Perna, S. Avanzato, I. Nichetti, M. D’Antona, G. Negro, M. & Rondanelli, M. 2017. Association between dietary patterns of Excluded, as the review had already been assessed in
meat and fish consumption with bone mineral density or fracture risk: a systematic literature. Nutrients, 9(9):1029. VKM 2022.
Albertazzi, P. & Coupland, K. 2002. Polyunsaturated fatty acids. Is there a role in postmenopausal osteoporosis Excluded based on inclusion and exclusion criteria: not a
prevention?. Maturitas, 42(1):13-22. systematic review
Molfino, A. Gioia, G. Fanelli, F.R. & Muscaritoli, M. 2014. The role for dietary omega-3 fatty acids supplementation in older Excluded based on inclusion and exclusion criteria: not a
adults. Nutrients, 6(10):4058-4072. systematic review
Pampaloni, B. Quattrini, S. & Brandi, M.L. 2018. The Mediterranean diet for bone health in osteoporosis. Children and Excluded based on inclusion and exclusion criteria: not a
adolescents. Clinical Cases in Mineral & Bone Metabolism, 15(1). systematic review

345
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.7 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “BONE HEALTH”
Study (n = 26) Reason for exclusion
Alveblom, A. K. et al. Incidence of hospitalized osteoporotic fractures in cohorts with high dietary intake of persistent Excluded based on inclusion and exclusion criteria:
organochlorine compounds. Int Arch Occup Environ Health, 76(3):246-248. retrospective cohort study
Langsetmo, L. et al. Dietary patterns in men and women are simultaneously determinants of altered glucose metabolism Excluded based on inclusion and exclusion criteria:
and bone metabolism. Nutr Res, 36(4):328-336. dietary patterns as exposure, not fish alone
Macdonald, H. M. et al. Vitamin D status in postmenopausal women living at higher latitudes in the UK in relation to bone Excluded based on inclusion and exclusion criteria: only
health, overweight, sunlight exposure and dietary vitamin D. Bone, 42(5):996-1003. vitamin d status, not fish as exposure alone
Mangano, K. M. et al. Dietary protein is associated with musculoskeletal health independently of dietary pattern: the Excluded based on inclusion and exclusion criteria:
Framingham Third Generation Study. Am J Clin Nutr, 105(3):714-722. protein intake, not fish as exposure alone
Melaku, Y. A. et al. Association between dietary patterns and low bone mineral density among adults aged 50 years and Excluded based on inclusion and exclusion criteria:
above: findings from the North West Adelaide Health Study (NWAHS). Br J Nutr, 116(8):1437-1446. cross-sectional study
Meyer, H. E. et al. Dietary factors and the incidence of hip fracture in middle-aged Norwegians - A prospective study. Excluded based on inclusion and exclusion criteria:
American Journal of Epidemiology, 145(2):117-123. protein intake, not fish as exposure alone
Paunescu, A. C. et al. 2013. Polyunsaturated fatty acids and calcaneal ultrasound parameters among Inuit women from Excluded based on inclusion and exclusion criteria: n-3
Nuuk (Greenland): a longitudinal study. Int J Circumpolar Health, 72:20988. and n-6, not fish as exposure alone
Rogers, T. S. et al. Dietary patterns and longitudinal change in hip bone mineral density among older men. Osteoporos Int, Excluded based on inclusion and exclusion criteria:
29(5):1135-1145. dietary patterns as exposure, not fish alone
Rosendahl-Riise, H. et al. Total and lean fish intake is positively associated with bone mineral density in older women in Excluded based on inclusion and exclusion criteria:
the community-based Hordaland Health Study. Eur J Nutr, 58(4):1403-1413. cross-sectional study
Shaw, S. C. et al. Diet Quality and Bone Measurements Using HRpQCT and pQCT in Older Community-Dwelling Adults from Excluded based on inclusion and exclusion criteria: diet
the Hertfordshire Cohort Study. Calcif Tissue Int,103(5):494-500. quality as exposure, not fish alone
Umaretiya, P. J. et al. Bone mineral density in Nigerian children after discontinuation of calcium supplementation. Bone, Excluded based on inclusion and exclusion criteria:
55(1):64-68. Calcium supplements, not fish
van den Hooven, E. H. et al. Identification of a dietary pattern prospectively associated with bone mass in Australian Excluded based on inclusion and exclusion criteria:
young adults. Am J Clin Nutr, 102(5):1035-1043. dietary patterns as exposure, not fish alone
Vatanparast, H. et al. Positive effects of vegetable and fruit consumption and calcium intake on bone mineral accrual in Excluded based on inclusion and exclusion criteria: fish
boys during growth from childhood to adolescence: the University of Saskatchewan Pediatric Bone Mineral Accrual Study. was not exposure variable
Am J Clin Nutr, 82(3):700-706.
Wallin, E. et al. Exposure to CB-153 and p,p'-DDE and bone mineral density and bone metabolism markers in middle- Excluded based on inclusion and exclusion criteria:
aged and elderly men and women. Osteoporos Int, 16(12):2085-2094. retrospective cohort study
Wallin, E. et al. Exposure to persistent organochlorine compounds through fish consumption and the incidence of Excluded based on inclusion and exclusion criteria:
osteoporotic fractures. Scand J Work Environ Health, 30(1):30-35. retrospective cohort study
Whiting, S. J. et al. Factors that affect bone mineral accrual in the adolescent growth spurt. J Nutr, 134(3):696s-700s. Excluded based on inclusion and exclusion criteria: no
fish as exposure
Wu, F. et al. Associations of dietary patterns with bone mass, muscle strength and balance in a cohort of Australian Excluded based on inclusion and exclusion criteria:
middle-aged women. Br J Nutr, 118(8):598-606. cross-sectional study
Appleby, P. et al. Comparative fracture risk in vegetarians and nonvegetarians in EPIC-Oxford. Eur J Clin Nutr, Excluded for further assessment, as the primary study
61(12):1400-1406. had already been assessed in VKM 2022
Chan, R. et al. Effects of food groups and dietary nutrients on bone loss in elderly Chinese population. J Nutr Health Excluded for further assessment, as the primary study
Aging, 15(4):287-294. had already been assessed in VKM 2022
Farina, E. K. et al. Dietary intakes of arachidonic acid and alpha-linolenic acid are associated with reduced risk of hip Excluded for further assessment, as the primary study
fracture in older adults. J Nutr, 141(6):1146-1153. had already been assessed in VKM 2022
Farina, E. K. et al. Protective effects of fish intake and interactive effects of long-chain polyunsaturated fatty acid intakes Excluded for further assessment, as the primary study
on hip bone mineral density in older adults: the Framingham Osteoporosis Study. Am J Clin Nutr, 93(5):1142-1151. had already been assessed in VKM 2022
Virtanen, J. K. et al. Dietary intake of polyunsaturated fatty acids and risk of hip fracture in men and women. Osteoporos Excluded for further assessment, as the primary study
Int, 23(11):2615-2624. had already been assessed in VKM 2022
Virtanen, J. K. et al. Fish consumption, bone mineral density, and risk of hip fracture among older adults: the Excluded for further assessment, as the primary study
cardiovascular health study. J Bone Miner Res, 25(9):1972-1979. had already been assessed in VKM 2022
Erkkilä, A. T. et al. Associations of Baltic Sea and Mediterranean dietary patterns with bone mineral density in elderly Excluded for further assessment, as the primary study
women. Public Health Nutr, 20(15):2735-2743. had already been assessed in a systematic review that
was included in VKM 2022
Feskanich, D. et al. Calcium, vitamin D, milk consumption, and hip fractures: a prospective study among postmenopausal Excluded for further assessment, as the primary study
women. Am J Clin Nutr, 77(2):504-511. had already been assessed in a systematic review that
was included in VKM 2022
Longo, A.B. & Ward, W.E. PUFAs, Bone Mineral Density, and Fragility Fracture: Findings from Human Studies. Adv Nutr, Excluded for further assessment, as the primary study
7(2): 299-312. had already been assessed in a systematic review that
was included in VKM 2022

346
APPENDICES

CANCER
TABLE A3.8 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CANCER”, BASED ON INCLUSION
AND EXCLUSION CRITERIA
Study (n = 17) Reason for exclusion
Wilson, K.M. & Mucci, L.A. 2019. Diet and Lifestyle in Prostate Cancer. Adv Exp Med Biol, 1210:1-27. Excluded based on inclusion and exclusion criteria: book
chapter, wrong publication type
Yoo, J.Y. Cho, H.J. Moon, S. Choi, J. Lee, S. Ahn, C. Yoo, K.Y. Kim, I. Ko, K.P. Lee, J.E. & Park, S.K. 2020. Pickled Vegetable Excluded based on inclusion and exclusion criteria:
and Salted Fish Intake and the Risk of Gastric Cancer: Two Prospective Cohort Studies and a Meta-Analysis. Cancers narrative review with meta-analysis
(Basel), 12.
Hu, S. Yu, J. Wang, Y. Li, Y. Chen, H. Shi, Y. & Ma, X. 2019. Fish consumption could reduce the risk of oral cancer in Excluded based on inclusion and exclusion criteria: 2
Europeans: A meta-analysis. Arch Oral Biol, 107:104494. cohort studies before 2018, and the rest was 13 case-
control studies
Jiang, W. Wang, M. Jiang, H.Z. Chen, G.C. and Hua, Y.F. 2019. Meta-analysis of fish consumption and risk of pancreatic Excluded based on inclusion and exclusion criteria: only
cancer in 13 prospective studies with 1.8 million participants. PLoS One, 14:e0222139. one study included after 2018, and this is also included
in the search of primary studies
Yang, L. Shi, W.Y. Xu, X.H. Wang, X.F. Zhou, L. & Wu, D.P. 2020. Fish consumption and risk of non-Hodgkin lymphoma: A Excluded based on inclusion and exclusion criteria: all
meta-analysis of observational studies. Hematology, 25:194-202. studies included are older than 2018
Zhang, Z.H. & Xin, J.Z. 2019. Dietary fresh fish and processed fish intake and the risk of glioma: A meta-analysis of Excluded based on inclusion and exclusion criteria: all
observational studies. Cellular and Molecular Biology, 65:48-53. studies included are older than 2018
Chapelle, N. Martel, M. Toes-Zoutendijk, E. Barkun, A.N. & Bardou, M. 2020. Recent advances in clinical practice: Excluded based on inclusion and exclusion criteria: all
colorectal cancer chemoprevention in the average-risk population. Gut, 69:2244-2255. relevant studies included are older than 2018
Fakhri, G. Al Assaad, M. & Tfayli, A. 2020. Association of various dietary habits and risk of lung cancer: an updated Excluded based on inclusion and exclusion criteria:
comprehensive literature review. Tumori, 106:445-456. all studies regarding fish intake and cancer are older
than 2018
Ghaffari, H.R. Yunesian, M. Nabizadeh, R. Nasseri, S. Sadjadi, A. Pourfarzi, F. Poustchi, H. & Eshraghian, A. 2019. Excluded based on inclusion and exclusion criteria:
Environmental etiology of gastric cancer in Iran: a systematic review focusing on drinking water, soil, food, radiation, and all studies regarding fish intake and cancer are older
geographical conditions. Environ Sci Pollut Res Int, 26:10487-10495. than 2018
Lei, H.C. To, C.H. & Lei, U.P. 2020. Association between fish intake and glioma risk: a systematic review and meta- Excluded based on inclusion and exclusion criteria: all
analysis. Journal of International Medical Research, 48. studies included are older than 2018
Li, N. Wu, X.T. Zhuang, W. Xia, L. Chen, Y. Wu, C.C. Rao, Z.Y. Du, L. Zhao, R. Yi, M.S. et al. 2020. Fish consumption and Excluded based on inclusion and exclusion criteria: only
multiple health outcomes: Umbrella review. Trends in Food Science & Technology, 99:273-283. one study after 2018, and this is also included in the
review Jayedi, 2020
Lv, D. Wang, R. Chen, M. Li, Y. & Cao, C. 2021. Fish Intake, Dietary Polyunsaturated Fatty Acids, and Lung Cancer: Excluded based on inclusion and exclusion criteria: main
Systematic Review and Dose-Response Meta-Analysis of 1.7 Million Men and Women. Nutr Cancer, 74(6): 1976-1985 focus on n-3 PUFAs, and fish-intake studies are older
than 2018
Okekpa, S.I. RB, S.M.N.M. Mangantig, E. Azmi, N.S.A. Zahari, S.N.S. Kaur, G. & Musa, Y. 2019. Nasopharyngeal Carcinoma Excluded based on inclusion and exclusion criteria: all
(NPC) Risk Factors: A Systematic Review and Meta-Analysis of the Association with Lifestyle, Diets, Socioeconomic and studies included are older than 2018 and only case-
Sociodemographic in Asian Region. Asian Pac J Cancer Prev, 20:3505-3514. control studies
Poorolajal, J. Moradi, L. Mohammadi, Y. Cheraghi, Z. & Gohari-Ensaf, F. 2020. Risk factors for stomach cancer: a Excluded based on inclusion and exclusion criteria:
systematic review and meta-analysis. Epidemiol Health, 42:e2020004. all studies regarding fish intake and cancer are older
than 2018
Sergentanis, T.N. Ntanasis-Stathopoulos, I. Tzanninis, I.G. Gavriatopoulou, M. Sergentanis, I.N. Dimopoulos, M.A. & Excluded based on inclusion and exclusion criteria: all
Psaltopoulou, T. 2019. Meat, fish, dairy products and risk of hematological malignancies in adults - a systematic review studies included are older than 2018
and meta-analysis of prospective studies. Leuk Lymphoma, 60:1978-1990.
Ubago-Guisado, E. Rodríguez-Barranco, M. Ching-López, A. Petrova, D. Molina-Montes, E. Amiano, P. Barricarte-Gurrea, A. Excluded based on inclusion and exclusion criteria: only
Chirlaque, M.D. Agudo, A. & Sánchez, M.J. 2021. Evidence Update on the Relationship between Diet and the Most Common one study included after 2018, and this is also included
Cancers from the European Prospective Investigation into Cancer and Nutrition (EPIC) Study: A Systematic Review. in our search of primary studies (Aglago, 2020)
Nutrients, 13 (10): 3582.
Cao, C. & Xu, N. 2019. Fish Intake, Dietary Polyunsaturated Fatty Acids, and Lung Cancer: Systematic Review and Excluded based on inclusion and exclusion criteria:
Dose-Response Meta-Analysis of 1.7 Million Men and Women. American Journal of Respiratory and Critical Care Medicine, conference abstract
199:A7277.
Lee, K.H. Seong, H.J. Kim, G. Jeong, G.H. Kim, J.Y. Park, H. Jung, E. Kronbichler, A. Eisenhut, M. Stubbs, B. et al. 2020. Excluded based on inclusion and exclusion criteria: all
Consumption of Fish and ω-3 Fatty Acids and Cancer Risk: An Umbrella Review of Meta-Analyses of Observational relevant studies included are older than 2018
Studies. Adv Nutr, 11: 1134-1149.

347
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.9 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CANCER”
Study (n = 10) Reason for exclusion
Dydjow-Bendek, D. & Zagoźdźon, P. 2020. Total Dietary Fats, Fatty Acids, and Omega-3/Omega-6 Ratio as Risk Factors of Excluded based on inclusion and exclusion criteria:
Breast Cancer in the Polish Population - a Case-Control Study. In Vivo, 34:423-431. case-control study
Golozar, A. Etemadi, A. Kamangar, F. Fazeltabar Malekshah, A. Islami, F. Nasrollahzadeh, D. Abedi-Ardekani, B. Khoshnia, Excluded based on inclusion and exclusion criteria:
M. Pourshams, A. Semnani, S. et al. 2018. Food preparation methods, drinking water source, and esophageal squamous case-control study
cell carcinoma in the high-risk area of Golestan, Northeast Iran. Eur J Cancer Prev, 25, 123-129.
Gonzalez, C.A. 2022. The European Prospective Investigation into Cancer and Nutrition (EPIC). Public Health Nutr, 9, Excluded based on inclusion and exclusion criteria:
124-126. wrong study design/article type
Liu, Z. Luo, Y. Ren, J. Yang, L. Li, J. Wei, Z. He, Y. Wang, J. Li, R. He, L. et al. 2022. Association between fish oil Excluded based on inclusion and exclusion criteria:
supplementation and cancer risk according to fatty fish consumption: A large prospective population-based cohort study main aim to investigate fish oil supplementation and
using UK Biobank. Int J Cancer,150 (4): 562-571 cancer risk
Oh, C.C. Jin, A.Z. Yuan, J.M. & Koh, W.P. 2020. Fish intake and risk of nonmelanoma skin cancer in a Chinese population: Excluded based on inclusion and exclusion criteria: not a
the Singapore Chinese Health Study. Clin Exp Dermatol, 45:461-463. research article, only a correspondence in the journal
McClain, K.M. Bradshaw, P.T. Khankari, N.K. Gammon, M.D. & Olshan, A.F. 2019. Fish/shellfish intake and the risk of head Excluded based on inclusion and exclusion criteria:
and neck cancer. Eur J Cancer Prev, 28:102-108. case-control study
Rada-Fern, ez de Jauregui, D. Evans, C.E.L. Jones, P. Greenwood, D.C. Hancock, N. & Cade, J.E. 2018. Common dietary Excluded based on inclusion and exclusion criteria:
patterns and risk of cancers of the colon and rectum: Analysis from the United Kingdom Women's Cohort Study (UKWCS). dietary patterns
Int J Cancer, 143:773-781.
Wang, Y. Jacobs, E.J. Shah, R.A. Stevens, V.L. Gansler, T. & McCullough, M.L. 2020. Red and Processed Meat, Poultry, Fish, Excluded based on inclusion and exclusion criteria:
and Egg Intakes and Cause-Specific and All-Cause Mortality among Men with Nonmetastatic Prostate Cancer in a U.S. outcome is mortality/survival
Cohort. Cancer Epidemiol Biomarkers Prev, 29:1029-1038.
Marcondes, L.H. Franco, O.H. Ruiter, R. Ikram, M.A. Mulder, M. Stricker, B.H. & Kiefte-de Jong, J.C. 2019. Animal foods and Excluded for further assessment, as the primary study
postmenopausal breast cancer risk: a prospective cohort study. Br J Nutr, 122:583-591. had already been assessed in the included systematic
review Kazemi et al., 2021.
McCullough, M.L. Jacobs, E.J. Shah, R. Campbell, P.T. Wang, Y. Hartman, T.J. & Gapstur, S.M. 2018. Meat consumption and Excluded for further assessment, as the primary study
pancreatic cancer risk among men and women in the Cancer Prevention Study-II Nutrition Cohort. Cancer Causes Control, had already been assessed in the included systematic
29:125-133. review Gao et al., 2022.

348
APPENDICES

CARDIOVASCULAR DISEASES AND OUTCOMES

TABLE A3.10 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES AND
OUTCOMES”, BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 13) Reason for exclusion
Alhassan, A. et al. 2017, Consumption of fish and vascular risk factors: A systematic review and meta-analysis of Excluded based on inclusion and exclusion criteria: only
intervention studies. Atherosclerosis, 266:87-94. measured biomarkers as outcome
Bechthold, A. et al. 2019. Food groups and risk of coronary heart disease, stroke and heart failure: A systematic review Excluded, as the review has already been assessed in
and dose-response meta-analysis of prospective studies. Critical Reviews in Food Science and Nutrition, 59(7):1071- VKM 2022.
1090.
Chen, C. et al. 2021. Fish consumption, long-chain omega-3 fatty acids intake and risk of stroke: An updated systematic Excluded, as the review has already been assessed in
review and meta-analysis. Asia Pacific Journal of Clinical Nutrition, 30(1):140-152. VKM 2022.
Djousse, L. et al. 2012, Fish consumption, omega-3 fatty acids and risk of heart failure: A meta-analysis. Clinical Excluded based on inclusion and exclusion criteria: not a
Nutrition, 31(6):846-853. systematic review, only meta-analysis.
Jayedi, A. & Shab-Bidar, S. 2020. Fish Consumption and the Risk of Chronic Disease: An Umbrella Review of Meta- Excluded, as the review has already been assessed in
Analyses of Prospective Cohort Studies. Advances in Nutrition, 11(5):1123-1133. VKM 2022.
Jayedi, A. et al. 2018. Fish consumption and risk of all-cause and cardiovascular mortality: a dose-response meta- Excluded, as the review has already been assessed in
analysis of prospective observational studies. Public Health Nutrition, 21(7):1297-1306. VKM 2022.
Jayedi, A. et al. 2021. Fish consumption and the risk of cardiovascular disease and mortality in patients with type 2 Excluded, as the review had already been assessed in
diabetes: a dose-response meta-analysis of prospective cohort studies. Critical Reviews in Food Science and Nutrition, VKM 2022.
61(10):1640-1650.
Jayedi, A. M.S. Zargar & S. Shab-Bidar, 2019. Fish consumption and risk of myocardial infarction: a systematic review Excluded, as the review had already been assessed in
and dose-response meta analysis suggests a regional difference. Nutrition Research, 62:1-12. VKM 2022.
Krittanawong, C. et al. 2021. Fish Consumption and Cardiovascular Health: A Systematic Review. American Journal of Excluded, as the review had already been assessed in
Medicine. 134(6): p. 713-720. VKM 2022.
Kwok, C.S. et al. 2019. Dietary components and risk of cardiovascular disease and all-cause mortality: a review of Excluded, as the review had already been assessed in
evidence from meta-analyses. European Journal of Preventive Cardiology, 26(13):1415-1429. VKM 2022.
Mattiuzzi, C. et al. 2016. Fish Intake and Venous Thromboembolism: A Systematic Literature Review. Clinical and Applied Excluded, as the review had already been assessed in
Thrombosis-Hemostasis, 22(4):309-313. VKM 2022.
Ness, A.R. et al. 1999. The long-term effect of advice to eat more fish on blood pressure in men with coronary disease: Excluded based on inclusion and exclusion criteria: not a
results from the Diet and Reinfarction Trial. Journal of Human Hypertension, 13(11):729-733. systematic review.
Wang, C.C. et al. 2006, n-3 fatty acids from fish or fish-oil supplements, but not alpha-linolenic acid, benefit Excluded based on inclusion and exclusion criteria: not
cardiovascular disease outcomes in primary- and secondary-prevention studies: a systematic review. American Journal of measured fish consumption, only n-3 biomarkers
Clinical Nutrition, 84(1):5-17.

349
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.11 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES
AND OUTCOMES”
Study (n = 139) Reason for exclusion
Amoah, J. et al. 2021. Effects of a school-based intervention to reduce cardiovascular disease risk factors among Excluded based on inclusion and exclusion criteria: fish
secondary school students: A cluster- randomized, controlled trial. PLoS One, 16(11):e0259581. consumption not included
Archer, S.L. et al. 1998. Association of dietary fish and n-3 fatty acid intake with hemostatic factors in the coronary artery Excluded based on inclusion and exclusion criteria:
risk development in young adults (CARDIA) study. Arterioscler Thromb Vasc Biol, 18(7): 1119-23. intermediate CVD biomarkers
Borgi, L. et al. 2015. Long-term intake of animal flesh and risk of developing hypertension in three prospective cohort Excluded based on inclusion and exclusion criteria:
studies. J Hypertens, 33(11): 2231-8. not CVD
Bravata, D.M. et al. 2007. Dietary fish or seafood consumption is not related to cerebrovascular disease risk in twin Excluded based on inclusion and exclusion criteria:
veterans. Neuroepidemiology, 28(3):186-190. information of fish consumption not given properly
Burke, V. et al. 2007. A lifestyle program for treated hypertensives improved health-related behaviors and cardiovascular Excluded based on inclusion and exclusion criteria
risk factors, a randomized controlled trial. J Clin Epidemiol, 60(2):133-41.
Burr, M.L. 2007. Secondary prevention of CHD in UK men: the Diet and Reinfarction Trial and its sequel. Proc Nutr Soc, Excluded based on inclusion and exclusion criteria: not
66(1):9-15. relevant
Chrysohoou, C. et al. 2007. Long-term fish consumption is associated with protection against arrhythmia in healthy Excluded based on inclusion and exclusion criteria: the
persons in a Mediterranean region - The ATTICA study. American Journal of Clinical Nutrition, 85(5):1385-1391. design was cross-sectional (exclusion criteria)
Damsgaard, C.T. et al. 2016. Effects of oily fish intake on cardiovascular risk markers, cognitive function, and behavior in Excluded based on inclusion and exclusion criteria:
school-aged children: study protocol for a randomized controlled trial. Trials, 17(1):510. study protocol
De Lorgeril, M. & Salen, P. 2002. Fish and N-3 fatty acids for the prevention and treatment of coronary heart disease: Excluded based on inclusion and exclusion criteria:
Nutrition is not pharmacology. American Journal of Medicine, 112(4):316-319. narrative review/opinion
Engell, R.E. et al. 2013. Seafood omega-3 intake and risk of coronary heart disease death: an updated meta-analysis with Excluded based on inclusion and exclusion criteria: not
implications for attributable burden. Lancet, 381:45-45. relevant
Eshak, E.S. et al. 2014. Modification of the excess risk of coronary heart disease due to smoking by seafood/fish intake. Excluded based on inclusion and exclusion criteria:
Am J Epidemiol, 179(10):1173-81. exposure focused on smoking
Gerhard, G.T. et al. 1991. Comparison of three species of dietary fish: effects on serum concentrations of low-density- Excluded based on inclusion and exclusion criteria: only
lipoprotein cholesterol and apolipoprotein in normotriglyceridemic subjects. Am J Clin Nutr, 54(2):334-9. markers, not CVD
Guasch-Ferre, M. et al. 2019. Meta-Analysis of Randomized Controlled Trials of Red Meat Consumption in Comparison Excluded based on inclusion and exclusion criteria:
With Various Comparison Diets on Cardiovascular Risk Factors. Circulation, 139(15):1828-1845. systematic review
Gunnarsdottir, I. et al. 2008. Inclusion of fish or fish oil in weight-loss diets for young adults: effects on blood lipids. Int J Excluded based on inclusion and exclusion criteria:
Obes (Lond), 32(7):1105-12. wrong outcome
Hallgren, C.G. et al. 2001. Markers of high fish intake are associated with decreased risk of a first myocardial infarction. Excluded based on inclusion and exclusion criteria:
Br J Nutr, 86(3):397-404. case-control study
Hallund, J. et al. 2010. The effect of farmed trout on cardiovascular risk markers in healthy men. Br J Nutr, 104(10):1528- Excluded based on inclusion and exclusion criteria:
36. wrong outcome
He, K. et al. 2004. Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-analysis of Excluded based on inclusion and exclusion criteria
cohort studies. Circulation, 109(22):2705-11.
He, K. et al. 2004. Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-analysis of Excluded based on inclusion and exclusion criteria
cohort studies. Circulation, 109(22):2705-11.
He, K. et al. 2009. Associations of dietary long-chain n-3 polyunsaturated fatty acids and fish with biomarkers Excluded based on inclusion and exclusion criteria: not
of inflammation and endothelial activation (from the Multi-Ethnic Study of Atherosclerosis [MESA]). Am J CVD, only biomarkers
Cardiol,103(9):1238-43.
He, K. et al. 2004. Fish consumption and incidence of stroke: a meta-analysis of cohort studies. Stroke, 35(7):1538-42. Excluded based on inclusion and exclusion criteria
Hou, L.N. et al. 2012. Fish intake and risk of heart failure: A meta-analysis of five prospective cohort studies. Exp Ther Excluded based on inclusion and exclusion criteria
Med, 4(3):481-486.
Hu, F.B. et al. 1999. Dietary saturated fats and their food sources in relation to the risk of coronary heart disease in Excluded based on inclusion and exclusion criteria
women. American Journal of Clinical Nutrition, 70(6):1001-1008.
Johansson, A. & Acosta, S. 2020. Diet and Lifestyle as Risk Factors for Carotid Artery Disease: A Prospective Cohort Study. Excluded based on inclusion and exclusion criteria: not
Cerebrovascular Diseases, 2020. 49(5):563-569. relevant
Johnsen, S.H. et al. 2018. Fish consumption, fish oil supplements and risk of atherosclerosis in the Tromso study. Nutr J, Excluded based on inclusion and exclusion criteria:
17(1):56. wrong study design
Kim, S.A. et al. 2019. Oily Fish Consumption and the Risk of Dyslipidemia in Korean Adults: A Prospective Cohort Study Excluded based on inclusion and exclusion criteria: not
Based on the Health Examinees Gem (HEXA-G) Study. Nutrients, 11(10). CVD, only biomarkers
Konig, A. et al. 2005. A quantitative analysis of fish consumption and coronary heart disease mortality. Am J Prev Med, Excluded based on inclusion and exclusion criteria:
29(4):335-46. narrative review (exclusion criteria)
Kris-Etherton, P.M. Harris, W.S. & Appel, L.J. 2002. Fish consumption, fish oil, omega-3 fatty acids, and cardiovascular Excluded based on inclusion and exclusion criteria:
disease. Arterioscler Thromb Vasc Biol, 23(2):e20-30. narrative review (exclusion criteria)
Lamlili, E.N.M. et al. 2016. Fish Consumption Impact on Coronary Heart Disease Mortality in Morocco: A Mathematical Excluded based on inclusion and exclusion criteria: not
Model with Optimal Control. Engineering Letters, 24(3):246-251. relevant

350
APPENDICES

TABLE A3.11 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES
AND OUTCOMES” (cont.)
Lapidus, L. et al. 1986. Dietary habits in relation to incidence of cardiovascular disease and death in women: a 12-year Excluded based on inclusion and exclusion criteria
follow-up of participants in the population study of women in Gothenburg, Sweden. Am J Clin Nutr, 44(4):444-8.
Lee, H.A. & Park, H. 2018. Diet-Related Risk Factors for Incident Hypertension During an 11-Year Follow-Up: The Korean Excluded based on inclusion and exclusion criteria:
Genome Epidemiology Study. Nutrients, 10(8):1077. not CVD
Lentjes, M.A.H. et al. 2016. Total (food and supplement) n-3 PUFA intake is associated with lower Coronary Heart Disease Excluded based on inclusion and exclusion criteria: just
mortality, independently of fish intake. Proceedings of the Nutrition Society, 75(Oce1):E42-E42. an abstract (summary from a winter meeting)
Li, Y.H. et al. 2013. Fish consumption and incidence of heart failure: a meta-analysis of prospective cohort studies. Chin Excluded based on inclusion and exclusion criteria:
Med J (Engl), 126(5):942-8. review article
Lilja, E. et al. 2019. The association between dietary intake, lifestyle and incident symptomatic peripheral arterial disease Excluded based on inclusion and exclusion criteria: not
among individuals with diabetes mellitus: insights from the Malmo Diet and Cancer study. Therapeutic Advances in general population, patients with diabetes
Endocrinology and Metabolism, 10:1-8
Lindqvist, H. et al. 2007. Herring (Clupea harengus) supplemented diet influences risk factors for CVD in overweight Excluded based on inclusion and exclusion criteria
subjects. Eur J Clin Nutr, 61(9):1106-13.
Mark, K. et al. 1998. Eating fish may reduce infarct size and the occurrence of Q wave infarcts. European Journal of Excluded based on inclusion and exclusion criteria:
Clinical Nutrition, 52(1):40-44. not CVD
Masson, S. et al. 2013. Plasma n-3 polyunsaturated fatty acids in chronic heart failure in the GISSI-Heart Failure Trial: Excluded based on inclusion and exclusion criteria: not
relation with fish intake, circulating biomarkers, and mortality. Am Heart J, 165(2):208-15 e4. general population
Matsumoto, C. et al. 2019. Fish and omega-3 fatty acid consumption and risk of hypertension. J Hypertens, 37(6):1223- Excluded based on inclusion and exclusion criteria:
1229. not CVD
Meng, L.X. et al. 2011. Association of Fish Consumption Factors with Stroke Mortality in The Multiethnic Cohort Study. Excluded based on inclusion and exclusion criteria: just
Stroke, 42(3):E276-E276. an abstract.
Mori, T.A. et al. 1994. Effects of Varying Dietary-Fat, Fish, and Fish Oils on Blood-Lipids in a Randomized Controlled Trial Excluded based on inclusion and exclusion criteria: only
in Men at Risk of Heart-Disease. American Journal of Clinical Nutrition, 59(5):1060-1068. biomarkers
Ness, A.R. et al. 2005. Diet in childhood and adult cardiovascular and all-cause mortality: the Boyd Orr cohort. Heart, Excluded based on inclusion and exclusion criteria:
91(7):894-898. not CVD
Panagiotakos, D.B. & Kastorini, C.M. 2011. Fish consumption and risk of stroke. Womens Health (Lond), 7(3):279-81. Excluded based on inclusion and exclusion criteria:
commentary article
Petsini, F. Fragopoulou, E. & Antonopoulou, S. 2019. Fish consumption and cardiovascular disease related biomarkers: A Excluded based on inclusion and exclusion criteria:
review of clinical trials. Crit Rev Food Sci Nutr, 59(13):2061-2071. narrative review (exclusion criteria)
Raisi-Estabragh, Z. et al. 2021. Associations of Meat and Fish Consumption With Conventional and Radiomics Excluded based on inclusion and exclusion criteria:
Cardiovascular Magnetic Resonance Phenotypes in the UK Biobank. Front Cardiovasc Med, 8:667849. biomarker
Ramel, A. et al. 2010. Moderate consumption of fatty fish reduces diastolic blood pressure in overweight and obese Excluded based on inclusion and exclusion criteria
European young adults during energy restriction. Nutrition, 26(2):168-74.
Rundblad, A. et al. 2018. Effects of krill oil and lean and fatty fish on cardiovascular risk markers: a randomised Excluded based on inclusion and exclusion criteria:
controlled trial. J Nutr Sci, 7:e3. not CVD
Salisbury, A.C. et al. 2011. Predictors of omega-3 index in patients with acute myocardial infarction. Mayo Clin Proc, Excluded based on inclusion and exclusion criteria:
86(7):626-32. omega-3 and not fish intake
Steur, M. et al. 2021. Dietary Fatty Acids, Macronutrient Substitutions, Food Sources and Incidence of Coronary Excluded based on inclusion and exclusion criteria:
Heart Disease: Findings From the EPIC-CVD Case-Cohort Study Across Nine European Countries. J Am Heart Assoc, case-cohort study (exclusion criteria)
10(23):e019814.
Sun, Y. et al. 2016. Plasma alpha-Linolenic and Long-Chain omega-3 Fatty Acids Are Associated with a Lower Risk of Excluded based on inclusion and exclusion criteria: no
Acute Myocardial Infarction in Singapore Chinese Adults. J Nutr, 146(2):275-82. food frequency questionnaire data
Vuholm, S. et al. 2019. Effects of oily fish intake on cardiometabolic markers in healthy 8- to 9-y-old children: the FiSK Excluded based on inclusion and exclusion criteria:
Junior randomized trial. Am J Clin Nutr, 110(6):1296-1305. biomarkers
Whelton, S.P. et al. 2004. Meta-analysis of observational studies on fish intake and coronary heart disease. Am J Cardiol, Excluded based on inclusion and exclusion criteria:
93(9):1119-23. review
Yinko, S.S.L.L. et al. 2014. Fish Consumption and Acute Coronary Syndrome: A Meta-Analysis. American Journal of Excluded based on inclusion and exclusion criteria:
Medicine,127(9): p. 848-+. review
Zhang, J. et al. 2010. Inclusion of Atlantic salmon in the Chinese diet reduces cardiovascular disease risk markers in Excluded based on inclusion and exclusion criteria:
dyslipidemic adult men. Nutr Res, 30(7): 447-54. wrong outcome
Zhang, Y. et al. 2020. Associations of Fish and Omega-3 Fatty Acids Consumption With the Risk of Venous Excluded based on inclusion and exclusion criteria:
Thromboembolism. A Meta-Analysis of Prospective Cohort Studies. Front Nutr, 7: 614784. review
Zhu, N.B. et al. 2019. Adherence to a healthy lifestyle and all-cause and cause-specific mortality in Chinese adults: a Excluded based on inclusion and exclusion criteria:
10-year prospective study of 0.5 million people. International Journal of Behavioral Nutrition and Physical Activity, 16: not CVD
1-13
Aadland, et al. 2016. Lean Seafood Intake Reduces Postprandial C-peptide and Lactate Concentrations in Healthy Adults Excluded based on inclusion and exclusion criteria:
in a Randomized Controlled Trial with a Crossover Design. Journal of Nutrition, 146(5):1027-1034. not CVD
Aadland, et al. 2015. Lean-seafood intake reduces cardiovascular lipid risk factors in healthy subjects: results from a Excluded based on inclusion and exclusion criteria:
randomized controlled trial with a crossover design. American Journal of Clinical Nutrition,102(3):582-592. not CVD

351
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.11 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES
AND OUTCOMES” (cont.)
Albert, C.M. et al. 1998. Fish consumption and risk of sudden cardiac death. JAMA, 279(1):23-8. Excluded for further assessment, as the primary study
had already been assessed in VKM 2022
Amiano, P. et al. 2016. No association between fish consumption and risk of stroke in the Spanish cohort of the European Excluded for further assessment, as the primary study
Prospective Investigation into Cancer and Nutrition (EPIC-Spain): a 13.8-year follow-up study. Public Health Nutr, had already been assessed in a systematic review that
19(4):674-81. is included in VKM 2022
Ascherio, A. et al. 1995. Dietary intake of marine n-3 fatty acids, fish intake, and the risk of coronary disease among men. Excluded for further assessment, as the primary study
N Engl J Med, 332(15):977-82. had already been assessed in VKM 2022
Atkinson, C. et al. 2011. Associations between types of dietary fat and fish intake and risk of stroke in the Caerphilly Excluded for further assessment, as the primary study
Prospective Study (CaPS). Public Health,125(6):345-8. had already been assessed in a systematic review that
is included in VKM 2022
Belin, R.J. et al. 2011. Fish intake and the risk of incident heart failure: the Women's Health Initiative. Circ Heart Fail, Excluded for further assessment, as the primary study
4(4):404-13. had already been assessed in a systematic review that
is included in VKM 2022
Bernstein, A.M. et al. 2012. Dietary protein sources and the risk of stroke in men and women. Stroke,43(3):637-44. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Bernstein, A.M. et al. 2010. Major dietary protein sources and risk of coronary heart disease in women. Excluded for further assessment, as the primary study
Circulation,122(9):876-83. had already been assessed in VKM 2022
Berry, J.D. et al. 2010. Dietary fish intake and incident atrial fibrillation (from the Women's Health Initiative). Am J Excluded for further assessment, as the primary study
Cardiol,105(6):844-8. had already been assessed in VKM 2022
Bierregaard, L.J. et al. 2010. Fish intake and acute coronary syndrome. European Heart Journal, 31(1):29-34. Excluded for further assessment, as the primary study
had already been assessed in VKM 2022
Bonaccio, M. et al. 2017. Fish intake is associated with lower cardiovascular risk in a Mediterranean population: Excluded for further assessment, as the primary study
Prospective results from the Moli-sani study. Nutrition Metabolism and Cardiovascular Diseases, 27(10):865-873. had already been assessed in VKM 2022
Bouzan, C. et al. 2005. A quantitative analysis of fish consumption and stroke risk. Am J Prev Med, 2005. 29(4): p. 347- Excluded for further assessment, as the primary study
52. had already been assessed in a systematic review that
is included in VKM 2022
Brouwer, I.A. et al. 2006. Intake of very long-chain n-3 fatty acids from fish and incidence of atrial fibrillation. The Excluded for further assessment, as the primary study
Rotterdam Study. American Heart Journal,151(4):857-862. had already been assessed in VKM 2022
Burr, M.L. & Fehily, A.M. 1991. Fatty fish and heart disease: a randomized controlled trial. World Rev Nutr Diet,66:306-12. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Burr, M.L. et al. 1989. Diet and reinfarction trial (DART): design, recruitment, and compliance. Eur Heart J, 10(6):558-67. Excluded for further assessment, as the primary study
had already been assessed in VKM 2022
Burr, M.L. et al. 1989. Effects of changes in fat, fish, and fibre intakes on death and myocardial reinfarction: diet and Excluded for further assessment, as the primary study
reinfarction trial (DART). Lancet, 2(8666):757-61. had already been assessed in VKM 2022
Burr, M.L. 1993. Fish and ischaemic heart disease. World Rev Nutr Diet, 72:49-60. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Daviglus, M.L. et al. 1997. Fish consumption and the 30-year risk of fatal myocardial infarction. N Engl J Med, Excluded for further assessment, as the primary study
336(15):1046-53. had already been assessed in VKM 2022
de Goede, J. et al. 2012. Gender-specific associations of marine n-3 fatty acids and fish consumption with 10-year Excluded for further assessment, as the primary study
incidence of stroke. PLoS One, 7(4):e33866. had already been assessed in a systematic review that
is included in VKM 2022
de Goede, J. et al. 2010. Marine (n-3) fatty acids, fish consumption, and the 10-year risk of fatal and nonfatal coronary Excluded for further assessment, as the primary study
heart disease in a large population of Dutch adults with low fish intake. J Nutr, 140(5):1023-8. had already been assessed in VKM 2022
Dijkstra, S.C. et al. 2009. Intake of very long chain n-3 fatty acids from fish and the incidence of heart failure: the Excluded for further assessment, as the primary study
Rotterdam Study. Eur J Heart Fail, 11(10):922-8. had already been assessed in a systematic review that
is included in VKM 2022
Erkkila, A.T. et al. 2004. Fish intake is associated with a reduced progression of coronary artery atherosclerosis in Excluded for further assessment, as the primary study
postmenopausal women with coronary artery disease. American Journal of Clinical Nutrition, 80(3):626-632. had already been assessed in VKM 2022
Gammelmark, A. et al. 2016. Association of fish consumption and dietary intake of marine n-3 PUFA with myocardial Excluded for further assessment, as the primary study
infarction in a prospective Danish cohort study. Br J Nutr, 116(1):167-77. had already been assessed in VKM 2022
Gillum, R.F. 1996. Fish consumption and stroke incidence. Stroke, 27(7):1254. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Gillum, R.F. M. Mussolino, & Madans, J.H. 2000. The relation between fish consumption, death from all causes, and Excluded for further assessment, as the primary study
incidence of coronary heart disease. the NHANES I Epidemiologic Follow-up Study. J Clin Epidemiol, 53(3):237-44. had already been assessed in VKM 2022
Gillum, R.F. Mussolino, M.E. & Madans, J.H. 1996.The relationship between fish consumption and stroke incidence - The Excluded for further assessment, as the primary study
NHANES I epidemiologic follow-up study. Archives of Internal Medicine, 156(5):537-542. had already been assessed in a systematic review that
is included in VKM 2022

352
APPENDICES

TABLE A3.11 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES
AND OUTCOMES” (cont.)
Gronroos, N.N. et al. 2012. Fish, fish-derived n-3 fatty acids, and risk of incident atrial fibrillation in the Atherosclerosis Excluded for further assessment, as the primary study
Risk in Communities (ARIC) study. PLoS One, 7(5):e36686. had already been assessed in VKM 2022
Hansen-Krone, I.J. et al. 2014. High fish plus fish oil intake is associated with slightly reduced risk of venous Excluded for further assessment, as the primary study
thromboembolism: the Tromso Study. J Nutr, 144(6):861-7. had already been assessed in a systematic review that
is included in VKM 2022
Haring, B. et al. 2014. Dietary protein intake and coronary heart disease in a large community based cohort: results from Excluded for further assessment, as the primary study
the Atherosclerosis Risk in Communities (ARIC) study [corrected]. PLoS One, 9(10):e109552. had already been assessed in VKM 2022
He, K. et al. 2002. Fish consumption and risk of stroke in men. JAMA, 288(24):3130-6. Excluded for further assessment, as the primary study
had already been assessed in VKM 2022
Hengeveld, L.M. et al. 2018. Fish consumption and risk of stroke, coronary heart disease, and cardiovascular mortality in Excluded for further assessment, as the primary study
a Dutch population with low fish intake. Eur J Clin Nutr, 72(7):942-950. had already been assessed in VKM 2022
Holmberg, S. Thelin, A. & Stiernstrom, E.L. 2009. Food choices and coronary heart disease: a population based cohort Excluded for further assessment, as the primary study
study of rural Swedish men with 12 years of follow-up. Int J Environ Res Public Health, 6(10):2626-38. had already been assessed in VKM 2022
Hu, F.B. et al. 2003. Fish and long-chain omega-3 fatty acid intake and risk of coronary heart disease and total mortality Excluded for further assessment, as the primary study
in diabetic women. Circulation, 107(14):1852-7. had already been assessed in VKM 2022
Hu, F.B. et al. 2002. Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. Jama, Excluded for further assessment, as the primary study
287(14):1815-21. had already been assessed in VKM 2022
Iso, H. et al. 2006. Intake of fish and n3 fatty acids and risk of coronary heart disease among Japanese: the Japan Public Excluded for further assessment, as the primary study
Health Center-Based (JPHC) Study Cohort I. Circulation, 113(2):195-202. had already been assessed in VKM 2022
Iso, H. et al. 2001. Intake of fish and omega-3 fatty acids and risk of stroke in women. JAMA, 285(3):304-12. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Jarvinen, R. et al. 2006. Intake of fish and long-chain n-3 fatty acids and the risk of coronary heart mortality in men and Excluded for further assessment, as the primary study
women. Br J Nutr, 95(4):824-9. had already been assessed in VKM 2022
Jiang, L. et al. 2021. Intake of Fish and Marine n-3 Polyunsaturated Fatty Acids and Risk of Cardiovascular Disease Excluded for further assessment, as the primary study
Mortality: A Meta-Analysis of Prospective Cohort Studies. Nutrients, 13(7). had already been assessed in VKM 2022
Key, T.J. et al. 2019. Consumption of Meat, Fish, Dairy Products, and Eggs and Risk of Ischemic Heart Disease A Excluded for further assessment, as the primary study
Prospective Study of 7198 Incident Cases Among 409 885 Participants in the Pan-European EPIC Cohort. Circulation, had already been assessed in VKM 2022
139(25):2835-2845.
Kuhn, T. et al. 2013. Fish consumption and the risk of myocardial infarction and stroke in the German arm of the Excluded for further assessment, as the primary study
European Prospective Investigation into Cancer and Nutrition (EPIC-Germany). British Journal of Nutrition, 110(6):1118- had already been assessed in VKM 2022
1125.
Larsson, S.C. & Wolk, A. 2017. Fish, long-chain omega-3 polyunsaturated fatty acid intake and incidence of atrial Excluded for further assessment, as the primary study
fibrillation: A pooled analysis of two prospective studies. Clin Nutr, 36(2):537-541. had already been assessed in VKM 2022
Larsson, S.C. & Orsini, N. 2011. Fish consumption and the risk of stroke: a dose-response meta-analysis. Stroke, Excluded for further assessment, as the primary study
42(12):3621-3. had already been assessed in VKM 2022
Levitan, E.B. Wolk, A. & Mittleman, M.A. 2010. Fatty fish, marine omega-3 fatty acids and incidence of heart failure. Excluded for further assessment, as the primary study
European Journal of Clinical Nutrition, 64(6):587-594. had already been assessed in a systematic review that
is included in VKM 2022
Levitan, E.B. Wolk, A. & Mittleman, M.A. 2009. Fish consumption, marine omega-3 fatty acids, and incidence of heart Excluded for further assessment, as the primary study
failure: a population-based prospective study of middle-aged and elderly men. Eur Heart J, 30(12):1495-500. had already been assessed in a systematic review that
is included in VKM 2022
Li, F.R. et al. 2017. Dietary Fish and Long-Chain n-3 Polyunsaturated Fatty Acids Intake and Risk of Atrial Fibrillation: A Excluded for further assessment, as the primary study
Meta-Analysis. Nutrients, 9(9). had already been assessed in VKM 2022
Manson, J.E. et al. 2020. Vitamin D, Marine n-3 Fatty Acids, and Primary Prevention of Cardiovascular Disease Current Excluded for further assessment, as the primary study
Evidence. Circ Res, 126(1):112-128. had already been assessed in VKM 2022
Micha, R. et al. 2017. Association Between Dietary Factors and Mortality From Heart Disease, Stroke, and Type 2 Diabetes Excluded for further assessment, as the primary study
in the United States. Jama-Journal of the American Medical Association, 317(9):912-924. had already been assessed in VKM 2022
Mohan, D. et al. 2021. Associations of Fish Consumption With Risk of Cardiovascular Disease and Mortality Among Excluded for further assessment, as the primary study
Individuals With or Without Vascular Disease From 58 Countries. Jama Internal Medicine, 181(5): 631-649. had already been assessed in VKM 2022
Montonen, J. et al. 2009. Fish consumption and the incidence of cerebrovascular disease. Br J Nutr, 102(5):750-6. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Morris, M.C. et al. 1995. Fish consumption and cardiovascular disease in the physicians' health study: a prospective Excluded for further assessment, as the primary study
study. Am J Epidemiol, 142(2):166-75. had already been assessed in VKM 2022
Mozaffarian, D. et al. 2003. Cardiac benefits of fish consumption may depend on the type of fish meal consumed: the Excluded for further assessment, as the primary study
Cardiovascular Health Study. Circulation, 107(10):1372-7. had already been assessed in VKM 2022
Mozaffarian, D. et al. 2005. Fish consumption and stroke risk in elderly individuals: the cardiovascular health study. Arch Excluded for further assessment, as the primary study
Intern Med, 165(2): 200-6. had already been assessed in a systematic review that
is included in VKM 2022

353
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.11 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES
AND OUTCOMES” (cont.)
Mozaffarian, D. et al. 2004. Fish intake and risk of incident atrial fibrillation. Circulation, 110(4):368-373. Excluded for further assessment, as the primary study
had already been assessed in a systematic review that
is included in VKM 2022
Mozaffarian, D. et al. 2011. Mercury exposure and risk of cardiovascular disease in two U.S. cohorts. N Engl J Med, Excluded for further assessment, as the primary study
364(12):1116-25. had already been assessed in a systematic review that
is included in VKM 2022
Myint, P.K. et al. 2006. Habitual fish consumption and risk of incident stroke: the European Prospective Investigation into Excluded for further assessment, as the primary study
Cancer (EPIC)-Norfolk prospective population study. Public Health Nutrition, 2006. 9(7):882-888. had already been assessed in the included systematic
review Chowdhury et al 2012.
Nahab, F. et al. 2016. Dietary fried fish intake increases risk of CVD: the Reasons for Geographic And Racial Differences Excluded for further assessment, as the primary study
in Stroke (REGARDS) study. Public Health Nutr, 19(18):3327-3336. had already been assessed in VKM 2022
Orencia, A.J. et al. 1996. Fish consumption and stroke in men. 30-year findings of the Chicago Western Electric Study. Excluded for further assessment, as the primary study
Stroke, 27(2):204-9. had already been assessed in a systematic review that
is included in VKM 2022
Osler, M. Andreasen, A.H. & Hoidrup, S. 2003. No inverse association between fish consumption and risk of death from Excluded for further assessment, as the primary study
all-causes, and incidence of coronary heart disease in middle-aged, Danish adults. Journal of Clinical Epidemiology, had already been assessed in VKM 2022
56(3):274-279.
Qin, Z.Z. et al. 2018. Effects of fatty and lean fish intake on stroke risk: a meta-analysis of prospective cohort studies. Excluded for further assessment, as the primary study
Lipids Health Dis, 17(1):264. had already been assessed in VKM 2022
Rhee, J.J. et al 2017. Fish Consumption, Omega-3 Fatty Acids, and Risk of Cardiovascular Disease. Am J Prev Med, Excluded for further assessment, as the primary study
52(1):10-19. had already been assessed in VKM 2022
Rodriguez, B.L. et al. 1995. Fish intake may limit the increase in risk of coronary heart disease morbidity and mortality Excluded for further assessment, as the primary study
among heavy smokers - The Honolulu Heart Program. Circulation, 94(5):952-956. had already been assessed in VKM 2022
Salonen, J.T. Nyyssonen, K. & Salonen, R. 1995. Fish intake and the risk of coronary disease. N Engl J Med, 333(14):937; Excluded for further assessment, as the primary study
author reply 938. had already been assessed in VKM 2022
Severinsen, M.T. et al. 2014. Fish intake and venous thromboembolism: a Danish follow-up study. Thromb Res, Excluded for further assessment, as the primary study
133(3):352-6. had already been assessed in a systematic review that
is included in VKM 2022
Shen, J. et al. 2011. Dietary factors and incident atrial fibrillation: the Framingham Heart Study. Am J Clin Nutr, Excluded for further assessment, as the primary study
93(2):261-6. had already been assessed in VKM 2022
Steffen, L.M. et al. 2007. Greater fish, fruit, and vegetable intakes are related to lower incidence of venous Excluded for further assessment, as the primary study
thromboembolism: the Longitudinal Investigation of Thromboembolism Etiology. Circulation, 115(2):188-95. had already been assessed in a systematic review that
is included in VKM 2022
Strom, M. et al. 2011. Fish consumption measured during pregnancy and risk of cardiovascular diseases later in life: an Excluded for further assessment, as the primary study
observational prospective study. PLoS One, 6(11):e27330. had already been assessed in VKM 2022
Strom, M. et al. 2012. Fish, n-3 fatty acids, and cardiovascular diseases in women of reproductive age: a prospective Excluded for further assessment, as the primary study
study in a large national cohort. Hypertension, 59(1):36-43. had already been assessed in VKM 2022
Virtanen, J.K. et al. 2008. Fish consumption and risk of major chronic disease in men. Am J Clin Nutr, 88(6):1618-25. Excluded for further assessment, as the primary study
had already been assessed in VKM 2022
Wallin, A. et al. 2018. Fish consumption in relation to myocardial infarction, stroke and mortality among women and men Excluded for further assessment, as the primary study
with type 2 diabetes: A prospective cohort study. Clinical Nutrition, 37(2):590-596. had already been assessed in a systematic review that
is included in VKM 2022
Ward, R.E. et al. 2020. Omega-3 supplement use, fish intake, and risk of non-fatal coronary artery disease and ischemic Excluded for further assessment, as the primary study
stroke in the Million Veteran Program. Clin Nutr, 39(2):574-579. had already been assessed in VKM 2022
Wennberg, M. et al. 2011. Fish consumption and myocardial infarction: a second prospective biomarker study from Excluded for further assessment, as the primary study
northern Sweden. Am J Clin Nutr, 93(1):27-36. had already been assessed in VKM 2022
Wennberg, M. et al. 2007. Fish intake, mercury, long-chain n-3 polyunsaturated fatty acids and risk of stroke in northern Excluded for further assessment, as the primary study
Sweden. Br J Nutr, 98(5):1038-45. had already been assessed in a systematic review that
is included in VKM 2022
Wennberg, M. et al. 2012. Myocardial infarction in relation to mercury and fatty acids from fish: a risk-benefit analysis Excluded for further assessment, as the primary study
based on pooled Finnish and Swedish data in men. Am J Clin Nutr, 96(4):706-13. had already been assessed in a systematic review that
is included in VKM 2022
Wilk, J.B. et al. 2012. Plasma and dietary omega-3 fatty acids, fish intake, and heart failure risk in the Physicians' Health Excluded for further assessment, as the primary study
Study. Am J Clin Nutr, 96(4):882-8. had already been assessed in a systematic review that
is included in VKM 2022
Wurtz, A.M. et al. 2016. Substitution of meat and fish with vegetables or potatoes and risk of myocardial infarction. Br J Excluded for further assessment, as the primary study
Nutr, 116(9):1602-1610. had already been assessed in a systematic review that
is included in VKM 2022
Wurtz, A.M. et al. 2016. Substitutions of red meat, poultry and fish and risk of myocardial infarction. Br J Nutr, Excluded for further assessment, as the primary study
115(9):1571-8. had already been assessed in a systematic review that
is included in VKM 2022

354
APPENDICES

TABLE A3.11 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “CARDIOVASCULAR DISEASES
AND OUTCOMES” (cont.)
Xun, P. et al. 2012. Fish consumption and risk of stroke and its subtypes: accumulative evidence from a meta-analysis of Excluded for further assessment, as the primary study
prospective cohort studies. Eur J Clin Nutr, 66(11):1199-207. had already been assessed in VKM 2022
Yamagishi, K. et al. 2008. Fish, omega-3 polyunsaturated fatty acids, and mortality from cardiovascular diseases in Excluded for further assessment, as the primary study
a nationwide community-based cohort of Japanese men and women the JACC (Japan Collaborative Cohort Study for had already been assessed in a systematic review that
Evaluation of Cancer Risk) Study. J Am Coll Cardiol, 52(12):988-96. is included in VKM 2022
Yuan, J.M. et al. 2001. Fish and shellfish consumption in relation to death from myocardial infarction among men in Excluded for further assessment, as the primary study
Shanghai, China. Am J Epidemiol, 154(9):809-16. had already been assessed in a systematic review that
is included in VKM 2022
Zhang, B. et al. 2020. Fish Consumption and Coronary Heart Disease: A Meta-Analysis. Nutrients, 12(8). Excluded for further assessment, as the primary study
had already been assessed in VKM 2022
Zhang, H. et al. 2021. Familial factors, diet, and risk of cardiovascular disease: a cohort analysis of the UK Biobank. Am J Excluded for further assessment, as the primary study
Clin Nutr, 114(5):1837-1846. had already been assessed in VKM 2022
Zhao, W. et al. 2019. Fish Consumption and Stroke Risk: A Meta-Analysis of Prospective Cohort Studies. J Stroke Excluded for further assessment, as the primary study
Cerebrovasc Dis, 28(3):604-611. had already been assessed in VKM 2022
Zheng, J. et al. 2012. Fish consumption and CHD mortality: an updated meta-analysis of seventeen cohort studies. Public Excluded for further assessment, as the primary study
Health Nutr, 2012. 15(4):725-37. had already been assessed in VKM 2022
Zhong, V.W. et al. 2020. Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Excluded for further assessment, as the primary study
Cardiovascular Disease and All-Cause Mortality. JAMA Intern Med, 180(4):503-512. had already been assessed in VKM 2022

355
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TYPE 2 DIABETES
TABLE A3.12 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TYPE 2 DIABETES”
DURING FULL-TEXT SCREENING BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 10) Reason for exclusion
Dàrcy et al. 2020 The Role of Diet in the Prevention of Diabetes among Women with Prior Gestational Diabetes: A Excluded based on inclusion and exclusion criteria:
Systematic Review of Intervention and Observational Studies. Journal of the Academy of Nutrition and Dietetics, 120(1), dietary pattern
69-85.
Franz et al. 2017. Academy of Nutrition and Dietetics Nutrition Practice Guideline for Type 1 and Type 2 Diabetes in Adults: Excluded based on inclusion and exclusion criteria: Not
Systematic Review of Evidence for Medical Nutrition Therapy Effectiveness and Recommendations for Integration into the relevant – the study investigates the effects of different
Nutrition Care Process. Journal of the Academy of Nutrition and Dietetics, 117(10), 1659-1679. medical nutrition therapies.
Karimi et al. 2020. A systematic review and meta-analysis of the association between fish consumption and risk of Excluded based on inclusion and exclusion criteria: does
metabolic syndrome. Nutrition, Metabolism and Cardiovascular Diseases, 30(5), 717-729. not meet health outcome criteria
Kim et al. 2015. Fish consumption, long-chain omega-3 polyunsaturated fatty acid intake and risk of metabolic Excluded based on inclusion and exclusion criteria: does
syndrome: a meta-analysis. Nutrients, 7(4), 2085-2100. not meet health outcome criteria
Schwab et al. 2014. Effect of the amount and type of dietary fat on cardiometabolic risk factors and risk of developing Excluded based on inclusion and exclusion criteria: does
type 2 diabetes, cardiovascular diseases, and cancer: a systematic review. Food & Nutrition research, 58(1), 25145. not meet intervention/exposure criteria
Tørris et al. 2014. Fish consumption and its possible preventive role on the development and prevalence of metabolic Excluded based on inclusion and exclusion criteria: does
syndrome - a systematic review. Diabetology & Metabolic Syndrome, 6, 1-11. not meet health outcome criteria
Schwingshackl et al. 2017. Food groups and risk of type 2 diabetes mellitus: a systematic review and meta-analysis of Excluded, as the review has already been assessed in
prospective studies. European Journal of Epidemiology, 32, 363-375. VKM 2022.
Namazi et al. 2019. The association between types of seafood intake and the risk of type 2 diabetes: a systematic review Excluded, as the review has already been assessed in
and meta-analysis of prospective cohort studies. Health Promotion Perspectives, 9(3), 164. VKM 2022.
Pastorino et al. 2021. Heterogeneity of Associations between Total and Types of Fish Intake and the Incidence of Type 2 Excluded, as the review has already been assessed in
Diabetes: Federated Meta-Analysis of 28 Prospective Studies Including 956,122 Participants. Nutrients, 13(4), 1223. VKM 2022.
Yang et al. 2020. Meat and fish intake and type 2 diabetes: Dose-response meta-analysis of prospective cohort studies. Excluded, as the review has already been assessed in
Diabetes & Metabolism, 46(5), 345-352. VKM 2022.

356
APPENDICES

TABLE A3.13 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TYPE 2 DIABETES”
Study (n = 40) Reason for exclusion
Amoah, J. et al. 2021. Effects of a school-based intervention to reduce cardiovascular disease risk factors among Excluded based on inclusion and exclusion criteria: not
secondary school students: A cluster- randomized, controlled trial. PloS One, 16(11):e0259581. general population
Abete, I. Parra, D. Crujeiras, A.B. Goyenechea, E. & Martinez, J.A. 2008. Specific insulin sensitivity and leptin responses to Excluded based on inclusion and exclusion criteria: type
a nutritional treatment of obesity via a combination of energy restriction and fatty fish intake. Journal of human nutrition 2 diabetes is not an endpoint
and dietetics, 21(6):591-600.
Adamsson, V. Reumark, A. Fredriksson, I.B. Hammarström, E. Vessby, B. Johansson, G. & Risérus, U. 2011. Effects of Excluded based on inclusion and exclusion criteria:
a healthy Nordic diet on cardiovascular risk factors in hypercholesterolaemic subjects: a randomized controlled trial excluded as fish is studied as a part of a Nordic diet
(NORDIET). Journal of internal medicine, 269(2):150-159.
Baik, I. Abbott, R.D. Curb, J.D. & Shin, C. 2010. Intake of fish and n-3 fatty acids and future risk of metabolic syndrome. Excluded based on inclusion and exclusion criteria: type
Journal of the American Dietetic Association, 110(7):1018-1026. 2 diabetes is not an endpoint
Brouwer-Brolsma, E.M. van Woudenbergh, G.J. Elferink, S.O. Singh-Povel, C.M. Hofman, A. Dehghan, A. Feskens, E. J. M. Excluded based on inclusion and exclusion criteria: no
et al. 2016. Intake of different types of dairy and its prospective association with risk of type 2 diabetes: the Rotterdam fish consumption measured
Study. Nutrition, Metabolism and Cardiovascular Diseases, 26(11):987-995.
Díaz-Rizzolo, D. A. Serra, A. Colungo, C. Sala-Vila, A. Sisó-Almirall, A. & Gomis, R. 2021. Type 2 diabetes preventive Excluded based on inclusion and exclusion criteria: the
effects with a 12-months sardine-enriched diet in elderly population with prediabetes: An interventional, randomized and population studied is prediabetic
controlled trial. Clinical Nutrition, 40(5):2587-2598.
Feskens, E. J. Virtanen, S. M. Räsänen, L. Tuomilehto, J. Stengård, J. Pekkanen, J. Kromhout, D. et al. 1995. Dietary factors Excluded based on inclusion and exclusion criteria: not
determining diabetes and impaired glucose tolerance: a 20-year follow-up of the Finnish and Dutch cohorts of the Seven analysed in a prospective manner
Countries Study. Diabetes care, 18(8):1104-1112.
Feskens, E.J. Bowles, C.H. & Kromhout, D. 1991. Inverse association between fish intake and risk of glucose intolerance in Excluded based on inclusion and exclusion criteria: not
normoglycemic elderly men and women. Diabetes care, 14(11):935-941. analysed in a prospective manner
Helland, A. Bratlie, M. Hagen, I.V. Mjøs, S.A. Sørnes, S. Halstensen, A.I. Gudbrandsen, O.A. et al. 2017. High intake of fatty Excluded based on inclusion and exclusion criteria: type
fish, but not of lean fish, improved postprandial glucose regulation and increased the n-3 PUFA content in the leucocyte 2 diabetes is not an endpoint
membrane in healthy overweight adults: a randomised trial. British Journal of Nutrition, 117(10):1368-1378.
Hustad, K.S. Ottestad, I. Hjorth, M. Dalen, K.T. Sæther, T. Sheikh, N.A. Holven, K.B. et al. 2021. No effect of salmon fish Excluded based on inclusion and exclusion criteria: RCT
protein on 2-h glucose in adults with increased risk of type 2 diabetes: a randomised controlled trial. British Journal of with salmon fish protein supplement
Nutrition, 126(9):1304-1313.
Ibsen, D.B. Steur, M. Imamura, F. Overvad, K. Schulze, M.B. Bendinelli, B. Wareham, N.J. et al. 2020. Replacement of red Excluded based on inclusion and exclusion criteria:
and processed meat with other food sources of protein and the risk of type 2 diabetes in European populations: the EPIC- wrong study design, case-cohort study
InterAct Study. Diabetes Care, 43(11):2660-2667.
Ibsen, D. Jakobsen, M. Halkjær, J. Parner, E. & Overvad, K. 2020. Replacing Red Meat with Alternative Food Sources of Excluded based on inclusion and exclusion criteria:
Protein on Risk of Type 2 Diabetes-Modeling Dietary Changes in a Causal Framework. Current Developments in Nutrition, wrong study design
4(Supplement_2):1418-1418.
Ibsen, D.B. Warberg, C.K. Würtz, A.M.L. Overvad, K. & Dahm, C.C. 2019. Substitution of red meat with poultry or fish and Excluded based on inclusion and exclusion criteria:
risk of type 2 diabetes: a Danish cohort study. European journal of nutrition, 58, 2705-2712. wrong study design
Ikeda, K. Sato, T. Nakayama, T. Tanaka, D. Nagashima, K. Mano, F. Nagahama Study Group et al. 2018. Dietary habits Excluded based on inclusion and exclusion criteria:
associated with reduced insulin resistance: The Nagahama study. Diabetes Research and Clinical Practice, 141, 26-34. wrong study design, cross-sectional analyses
Kim, Y. S. Xun, P. Iribarren, C. Van Horn, L. Steffen, L. Daviglus, M.L. He, K. et al. 2016. Intake of fish and long-chain Excluded based on inclusion and exclusion criteria: T2D
omega-3 polyunsaturated fatty acids and incidence of metabolic syndrome among American young adults: a 25-year is not an endpoint
follow-up study. European journal of nutrition, 55, 1707-1716.
Lankinen, M. Schwab, U. Kolehmainen, M. Paananen, J. Poutanen, K. Mykkänen, H. Orešič, M. et al. 2011. Whole grain Excluded based on inclusion and exclusion criteria: RCT,
products, fish and bilberries alter glucose and lipid metabolism in a randomized, controlled trial: the Sysdimet study. PloS high-risk persons
One, 6(8):e22646.
Mori, T.A. Bao, D.Q. Burke, V. Puddey, I.B. Watts, G.F. & Beilin, L.J. et al. 1999. Dietary fish as a major component of Excluded based on inclusion and exclusion criteria:
a weight-loss diet: effect on serum lipids, glucose, and insulin metabolism in overweight hypertensive subjects. The wrong publication type, type 2 diabetes not an endpoint
American journal of clinical nutrition, 70(5):817-825.
Nanri, A. 2013. Nutritional epidemiology of type 2 diabetes and depressive symptoms. Journal of Epidemiology, 23(4):243- Excluded based on inclusion and exclusion criteria:
250. wrong study design, review paper
Navas-Carretero, S. Pérez-Granados, A.M. Schoppen, S. & Vaquero, M.P. 2009. An oily fish diet increases insulin sensitivity Excluded based on inclusion and exclusion criteria: RCT,
compared to a red meat diet in young iron-deficient women. British journal of nutrition, 102(4):546-553. type 2 diabetes is not an endpoint
Ouellet, V. Marois, J. Weisnagel, S.J. & Jacques, H. 2007. Dietary cod protein improves insulin sensitivity in insulin- Excluded based on inclusion and exclusion criteria: RCT,
resistant men and women: a randomized controlled trial. Diabetes Care, 30(11):2816-2821. type 2 diabetes is not an endpoint
Ramel, A. Martinez, A. Kiely, M. Morais, G. Bandarra, N.M. & Thorsdottir, I. 2008. Beneficial effects of long-chain n-3 Excluded based on inclusion and exclusion criteria: RCT,
fatty acids included in an energy-restricted diet on insulin resistance in overweight and obese European young adults. obese individuals, type 2 diabetes is not an endpoint
Diabetologia, 51, 1261-1268.
Ruidavets, J.B. Bongard, V. Dallongeville, J. Arveiler, D. Ducimetière, P. Perret, B. Ferrières, J. et al. 2007. High Excluded based on inclusion and exclusion criteria: T2D
consumptions of grain, fish, dairy products and combinations of these are associated with a low prevalence of metabolic is not an endpoint
syndrome. Journal of Epidemiology & Community Health, 61(9):810-817.
Torjesen, P.A. Birkeland, K.I. Anderssen, S.A. Hjermann, I. Holme, I. & Urdal, P. 1997. Lifestyle changes may reverse Excluded based on inclusion and exclusion criteria: RCT,
development of the insulin resistance syndrome. The Oslo Diet and Exercise Study: a randomized trial. Diabetes care, type 2 diabetes is not an endpoint
20(1):26-31.

357
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.13 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TYPE 2 DIABETES” (cont.)
Würtz, A.M.L. Jakobsen, M.U. Bertoia, M.L. Hou, T. Schmidt, E.B. Willett, W.C. Rimm, E.B. et al. 2021. Replacing the Excluded based on inclusion and exclusion criteria:
consumption of red meat with other major dietary protein sources and risk of type 2 diabetes mellitus: a prospective wrong study design
cohort study. The American Journal of Clinical Nutrition, 113(3):612-621.
Aadland, E.K. Graff, I.E. Lavigne, C. Eng, Ø. Paquette, M. Holthe, A. Liaset, B. et al. 2016. Lean seafood intake reduces Excluded based on inclusion and exclusion criteria: RCT,
postprandial C-peptide and lactate concentrations in healthy adults in a randomized controlled trial with a crossover type 2 diabetes is not an endpoint
design. The Journal of nutrition, 146(5):1027-1034.
Chen, G.C. Arthur, R. Qin, L.Q. Chen, L.H. Mei, Z. Zheng, Y. Qi, Q. 2021. Association of oily and nonoily fish consumption Excluded for further assessment as the primary study
and fish oil supplements with incident type 2 diabetes: a large population-based prospective study. Diabetes Care, had already been assessed in VKM 2022
44(3):672-680.
Djousse, L. Gaziano, J.M. Buring, J.E. & Lee, I.M. 2011. Dietary omega-3 fatty acids and fish consumption and risk of type Excluded for further assessment as the primary study
2 diabetes. The American journal of clinical nutrition, 93(1):143-150. had already been assessed in VKM 2022
Du, H. Guo, Y. Bennett, D.A. Bragg, F. Bian, Z. Chadni, M. China Kadoorie Biobank collaborative group. et al. 2020. Red Excluded for further assessment as the primary study
meat, poultry and fish consumption and risk of diabetes: a 9 year prospective cohort study of the China Kadoorie Biobank. had already been assessed in VKM 2022
Diabetologia, 63, 767-779.
Kaushik, M. Mozaffarian, D. Spiegelman, D. Manson, J.E. Willett, W.C. & Hu, F.B. 2009. Long-chain omega-3 fatty acids, Excluded for further assessment as the primary study
fish intake, and the risk of type 2 diabetes mellitus. The American journal of clinical nutrition, 90(3):613-620. had already been assessed in VKM 2022
Löfvenborg, J.E. Carlsson, S. Andersson, T. Hampe, C.S. Koulman, A. Chirlaque Lopez, M.D., Wareham, N. J. et al. 2021. Excluded for further assessment as the primary study
Interaction between GAD65 antibodies and dietary fish intake or plasma phospholipid n-3 polyunsaturated fatty acids on had already been assessed in VKM 2022
incident adult-onset diabetes: the EPIC-InterAct study. Diabetes care, 44(2):416-424.
Nanri, A. Mizoue, T. Noda, M. Takahashi, Y. Matsushita, Y. Poudel-Tandukar, K. Japan Public Health Center–based Excluded for further assessment as the primary study
Prospective Study Group. et al. 2011. Fish intake and type 2 diabetes in Japanese men and women: the Japan Public had already been assessed in VKM 2022
Health Center–based Prospective Study–. The American journal of clinical nutrition, 94(3):884-891.
Patel, P.S. Sharp, S.J. Luben, R.N. Khaw, K.T. Bingham, S.A. Wareham, N.J. & Forouhi, N.G. 2009. Association between Excluded for further assessment as the primary study
type of dietary fish and seafood intake and the risk of incident type 2 diabetes: the European prospective investigation of had already been assessed in a systematic review that
cancer (EPIC)-Norfolk cohort study. Diabetes care, 32(10):1857-1863. is included in VKM 2022
Patel et al. 2012. The prospective association between total and type of fish intake and type 2 diabetes in 8 European Excluded for further assessment as the primary study
countries: EPIC-InterAct Study. The American journal of clinical nutrition, 95.6: 1445-1453. had already been assessed in a systematic review that
is included in VKM 2022
Rylander, C. Sandanger, T.M. Engeset, D. & Lund, E. 2014. Consumption of lean fish reduces the risk of type 2 diabetes Excluded for further assessment as the primary study
mellitus: a prospective population based cohort study of Norwegian women. PloS One, 9(2):e89845. had already been assessed in VKM 2022
Talaei, M. Wang, Y.L. Yuan, J.M. Pan, A. & Koh, W.P. 2017. Meat, dietary heme iron, and risk of type 2 diabetes mellitus: the Excluded for further assessment as the primary study
Singapore Chinese Health Study. American journal of epidemiology, 186(7):824-833. had already been assessed in VKM 2022
Van Woudenbergh, G.J. van Ballegooijen, A.J. Kuijsten, A. Sijbrands, E.J. van Rooij, F.J. Geleijnse, J.M. Feskens, E.J. Excluded for further assessment as the primary study
et al. 2009. Eating fish and risk of type 2 diabetes: a population-based, prospective follow-up study. Diabetes care, had already been assessed in VKM 2022
32(11):2021-2026.
Villegas, R. Xiang, Y.B. Elasy, T. Li, H.L. Yang, G. Cai, H. Shu, X.O. et al. 2011. Fish, shellfish, and long-chain n− 3 fatty Excluded for further assessment as the primary study
acid consumption and risk of incident type 2 diabetes in middle-aged Chinese men and women. The American journal of had already been assessed in VKM 2022
clinical nutrition, 94(2):543-551.
Virtanen, J.K. Mursu, J. Voutilainen, S. Uusitupa, M. & Tuomainen, T.P. 2014. Serum omega-3 polyunsaturated fatty Excluded for further assessment as the primary study
acids and risk of incident type 2 diabetes in men: the Kuopio Ischemic Heart Disease Risk Factor study. Diabetes care, had already been assessed in VKM 2022
37(1):189-196.
Wallin, A. Di Giuseppe, D. Orsini, N. Åkesson, A. Forouhi, N.G. & Wolk, A. 2017. Fish consumption and frying of fish in Excluded for further assessment as the primary study
relation to type 2 diabetes incidence: a prospective cohort study of Swedish men. European journal of nutrition, 56, had already been assessed in VKM 2022
843-852.
Zhang, Y. Zhuang, P. Mao, L. Chen, X. Wang, J. Cheng, L. Jiao, J. et al. 2019. Current level of fish and omega-3 fatty acid Excluded for further assessment as the primary study
intakes and risk of Type 2 diabetes in China. The Journal of nutritional biochemistry, 74, 108249. had already been assessed in VKM 2022
Øyen, J. Brantsæter, A.L. Nøstbakken, O. . Birkeland, K.I. Haugen, M. Madsen, L. & Egeland, G.M. 2021. Intakes of fish and Excluded for further assessment, as the primary study
long-chain n-3 polyunsaturated fatty acid supplements during pregnancy and subsequent risk of type 2 diabetes in a had already been assessed in VKM 2022
large prospective cohort study of Norwegian women. Diabetes Care, 44(10):2337-2345.

358
APPENDICES

NEURODEVELOPMENT AND NEUROLOGICAL DISORDERS

TABLE A3.14 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “NEURODEVELOPMENT AND
NEUROLOGICAL DISORDERS” BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 20) Reason for exclusion
Appleton, K.M. Hayward, R.C. Gunnell, D. Peters, T.J. Rogers, P.J. Kessler, D. & Ness, A.R. 2006. Effects of n–3 long-chain Excluded based on inclusion and exclusion criteria: focus
polyunsaturated fatty acids on depressed mood: systematic review of published trials. The American journal of clinical on n-3 fatty acids.
nutrition, 84(6):1308-1316.
Appleton, K.M. Rogers, P.J. & Ness, A.R. 2010. Updated systematic review and meta-analysis of the effects of n−3 long- Excluded based on inclusion and exclusion criteria: focus
chain polyunsaturated fatty acids on depressed mood. The American journal of clinical nutrition, 91(3):757-770. on n-3 fatty acids.
Bakre, A.T. Chen, R. Khutan, R. Wei, L. Smith, T. Qin, G. Ni, J. et al. 2018. Association between fish consumption and risk Excluded, as the review has already been assessed in
of dementia: a new study from China and a systematic literature review and meta-analysis. Public health nutrition, VKM 2022.
21(10):1921-1932.
Fernandes, A.C. Medeiros, C.O. Bernardo, G.L. Ebone, M.V. Di Pietro, P.F. Assis, M.A.A.D. & Vasconcelos, F. 2012. Benefits Excluded based on inclusion and exclusion criteria:
and risks of fish consumption for the human health. Revista de Nutrição, 25, 283-295. review
Fotuhi, M. Mohassel, P. & Yaffe, K. 2009. Fish consumption, long-chain omega-3 fatty acids and risk of cognitive decline Excluded based on inclusion and exclusion criteria:
or Alzheimer disease: a complex association. Nature Reviews Neurology, 5(3):140-152. focus on use of n-3 in relation to cognitive impairment
and dementia
Grosso, G. Micek, A. Marventano, S. Castellano, S. Mistretta, A. Pajak, A. & Galvano, F. 2016. Dietary n-3 PUFA, fish Excluded, as the review has already been assessed in
consumption and depression: a systematic review and meta-analysis of observational studies. Journal of Affective VKM 2022.
Disorders, 205, 269-281.
Hibbeln, J. R. Spiller, P. Brenna, J. T. Golding, J. Holub, B. J. Harris, W. S., Carlson, S. E. et al. 2019. Relationships between Excluded, as the review has already been assessed in
seafood consumption during pregnancy and childhood and neurocognitive development: Two systematic reviews. VKM 2022.
Prostaglandins, Leukotrienes and Essential Fatty Acids, 151, 14-36.
Ishihara, L. & Brayne, C. 2005. A systematic review of nutritional risk factors of Parkinson's disease. Nutrition research Excluded based on inclusion and exclusion criteria: focus
reviews, 18(2):259-282. on nutritional factors and not fish intake alone.
Kosti, R. I. Kasdagli, M. I. Kyrozis, A. Orsini, N. Lagiou, P. Taiganidou, F. & Naska, A. 2022. Fish intake, n-3 fatty acid Excluded, as the review has already been assessed in
body status, and risk of cognitive decline: a systematic review and a dose–response meta-analysis of observational and VKM 2022.
experimental studies. Nutrition Reviews, 80(6):1445-1458.
Lee, Y. Back, J.H. Kim, J. Kim, S.H. Na, D.L. Cheong, H.K. Kim, Y.G. et al. 2010. Systematic review of health behavioral risks Excluded based on inclusion and exclusion criteria:
and cognitive health in older adults. International psychogeriatrics, 22(2):174-187. health behaviour as outcome, and only 3 studies of the
37 included reported on fish intake.
Li, F. Liu, X. & Zhang, D. 2016. Fish consumption and risk of depression: a meta-analysis. J Epidemiol Community Health, Excluded, as the review has already been assessed in
70(3):299-304. VKM 2022.
Matison, A.P. Mather, K.A. Flood, V.M. & Reppermund, S. 2021. Associations between nutrition and the incidence of Excluded, as the review has already been assessed in
depression in middle-aged and older adults: A systematic review and meta-analysis of prospective observational VKM 2022.
population-based studies. Ageing Research Reviews, 70, 101403.
Molendijk, M. Molero, P. Sánchez-Pedreño, F.O. Van der Does, W. & Martínez-González, M.A. 2018. Diet quality and Excluded, as the review has already been assessed in
depression risk: a systematic review and dose-response meta-analysis of prospective studies. Journal of affective VKM 2022.
disorders, 226, 346-354.
Solfrizzi, V. Custodero, C. Lozupone, M. Imbimbo, B. P. Valiani, V. Agosti, P., Panza, F. et al. 2017. Relationships of dietary Excluded, as the review has already been assessed in
patterns, foods, and micro-and macronutrients with Alzheimer’s disease and late-life cognitive disorders: a systematic VKM 2022.
review. Journal of Alzheimer's Disease, 59(3):815-849.
Sparling, T.M. Henschke, N. Nesbitt, R.C. & Gabrysch, S. 2017. The role of diet and nutritional supplementation in Excluded based on inclusion and exclusion criteria: focus
perinatal depression: a systematic review. Maternal & child nutrition, 13(1). on nutritional supplementation and diet overall.
Wu, S. Ding, Y. Wu, F. Li, R. Hou, J. & Mao, P. 2015. Omega-3 fatty acids intake and risks of dementia and Alzheimer's Excluded based on inclusion and exclusion criteria: focus
disease: a meta-analysis. Neuroscience & Biobehavioral Reviews, 48, 1-9. on n-3 fatty acids.
Zhao, X.H.& Zhang, Z.H. 2020. Risk factors for postpartum depression: An evidence-based systematic review of Excluded based on inclusion and exclusion criteria:
systematic reviews and meta-analyses. Asian journal of psychiatry, 53, 102353. review of risk factors and not on fish intake.
Yang, Y. Kim, Y. & Je, Y. 2018. Fish consumption and risk of depression: Epidemiological evidence from prospective Excluded, as the review has already been assessed in
studies. Asia‐Pacific Psychiatry, 10(4):e12335. VKM 2022.
Zeng, L.F. Cao, Y. Liang, W.X. Bao, W.H. Pan, J.K. Wang, Q., Wang, N.S. et al. 2017. An exploration of the role of a Excluded, as the review has already been assessed in
fish-oriented diet in cognitive decline: a systematic review of the literature. Oncotarget, 8(24):39877. VKM 2022.
Zhang, Y. Chen, J. Qiu, J. Li, Y. Wang, J. & Jiao, J. 2015. Intakes of fish and polyunsaturated fatty acids and mild-to-severe Excluded, as the review has already been assessed in
cognitive impairment risks: a dose-response meta-analysis of 21 cohort studies–3. The American Journal of Clinical VKM 2022.
Nutrition, 103(2):330-340.

359
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.15 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “NEURODEVELOPMENT
AND NEUROLOGICAL DISORDERS” BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 80) Reason for exclusion
Batty, G.D. Deary, I.J. Schoon, I. & Gale, C.R. 2007. Childhood mental ability in relation to food intake and physical activity Excluded based on inclusion and exclusion criteria: cross
in adulthood: The 1970 British cohort study. Pediatrics, 119(1), e38-e45. sectional study.
Belfort, M.B. Rifas-Shiman, S.L. Kleinman, K.P. Guthrie, L.B. Bellinger, D.C. Taveras, E.M. Gillman, M.W. & Oken, E. 2013. Excluded based on inclusion and exclusion criteria: no
Infant Feeding and Childhood Cognition at Ages 3 and 7 Years Effects of Breastfeeding Duration and Exclusivity. JAMA data on fish (breastfeeding related to cognition at 3
Pediatrics, 167(9), 836-844. and 7 years)
Dangour, A. D. Allen, E. Elbourne, D. Fletcher, A. Richards. M. & Uauy, R. 2009. Fish consumption and cognitive function Excluded based on inclusion and exclusion criteria: cross
among older people in the UK: baseline data from the Opal Study. The Journal of Nutrition, Health and Aging, 13(3), sectional data
198-202.
Eskelinen, M.H. Ng, T.U., Helkala, E.L. Tuomilehto, J., Nissinen, A. Soininen, H. & Kivipelto, M. 2008. Fat intake at midlife Excluded based on inclusion and exclusion criteria: no
and cognitive impairment later in life: a population-based CAIDE study. International Journal of Geriatric Psychiatry: A data on fish
journal of the psychiatry of late life and allied sciences: 741-747.
Gatto, N.M. Garcia-Cano, J. Irani, C. Jaceldo-Siegl, K. Liu, T. Chen, Z. Paul, J. Fraser, G. Wang, C. & Lee, G.J. 2021. Excluded based on inclusion and exclusion criteria:
Vegetarian Dietary Patterns and Cognitive Function among Older Adults: The Adventist Health Study-2. Journal of dietary pattern study
Nutrition in Gerontology and Geriatrics, 40(4), 197-214.
Gonzalez, S. Huerta, J.M. Fernandez, Patterson, A.M. & Lasheras, C. 2010. The relationship between dietary lipids and Excluded based on inclusion and exclusion criteria:
cognitive performance in an elderly population. International Journal of Food Sciences and Nutrition, 61(2), 217-225. dietary lipids and no fish and cross sectional study
Handeland, K. , Skotheim, S. Baste, V. Graff, I.E. Froyl, L., Lie, O. Kjellevold, M. Markhus, M. Stormark, K.M. Oyen, J. & Dahl, Excluded based on inclusion and exclusion criteria:
L. 2018.The effects of fatty fish intake on adolescents' nutritional status and associations with attention performance: nutrient status in relation to attention
results from the FINS-TEENS randomized controlled trial. Nutrition Journal, 17, 1-12.
Hansen, A.L. Ambroziak, G. Thornton, D. Dahl, L. & Grung, B. 2018. Age and IQ Explained Working Memory Performance in Excluded based on inclusion and exclusion criteria:
a RCT with Fatty Fish in a Group of Forensic Inpatients. The Journal of Nutrition, Health and Aging, 22(4), 513-518. specific population group
Hansen, A.L. Olson, G. Dahl, L. Thornton, D. Grung, B. Graff, I.E. Frøyl, L. & Thayer, J.F. 2014. Reduced anxiety in forensic Excluded based on inclusion and exclusion criteria:
inpatients after a long-term intervention with Atlantic salmon. Nutrients, 6(12), 5405-5418. specific population group
Kwok, T.C.Y. Lam, L.C.W. Sea, M.M.M. Goggins, W. & Woo, J. 2012. A randomized controlled trial of dietetic interventions to Excluded based on inclusion and exclusion criteria: diet
prevent cognitive decline in old age hostel residents. European Journal of Clinical Nutrition, 66(10), 1135-1140 in general and primary outcome was cognitive decline
Luxwolda, M.F. Kuipers, R.S. Boersma, E.R. van Goor, S.A. Dijck-Brouwer, D.A. Bos, A.F. & Muskiet, F.A. 2014. DHA status Excluded based on inclusion and exclusion criteria: DHA
is positively related to motor development in breastfed African and Dutch infants. Nutritional Neuroscience, 17(3), in RBC related to motor development
97-103.
Miyake, Y. Sasaki, S,. Yokoyama, T. Tanaka, K. Ohya, Y. Fukushima, W. Saito, K. Ohfuji, S. & Kiyohara, C. 2006. Risk of Excluded based on inclusion and exclusion criteria:
postpartum depression in relation to dietary fish and fat intake in Japan: the Osaka Maternal and Child Health Study. investigated the relationship of high-fat foods
Psychological Medicine, 36(12), 1727-1735. and specific types of fatty acids with the risk of PP
depression.
Nisevic, J.R. Prpic, I. Kolic, I. Bazdaric, K. Tratnik, J.S. Prpic, I.S. Mazej, D. Spiric, Z. Barbone, F. & Horvat, M. 2019. Excluded based on inclusion and exclusion criteria:
Combined prenatal exposure to mercury and LCPUFA on newborn's brain measures and neurodevelopment at the age of mercury and n-3
18 months. Environmental Research, 178, 108682.
Poudel-Tandukar, Kalpana, et al. 2011. Long chain n-3 fatty acids intake, fish consumption and suicide in a cohort Excluded based on inclusion and exclusion criteria: n-3
of Japanese men and women—The Japan Public Health Center-based (JPHC) Prospective Study. Journal of Affective and fish in relation to suicide; wrong outcome
Disorders 129.1-3: 282-288.
Reeves, J. L., Otahal, P., Magnussen, C. G., Dwyer, T., Kangas, A. J., Soininen, P., & Smith, K. J. 2017. DHA mediates the Excluded based on inclusion and exclusion criteria: n-3
protective effect of fish consumption on new episodes of depression among women. British Journal of Nutrition, 118(9), and tyrosine in relation to depression
743-749.
Tiainen, A. M. K., Männistö, S., Lahti, M., Blomstedt, P. A., Lahti, J., Perälä, M. M., & Eriksson, J. G. 2013. Personality and Excluded based on inclusion and exclusion criteria:
dietary intake–findings in the Helsinki birth cohort study. PloS One, 8(7), e68284. dietary pattern and personality, cross sectional study
Timonen, M., Horrobin, D., Jokelainen, J., Laitinen, J., Herva, A., & Räsänen, P. 2004. Fish consumption and depression: the Excluded based on inclusion and exclusion criteria:
Northern Finland 1966 birth cohort study. Journal of Affective Disorders, 82(3), 447-452. brief report
Tsai, A. C., Lucas, M., Okereke, O. I., O'Reilly, É. J., Mirzaei, F., Kawachi, I., & Willett, W. C. 2014. Suicide mortality in Excluded based on inclusion and exclusion criteria: n-3
relation to dietary intake of n-3 and n-6 polyunsaturated fatty acids and fish: equivocal findings from 3 large US cohort and suicide
studies. American Journal of Epidemiology, 179(12), 1458-1466.
Vaz, J. D. S., Kac, G., Emmett, P., Davis, J. M., Golding, J., & Hibbeln, J. R. 2013. Dietary patterns, n-3 fatty acids intake Excluded based on inclusion and exclusion criteria:
from seafood and high levels of anxiety symptoms during pregnancy: findings from the Avon Longitudinal Study of dietary patterns and n-3 from seafood and anxiety
Parents and Children. PLoS One, 8(7), e67671.
Vuholm, S., Teisen, M. N., Mølgaard, C., Lauritzen, L., & Damsgaard, C. T. 2021. Sleep and physical activity in healthy Excluded based on inclusion and exclusion criteria: sleep
8–9-year-old children are affected by oily fish consumption in the FiSK Junior randomized trial. European Journal of is not an included outcome
Nutrition, 1-12.
Winpenny, E. M., van Harmelen, A. L., White, M., van Sluijs, E. M., & Goodyer, I. M. 2018. Diet quality and depressive Excluded based on inclusion and exclusion criteria: diet
symptoms in adolescence: no cross-sectional or prospective associations following adjustment for covariates. Public and fish data is cross-sectional data and depressive
Health Nutrition, 21(13), 2376-2384. symptoms
Yang, Y., Kim, Y., & Je, Y. 2018. Fish consumption and risk of depression: Epidemiological evidence from prospective Excluded based on inclusion and exclusion criteria:
studies. Asia‐Pacific Psychiatry, 10(4), e12335. review
Åberg, M. A., Åberg, N., Brisman, J., Sundberg, R., Winkvist, A., & Torén, K. 2009. Fish intake of Swedish male adolescents Excluded for further assessment, as the primary study
is a predictor of cognitive performance. Acta Paediatrica, 98(3), 555-560. had already been assessed in VKM 2022

360
APPENDICES

TABLE A3.15 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “NEURODEVELOPMENT
AND NEUROLOGICAL DISORDERS” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Appleton, K. M., Woodside, J. V., Yarnell, J. W. G., Arveiler, D., Haas, B., Amouyel, P. & PRIME Study Group. 2007. Depressed Excluded for further assessment, as the primary study
mood and dietary fish intake: direct relationship or indirect relationship as a result of diet and lifestyle?. Journal of had already been assessed in one of the included
Affective Disorders, 104(1-3), 217-223. systematic reviews
Appleton, K. M., Peters, T. J., Hayward, R. C., Heatherley, S. V., McNaughton, S. A., Rogers, P. J., & Kessler, D. 2007. Excluded for further assessment, as the primary study
Depressed mood and n-3 polyunsaturated fatty acid intake from fish: non-linear or confounded association?. Social had already been assessed in one of the included
Psychiatry and Psychiatric Epidemiology, 42, 100-104. systematic reviews
Astorg, P., Couthouis, A., Bertrais, S., Arnault, N., Meneton, P., Guesnet, P., & Hercberg, S. 2008. Association of fish and Excluded for further assessment, as the primary study
long-chain n-3 polyunsaturated fatty acid intakes with the occurrence of depressive episodes in middle-aged French men had already been assessed in VKM 2022
and women. Prostaglandins, Leukotrienes and Essential Fatty Acids, 78(3), 171-182.
Chuang, S. Y., Lo, Y. L., Wu, S. Y., Wang, P. N., & Pan, W. H. 2019. Dietary patterns and foods associated with cognitive Excluded for further assessment, as the primary study
function in Taiwanese older adults: the cross-sectional and longitudinal studies. Journal of the American Medical had already been assessed in VKM 2022
Directors Association, 20(5), 544-550.
Colangelo, L. A., He, K., Whooley, M. A., Daviglus, M. L., & Liu, K. 2009. Higher dietary intake of long-chain ω-3 Excluded for further assessment, as the primary study
polyunsaturated fatty acids is inversely associated with depressive symptoms in women. Nutrition, 25(10), 1011-1019. had already been assessed in VKM 2022
Daniels, J. L., Longnecker, M. P., Rowland, A. S., Golding, J., & ALSPAC Study Team. 2004. Fish intake during pregnancy Excluded for further assessment, as the primary study
and early cognitive development of offspring. Epidemiology, 15(4), 394-402. had already been assessed in VKM 2022
Davidson, P. W., Cory-Slechta, D. A., Thurston, S. W., Huang, L. S., Shamlaye, C. F., Gunzler, D., Myers, G.J. et al. 2011. Fish Excluded for further assessment, as the primary study
consumption and prenatal methylmercury exposure: cognitive and behavioral outcomes in the main cohort at 17 years had already been assessed in VKM 2022
from the Seychelles child development study. Neurotoxicology, 32(6), 711-717.
Demmelmair, H., Øyen, J., Pickert, T., Rauh-Pfeiffer, A., Stormark, K. M., Graff, I. E., Koletzko, B. et al. 2019. The effect of Excluded for further assessment, as the primary study
Atlantic salmon consumption on the cognitive performance of preschool children–a randomized controlled trial. Clinical had already been assessed in VKM 2022
Nutrition, 38(6), 2558-2568.
Devore, E. E., Grodstein, F., van Rooij, F. J., Hofman, A., Rosner, B., Stampfer, M. J., Breteler, M.M. et al. Dietary intake of Excluded for further assessment, as the primary study
fish and omega-3 fatty acids in relation to long-term dementia risk. The American Journal of Clinical Nutrition, 90(1), had already been assessed in VKM 2022
170-176.
Elstgeest, L. E., Visser, M., Penninx, B. W., Colpo, M., Bandinelli, S., & Brouwer, I. A. 2019. Bidirectional associations Excluded for further assessment, as the primary study
between food groups and depressive symptoms: longitudinal findings from the Invecchiare in Chianti (InCHIANTI) study. had already been assessed in VKM 2022
British Journal of Nutrition, 121(4), 439-450.
Fischer, K., Melo van Lent, D., Wolfsgruber, S., Weinhold, L., Kleineidam, L., Bickel, H., ... & Wagner, M. 2018. Prospective Excluded for further assessment, as the primary study
associations between single foods, Alzheimer’s dementia and memory decline in the elderly. Nutrients, 10(7), 852. had already been assessed in VKM 2022
Gale, C. R., Robinson, S. M., Godfrey, K. M., Law, C. M., Schlotz, W., & O’Callaghan, F. J. 2008. Oily fish intake during Excluded for further assessment, as the primary study
pregnancy–association with lower hyperactivity but not with higher full‐scale IQ in offspring. Journal of Child Psychology had already been assessed in VKM 2022
and Psychiatry, 49(10), 1061-1068.
Golding, J., Steer, C., Emmett, P., Davis, J. M., & Hibbeln, J. R. 2009. High levels of depressive symptoms in pregnancy with Excluded for further assessment, as the primary study
low omega-3 fatty acid intake from fish. Epidemiology, 20(4), 598-603. had already been assessed in one of the included
systematic reviews
Hamazaki, K., Matsumura, K., Tsuchida, A., Kasamatsu, H., Tanaka, T., Ito, M., & Inadera, H. 2020. Maternal dietary intake Excluded for further assessment, as the primary study
of fish and PUFAs and child neurodevelopment at 6 months and 1 year of age: a nationwide birth cohort—the Japan had already been assessed in VKM 2022
Environment and Children’s Study (JECS). American Journal of Clinical Nutrition, 112(5), 1295-1303.
Hamazaki, K., Matsumura, K., Tsuchida, A., Kasamatsu, H., Tanaka, T., Ito, M., & Inadera, H. 2020. Dietary intake of fish Excluded for further assessment, as the primary study
and n-3 polyunsaturated fatty acids and risk of postpartum depression: a nationwide longitudinal study–the Japan had already been assessed in VKM 2022
Environment and Children's Study (JECS). Psychological Medicine, 50(14), 2416-2424.
Handeland, K., Øyen, J., Skotheim, S., Graff, I. E., Baste, V., Kjellevold, M., Stormark, K. M. et al. 2017. Fatty fish intake Excluded for further assessment, as the primary study
and attention performance in 14–15 year old adolescents: FINS-TEENS-a randomized controlled trial. Nutrition Journal, had already been assessed in VKM 2022
16, 1-10.
Hansen, A. L., Dahl, L., Olson, G., Thornton, D., Grung, B., & Thayer, J. F. 2015. A long‐term fatty fish intervention improved Excluded for further assessment, as the primary study
executive function in inpatients with antisocial traits and a history of alcohol and drug abuse. Scandinavian Journal of had already been assessed in VKM 2022
Psychology, 56(5), 467-474.
Hedelin, M., Löf, M., Olsson, M., Lewander, T., Nilsson, B., Hultman, C. M., & Weiderpass, E. 2010. Dietary intake of fish, Excluded for further assessment, as the primary study
omega-3, omega-6 polyunsaturated fatty acids and vitamin D and the prevalence of psychotic-like symptoms in a cohort had already been assessed in VKM 2022
of 33 000 women from the general population. BMC Psychiatry, 10, 1-13.
Hibbeln, J. R., Davis, J. M., Steer, C., Emmett, P., Rogers, I., Williams, C., & Golding, J. 2007. Maternal seafood Excluded for further assessment, as the primary study
consumption in pregnancy and neurodevelopmental outcomes in childhood (ALSPAC study): an observational cohort study. had already been assessed in VKM 2022
The Lancet, 369(9561), 578-585.
Huang, T. L., Zandi, P. P., Tucker, K. L., Fitzpatrick, A. L., Kuller, L. H., Fried, L. P., Carlson, M.C. et al. 2005. Benefits of fatty Excluded for further assessment, as the primary study
fish on dementia risk are stronger for those without APOE ε4. Neurology, 65(9), 1409-1414. had already been assessed in VKM 2022
Hysing, M., Kvestad, I., Kjellevold, M., Kolden Midtbø, L., Graff, I. E., Lie, Ø., Øyen, J. et al. 2018. Fatty fish intake and the Excluded for further assessment, as the primary study
effect on mental health and sleep in preschool children in FINS-KIDS, a randomized controlled trial. Nutrients, 10(10), had already been assessed in VKM 2022
1478.
Julvez, J., Fernández-Barrés, S., Gignac, F., López-Vicente, M., Bustamante, M., Garcia-Esteban, R., Sunyer, J. et al. Excluded for further assessment, as the primary study
2020. Maternal seafood consumption during pregnancy and child attention outcomes: a cohort study with gene effect had already been assessed in VKM 2022
modification by PUFA-related genes. International Journal of Epidemiology, 49(2), 559-571.

361
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.15 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “NEURODEVELOPMENT
AND NEUROLOGICAL DISORDERS” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Julvez, J., Méndez, M., Fernandez-Barres, S., Romaguera, D., Vioque, J., Llop, S., Sunyer, J. et al. 2016. Maternal Excluded for further assessment, as the primary study
consumption of seafood in pregnancy and child neuropsychological development: a longitudinal study based on a had already been assessed in VKM 2022
population with high consumption levels. American Journal of Epidemiology, 183(3), 169-182.
Kalmijn, S., Launer, L. J., Ott, A., Witteman, J. C., Hofman, A., & Breteler, M. M. 1997. Dietary fat intake and the risk of Excluded for further assessment, as the primary study
incident dementia in the Rotterdam Study. Annals of Neurology, 42(5), 776-782. had already been assessed in VKM 2022
Kesse-Guyot, E., Peneau, S., Ferry, M., Jeandel, C., Hercberg, S., Galan, P.; SU.VI.Max 2 Research Group. 2011. Thirteen- Excluded for further assessment, as the primary study
year prospective study between fish consumption, long-chain n-3 fatty acids intakes and cognitive function. The Journal had already been assessed in VKM 2022
of Nutrition, Health & Aging, 15, 115-120.
Kim, D. H., Grodstein, F., Rosner, B., Kang, J. H., Cook, N. R., Manson, J. E., Okereke, O. I. et al. 2013. Seafood types and Excluded for further assessment, as the primary study
age-related cognitive decline in the Women’s Health Study. Journals of Gerontology Series A: Biomedical Sciences and had already been assessed in VKM 2022
Medical Sciences, 68(10), 1255-1262.
Kosti, R. I., Kasdagli, M. I., Kyrozis, A., Orsini, N., Lagiou, P., Taiganidou, F., & Naska, A. 2022. Fish intake, n-3 fatty acid Excluded for further assessment, as the primary study
body status, and risk of cognitive decline: a systematic review and a dose–response meta-analysis of observational and had already been assessed in a systematic review that
experimental studies. Nutrition Reviews, 80(6), 1445-1458. is included in VKM 2022
Kvestad, I., Hysing, M., Kjellevold, M., Næss, S., Dahl, L., & Markhus, M. W. 2021. Maternal cod intake during pregnancy Excluded for further assessment, as the primary study
and infant development in the first year of life: secondary analyses from a randomized controlled trial. The Journal of had already been assessed in VKM 2022
Nutrition, 151(7), 1879-1885.
Larrieu, S., Letenneur, L., Helmer, C., Dartigues, J. F., & Barberger-Gateau, P. 2004. Nutritional factors and risk of incident Excluded for further assessment, as the primary study
dementia in the PAQUID longitudinal cohort. The Journal of Nutrition, Health & Aging, 8(3), 150-154. had already been assessed in VKM 2022
Li, Y., Dai, Q., Ekperi, L. I., Dehal, A., & Zhang, J. 2011. Fish consumption and severely depressed mood, findings from the Excluded for further assessment, as the primary study
first national nutrition follow-up study. Psychiatry Research, 190(1), 103-109. had already been assessed in VKM 2022
Liu, J., Cui, Y., Li, L., Wu, L., Hanlon, A., Pinto-Martin, J., Hibbeln, J. R. et al. 2017. The mediating role of sleep in the fish Excluded for further assessment, as the primary study
consumption–cognitive functioning relationship: a cohort study. Scientific Reports, 7(1), 1-9. had already been assessed in VKM 2022
Lopez, L. B., Kritz-Silverstein, D., & Barrett-Connor, E. 2011. High dietary and plasma levels of the omega-3 fatty acid Excluded for further assessment, as the primary study
docosahexaenoic acid are associated with decreased dementia risk: the Rancho Bernardo study. The Journal of Nutrition, had already been assessed in VKM 2022
Health & Aging, 15, 25-31.
Lucas, M., Mirzaei, F., O’Reilly, E. J., Pan, A., Willett, W. C., Kawachi, I., Ascherio, A. et al. 2011. Dietary intake of n− 3 and Excluded for further assessment, as the primary study
n− 6 fatty acids and the risk of clinical depression in women: a 10-y prospective follow-up study. The American Journal of had already been assessed in one of the included
Clinical Nutrition, 93(6), 1337-1343. systematic reviews
Markhus, M. W., Hysing, M., Midtbø, L. K., Nerhus, I., Næss, S., Aakre, I., Kjellevold, M. et al. 2021. Effects of two weekly Excluded for further assessment, as the primary study
servings of cod for 16 weeks in pregnancy on maternal iodine status and infant neurodevelopment: Mommy's Food, a had already been assessed in VKM 2022
randomized-controlled trial. Thyroid, 31(2), 288-298.
Mendez, M. A., Torrent, M., Julvez, J., Ribas-Fitó, N., Kogevinas, M., & Sunyer, J. 2009. Maternal fish and other seafood Excluded for further assessment, as the primary study
intakes during pregnancy and child neurodevelopment at age 4 years. Public Health Nutrition, 12(10), 1702-1710. had already been assessed in VKM 2022
Morris, M.C., Evans, D.A., Tangney, C.C., Bienias, J.L., & Wilson, R.S. 2005. Fish consumption and cognitive decline with Excluded for further assessment, as the primary study
age in a large community study. Archives of Neurology, 62(12), 1849-1853. had already been assessed in VKM 2022
Morris, M. C., Evans, D. A., Bienias, J. L., Tangney, C. C., Bennett, D. A., Wilson, R. S., Schneider, J. et al. 2003. Excluded for further assessment, as the primary study
Consumption of fish and n-3 fatty acids and risk of incident Alzheimer disease. Archives of Neurology, 60(7), 940-946. had already been assessed in VKM 2022
Ngabirano, L., Samieri, C., Feart, C., Gabelle, A., Artero, S., Duflos, C., Mura, T. et al. 2019. Intake of meat, fish, fruits, and Excluded for further assessment, as the primary study
vegetables and long-term risk of dementia and Alzheimer’s disease. Journal of Alzheimer's Disease, 68(2), 711-722. had already been assessed in VKM 2022
Nooyens, A. C., Van Gelder, B. M., Bueno-de-Mesquita, H. B., Van Boxtel, M. P., & Verschuren, W. M. 2018. Fish Excluded for further assessment, as the primary study
consumption, intake of fats and cognitive decline at middle and older age: the Doetinchem Cohort Study. European had already been assessed in VKM 2022
Journal of Nutrition, 57, 1667-1675.
Nozaki, S., Sawada, N., Matsuoka, Y. J., Shikimoto, R., Mimura, M., & Tsugane, S. 2021. Association between dietary fish Excluded for further assessment, as the primary study
and PUFA intake in midlife and dementia in later life: the JPHC Saku Mental Health Study. Journal of Alzheimer's Disease, had already been assessed in VKM 2022
79(3), 1091-1104..
Oken, E., Østerdal, M. L., Gillman, M. W., Knudsen, V. K., Halldorsson, T. I., Strøm, M., Olsen, S. F. et al. 2008. Associations Excluded for further assessment, as the primary study
of maternal fish intake during pregnancy and breastfeeding duration with attainment of developmental milestones in had already been assessed in VKM 2022
early childhood: a study from the Danish National Birth Cohort. The American Journal of Clinical Nutrition, 88(3),
789-796.
Øyen, J., Kvestad, I., Midtbø, L. K., Graff, I. E., Hysing, M., Stormark, K. M., Kjellevold, M. et al. 2018. Fatty fish intake and Excluded for further assessment, as the primary study
cognitive function: FINS-KIDS, a randomized controlled trial in preschool children. BMC Medicine, 16, 1-15. had already been assessed in VKM 2022
Qin, B., Plassman, B. L., Edwards, L. J., Popkin, B. M., Adair, L. S., & Mendez, M. A. 2014. Fish intake is associated with Excluded for further assessment, as the primary study
slower cognitive decline in Chinese older adults. The Journal of Nutrition, 144(10), 1579-1585. had already been assessed in VKM 2022
Samieri, C., Morris, M. C., Bennett, D. A., Berr, C., Amouyel, P., Dartigues, J. F., Grodstein, F. et al. 2018. Fish intake, Excluded for further assessment, as the primary study
genetic predisposition to Alzheimer disease, and decline in global cognition and memory in 5 cohorts of older persons. had already been assessed in VKM 2022
American Journal of Epidemiology, 187(5), 933-940.
Sanchez-Villegas, A., Henríquez, P., Figueiras, A., Ortuño, F., Lahortiga, F., & Martínez-González, M. A. 2007. Long chain Excluded for further assessment, as the primary study
omega-3 fatty acids intake, fish consumption and mental disorders in the SUN cohort study. European Journal of had already been assessed in VKM 2022
Nutrition, 46, 337-346..

362
APPENDICES

TABLE A3.15 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “NEURODEVELOPMENT
AND NEUROLOGICAL DISORDERS” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Skotheim, S., Handeland, K., Kjellevold, M., Øyen, J., Frøyland, L., Lie, Ø., Dahl, L. et al. 2017. The effect of school Excluded for further assessment, as the primary study
meals with fatty fish on adolescents’ self-reported symptoms for mental health: FINS-TEENS-a randomized controlled had already been assessed in VKM 2022
intervention trial. Food & Nutrition Research, 12;61(1)
Smith, K. J., Sanderson, K., McNaughton, S. A., Gall, S. L., Dwyer, T., & Venn, A. J. 2014. Longitudinal associations between Excluded for further assessment, as the primary study
fish consumption and depression in young adults. American Journal of Epidemiology, 179(10), 1228-1235. had already been assessed in VKM 2022
Solfrizzi, V., Custodero, C., Lozupone, M., Imbimbo, B. P., Valiani, V., Agosti, P., Panza, F. et al. 2017. Relationships of Excluded for further assessment, as the primary study
dietary patterns, foods, and micro-and macronutrients with Alzheimer’s disease and late-life cognitive disorders: a had already been assessed in VKM 2022
systematic review. Journal of Alzheimer's Disease, 59(3), 815-849.
Strøm, M., Mortensen, E. L., Halldorsson, T. I., Thorsdottir, I., & Olsen, S. F. 2009. Fish and long-chain n-3 polyunsaturated Excluded for further assessment, as the primary study
fatty acid intakes during pregnancy and risk of postpartum depression: a prospective study based on a large national had already been assessed in VKM 2022
birth cohort. The American Journal of Clinical Nutrition, 90(1), 149-155.
Teisen, M. N., Vuholm, S., Niclasen, J., Aristizabal-Henao, J. J., Stark, K. D., Geertsen, S. S., Lauritzen, L. et al. 2020. Excluded for further assessment, as the primary study
Effects of oily fish intake on cognitive and socioemotional function in healthy 8–9-year-old children: The FiSK Junior had already been assessed in VKM 2022
randomized trial. The American Journal of Clinical Nutrition, 112(1), 74-83.
Tsurumaki, N., Zhang, S., Tomata, Y., Abe, S., Sugawara, Y., Matsuyama, S., & Tsuji, I. 2019. Fish consumption and risk of Excluded for further assessment, as the primary study
incident dementia in elderly Japanese: the Ohsaki cohort 2006 study. British Journal of Nutrition, 122(10), 1182-1191. had already been assessed in VKM 2022
Valent, F., Mariuz, M., Bin, M., Mazej, D., Tognin, V., Tratnik, J., Barbone, F. et al. 2013. Associations of prenatal mercury Excluded for further assessment, as the primary study
exposure from maternal fish consumption and polyunsaturated fatty acids with child neurodevelopment: a prospective had already been assessed in VKM 2022
cohort study in Italy. Journal of Epidemiology, 23(5), 360-370.
van de Rest, O., Spiro III, A., Krall-Kaye, E., Geleijnse, J. M., de Groot, L. C., & Tucker, K. L. 2009. Intakes of (n-3) fatty Excluded for further assessment, as the primary study
acids and fatty fish are not associated with cognitive performance and 6-year cognitive change in men participating in had already been assessed in VKM 2022
the Veterans Affairs Normative Aging Study. The Journal of Nutrition, 139(12), 2329-2336.
van Gelder, B. M., Tijhuis, M., Kalmijn, S., & Kromhout, D. 2007. Fish consumption, n− 3 fatty acids, and subsequent 5-y Excluded for further assessment, as the primary study
cognitive decline in elderly men: the Zutphen Elderly Study. The American Journal of Clinical Nutrition, 85(4), 1142-1147. had already been assessed in VKM 2022
Vecchione, R., Vigna, C., Whitman, C., Kauffman, E. M., Braun, J. M., Chen, A., Lyall, K. et al. 2021. The association Excluded for further assessment, as the primary study
between maternal prenatal fish intake and child autism-related traits in the EARLI and HOME Studies. Journal of Autism had already been assessed in VKM 2022
and Developmental Disorders, 51, 487-500.
Yang, Y., Kim, Y., & Je, Y. 2018. Fish consumption and risk of depression: Epidemiological evidence from prospective Excluded for further assessment, as the primary study
studies. Asia‐Pacific Psychiatry, 10(4), e12335. had already been assessed in a systematic review that
is included in VKM 2022
Zhou, F., Wu, F., Zou, S., Chen, Y., Feng, C., & Fan, G. 2016. Dietary, nutrient patterns and blood essential elements in Excluded for further assessment, as the primary study
Chinese children with ADHD. Nutrients, 8(6), 352. had already been assessed in VKM 2022

363
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

MORTALITY
TABLE A3.16 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “MORTALITY” BASED ON
INCLUSION AND EXCLUSION CRITERIA
Study (n = 19) Reason for exclusion
Zhang, B. Xiong, K. Cai, J. & Ma, A.G. 2020. Fish Consumption and Coronary Heart Disease: A Meta-Analysis. Nutrients, Excluded, as the review has already been assessed in
12(8), 2278 VKM 2022
Zhang, Z. Chen, G.C. Qin, Z.Z. Tong, X. Li, D.P. & Qin, L.Q. 2018. Poultry and Fish Consumption in Relation to Total Cancer Excluded, as the review has already been assessed in
Mortality: A Meta-Analysis of Prospective Studies. Nutrition and Cancer-an International Journal, 70(2), 204-212. VKM 2022
Zhao, L.G. Sun, J.W. Yang, Y. Ma, X. Wang, Y.Y. & Xiang, Y.B. 2016. Fish consumption and all-cause mortality: a meta- Excluded, as the review has already been assessed in
analysis of cohort studies. European Journal of Clinical Nutrition, 70(2):155-161. VKM 2022
Jiang, L. Wang, J.Y. Xiong, K. Xu, L. Zhang, B. & Ma, A.G. 2021. Intake of Fish and Marine n-3 Polyunsaturated Fatty Acids Excluded, as the review has already been assessed in
and Risk of Cardiovascular Disease Mortality: A Meta-Analysis of Prospective Cohort Studies. Nutrients, 13(7):2342. VKM 2022
Jayedi, A. Shab-Bidar, S. Eimeri, S. & Djafarian, K. 2018. Fish consumption and risk of all-cause and cardiovascular Excluded, as the review has already been assessed in
mortality: a dose-response meta-analysis of prospective observational studies. Public Health Nutrition, 21(7):1297-1306. VKM 2022
Schwingshackl, L. Schwedhelm, C. Hoffmann, G. Lampousi, A.M. Knuppel, S. Iqbal, K. Bechthold, A. Schlesinger, S. & Excluded, as the review has already been assessed in
Boeing, H. 2017. Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective VKM 2022
studies. American Journal of Clinical Nutrition, 105(6):1462-1473.
Kwok, C.S. Gulati, M. Michos, E.D. Potts, J. Wu, P. Watson, L. Loke, Y.K. Mallen, C. & Mamas, M.A. 2019. Dietary Excluded, as the review has already been assessed in
components and risk of cardiovascular disease and all-cause mortality: a review of evidence from meta-analyses. VKM 2022
European Journal of Preventive Cardiology, 26(13):1415-1429.
Li, N. Wu, X.T. Zhuang, W. Xia, L. Chen, Y. Wu, C.C. Rao, Z.Y. Du, L. Zhao, R. Yi, M.S. et al. 2020. Fish consumption and Excluded, as the review has already been assessed in
multiple health outcomes: Umbrella review. Trends in Food Science & Technology, 99:273-283. VKM 2022
Mozaffarian, D. & Rimm, E.B. 2006. Fish intake, contaminants, and human health - Evaluating the risks and the benefits. Excluded, as the review has already been assessed in
Jama-Journal of the American Medical Association, 296(15):1885-1899. VKM 2022
Wan, Y. Zheng, J.S. Wang, F.L. & Li, D. 2017. Fish, long chain omega-3 polyunsaturated fatty acids consumption, and risk Excluded, as the review has already been assessed in
of all-cause mortality: a systematic review and dose-response meta-analysis from 23 independent prospective cohort VKM 2022
studies. Asia Pacific Journal of Clinical Nutrition, 26(5):939-956.
Jayedi, A. & Shab-Bidar, S. 2020. Fish Consumption and the Risk of Chronic Disease: An Umbrella Review of Meta- Excluded, as the review has already been assessed in
Analyses of Prospective Cohort Studies. Advances in Nutrition, 11(5):1123-1133. VKM 2022
Wang, C.C. Harris, W.S. Chung, M. Lichtenstein, A.H. Balk, E.M. Kupelnick, B. Jordan, H.S. & Lau, J. 2006. N-3 fatty acids Excluded based on inclusion and exclusion criteria:
from fish or fish-oil supplements, but not alpha-linolenic acid, benefit cardiovascular disease outcomes in primary- and fish oil
secondary-prevention studies: a systematic review. American Journal of Clinical Nutrition, 84(1):5-17.
Seyedrezazadeh, E. Moghaddam, M.P. Ansarin, K. Asghari Jafarabadi, M. Sharifi, A. Sharma, S. & Kolahdooz, F. 2019. Excluded based on inclusion and exclusion criteria:
Dietary Factors and Risk of Chronic Obstructive Pulmonary Disease: a Systemic Review and Meta-Analysis. Tanaffos, wrong outcome (COPD)
18(4):294-309.
Mozaffarian, D. 2008. Fish and n-3 fatty acids for the prevention of fatal coronary heart disease and sudden cardiac Excluded based on inclusion and exclusion criteria:
death. American Journal of Clinical Nutrition, 87(6):1991S-1996S.
de Lorgeril, M. Salen, P. Defaye, P. Mabo, P. & Paillard, F. 2001. Dietary prevention of sudden cardiac death. European Excluded based on inclusion and exclusion criteria:
Heart Journal, 23(4):277-285.
Aucoin, M. Cooley, K. Knee, C. Fritz, H. Balneaves, L.G. Breau, R. Fergusson, D. Skidmore, B. Wong, R. & Seely, D. 2016. Excluded based on inclusion and exclusion criteria:
Fish-Derived Omega-3 Fatty Acids and Prostate Cancer: A Systematic Review. Integrative Cancer Therapies, 16(1):32-62. omega-3
Bucher, H.C. Hengstler, P. Schindler, C. & Meier, G. 2002. N-3 polyunsaturated fatty acids in coronary heart disease: A Excluded based on inclusion and exclusion criteria:
meta-analysis of randomized controlled trials. American Journal of Medicine, 112(4):298-304. omega-3
Zheng, J.S. Huang, T. Yu, Y.H. Hu, X.J. Yang, B. & Li, D. 2011. Fish consumption and CHD mortality: an updated meta- Excluded based on inclusion and exclusion criteria:
analysis of seventeen cohort studies. Public Health Nutrition, 15(4):725-737. review article
Nutrition Reviews. 1985. Mortality from Coronary Heart Disease Is Inversely Related to Fish Consumption in the Excluded based on inclusion and exclusion criteria:
Netherlands, Nutrition Reviews, 43(9):271–273. review

364
APPENDICES

TABLE A3.17 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “MORTALITY”
Study (n = 85) Reason for exclusion
Appleby, P.N., Key, T.J., Thorogood, M., Burr, M.L. & Mann, J. 2002. Mortality in British vegetarians. Public Health Nutr, Excluded based on inclusion and exclusion criteria:
5(1): 29-36. specific population group
Bergdahl, I.A., Ahlqwist, M., Barregard, L., Bjorkelund, C., Blomstrand, A., Skerfving, S., Sundh, V., Wennberg, M. Excluded based on inclusion and exclusion criteria: main
& Lissner, L. 2013. Mercury in serum predicts low risk of death and myocardial infarction in Gothenburg women. outcome is mortality related to mercury in serum, and
International Archives of Occupational and Environmental Health, 86(1): 71-77. fish consumption is only a covariate
Biesbroek, S., Bueno-de-Mesquita, H.B., Peeters, P.H.M., Verschuren, W.M.M., van der Schouw, Y.T., Kramer, G.F.H., Tyszler, Excluded based on inclusion and exclusion criteria:
M. & Temme, E.H.M. 2014. Reducing our environmental footprint and improving our health: greenhouse gas emission and mortality is only directly associated with greenhouse gas
land use of usual diet and mortality in EPIC-NL: a prospective cohort study. Environmental Health, 13. emission, and not fish intake
Bonaccio, M., Ruggiero, E., Di Castelnuovo, A., Costanzo, S., Persichillo, M., De Curtis, A., Cerletti, C. et al. 2017. Fish Excluded based on inclusion and exclusion criteria:
intake is associated with lower cardiovascular risk in a Mediterranean population: Prospective results from the Moli-sani main outcome of fish intake is associated with CHD and
study. Nutrition Metabolism and Cardiovascular Diseases, 27(10): 865-873. stroke, not mortality
Breslow, R.A., Graubard, B.I., Sinha, R. & Subar, A.F. 2000. Diet and lung cancer mortality: a 1987 National Health Excluded based on inclusion and exclusion criteria: study
Interview Survey cohort study. Cancer Causes & Control, 11(5): 419-431. does not discriminate between meat/poultry/fish
de Goede, J., Verschuren, W.M.M., Boer, J.M.A., Kromhout, D. & Geleijnse, J.M. 2012. Gender-Specific Associations of Marine Excluded based on inclusion and exclusion criteria: main
n-3 Fatty Acids and Fish Consumption with 10-Year Incidence of Stroke. Plos One, 7(4). outcome is fish intake associated with non-fatal stroke
Donat-Vargas, C., Bellavia, A., Berglund, M., Glynn, A., Wolk, A. & Akesson, A. 2020. Cardiovascular and cancer mortality Excluded based on inclusion and exclusion criteria:
in relation to dietary polychlorinated biphenyls and marine polyunsaturated fatty acids: a nutritional-toxicological aspect mortality is associated with dietary polychlorinated
of fish consumption. Journal of Internal Medicine, 287(2): 197-209. biphenyls and marine polyunsaturated fatty acids
Erkkila, A.T., Lehto, S., Pyorala, K. & Uusitupa, M.I.J. 2003. n-3 fatty acids and 5-y risks of death and cardiovascular Excluded based on inclusion and exclusion criteria:
disease events in patients with coronary artery disease. American Journal of Clinical Nutrition, 78(1): 65-71. mortality is associated with n-3 fatty acids in serum
lipids, not fish consumption
Fresan, U., Martinez-Gonzalez, M.A., Segovia-Siapco, G., Sabate, J. & Bes-Rastrollo, M. 2020. A three-dimensional dietary Excluded based on inclusion and exclusion criteria:
index (nutritional quality, environment and price) and reduced mortality: The "Seguimiento Universidad de Navarra" based on dietary pattern, the sustainable diet index
cohort. Preventive Medicine, 137.
Holmberg, S., Thelin, A. & Stiernstrom, E.L. 2009. Food Choices and Coronary Heart Disease: A Population Based Cohort Excluded based on inclusion and exclusion criteria: main
Study of Rural Swedish Men with 12 Years of Follow-up. International Journal of Environmental Research and Public outcome is risk of coronary heart disease, and it is not
Health, 6(10): 2626-2638. specified by mortality
Ikeda, M., Yoshimoto, K., Yoshimura, T., Kono, S., Kato, H. & Kuratsune, M. 1983. A cohort study on the possible Excluded based on inclusion and exclusion criteria: study
association between broiled fish intake and cancer. Gan, 74(5): 640-8. only investigate intake of broiled fish
Khankari, N.K., Bradshaw, P.T., Steck, S.E., He, K., Olshan, A.F., Shen, J., Ahn, J. et al. 2015. Dietary intake of fish, Excluded based on inclusion and exclusion criteria:
polyunsaturated fatty acids, and survival after breast cancer: A population-based follow-up study on Long Island, New breast cancer patients
York. Cancer, 121(13): 2244-52.
Konig, A., Bouzan, C., Cohen, J.T., Connor, W.E., Kris-Etherton, P.M., Gray, G.M., Lawrence, R.S., Savitz, D.A. & Teutsch, S.M. Excluded based on inclusion and exclusion criteria:
2005. A quantitative analysis of fish consumption and coronary heart disease mortality. American Journal of Preventive narrative review
Medicine, 29(4): 335-346.
Krieger, J.P., Cabaset, S., Pestoni, G., Rohrmann, S., Faeh, D. & Swiss Natl Cohort Study, G. 2018. Dietary Patterns Are Excluded based on inclusion and exclusion criteria:
Associated with Cardiovascular and Cancer Mortality among Swiss Adults in a Census-Linked Cohort. Nutrients, 10(3). dietary patterns
Kromhout, D., Bloemberg, B.P.M., Feskens, E.J.M., Hertog, M.G.L., Menotti, A. & Blackburn, H. 1996. Alcohol, fish, fibre Excluded based on inclusion and exclusion criteria:
and antioxidant vitamins intake do not explain population differences in coronary heart disease mortality. International dietary intake represents the availability of foods per
Journal of Epidemiology, 25(4): 753-759. country and not the foods actually consumed
Kutner, N.G., Clow, P.W., Zhang, R. & Aviles, X. 2002. Association of fish intake and survival in a cohort of incident dialysis Excluded based on inclusion and exclusion criteria:
patients. American Journal of Kidney Diseases, 39(5): 1018-1024. specific patient group, survival in dialysis patients is
investigated
Leo, Q.J.N., Ollberding, N.J., Wilkens, L.R., Kolonel, L.N., Henderson, B.E., Le Marchand, L. & Maskarinec, G. 2016. Excluded based on inclusion and exclusion criteria: non-
Nutritional factors and non-Hodgkin lymphoma survival in an ethnically diverse population: the Multiethnic Cohort. Hodgkin’s lymphoma (NHL) survival is investigated
European Journal of Clinical Nutrition, 70(1): 41-46.
Levitan, E.B., Wolk, A. & Mittleman, M.A. 2009. Fish consumption, marine omega-3 fatty acids, and incidence of heart Excluded based on inclusion and exclusion criteria: main
failure: a population-based prospective study of middle-aged and elderly men. European Heart Journal, 30(12): 1495- outcome is heart failure, and it does not discriminate
1500. between death and not
Levitan, E.B., Wolk, A. & Mittleman, M.A. 2010. Fatty fish, marine omega-3 fatty acids and incidence of heart failure. Excluded based on inclusion and exclusion criteria: main
European Journal of Clinical Nutrition, 64(6): 587-594. outcome is heart failure, and it does not discriminate
between death and not
Menotti, A., Kromhout, D., Blackburn, H., Fidanza, F., Buzina, R., Nissinen, A. & Seven Countries Study Res, G. 1999. Food Excluded based on inclusion and exclusion criteria:
intake patterns and 25-year mortality from coronary heart disease: Cross-cultural correlations in the Seven Countries dietary patterns
Study. European Journal of Epidemiology, 15(6): 507-515.
Morris, M.C. Manson, J.E. Rosner, B. Buring, J.E. Willett, W.C. & Hennekens, C.H. 1995. Fish consumption and Excluded based on inclusion and exclusion criteria:
cardiovascular disease in the Physician’s Health Study – a prospective study. American Journal of Epidemiology, cardiovascular disease and not mortality
142(2):166-175.
Nettleton, J.A., Steffen, L.M., Loehr, L.R., Rosamond, W.D. & Folsom, A.R. 2008. Incident Heart Failure Is Associated with Excluded based on inclusion and exclusion criteria:
Lower Whole-Grain Intake and Greater High-Fat Dairy and Egg Intake in the Atherosclerosis Risk in Communities (ARIC) outcome is heart failure and does not discriminate
Study. Journal of the American Dietetic Association, 108(11): 1881-1887. between fatal and non-fatal

365
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.17 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “MORTALITY” (cont.)
Ohira, T., Iso, H., Yamagishi, K. & Tamakoshi, A. 2011. Fish, vegetable, and fruit intakes and mortality from pulmonary Excluded based on inclusion and exclusion criteria
embolism among japanese men and women: the jacc (japan collaborative cohort study for evaluation of cancer risk)
study. Journal of Epidemiology and Community Health, 65: A284-A284.
Ohira, T., Iso, H., Yamagishi, K., Tamakoshi, A. & Grp, J.S. 2018. Fish Intake and Death From Pulmonary Embolisms Among Excluded based on inclusion and exclusion criteria:
Japanese Men and Women - The Japan Collaborative Cohort (JACC) Study. Circulation Journal, 82(8): 2063-2070. abstract from congress
Outzen, M., Tjonneland, A., Christensen, J. & Olsen, A. 2018. Fish consumption and prostate cancer risk and mortality in a Excluded based on inclusion and exclusion criteria:
Danish cohort study. European Journal of Cancer Prevention, 27(4): 355-360. mortality in persons diagnosed with prostate cancer
Pan, A., Sun, Q., Bernstein, A.M., Schulze, M.B., Manson, J.E., Stampfer, M.J., Willett, W.C. & Hu, F.B. 2012. Red Meat Excluded based on inclusion and exclusion criteria:
Consumption and Mortality Results From 2 Prospective Cohort Studies. Archives of Internal Medicine, 172(7): 555-563. investigates red meat and the replacement of red meat
with fish, not fish intake solely
Papier, K., Appleby, P.N., Fensom, G.K., Knuppel, A., Perez-Cornago, A., Schmidt, J.A., Tong, T.Y.N. & Key, T.J. 2019. Excluded based on inclusion and exclusion criteria:
Vegetarian diets and risk of hospitalisation or death with diabetes in British adults: results from the EPIC-Oxford study. vegetarians
Nutrition & Diabetes, 9.
Petermann-Rocha, F., Parra-Soto, S., Gray, S., Anderson, J., Welsh, P., Gill, J., Sattar, N., Ho, F.K., Celis-Morales, C. & Excluded based on inclusion and exclusion criteria:
Pell, J.P. 2021. Vegetarians, fish, poultry, and meat-eaters: who has higher risk of cardiovascular disease incidence and vegetarians
mortality? A prospective study from UK Biobank. European Heart Journal, 42(12): 1136-1143.
Playdon, M.C., Nagle, C.M., Ibiebele, T.I., Ferrucci, L.M., Protani, M.M., Carter, J., Hyde, S.E. et al. 2017. Pre-diagnosis diet Excluded based on inclusion and exclusion criteria:
and survival after a diagnosis of ovarian cancer. British Journal of Cancer, 116(12): 1627-1637. survival after ovarian cancer diagnosis
Poudel-Tandukar, K., Nanri, A., Iwasaki, M., Mizoue, T., Matsushita, Y., Takahashi, Y., Noda, M., Inoue, M., Tsugane, S. Excluded based on inclusion and exclusion criteria:
& Japan Public Hlth Ctr-Based, P. 2011. Long chain n-3 fatty acids intake, fish consumption and suicide in a cohort suicide as outcome
of Japanese men and women - The Japan Public Health Center-based (JPHC) Prospective Study. Journal of Affective
Disorders, 129(1-3): 282-288.
Rodriguez, B.L., Sharp, D.S., Abbott, R.D., Burchfiel, C.M., Masaki, K., Chyou, P.H., Huang, B., Yano, K. & Curb, J.D. 1996. Excluded based on inclusion and exclusion criteria
Fish intake may limit the increase in risk of coronary heart disease morbidity and mortality among heavy smokers. The
Honolulu Heart Program. Circulation, 94(5): 952-6.
Sala-Vila, A., Guasch-Ferré, M., Hu, F.B., Sánchez-Tainta, A., Bulló, M., Serra-Mir, M., López-Sabater, C. et al. 2016. Dietary Excluded based on inclusion and exclusion criteria: focus
α-Linolenic Acid, Marine ω-3 Fatty Acids, and Mortality in a Population With High Fish Consumption: Findings From the on LCn-3 PUFA
PREvención con DIeta MEDiterránea (PREDIMED) Study. J Am Heart Assoc, 5(1).
Sasakabe, T., Wakai, K., Ukawa, S., Ando, M., Kawamura, T., Okabayashi, S., Tsushita, K., Ohira, H. & Tamakoshi, A. Excluded based on inclusion and exclusion criteria: fish
2021. Food group intakes and all-cause mortality among a young older Japanese population of the same age: the New intake in relation to smoking
Integrated Suburban Seniority Investigation Project. Nagoya Journal of Medical Science, 83(1): 169-182.
Sotomayor, C.G., Gomes-Neto, A.W., Gans, R.O.B., de Borst, M.H., Berger, S.P., Rodrigo, R., Navis, G.J., Touw, D.J. & Bakker, Excluded based on inclusion and exclusion criteria:
S.J.L. 2018. Fish Intake, Circulating Mercury and Mortality in Renal Transplant Recipients. Nutrients, 10(10). evaluates the impact from circulating mercury on
impact from fish intake on mortality in renal transplant
recipients
Sun, Y.B., Liu, B.Y., Snetselaar, L.G., Robinson, J.G., Wallace, R.B., Peterson, L.L. & Bao, W. 2019. Association of fried food Excluded based on inclusion and exclusion criteria:
consumption with all cause, cardiovascular, and cancer mortality: prospective cohort study. British Medical Journal, 364. investigates the impact of fried food consumption
Tsai, A.C., Lucas, M., Okereke, O.I., O'Reilly, E.J., Mirzaei, F., Kawachi, I., Ascherio, A. & Willett, W.C. 2014. Suicide Mortality Excluded based on inclusion and exclusion criteria:
in Relation to Dietary Intake of n-3 and n-6 Polyunsaturated Fatty Acids and Fish: Equivocal Findings From 3 Large US suicide as outcome
Cohort Studies. American Journal of Epidemiology, 179(12): 1458-1466.
Van Blarigan, E.L., Fuchs, C.S., Niedzwiecki, D., Ye, X., Zhang, S., Song, M., Saltz, L.B. et al. 2018. Marine ω-3 Excluded based on inclusion and exclusion criteria: colon
Polyunsaturated Fatty Acid and Fish Intake after Colon Cancer Diagnosis and Survival: CALGB 89803 (Alliance). Cancer cancer recurrence/survival
Epidemiol Biomarkers Prev, 27(4): 438-445.
Wang, Y., Jacobs, E.J., Shah, R.A., Stevens, V.L., Gansler, T. & McCullough, M.L. 2020. Red and Processed Meat, Poultry, Excluded based on inclusion and exclusion criteria:
Fish, and Egg Intakes and Cause-Specific and All-Cause Mortality among Men with Nonmetastatic Prostate Cancer in a cancer survival
US Cohort. Cancer Epidemiology Biomarkers & Prevention, 29(5): 1029-1038.
Whelton, S.P., He, J., Whelton, P.K. & Muntner, P. 2004. Meta-analysis of observational studies on fish intake and coronary Excluded based on inclusion and exclusion criteria: no
heart disease. American Journal of Cardiology, 93(9): 1119-1123. CVD patients
Zhong, V.W., Allen, N.B., Greenland, P., Carnethon, M.R., Ning, H.Y., Wilkins, J.T., Lloyd-Jones, D.M. & Van Horn, L. 2021. Excluded based on inclusion and exclusion criteria:
Protein foods from animal sources, incident cardiovascular disease and all-cause mortality: a substitution analysis. pooled analysis of six prospective cohort studies of 29
International Journal of Epidemiology, 50(1): 223-233. 682 US participants
Albert, C.M., Hennekens, C.H., O'Donnell, C.J., Ajani, U.A., Carey, V.J., Willett, W.C., Ruskin, J.N. & Manson, J.E. 1998. Fish Excluded, as the review has already been assessed in
consumption and risk of sudden cardiac death. JAMA, 279(1): 23-28. VKM 2022.
Ascherio, A., Rimm, E.B., Stampfer, M.J., Giovannucci, E.L. & Willett, W.C. 1995. Dietary intake of marine n-3 fatty acids, Excluded, as the review has already been assessed in
fish intake, and the risk of coronary disease among men. N Engl J Med, 332(15): 977-82. VKM 2022.
Bellavia et al. 2017. Fish consumption and all-cause mortality in a cohort of Swedish men and women. Excluded, as the review has already been assessed in
VKM 2022.
Chavarro, J.E., Stampfer, M.J., Hall, M.N., Sesso, H.D. & Ma, J. 2008. A 22-y prospective study of fish intake in relation to Excluded for further assessment, as the primary study
prostate cancer incidence and mortality. American Journal of Clinical Nutrition, 88(5): 1297-1303 had already been assessed in one of the included
systematic reviews (Szymanski et al., 2010)
Daviglus, M.L., Stamler, J., Orencia, A.J., Dyer, A.R., Liu, K., Greenland, P., Walsh, M.K., Morris, D. & Shekelle, R.B. 1997. Excluded, as the review has already been assessed in
Fish consumption and the 30-year risk of fatal myocardial infarction. N Engl J Med, 336(15): 1046-53. VKM 2022.

366
APPENDICES

TABLE A3.17 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “MORTALITY” (cont.)
de Goede, J., Geleijnse, J.M., Boer, J.M.A., Kromhout, D. & Verschuren, W.M.M. 2010. Marine (n-3) Fatty Acids, Fish Excluded, as the review has already been assessed in
Consumption, and the 10-Year Risk of Fatal and Nonfatal Coronary Heart Disease in a Large Population of Dutch Adults VKM 2022.
with Low Fish Intake. Journal of Nutrition, 140(5): 1023-1028
Deng, A., Pattanaik, S., Bhattacharya, A., Yin, J., Ross, L., Liu, C. & Zhang, J. 2018. Fish consumption is associated with Excluded, as the review has already been assessed in
a decreased risk of death among adults with diabetes: 18-year follow-up of a national cohort. Nutrition Metabolism and VKM 2022.
Cardiovascular Diseases, 28(10): 1012-1020
Engeset, D., Braaten, T., Teucher, B., Kuhn, T., Bueno-de-Mesquita, H., Leenders, M., Agudo, A. et al. 2015. Fish Excluded, as the review has already been assessed in
consumption and mortality in the European Prospective Investigation into Cancer and Nutrition cohort. European Journal VKM 2022.
of Epidemiology, 30(1): 57-70.
Farvid, M.S., Malekshah, A.F., Pourshams, A., Poustchi, H., Sepanlou, S.G., Sharafkhah, M., Khoshnia, M. et al. 2017. Excluded, as the review has already been assessed in
Dietary Protein Sources and All-Cause and Cause-Specific Mortality: The Golestan Cohort Study in Iran. American Journal VKM 2022.
of Preventive Medicine, 52(2): 237-248..
Feskens, E.J., Bowles, C.H. & Kromhout, D. 1993. Association between fish intake and coronary heart disease mortality. Excluded for further assessment, as the primary study
Differences in normoglycemic and glucose intolerant elderly subjects. Diabetes Care, 16(7): 1029-34. had already been assessed in a systematic review that
is included in VKM 2022 (Jayedi et al., 2020)
Folsom, A.R. & Demissie, Z. 2004. Fish intake, marine omega-3 fatty acids, and mortality in a cohort of postmenopausal Excluded, as the review has already been assessed in
women. American Journal of Epidemiology, 160(10): 1005-1010. VKM 2022.
Gillum, R.F., Mussolino, M.E. & Madans, J.H. 1996. The relationship between fish consumption and stroke incidence – The Excluded, as the review has already been assessed in
NHANES I epidemiologic follow-up study. Archives of Internal Medicine, 156(5): 537-542. VKM 2022.
Hengeveld, L.M., Praagman, J., Beulens, J.W.J., Brouwer, I.A., van der Schouw, Y.T. & Sluijs, I. 2018. Fish consumption and Excluded, as the review has already been assessed in
risk of stroke, coronary heart disease, and cardiovascular mortality in a Dutch population with low fish intake. European VKM 2022.
Journal of Clinical Nutrition, 72(7): 942-950.
Hu, F.B. & Willett, W.C. 2002. Optimal diets for prevention of coronary heart disease. Jama-Journal of the American Excluded, as the review has already been assessed in
Medical Association, 288(20): 2569-2578.. VKM 2022.
Jarvinen, R., Knekt, P., Rissanen, H. & Reunanen, A. 2006. Intake of fish and long-chain n-3 fatty acids and the risk of Excluded, as the review has already been assessed in
coronary heart mortality in men and women. British Journal of Nutrition, 95(4): 824-829 VKM 2022.
Jayedi, A., Soltani, S., Abdolshahi, A. & Shab-Bidar, S. 2021. Fish consumption and the risk of cardiovascular disease and Excluded, as the review has already been assessed in
mortality in patients with type 2 diabetes: a dose-response meta-analysis of prospective cohort studies. Critical Reviews VKM 2022.
in Food Science and Nutrition, 61(10): 1640-1650.
Kinjo, Y., Beral, V., Akiba, S., Key, T., Mizuno, S., Appleby, P., Yamaguchi, N., Watanabe, S. & Doll, R. 1999. Possible Excluded, as the review has already been assessed in
protective effect of milk, meat and fish for cerebrovascular disease mortality in Japan. J Epidemiol, 9(4): 268-74 VKM 2022.
Kondo, K., Miura, K., Tanaka-Mizuno, S., Kadota, A., Arima, H., Okuda, N., Fujiyoshi, A. et al. 2019. Cardiovascular Risk Excluded, as the review has already been assessed in
Assessment Chart by Dietary Factors in Japan – NIPPON DATA80. Circ J, 83(6): 1254-1260. VKM 2022.
Kromhout, D., Bosschieter, E.B. & de Lezenne Coulander, C. 1985. The inverse relation between fish consumption and Excluded, as the review has already been assessed in
20-year mortality from coronary heart disease. N Engl J Med, 312(19): 1205-9. VKM 2022.
Letois, F., Mura, T., Scali, J., Gutierrez, L.A., Feart, C. & Berr, C. 2016. Nutrition and mortality in the elderly over 10 years of Excluded for further assessment, as the primary study
follow-up: the Three-City study. British Journal of Nutrition, 116(5): 882-889 had already been assessed in a systematic review that is
included in VKM 2022 (Schwingshackl et al., 2017)
Manger, M.S., Strand, E., Ebbing, M., Seifert, R., Refsum, H., Nordrehaug, J.E., Nilsen, D.W. et al. 2010. Dietary intake of Excluded, as the review has already been assessed in
n-3 long-chain polyunsaturated fatty acids and coronary events in Norwegian patients with coronary artery disease. Am J VKM 2022.
Clin Nutr, 92(1): 244-51.
Mohan, D., Mente, A., Dehghan, M., Rangarajan, S., O'Donnell, M., Hu, W.H., Dagenais, G. et al. 2021. Associations of Fish Excluded, as the review has already been assessed in
Consumption With Risk of Cardiovascular Disease and Mortality Among Individuals With or Without Vascular Disease VKM 2022.
From 58 Countries. Jama Internal Medicine, 181(5): 631-649.
Mozaffarian, D., Lemaitre, R.N., Kuller, L.H., Burke, G.L., Tracy, R.P. & Siscovick, D.S. 2003. Cardiac benefits of fish Excluded, as the review has already been assessed in
consumption may depend on the type of fish meal consumed: the Cardiovascular Health Study. Circulation, 107(10): VKM 2022.
1372-7.
Nagata, C., Takatsuka, N. & Shimizu, H. 2002. Soy and fish oil intake and mortality in a Japanese community. American Excluded for further assessment, as the primary study
Journal of Epidemiology, 156(9): 824-831. had already been assessed in a systematic review that is
included in VKM 2022 (Schwingshackl et al., 2017)
Nahab, F., Pearson, K., Frankel, M.R., Ard, J., Safford, M.M., Kleindorfer, D., Howard, V.J. & Judd, S. 2016. Dietary fried fish Excluded, as the review has already been assessed in
intake increases risk of CVD: the Reasons for Geographic And Racial Differences in Stroke (REGARDS) study. Public Health VKM 2022.
Nutrition, 19(18): 3327-3336.
Nakamura, Y., Hozawa, A., Turin, T.C., Takashima, N., Okamura, T., Hayakawa, T., Kita, Y. et al. 2009. Dietary Habits in Excluded for further assessment, as the primary study
Middle Age and Future Changes in Activities of Daily Living – NIPPON DATA80. Gerontology, 55(6): 707-713 had already been assessed in a systematic review that
is included in VKM 2022 (Nakamura et al., 2005).
Ness, A.R., Maynard, M., Frankel, S., Smith, G.D., Frobisher, C., Leary, S.D., Emmett, P.M. & Gunnell, D. 2005. Diet in Excluded for further assessment, as the primary study
childhood and adult cardiovascular and all-cause mortality: the Boyd Orr cohort. Heart, 91(7): 894-898.. had already been assessed in a systematic review that
is included in VKM 2022 (Wan et al., 2017 and Zhao et
al., 2016)
Oomen et al. 2000. Fish consumption and coronary heart disease mortality in Finland, Italy, and The Netherlands. Excluded, as the review has already been assessed in
VKM 2022.
Orencia, A.J. Daviglus, M.L. Dyer, A.R. Shekelle, R.B. & Stamler, J. 1996. Fish consumption and stroke in men - 30-year Excluded, as the review has already been assessed in
findings of the Chicago Western Electric Study. Stroke, 27:204-209. VKM 2022.

367
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.17 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “MORTALITY” (cont.)
Osler, M., Andreasen, A.H. & Hoidrup, S. 2003. No inverse association between fish consumption and risk of death from Excluded, as the review has already been assessed in
all-causes, and incidence of coronary heart disease in middle-aged, Danish adults. Journal of Clinical Epidemiology, VKM 2022.
56(3): 274-279.
Otsuka, R., Tange, C., Nishita, Y., Tomida, M., Kato, Y., Imai, T., Ando, F. & Shimokata, H. 2019. Fish and Meat Intake, Excluded, as the review has already been assessed in
Serum Eicosapentaenoic Acid and Docosahexaenoic Acid Levels, and Mortality in Community-Dwelling Japanese Older VKM 2022.
Persons. International Journal of Environmental Research and Public Health, 16(10).
Owen, A.J., Magliano, D.J., O'Dea, K., Barr, E.L.M. & Shaw, J.E. 2016. Polyunsaturated fatty acid intake and risk of Excluded, as the review has already been assessed in
cardiovascular mortality in a low fish-consuming population: a prospective cohort analysis. European Journal of Nutrition, VKM 2022.
55(4): 1605-1613.
Sauvaget, C., Nagano, J., Allen, N., Grant, E.J. & Beral, V. 2003. Intake of animal products and stroke mortality in the Excluded, as the review has already been assessed in
Hiroshima/Nagasaki Life Span Study. International Journal of Epidemiology, 32(4): 536-543. VKM 2022.
Shao, M.Y., Jiang, C.Q., Zhang, W.S., Zhu, F., Jin, Y.L., Woo, J., Cheng, K.K., Lam, T.H. & Xu, L. 2021. Association of fish Excluded, as the review has already been assessed in
consumption with risk of all-cause and cardiovascular disease mortality: an 11-year follow-up of the Guangzhou Biobank VKM 2022.
Cohort Study. European Journal of Clinical Nutrition
Takata, Y., Zhang, X., Li, H., Gao, Y.T., Yang, G., Gao, J., Cai, H., Xiang, Y.B., Zheng, W. & Shu, X.O. 2013. Fish intake Excluded, as the review has already been assessed in
and risks of total and cause-specific mortality in 2 population-based cohort studies of 134,296 men and women. Am J VKM 2022.
Epidemiol, 178(1): 46-57.
Tomasallo, C., Anderson, H., Haughwout, M., Imm, P. & Knobeloch, L. 2010. Mortality among frequent consumers of Great Excluded for further assessment, as the primary study
Lakes sport fish. Environmental Research, 110(1): 62-69. had already been assessed in a systematic review that
is included in VKM 2022 (Wan et al., 2017)
van den Brandt, P.A. 2019. Red meat, processed meat, and other dietary protein sources and risk of overall and cause- Excluded, as the review has already been assessed in
specific mortality in The Netherlands Cohort Study. European Journal of Epidemiology, 34(4): 351-369 VKM 2022.
Villegas, R., Takata, Y., Murff, H. & Blot, W.J. 2015. Fish, omega-3 long-chain fatty acids, and all-cause mortality in a Excluded, as the review has already been assessed in
low-income US population: Results from the Southern Community Cohort Study. Nutrition Metabolism and Cardiovascular VKM 2022.
Diseases, 25(7): 651-658.
Virtanen, H.E.K., Voutilainen, S., Koskinen, T.T., Mursu, J., Kokko, P., Ylilauri, M.P.T., Tuomainen, T.P., Salonen, J.T. & Excluded, as the review has already been assessed in
Virtanen, J.K. 2019. Dietary proteins and protein sources and risk of death: the Kuopio Ischaemic Heart Disease Risk VKM 2022.
Factor Study. Am J Clin Nutr, 109(5): 1462-1471
Wallin, A., Orsini, N., Forouhi, N.G. & Wolk, A. 2018. Fish consumption in relation to myocardial infarction, stroke and Excluded, as the review has already been assessed in
mortality among women and men with type 2 diabetes: A prospective cohort study. Clinical Nutrition, 37(2): 590-596 VKM 2022.
Yamagishi, K., Iso, H., Date, C., Fukui, M., Wakai, K., Kikuchi, S., Inaba, Y., Tanabe, N., Tamakoshi, A. & Grp, J.S. 2008. Excluded, as the review has already been assessed in
Fish, omega-3 polyunsaturated fatty acids, and mortality from cardiovascular diseases in a nationwide community-based VKM 2022.
cohort of Japanese men and women – The JACC (Japan Collaborative Cohort Study for Evaluation of Cancer Risk) study.
Journal of the American College of Cardiology, 52(12): 988-996.
Yuan, J.M., Ross, R.K., Gao, Y.T. & Yu, M.C. 2001. Fish and shellfish consumption in relation to death from myocardial Excluded, as the review has already been assessed in
infarction among men in Shanghai, China. Am J Epidemiol, 154(9): 809-16 VKM 2022.
Zhang, Y., Zhuang, P., He, W., Chen, J.N., Wang, W.Q., Freedman, N.D., Abnet, C.C., Wang, J.B. & Jiao, J.J. 2018. Association Excluded for further assessment, as the primary study
of fish and long-chain omega-3 fatty acids intakes with total and cause-specific mortality: prospective analysis of 421 had already been assessed in a systematic review that
309 individuals. J Intern Med, 284(4): 399-417. is included in VKM 2022 (Jayedi et al., 2020)
Zhong, V.T.W., Van Horn, L., Greenland, P., Carnethon, M.R., Ning, H.Y., Wilkins, J.T., Lloyd-Jones, D.M. & Allen, N.B. 2020. Excluded, as the review has already been assessed in
Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Cardiovascular Disease and VKM 2022.
All-Cause Mortality. Jama Internal Medicine, 180(4): 503-512
Zhuang, P., Wang, W.Q., Wang, J., Zhang, Y. & Jiao, J.J. 2018. Current Level of Fish Consumption is Associated with Excluded, as the review has already been assessed in
Mortality in Chinese but not US Adults: New Findings From Two Nationwide Cohort Studies With 14 and 9.8 Years of VKM 2022.
Follow-Up. Molecular Nutrition & Food Research, 62(8)

368
APPENDICES

OVERWEIGHT AND OBESITY

TABLE A3.18 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “OBESITY AND OVERWEIGHT”
DURING FULL-TEXT SCREENING BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 2) Reason for exclusion
Liberali, R. Kupek, E. & Assis, M.A.A.D. 2020. Dietary patterns and childhood obesity risk: a systematic review. Childhood Excluded based on inclusion and exclusion criteria:
Obesity, 16(2):70-85. excluded since study focuses on childhood obesity and
not general population. In addition, effects of seafood
were not emphasized or specifically mentioned.
Schlesinger, S. Neuenschwander, M. Schwedhelm, C. Hoffmann, G. Bechthold, A. Boeing, H. & Schwingshackl, L. 2019. Excluded, as the review has already been assessed in
Food groups and risk of overweight, obesity, and weight gain: a systematic review and dose-response meta-analysis of VKM 2022.
prospective studies. Advances in Nutrition, 10(2):205-218.
Bender, N. Portmann, M. Heg, Z. Hofmann, K. Zwahlen, M. & Egger, M. 2014. Fish or n3‐PUFA intake and body Excluded, as the review has already been assessed in
composition: a systematic review and meta‐analysis. Obesity reviews, 15(8):657-665. VKM 2022.

TABLE A3.19 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “OBESITY AND OVERWEIGHT”
Study (n = 28) Reason for exclusion
Abete, I., Parra, D., Crujeiras, A.B., Goyenechea, E. & Martinez, J.A. 2008. Specific insulin sensitivity and leptin responses Excluded based on inclusion and exclusion criteria
to a nutritional treatment of obesity via a combination of energy restriction and fatty fish intake. J Hum Nutr Diet, 21(6):
591-600
Alkerwi, A., Sauvageot, N., Buckley, J.D., Donneau, A.F., Albert, A., Guillaume, M. & Crichton, G.E. 2015. The potential Excluded based on inclusion and exclusion criteria:
impact of animal protein intake on global and abdominal obesity: evidence from the Observation of Cardiovascular Risk cross-sectional study
Factors in Luxembourg (ORISCAV-LUX) study. Public Health Nutrition, 18(10): 1831-1838.
Celis-Morales, C., Livingstone, K.M., Affleck, A., Navas-Carretero, S., San-Cristobal, R., Martinez, J.A., Marsaux, C.F.M. et Excluded based on inclusion and exclusion criteria:
al. 2018. Correlates of overall and central obesity in adults from seven European countries: findings from the Food4Me cross-sectional data
Study. Eur J Clin Nutr, 72(2): 207-219
Fisk, H.L., Irvine, M., Miles, E.A., Lietz, G., Mathers, J.C., Packard, C.J., Armah, C.K. et al. 2018. Association of oily fish Excluded based on inclusion and exclusion criteria: APOE
intake, sex, age, BMI and APOE genotype with plasma long-chain n-3 fatty acid composition. Br J Nutr, 120(1): 23-32. genotype
Gunnarsdottir, I., Tomasson, H., Kiely, M., Martinez, J.A., Bandarra, N.M., Morais, M.G. & Thorsdottir, I. 2008. Inclusion Excluded based on inclusion and exclusion criteria: only
of fish or fish oil in weight-loss diets for young adults: effects on blood lipids. International Journal of Obesity, 32(7): biomarker
1105-1112
Ilesanmi-Oyelere, B.L., Coad, J., Roy, N.C. & Kruger, M.C. 2020. Dietary Patterns, Body Composition, and Bone Health in Excluded based on inclusion and exclusion criteria:
New Zealand Postmenopausal Women. Frontiers in Nutrition, 7 dietary patterns
Inelmen, E.M., Toffanello, E.D., Enzi, G., Sergi, G., Coin, A., Busetto, L. & Manzato, E. 2008. Differences in dietary patterns Excluded based on inclusion and exclusion criteria:
between older and younger obese and overweight outpatients. J Nutr Health Aging, 12(1): 3-8 focuses only on frequency of food pattern
Mori, T.A., Bao, D.Q., Burke, V., Puddey, I.B., Watts, G.F. & Beilin, L.J. 1999. Dietary fish as a major component of a weight- Excluded based on inclusion and exclusion criteria: only
loss diet: effect on serum lipids, glucose, and insulin metabolism in overweight hypertensive subjects. American Journal biomarker
of Clinical Nutrition, 70(5): 817-825
Mori, T.A., Burke, V., Puddey, I.B., Shaw, J.E. & Beilin, L.J. 2004. Effect of fish diets and weight loss on serum leptin Excluded based on inclusion and exclusion criteria: only
concentration in overweight, treated-hypertensive subjects. J Hypertens, 22(10): 1983-90. biomarker
Panagiotakos, D.B., Zeimbekis, A., Boutziouka, V., Economou, M., Kourlaba, G., Toutouzas, P. & Polychronopoulos, E. 2007. Excluded based on inclusion and exclusion criteria:
Long-term fish intake is associated with better lipid profile, arterial blood pressure, and blood glucose levels in elderly only includes the effect of fish consumption on several
people from Mediterranean islands (MEDIS epidemiological study). Medical Science Monitor, 13(7):CR307-CR312. biomarkers of cardiovascular-disease risk
Parra, D., Bandarra, N.M., Kiely, M., Thorsdottir, I. & Martinez, J.A. 2007. Impact of fish intake on oxidative stress when Excluded based on inclusion and exclusion criteria
included into a moderate energy-restricted program to treat obesity. European Journal of Nutrition, 46(8): 460-467
Ramel, A., Martinez, J.A., Kiely, M., Bandarra, N.M. & Thorsdottir, I. 2010. Effects of weight loss and seafood consumption Excluded based on inclusion and exclusion criteria
on inflammation parameters in young, overweight and obese European men and women during 8 weeks of energy
restriction. European Journal of Clinical Nutrition, 64(9): 987-993
Ramel, A., Martinez, J.A., Kiely, M., Bandarra, N.M. & Thorsdottir, I. 2010. Moderate consumption of fatty fish reduces Excluded based on inclusion and exclusion criteria
diastolic blood pressure in overweight and obese European young adults during energy restriction. Nutrition, 26(2):
168-174.
Ramel, A. Jonsdottir, M.T. & Thorsdottir, I. 2009. Consumption of cod and weight loss in young overweight and obese Excluded based on inclusion and exclusion criteria
adults on an energy reduced diet for 8-weeks. Nutr Metab Cardiovasc Dis, 19(10):690-6.
Satija, A., Hu, F.B., Bowen, L., Bharathi, A.V., Vaz, M., Prabhakaran, D., Reddy, K.S. et al. 2015. Dietary patterns in India Excluded based on inclusion and exclusion criteria:
and their association with obesity and central obesity. Public Health Nutrition, 18(16): 3031-3041.. dietary patterns linked to obesity
Schulze, M.B., Fung, T.T., Manson, J.E., Willett, W.C. & Hu, F.B. 2006. Dietary patterns and changes in body weight in Excluded based on inclusion and exclusion criteria:
women. Obesity (Silver Spring), 14(8): 1444-53. dietary patterns and changes in body weight
Spencer, E.A., Appleby, P.N., Davey, G.K. & Key, T.J. 2003. Diet and body mass index in 38,000 EPIC-Oxford meat-eaters, Excluded based on inclusion and exclusion criteria:
fish-eaters, vegetarians and vegans. International Journal of Obesity, 27(6): 728-734 cross-sectional analysis

369
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.19 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “OBESITY AND OVERWEIGHT” (cont.)
St-Jules, D.E. Watters, C.A. & Novotny, R. 2014. Estimation of fish intake in Asian and white female adolescents, and Excluded based on inclusion and exclusion criteria:
association with 2-year changes in body fatness and body fat distribution: the female adolescent maturation study. cross-sectional analysis
J Acad Nutr Diet, 114(4):543-51.
Vázquez, C., Botella-Carretero, J.I., Corella, D., Fiol, M., Lage, M., Lurbe, E., Richart, C. et al. 2014. White fish reduces Excluded based on inclusion and exclusion criteria: only
cardiovascular risk factors in patients with metabolic syndrome: the WISH-CARE study, a multicenter randomized clinical biomarker
trial. Nutr Metab Cardiovasc Dis, 24(3): 328-35.
Xu, X.Y., Shi, Z.M., Liu, G., Chang, D.N., Inglis, S.C., Hall, J.J., Schutte, A.E., Byles, J.E. & Parker, D. 2021. The Joint Excluded based on inclusion and exclusion criteria:
Effects of Diet and Dietary Supplements in Relation to Obesity and Cardiovascular Disease over a 10-Year Follow-Up: A fish oil
Longitudinal Study of 69,990 Participants in Australia. Nutrients, 13(3)
Abete, I., Parra, D. & Martinez, J.A. 2009. Legume-, fish-, or high-protein-based hypocaloric diets: effects on weight loss Excluded for further assessment, as the primary study
and mitochondrial oxidation in obese men. J Med Food, 12(1): 100-8 had already been assessed in one of the included
systematic reviews
Huang, T., Wang, T.G., Heianza, Y., Wiggs, J., Sun, D.J.Y., Choi, H.K., Chai, J.F. et al. 2019. Fish and marine fatty acids Excluded for further assessment, as the primary study
intakes, the FADS genotypes and long-term weight gain: a prospective cohort study. BMJ Open, 9(7). had already been assessed in VKM 2022
Huang, T., Wang, T.G., Heianza, Y., Zheng, Y., Sun, D.J.Y., Kang, J.H., Pasquale, L.R. et al. 2019. Habitual consumption of Excluded for further assessment, as the primary study
long-chain n-3 PUFAs and fish attenuates genetically associated long-term weight gain. American Journal of Clinical had already been assessed in VKM 2022
Nutrition, 109(3): 665-673
Jakobsen, M.U., Due, K.M., Dethlefsen, C., Halkjaer, J., Holst, C., Forouhi, N.G., Tjønneland, A. et al. 2012. Fish Excluded for further assessment, as the primary study
consumption does not prevent increase in waist circumference in European women and men. Br J Nutr, 108(5): 924-31.. had already been assessed in VKM 2022
Jakobsen, M.U., Dethlefsen, C., Due, K.M., May, A.M., Romaguera, D., Vergnaud, A.C., Norat, T. et al. 2013. Fish Excluded for further assessment, as the primary study
consumption and subsequent change in body weight in European women and men. British Journal of Nutrition, 109(2): had already been assessed in a systematic review that
353-362.. is included in VKM 2022
Ramel, A., Parra, D., Martinéz, J.A., Kiely, M. & Thorsdottir, I. 2009. Effects of seafood consumption and weight loss on Excluded for further assessment, as the primary study
fasting leptin and ghrelin concentrations in overweight and obese European young adults. Eur J Nutr, 48(2): 107-14 had already been assessed in VKM 2022
Tapsell, L.C., Batterham, M.J., Charlton, K.E., Neale, E.P., Probst, Y.C., O'Shea, J.E., Thorne, R.L., Zhang, Q.S. & Louie, J.C.Y. Excluded for further assessment, as the primary study
2013. Foods, nutrients or whole diets: effects of targeting fish and LCn3PUFA consumption in a 12mo weight loss trial. had already been assessed in VKM 2022
Bmc Public Health, 13
Thorsdottir, I., Tomasson, H., Gunnarsdottir, I., Gisladottir, E., Kiely, M., Parra, M.D., Bandarra, N.M., Schaafsma, G. Excluded for further assessment, as the primary study
& Martinez, J.A. 2007. Randomized trial of weight-loss-diets for young adults varying in fish and fish oil content. had already been assessed in one of the included
International Journal of Obesity, 31(10): 1560-1566 systematic reviews

370
APPENDICES

QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS AND PRIMARY STUDIES

ALLERGY AND IMMUNOLOGY
TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8)
Di Giuseppe, D., Crippa, A. & Orsini, N. 2014. Fish consumption and risk of rheumatoid arthritis: a dose-response meta-analysis. Arthritis Research & Therapy, 16(5):446
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no yes 0.00
Q2 Protocol yes/partial yes/no yes 0.50
Q3 Explanation of included study design yes/no yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no yes 1.00
Q5 Paired study selection yes/no yes 0.00
Q6 Paired data extraction yes/no yes 1.00
Q7 List of excluded studies yes/partial yes/no no 0.50
Q8 Description of included studies yes/partial yes/no partial yes 1.00
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI partial yes 1.00
Q10 Sources of funding for included studies yes/no yes 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q13 Impact of risk of bias in individual studies when yes/no yes 0.00
interpreting results
Q14 Heterogeneity assessed yes/no yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted yes 1.00
Q16 Conflict of interest included yes/no yes 1.00
Total score 11.00
Percent 68.75
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

371
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Ierodiakonou, D., Garcia-Larsen, V., Logan, A., Groome, A., Cunha, S., Chivinge, J., Boyle, R. J. et al. 2016. Timing of allergenic food introduction to the infant diet and risk of
allergic or autoimmune disease: a systematic review and meta-analysis. Jama, 316(11):1181-1192.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, they included PICO 1.00
Q2 Protocol yes/partial yes/no Yes, their protocol was registered in PROSPERO 1.00
Q3 Explanation of included study design yes/no Yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no Yes they searched the Cochrane Library, EMBASE, 1.00
LILACS, MEDLINE, Web of Science, and http://
apps.who.int/trialsearch from 1 January 1946 to
8 March 2016
Q5 Paired study selection yes/no No, it is not mentioned 0.00
Q6 Paired data extraction yes/no Yes, data were extracted in duplicate 1.00
Q7 List of excluded studies yes/partial yes/no No list was provided 0.00
Q8 Description of included studies yes/partial yes/no Yes, they explained the included articles in detail 1.00
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI Yes, they used Cochrane Risk of Bias tool and 1.00
the National Institute for Clinical Excellence
methodological checklists for intervention and
observational studies
Q10 Sources of funding for included studies yes/no No, they did not mention it 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes, random effects meta-analyses used generic 1.00
inverse variance and Mantel-Haenszel methods
for observational and intervention studies,
respectively
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Yes, the considered RoB in individual studies 1.00
Q13 Impact of risk of bias in individual studies when yes/no Yes, the considered RoB in single studies 1.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes, they explained the reasons for heterogeneity 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted Yes, publication bias was assessed using funnel 1.00
plots and the Egger test
Q16 Conflict of interest included yes/no Yes, all authors have completed and submitted 1.00
the ICMJE Form for Disclosure of Potential
Conflicts of Interest
Total score 13
Percent 81%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

372
APPENDICES

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Li, N., Wu, X., Zhuang, W., Xia, L., Chen, Y., Wu, C., Zhou, Y. et al. 2020. Fish consumption and multiple health outcomes: Umbrella review. Trends in Food Science & Technology,
99:273-283.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, they included PICO in their study. 1.00
Q2 Protocol yes/partial yes/no Partially yes, protocol was not established before 0.50
study, but they had a methodology.
Q3 Explanation of included study design yes/no Yes, they performed eligibility criteria (2.3) 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no Yes, they searched Embase, Medline, the 1.00
Cochrane Database of Systematic Reviews and
Web of Science.
Q5 Paired study selection yes/no No, not specified 0.00
Q6 Paired data extraction yes/no Yes, two independent investigators (NL and 1.00
XW) extracted the following data separately for
eligible articles
Q7 List of excluded studies yes/partial yes/no They did not provide the list of excluded articles. 0.00
Q8 Description of included studies yes/partial yes/no Yes, included studies with relevant reasoning are 1.00
presented in tables.
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI Yes, they used AMSTAR. 1.00
Q10 Sources of funding for included studies yes/no No, they did not. 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted N/A, no meta-analysis conducted 0.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted N/A, no meta-analysis conducted 0.00
Q13 Impact of risk of bias in individual studies when yes/no They did not discuss RoB. 0.00
interpreting results
Q14 Heterogeneity assessed yes/no No, they did not provide. explanation for 0.00
heterogeneity.
Q15 Publication bias assessed yes/no/no meta-analysis conducted N/A, no meta-analysis conducted 0.00
Q16 Conflict of interest included yes/no Yes, the authors declare no competing interests. 1.00
Total score 7.50
Percent 47%
Percent (exclude n/a question) 58%
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

373
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Venter, C., Agostoni, C., Arshad, S.H., Ben‐Abdallah, M., Du Toit, G., Fleischer, D.M., O’Mahony, L. et al. 2020. Dietary factors during pregnancy and atopic outcomes in childhood:
A systematic review from the European Academy of Allergy and Clinical Immunology. Pediatric Allergy and Immunology, 31(8):889-912.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, authors used PICO approach in their search 1.00
and inclusion criteria.
Q2 Protocol yes/partial yes/no Partially yes, protocol was not established before 0.50
study, but they had a methodology.
Q3 Explanation of included study design yes/no Yes, they used a flow chart. 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no Yes, they searched three bibliographic databases 1.00
(MEDLINE, EMBASE, and Web of Science).
Q5 Paired study selection yes/no Yes, for studies considered potentially relevant, 1.00
full-text copies were obtained, and inclusion
of the studies was discussed by CV, MBA, MP,
and AM.
Q6 Paired data extraction yes/no Yes, data were extracted by three authors (AM, 1.00
MBA, and MP).
Q7 List of excluded studies yes/partial yes/no Yes, reasons are mentioned. 1.00
Q8 Description of included studies yes/partial yes/no Yes, they described populations, interventions, 1.00
comparators, outcomes and research designs.
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI Yes, they used Cochrane Collaboration Risk of 1.00
Bias tool for intervention trials and the National
Institute for Clinical Excellence methodological
checklist for cohort and case-control studies.
Q10 Sources of funding for included studies yes/no Yes, conflicts of interest were noted if industry 1.00
was involved in any aspect of the study or if
authors received funding.
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes, relevant statistics were used. 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Yes, risk of bias for cohort and case-control 1.00
studies included assessment of: (a) selection
bias, which was considered low if cases and
controls were recruited from similar populations
and had a similar attrition rate <20%; (b)
assessment bias, which included blinding
of outcome assessors and use of validated
assessment tools; and (c) confounding bias (did
study design and analysis account for relevant
confounders?
Q13 Impact of risk of bias in individual studies when yes/no No. 0.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes, the heterogeneity of the studies was 1.00
calculated using the Cochran χ2 and I2
statistics.
Q15 Publication bias assessed yes/no/no meta-analysis conducted Yes, in order to minimize publication bias, 1.00
they performed a comprehensive search of the
literature and included experts in the field to
ensure that no relevant study was missed. They
performed the Egger's test.
Q16 Conflict of interest included yes/no Yes, they declared no conflict of interest. 1.00
Total score 14.50
Percent 90.63
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

374
APPENDICES

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Netting, M. J., Middleton, P. F. & Makrides, M. 2014. Does maternal diet during pregnancy and lactation affect outcomes in offspring? A systematic review of food-based
approaches. Nutrition, 30(11-12):1225-1241.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, they included ICO but they did not explain the 1.00
population considered in this study.
Q2 Protocol yes/partial yes/no Partially yes. Protocol was not established before 0.50
study, but they had a methodology.
Q3 Explanation of included study design yes/no Yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no Partially yes, searched MEDLINE, EMBASE, and 0.50
the Cochrane Library (last searched end of August
2011), more than 24 months until publication
finished at 2014.
Q5 Paired study selection yes/no Yes, two authors independently assessed search 1.00
results against study eligibility criteria.
Q6 Paired data extraction yes/no Yes, two authors also independently conducted 1.00
data extraction for each included study.
Q7 List of excluded studies yes/partial yes/no Yes, one intervention trial and seven other studies 1.00
were excluded [18–25].
Q8 Description of included studies yes/partial yes/no Yes, this systematic review included 42 studies 1.00
(Table 1).
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI Yes, they assessed the risk for bias using the 1.00
methods outlined in the Cochrane Handbook for
Reviews of Interventions.
Q10 Sources of funding for included studies yes/no No, they did not. 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes, where possible, the results of randomized 1.00
controlled trials (RCTs) were pooled, using the
meta-analysis program RevMan.
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Yes, the risk for bias assessments are described 1.00
in Table 1 (association studies [26–76]) and in
Table 2 (intervention studies. [26,29,32,34,36,39,
41,42,44,45,78]).
Q13 Impact of risk of bias in individual studies when yes/no Yes, they did consider risk of bias in 1.00
interpreting results interpretation.
Q14 Heterogeneity assessed yes/no Yes, they performed heterogeneity test but with no 1.00
explanation on reasons.
Q15 Publication bias assessed yes/no/no meta-analysis conducted No, they did not mention that. 0.00
Q16 Conflict of interest included yes/no Yes, they explained the funding source in the 1.00
acknowledgments.
Total score 13.00
Percent 81.25
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

375
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Kremmyda, L.S., Vlachava, M., Noakes, P.S., Diaper, N.D., Miles, E.A. & Calder, P.C. 2011. Atopy risk in infants and children in relation to early exposure to fish, oily fish, or long-
chain omega-3 fatty acids: a systematic review. Clinical reviews in allergy & immunology, 41(1):36-66.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, they investigated PICO. 1.00
Q2 Protocol yes/partial yes/no No methodology 0.00
Q3 Explanation of included study design yes/no No explanation 0.00
Q4 Comprehensive literature search strategy yes/partial yes/no Yes, the studies were identified through Ovid 1.00
Medline (1950– 2009) and PubMed (1950–2009)
databases.
Q5 Paired study selection yes/no No, not mentioned. 0.00
Q6 Paired data extraction yes/no No, not mentioned. 0.00
Q7 List of excluded studies yes/partial yes/no No, not mentioned. 0.00
Q8 Description of included studies yes/partial yes/no Partial yes. They described them in tables but no 0.50
follow up.
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI No RoB was performed. 0.00
Q10 Sources of funding for included studies yes/no No, they did not mention. 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted N/A, no meta analyses. 0.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted N/A, no meta analyses. 0.00
Q13 Impact of risk of bias in individual studies when yes/no No RoB was performed. 0.00
interpreting results
Q14 Heterogeneity assessed yes/no N/A, no meta analyses. 0.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted No, not mentioned. 0.00
Q16 Conflict of interest included yes/no No declaration of conflict of interest. 0.00
Total score 2.50
Percent 15.63
Percent (exclude N/A question) 19.23
Overall AMSTAR 2 judgement (confidence in the results) Critically low
Include/exclude Excluded
(narrative)

376
APPENDICES

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Pattison, D.J., Harrison, R.A. & Symmons, D.P. 2004. The role of diet in susceptibility to rheumatoid arthritis: a systematic review. The Journal of rheumatology, 31(7):1310-1319.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, they included PICO in their criteria. 1.00
Q2 Protocol yes/partial yes/no Yes, the criteria were based on methodological 1.00
issues relevant to nutritional epidemiology, expert
knowledge, and published guidelines.
Q3 Explanation of included study design yes/no Yes, they used following criteria (1) an a priori 1.00
hypothesis was given; (2) cases were ascertained
using the ARA 19588 or 19879 criteria for
diagnosis of RA; (3) cases and controls were
comparable at baseline; (4) controls were
randomly selected from the source of the
population of the cases; (5) dietary assessment
was undertaken prior to onset of symptoms, using
a “validated” method of assessment and the
same method was used for cases and controls;
and (6) potential confounding factors were
accounted for.
Q4 Comprehensive literature search strategy yes/partial yes/no Yes, they searched the Cochrane Database of 1.00
Systematic Reviews, Medline OVID citations (1966
to 2003), Embase (1980 to 2003), and the ISI Web
of Science (1981 to 2003).
Q5 Paired study selection yes/no Yes, two authors selected the articles. 1.00
Q6 Paired data extraction yes/no Yes, two authors extracted the data. 1.00
Q7 List of excluded studies yes/partial yes/no Yes, they listed excluded articles with 1.00
justification.
Q8 Description of included studies yes/partial yes/no Yes, they described the included articles and 1.00
explained them in details (see Table 1 and Table
2).
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Yes, explained in quality assessment and data 1.00
extraction.
Q10 Sources of funding for included studies yes/no No, funding source was not investigated. 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted N/A – no meta-analysis conducted. N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted N/A – no meta-analysis conducted. N/A
Q13 Impact of risk of bias in individual studies when yes/no They did not perform RoB. 0.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes, they explained the heterogeneity. 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted N/A – no meta-analysis conducted N/A
Q16 Conflict of interest included yes/no They did not declare conflict of interest nor the 0.00
funding source.
Total score 10.00
Percent 62.50
Percent (exclude n/a question) 76.9
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

377
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.20 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “ALLERGY AND IMMUNOLOGY”
(N = 8) (cont.)
Solman, L., Lloyd‐Lavery, A., Grindlay, D.J.C., Rogers, N.K., Thomas, K.S. & Harman, K.E. 2019. What's new in atopic eczema? An analysis of systematic reviews published in 2016.
Part 1: treatment and prevention. Clinical and experimental dermatology, 44(4):363-369.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no No clear inclusion criteria 0.00
Q2 Protocol yes/partial yes/no No protocol was used for review. 0.00
Q3 Explanation of included study design yes/no No clear inclusion criteria 0.00
Q4 Comprehensive literature search strategy yes/partial yes/no Not mentioned 0.00
Q5 Paired study selection yes/no Not mentioned 0.00
Q6 Paired data extraction yes/no Not mentioned 0.00
Q7 List of excluded studies yes/partial yes/no Not mentioned 0.00
Q8 Description of included studies yes/partial yes/no Yes, they explain the included studies. 1.00
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI No RoB was performed. 0.00
Q10 Sources of funding for included studies yes/no Yes, they investigated source of funding. 1.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted N/A – no meta-analysis conducted 0.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted N/A – no meta-analysis conducted 0.00
Q13 Impact of risk of bias in individual studies when yes/no No RoB was performed. 0.00
interpreting results
Q14 Heterogeneity assessed yes/no N/A – no meta-analysis conducted 0.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted N/A – no meta-analysis conducted 0.00
Q16 Conflict of interest included yes/no Yes, they did declare conflict of interest and the 1.00
funding source.
Total score 3.00
Percent 18.75
Percent (exclude N/A question) 23.08
Overall AMSTAR 2 judgement (confidence in the results) Critically low
Include/exclude Exclude

378
APPENDICES

TABLE A3.21 QUALITY ASSESSMENT (RISK OF BIAS) PF PRIMARY STUDIES FOR THE THEME “ALLERGY AND IMMUNOLOGY” (N = 22)
Quality assessment
Reference primary study (n = 22) Score
(risk of bias judgement)
Acevedo, N., Frumento, P., Harb, H., Alhamwe, B.A., Johansson, C No details on fish intake. Maternal fish consumption was not clearly
C., Eick, L., Alm, J., Renz, H., Scheynius, A. & Potaczek, D.P. 2019. described in the analysis. Yes/no question if mother consumed fish
Histone Acetylation of Immune Regulatory Genes in Human during pregnancy.
Placenta in Association with Maternal Intake of Olive Oil and Fish
Consumption. International Journal of Molecular Sciences, 20(5).
Dunlop, A.L., Reichrtova, E., Palcovicova, L., Ciznar, P., Adamcakova- C No clear research question was formulated; dietary pattern study; no
Dodd, A., Smith, S.J. & McNabb, S.J. 2006. Environmental and details on fish intake
dietary risk factors for infantile atopic eczema among a Slovak birth
cohort. Pediatr Allergy Immunol, 17(2): 103-11.
Goksor, E., Alm, B., Pettersson, R., Mollborg, P., Erdes, L., Aberg, C Unclear dietary information
N. & Wennergren, G. 2013. Early fish introduction and neonatal
antibiotics affect the risk of asthma into school age. Pediatric
Allergy and Immunology, 24(4): 339-344.
Goksor, E., Alm, B., Thengilsdottir, H., Pettersson, R., Aberg, N. C Unclear dietary information, specific groups and use of antibiotics
& Wennergren, G. 2011. Preschool wheeze – impact of early fish
introduction and neonatal antibiotics. Acta Paediatrica, 100(12):
1561-1566.
González-Delgado, P., Caparrós, E., Moreno, M.V., Clemente, F., C In vitro study. In vitro measurement of both cytokine production
Flores, E., Velásquez, L., Rubio, G. & Fernández, J. 2016. Clinical in peripheral blood mononuclear cells (PBMCs) and expression of
and immunological characteristics of a pediatric population with HLADR in monocyte-derived dendritic cells stimulated with fish
food protein-induced enterocolitis syndrome (FPIES) to fish. Pediatr extracts.
Allergy Immunol, 27(3): 269-75.
Grieger, J.A., Pelecanos, A.M., Hurst, C., Tai, A. & Clifton, V.L. 2019. C Missing to control for confounding factors
Pre-Conception Maternal Food Intake and the Association with
Childhood Allergies. Nutrients, 11(8).
Hesselmar, B., Saalman, R., Rudin, A., Adlerberth, I. & Wold, A. C Fish intake not reported
2010. Early fish introduction is associated with less eczema, but not
sensitization, in infants. Acta Paediatr, 99(12): 1861-7.
Papamichael, M.M., Katsardis, C., Lambert, K., Tsoukalas, D., C The study group has asthma
Koutsilieris, M., Erbas, B. & Itsiopoulos, C. 2019. Efficacy of a
Mediterranean diet supplemented with fatty fish in ameliorating
inflammation in paediatric asthma: a randomised controlled trial.
Journal of Human Nutrition and Dietetics, 32(2): 185-197
Papamichael, M.M., Itsiopoulos, C., Lambert, K., Katsardis, C., C Investigation of the vitamin D status on lung function using
Tsoukalas, D. & Erbas, B. 2020. Sufficient vitamin D status participants in a RCT with fatty fish in asthma children.
positively modified ventilatory function in asthmatic children
following a Mediterranean diet enriched with fatty fish intervention
study. Nutr Res, 82: 99-109.
Roberts, G., Golder, N. & Lack, G. 2002. Bronchial challenges with C Patient group.
aerosolized food in asthmatic, food-allergic children. Allergy, 57(8):
713-7.
Storrø, O., Oien, T., Dotterud, C.K., Jenssen, J.A. & Johnsen, R. 2010. C Mix of study designs; prevention study; the study aimed to evaluate
A primary health-care intervention on pre- and postnatal risk factor the impact of a primary prevention intervention program on risk
behavior to prevent childhood allergy. The Prevention of Allergy behaviour for allergic diseases among children in a pre- and
among Children in Trondheim (PACT) study. BMC Public Health, postnatal primary healthcare setting.
10: 443.
Uddenfeldt, M., Janson, C., Lampa, E., Leander, M., Norback, D., C Information about the fish was unclear. The aim was to investigate
Larsson, L. & Rask-Andersen, A. 2010. High BMI is related to higher risk factors for asthma development in three different age groups
incidence of asthma, while a fish and fruit diet is related to a (n = 8 150).
lower-Results from a long-term follow-up study of three age groups
in Sweden. Respiratory Medicine, 104(7): 972-980.
Ukleja-Sokolowska, N., Zbikowska-Gotz, M., Lis, K., Adamczak, R. C Use of patient allergic to shrimps to assess allergic intermediate
& Bartuzi, Z. 2021. Assessment of TSLP, IL 25 and IL 33 in patients factors in blood
with shrimp allergy. Allergy Asthma and Clinical Immunology, 17(1).
Urwin, H.J., Miles, E.A., Noakes, P.S., Kremmyda, L.S., Vlachava, C Wrong outcome: faecal microbiota, IgA and calprotectin
M., Diaper, N.D., Godfrey, K.M., Calder, P.C., Vulevic, J. & Yaqoob, P.
2014. Effect of salmon consumption during pregnancy on maternal
and infant faecal microbiota, secretory IgA and calprotectin. Br J
Nutr, 111(5): 773-84.
Varraso, R., Fung, T.T., Barr, R.G., Hu, F.B., Willett, W. & Camargo, C Dietary patterns studied in patient with chronic obstructive
C.A. 2007. Prospective study of dietary patterns and chronic pulmonary disease
obstructive pulmonary disease among US women. American Journal
of Clinical Nutrition, 86(2): 488-495.

379
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.21 QUALITY ASSESSMENT (RISK OF BIAS) PF PRIMARY STUDIES FOR THE THEME “ALLERGY AND IMMUNOLOGY” (N = 22)
(cont.)
Quality assessment
Reference primary study (n = 22) Score
(risk of bias judgement)
Vasileiadou, S., Wennergren, G., Celind, F.S., Aberg, N., Pettersson, C Study analyses the prevalence of risk factors and protective factors
R., Alm, B. & Goksor, E. 2018. Eating fish and farm life reduce for allergic rhinitis. Only one question on fish intake.
allergic rhinitis at the age of twelve. Pediatric Allergy and
Immunology, 29(3): 283-289.
Virtanen, S.M., Kaila, M., Pekkanen, J., Kenward, M.G., Uusitalo, U., C Participants have increased risk of type 1 diabetes.
Pietinen, P., Kronberg-Kippilä, C. et al. 2010. Early introduction of
oats associated with decreased risk of persistent asthma and early
introduction of fish with decreased risk of allergic rhinitis. Br J Nutr,
103(2): 266-73.
Wijga, A.H., Smit, H.A., Kerkhof, M., de Jongste, J.C., Gerritsen, J., C Mothers were allergic. Children were examined at 2 years of food
Neijens, H.J., Boshuizen, H.C. & Brunekreef, B. 2003. Association of consumption and at 3 years for asthma symptoms. No result on fish
consumption of products containing milk fat with reduced asthma is reported.
risk in pre-school children: the PIAMA birth cohort study. Thorax,
58(7): 567-572.
Woodman, R.J., Baghdadi, L.R., Shanahan, E.M., de Silva, I., C Participants have rheumatoid arthritis. Data on nutrients intake and
Hodgson, J.M. & Mangoni, A.A. 2019. Diets high in n-3 fatty not fish. Aim to identify patterns of fatty acids intake in RA patients.
acids are associated with lower arterial stiffness in patients with
rheumatoid arthritis: a latent profile analysis. British Journal of
Nutrition, 121(2): 182-194
Zapatero Remón, L., Alonso Lebrero, E., Martín Fernández, E. & C Report on 14 children with enterocolitis syndrome due to fish protein
Martínez Molero, M.I. 2005. Food-protein-induced enterocolitis
syndrome caused by fish. Allergol Immunopathol (Madr), 33(6):
312-6.
Zeng, J., Wu, W., Tang, N., Chen, Y., Jing, J. & Cai, L. 2021. Maternal C Study aim was to investigate the association between maternal
Dietary Protein Patterns During Pregnancy and the Risk of Infant dietary protein patterns during pregnancy and the risk of infant
Eczema: A Cohort Study. Front Nutr, 8: 608972 eczema. Dietary pattern study. Fish and red meat results are given
together.
Zinn, C., Lopata, A., Visser, M. & Potter, P.C. 1997. The spectrum of C Study tested different fish species by ingestion, skin prick test and
allergy to South African bony fish (Teleosti). Evaluation by double- RAST test in the participants with suspected fish allergy.
blind, placebo-controlled challenge. S Afr Med J, 87(2): 146-52.

380
APPENDICES

BIRTH AND GROWTH OUTCOMES

TABLE A3.22 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “BIRTH AND GROWTH” (N = 3)
Quality assessment
Reference primary study (n = 3) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Oken, E. Kleinman, K. P. Olsen, S. F. Rich-Edwards, J. W. & Gillman, B
M. W. 2004. Associations of seafood and elongated n-3 fatty acid
intake with fetal growth and length of gestation: results from a US
pregnancy cohort. American journal of epidemiology, 160(8):774-
783.
Olsen, S. F. Grandjean, P. Weihe, P. & Viderø, T. 1993. Frequency C Whale consumption is aggregated with fish consumption in
of seafood intake in pregnancy as a determinant of birth weight: exposure, and it is not possible to study the effects of fish
evidence for a dose dependent relationship. Journal of Epidemiology consumption on health outcomes alone.
& Community Health, 47(6):436-440.
Zhao, R. Gao, Q. Xiong, T. Zhou, J. Wang, S. Zhang, Z. Hao, L. et al. B
2022. Moderate freshwater fish intake, but not n-3 polyunsaturated
fatty acids, is associated with a reduced risk of small for
gestational age in a prospective cohort of Chinese pregnant women.
Journal of the Academy of Nutrition and Dietetics, 122(4):722-730.

381
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

BONE HEALTH
TABLE A3.23 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “BONE HEALTH” (N = 11)
Quality assessment
Reference primary study (n = 11) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Hirota, T., Kusu, T. & Hirota, K. 2005. Improvement of nutrition B
stimulates bone mineral gain in Japanese school children and
adolescents. Osteoporos Int, 16(9): 1057-64.
Tong, T. Y., Appleby, P. N., Armstrong, M. E., Fensom, G. K., Knuppel, B
A., Papier, K., Key, T. J. et al. 2020. Vegetarian and vegan diets
and risks of total and site-specific fractures: results from the
prospective EPIC-Oxford study. BMC medicine, 18, 1-15.
Lucey, A. J., Paschos, G. K., Cashman, K. D., Martínéz, J. A., B
Thorsdottir, I., & Kiely, M. 2008. Influence of moderate energy
restriction and seafood consumption on bone turnover in overweight
young adults. The American Journal of Clinical Nutrition, 87(4),
1045-1052.
Thacher, T. D., Bommersbach, T. J., Pettifor, J. M., Isichei, C. O., & B
Fischer, P. R. 2015. Comparison of limestone and ground fish for
treatment of nutritional rickets in children in Nigeria. The Journal of
Pediatrics, 167(1), 148-154.
de Jonge, E. A., Rivadeneira, F., Erler, N. S., Hofman, A., Uitterlinden, C Use dietary pattern as exposure and not fish consumption explicitly
A. G., Franco, O. H., & Kiefte-de Jong, J. C. 2018. Dietary patterns in
an elderly population and their relation with bone mineral density:
the Rotterdam Study. European Journal of Nutrition, 57, 61-73.
De Jonge, E. A., Kiefte-de Jong, J. C., De Groot, L. C., Voortman, C Does not have fish intake as exposure alone, but within “a bone
T., Schoufour, J. D., Zillikens, M. C., ... & Rivadeneira, F. 2015. mineral density" diet consisting of several food groups.
Development of a food group-based diet score and its association
with bone mineral density in the elderly: the Rotterdam study.
Nutrients, 7(8), 6974-6990.
Ishikawa-Takata, K. & Ohta, T. 2003. Relationship of lifestyle factors C Methodological weakness in the description of study design and
to bone mass in Japanese women. J Nutr Health Aging, 7(1):44-53. population
Nieves, J. W., Barrett-Connor, E., Siris, E. S., Zion, M., Barlas, C Fish intake used to estimate vitamin D intake, along with intake of
S., & Chen, Y. T. 2008. Calcium and vitamin D intake influence milk, supplements and sun exposure. Intake of fish is not explicitly
bone mass, but not short-term fracture risk, in Caucasian stated further in the paper, or used as an individual exposure
postmenopausal women from the National Osteoporosis Risk variable.
Assessment (NORA) study. Osteoporosis International, 19, 673-679.
Samieri, C., Ginder Coupez, V., Lorrain, S., Letenneur, L., Allès, B., C Excluded – dietary patterns
Féart, C., Barberger-Gateau, P. et al. 2013. Nutrient patterns and
risk of fracture in older subjects: results from the Three-City Study.
Osteoporosis International, 24, 1295-1305.
Graff, I. E., Øyen, J., Kjellevold, M., Frøyland, L., Gjesdal, C. G., C Tailor-made salmon; enriched with vitamin D
Almås, B., Lie, Ø. et al. 2016. Reduced bone resorption by intake
of dietary vitamin D and K from tailor-made Atlantic salmon: A
randomized intervention trial. Oncotarget, 7(43), 69200.
Kerstetter, J. E., Mitnick, M. E., Gundberg, C. M., Caseria, D. M., C Exposure was different levels of protein; no mention of fish
Ellison, A. F., Carpenter, T. O., & Insogna, K. L. 1999. Changes in
bone turnover in young women consuming different levels of dietary
protein. The Journal of Clinical Endocrinology & Metabolism, 84(3),
1052-1055.

382
APPENDICES

CANCER
TABLE A3.24 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “CANCER” (N = 3)
Gao, Y., Ma, Y., Yu, M., Li, G., Chen, Y., Li, X., Wang, X. et al. 2022. Poultry and Fish Intake and Pancreatic Cancer Risk: A Systematic Review and Meta-Analysis. Nutrition and
Cancer, 74(1):55-67.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no yes 1.00
Q2 Protocol yes/partial yes/no partial yes 0.50
Q3 Explanation of included study design yes/no yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no partial yes 0.50
Q5 Paired study selection yes/no yes 1.00
Q6 Paired data extraction yes/no yes 1.00
Q7 List of excluded studies yes/partial yes/no no 0.00
Q8 Description of included studies yes/partial yes/no partial yes 0.50
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI partial yes 0.50
Q10 Sources of funding for included studies yes/no no 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q13 Impact of risk of bias in individual studies when yes/no no 0.00
interpreting results
Q14 Heterogeneity assessed yes/no yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted yes 1.00
Q16 Conflict of interest included yes/no yes 1.00
Total score 11.00
Percent 69%
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

383
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.24 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “CANCER” (N = 3) (cont.)
Jayedi, A. & Shab-Bidar, S. 2020. Fish consumption and the risk of chronic disease: an umbrella review of meta-analyses of prospective cohort studies. Advances in Nutrition,
11(5):1123-1133.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no yes 1.00
Q2 Protocol yes/partial yes/no partial yes 0.50
Q3 Explanation of included study design yes/no yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no partial yes 0.50
Q5 Paired study selection yes/no yes 1.00
Q6 Paired data extraction yes/no yes 1.00
Q7 List of excluded studies yes/partial yes/no yes 1.00
Q8 Description of included studies yes/partial yes/no partial yes 0.50
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI partial yes 0.50
Q10 Sources of funding for included studies yes/no yes 1.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q13 Impact of risk of bias in individual studies when yes/no yes 1.00
interpreting results
Q14 Heterogeneity assessed yes/no yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted yes 1.00
Q16 Conflict of interest included yes/no yes 1.00
Total score 14.00
Percent 87.50
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

384
APPENDICES

TABLE A3.24 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “CANCER” (N = 3) (cont.)
Kazemi, A., Barati-Boldaji, R., Soltani, S., Mohammadipoor, N., Esmaeilinezhad, Z., Clark, C. C., Akbarzadeh, M. et al. 2021. Intake of various food groups and risk of breast
cancer: A systematic review and dose-response meta-analysis of prospective studies. Advances in Nutrition, 12(3):809-849.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no yes 1.00
Q2 Protocol yes/partial yes/no yes 1.00
Q3 Explanation of included study design yes/no yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no yes 1.00
Q5 Paired study selection yes/no yes 1.00
Q6 Paired data extraction yes/no yes 1.00
Q7 List of excluded studies yes/partial yes/no no 0.00
Q8 Description of included studies yes/partial yes/no partial yes 0.50
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI partial yes 0.50
Q10 Sources of funding for included studies yes/no yes 1.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q13 Impact of risk of bias in individual studies when yes/no yes 1.00
interpreting results
Q14 Heterogeneity assessed yes/no yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted yes 1.00
Q16 Conflict of interest included yes/no yes 1.00
Total score 14.00
Percent 87.50
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

385
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.25 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “CANCER” (N = 10)
Quality assessment
Reference primary study (n = 10) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Aglago, E.K. Huybrechts, I. Murphy, N. Casagr, E., C. Nicolas, G. B
Pischon, T. Fedirko, V. Severi, G. Boutron-Ruault, M.C. et al. 2020.
Consumption of Fish and Long-chain n-3 Polyunsaturated Fatty
Acids Is Associated With Reduced Risk of Colorectal Cancer in a
Large European Cohort. Clin Gastroenterol Hepatol 18:654-666.
e656.
Bradbury, K.E. Murphy, N. & Key, T.J. 2020. Diet and colorectal cancer B
in UK Biobank: a prospective study. Int J Epidemiol, 49:246-258.
Cai, H. Sobue, T. Kitamura, T. Ishihara, J. Sawada, N. Iwasaki, M. B
Shimazu, T. & Tsugane, S. 2020. Association between meat and
saturated fatty acid intake and lung cancer risk: The Japan Public
Health Center-based prospective study. Int J Cancer, 147:3019-
3028.
Dianatinasab, M. Wesselius, A. de Loeij, T. Salehi-Abargouei, A. B
Yu, E.Y.W. Fararouei, M. Brinkman, M. et al. 2021. The association
between meat and fish consumption and bladder cancer risk: a
pooled analysis of 11 cohort studies. Eur J Epidemiol, 36:781-792.
Etemadi, A. Abnet, C.C. Graubard, B.I. Beane-Freeman, L. Freedman, B
N.D. Liao, L. Dawsey, S.M. & Sinha, R. 2018. Anatomical subsite can
modify the association between meat and meat compounds and risk
of colorectal adenocarcinoma: Findings from three large US cohorts.
Int J Cancer, 143:2261-2270.
Hermans, K. van den Brandt, P.A. Loef, C. Jansen, R.L.H. & Schouten, B
L.J. 2021. Meat consumption and cancer of unknown primary (CUP)
risk: results from The Netherlands cohort study on diet and cancer.
Eur J Nutr, 60:4579-4593.
Ma, Y. Yang, W. Li, T. Liu, Y. Simon, T.G. Sui, J. Wu, K. Giovannucci, B
E.L. Chan, A.T. & Zhang, X. 2019. Meat intake and risk of
hepatocellular carcinoma in two large US prospective cohorts of
women and men. Int J Epidemiol, 48:1863-1871.
Makiuchi, T. Sobue, T. Kitamura, T. Ishihara, J. Sawada, N. Iwasaki, B
M. Yamaji, T. Shimazu, T. & Tsugane, S. 2019. Relationship between
Meat/Fish Consumption and Biliary Tract Cancer: The Japan
Public Health Center-Based Prospective Study. Cancer Epidemiol
Biomarkers Prev, 29:95-102.
Outzen, M. Tjønnel, A. Christensen, J. & Olsen, A. 2018. Fish B
consumption and prostate cancer risk and mortality in a Danish
cohort study. Eur J Cancer Prev, 27:355-360.
Zamani, S.A. McClain, K.M. Graubard, B.I. Liao, L.M. Abnet, C.C. B
Cook, M.B. & Petrick, J.L. 2020. Dietary Polyunsaturated Fat Intake
in Relation to Head and Neck, Esophageal, and Gastric Cancer
Incidence in the National Institutes of Health-AARP Diet and Health
Study. Am J Epidemiol, 189:1096-1113.

386
APPENDICES

CARDIOVASCULAR DISEASES AND OUTCOMES

TABLE A3.26 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “CARDIOVASCULAR DISEASES
AND OUTCOMES” (N= 3)
Chowdhury, R. et al. 2012. Association between fish consumption, long chain omega 3 fatty acids, and risk of cerebrovascular disease: systematic review and meta-analysis.
Bmj-British Medical Journal, 345.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no yes 1.00
Q2 Protocol yes/partial yes/no yes 1.00
Q3 Explanation of included study design yes/no yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no yes 1.00
Q5 Paired study selection yes/no yes 1.00
Q6 Paired data extraction yes/no yes 1.00
Q7 List of excluded studies yes/partial yes/no no 0.00
Q8 Description of included studies yes/partial yes/no yes 1.00
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI no 0.00
Q10 Sources of funding for included studies yes/no no 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted no 0.00
Q13 Impact of risk of bias in individual studies when yes/no no 0.00
interpreting results
Q14 Heterogeneity assessed yes/no yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted yes 1.00
Q16 Conflict of interest included yes/no yes 1.00
Total score 11.00
Percent 68.75
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

387
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.26 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “CARDIOVASCULAR DISEASES
AND OUTCOMES” (N= 3) (cont.)
Li, N. et al. 2020. Fish consumption and multiple health outcomes: Umbrella review. Trends in Food Science & Technology, 99:273-283.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no no 0.00
Q2 Protocol yes/partial yes/no yes 1.00
Q3 Explanation of included study design yes/no no 0.00
Q4 Comprehensive literature search strategy yes/partial yes/no yes 1.00
Q5 Paired study selection yes/no no 0.00
Q6 Paired data extraction yes/no yes 1.00
Q7 List of excluded studies yes/partial yes/no yes 1.00
Q8 Description of included studies yes/partial yes/no no 0.00
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI no 0.00
Q10 Sources of funding for included studies yes/no no 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted no meta-analysis conducted
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted no meta-analysis conducted
Q13 Impact of risk of bias in individual studies when yes/no no 0.00
interpreting results
Q14 Heterogeneity assessed yes/no no 0.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted no meta-analysis conducted
Q16 Conflict of interest included yes/no yes 1.00
Total score 5.00
Percent 31.25
Percent (exclude N/A question) 38.46
Overall AMSTAR 2 judgement (confidence in the results) Critically low
Include/exclude Exclude

388
APPENDICES

TABLE A3.26 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “CARDIOVASCULAR DISEASES
AND OUTCOMES” (N= 3) (cont.)
Mente, A. et al. 2009. A Systematic Review of the Evidence Supporting a Causal Link Between Dietary Factors and Coronary Heart Disease. Archives of Internal Medicine,
169(7):659-669.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no yes 1.00
Q2 Protocol yes/partial yes/no partial yes 0.50
Q3 Explanation of included study design yes/no yes 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no partial yes 0.50
Q5 Paired study selection yes/no no 1.00
Q6 Paired data extraction yes/no no 0.00
Q7 List of excluded studies yes/partial yes/no yes 1.00
Q8 Description of included studies yes/partial yes/no yes 1.00
Q9 Risk of bias tool yes/partial yes/no/includes only RCTs or NRSI yes 1.00
Q10 Sources of funding for included studies yes/no no 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted yes 1.00
Q13 Impact of risk of bias in individual studies when yes/no yes 1.00
interpreting results
Q14 Heterogeneity assessed yes/no yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted yes 1.00
Q16 Conflict of interest included yes/no yes 1.00
Total score 13.00
Percent 81.25
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

389
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.27 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “CARDIOVASCULAR DISEASES AND
OUTCOMES” (N = 10)
Quality assessment
Reference primary study (n = 10) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Acosta, S., Johansson, A., & Drake, I. 2021. Diet and lifestyle factors B
and risk of atherosclerotic cardiovascular disease—a prospective
cohort study. Nutrients, 13(11), 3822.
Frost, L. & Vestergaard, P. 2005. n−3 Fatty acids consumed from B
fish and risk of atrial fibrillation or flutter: the Danish Diet, Cancer,
and Health Study1–3. The American Journal of Clinical Nutrition,
81(1), 50-54.
Gammelmark, A., Nielsen, M. S., Bork, C. S., Lundbye-Christensen, B
S., Overvad, K., & Schmidt, E. B. 2015. Fish Consumption and
Adipose Tissue Content of Marine n-3 PUFA is Inversely Associated
With Myocardial Infarction: A Danish Prospective Cohort Study.
Circulation, 132:A12418-A12418.
Lajous, M., Willett, W. C., Robins, J., Young, J. G., Rimm, E., B
Mozaffarian, D., & Hernán, M. A. 2013. Changes in fish consumption
in midlife and the risk of coronary heart disease in men and women.
American Journal of Epidemiology, 178(3), 382-391.
Lasota, A. N., Grønholdt, M. L. M., Bork, C. S., Lundbye-Christensen, B
S., Schmidt, E. B., & Overvad, K. 2019. Substitution of poultry and
red meat with fish and the risk of peripheral arterial disease: a
Danish cohort study. European Journal of Nutrition, 58, 2731-2739.
Matheson, E. M., Mainous III, A. G., Hill, E. G., & Carnemolla, M. A. B
2009. Shellfish consumption and risk of coronary heart disease.
Journal of the American Dietetic Association, 109(8), 1422-1426.
Petermann-Rocha, F., Parra-Soto, S., Gray, S., Anderson, J., Welsh, B
P., Gill, J., Pell, J. P. et al. 2021. Vegetarians, fish, poultry, and meat-
eaters: who has higher risk of cardiovascular disease incidence and
mortality? A prospective study from UK Biobank. European Heart
Journal, 42(12), 1136-1143.
Tong, T. Y., Appleby, P. N., Bradbury, K. E., Perez-Cornago, A., Travis, B
R. C., Clarke, R., & Key, T. J. 2019. Risks of ischaemic heart disease
and stroke in meat eaters, fish eaters, and vegetarians over 18
years of follow-up: results from the prospective EPIC-Oxford study.
BMJ, 366.
Venø, S. K., Bork, C. S., Jakobsen, M. U., Lundbye-Christensen, S., B
Bach, F. W., McLennan, P. L., Overvad, K. et al. 2018. Substitution of
fish for red meat or poultry and risk of ischemic stroke. Nutrients,
10(11), 1648.
Zhong, V. W., Allen, N. B., Greenland, P., Carnethon, M. R., Ning, H., B
Wilkins, J. T., Van Horn, L. et al. 2021. Protein foods from animal
sources, incident cardiovascular disease and all-cause mortality: a
substitution analysis. International Journal of Epidemiology, 50(1),
223-233.

390
APPENDICES

TYPE 2 DIABETES
TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9)
Muley, A. Muley, P. & Shah, M. 2014. ALA, fatty fish or marine n-3 fatty acids for preventing DM?: a systematic review and meta-analysis. Current diabetes reviews, 10(3):158-
165.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the population of >18 years (P) 1
to find out if dietary intake of fish and n-3 PUFA
(I) as compared to none or in lower quantiIe (C)
reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Not mentioned 0
Q3 Explanation of included study design yes/no They included prospective cohort studies only. 0
They stated RCTs were not included as no studies
were available.
Q4 Comprehensive literature search strategy yes/partial yes/no Searched three databases (PubMed, EMBASE and 1
GOOGLE), and cross references from first yield
were hand searched.
Q5 Paired study selection yes/no In triplicate 1
Q6 Paired data extraction yes/no In triplicate 1
Q7 List of excluded studies yes/partial yes/no The reasons are given, but the studies are not 0.5
listed.
Q8 Description of included studies yes/partial yes/no Country, follow-up, age, sex, exposure 1
assessment, exposure type is explained.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI The internal validity of studies was assessed 1
based on the Cochrane collaboration’s tool for
assessment of bias but no study was excluded for
its quality regarding RoB.
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted The internal validity of studies was assessed 1
based on the Cochrane Collaboration’s tool for
assessment of bias, but no study was excluded
for its quality regarding RoB.
Q13 Impact of risk of bias in individual studies when yes/no No 0
interpreting results
Q14 Heterogeneity assessed yes/no I2 was used as a measure for heterogeneity, and 1
is discussed.
Q15 Publication bias assessed yes/no/no meta-analysis conducted No 0
Q16 Conflict of interest included yes/no The authors confirm that they have no conflict 1
of interest.
Total score 10.5
Percent 66%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

391
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Tian, S., Xu, Q., Jiang, R., Han, T., Sun, C. & Na, L. 2017. Dietary protein consumption and the risk of type 2 diabetes: a systematic review and meta-analysis of cohort studies.
Nutrients, 9(9):982.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to find out if dietary 1
intake of fish protein (I) as compared to other
protein sources (C) reduced the risk of diabetes
(O).
Q2 Protocol yes/partial yes/no Not mentioned 0
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Searched two databases (PubMed, EMBASE). 1
Q5 Paired study selection yes/no Not mentioned 0
Q6 Paired data extraction yes/no Not mentioned 0
Q7 List of excluded studies yes/partial yes/no The reasons are given, but the studies are not 0.5
listed.
Q8 Description of included studies yes/partial yes/no Adequately described in supplementary 1
information
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Not mentioned 0
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Not mentioned 0
Q13 Impact of risk of bias in individual studies when yes/no Not mentioned 0
interpreting results
Q14 Heterogeneity assessed yes/no I2 was used as a measure for heterogeneity. 1
Q15 Publication bias assessed yes/no/no meta-analysis conducted Egger linear regression test and Begg rank 1
correlation test were used to search for
publication bias.
Q16 Conflict of interest included yes/no The authors declare no conflict of interest. 1
Total score 8.50
Percent 53
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

392
APPENDICES

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Wallin, A., Di Giuseppe, D., Orsini, N., Patel, P.S., Forouhi, N.G. & Wolk, A. 2012. Fish consumption, dietary long-chain n-3 fatty acids, and risk of type 2 diabetes: systematic
review and meta-analysis of prospective studies. Diabetes care, 35(4):918-929.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
fish consumption (I) as compared to other food
(C) reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Yes – adhering to the PRISMA statement. 1
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Searched two databases (PubMed, EMBASE), and 1
used reference lists from retrieved articles.
Q5 Paired study selection yes/no In duplicate 1
Q6 Paired data extraction yes/no In duplicate 1
Q7 List of excluded studies yes/partial yes/no The reasons are given, and articles are found in 0.5
the reference list.
Q8 Description of included studies yes/partial yes/no Country, follow-up, age, sex, exposure 1
assessment, exposure type listed.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Newcastle-Ottawa Quality Assessment scale 1
was used.
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Quality score by Newcastle-Ottawa Quality 1
Assessment scale was used.
Q13 Impact of risk of bias in individual studies when yes/no Discussed 1
interpreting results
Q14 Heterogeneity assessed yes/no I2 and Cochrane Q-test were used to evaluate 1
heterogeneity.
Q15 Publication bias assessed yes/no/no meta-analysis conducted Potential publication bias was assessed using 1
Egger regression asymmetry test.
Q16 Conflict of interest included yes/no The authors declare no conflict of interest. 1
Total score 14.5
Percent 91%
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

393
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Wu, J.H., Micha, R., Imamura, F., Pan, A., Biggs, M.L., Ajaz, O., Mozaffarian, D. et al. 2012. Omega-3 fatty acids and incident type 2 diabetes: a systematic review and meta-
analysis. British journal of nutrition, 107(S2):S214-S227.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
fish consumption (I) as compared other food (C)
reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Yes. The Meta-analyses of Observational studies 1
in Epidemiology guidelines (MOOSE) was used.
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Searched three databases (MEDLINE, EMBASE, 1
LILACS). Related articles were hand searched.
Q5 Paired study selection yes/no Title and abstracts were screened by one 1
investigator. Two investigators assessed
independently and in duplicate the full text.
Q6 Paired data extraction yes/no Data were extracted independently and in 1
duplicate.
Q7 List of excluded studies yes/partial yes/no The reasons are given, and articles are found in 0.5
the reference list.
Q8 Description of included studies yes/partial yes/no Country, follow-up, age, sex, exposure 1
assessment, exposure type listed.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Five criteria are described. 1
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted MOOSE was used. 1
Q13 Impact of risk of bias in individual studies when yes/no Not mentioned 0
interpreting results
Q14 Heterogeneity assessed yes/no I2 was used to evaluate heterogeneity. 1
Q15 Publication bias assessed yes/no/no meta-analysis conducted Potential publication bias was assessed by visual 1
inspection of funnel plots and using Begg’s test.
Q16 Conflict of interest included yes/no One of the authors reports research grant from 1
the industry.
Total score 13.5
Percent 84%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

394
APPENDICES

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Xun, P. & He, K. 2012. Fish consumption and incidence of diabetes: meta-analysis of data from 438,000 individuals in 12 independent prospective cohorts with an average
11-year follow-up. Diabetes care, 35(4):930-938.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if fish 1
consumption (I) as compared to other food (C)
reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Yes. The Meta-analyses of Observational studies 1
in Epidemiology guidelines (MOOSE) was used.
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Only one database (PubMed) was used, but 0.5
articles were also retrieved through Google and
hand search of the references from retrieved
articles.
Q5 Paired study selection yes/no Two investigators assessed independently. 1
Q6 Paired data extraction yes/no Not stated 0
Q7 List of excluded studies yes/partial yes/no The reasons are given, and articles are found in 1
the reference list.
Q8 Description of included studies yes/partial yes/no Country, follow-up, age, sex, exposure 1
assessment, exposure type listed.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI MOOSE scoring was used. 1
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Five criteria are described 1
Q13 Impact of risk of bias in individual studies when yes/no Discussed 1
interpreting results
Q14 Heterogeneity assessed yes/no Study selection is described 1
Q15 Publication bias assessed yes/no/no meta-analysis conducted Potential publication bias was assessed using 1
Egger test.
Q16 Conflict of interest included yes/no The authors report no conflict of interest. 1
Total score 13.5
Percent 84%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

395
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Yanai, H., Hamasaki, H., Katsuyama, H., Adachi, H., Moriyama, S. & Sako, A. 2015. Effects of intake of fish or fish oils on the development of diabetes. Journal of clinical
medicine research, 7(1):8.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
fish consumption (I) as compared to other food
(C) reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no No information 0
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Not described 0
Q5 Paired study selection yes/no Not described 0
Q6 Paired data extraction yes/no Not described 0
Q7 List of excluded studies yes/partial yes/no Not described 0
Q8 Description of included studies yes/partial yes/no Country, age, sex, exposure assessment, exposure 0.5
type are listed. Follow-up is missing.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Not described 0
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted N/A
Q13 Impact of risk of bias in individual studies when yes/no Not mentioned 0
interpreting results
Q14 Heterogeneity assessed yes/no Heterogeneity is discussed 1
Q15 Publication bias assessed yes/no/no meta-analysis conducted N/A
Q16 Conflict of interest included yes/no The authors report no competing interest. 1
Total score 4.50
Percent 28.13
Percent (exclude n/a question) 34.62
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

396
APPENDICES

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Zhang, M., Picard-Deland, E. & Marette, A. 2013. Fish and marine omega-3 polyunsaturated fatty acid consumption and incidence of type 2 diabetes: a systematic review and
meta-analysis. International journal of endocrinology, 2013.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
fish consumption (I) as compared to other food
(C) reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Not described 0
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Three databases (PubMed, OViD and EMBASE) 1
were used, and cross references examined.
Q5 Paired study selection yes/no Two investigators assessed independently. 1
Q6 Paired data extraction yes/no Two investigators assessed independently. 1
Q7 List of excluded studies yes/partial yes/no The reasons are given, but articles are not in the 0.5
reference list.
Q8 Description of included studies yes/partial yes/no Country, follow-up, age, sex, exposure 1
assessment, exposure type are listed.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Scoring system is described. 1
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Three criteria are described. 1
Q13 Impact of risk of bias in individual studies when yes/no Discussed 1
interpreting results
Q14 Heterogeneity assessed yes/no An I2 was used to evaluate heterogeneity and 1
heterogeneity is discussed.
Q15 Publication bias assessed yes/no/no meta-analysis conducted Potential publication bias was assessed Egger’s 1
regression test and funnel plots.
Q16 Conflict of interest included yes/no The authors declare that there is no conflict of 1
interest.
Total score 13.5
Percent 84%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

397
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Zheng, J.S., Huang, T., Yang, J., Fu, Y.Q,. & Li, D. 2012. Marine N-3 polyunsaturated fatty acids are inversely associated with risk of type 2 diabetes in Asians: a systematic review
and meta-analysis. PLoS One. 7(9)
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
fish consumption (I) as compared to other food
(C) reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Followed the MOOSE protocol. 1
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Five databases (PubMed, Embase, Cochrane, 1
CNK1 and Chinese VIP database) were used.
Q5 Paired study selection yes/no Two investigators assessed independently. 1
Q6 Paired data extraction yes/no Not described 0
Q7 List of excluded studies yes/partial yes/no The reasons are given, but articles are not in the 0.5
reference list.
Q8 Description of included studies yes/partial yes/no Follow-up, age, exposure assessment, exposure 0.5
type and range are listed. Country and sex are
missing.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Stated in the supplementary Table 1 that this was 0
done, but not described in the paper.
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described. 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Not described 0
Q13 Impact of risk of bias in individual studies when yes/no Not mentioned 0
interpreting results
Q14 Heterogeneity assessed yes/no An I2 was used to evaluate heterogeneity. 1
Heterogeneity is discussed.
Q15 Publication bias assessed yes/no/no meta-analysis conducted Potential publication bias was assessed. Egger’s 1
regression test and Begg’s funnel plots.
Q16 Conflict of interest included yes/no The authors declare that there is no conflict of 1
interest.
Total score 10.0
Percent 63%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

398
APPENDICES

TABLE A3.28 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “DIABETES” (N = 9) (cont.)
Zhou, Y., Tian, C. & Jia, C. 2012. Association of fish and n-3 fatty acid intake with the risk of type 2 diabetes: a meta-analysis of prospective studies. British Journal of Nutrition,
108(3):408-417.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
high fish consumption (I) as compared to low
consumption (C) reduced the risk of diabetes (O).
Q2 Protocol yes/partial yes/no Not described 0
Q3 Explanation of included study design yes/no Study selection is described. 1
Q4 Comprehensive literature search strategy yes/partial yes/no Five databases (PubMed, Web of Science, CBM, 1
VIP and CNK1) were used. References from
retrieved references were hand searched.
Q5 Paired study selection yes/no Two investigators assessed independently. 1
Q6 Paired data extraction yes/no Two investigators assessed independently (stated 1
in author contribution).
Q7 List of excluded studies yes/partial yes/no Not described 0
Q8 Description of included studies yes/partial yes/no Country, age, follow-up, sex, exposure 1
assessment, exposure type and range are listed.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Newcastle-Ottawa scale was used. 1
Q10 Sources of funding for included studies yes/no Not mentioned 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Adequately described 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Newcastle-Ottawa scale was used. 1
Q13 Impact of risk of bias in individual studies when yes/no Discussed 1
interpreting results
Q14 Heterogeneity assessed yes/no An I2 and Q-test were used to evaluate 1
heterogeneity. Heterogeneity is discussed.
Q15 Publication bias assessed yes/no/no meta-analysis conducted Potential publication bias was assessed Egger’s 1
test.
Q16 Conflict of interest included yes/no The authors declare that there is no conflict of 1
interest.
Total score 13
Percent 81%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

399
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.29 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “TYPE 2 DIABETES” (N = 1)
Quality assessment
Reference primary study (n = 1) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Chen, Z. Franco, O.H. Lamballais, S. Ikram, M.A. Schoufour, J.D. B
Muka, T. & Voortman, T. 2020. Associations of specific dietary
protein with longitudinal insulin resistance, prediabetes and type 2
diabetes: The Rotterdam Study. Clinical nutrition, 39(1):242-249.

400
APPENDICES

NEURODEVELOPMENT AND NEUROLOGICAL DISORDERS

TABLE A3.30 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “NEURODEVELOPMENT AND
NEUROLOGICAL DISORDERS” (N = 3)
Ernst, E. 1999. Diet and dementia, is there a link? A systematic review. Nutritional Neuroscience, 2(1):1-6.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no No 0
Q2 Protocol yes/partial yes/no No 0
Q3 Explanation of included study design yes/no No 0
Q4 Comprehensive literature search strategy yes/partial yes/no Partial yes 0.5
Q5 Paired study selection yes/no No 0
Q6 Paired data extraction yes/no No 0
Q7 List of excluded studies yes/partial yes/no No 0
Q8 Description of included studies yes/partial yes/no No 0
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI No 0
Q10 Sources of funding for included studies yes/no No 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted No meta-analysis N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No meta-analysis N/A
Q13 Impact of risk of bias in individual studies when yes/no Yes 1
interpreting results
Q14 Heterogeneity assessed yes/no No 0
Q15 Publication bias assessed yes/no/no meta-analysis conducted No meta-analysis N/A
Q16 Conflict of interest included yes/no No 0
Total score 1.5
Percent 10%
Percent (exclude n/a question) 13%
Overall AMSTAR 2 judgement (confidence in the results) Critically low
Include/exclude Exclude

401
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.30 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “NEURODEVELOPMENT AND
NEUROLOGICAL DISORDERS” (N = 3) (cont.)
Murakami, K. & Sasaki, S. 2010. Dietary intake and depressive symptoms: a systematic review of observational studies. Molecular nutrition & food research, 54(4):471-488.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes
Q2 Protocol yes/partial yes/no No, no risk of bias assessment 0
Q3 Explanation of included study design yes/no Yes 1
Q4 Comprehensive literature search strategy yes/partial yes/no No, 1 database only 0
Q5 Paired study selection yes/no No 0
Q6 Paired data extraction yes/no No 0
Q7 List of excluded studies yes/partial yes/no No 0
Q8 Description of included studies yes/partial yes/no Partial yes
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Yes 1
Q10 Sources of funding for included studies yes/no No 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted No meta-analysis N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No meta-analysis N/A
Q13 Impact of risk of bias in individual studies when yes/no No 0
interpreting results
Q14 Heterogeneity assessed yes/no No 0
Q15 Publication bias assessed yes/no/no meta-analysis conducted No meta-analysis N/A
Q16 Conflict of interest included yes/no Yes 1
Total score 3
Percent 19%
Percent (exclude n/a question) 23%
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

402
APPENDICES

TABLE A3.30 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “NEURODEVELOPMENT AND
NEUROLOGICAL DISORDERS” (N = 3) (cont.)
Starling, P., Charlton, K., McMahon, A.T. & Lucas, C. 2015. Fish intake during pregnancy and foetal neurodevelopment—A systematic review of the evidence. Nutrients,
7(3):2001-2014.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Partial yes 0.5
Q2 Protocol yes/partial yes/no No 0
Q3 Explanation of included study design yes/no Yes 1
Q4 Comprehensive literature search strategy yes/partial yes/no Partial yes 0.5
Q5 Paired study selection yes/no No 0
Q6 Paired data extraction yes/no No 0
Q7 List of excluded studies yes/partial yes/no No 0
Q8 Description of included studies yes/partial yes/no Partial yes 0.5
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI No, no selection bias assessed 0
Q10 Sources of funding for included studies yes/no No 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted No meta-analysis N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No meta-analysis N/A
Q13 Impact of risk of bias in individual studies when yes/no Yes 1
interpreting results
Q14 Heterogeneity assessed yes/no No 0
Q15 Publication bias assessed yes/no/no meta-analysis conducted No meta-analysis N/A
Q16 Conflict of interest included yes/no Yes 1
Total score 1.5
Percent 28%
Percent (exclude n/a question) 34%
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

403
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.31 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “NEURODEVELOPMENT AND
NEUROLOGICAL DISORDERS” (N = 14)
Quality assessment
Reference primary study (n = 14) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Al-Ghannami, S.S. Al-Adawi, S. Ghebremeskel, K. Hussein, I.S. Min, B
Y. Jeyaseelan, L. Al-Oufi, H. S. et al. 2019. Randomized open-label
trial of docosahexaenoic acid–enriched fish oil and fish meal on
cognitive and behavioral functioning in Omani children. Nutrition,
57:167-172.
Almeida, O.P. Norman, P. Hankey, G. Jamrozik, K. & Flicker, L. 2006. C Aim of determining the lifestyle and clinical factors associated with
Successful mental health aging: results from a longitudinal study of successful mental health aging in a cohort
older Australian men. The American journal of geriatric psychiatry,
14(1):27-35.
Browne, J.C. Scott, K.M. & Silvers, K.M. 2006. Fish consumption in C Investigates n-3 status postpartum and depression. Fish intake in
pregnancy and omega-3 status after birth are not associated with pregnancy was dichotomized.
postnatal depression. Journal of affective disorders, 90(2-3):131-
139.
Danthiir, V. Hosking, D. Burns, N.R. Wilson, C. Nettelbeck, T. C Study design
Calvaresi, E. Wittert, G.A. et al. 2014. Cognitive performance in
older adults is inversely associated with fish consumption but not
erythrocyte membrane n–3 fatty acids. The Journal of nutrition,
144(3):311-320.
Emmett, P.M. Jones, L.R. & Golding, J. 2015. Pregnancy diet and C Reviews publications that have used ALSPAC data to report on diet
associated outcomes in the Avon Longitudinal Study of Parents and during pregnancy relative to the growth and development of the
Children. Nutrition reviews, 73(suppl_3):154-174. offspring, as well as to some maternal outcomes
García-Esquinas, E. Ortolá, R. Banegas, J.R. Lopez-García, E. & C Response rate not included. Outcome/endpoint not relevant
Rodríguez-Artalejo, F. 2019. Dietary n-3 polyunsaturated fatty
acids, fish intake and healthy ageing. International Journal of
Epidemiology, 48(6):1914-1924.
Lehner, A. Staub, K. Aldakak, L. Eppenberger, P. Rühli, F. Martin, C Outcome not clearly formulated. A cohort-study but a cross-
R.D. & Bender, N. 2020. Fish consumption is associated with school sectional study (not prospective).
performance in children in a non-linear way: results from the
German cohort study KiGGS. Evolution, medicine, and public health,
(1):2-11.
Matsuoka, Y.J. Sawada, N. Mimura, M. Shikimoto, R. Nozaki, C Food frequency questionnaire not used at baseline, important
S. Hamazaki, K. Tsugane, S. et al. 2017. Dietary fish, n-3 confounders not identified and considered by the authors
polyunsaturated fatty acid consumption, and depression risk in
Japan: a population-based prospective cohort study. Translational
psychiatry, 7(9):e1242-e1242.
Mesirow, M.S. Cecil, C. Maughan, B. & Barker, E. D. 2017. B
Associations between prenatal and early childhood fish and
processed food intake, conduct problems, and co-occurring
difficulties. Journal of abnormal child psychology, 45:1039-1049.
Nathanson, R. Hill, B. Skouteris, H. & Bailey, C. 2018. Antenatal C Response rate not acceptable; fish intake question not included at
diet and postpartum depressive symptoms: A prospective study. baseline
Midwifery, 62:69-76.
Schiepers, O.J. de Groot, R.H. Jolles, J. & van Boxtel, M.P. 2009. C Response rate at follow-up was very low (<10%); fish intake
Plasma phospholipid fatty acid status and depressive symptoms: question not included at baseline
association only present in the clinical range. Journal of Affective
Disorders, 118(1-3):209-214.
Shapouri-Moghaddam, A. Bagherniya, M. Ehteshamfar, S.M. Rahimi, C Power calculation not reported; low participant number; study
H. & Safarian, M. 2017. High fish consumption decreased the design maybe not suitable for the research hypothesis
likelihood of depressive symptoms in community-living older people:
a randomized-controlled trial. Journal of gerontology and geriatrics,
65:232-237.
Sharifan, P. Hosseini, M.S. & Sharifan, A. 2017. The interventional C Recruited patients (have depression severity 10-29 BDI), included
relationship between frequent fish consumption and depression fish oil in intervention
symptoms in aging adults: A randomized controlled trial.
International Journal of Geriatric Psychiatry, 32(12):e116-e122.
van de Rest, O. Wang, Y. Barnes, L.L. Tangney, C. Bennett, D.A. & C Investigated if APOE e4 modifies the association between seafood
Morris, M.C. 2016. APOE ε4 and the associations of seafood and and n-3 fatty acid intakes and domain-specific cognitive decline in
long-chain omega-3 fatty acids with cognitive decline. Neurology, a community-based, prospective study
86(22):2063-2070.

404
APPENDICES

MORTALITY
TABLE A3.32 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “MORTALITY” (N = 5)
He, K., Song, Y., Daviglus, M.L., Liu, K., Van Horn, L., Dyer, A.R. et al. 2004. Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-analysis of
cohort studies. Circulation, 109(22):2705-11.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1.00
fish consumption (>1-3/month, 1/week etc. (I) as
compared to less (<1mo) (C) reduced the risk of
CHD mortality (O).
Q2 Protocol yes/partial yes/no No, not mentioned. 0.00
Q3 Explanation of included study design yes/no Yes, study selection is described (included 1.00
prospective cohort studies)
Q4 Comprehensive literature search strategy yes/partial yes/no partial, two databases (Medline + EMBASE + 0.50
language + search terms).
Q5 Paired study selection yes/no Yes, duplicate 1.00
Q6 Paired data extraction yes/no Yes, duplicate 1.00
Q7 List of excluded studies yes/partial yes/no Partial – the reasons are given, and articles are 0.50
found in the reference list.
Q8 Description of included studies yes/partial yes/no Yes – country, follow-up, age, sex, exposure 0.50
assessment, exposure type listed
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Yes – Beggs + Egger’s. 0.00
Q10 Sources of funding for included studies yes/no No, not mentioned 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes – adequately described 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Yes Beggs + Egger’s 0.00
Q13 Impact of risk of bias in individual studies when yes/no Yes 1.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes, by a meta regression analysis. 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted Yes, adequately described 1.00
Q16 Conflict of interest included yes/no No, not mentioned 0.00
Total score 9.50
Percent 59
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

405
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.32 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “MORTALITY” (N = 5) (cont.)
Szymanski, K.M., Wheeler, D.C. & Mucci, L.A. 2010. Fish consumption and prostate cancer risk: a review and meta-analysis. The American journal of clinical nutrition,
92(5):1223-33.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1.00
higher fish consumption (C) reduced the risk of
prostate cancer mortality (O).
Q2 Protocol yes/partial yes/no Yes – protocol and guidelines 1.00
Q3 Explanation of included study design yes/no Yes. Study selection is described (prospective 1.00
cohort studies).
Q4 Comprehensive literature search strategy yes/partial yes/no Partial – three databases (Medline + EMBASE + 0.50
ProQuest + search terms + ref. list).
Q5 Paired study selection yes/no Yes – duplicate 1.00
Q6 Paired data extraction yes/no Yes – duplicate 1.00
Q7 List of excluded studies yes/partial yes/no Partial. The reasons are given, and articles are 0.50
found in the reference list.
Q8 Description of included studies yes/partial yes/no Yes. Country, follow-up, age, sex, exposure 1.00
assessment, exposure type listed.
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI No, not mentioned 0.00
Q10 Sources of funding for included studies yes/no No, not mentioned 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes, adequately described 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No, not mentioned 0.00
Q13 Impact of risk of bias in individual studies when yes/no No, not mentioned 0.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes, adequately described 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted Yes, by Egger and Begg + funnel plot 1.00
Q16 Conflict of interest included yes/no Yes. N conflict of interest is stated. 1.00
Total score 11.00
Percent 68.75
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

406
APPENDICES

TABLE A3.32 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “MORTALITY” (N = 5) (cont.)
Lovegrove, C., Ahmed, K., Challacombe, B., Khan, M.S., Popert, R. & Dasgupta, P. 2015. Systematic review of prostate cancer risk and association with consumption of fish and
fish-oils: analysis of 495,321 participants. Int J Clin Pract, 69(1):87-105.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1.00
higher fish consumption (C) reduced the risk of
prostate cancer mortality (O).
Q2 Protocol yes/partial yes/no No, not mentioned 0.00
Q3 Explanation of included study design yes/no No 0.00
Q4 Comprehensive literature search strategy yes/partial yes/no Partial, PubMed and Ovid, Medline 0.50
Q5 Paired study selection yes/no Yes, duplicate 0.00
Q6 Paired data extraction yes/no No, no analysis performed 0.00
Q7 List of excluded studies yes/partial yes/no No 0.00
Q8 Description of included studies yes/partial yes/no Partial 0.50
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI No 0.50
Q10 Sources of funding for included studies yes/no No 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted No meta-analysis performed N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No, not performed N/A
Q13 Impact of risk of bias in individual studies when yes/no No 0.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes 0.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted No 0.00
Q16 Conflict of interest included yes/no Yes 1.00
Total score 3.50
Percent 25.00
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

407
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.32 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “MORTALITY” (N = 5) (cont.)
Marckmann, P. & Grønbæk, M. 1999. Fish consumption and coronary heart disease mortality. A systematic review of prospective cohort studies. Eur J Clin Nutr, 53(8):585-90.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1.00
higher fish consumption (C) reduced the risk CHD
mortality (O).
Q2 Protocol yes/partial yes/no No, not mentioned 0.00
Q3 Explanation of included study design yes/no No 0.00
Q4 Comprehensive literature search strategy yes/partial yes/no No, only MEDLINE 0.00
Q5 Paired study selection yes/no No 0.00
Q6 Paired data extraction yes/no No 0.00
Q7 List of excluded studies yes/partial yes/no No 0.00
Q8 Description of included studies yes/partial yes/no Partial 0.50
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI No 0.50
Q10 Sources of funding for included studies yes/no No 0.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted No meta-analysis performed. N/A
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No meta-analysis performed. N/A
Q13 Impact of risk of bias in individual studies when yes/no Yes, each study was scored for scientific quality 1.00
interpreting results 0-6 points.
Q14 Heterogeneity assessed yes/no No 0.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted No 0.00
Q16 Conflict of interest included yes/no No 0.00
Total score 3.00
Percent 21.43
Percent (exclude N/A question)
Overall AMSTAR 2 judgement (confidence in the results) Low
Include/exclude Exclude

408
APPENDICES

TABLE A3.32 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS WITH AMSTAR 2 FOR “MORTALITY” (N = 5) (cont.)
Geelen, A., Schouten, J.M., Kamphuis, C., Stam, B.E., Burema, J., Renkema, J.M., Kampman, E. et al. 2007. Fish consumption, n-3 fatty acids, and colorectal cancer: a meta-
analysis of prospective cohort studies. American journal of epidemiology, 166(10):1116-1125.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, assessed the general (P) to investigate if 1
higher fish consumption (C) reduced the risk of
cancer mortality (O).
Q2 Protocol yes/partial yes/no No 0
Q3 Explanation of included study design yes/no Yes 1
Q4 Comprehensive literature search strategy yes/partial yes/no Yes 0.5
Q5 Paired study selection yes/no No 0
Q6 Paired data extraction yes/no No 1
Q7 List of excluded studies yes/partial yes/no No 0
Q8 Description of included studies yes/partial yes/no Yes 0.5
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI No 0
Q10 Sources of funding for included studies yes/no No 0
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes 1
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted Yes 0
Q13 Impact of risk of bias in individual studies when yes/no No 0
interpreting results
Q14 Heterogeneity assessed yes/no Yes 1
Q15 Publication bias assessed yes/no/no meta-analysis conducted Yes 1
Q16 Conflict of interest included yes/no Yes 1
Total score 8
Percent 50%
Percent (exclude n/a question)
Overall AMSTAR 2 judgement (confidence in the results) Moderate
Include/exclude Include

409
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.33 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “MORTALITY” (N = 16)
Quality assessment
Reference primary study (n = 16) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Iso, H., Kobayashi, M., Ishihara, J., Sasaki, S., Okada, K., Kita, Y., B
Kokubo, Y. & Tsugane, S. 2006. Intake of fish and n3 fatty acids and
risk of coronary heart disease among Japanese: the Japan Public
Health Center-Based (JPHC) Study Cohort I. Circulation, 113(2):
195-202.
Iso, H. & Kubota, Y. 2007. Nutrition and Disease in the Japan C The study mentions mortality, but data on this is not found in any of
Collaborative Cohort Study for Evaluation of Cancer. Asian Pacific the tables, nor is the hazard ratio for fish intake and mortality found
Journal of Cancer Prevention, 8: 35-80. in the study.
Jin, X., Xiong, S., Yuan, C., Gong, E., Zhang, X., Yao, Y., Leng, Y., C Energy intake not included/adjusted for + no response rate
Niu, Z., Zeng, Y. & Yan, L.L. 2021. Apolipoprotein E Genotype, Meat,
Fish, and Egg Intake in Relation to Mortality Among Older Adults: A
Longitudinal Analysis in China. Front Med (Lausanne), 8: 697389.
Kromhout, D., Feskens, E.J. & Bowles, C.H. 1995. The protective C The study design is not suited to test the research hypothesis, and
effect of a small amount of fish on coronary heart disease mortality number of participants too low
in an elderly population. Int J Epidemiol, 24(2): 340-5.
Kurozawa, Y., Ogimoto, I., Shibata, A., Nose, T., Yoshimura, T., Suzuki, C No confounders were adjusted for; energy level was not included;
H., Sakata, R. et al. 2004. Dietary habits and risk of death due to multiple testing was not adjusted for
hepatocellular carcinoma in a large scale cohort study in Japan.
Univariate analysis of JACC study data. Kurume Med J, 51(2): 141-9.
Lapidus, L., Andersson, H., Bengtsson, C. & Bosaeus, I. 1986. C Study design not suited to test the research hypothesis; study
Dietary habits in relation to incidence of cardiovascular disease mainly designed to test the impact of diet on myocardial infarction;
and death in women: a 12-year follow-up of participants in the and results on death only an additional endpoint and not presented
population study of women in Gothenburg, Sweden. Am J Clin Nutr, other than mentioned in the result briefly
44(4): 444-8.
Lv, Y.B., Kraus, V.B., Gao, X., Yin, Z.X., Zhou, J.H., Mao, C., Duan, J. C No detailed quantitative dietary intake evaluation performed; thus
et al. 2020. Higher dietary diversity scores and protein-rich food impossible to adjust for energy intake in the analyses
consumption were associated with lower risk of all-cause mortality
in the oldest old. Clinical Nutrition, 39(7): 2246-2254.
Mozaffarian, D., Stein, P.K., Prineas, R.J. & Siscovick, D.S. 2008. C Study design not suited to test the research hypothesis; incidence of
Dietary fish and omega-3 fatty acid consumption and heart rate fatal CHD according to differences in HRV (heart rate variability)
variability in US adults. Circulation, 117(9): 1130-1137.
Pertiwi, K., Küpers, L.K., de Goede, J., Zock, P.L., Kromhout, D. & B
Geleijnse, J.M. 2021. Dietary and Circulating Long-Chain Omega-3
Polyunsaturated Fatty Acids and Mortality Risk After Myocardial
Infarction: A Long-Term Follow-Up of the Alpha Omega Cohort. J Am
Heart Assoc: e022617.
Smit, E., Garcia-Palmieri, M.R., Figueroa, N.R., McGee, D.L., C Study participants had very low intake/null consumers of seafood
Messina, M., Freudenheim, J.L. & Crespo, C.J. 2007. Protein and intake, and not suitable to determine the impact from seafood
legume intake and prostate cancer mortality in Puerto Rican men. intake
Nutrition and Cancer-an International Journal, 58(2): 146-152.
Streppel, M.T., Ocke, M.C., Boshuizen, H.C., Kok, F.J. & Kromhout, B
D. 2008. Long-term fish consumption and n-3 fatty acid intake
in relation to (sudden) coronary heart disease death: the Zutphen
study. European Heart Journal, 29(16): 2024-2030.
Sun, Y., Liu, B., Rong, S., Zhang, J., Du, Y., Xu, G., Snetselaar, B
L.G., Wallace, R.B., Lehmler, H.J. & Bao, W. 2021. Association of
Seafood Consumption and Mercury Exposure With Cardiovascular
and All-Cause Mortality Among US Adults. JAMA Netw Open, 4(11):
e2136367.
Tabak, C., Feskens, E.J.M., Heederik, D., Kromhout, D., Menotti, A., C Dietary information collected only in small random samples (8 – 49)
Blackburn, H.W. & Seven Countries Study, G. 1998. Fruit and fish of each cohort, not for each individual
consumption: a possible explanation for population differences in
COPD mortality (the seven countries study). European Journal of
Clinical Nutrition, 52(11): 819-825.
Truong-Minh, P., Fujino, Y., Kubo, T., Ide, R., Tokui, N., Mizoue, T., C BMI and energy intake not included, nor mentioned as confounders
Ogimoto, I., Matsuda, S. & Yoshimura, T. 2009. Fish intake and the
risk of fatal prostate cancer: findings from a cohort study in Japan.
Public Health Nutrition, 12(5): 609-613.
Walda, I.C., Tabak, C., Smit, H.A., Rasanen, L., Fidanza, F., Menotti, B
A., Nissinen, A., Feskens, E.J.M. & Kromhout, D. 2002. Diet and 20-
year chronic obstructive pulmonary disease mortality in middle-aged
men from three European countries. European Journal of Clinical
Nutrition, 56(7): 638-643.
Yamagishi, K., Iso, H., Shimazu, T., Tamakoshi, A., Sawada, N., C Important confounders not identified/ascertained and considered
Matsuo, K., Ito, H. et al. 2019. Fish intake and risk of mortality due by authors
to aortic dissection and aneurysm: A pooled analysis of the Japan
cohort consortium. Clinical Nutrition, 38(4): 1678-1683.

410
APPENDICES

OVERWEIGHT AND OBESITY

TABLE A3.34 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “OVERWEIGHT AND OBESITY” (N = 7)
Quality assessment
Reference primary study (n = 7) Eventual reason for grade C
(risk of bias judgement) (A, B or C)
Beulen, Y. et al. 2018. Quality of Dietary Fat Intake and Body Weight B
and Obesity in a Mediterranean Population: Secondary Analyses
within the PREDIMED Trial. Nutrients, 10(12).
Cade, J.E. et al. 2004. The UK Women's Cohort Study: comparison of C Statistical analysis is not fully described; relevant confounders were
vegetarians, fish-eaters and meat-eaters. Public Health Nutrition, not adequately handled.
7(7):871-878.
Fatahi, S. et al. 2019. Effect of Weight Reduction Diets Containing C Research question is not clearly formulated; objective is to compare
Fish, Walnut or Fish plus Walnut on Cardiovascular Risk Factors the effects of walnuts, fish and the combination of the two on
in Overweight and Obese Women. Archives of Iranian Medicine, cardiovascular risk factors
22(10):574-583.
Riseberg, E. et al. 2022. Specific Dietary Protein Sources Are C Research question not clearly formulated
Associated with Cardiometabolic Risk Factors in the Boston Puerto
Rican Health Study. J Acad Nutr Diet, 122(2):298-308.
Rosell, M. et al. 2006. Weight gain over 5 years in 21 966 meat- C Analyses not adjusted for possible confounding factors
eating, fish-eating, vegetarian, and vegan men and women in
EPIC-Oxford. International Journal of Obesity, 30(9):1389-1396.
Smith, J.D. et al. 2015. Changes in intake of protein foods, B
carbohydrate amount and quality, and long-term weight change:
results from 3 prospective cohorts. American Journal of Clinical
Nutrition, 101(6):1216-1224.
Tørris, C. Molin, M. & Småstuen, M.C. 2017. Lean Fish Consumption B
Is Associated with Beneficial Changes in the Metabolic Syndrome
Components: A 13-Year Follow-Up Study from the Norwegian Tromsø
Study. Nutrients, 9(3).

411
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

MORTALITY
TABLE A3.35 QUALITY ASSESSMENT (RISK OF BIAS) OF THE 2022 VKM REPORT WITH AMSTAR-2
VKM (Norwegian Scientific Committee for Food and Environment). 2022. Benefit and risk assessment of fish in the Norwegian diet. Scientific Opinion of the Steering Committee
of the Norwegian Scientific Committee for Food and Environment. Oslo.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, page 50 in report. 1.00
Q2 Protocol yes/partial yes/no Yes, published in 2020. 1.00
Q3 Explanation of included study design yes/no Yes, pages 53-54 in report. 1.00
Q4 Comprehensive literature search strategy yes/partial yes/no Yes, pages 51-52 in report 1.00
Q5 Paired study selection yes/no Yes, page 52 in report 1.00
Q6 Paired data extraction yes/no No 0.00
Q7 List of excluded studies yes/partial yes/no No 0.00
Q8 Description of included studies yes/partial yes/no Yes 1.00
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Yes 1.00
Q10 Sources of funding for included studies yes/no Yes 1.00
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted Yes 1.00
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No 0.00
Q13 Impact of risk of bias in individual studies when yes/no Yes, only included studies graded A or B 1.00
interpreting results
Q14 Heterogeneity assessed yes/no Yes 1.00
Q15 Publication bias assessed yes/no/no meta-analysis conducted No 0.00
Q16 Conflict of interest included yes/no Yes 1.00
Total score 12.00
Percent 75%
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

412
APPENDICES

LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE”

TABLE A3.36 OVERVIEW OF THE LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE” FROM THE 2022 VKM REPORT AND
THE SYSTEMATIC LITERATURE SEARCH FOR ALL THE HEALTH OUTCOMES IN “EVIDENCE OF HEALTH BENEFITS FROM
FISH CONSUMPTION”
Health outcome Fish consumption Literature included in 2022 VKM Conclusion “weight of Literature included in the Final conclusion
report in the “weight of evidence” evidence” 2022 VKM systematic literature search in “weight of evidence”
conclusion report the final “weight of evidence” in background report
(number of systematic reviews and conclusion (including VKM, 2022
primary studies in VKM, 2022) (number of systematic reviews and and updated literature
primary studies) search)
Allergy and immunology
Allergic rhinitis in Maternal total fish VKM, 2022 Section 4.32: systematic Limited, no conclusion Systematic reviews, n = 1; primary Limited, no conclusion
children intake in pregnancy reviews, n = 2; primary studies, studies, n = 0
n=3
Allergic rhinitis in Early fish introduction VKM, 2022 Section 4.32: systematic Limited, no conclusion Systematic reviews, n = 1; primary Limited, no conclusion
children reviews, n = 1; primary studies, studies, n = 0
n=3
Allergic sensitization Maternal total fish VKM, 2022 Section 4.33: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
in children intake in pregnancy reviews, n = 2; primary studies, studies, n = 0
n=2
Allergic sensitization Child total fish intake VKM, 2022 Section 4.33: systematic Limited, no conclusion Systematic reviews, n = 1; primary Limited, no conclusion
in children reviews, n = 0; primary studies, studies, n = 0
n=2
Asthma in children Maternal total, fatty VKM, 2022 Section 4.31: systematic Limited, no conclusion Systematic reviews, n = 2; primary Limited, no conclusion
and lean intake in reviews, n = 3; primary studies, studies, n = 0
pregnancy n=4
Eczema in children Maternal total fish VKM, 2022 Section 4.29: systematic Limited, suggestive Systematic reviews, n = 2; primary Limited, no conclusion
intake in pregnancy reviews, n = 2; primary studies, (protective) studies, n = 0 (downgraded from
n=8 VKM, 2022)
Eczema in children Child total fish intake VKM, 2022 Section 4.29: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 1; primary studies, (protective for intake studies, n = 0 (protective for intake
n=2 in the first year of life, in the first year of life,
but not later) but not later)
Multiple sclerosis Total fish intake VKM, 2022 Section 4.34: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 1; primary studies, (protective) studies, n = 0 (protective)
n=2
Rheumatoid arthritis Total fish intake VKM, 2022 Section 4.16: systematic Limited, suggestive Systematic reviews, n = 2; primary Limited, no conclusion
reviews, n = 1; primary studies, (protective) studies, n = 0 (downgraded from VKM
n=6 2022)
Birth and growth outcomes
Preterm birth Maternal total fish VKM, 2022 Section 4.23: systematic Probable (protective Systematic reviews, n = 0; primary Probable (protective
intake in pregnancy reviews, n = 1; primary studies, effect) studies, n = 0 effect)
n = 11 (including one pooled
analysis consisting of n = 13 unique
European cohort studies)
Preterm birth Maternal fatty fish VKM, 2022 Section 4.23: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
intake in pregnancy reviews, n = 1; primary studies, n studies, n = 0
= 4 (including one pooled analysis
consisting of n = 11 European
unique cohort studies)
Preterm birth Maternal lean fish VKM, 2022 Section 4.23: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
intake in pregnancy reviews, n = 1; primary studies, n studies, n = 0
= 4 (including one pooled analysis
consisting of n = 10 European
unique cohort studies)
Small for gestational Maternal total fish VKM, 2022 Section 4.24: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
age intake in pregnancy reviews, n = 1; primary studies, n (protective) studies, n = 1 (protective)
= 9 (including one pooled analysis
consisting of n = 11 European cohort
studies)
Small for gestational Maternal fatty fish VKM, 2022 Section 4.24: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
age intake in pregnancy reviews, n = 1; primary studies, n studies, n = 0
= 5 (including one pooled analysis
consisting of n = 11 European
cohort studies)

413
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.36 OVERVIEW OF THE LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE” FROM THE 2022 VKM REPORT AND
THE SYSTEMATIC LITERATURE SEARCH FOR ALL THE HEALTH OUTCOMES IN “EVIDENCE OF HEALTH BENEFITS FROM
FISH CONSUMPTION” (cont.)
Health outcome Fish consumption Literature included in 2022 VKM Conclusion “weight of Literature included in the Final conclusion
report in the “weight of evidence” evidence” 2022 VKM systematic literature search in “weight of evidence”
conclusion report the final “weight of evidence” in background report
(number of systematic reviews and conclusion (including VKM, 2022
primary studies in VKM, 2022) (number of systematic reviews and and updated literature
primary studies) search)
Small for gestational Maternal lean fish VKM, 2022 Section 4.24: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
age intake in pregnancy reviews, n = 1; primary studies, n studies, n = 0
= 4 (including one pooled analysis
consisting of n = 10 European
cohort studies)
Birth weight Maternal total, fatty VKM, 2022 Section 4.26: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
and lean fish intake reviews, n = 0; primary studies, n (positive association) studies, n = 1 (positive association)
in pregnancy = 13 (including one pooled analysis
consisting of n = 13 European
cohort studies)
Birth weight Maternal fatty fish VKM, 2022 Section 4.26: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
intake in pregnancy reviews, n = 0; primary studies, n (positive association) studies, n = 1 (positive association)
= 4 (including one pooled analysis
consisting of n = 11 European
cohort studies)
Birth weight Maternal lean fish VKM, 2022 Section 4.26: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
intake in pregnancy reviews, n = 0; primary studies, n (positive association) studies, n = 1 (positive association)
= 3 (including one pooled analysis
consisting of n = 10 European
cohort studies)
Low birth weight Maternal total fish VKM, 2022 Section 4.25: systematic Probable (protective Systematic reviews, n = 0; primary Probable (protective
intake in pregnancy reviews, n = 1; primary studies, n effect) studies, n = 0 effect)
= 10 (including one pooled analysis
consisting of n = 13 European
cohort studies)
Low birth weight Maternal fatty fish VKM, 2022 Section 4.25: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
intake in pregnancy reviews, n = 1; primary studies, n studies, n = 0
= 4 (including one pooled analysis
consisting of n = 13 European
cohort studies)
Low birth weight Maternal lean fish VKM, 2022 Section 4.25: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
intake in pregnancy reviews, n = 1; primary studies, n studies, n = 0
= 3 (including one pooled analysis
consisting of n = 12 European
cohort studies)
High birth weight Maternal total, fatty VKM, 2022 Section 4.25: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
and lean fish intake reviews, n = 0; primary studies, n (increased risk) studies, n = 0 (increased risk)
in pregnancy = 1 (including one pooled analysis
consisting of n = 13 European
cohort studies)
Birth length Maternal total, fatty VKM, 2022 Section 4.27: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
and lean fish intake reviews, n = 0; primary studies, studies, n = 0
in pregnancy n= 6
Head circumference Maternal total, fatty VKM, 2022 Section 4.27: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
and lean fish intake reviews, n = 0; primary studies, studies, n = 0
in pregnancy n= 6
Bone health
Hip fracture Total fish intake VKM, 2022 Section 4.21: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 1; primary studies, (protective) studies, n = 1 (protective)
n= 5
Cancer
Liver cancer Total fish intake WCRF, 2018 Limited, suggestive Systematic reviews, n = 1; primary Limited, suggestive
(protective) (from studies, n = 0 (protective)
WCRF, 2018)
Colorectal cancer Total fish intake WCRF, 2018 Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
(protective) (from studies, n = 3 (protective)
WCRF, 2018)
Nasopharyngeal Cantonese-style WCRF, 2018 Not included in VKM, Systematic reviews, n = 0; primary Strong evidence
cancer salted fish 2022 studies, n = 0 (increased risk)

414
APPENDICES

TABLE A3.36 OVERVIEW OF THE LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE” FROM THE 2022 VKM REPORT AND
THE SYSTEMATIC LITERATURE SEARCH FOR ALL THE HEALTH OUTCOMES IN “EVIDENCE OF HEALTH BENEFITS FROM
FISH CONSUMPTION” (cont.)
Health outcome Fish consumption Literature included in 2022 VKM Conclusion “weight of Literature included in the Final conclusion
report in the “weight of evidence” evidence” 2022 VKM systematic literature search in “weight of evidence”
conclusion report the final “weight of evidence” in background report
(number of systematic reviews and conclusion (including VKM, 2022
primary studies in VKM, 2022) (number of systematic reviews and and updated literature
primary studies) search)
Pancreatic cancer Total fish intake Not included in VKM, 2022 Not included in VKM, Systematic reviews, n = 1; primary Limited, no conclusion
2022 studies, n = 0
Breast cancer Total fish intake Not included in VKM, 2022 Not included in VKM, Systematic reviews, n = 1; primary Limited, no conclusion
2022 studies, n = 0
CVD
Total cardiovascular Total fish intake VKM, 2022 Section 4.2: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
disease reviews, n = 0; primary studies, (protective effect) studies, n = 2 (protective effect)
n=8
Total cardiovascular Fatty fish intake VKM, 2022 Section 4.2: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
disease reviews, n = 0; primary studies, studies, n = 0
n=3
Total cardiovascular Lean fish intake VKM, 2022 Section 4.2: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
disease reviews, n = 0; primary studies, studies, n = 0
n=2
CHD Total fish intake VKM, 2022 Section 4.3: systematic Probable (protective Systematic reviews, n = 1; primary Probable (protective
reviews, n = 3; Umbrella reviews, n effect) studies, n = 5 effect)
= 2; primary studies, n = 9
CHD Fatty fish intake VKM, 2022 Section 4.3: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 0; primary studies, (protective effect) studies, n = 0 (protective effect)
n=4
CHD Lean fish intake VKM, 2022 Section 4.3: systematic Limited, suggestive (no Systematic reviews, n = 0; primary Limited, suggestive (no
reviews, n = 0; primary studies, effect) studies, n = 0 effect)
n=4
CHD Shellfish Not included in VKM, 2022 Not included in VKM, Systematic reviews, n = 0; primary Limited, no conclusion
2022 studies, n = 1
Myocardial infarction Total fish intake VKM, 2022 Section 4.4: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 2; Umbrella review, n = (protective effect) studies, n = 2 (protective effect)
1; primary studies, n = 8
Myocardial infarction Fatty fish intake VKM, 2022 Section 4.4: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 0; primary studies, (protective effect) studies, n = 1 (protective effect)
n=4
Myocardial infarction Lean fish intake VKM, 2022 Section 4.4: systematic Limited, suggestive (no Systematic reviews, n = 0; primary Limited, suggestive (no
reviews, n = 0; primary studies, effect) studies, n = 1 effect)
n =4
Total stroke Total fish intake VKM, 2022 Section 4.5: systematic Probably (protective Systematic reviews, n = 1; primary Probably (protective
reviews, n = 5; umbrella reviews, n effect) studies, n = 2 effect)
= 4; primary studies, n = 14
Total stroke Fatty fish intake VKM, 2022 Section 4.5: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 2; primary studies, (protective effect) studies, n = 0 (protective effect)
n=7
Total stroke Lean fish intake VKM, 2022 Section 4.5: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 2; primary studies, (protective effect) studies, n = 0 (protective effect)
n=7
Total stroke Shellfish VKM, 2022 Section 4.5: systematic Not included in VKM, Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 1; primary studies, 2022 studies, n = 0
n=0
Ischemic stroke Total fish intake VKM, 2022 Section 4.5: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 3; primary studies, (protective effect) studies, n = 2 (protective effect)
n=8
Haemorrhagic stroke Total fish intake VKM, 2022 Section 4.5: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 3; primary studies, (protective effect) studies, n = 1 (protective effect)
n=8
Atrial fibrillation Total fish intake VKM, 2022 Section 4.6: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 1; primary studies, (adverse effect) studies = 1 (adverse effect)
n=5

415
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.36 OVERVIEW OF THE LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE” FROM THE 2022 VKM REPORT AND
THE SYSTEMATIC LITERATURE SEARCH FOR ALL THE HEALTH OUTCOMES IN “EVIDENCE OF HEALTH BENEFITS FROM
FISH CONSUMPTION” (cont.)
Health outcome Fish consumption Literature included in 2022 VKM Conclusion “weight of Literature included in the Final conclusion
report in the “weight of evidence” evidence” 2022 VKM systematic literature search in “weight of evidence”
conclusion report the final “weight of evidence” in background report
(number of systematic reviews and conclusion (including VKM, 2022
primary studies in VKM, 2022) (number of systematic reviews and and updated literature
primary studies) search)
Atrial fibrillation Fatty fish intake VKM, 2022 Section 4.6: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 1; primary studies, studies, n = 0
n=4
Atrial fibrillation Lean fish intake VKM, 2022 Section 4.6: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 1; primary studies, (protective effect) studies, n = 0 (protective effect)
n=3
Heart failure Total fish intake VKM, 2022 Section 4.6: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
reviews, n = 1; primary studies, (protective effect) studies, n = 1 (protective effect)
n=4
Venous Total fish intake VKM, 2022 Section 4.6: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
thromboembolism reviews, n = 0; primary studies, studies, n = 0
n=3
Peripheral arterial Total fish intake Not included in VKM, 2022 Not included in VKM, Systematic reviews, n = 0; primary Limited, no conclusion
disease 2022 studies, n = 1
Peripheral arterial Fatty fish intake Not included in VKM, 2022 Not included in VKM, Systematic reviews, n = 0; primary Limited, no conclusion
disease 2022 studies, n = 1
Peripheral arterial Lean fish intake Not included in VKM, 2022 Not included in VKM, Systematic reviews, n = 0; primary Limited, no conclusion
disease 2022 studies, n = 1
Type 2 diabetes
Type 2 diabetes Total fish intake VKM, 2022 Section 4.15: systematic Limited, no conclusion Systematic reviews, n = 6; primary Limited, no conclusion
reviews, n = 3; umbrella review, n = studies, n = 1
1; primary studies, n = 16
Type 2 diabetes Fatty fish intake VKM, 2022 Section 4.15: systematic Limited, no conclusion Systematic reviews, n = 2; primary Limited, no conclusion
reviews, n = 2; primary studies, studies, n = 0
n=7
Type 2 diabetes Lean fish intake VKM, 2022 Section 4.15: systematic Limited, suggestive (no Systematic reviews, n = 2; primary Limited, suggestive (no
reviews, n = 2; primary studies, association) studies, n = 0 association)
n=7
Mortality
Alzheimer’s disease Total fish intake VKM, 2022 Section 4.7: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
mortality reviews, n = 0; primary studies, studies, n = 0
n=2
CVD mortality Total fish intake VKM, 2022 Section 4.7: systematic Probable (protective) Systematic reviews, n = 0; primary Probable (protective)
reviews, n = 2; primary studies, n studies, n = 4
= 18
Total heart disease Total fish intake VKM, 2022 Section 4.7: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
mortality reviews, n = 2; primary studies, studies, n = 0
n=2
CHD mortality Total fish intake VKM, 2022 Section 4.7: systematic Probable (protective) Systematic reviews, n = 1; primary Probable (protective)
reviews, n = 2; primary studies, studies, n = 3
n = 18
CHD mortality Total fatty fish intake VKM, 2022 Section 4.7: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 0; primary studies, studies, n = 0
n=2
CHD mortality Total lean fish intake VKM, 2022 Section 4.7: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 0; primary studies, studies, n = 0
n=2
Myocardial infarction Total fish intake VKM, 2022 Section 4.7: systematic Probable (protective) Systematic reviews, n = 0; primary Probable (protective)
(MI) mortality reviews, n = 0; primary studies, studies, n = 0
n=5
Stroke mortality Total fish intake VKM, 2022 Section 4.7: systematic Probable (protective) Systematic reviews, n = 0; primary Probable (protective)
reviews, n = 0; primary studies, studies, n = 0
n = 12
Ischemic stroke Total fish intake VKM, 2022 Section 4.7: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
mortality reviews, n = 0; primary studies, (protective) studies, n = 0 (protective)
n=6

416
APPENDICES

TABLE A3.36 OVERVIEW OF THE LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE” FROM THE 2022 VKM REPORT AND
THE SYSTEMATIC LITERATURE SEARCH FOR ALL THE HEALTH OUTCOMES IN “EVIDENCE OF HEALTH BENEFITS FROM
FISH CONSUMPTION” (cont.)
Health outcome Fish consumption Literature included in 2022 VKM Conclusion “weight of Literature included in the Final conclusion
report in the “weight of evidence” evidence” 2022 VKM systematic literature search in “weight of evidence”
conclusion report the final “weight of evidence” in background report
(number of systematic reviews and conclusion (including VKM, 2022
primary studies in VKM, 2022) (number of systematic reviews and and updated literature
primary studies) search)
Haemorrhagic stroke Total fish intake VKM, 2022 Section 4.7: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
mortality reviews, n = 0; primary studies, (protective) studies, n = 0 (protective)
n=6
Type 2 diabetes Total fish intake VKM, 2022 Section 4.7: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
mortality reviews, n = 0; primary studies, studies, n = 0
n=4
Colorectal cancer Total fish intake Not included in VKM report Not included in VKM Systematic reviews, n = 1; primary Limited, no conclusion
mortality report studies, n = 0
Prostate cancer- Total fish intake Not included in VKM report Not included in VKM Systematic reviews, n = 1; primary Limited, no conclusion
specific mortality report studies, n = 0
All-cause mortality Total fish intake VKM, 2022 Section 4.8: systematic Probable (protective) Systematic reviews, n = 0; primary Probable (protective)
reviews, n = 5; primary studies, studies, n = 2
n = 23
All-cause mortality Fatty fish intake VKM, 2022 Section 4.8: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 0; primary studies, studies, n = 0
n=1
All-cause mortality Fatty lean fish intake VKM, 2022 Section 4.8: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 0; primary studies, studies, n = 0
n=1
Neurodevelopment and neurological diseases
Neurodevelopment in Maternal total fish VKM, 2022 Section 4.9: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
children intake in pregnancy reviews, n = 1; primary studies, (protective) studies, n = 1 (protective)
n = 22
Neurodevelopment in Maternal fatty fish VKM, 2022 Section 4.9: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
children intake in pregnancy reviews, n = 0; primary studies, studies, n = 0
n=4
Neurodevelopment in Maternal lean fish VKM, 2022 Section 4.9: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
children intake in pregnancy reviews, n = 0; primary studies, studies, n = 0
n=5
Neurodevelopment in Child total fish intake VKM, 2022 Section 4.9: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
children reviews, n = 1; primary studies, (protective) studies, n = 1 (protective)
n=4
Neurodevelopment in Child fatty fish intake VKM, 2022 Section 4.9: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
children reviews, n = 1; primary studies, (protective) studies, n = 1 (protective)
n=6
Neurodevelopment in Child lean fish intake VKM, 2022 Section 4.9: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
children reviews, n = 1; primary studies, studies, n = 0
n=0
Neurocognitive and Total fish intake VKM, 2022 Section 4.13: systematic Probable (protective Systematic reviews, n = 0; primary Probable (protective
psychiatric endpoints reviews, n = 4; umbrella review, n = effect) studies, n = 0 effect)
in adults (dementia, 1; primary studies, n = 21
Alzheimer’s disease,
and cognitive decline)
Neurocognitive and Fatty fish intake VKM, 2022 Section 4.13: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
psychiatric endpoints reviews, n = 0; primary studies, studies, n = 0
in adults (dementia, n=2
Alzheimer’s disease,
and cognitive decline)
Neurocognitive and Lean fish intake VKM, 2022 Section 4.13: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
psychiatric endpoints reviews, n = 0; primary studies, studies, n = 0
in adults (dementia, n=1
Alzheimer’s disease,
and cognitive decline)
Depression and post- Total fish intake VKM, 2022 Section 4.14: systematic Limited, suggestive Systematic reviews, n = 0; primary Limited, suggestive
partum depression reviews, n = 4; umbrella reviews, n studies, n = 0
= 2; primary studies, n = 13

417
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A3.36 OVERVIEW OF THE LITERATURE INCLUDED IN THE FINAL “WEIGHT OF EVIDENCE” FROM THE 2022 VKM REPORT AND
THE SYSTEMATIC LITERATURE SEARCH FOR ALL THE HEALTH OUTCOMES IN “EVIDENCE OF HEALTH BENEFITS FROM
FISH CONSUMPTION” (cont.)
Health outcome Fish consumption Literature included in 2022 VKM Conclusion “weight of Literature included in the Final conclusion
report in the “weight of evidence” evidence” 2022 VKM systematic literature search in “weight of evidence”
conclusion report the final “weight of evidence” in background report
(number of systematic reviews and conclusion (including VKM, 2022
primary studies in VKM, 2022) (number of systematic reviews and and updated literature
primary studies) search)
Depression and post- Fatty fish intake VKM, 2022 Section 4.14: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
partum depression reviews, n = 0; primary studies, studies, n = 0
n=1
Depression and post- Lean fish intake VKM, 2022 Section 4.14: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
partum depression reviews, n = 0; primary studies, studies, n = 0
n=0
Obesity in adults
Obesity in adults Total fish intake VKM, 2022 Section 4.18: systematic Limited, no conclusion Systematic reviews, n = 0; primary Limited, no conclusion
reviews, n = 1; primary studies, studies, n = 3
n=3

418
APPENDICES

APPENDIX 4
TOXIC EFFECTS OF
DIOXINS AND dl-PCBs

LITERATURE SEARCH STRATEGY

TABLE A4.1 LITERATURE SEARCH STRATEGY FOR THE SYSTEMATIC REVIEW “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”
IN PUBMED
Database: PubMed
Date of literature search: 16 December 2021
Literature search string:
(((((((((("journal article"[Publication Type]) OR "review"[Publication Type]) OR "scientific integrity review"[Publication Type]) OR "meta analysis"[Publication Type]) OR
research[Publication Type]) OR review, systematic[MeSH Terms]) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ) AND Humans[Mesh])) AND ((((((((((((((((tetrachlorodibenzodiox
in[MeSH Terms]) OR 2,3,7,8 tetrachlorodibenzo p dioxin[MeSH Terms]) OR tcdd[MeSH Terms]) OR dioxins[MeSH Terms]) OR polychlorinated biphenyls[MeSH Terms]) OR pcbs[MeSH
Terms]) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ))) OR ((((((((((Tetrachlorodibenzodioxin[Title/Abstract]) OR "2,3,7,8 tetrachlorodibenzo p dioxin"[Title/Abstract]) OR
TCDD*[Title/Abstract]) OR PCDD*[Title/Abstract]) OR PCDF*[Title/Abstract]) OR “Polychlorinated dibenzofuran”[Title/Abstract]) OR dioxin*[Title/Abstract]) OR "polychlorinated
biphenyl*"[Title/Abstract]) OR PCB[Title/Abstract]) OR PCBs[Title/Abstract]) OR (TEQ[Title/Abstract] OR "total equivalen*"[Title/Abstract])) OR coplanar[Title/Abstract] OR
“Polychlorinated dibenzodioxin”[Title/Abstract])) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ))) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ))) AND (english[Language]
AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ) AND Humans[Mesh])) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ) AND Humans[Mesh])))) AND ((((((((cohort study OR
cohort studies[MeSH Terms])) OR (case control study OR case control studies[MeSH Terms])) OR adverse effects[MeSH Terms]) OR (cross sectional study OR cross sectional
studies[MeSH Terms])) OR case reports[MeSH Terms] OR blood [MeSH Terms] OR hormones [MeSH Terms] OR serum [MeSH Terms] OR urine [MeSH Terms] OR semen[MeSH Terms]))
OR (epidemiolog*[Title/Abstract] OR "cohort study"[Title/Abstract] OR "cohort studies"[Title/Abstract] OR "case control study"[Title/Abstract] OR "case control studies"[Title/Abstract]
OR "adverse effect"[Title/Abstract] OR "adverse effects"[Title/Abstract] OR "observational study"[Title/Abstract] OR "observational studies"[Title/Abstract] OR "case series"[Title/
Abstract]OR "cross sectional study"[Title/Abstract] OR "cross sectional studies"[Title/Abstract] OR "case report"[Title/Abstract]OR "case reports"[Title/Abstract] OR urine[Title/
Abstract] OR serum[Title/Abstract] OR plasma [Title/Abstract] OR haema*[Title/Abstract] OR hema [Title/Abstract] OR blood [Title/Abstract] OR sperm [Title/Abstract] OR semen[Title/
Abstract] OR hormone*[Title/Abstract] OR reproduct*[Title/Abstract] ))) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ) AND Humans[Mesh])) AND (english[Language] AND (
"2016/07/05"[PDat] : "2021/12/16"[PDat] ) AND Humans[Mesh])) AND ( "2016/07/05"[PDat] : "2021/12/16"[PDat] ) AND Humans[Mesh])))
Total hits: 1 193

419
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A4.2 LITERATURE SEARCH STRATEGY FOR THE SYSTEMATIC REVIEW “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”
IN PUBMED
Database: Web of Science
Date of literature search: 16 December 2021
# Search group Literature search string Search history and limitations in search Total hits
#1 Exposure (Tetrachlorodibenzodioxin OR (Tetrachlorodibenzodioxin OR "2,3,7,8-Tetrachlorodibenzo-p-dioxin" OR TCDD OR dioxin* OR "polychlorinated 19 995
"2,3,7,8-Tetrachlorodibenzo- biphenyl*" OR PCB$ OR TEQ OR "total equivalen*" OR coplanar OR PCDD$ OR PCDF$ OR "Polychlorinated
p-dioxin" OR TCDD OR dioxin* dibenzofuran" OR Polychlorinated dibenzodioxins) (Topic) and Proceedings Papers or Meeting Abstracts or
OR "polychlorinated biphenyl*" Editorial Materials or Letters or Book Chapters or News Items or Poetry or Retractions or Biographical-Items
OR PCB$ OR TEQ OR "total or Book Reviews or Reprints or Retracted Publications or Withdrawn Publication (Exclude – Document Types)
equivalen*" OR coplanar and Chinese or Korean or Spanish or Portuguese or German or Polish or French or Japanese or Russian or
OR PCDD$ OR PCDF$ OR Italian or Turkish or Malay or Czech or Ukrainian or Croatian or Indonesian or Slovak (Exclude – Languages)
"Polychlorinated dibenzofuran" Timespan: 2016-07-05 to 2021-12-13 (publication date)
OR Polychlorinated
dibenzodioxins) (Topic)
#2 Population (human OR women OR men (human OR women OR men OR child*) (Topic) and Meeting Abstracts or Editorial Materials or Book Reviews 1 894 550
OR child*) or Letters or Proceedings Papers or Book Chapters or News Items or Retractions or Biographical-Items or
Poetry or Retracted Publications or Art Exhibit Reviews or Film Reviews or Reprints or Fiction, Creative Prose
or Record Reviews or Music Performance Review or Theater Reviews or TV Review, Radio Review Videos or
Dance Performance Reviews or Item Withdrawal or Bibliographies or Withdrawn Publication or Hardware
Reviews or Excerpts or Software Reviews or Database Reviews (Exclude – Document Types) and Spanish or
Portuguese or Russian or German or French or Turkish or Chinese or Italian or Polish or Korean or Czech or
Croatian or Hungarian or Ukrainian or Japanese or Slovenian or Slovak or Unspecified or Malay or Indonesian
or Norwegian or Greek or Bulgarian or Afrikaans or Catalan or Dutch or Lithuanian or Arabic or Icelandic or
Estonian or Serbian or Swedish or Persian or Danish or Galician or Welsh or Basque or Eskimo or Georgian or
Latin or Samoan or Zulu or Hebrew or Urdu (Exclude – Languages)
Timespan: 2016-07-05 to 2021-12-13 (publication date)
#3 #1 AND #2 Timespan: 2016-07-05 to 2021-12-13 (Publication Date) 3 692
#4 Outcome (epidemiolog* OR "cohort (epidemiolog* OR "cohort stud*" OR "case control stud* " OR "adverse effect*" OR "observational stud*" 1 739 115
stud*" OR "case control stud* OR "case serie*" OR "case report*" OR "cross sectional stud*" OR urine OR serum OR plasma OR haema*
" OR "adverse effect*" OR OR hema* OR blood OR sperm OR semen OR hormone* OR reproduct*) (Topic) and Meeting Abstracts
"observational stud*" OR "case or Editorial Materials or Letters or Proceedings Papers or Book Chapters or Book Reviews or News Items
serie*" OR "case report*" OR or Retractions or Retracted Publications or Biographical-Items or Reprints or Poetry or Film Reviews or
"cross sectional stud*" OR Hardware Reviews or Withdrawn Publication or Art Exhibit Reviews or Bibliographies or Music Performance
urine OR serum OR plasma Review or Fiction, Creative Prose or Item Withdrawal or Record Reviews or Excerpts or TV Review, Radio
OR haema* OR hema* OR Review Videos or Meeting Summary or Software Reviews or Theater Reviews (Exclude – Document Types)
blood OR sperm OR semen OR and Spanish or German or French or Russian or Portuguese or Chinese or Turkish or Polish or Korean or
hormone* OR reproduct*) Italian or Hungarian or Czech or Japanese or Ukrainian or Indonesian or Greek or Slovak or Croatian or
Icelandic or Norwegian or Arabic or Malay or Dutch or Persian or Slovenian or Unspecified or Catalan or
Danish or Welsh or Esperanto or Afrikaans or Bulgarian or Estonian or Serbian or Galician or Lithuanian or
Eskimo or Swedish or Serbo Croatian or Basque or Hebrew or Samoan (Exclude – Languages)
Timespan: 2016-07-05 to 2021-12-13 (publication date)
#5 #3 AND #4 Timespan: 2016-07-05 to 2021-12-13 (publication date) 1 577

420
APPENDICES

LIST OF EXCLUDED RECORDS DURING FULL-TEXT SCREENING

TABLE A4.3 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECTS OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 32) Reason for exclusion
Aghaei, M., Janjani, H., Yousefian, F., Jamal, A., & Yunesian, M. 2019. Association None of the included studies published after 2016 were relevant for our purpose.
between ambient gaseous and particulate air pollutants and attention deficit
hyperactivity disorder (ADHD) in children; a systematic review. Environmental Research,
173, 135-156.
Cano-Sancho, G., Ploteau, S., Matta, K., Adoamnei, E., Louis, G. B., Mendiola, J., Only one of the included 17 studies in the review was relevant for our purpose (and this
Antignac, J.P. et al. 2019. Human epidemiological evidence about the associations was included in the primary studies).
between exposure to organochlorine chemicals and endometriosis: Systematic review and
meta-analysis. Environment International, 123, 209-223.
Edwards, D., Voronina, A., Attwood, K., & Grand’Maison, A. 2021. Association between Studies that are relevant for our purpose were published before 2009.
occupational exposures and sarcoma incidence and mortality: systematic review and
meta-analysis. Systematic Reviews, 10, 1-19.
Ennour-Idrissi, K., Ayotte, P., & Diorio, C. 2019. Persistent organic pollutants and breast Only one of the included studies in the review was relevant for our purpose (and this was
cancer: a systematic review and critical appraisal of the literature. Cancers, 11(8), 1063. included in the primary studies).
Fernández-Martínez, N. F., Ching-Lopez, A., de Labry Lima, A. O., Salamanca-Fernández, None of the included studies were relevant for our purpose.
E., Pérez-Gómez, B., Jiménez-Moleón, J. J., Rodríguez-Barranco, M. et al. 2020.
Relationship between exposure to mixtures of persistent, bioaccumulative, and toxic
chemicals and cancer risk: A systematic review. Environmental Research, 188, 109787.
Fiolet, T., Mahamat-Saleh, Y., Frenoy, P., Kvaskoff, M., & Mancini, F.R. 2021. Background None of the included studies were relevant for our purpose (mono-ortho only).
exposure to polychlorinated biphenyls and all-cause, cancer-specific, and cardiovascular-
specific mortality: A systematic review and meta-analysis. Environment International,
154, 106663.
Kadawathagedara, M, de Lauzon-Guillain, B. & Botton, J. 2018. Environmental Not a systematic review.
contaminants and child's growth. J Dev Orig Health Dis, 9:632-641.
Kahn, L. G., Harley, K. G., Siegel, E. L., Zhu, Y., Factor-Litvak, P., Porucznik, C. A., Program None of the included studies were relevant for our purpose.
Collaborators for Environmental Influences on Child Health Outcomes Program. et al.
2021. Persistent organic pollutants and couple fecundability: a systematic review.
Human Reproduction Update, 27(2), 339-366.
Lefebvre, T., Fréour, T., Ploteau, S., Le Bizec, B., Antignac, J. P., & Cano-Sancho, G. 2021. Only one of the included studies in the review was relevant for our purpose (and this is
Associations between human internal chemical exposure to Persistent Organic Pollutants included in the primary studies).
(POPs) and In Vitro Fertilization (IVF) outcomes: Systematic review and evidence map of
human epidemiological evidence. Reproductive Toxicology, 105, 184-197.
Mallozzi, M., Leone, C., Manurita, F., Bellati, F., & Caserta, D. 2017. Endocrine disrupting All studies included in the review were published before 2016.
chemicals and endometrial cancer: an overview of recent laboratory evidence and
epidemiological studies. International Journal of Environmental Research and Public
Health, 14(3), 334.
Matta, K., Koual, M., Ploteau, S., Coumoul, X., Audouze, K., Le Bizec, B., ... & Cano- Only included animal studies.
Sancho, G. 2021. Associations between exposure to organochlorine chemicals and
endometriosis: a systematic review of experimental studies and integration of
epidemiological evidence. Environmental Health Perspectives, 129(7), 076003.
Meltzer, G. Y., Watkins, B. X., Vieira, D., Zelikoff, J. T., & Boden-Albala, B. 2020. A Only one of the included studies in the review was relevant for our purpose (and this is
systematic review of environmental health outcomes in selected American Indian and included in the primary studies).
Alaska Native populations. Journal of racial and ethnic health disparities, 7, 698-739.
Mouly, T. A., & Toms, L. M. L. 2016. Breast cancer and persistent organic pollutants Includes only studies published between 2006 and 2015.
(excluding DDT): a systematic literature review. Environmental Science and Pollution
Research, 23, 22385-22407.
Nelson, W., Liu, D. Y., Yang, Y., Zhong, Z. H., Wang, Y. X., & Ding, Y. B. 2020. In utero Studies that were relevant for our purpose were published before 2016.
exposure to persistent and nonpersistent endocrine-disrupting chemicals and anogenital
distance. A systematic review of epidemiological studies. Biology of Reproduction,
102(2), 276-291.
Peinado, F. M., Artacho-Cordón, F., Barrios-Rodríguez, R., & Arrebola, J. P. 2020. Influence Not relevant papers according to criteria.
of polychlorinated biphenyls and organochlorine pesticides on the inflammatory milieu.
A systematic review of in vitro, in vivo and epidemiological studies. Environmental
Research, 186, 109561.
Poole, C.J.M. & Basu, S. 2017. Systematic Review: Occupational illness in the waste and Includes only studies published between 1997 and 2015.
recycling sector. Occup Med (Lond), 67:626-636.
Raffetti, E., Donat-Vargas, C., Mentasti, S., Chinotti, A., & Donato, F. 2020. Association Only four of the included studies were relevant for our purpose (and these were included
between exposure to polychlorinated biphenyls and risk of hypertension: A systematic in the primary studies).
review and meta-analysis. Chemosphere, 255, 126984.
Ribeiro, C. M., Beserra, B. T. S., Silva, N. G., Lima, C. L., Rocha, P. R. S., Coelho, M. S., Only two of the included studies were relevant for our purpose (and these are included in
Amato, A. A. et al. 2020. Exposure to endocrine-disrupting chemicals and anthropometric the primary studies).
measures of obesity: a systematic review and meta-analysis. BMJ Open, 10(6), e033509.

421
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

LIST OF EXCLUDED RECORDS DURING FULL-TEXT SCREENING

TABLE A4.3 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECTS OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Study (n = 32) Reason for exclusion
Rivollier, F. Krebs, M.O. & Kebir, O. 2019. Perinatal Exposure to Environmental Endocrine Only two of the included 47 studies in the review were relevant for our purpose (and these
Disruptors in the Emergence of Neurodevelopmental Psychiatric Diseases: A Systematic were included in the primary studies).
Review. International Journal of Environmental Research and Public Health, 16:19.
Rocha, P. R. S., Oliveira, V. D., Vasques, C. I., Dos Reis, P.E.D., & Amato, A.A. 2021. Only one of the included studies in the review was relevant for our purpose (and this is
Exposure to endocrine disruptors and risk of breast cancer: a systematic review. Critical included in the primary studies).
Reviews In Oncology/Hematology, 161, 103330.
Singh, N. Ogunseitan, O.A. & Tang, Y.Y. 2021. Systematic review of pregnancy and Only two of the included 27 studies in the review were relevant for our purpose (and these
neonatal health outcomes associated with exposure to e-waste disposal. Critical Reviews are included in the primary studies).
in Environmental Science and Technology, 51:2424-2448.
Sirohi, D. Al Ramadhani, R. & Knibbs, L.D. 2021. Environmental exposures to endocrine Only one of the included 29 studies in the review was relevant for our purpose (and this
disrupting chemicals (EDCs) and their role in endometriosis: a systematic literature is included in the primary studies).
review. Rev Environ Health, 36:101-115.
Tsai, M. S., Chen, M. H., Lin, C. C., Liu, C. Y., & Chen, P.C. 2019. Children's environmental No relevant studies included, as the study did not cover dioxins and dl-PCBs.
health based on birth cohort studies of Asia (2)–air pollution, pesticides, and heavy
metals. Environmental research, 179, 108754.
Tsai, M. S., Chen, M. H., Lin, C. C., Ng, S., Hsieh, C. J., Liu, C. Y., Chen, P. C. et al. 2017. Includes only studies published before 2015.
Children's environmental health based on birth cohort studies of Asia. Science of the
Total Environment, 609, 396-409.
Wan, M.L.Y. Co, V.A. & El-Nezami, H. 2021. Endocrine disrupting chemicals and breast None of the included studies in the review was relevant for our purpose.
cancer: a systematic review of epidemiological studies. Critical Reviews in Food Science
and Nutrition, 28.
Wang, Y., Hollis‐Hansen, K., Ren, X., Qiu, Y., & Qu, W. J. O. R. 2016. Do environmental Only included studies published before 2016.
pollutants increase obesity risk in humans?. Obesity reviews, 17(12), 1179-1197.
Wikoff, D. S., Urban, J. D., Ring, C., Britt, J., Fitch, S., Budinsky, R., & Haws, L. C. Mainly focuses on animal studies.
2021. Development of a range of plausible noncancer toxicity values for 2, 3, 7,
8-tetrachlorodibenzo-p-dioxin based on effects on sperm count: Application of systematic
review methods and quantitative integration of dose response using meta-regression.
Toxicological Sciences, 179(2), 162-182.
Xu, J., Ye, Y., Huang, F., Chen, H., Wu, H., Huang, J., Wu, Y. et al. 2016. Association All studies were published before 2015.
between dioxin and cancer incidence and mortality: a meta-analysis. Scientific Reports,
6(1), 1-17.
Zani, C., Ceretti, E., Covolo, L., & Donato, F. 2017. Do polychlorinated biphenyls cause All studies were published before 2015.
cancer? A systematic review and meta-analysis of epidemiological studies on risk of
cutaneous melanoma and non-Hodgkin lymphoma. Chemosphere, 183, 97-106.
Zeinomar, N., Oskar, S., Kehm, R. D., Sahebzeda, S., & Terry, M. B. 2020. Environmental Only two of the included studies were relevant for our purpose (and these are included in
exposures and breast cancer risk in the context of underlying susceptibility: A systematic the primary studies).
review of the epidemiological literature. Environmental research, 187, 109346.
Zeng, Z., Ngai, S., Wang, Q., Liang, W., & Huo, X. 2021. Early-life exposure to widespread No relevant studies were included.
environmental toxicants and children's health risks: A focus on the post-vaccination
antibody potency or immunoglobulin levels. Science of The Total Environment, 781,
146714.
Zou, H., Lin, Y., Yang, L., Ou, C., Geng, F., Wang, Y., Sun, Y. et al. 2019. Neonatal weight All studies were published before 2015.
and prenatal exposure to polychlorinated biphenyls: a meta-analysis. Asian Pacific
Journal of Cancer Prevention: APJCP, 20(11), 3251.

422
APPENDICES

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 283)
Abellan, A. et al. 2019. Prenatal exposure to organochlorine compounds and lung function during childhood. Environ Int, 131:105049.
Alampi, J.D. et al. 2021. Association Between Gestational Exposure to Toxicants and Autistic Behaviors Using Bayesian Quantile Regression. American Journal of Epidemiology,
190(9):1803-1813.
Aminov, Z. et al. 2016. Diabetes Prevalence in Relation to Serum Concentrations of Polychlorinated Biphenyl (PCB) Congener Groups and Three Chlorinated Pesticides in a Native
American Population. Environ Health Perspect, 124(9):1376-83.
Arias-Ortiz, N.E. G. Icaza-Noguera, and P. Ruiz-Rudolph, 2018. Thyroid cancer incidence in women and proximity to industrial air pollution sources: A spatial analysis in a middle size
city in Colombia. Atmospheric Pollution Research, 9(3):464-475.
Arisi, M. et al. 2021. Neoplastic and inflammatory skin disorders and serum levels of polychlorinated biphenyls in a population living in a highly polluted area. Eur J Dermatol,
31(1):41-47.
Attfield, K.R. et al. 2019. Longitudinal study of age of menarche in association with childhood concentrations of persistent organic pollutants. Environmental Research, 176:8.
Bach, M.A. et al. 2020. Association of polychlorinated biphenyls and organochlorine pesticides with autism spectrum disorder in Jamaican children. Research in Autism Spectrum
Disorders, 76:14.
Bae, J. et al. 2018. Maternal and paternal serum concentrations of persistent organic pollutants and the secondary sex ratio: A population-based preconception cohort study. Environ
Res, 161:9-16.
Barrios-Rodriguez, R. et al. 2018. Associations of accumulated selected persistent organic pollutants in adipose tissue with insulin sensitivity and risk of incident type-2 diabetes.
Environment International, 155:10.
Bassig, B.A. et al. 2019. Pre-diagnostic serum concentrations of organochlorines and risk of acute myeloid leukemia: A nested case-control study in the Norwegian Janus Serum Bank
Cohort. Environ Int, 125:229-235.
Benson, K. et al. 2018. Polychlorinated biphenyls, indicators of thyroid function and thyroid autoantibodies in the Anniston Community Health Survey I (ACHS-I). Chemosphere,
195:156-165.
Berg, V. et al. 2017. Persistent Organic Pollutants and the Association with Maternal and Infant Thyroid Homeostasis: A Multipollutant Assessment. Environmental Health
Perspectives, 125(1):127-133.
Berg, V. et al. 2021. Pre- and post-diagnostic blood profiles of chlorinated persistent organic pollutants and metabolic markers in type 2 diabetes mellitus cases and controls; a pilot
study. Environ Res, 195:110846.
Bernardo, B.A. et al. 2019. Assessing the Relation between Plasma PCB Concentrations and Elevated Autistic Behaviours using Bayesian Predictive Odds Ratios. Int J Environ Res
Public Health, 16(3).
Beszterda, M. & Frański, R. 2018. Endocrine disruptor compounds in environment: As a danger for children health. Pediatr Endocrinol Diabetes Metab, 2018. 24(2):88-95.
Bloom, M.S. et al. 2017. Persistent organic pollutants (POPs) in human follicular fluid and in vitro fertilization outcomes, a pilot study. Reprod Toxicol, 2017. 67:165-173.
Bornstein, S.R. et al. 2020. Is There a Role for Environmental and Metabolic Factors Predisposing to Severe COVID-19? Horm Metab Res, 2020. 52(7):540-546.
Brown, A.S. et al. 2018. Association of Maternal Insecticide Levels With Autism in Offspring From a National Birth Cohort. American Journal of Psychiatry, 175(11):1094-1101.
Buck Louis, G.M. et al. 2018. Endocrine disruptors and neonatal anthropometry, NICHD Fetal Growth Studies - Singletons. Environ Int, 119:515-526.
Burns, J.S. et al. 2016. Associations of Peripubertal Serum Dioxin and Polychlorinated Biphenyl Concentrations with Pubertal Timing among Russian Boys. Environ Health Perspect,
124(11):1801-1807.
Cabrera-Rodríguez, R. et al. 2019. Association between prenatal exposure to multiple persistent organic pollutants (POPs) and growth indicators in newborns. Environ Res, 171:285-
292.
Callahan, C.L. et al. 2017. Serum polychlorinated biphenyls and leukocyte telomere length in a highly-exposed population: The Anniston Community Health Survey. Environ Int,
108:212-220.
Callan, A.C. et al. 2016. Sex specific influence on the relationship between maternal exposures to persistent chemicals and birth outcomes. Int J Hyg Environ Health, 219(8):734-741.
Cao, J. et al. 2019. Association study between plasma levels of polychlorinated biphenyls and risk of cutaneous malignant melanoma. Environ Int, 126:298-301.
Carrizo, D. et al. 2017. Untargeted metabolomic analysis of human serum samples associated with exposure levels of persistent organic pollutants indicate important perturbations
in Sphingolipids and Glycerophospholipids levels. Chemosphere, 168:731-738.
Caspersen, I.H. et al. 2016. The influence of maternal dietary exposure to dioxins and PCBs during pregnancy on ADHD symptoms and cognitive functions in Norwegian preschool
children. Environment International, 94:649-660.
Catalani, S. et al. 2019. Occupational and environmental exposure to polychlorinated biphenyls and risk of non-Hodgkin lymphoma: a systematic review and meta-analysis of
epidemiology studies. Eur J Cancer Prev, 28(5):441-450.
Chung, M.K. et al. 2019. Exposome-wide association study of semen quality: Systematic discovery of endocrine disrupting chemical biomarkers in fertility require large sample sizes.
Environment International, 125:505-514.
Clair, H.B. et al. 2018. Liver Disease in a Residential Cohort With Elevated Polychlorinated Biphenyl Exposures. Toxicol Sci, 164(1):39-49.
Cordier, S. et al. 2020. Association between exposure to persistent organic pollutants and mercury, and glucose metabolism in two Canadian Indigenous populations. Environ Res,
184:109345.
Criswell, R. et al. 2017. Persistent Environmental Toxicants in Breast Milk and Rapid Infant Growth. Ann Nutr Metab, 70(3):210-216.
Curtis, S.W. et al. 2019. Thyroid hormone levels associate with exposure to polychlorinated biphenyls and polybrominated biphenyls in adults exposed as children. Environ Health,
18(1):75.
Cypel, Y. et al. 2019. Spirometric Pulmonary Restriction in Herbicide-Exposed U.S. Vietnam War Veterans. Int J Environ Res Public Health, 16(17).
Cypel, Y.S. et al. 2016. Herbicide Exposure, Vietnam Service, and Hypertension Risk in Army Chemical Corps Veterans. J Occup Environ Med, 58(11):1127-1136.

423
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Dai, Y. et al. 2020. Early-life exposure to widespread environmental toxicants and maternal-fetal health risk: A focus on metabolomic biomarkers. Sci Total Environ, 739:139626.
Danjou, A.M.N. et al. 2019. Long-term airborne dioxin exposure and breast cancer risk in a case-control study nested within the French E3N prospective cohort. Environ Int, 124:236-
248.
Darvishian, M. et al. 2022. Persistent organic pollutants and risk of cutaneous malignant melanoma among women. Cancer Reports:12.
de Cock, M. et al. 2017. Thyroid-stimulating hormone levels in newborns and early life exposure to endocrine-disrupting chemicals: analysis of three European mother-child cohorts.
Pediatr Res, 82(3):429-437.
de Prado-Bert, P. et al. 2021. The early-life exposome and epigenetic age acceleration in children. Environ Int, 155:106683.
Deziel, N.C. et al. 2021. Exposure to polychlorinated biphenyls and organochlorine pesticides and thyroid cancer in Connecticut women. Environmental Research, 192:9.
Donato, F. et al. 2021. Polychlorinated biphenyls and risk of hepatocellular carcinoma in the population living in a highly polluted area in Italy. Sci Rep, 11(1):3064.
Donat-Vargas, C. et al. 2020. Cardiovascular and cancer mortality in relation to dietary polychlorinated biphenyls and marine polyunsaturated fatty acids: a nutritional-toxicological
aspect of fish consumption. Journal of Internal Medicine, 287(2):197-209.
Donat-Vargas, C. et al. 2016. Dietary exposure to polychlorinated biphenyls and risk of breast, endometrial and ovarian cancer in a prospective cohort. British Journal of Cancer,
115(9):1113-1121.
Donat-Vargas, C. et al. 2017. Dietary polychlorinated biphenyls, long-chain n-3 polyunsaturated fatty acids and incidence of malignant melanoma. European Journal of Cancer,
72:137-143.
Donat-Vargas, C. et al. 2018. Persistent Organochlorine Pollutants in Plasma, Blood Pressure, and Hypertension in a Longitudinal Study. Hypertension, 71(6):1258-1268.
Dufour, P. et al. 2018. Association between organohalogenated pollutants in cord blood and thyroid function in newborns and mothers from Belgian population. Environ Pollut,
238:389-396.
Dusanov, S. et al. 2018. Associations between persistent organic pollutants and metabolic syndrome in morbidly obese individuals. Nutrition Metabolism and Cardiovascular
Diseases, 28(7):735-742.
Dusanov, S. et al. 2020. Effect of fatty fish or nut consumption on concentrations of persistent organic pollutants in overweight or obese men and women: A randomized controlled
clinical trial. Nutr Metab Cardiovasc Dis, 30(3):448-458.
Eguchi, A. et al. 2019. An Altered DNA Methylation Status in the Human Umbilical Cord Is Correlated with Maternal Exposure to Polychlorinated Biphenyls. Int J Environ Res Public
Health, 16(15).
Eguchi, A. et al. 2017. Exploration of potential biomarkers and related biological pathways for PCB exposure in maternal and cord serum: A pilot birth cohort study in Chiba, Japan.
Environ Int, 102:157-164.
Eguchi, A. et al. 2019. The relationship of maternal PCB, toxic, and essential trace element exposure levels with birth weight and head circumference in Chiba, Japan. Environ Sci
Pollut Res Int, 26(15):15677-15684.
Emecen-Huja, P. et al. 2019. Epidemiologic evaluation of Nhanes for environmental factors and periodontal disease. Sci Rep, 9(1):8227.
Eskenazi, B. et al. 2017. In utero and childhood DDT, DDE, PBDE and PCBs exposure and sex hormones in adolescent boys: The CHAMACOS study. International Journal of Hygiene and
Environmental Health, 220(2):364-372.
Eslami, B. et al. 2016. Association between serum concentrations of persistent organic pollutants and gestational diabetes mellitus in primiparous women. Environ Res, 151:706-712.
Eslami, B. et al. 2016. Association of serum concentrations of persistent organic pollutants (POPs) and risk of pre-eclampsia: a case-control study. Journal of Environmental Health
Science and Engineering, 14:8.
Esser, A. et al. 2016. Association between polychlorinated biphenyls and diabetes mellitus in the German HELPcB cohort. Int J Hyg Environ Health, 219(6):557-65.
Etheridge, T. et al. 2019. The Impact of Agent Orange Exposure on Prostate Cancer Outcomes. J Urol, 201(4):742-750.
Forns, J. et al. 2016. Novel application of statistical methods for analysis of multiple toxicants identifies DDT as a risk factor for early child behavioral problems. Environ Res, 151:91-
100.
Forte, I.M. et al. 2020. Blood screening for heavy metals and organic pollutants in cancer patients exposed to toxic waste in southern Italy: A pilot study. Journal of Cellular Physiology,
235(6):5213-5222.
Fracchiolla, N.S. et al. 2016. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) role in hematopoiesis and in hematologic diseases: A critical review. Toxicology, 374:60-68.
Fry, K. & Power, M.C. 2017. Persistent organic pollutants and mortality in the United States, NHANES 1999-2011. Environ Health, 16(1):105.
Fu, X.J. et al. 2020. The association between environmental endocrine disruptors and cardiovascular diseases: A systematic review and meta-analysis. Environmental Research,
187:19.
Gallo, M.V. et al. 2016. Endocrine disrupting chemicals and ovulation: Is there a relationship? Environmental Research, 151:410-418.
Gallo, M.V. et al. 2018. Persistent organic pollutants as predictors of increased FSH:LH ratio in naturally cycling, reproductive age women. Environmental Research, 164:556-564.
Gasull, M. et al. 2018. Blood Concentrations of Persistent Organic Pollutants and Unhealthy Metabolic Phenotypes in Normal-Weight, Overweight, and Obese Individuals. Am J
Epidemiol, 187(3):494-506.
Gaum, P.M. et al. 2021. Adverse health effects of PCBs on fine motor performance - Analysis of a neurophysiological pathway in the HELPcB surveillance program. Neurotoxicology,
84:146-154.
Gaum, P.M. et al. 2020. Cross-Sectional and Longitudinal Effects of PCB Exposure on Human Stress Hormones in the German HELPcB Surveillance Program. International Journal of
Environmental Research and Public Health, 17(13):19.
Gaum, P.M. et al. 2019. Depressive Symptoms After PCB Exposure: Hypotheses for Underlying Pathomechanisms via the Thyroid and Dopamine System. International Journal of
Environmental Research and Public Health, 16(6):18.

424
APPENDICES

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Gaum, P.M. et al. 2017. Polychlorinated biphenyls and depression: cross-sectional and longitudinal investigation of a dopamine-related neurochemical path in the German HELPcB
surveillance program. Environ Health, 16(1):106.
Georgiadis, P. et al. 2019. DNA methylation profiling implicates exposure to PCBs in the pathogenesis of B-cell chronic lymphocytic leukemia. Environ Int, 126:24-36.
Gibson, E.A. et al. 2019. An overview of methods to address distinct research questions on environmental mixtures: an application to persistent organic pollutants and leukocyte
telomere length. Environ Health, 18(1):76.
Goutman, S.A. et al. 2019. High plasma concentrations of organic pollutants negatively impact survival in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry, 90(8):907-912.
Graceli, J.B. et al. 2020. The impact of endocrine-disrupting chemical exposure in the mammalian hypothalamic-pituitary axis. Mol Cell Endocrinol, 518:110997.
Granillo, L. et al. 2019. Polychlorinated biphenyls influence on autism spectrum disorder risk in the MARBLES cohort. Environmental Research, 171:177-184.
Grice, B.A. et al. 2017. Associations between persistent organic pollutants, type 2 diabetes, diabetic nephropathy and mortality. Occup Environ Med, 74(7):521-527.
Gross, R.S. et al. 2020. Persistent organic pollutants exposure in newborn dried blood spots and infant weight status: A case-control study of low-income Hispanic mother-infant
pairs. Environmental Pollution, 267:9.
Guercio, V. et al. 2019. Plasma levels of polychlorinated biphenyls (PCB) and the risk of soft tissue sarcoma. Med Lav, 110(5):342-352.
Guo, J.Y. et al. 2020. The Undervalued Effects of Polychlorinated Biphenyl Exposure on Breast Cancer. Clin Breast Cancer, 20(1):12-18.
Guo, L.C. et al. 2020. Changes in thyroid hormone related proteins and gene expression induced by polychlorinated biphenyls and halogen flame retardants exposure of children in a
Chinese e-waste recycling area. Sci Total Environ, 2020. 742:140597.
Guo, L.C. et al. 2019. Disruption of thyroid hormone regulated proteins and gene expression by polychlorinated biphenyls, polybrominated diphenyl ethers and new flame retardants in
residents of an e-waste region. Environmental Pollution, 254:9.
Güil-Oumrait, N. et al. 2021. Prenatal exposure to persistent organic pollutants and markers of obesity and cardiometabolic risk in Spanish adolescents. Environ Int, 151:106469.
Hamid, E.R.A. et al. 2016. In utero exposure to organochlorine pesticide residues and their potential impact on birth outcomes and fetal gender. Environmental Science and Pollution
Research, 27(27):33703-33711.
Han, L. et al. 2016. In utero exposure to polychlorinated biphenyls is associated with decreased fecundability in daughters of Michigan female fish eaters: a cohort study. Environ
Health, 15(1):92.
Han, X. et al. 2019. Associations between Exposure to Persistent Organic Pollutants and Thyroid Function in a Case-Control Study of East China. Environmental Science & Technology,
53(16):9866-9875.
Han, X. et al. 2020. Associations between the exposure to persistent organic pollutants and type 2 diabetes in East China: A case-control study. Chemosphere, 241:125030.
Henríquez-Hernández, L.A. et al. 2017. Determinants of increasing serum POPs in a population at high risk for cardiovascular disease. Results from the PREDIMED-CANARIAS study.
Environ Res, 156:477-484.
Henríquez-Hernández, L.A. et al. 2021. Human biomonitoring of persistent organic pollutants in elderly people from the Canary Islands (Spain): A temporal trend analysis from the
PREDIMED and PREDIMED-Plus cohorts. Sci Total Environ, 758:143637.
Henríquez-Hernández, L.A. et al. 2017. Persistent organic pollutants and risk of diabetes and obesity on healthy adults: Results from a cross-sectional study in Spain. Sci Total
Environ, 607:1096-1102.
Hens, B. & Hens, L. 2018. Persistent Threats by Persistent Pollutants: Chemical Nature, Concerns and Future Policy Regarding PCBs-What Are We Heading For? Toxics, 6(1):21.
Heyer, D.B. & Meredith, R.M. 2017. 2020. Environmental toxicology: Sensitive periods of development and neurodevelopmental disorders. Neurotoxicology, 58:23-41.
Hjermitslev, M.H. et al. 2020. Persistent organic pollutants in Greenlandic pregnant women and indices of foetal growth: The ACCEPT study. Science of the Total Environment, 698:12.
Hong, P., Song, Y.G. & Paek, S. 2021. Possible effects of agent orange and posttraumatic stress disorder on hyperglycemia in Korean veterans from the US-Vietnam war. Medicine
(Baltimore), 100(25):e26508.
Houston, T.J. & Ghosh, R. 2020. Untangling the association between environmental endocrine disruptive chemicals and the etiology of male genitourinary cancers. Biochemical
Pharmacology, 172:8.
Hsu, P.C. et al. 2016. Polychlorinated biphenyls and dibenzofurans increased abnormal sperm morphology without alterations in aneuploidy: The Yucheng study. Chemosphere,
165:294-297.
Hu, X. et al. 2021. Environmental chemicals and metabolic disruption in primary and secondary human parathyroid tumors. Surgery, 169(1):102-108.
Hui, L.L. et al. 2016. Prenatal dioxin exposure and neurocognitive development in Hong Kong 11-year-old children. Environ Res, 150:205-212.
Hwang, B.S. et al. 2019. A Bayesian multi-dimensional couple-based latent risk model with an application to infertility. Biometrics, 75(1):315-325.
Ikeno, T. et al. 2018. Effects of low-level prenatal exposure to dioxins on cognitive development in Japanese children at 42 months. Sci Total Environ, 618:1423-1430.
Iszatt, N. et al. 2019. Environmental toxicants in breast milk of Norwegian mothers and gut bacteria composition and metabolites in their infants at 1 month. Microbiome, 7(1):34.
Iszatt, N. et al. 2016. Perinatal exposure to dioxins and dioxin-like compounds and infant growth and body mass index at seven years: A pooled analysis of three European birth
cohorts. Environ Int, 94:399-407.
Itoh, S. et al. 2018. Association of maternal serum concentration of hydroxylated polychlorinated biphenyls with maternal and neonatal thyroid hormones: The Hokkaido birth cohort
study. Environ Res, 167:583-590.
Jacobson, M.H. et al. 2017. Serum Polybrominated Biphenyls (PBBs) and Polychlorinated Biphenyls (PCBs) and Thyroid Function among Michigan Adults Several Decades after the
1973-1974 PBB Contamination of Livestock Feed. Environ Health Perspect, 125(9):097020.
Jain, R.B. 2021. Trends in concentrations of selected dioxins and furans across various stages of kidney function for US adults. Environ Sci Pollut Res Int, 28(32):43763-43776.
Jansen, A. et al. 2020. The Influence of Persistent Organic Pollutants on Thyroidal, Reproductive and Adrenal Hormones After Bariatric Surgery. Obes Surg, 30(4):1368-1378.
Ju, Q. & Zouboulis, C.C. 2016. Endocrine-disrupting chemicals and skin manifestations. Reviews in Endocrine & Metabolic Disorders, 17(3):449-457.

425
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Jaacks, L.M. et al. 2016. Pre-Pregnancy Maternal Exposure to Persistent Organic Pollutants and Gestational Weight Gain: A Prospective Cohort Study. International Journal of
Environmental Research and Public Health, 13(9):12.
Kaifie, A. et al. 2019. Functional and structural liver abnormalities in former PCB exposed workers - analyses from the HELPcB cohort. J Toxicol Environ Health A, 82(1):52-61.
Kalloo, G. et al. 2021. Chemical mixture exposures during pregnancy and cognitive abilities in school-aged children. Environmental Research, 197:9.
Kalloo, G. et al. 2020. Exposures to chemical mixtures during pregnancy and neonatal outcomes: The HOME study. Environ Int, 134:105219.
Kamio, Y. et al. 2020. Insight into innate immune response in "Yusho": The impact of natural killer cell and regulatory T cell on inflammatory prone diathesis of Yusho patients.
Environ Res, 185:109415.
Kao, C.C. et al. 2019. Residue Levels of Organochlorine Pesticides in Breast Milk and Its Associations with Cord Blood Thyroid Hormones and the Offspring's Neurodevelopment.
International Journal of Environmental Research and Public Health, 16(8):19.
Karimi, B.R., Nabizadeh N. R., & Yunesian, M. 2020. Serum level of PCBs and OCPs and leukocyte telomere length among adults in Tehran, Iran. Chemosphere, 248:126092.
Karimi, B.R., Nabizadeh N. R., & Yunesian, M. 2020. Association Between Leukocyte Telomere Length and Serum Concentrations of PCBs and Organochlorine Pesticides. Arch Environ
Contam Toxicol, 79(1):122-130.
Karlsen, M. et al. 2017. Early-life exposures to persistent organic pollutants in relation to overweight in preschool children. Reprod Toxicol, 68:145-153.
Karwacka, A. et al. 2019. Exposure to modern, widespread environmental endocrine disrupting chemicals and their effect on the reproductive potential of women: an overview of
current epidemiological evidence. Hum Fertil (Camb), 22(1):2-25.
Kawasaki, G. & Yoshitomi, I. 2019. Effect of dioxin-related compounds on oral pigmentation in patients affected by the Yusho incident. Arch Oral Biol, 102:244-248.
Kern, J.K. et al. 2017. Developmental neurotoxicants and the vulnerable male brain: a systematic review of suspected neurotoxicants that disproportionally affect males. Acta
Neurobiol Exp (Wars), 77(4):269-296.
Khan, M.W. et al. 2019. Impact of brick kilns industry on environment and human health in Pakistan. Sci Total Environ, 678:383-389.
Khodavandi, A. Alizadeh F., & Razis, A.F.A. 2021. Association between dietary intake and risk of ovarian cancer: a systematic review and meta-analysis. Eur J Nutr, 60(4):1707-1736.
Kim, M. et al. 2020. Plasma levels of polychlorinated biphenyl, genetic polymorphisms, and the risk of advanced stage endometriosis. Gynecol Endocrinol, 36(7):636-640.
Kim, S. et al. 2018. Association between maternal exposure to major phthalates, heavy metals, and persistent organic pollutants, and the neurodevelopmental performances of their
children at 1 to 2 years of age- CHECK cohort study. Sci Total Environ, 624:377-384.
Kim, S. et al. 2019. Maternal exposures to persistent organic pollutants are associated with DNA methylation of thyroid hormone-related genes in placenta differently by infant sex.
Environ Int, 130:104956.
Kim, S. et al. 2018. Prenatal exposure to persistent organic pollutants and methylation of LINE-1 and imprinted genes in placenta: A CHECK cohort study. Environ Int, 119:398-406.
Kippler, M. et al. 2016. Associations of dietary polychlorinated biphenyls and long-chain omega-3 fatty acids with stroke risk. Environ Int, 94:706-711.
Kishi, R. et al. 2021. Hokkaido birth cohort study on environment and children's health: cohort profile 2021. Environmental Health and Preventive Medicine, 26(1):20.
Kishi, R. et al. 2018. The Hokkaido Birth Cohort Study on Environment and Children's Health: cohort profile-updated 2017. Environmental Health and Preventive Medicine, 22(1):16.
Klil-Drori, A.J. et al. 2018. Serum organochlorines and non-Hodgkin lymphoma: A case-control study in Israeli Jews and Palestinians. Chemosphere, 213:395-402.
Knudsen, A.S. et al. 2018. Persistent organic pollutants and haematological markers in Greenlandic pregnant women: the ACCEPT sub-study. Int J Circumpolar Health,
77(1):1456303.
Kobayashi, N. et al. 2017. Factors associated with aberrant imprint methylation and oligozoospermia. Sci Rep, 7:42336.
Kornvig, S. et al. 2021. Prenatal exposure to persistent organic pollutants and metals and problematic child behavior at 3-5 years of age: a Greenlandic cohort study. Scientific
Reports, 11(1):18.
Kronke, A.A. et al. 2022.Persistent organic pollutants in pregnant women potentially affect child development and thyroid hormone status. Pediatric Research, 91:690-698
Kyriklaki, A. et al. 2016. Prenatal exposure to persistent organic pollutants in association with offspring neuropsychological development at 4years of age: The Rhea mother-child
cohort, Crete, Greece. Environ Int, 97:204-211.
La Merrill, M.A. et al. 2019. Exposure to Persistent Organic Pollutants (POPs) and Their Relationship to Hepatic Fat and Insulin Insensitivity among Asian Indian Immigrants in the
United States. Environmental Science & Technology, 53(23):13906-13918.
Lamoureux-Tremblay, V. et al. 2020. Risk factors associated with developing anxiety in Inuit adolescents from Nunavik. Neurotoxicol Teratol, 81:106903.
Lauritzen, H.B. et al. 2017. Maternal serum levels of perfluoroalkyl substances and organochlorines and indices of fetal growth: a Scandinavian case-cohort study. Pediatr Res,
81(1):33-42.
Lauritzen, H.B. et al. 2018. Prenatal exposure to persistent organic pollutants and child overweight/obesity at 5-year follow-up: a prospective cohort study. Environ Health, 17(1):9.
Lee, E. et al. 2020. Serum Levels of Commonly Detected Persistent Organic Pollutants and Per- and Polyfluoroalkyl Substances (PFASs) and Mammographic Density in Postmenopausal
Women. International Journal of Environmental Research and Public Health, 17(2):10.
Lee, J.Y. et al. 2018. Association of low-dose exposure to persistent organic pollutants with E-cadherin promoter methylation in healthy Koreans. Biomarkers, 23(3):293-298.
Lee, M.H. et al. 2017. Association between serum persistent organic pollutants and DNA methylation in Korean adults. Environ Res, 158:333-341.
Lee, Y.M. et al. 2018. Association of colorectal polyps and cancer with low-dose persistent organic pollutants: A case-control study. PLoS One, 13(12):e0208546.
Lee, Y.M. et al. 2017. Low-Dose Persistent Organic Pollutants Impair Insulin Secretory Function of Pancreatic beta-Cells: Human and In Vitro Evidence. Diabetes, 66(10):2669-2680.
Leemans, M. et al. 2019. Pesticides With Potential Thyroid Hormone-Disrupting Effects: A Review of Recent Data. Frontiers in Endocrinology, 10:29.
Leijs, M.M. et al. 2017. Alterations in the programming of energy metabolism in adolescents with background exposure to dioxins, dl-PCBs and PBDEs. PLoS One, 12(9):e0184006.
Leisegang, K. & Dutta, S. 2021. Do lifestyle practices impede male fertility? Andrologia,13.

426
APPENDICES

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Lemaitre, M. et al. 2021. Dietary exposure to polychlorinated biphenyls (PCB) and risk of Non-Hodgkin's lymphoma: Evidence from the French E3N prospective cohort. Environ Res,
197:111005.
Lenters, V. et al. 2019. Early-life exposure to persistent organic pollutants (OCPs, PBDEs, PCBs, PFASs) and attention-deficit/hyperactivity disorder: A multi-pollutant analysis of a
Norwegian birth cohort. Environ Int, 125:33-42.
Lerro, C.C. et al. 2018. A nested case-control study of polychlorinated biphenyls, organochlorine pesticides, and thyroid cancer in the Janus Serum Bank cohort. Environ Res, 165:125-
132.
Leung, Y.K. et al. 2018. Identification of sex-specific DNA methylation changes driven by specific chemicals in cord blood in a Faroese birth cohort. Epigenetics, 13(3):290-300.
Li, P. et al. 2021. Serum levels of polychlorinated biphenyls and stroke risk among Chinese: a hospital-based case-control study. Acta Neurologica Belgica, 121(5):1217-1224.
Li, S.Z. et al. 2020. Understanding mixed environmental exposures using metabolomics via a hierarchical community network model in a cohort of California women in 1960's.
Reproductive Toxicology, 92:57-65.
Liberda, E.N., Zuk, A.M. & Tsuji, L.J.S. 2021. Heart rate variation and human body burdens of environmental mixtures in the Cree First Nation communities of Eeyou Istchee, Canada.
Environ Int, 146:106220.
Lin, B.G. et al. 2021. Effects of organochlorine exposure on male reproductive disorders in an electronic waste area of South China. Environ Int, 147:106318.
Lind, L. et al. 2017. Mixture effects of 30 environmental contaminants on incident metabolic syndrome-A prospective study. Environ Int, 107:8-15.
Lind, P.M. et al. 2019. Association of Exposure to Persistent Organic Pollutants With Mortality Risk: An Analysis of Data From the Prospective Investigation of Vasculature in Uppsala
Seniors (PIVUS) Study. JAMA Netw Open, 2(4):e193070.
Ljunggren, S.A. et al. 2017. Alterations in high-density lipoprotein proteome and function associated with persistent organic pollutants. Environ Int, 98:204-211.
Louis, G.M.B. et al. 2018. Endocrine disruptors and neonatal anthropometry, NICHD Fetal Growth Studies - Singletons. Environment International, 119:515-526.
Louzon, M. et al. 2019. Telomere dynamic in humans and animals: Review and perspectives in environmental toxicology. Environ Int, 131:105025.
Lucchini, R.G. et al. 2017. A comparative assessment of major international disasters: the need for exposure assessment, systematic emergency preparedness, and lifetime health
care. Bmc Public Health,. 17:12.
Lundin, M.S., Abro, C. & Laird-Fick, H. 2019. Intra-abdominal follicular lymphoma masquerading as severe cardiopulmonary disease in a Vietnam War veteran exposed to agent
orange. BMJ Case Rep, 12(3).
Luo, D. et al. 2017. Association of in utero exposure to organochlorine pesticides with thyroid hormone levels in cord blood of newborns. Environmental Pollution, 231:78-86.
Lyall, K. et al. 2017. Polychlorinated Biphenyl and Organochlorine Pesticide Concentrations in Maternal Mid-Pregnancy Serum Samples: Association with Autism Spectrum Disorder
and Intellectual Disability. Environ Health Perspect, 125(3):474-480.
Magliano, D.J. et al. 2021. Exposure to persistent organic pollutants and the risk of type 2 diabetes: a case-cohort study. Diabetes & Metabolism, 47(5):8.
Magnus, P. 2017. Looking for effects of environmental contaminants in a large birth cohort: Summarizing results of the Norwegian Mother and Child Cohort Study (MoBa).
International Journal of Hygiene and Environmental Health, 220(2):71-76.
Magoni, M. et al. 2018. Plasma levels of polychlorinated biphenyls and risk of cutaneous malignant melanoma: A hospital-based case-control study. Environment International,
113:20-25.
Magoni, M. et al. 2019. Serum levels of polychlorinated biphenyls and risk of non-Hodgkin lymphoma: A hospital-based case-control study. Chemosphere, 235:969-975.
Maitre, L. et al. 2021. Early-life environmental exposure determinants of child behavior in Europe: A longitudinal, population-based study. Environ Int, 153:106523.
Maitre, L. et al. 2018. Urine Metabolic Signatures of Multiple Environmental Pollutants in Pregnant Women: An Exposome Approach. Environ Sci Technol, 52(22):13469-13480.
Mallon, C.T. et al. 2016. Introduction to Department of Defense Research on Burn Pits, Biomarkers, and Health Outcomes Related to Deployment in Iraq and Afghanistan. J Occup
Environ Med, 58(8):S3-s11.
Mannetje, A.T. et al. 2017. Sex ratio of the offspring of New Zealand phenoxy herbicide producers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Occup Environ Med, 74(1):24-29.
Mansouri, E.H. & Reggabi, M. 2021. Association between type 2 diabetes and exposure to chlorinated persistent organic pollutants in Algeria: A case-control study. Chemosphere,
264:128596.
Marks, K.J. et al. 2021. Prenatal exposure to mixtures of persistent endocrine disrupting chemicals and early menarche in a population-based cohort of British girls. Environ Pollut,
276:116705.
Marks, K.J. et al. 2021. Prenatal exposure to mixtures of persistent endocrine disrupting chemicals and postnatal body size in British girls. Early Human Development, 161:9.
Marks, K.J. et al. 2021. Prenatal Exposure to Mixtures of Persistent Endocrine-disrupting Chemicals and Birth Size in a Population-based Cohort of British Girls. Epidemiology,
32(4):573-582.
Marotta, V. et al. 2020. Fathoming the link between anthropogenic chemical contamination and thyroid cancer. Crit Rev Oncol Hematol, 150:102950.
Marushka, L. et al. 2021. The relationship between dietary exposure to persistent organic pollutants from fish consumption and type 2 diabetes among First Nations in Canada.
Canadian Journal of Public Health-Revue Canadienne De Sante Publique, 112:168-182.
Marushka, L. et al. 2018. The Relationship between Persistent Organic Pollutants Exposure and Type 2 Diabetes among First Nations in Ontario and Manitoba, Canada: A Difference in
Difference Analysis. Int J Environ Res Public Health, 15(3).
Medehouenou, T.C.M. et al. 2019. Exposure to polychlorinated biphenyls and organochlorine pesticides and risk of dementia, Alzheimer's disease and cognitive decline in an older
population: a prospective analysis from the Canadian Study of Health and Aging. Environ Health, 18(1):57.
Mehta, S.S. et al. 2021. Persistent organic pollutants and maternal glycemic outcomes in a diverse pregnancy cohort of overweight women. Environmental Research, 193:7.
Mínguez-Alarcón, L. et al. 2017. A Longitudinal Study of Peripubertal Serum Organochlorine Concentrations and Semen Parameters in Young Men: The Russian Children's Study.
Environ Health Perspect, 2017. 125(3):460-466.
Miri, M. et al. 2019. Air pollution and telomere length in adults: A systematic review and meta-analysis of observational studies. Environ Pollut, 244:636-647.

427
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Mohanto, N.C. et al. 2021. Life-Time Environmental Chemical Exposure and Obesity: Review of Epidemiological Studies Using Human Biomonitoring Methods. Frontiers in
Endocrinology, 12:26.
Morgan, M. et al. 2017. Environmental estrogen-like endocrine disrupting chemicals and breast cancer. Mol Cell Endocrinol, 457:89-102.
Mowery, A.M. Conlin, & Clayburgh, D. 2020. Increased risk of head and neck cancer in Agent Orange exposed Vietnam Era veterans. Oral Oncol, 100:104483.
Murinova, L.P. et al. 2016. PCB exposure and cochlear function at age 6 years. Environmental Research, 151:428-435.
Mustieles, V. et al. 2021. Adipose Tissue Redox Microenvironment as a Potential Link between Persistent Organic Pollutants and the 16-Year Incidence of Non-hormone-Dependent
Cancer. Environ Sci Technol, 55(14):9926-9937.
Nakajima, S. et al. 2017. Sex-specific differences in effect of prenatal exposure to dioxin-like compounds on neurodevelopment in Japanese children: Sapporo cohort study. Environ
Res, 159:222-231.
Namulanda, G. et al. 2016. In utero exposure to organochlorine pesticides and early menarche in the Avon Longitudinal Study of Parents and Children. Environment International,
94:467-472.
Nandipati, S. & Litvan, I. 2016. Environmental Exposures and Parkinson's Disease. Int J Environ Res Public Health, 13(9).
Naqvi, A. et al. 2018. Quantification of polychlorinated biphenyl contamination using human placenta as biomarker from Punjab Province, Pakistan. Environmental Science and
Pollution Research, 25(15):14551-14562.
Neblett, M.F. et al. 2020. 2019. Examining Reproductive Health Outcomes in Females Exposed to Polychlorinated Biphenyl and Polybrominated Biphenyl. Sci Rep, 10(1):3314.
Ngoc, V.T.N. et al. 2019. The higher prevalence of developmental defects of enamel in the dioxin-affected region than non-dioxin-affected region: result from a cross-sectional study
in Vietnam. Odontology, 107(1):17-22.
Nguyen, C.H. et al. 2017. Expression of aryl hydrocarbon receptor, inflammatory cytokines, and incidence of rheumatoid arthritis in Vietnamese dioxin-exposed people. J
Immunotoxicol, 14(1):196-203.
Niehoff, N.M. et al. 2020. Prediagnostic serum polychlorinated biphenyl concentrations and primary liver cancer: A case-control study nested within two prospective cohorts. Environ
Res, 187:109690.
Nowak, K.E. Jablonska & Ratajczak-Wrona, W. 2019. Immunomodulatory effects of synthetic endocrine disrupting chemicals on the development and functions of human immune
cells. Environment International, 125:350-364.
Nowak, K. E. Jabłońska & Ratajczak-Wrona, W. 2019. Neutrophils life under estrogenic and xenoestrogenic control. J Steroid Biochem Mol Biol, 186:203-211.
Odutola, M.K. et al. 2021. A systematic review and meta-analysis of occupational exposures and risk of follicular lymphoma. Environ Res, 197:110887.
Onozuka, D. et al. 2020. Mortality in Yusho patients exposed to polychlorinated biphenyls and polychlorinated dibenzofurans: a 50-year retrospective cohort study. Environ Health,
19(1):119.
Ouidir, M. et al. 2020. Association of Maternal Exposure to Persistent Organic Pollutants in Early Pregnancy With Fetal Growth. Jama Pediatrics, 174(2):149-161.
Oulhote, Y. et al. 2017. Children's white blood cell counts in relation to developmental exposures to methylmercury and persistent organic pollutants. Reprod Toxicol, 68:207-214.
Palkovičová Murínová, Ľ. et al. 2016. PCB exposure and cochlear function at age 6 years. Environ Res, 151:428-435.
Pan, W. et al. 2019. Selected persistent organic pollutants associated with the risk of primary ovarian insufficiency in women. Environ Int, 129:51-58.
Parada, H. Jr. et al. 2021. A Congener-specific and Mixture Analysis of Plasma Polychlorinated Biphenyl Levels and Incident Breast Cancer. Epidemiology, 32(4):499-507.
Parada, H. Jr. et al. 2020. Plasma levels of polychlorinated biphenyls (PCBs) and breast cancer mortality: The Carolina Breast Cancer Study. Int J Hyg Environ Health, 227:113522.
Parazzini, F. et al. 2017. Epidemiology of endometriosis and its comorbidities. Eur J Obstet Gynecol Reprod Biol, 209:3-7.
Park, E.Y. et al. 2020. Impact of environmental exposure to persistent organic pollutants on lung cancer risk. Environ Int, 143:105925.
Park, S.H. et al. 2016. Serum Levels of Persistent Organic Pollutants and Insulin Secretion among Children Age 7-9 Years: A Prospective Cohort Study. Environ Health Perspect,
124(12):1924-1930.
Pearce, J.L. et al. 2021. Exploring associations between prenatal exposure to multiple endocrine disruptors and birth weight with exposure continuum mapping. Environ Res,
200:111386.
Pelletier, M. et al. 2018. Chemical-by-chemical and cumulative risk assessment of residential indoor exposure to semivolatile organic compounds in France. Environ Int, 117:22-32.
Peters, J.L., Fabian, M.P. & Levy, J.I. 2019. Epidemiologically-informed cumulative risk hypertension models simulating the impact of changes in metal, organochlorine, and non-
chemical exposures in an environmental justice community. Environ Res,176:108544.
Petersen, M.S. et al. 2018. Reproductive Function in a Population of Young Faroese Men with Elevated Exposure to Polychlorinated Biphenyls (PCBs) and Perfluorinated Alkylate
Substances (PFAS). International Journal of Environmental Research and Public Health, 15(9):14.
Pittman, G.S. et al. 2020. Polychlorinated biphenyl exposure and DNA methylation in the Anniston Community Health Survey. Epigenetics, 15(4):337-357.
Ploteau, S. et al. 2017. Associations between internal exposure levels of persistent organic pollutants in adipose tissue and deep infiltrating endometriosis with or without concurrent
ovarian endometrioma. Environ Int, 108:195-203.
Ploteau, S. et al. 2016. Distribution of persistent organic pollutants in serum, omental, and parietal adipose tissue of French women with deep infiltrating endometriosis and
circulating versus stored ratio as new marker of exposure. Environ Int, 97:125-136.
Polevoy, C. et al. 2020. Prenatal exposure to legacy contaminants and visual acuity in Canadian infants: a maternal-infant research on environmental chemicals study (MIREC-ID).
Environ Health, 19(1):14.
Pollack, A.Z. et al. 2021. Adipose to serum ratio and mixtures of persistent organic pollutants in relation to endometriosis: Findings from the ENDO Study. Environ Res, 2021.
195:110732.
Post, C.M. et al. 2019. The Ancestral Environment Shapes Antiviral CD8(+) T cell Responses across Generations. Iscience, 20:168.
Przybyla, J. et al. 2017. A path analysis of multiple neurotoxic chemicals and cognitive functioning in older US adults (NHANES 1999-2002). Environ Health, 16(1):19.

428
APPENDICES

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Qi, Z.H. et al. 2020. Chemical identity and cardiovascular toxicity of hydrophobic organic components in PM2.5. Ecotoxicology and Environmental Safety, 201:17.
Raffetti, E. et al. 2020. Polychlorinated biphenyls (PCBs) and risk of dementia and Parkinson disease: A population-based cohort study in a North Italian highly polluted area.
Chemosphere, 261:127522.
Raffetti, E. et al. 2020. Polychlorinated biphenyls (PCBs) and risk of hypertension: A population-based cohort study in a North Italian highly polluted area. Sci Total Environ,
714:136660.
Raffetti, E. et al. 2018. Polychlorinated biphenyls (PCBs) exposure and cardiovascular, endocrine and metabolic diseases: A population-based cohort study in a North Italian highly
polluted area. Environ Int, 120:215-222.
Rahman, M.L. et al. 2019. Persistent organic pollutants and gestational diabetes: A multi-center prospective cohort study of healthy US women. Environment International, 124:249-
258.
Ramalingam, S. et al. 2021. Persistent organic pollutants-environmental risk factors for diabetes mellitus? - A population-based study. Indian Journal of Occupational and
Environmental Medicine, 25(3):157-162.
Rodgers, K.M. et al. 2018. Environmental chemicals and breast cancer: An updated review of epidemiological literature informed by biological mechanisms. Environ Res, 160:152-
182.
Rosenbaum, P.F. et al. 2017. Metabolic syndrome is associated with exposure to organochlorine pesticides in Anniston, AL, United States. Environ Int, 108:11-21.
Roswall, N. et al. 2018. No Association between Organochlorine Concentrations in Adipose Tissue and Survival after Non-Hodgkin Lymphoma. Cancer Epidemiol Biomarkers Prev,
27(2):224-226.
Roswall, N. et al. 2018. Organochlorine concentrations in adipose tissue and survival in postmenopausal, Danish breast cancer patients. Environmental Research, 163:237-248.
Ruder, A.M. et al. 2017. Cancer incidence among capacitor manufacturing workers exposed to polychlorinated biphenyls. Am J Ind Med, 60(2):198-207.
Rusiecki, J.A. et al. 2020. Serum concentrations of DDE, PCBs, and other persistent organic pollutants and mammographic breast density in Triana, Alabama, a highly exposed
population. Environ Res, 182:109068.
Schildroth, S. et al. 2021. Correlates of Persistent Endocrine-Disrupting Chemical Mixtures among Reproductive-Aged Black Women. Environ Sci Technol, 55(20):14000-14014.
Schug, T.T. et al. 2016. Minireview: Endocrine Disruptors: Past Lessons and Future Directions. Molecular Endocrinology, 30(8):833-847.
Schwarz, M. et al. 2021. Association of persistent organic pollutants with sensorimotor neuropathy in participants with and without diabetes or prediabetes: Results from the
population-based KORA FF4 study. Int J Hyg Environ Health, 235:113752.
Schæbel, L.K. et al. 2017. The influence of persistent organic pollutants in the traditional Inuit diet on markers of inflammation. PLoS One, 12(5):e0177781.
Sen, A. & Sellix, M.T. 2016. The Circadian Timing System and Environmental Circadian Disruption: From Follicles to Fertility. Endocrinology, 157(9):3366-3373.
Sergeyev, O. et al. 2017. The association of peripubertal serum concentrations of organochlorine chemicals and blood lead with growth and pubertal development in a longitudinal
cohort of boys: a review of published results from the Russian Children's Study. Rev Environ Health, 32(1):83-92.
Singh, K. & Chan, H.M. 2018. Association of blood polychlorinated biphenyls and cholesterol levels among Canadian Inuit. Environ Res, 160:298-305.
Singh, K. & Chan, H.M. 2017. Chan, Persistent organic pollutants and diabetes among Inuit in the Canadian Arctic. Environ Int, 101:183-189.
Smarr, M.M. et al. 2021. A multi-pollutant assessment of preconception persistent endocrine disrupting chemicals and incident pregnancy loss. Environ Int, 157:106788.
Soechitram, S.D. et al. 2017. Polychlorinated biphenyl exposure and deiodinase activity in young infants. Sci Total Environ, 574:1117-1124.
Steinholt, M. et al. 2020. Serum Concentrations of Selected Organochlorines in Pregnant Women and Associations with Pregnancy Outcomes. A Cross-Sectional Study from Two Rural
Settings in Cambodia. Int J Environ Res Public Health, 17(20).
Su, K.Y. et al. 2019. Perinatal polychlorinated biphenyls and polychlorinated dibenzofurans exposure are associated with DNA methylation changes lasting to early adulthood: Findings
from Yucheng second generation. Environ Res, 170:481-486.
Suarez-Lopez, J.R. et al. 2019. Organochlorine pesticides and polychlorinated biphenyls (PCBs) in early adulthood and blood lipids over a 23-year follow-up. Environ Toxicol
Pharmacol, 66:24-35.
Svensson, K. et al. 2021. Prenatal exposures to mixtures of endocrine disrupting chemicals and children's weight trajectory up to age 5.5 in the SELMA study. Scientific Reports,
11(1):12.
Tanner, E.M. et al. 2020. A longitudinal study of polychlorinated biphenyls and neuropsychological function among older adults from New York State. Int J Hyg Environ Health,
223(1):1-9.
Tatsuta, N. et al. 2017. Effects of intrauterine exposures to polychlorinated biphenyls, methylmercury, and lead on birth weight in Japanese male and female newborns. Environ Health
Prev Med, 22(1):39.
Timmermann, C.A.G. et al. 2022. Concentrations of tetanus and diphtheria antibodies in vaccinated Greenlandic children aged 7-12 years exposed to marine pollutants, a cross
sectional study. Environ Res, 203:111712.
Timmermann, C.A.G. et al. 2017. Secondary sex ratio in relation to exposures to polychlorinated biphenyls, dichlorodiphenyl dichloroethylene and methylmercury. International Journal
of Circumpolar Health, 76:8.
Trasande, L. et al. 2017. Population attributable risks and costs of diabetogenic chemical exposures in the elderly. J Epidemiol Community Health, 71(2):111-114.
Tuomisto, J. et al. 2016. A pharmacokinetic analysis and dietary information are necessary to confirm or reject the hypothesis on persistent organic pollutants causing type 2
diabetes. Toxicol Lett, 261:41-48.
Tuomisto, J. et al. 2017. Comparison of questionnaire data and analyzed dioxin concentrations as a measure of exposure in soft-tissue sarcoma studies. Toxicol Lett, 270:8-11.
Tuomisto, J.T. et al. 2020. 2Health effects of nutrients and environmental pollutants in Baltic herring and salmon: a quantitative benefit-risk assessment. Bmc Public Health,
20(1):18.

429
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A4.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECT OF DIOXINS
AND dl-PCBs” BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Tyagi, S. et al. 2021. Level of Organochlorine Pesticide in Prediabetic and Newly Diagnosed Diabetes Mellitus Patients with Varying Degree of Glucose Intolerance and Insulin
Resistance among North Indian Population. Diabetes & Metabolism Journal, 45(4):558-568.
Vafeiadi, M. et al. 2017. Persistent organic pollutants in early pregnancy and risk of gestational diabetes mellitus. Environ Int, 98:89-95.
Valvi, D. et al. 2020. Environmental chemical burden in metabolic tissues and systemic biological pathways in adolescent bariatric surgery patients: A pilot untargeted metabolomic
approach. Environ Int, 143:105957.
Valvi, D. et al. 2017. Gestational diabetes and offspring birth size at elevated environmental pollutant exposures. Environment International, 107:205-215.
van den Dungen, M.W. et al. 2017. Association between DNA methylation profiles in leukocytes and serum levels of persistent organic pollutants in Dutch men. Environmental
Epigenetics, 3(1):11.
Vinceti, M. et al. 2017. Pesticides, polychlorinated biphenyls and polycyclic aromatic hydrocarbons in cerebrospinal fluid of amyotrophic lateral sclerosis patients: a case-control
study. Environ Res, 155:261-267.
Vuong, A.M. et al. 2020. Prenatal exposure to a mixture of persistent organic pollutants (POPs) and child reading skills at school age. Int J Hyg Environ Health, 228:113527.
Wahlang, B. et al. 2020. Insecticide and metal exposures are associated with a surrogate biomarker for non-alcoholic fatty liver disease in the National Health and Nutrition
Examination Survey 2003-2004. Environmental Science and Pollution Research, 27(6):6476-6487.
Wallin, A. et al. 2017. Fish consumption and frying of fish in relation to type 2 diabetes incidence: a prospective cohort study of Swedish men. European Journal of Nutrition,
56(2):843-852.
Wang, S.S. et al. 2021. Association between prenatal exposure to persistent organic pollutants and neurodevelopment in early life: A mother-child cohort (Shanghai, China).
Ecotoxicology and Environmental Safety, 208:8.
Wang, S.S. et al. 2022. Effects of prenatal exposure to persistent organic pollutants on neonatal Outcomes:A mother-child cohort (Shanghai, China). Environ Res, 203:111767.
Wang, Y. et al. 2018. Genomic instability in adult men involved in processing electronic waste in Northern China. Environ Int, 117:69-81.
Wania, F., Binnington, M.J. & Curren, M.S. 2017. Mechanistic modeling of persistent organic pollutant exposure among indigenous Arctic populations: motivations, challenges, and
benefits. Environmental Reviews, 25(4):396-407.
Warembourg, C. et al. 2019. Early-Life Environmental Exposures and Blood Pressure in Children. J Am Coll Cardiol, 74(10):1317-1328.
Weber, L. et al. 2018. Organochlorine Levels in Plasma and Risk of Multiple Myeloma. J Occup Environ Med, 60(10):911-916.
Wen, X. et al. 2019. The risk of endometriosis after exposure to endocrine-disrupting chemicals: a meta-analysis of 30 epidemiology studies. Gynecol Endocrinol, 35(8):645-650.
Wesselink, A.K. et al. 2021. A Prospective Ultrasound Study of Plasma Polychlorinated Biphenyl Concentrations and Incidence of Uterine Leiomyomata. Epidemiology, 32(2):259-267.
Wielsoe, M., Kern P. & Bonefeld-Jorgensen, E.C. 2017. Serum levels of environmental pollutants is a risk factor for breast cancer in Inuit: a case control study. Environmental Health,
16:16.
Wolf, K. et al. 2019. Persistent organic pollutants and the incidence of type 2 diabetes in the CARLA and KORA cohort studies. Environ Int, 129:221-228.
Woods, M.M. et al. 2017. Gestational exposure to endocrine disrupting chemicals in relation to infant birth weight: a Bayesian analysis of the HOME Study. Environ Health, 16(1):115.
Xu, C. et al. 2019. Exploring the associations of serum concentrations of PCBs, PCDDs, and PCDFs with walking speed in the U.S. general population: Beyond standard linear models.
Environ Res, 178:108666.
Yamazaki, K. et al. 2018. Association between prenatal exposure to organochlorine pesticides and the mental and psychomotor development of infants at ages 6 and 18 months: The
Hokkaido Study on Environment and Children's Health. Neurotoxicology, 69:201-208.
Yamazaki, K. et al. 2020. Associations between prenatal exposure to organochlorine pesticides and thyroid hormone levels in mothers and infants: The Hokkaido study on environment
and children's health. Environmental Research, 189:10.
Zani, C. et al. 2019. Polychlorinated biphenyl serum levels, thyroid hormones and endocrine and metabolic diseases in people living in a highly polluted area in North Italy: A
population-based study. Heliyon, 5(6):7.
Zhang, H. et al. 2017. Prenatal PBDE and PCB Exposures and Reading, Cognition, and Externalizing Behavior in Children. Environ Health Perspect, 125(4):746-752.
Zheng, J. et al. 2017. Disruption of thyroid hormone (TH) levels and TH-regulated gene expression by polybrominated diphenyl ethers (PBDEs), polychlorinated biphenyls (PCBs), and
hydroxylated PCBs in e-waste recycling workers. Environ Int, 102:138-144.
Ziegler, S. et al. 2017. Accelerated telomere shortening in peripheral blood lymphocytes after occupational polychlorinated biphenyls exposure. Arch Toxicol, 91(1):289-300.
Zong, G. et al. 2016. Lactation history, serum concentrations of persistent organic pollutants, and maternal risk of diabetes. Environ Res, 150:282-288.
Zong, G. et al. 2018. Persistent organic pollutants and risk of type 2 diabetes: A prospective investigation among middle-aged women in Nurses' Health Study II. Environment
International, 114:334-342.
Zota, A.R. et al. 2018. Association between persistent endocrine-disrupting chemicals (PBDEs, OH-PBDEs, PCBs, and PFASs) and biomarkers of inflammation and cellular aging
during pregnancy and postpartum. Environ Int, 115:9-20.
Zuk, A.M. et al. 2019. Examining environmental contaminant mixtures among adults with type 2 diabetes in the Cree First Nation communities of Eeyou Istchee, Canada. Sci Rep,
9(1):15909.
Åkesson, A. et al. 2019. Dietary exposure to polychlorinated biphenyls and risk of heart failure - A population-based prospective cohort study. Environ Int, 126:1-6.

430
APPENDICES

TABLE A4.5 PRIMARY STUDIES EXCLUDED AS THEY HAD ALREADY BEEN ASSESSED IN THE DIOXIN RISK ASSESSMENTS
IN EFSA 2018
Study (n = 11)
Caspersen, I.H. et al. 2016. The influence of maternal dietary exposure to dioxins and PCBs during pregnancy on ADHD symptoms.and cognitive functions in Norwegian preschool
children. Environment International, 94:649-660.
Cypel, Y.S. et al. 2016. Herbicide Exposure, Vietnam Service, and Hypertension Risk in Army Chemical Corps Veterans. J Occup Environ Med, 58(11):1127-1136.
Hsu, P.C. et al. 2016. Polychlorinated biphenyls and dibenzofurans increased abnormal sperm morphology without alterations in aneuploidy: The Yucheng study. Chemosphere,
165:294-297.
Hui, L.L. et al. 2016. Prenatal dioxin exposure and neurocognitive development in Hong Kong 11-year-old children. Environ Res, 150:205-212.
Ikeno, T. et al. 2018. Effects of low-level prenatal exposure to dioxins on cognitive development in Japanese children at 42months. Sci Total Environ, 618:1423-1430.
Iszatt, N. et al. 2016. Perinatal exposure to dioxins and dioxin-like compounds and infant growth and body mass index at seven years: A pooled analysis of three European birth
cohorts. Environ Int, 94:399-407.
Leijs, M.M. et al. 2017. Alterations in the programming of energy metabolism in adolescents with background exposure to dioxins, dl-PCBs and PBDEs. PLoS One, 12(9):e0184006.
Mannetje, A.T. et al. 2017. Sex ratio of the offspring of New Zealand phenoxy herbicide producers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Occup Environ Med, 74(1):24-29.
Mínguez-Alarcón, L. et al. 2017. A Longitudinal Study of Peripubertal Serum Organochlorine Concentrations and Semen Parameters in Young Men: The Russian Children's Study.
Environ Health Perspect, 125(3):460-466.
Nakajima, S. et al. 2017. Sex-specific differences in effect of prenatal exposure to dioxin-like compounds on neurodevelopment in Japanese children: Sapporo cohort study. Environ
Res, 159:222-231.
Ploteau, S. et al. 2017. Associations between internal exposure levels of persistent organic pollutants in adipose tissue and deep infiltrating endometriosis with or without concurrent
ovarian endometrioma. Environ Int,108:195-203.

431
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS IN “TOXIC EFFECTS OF

DIOXINS AND dl-PCBs”
TABLE A4.6 QUALITY ASSESSMENT (RISK OF BIAS) OF THE RISK ASSESSMENT REPORT OF DIOXINS FROM (EFSA, 2018) WITH THE
QUALITY ASSESSMENT TOOL AMSTAR 2
EFSA. 2018. Risk for animal and human health related to the presence of dioxins and dioxin‐like PCBs in feed and food. EFSA Journal, 16(11):e05333.
Question domain Categories of answers Judgement Score
Q1 Inclusion of PICO yes/no Yes, defined population, intervention (exposure), 1
control group and outcome
Q2 Protocol yes/partial yes/no Yes, protocol published in 2016. 1
Q3 Explanation of included study design yes/no Yes, included both human and animal studies 1
as there is a low number of human studies
available.
Q4 Comprehensive literature search strategy yes/partial yes/no Partial yes 0.5
Q5 Paired study selection yes/no Yes 1
Q6 Paired data extraction yes/no Yes 1
Q7 List of excluded studies yes/partial yes/no Yes 1
Q8 Description of included studies yes/partial yes/no Partial yes 0.5
Q9 Risk-of-bias tool yes/partial yes/no/includes only RCTs or NRSI Partial yes 0.5
Q10 Sources of funding for included studies yes/no Yes 1
Q11 Appropriate statistical methods in meta-analysis yes/no/no meta-analysis conducted No meta-analyses conducted n/a
Q12 Impact of risk of bias in meta-analysis yes/no/no meta-analysis conducted No meta-analyses conducted n/a
Q13 Impact of risk of bias in individual studies when yes/no Yes, OHAT analyses included. 1
interpreting results
Q14 Heterogeneity assessed yes/no Not mentioned in the report. 0
Q15 Publication bias assessed yes/no/no meta-analysis conducted No meta-analyses conducted n/a
Q16 Conflict of interest included yes/no Yes, implicit in being an EFSA expert. 1
Total score 10.5
Percent 66%
Percent (exclude n/a questions) 81%
Overall AMSTAR 2 judgement (confidence in the results) High
Include/exclude Include

432
 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES
TABLE A4.7 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”, USING THE QUALITY ASSESSMENT TOOL
APPENDICES

OF THE OHAT (OFFICE OF HEALTH ASSESSMENT AND TRANSLATION)

Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Alvarez-Silvares E, Fernández-Cruz T, Domínguez-Vigo P, et al. 2021. Association between placenta concentrations polybrominated and polychlorinated
-- ++ ++ - - - - Tier 2
biphenyls and gestational diabetes mellitus: a case-control study in northwestern Spain. Environ Sci Pollut Res Int, 28:10292-10301.
Alvarez-Silvares E, Rubio-Cid P, Gonzalez-Gomez X, et al. 2021. Determination of organic pollutants in meconium and its relationship with
- - ++ + + - -- Tier 2
fetal growth. Case control study in Northwestern Spain. Journal of Perinatal Medicine, 49:884-896.
Ames J, Warner M, Brambilla P, et al. 2018. Neurocognitive and physical functioning in the Seveso Women's Health Study. Environ Res, 162:55-
- - + + - + - Tier 2
62.
Ames, J., Warner, M., Mocarelli, P., et al. 2018. AHR gene-dioxin interactions and birthweight in the Seveso Second Generation Health Study. Int
- + + + ++ ++ + Tier 2
J Epidemiol, 7:1992-2004.
Ames, J, Warner, M, Siracusa C, et al. 2019. Prenatal dioxin exposure and neuropsychological functioning in the Seveso Second Generation
- + ++ + ++ ++ ++ Tier 2
Health Study. Int J Hyg Environ Health, 222:425-433.
Aminov, Z. & Carpenter, D.O. 2020. Serum concentrations of persistent organic pollutants and the metabolic syndrome in Akwesasne Mohawks,
+ ++ -- + ++ ++ ++ Tier 2
a Native American community. Environ Pollut, 260:114004.
Anh LT, Kido T, Seijiro HB, et al. 2017. A relationship in adrenal androgen levels between mothers and their children from a dioxin-exposed
-- - ++ - ++ ++ - Tier 2
region in Vietnam. Science of the Total Environment, 607:32-41.
Baba T, Ito S, Yuasa M, et al. 2018. Association of prenatal exposure to PCDD/Fs and PCBs with maternal and infant thyroid hormones: The
++ + + + - ++ + Tier 1
Hokkaido Study on Environment and Children's Health. Sci Total Environ, 615:1239-1246.
Boda H, Nghi TN, Nishijo M, et al. 2018. Prenatal dioxin exposure estimated from dioxins in breast milk and sex hormone levels in umbilical
+ ++ ++ + + ++ ++ Tier 1
cord blood in Vietnamese newborn infants. Science of the Total Environment, 615:1312-1318.
Burns JS, Williams PL, Sergeyev O, et al. 2020. Associations of peri-pubertal serum dioxins and polychlorinated biphenyls with growth and
- + + + + + - Tier 2
body composition among Russian boys in a longitudinal cohort. Int J Hyg Environ Health, 223:228-237.
Chessa MA, La Placa M, Patrizi A, et al. 2021. Chloracne: a case series on cutaneous expression of CYP1A1 as diagnostic biomarker. Clin Exp
-- - - - + ++ - Tier 3
Dermatol, 46:896-900.
Chu CP, Wu SW, Huang YJ, et al. 2019. Neuroimaging signatures of brain plasticity in adults with prenatal exposure to polychlorinated
- - + + + ++ - Tier 2
biphenyls: Altered functional connectivity on functional MRI. Environ Pollut, 250:960-968.
Collins JJ, Bodner KM, Aylward LL, et al. 2016. Mortality risk among workers with exposure to dioxins. Occup Med (Lond), 66:706-712.
-- - - - - - - Tier 3

Donat-Vargas C, Moreno-Franco B, Laclaustra M, et al. 2020. Exposure to dietary polychlorinated biphenyls and dioxins, and its relationship
+ - ++ + + + + Tier 2
with subclinical coronary atherosclerosis: The Aragon Workers' Health Study. Environ Int, 136:105433.
Dong JJ, Ruan MC, Hang JG, et al. 2020. The relationship between perinatal exposure to dioxins and serum steroid hormone levels in preschool-
- - ++ + + + - Tier 2
aged children at an e-waste region in China. Int J Hyg Environ Health, 229:113580.
Eskenazi B, Ames J, Rauch S, et al. 2021. Dioxin exposure associated with fecundability and infertility in mothers and daughters of Seveso,
+ + - ++ ++ - + Tier 2
Italy. Hum Reprod, 36:794-807.
Fukushi J, Tsushima H, Matsumoto Y, et al. 2019. Influence of dioxin-related compounds on physical function in Yusho incident victims.
- ++ + + ++ ++ - Tier 2
Heliyon, 5:7.

433
434
TABLE A4.7 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”, USING THE QUALITY ASSESSMENT TOOL
OF THE OHAT (OFFICE OF HEALTH ASSESSMENT AND TRANSLATION) (cont.)
Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Henriquez-Hernandez LA, Boada LD, Perez-Arellano JL, et al. 2016. Relationship of polychlorinated biphenyls (PCBs) with parasitism, iron homeostasis,
- - - - + + - Tier 3
and other health outcomes: Results from a cross-sectional study on recently arrived African immigrants. Environmental Research, 50:549-556.
Huang CY, Lee CC, Chang JW, et al. 2017. Association Between Dioxin and Metabolic Syndrome by Age and Sex in an Endemic Area of Exposure
- ++ + + ++ ++ - Tier 2
in Taiwan. Epidemiology, 28:S82-s88.
Hui LL, Lam HS, Lau EYY, et al. 2016. Prenatal dioxin exposure and neurocognitive development in Hong Kong 11-year-old children. Environ
+ - + - + ++ - Tier 2
Res, 150:205-212.
Kelsey KT, Rytel M, Dere E, et al. 2019. Serum dioxin and DNA methylation in the sperm of operation ranch hand veterans exposed to Agent
+ - - + ++ + - Tier 2
Orange. Environ Health, 18:91.
Kobayashi S, Sata F, Miyashita C, et al. 2017. Dioxin-metabolizing genes in relation to effects of prenatal dioxin levels and reduced birth size:
++ ++ ++ ++ ++ ++ ++ Tier 1
The Hokkaido study. Reprod Toxicol, 67:111-116.
Kondo H, Tanio K, Nagaura Y, et al. 2018. Sleep disorders among Yusho patients highly intoxicated with dioxin-related compounds: A 140-case
- ++ + ++ + + - Tier 2
series. Environ Res, 166:261-268.
Koual M, Cano-Sancho G, Bats AS, et al. 2019. Associations between persistent organic pollutants and risk of breast cancer metastasis.
+ ++ ++ ++ ++ ++ ++ Tier 1
Environ Int, 132:105028.
Lambertino A, Persky V, Freels S, et al. 2021. Associations of PCBS, dioxins and furans with follicle-stimulating hormone and luteinizing
+ ++ + + + ++ - Tier 1
hormone in postmenopausal women: National Health and Nutrition Examination Survey 1999-2002. Chemosphere, 262:128309.
Leijs MM, Esser A, Amann PM, et al. 2018. Hyperpigmentation and higher incidence of cutaneous malignancies in moderate-high PCB- and
- + + + - + + Tier 2
dioxin exposed individuals. Environ Res, 164:221-228.
Leijs MM, Koppe JG, Olie K, et al. 2018. Exposure to Environmental Contaminants and Lung Function in Adolescents-Is There a Link? Int J
- + + - - + - Tier 2
Environ Res Public, Health 2018.
Li ZM, Albrecht M, Fromme H, et al. 2020. Persistent Organic Pollutants in Human Breast Milk and Associations with Maternal Thyroid Hormone
+ + + + + ++ + Tier 1
Homeostasis. Environ Sci Technol, 54:1111-1119.
Li ZM, Hernandez-Moreno D, Main KM, et al. 2018. Association of In Utero Persistent Organic Pollutant Exposure With Placental Thyroid
+ + + + + + - Tier 1
Hormones. Endocrinology, 159:3473-3481.
Liang YS, Tang Z, Jiang YS, et al. 2021. Lipid metabolism disorders associated with dioxin exposure in a cohort of Chinese male workers
+ - + + + - - Tier 2
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

revealed by a comprehensive lipidomics study. Environment International, 155:10.

Liang YS, Tang Z, Jiang YS, et al. 2020. Serum metabolic changes associated with dioxin exposure in a Chinese male cohort. Environment
+ - + + + - - Tier 2
International, 143:9.
Lim JE, Lee S, Lee S, et al. 2018. Serum persistent organic pollutants levels and stroke risk. Environ Pollut, 233:855-861.
- + ++ - + - - Tier 2

Lim JE, Nam C, Yang J, et al. 2017. Serum persistent organic pollutants (POPs) and prostate cancer risk: A case-cohort study. Int J Hyg Environ
+ ++ ++ - ++ - - Tier 1
Health, 220:849-856.
Liu X, Zhang L, Li JG, et al. 2019. Relative Effect Potency Estimates for Dioxin-Like Compounds in Pregnant Women with Gestational Diabetes
++ ++ ++ ++ + ++ ++ Tier 1
Mellitus and Blood Glucose Outcomes Based on a Nested Case-control Study. Environmental Science & Technology, 53:7792-7802.
Luong HV, Tai P, Nishijo M, et al. 2018. Association of dioxin exposure and reproductive hormone levels in men living near the Bien Hoa airbase,
+ ++ + ++ ++ ++ - Tier 1
Vietnam. Science of the Total Environment, 628:484-489.
 TABLE A4.7 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”, USING THE QUALITY ASSESSMENT TOOL
OF THE OHAT (OFFICE OF HEALTH ASSESSMENT AND TRANSLATION) (cont.)
APPENDICES

Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Mannetje AT, Eng A, Walls C, et al. 2018. Morbidity in New Zealand pesticide producers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD).
- + + ++ - ++ - Tier 2
Environ Int, 110:22-31.
Matovu H, Li ZM, Henkelmann B, et al. 2021. Multiple persistent organic pollutants in mothers' breastmilk: Implications for infant dietary
- + + + - ++ - Tier 2
exposure and maternal thyroid hormone homeostasis in Uganda, East Africa. Sci Total Environ, 770:145262.
Matta K, Vigneau E, Cariou V, et al. 2020. Associations between persistent organic pollutants and endometriosis: A multipollutant assessment
- + ++ + ++ ++ - Tier 2
using machine learning algorithms. Environ Pollut, 260:114066.
Mattonet K, Nowack-Weyers N, Vogel V, et al. 2022. Prenatal exposure to endocrine disrupting chemicals is associated with altered DNA
- + + ++ ++ ++ - Tier 2
methylation in cord blood. Epigenetics, 17(9):935-952.
McBride D, Lovelock K, Shepherd D, et al. 2016. Community exposure to hazardous site remediation in rural New Zealand: an exposed-referent
+ -- + ++ + - - Tier 2
study of serum dioxins and health effects. Aust N Z J Public Health, 40:412-417.
McBride DI, Collins JJ, Bender TJ, et al. 2018. Cohort study of workers at a New Zealand agrochemical plant to assess the effect of dioxin
- - + + + + + Tier 2
exposure on mortality. BMJ Open, 8:e019243.
Miyashita C, Araki A, Mitsui T, et al. 2018. Sex-related differences in the associations between maternal dioxin-like compounds and
++ ++ + - ++ ++ ++ Tier 1
reproductive and steroid hormones in cord blood: The Hokkaido study. Environ Int, 117:175-185.
Miyashita C, Bamai YA, Araki A, et al. 2018. Prenatal exposure to dioxin-like compounds is associated with decreased cord blood IgE and
- ++ - - ++ ++ + Tier 2
increased risk of wheezing in children aged up to 7years: The Hokkaido study. Sci Total Environ, 610:191-199.
Nghiem GT, Nishijo M, Pham TN, et al. 2019. Adverse effects of maternal dioxin exposure on fetal brain development before birth assessed by
- ++ ++ - ++ ++ - Tier 2
neonatal electroencephalography (EEG) leading to poor neurodevelopment; a 2-year follow-up study. Sci Total Environ, 667:718-729.
Nguyen ATN, Nishijo M, Pham TT, et al. 2018. Sex-specific effects of perinatal dioxin exposure on eating behavior in 3-year-old Vietnamese
- ++ + - + ++ - Tier 2
children. BMC Pediatr, 18:213.
Nwanaji-Enwerem JC, Jenkins TG, Colicino E, et al. 2020. Serum dioxin levels and sperm DNA methylation age: Findings in Vietnam war
- - - -- ++ ++ - Tier 3
veterans exposed to Agent Orange. Reproductive Toxicology, 96:27-35.
Oanh NTP, Kido T, Honma S, et al. 2018. Androgen disruption by dioxin exposure in 5-year-old Vietnamese children: Decrease in serum
+ + ++ - + ++ + Tier 1
testosterone level. Sci Total Environ, 640:466-474.
Oyama Y, Phuc HD, Honma S, et al. 2021. Decreased serum testosterone levels associated with 17 beta-hydroxysteroid dehydrogenase activity
+ + + - + ++ + Tier 1
in 7-year-old children from a dioxin-exposed area of Vietnam. Science of the Total Environment, 783:11.
Pan WY, Yin SS, Ye XQ, et al. 2019. Supporting dataset and methods for serum concentrations of selected persistent organic pollutants
-- ++ ++ + ++ ++ ++ Tier 2
measured in women with primary ovarian insufficiency. Data in Brief, 26:11.
Patel CJ, Manrai AK, Corona E, et al. 2017. Systematic correlation of environmental exposure and physiological and self-reported behaviour
+ - + - + ++ -- Tier 2
factors with leukocyte telomere length. Int J Epidemiol, 46:44-56.
Paul R, Moltó J, Ortuño N, et al. 2017. Relationship between serum dioxin-like polychlorinated biphenyls and post-testicular maturation in
+ ++ ++ + ++ ++ - Tier 1
human sperm. Reprod Toxicol, 73:312-321.
Pavuk M, Serio TC, Cusack C, et al. 2019. Hypertension in Relation to Dioxins and Polychlorinated Biphenyls from the Anniston Community
- + + - - + - Tier 2
Health Survey Follow-Up. Environ Health Perspect, 127:127007.
Pelcl T, Skrha J, Jr. Prazny M, et al. 2018. Diabetes, Cardiovascular Disorders and 2,3,7,8-Tetrachlorodibenzo-p-Dioxin Body Burden in Czech
- + + - + ++ -- Tier 2
Patients 50 Years After the Intoxication. Basic Clin Pharmacol Toxicol, 123:356-359.
Pelclova D, Navratil T, Vlckova S, et al. 2018. Exhaled breath condensate biomarkers reflect systemic changes in patients with chronic dioxin
- + + - + ++ -- Tier 2
intoxication. Monatshefte Fur Chemie, 149:1579-1586.

435
436
TABLE A4.7 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”, USING THE QUALITY ASSESSMENT TOOL
OF THE OHAT (OFFICE OF HEALTH ASSESSMENT AND TRANSLATION) (cont.)
Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Pelclova D, Urban P, Fenclova Z, et al. 2018. Neurological and Neurophysiological Findings in Workers with Chronic 2,3,7,8-Tetrachlorodibenzo-
- + + - + ++ -- Tier 2
p-Dioxin Intoxication 50 Years After Exposure. Basic Clin Pharmacol Toxicol, 122:271-277.
Petriello MC, Charnigo R, Sunkara M, et al. 2018. Relationship between serum trimethylamine N-oxide and exposure to dioxin-like pollutants.
+ + - + ++ - - Tier 2
Environ Res, 162:211-218.
Petrosino V, Motta G, Tenore G, et al. 2018. The role of heavy metals and polychlorinated biphenyls (PCBs) in the oncogenesis of head and neck
-- -- - -- - -- -- Tier 3
tumors and thyroid diseases: a pilot study. Biometals, 31:285-295.
Pham NT, Nishijo M, Nghiem TTG, et al. 2021. Effects of perinatal dioxin exposure on neonatal electroencephalography (EEG) activity of the
+ + + + + + - Tier 1
quiet sleep stage in the most contaminated area from Agent Orange in Vietnam. Int J Hyg Environ Health, 232:113661.
Pham The T, Pham Ngoc T, Hoang Van T, et al. 2020. Effects of perinatal dioxin exposure on learning abilities of 8-year-old children in Vietnam.
- - - + + - - Tier 3
Int J Hyg Environ Health, 223:132-141.
Pham TN, Nishijo M, Pham TT, et al. 2020. Dioxin exposure and sexual dimorphism of gaze behavior in prepubertal Vietnamese children living
- + - + + + - Tier 2
in Da Nang, a hot spot for dioxin contamination. Science of the Total Environment, 749:10.
Pilsner JR, Shershebnev A, Medvedeva YA, et al. 2018. Peripubertal serum dioxin concentrations and subsequent sperm methylome profiles of
- - - - - - -- Tier 3
young Russian adults. Reprod Toxicol, 78:40-49.
Rahbar MH, Swingle HM, Christian MA, et al. 2017. Environmental Exposure to Dioxins, Dibenzofurans, Bisphenol A, and Phthalates in Children
-- ++ ++ - ++ ++ -- Tier 2
with and without Autism Spectrum Disorder Living near the Gulf of Mexico. Int J Environ Res Public Health, 14(11):1425
Rytel MR, Butler R, Eliot M, et al. 2021. DNA methylation in the adipose tissue and whole blood of Agent Orange-exposed Operation Ranch
- + + -- ++ + - Tier 2
Hand veterans: a pilot study. Environ Health, 20:43.
Samer CF, Gloor Y, Rollason V, et al. 2020. Cytochrome P450 1A2 activity and incidence of thyroid disease and cancer after chronic or acute
- -- - + + + -- Tier 3
exposure to dioxins. Basic Clin Pharmacol Toxicol, 126:296-303.
Shi LL, Wang MQ, Nakayama SF, et al. 2020. The association between dioxins and steroid hormones in general adult males: a cross-sectional
+ ++ ++ + + + + Tier 1
study in an e-waste region of China. Environ Sci Pollut Res Int, 27:26511-26519.
Slama N, Warner M, Mocarelli P, et al. 2019. The 2nd to 4th digit length ratio (2D:4D) among children of Seveso women exposed to
- - -- + + ++ + Tier 3
2,3,7,8-tetrachlorodibenzo-p-dioxin. Early Hum Dev, 131:45-50.
Sun XL, Kido T, Honma S, et al. 2017. The relationship between dioxins exposure and risk of prostate cancer with steroid hormone and age in
+ + ++ ++ + ++ - Tier 1
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Vietnamese men. Sci Total Environ, 595:842-848.

Sun XL, Okamoto R, Kido T, et al. 2020. Association of dioxin in maternal breast milk and salivary steroid hormone levels in preschool children:
- ++ + - + + - Tier 2
A five-year follow-up study of a Vietnam cohort. Chemosphere, 241:124899.
Tai PT, Nishijo M, Nghi TN, et al. 2016. Effects of Perinatal Dioxin Exposure on Development of Children during the First 3 Years of Life.
- + + - + + - Tier 2
J Pediatr, 175:159-166.e152.
Van Luong H, Tai PT, Nishijo M, et al. 2018. Association of dioxin exposure and reproductive hormone levels in men living near the Bien Hoa
- ++ + + + + - Tier 2
airbase, Vietnam. Sci Total Environ, 628:484-489.
Vermeir G, Covaci A, Van Larebeke N, et al. 2021. Neurobehavioural and cognitive effects of prenatal exposure to organochlorine compounds in
+ + - - + + - Tier 2
three year old children. BMC Pediatr, 21:99.
Vu HT, Nishijo M, Pham TN, et al. 2021. Effects of perinatal dioxin exposure on mirror neuron activity in 9-year-old children living in a hot spot
- + - ++ - + - Tier 2
of dioxin contamination in Vietnam. Neuropsychologia, 161:108001.
Wang Z, Hang JG, Feng H, et al. 2019. Effects of perinatal dioxin exposure on development of children: a 3-year follow-up study of China
- - ++ ++ - + - Tier 2
cohort. Environ Sci Pollut Res Int, 26:20780-20786.
 TABLE A4.7 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FOR THE THEME “TOXIC EFFECTS OF DIOXINS AND dl-PCBs”, USING THE QUALITY ASSESSMENT TOOL
OF THE OHAT (OFFICE OF HEALTH ASSESSMENT AND TRANSLATION) (cont.)
APPENDICES

Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Warner M, Rauch S, Ames J, et al. 2019. In utero dioxin exposure and cardiometabolic risk in the Seveso Second Generation Study. Int J Obes
+ + + + + + + Tier 1
(Lond), 43:2233-2243.
Warner M, Rauch S, Ames J, et al. 2020. Prenatal dioxin exposure and thyroid hormone levels in the Seveso second generation study. Environ
+ + + + + + + Tier 1
Res, 183:109280.
Warner M, Rauch S, Brambilla P, et al. 2020. Prenatal dioxin exposure and glucose metabolism in the Seveso Second Generation study. Environ
+ + + + + + + Tier 1
Int, 134:105286.
Xu P, Chen Z, Wu L, et al. 2019. Health risk of childhood exposure to PCDD/Fs emitted from a municipal waste incinerator in Zhejiang, China.
-- + + + + + - Tier 2
Sci Total Environ, 689:937-944.
Ye M, Warner M, Mocarelli P, et al. 2018. Prenatal exposure to TCDD and atopic conditions in the Seveso second generation: a prospective
- - + ++ + + + Tier 2
cohort study. Environ Health, 17:22.
Zhang Z, He J, Shi T, et al. 2019. Associations between polychlorinated dibenzo-dioxins and polychlorinated dibenzo-furans exposure and
- - + + + + - Tier 2
oxidatively generated damage to DNA and lipid. Chemosphere, 227:237-246.
Zhang Z, Zhou M, He J, et al. 2020. Polychlorinated dibenzo-dioxins and polychlorinated dibenzo-furans exposure and altered lung function:
- - + + + + - Tier 2
The mediating role of oxidative stress. Environ Int, 137:105521.

437
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

APPENDIX 5
TOXIC EFFECTS OF MeHg

LITERATURE SEARCH STRATEGY

TABLE A5.1 LITERATURE SEARCH STRATEGY FOR THE REVIEW “TOXIC EFFECTS OF MeHg” IN PUBMED AND WEB OF SCIENCE
Database: PubMed
Date of literature search: 15 December 2021
Literature search string:
((((((((("journal article"[Publication Type]) OR "review"[Publication Type]) OR "scientific integrity review"[Publication Type]) OR "meta analysis"[Publication Type]) OR "systematic
review"[Publication Type])) AND ("2010/01/01"[PDat] : "2022/12/31"[PDat])) AND Humans[Mesh]) AND (("methylmercury compounds"[MeSH Terms]) OR (methylmercury[Title/
Abstract] OR MeHg[Title/Abstract] OR methyl-Hg[Title/Abstract] OR CH3Hg[Title/Abstract] OR mercury[MeSH Terms] OR mercury[Title/Abstract]) AND ("2010/01/01"[PDat] :
"2022/12/31"[PDat]))) AND (epidemiolog*[Title/Abstract] OR "cohort study"[Title/Abstract] OR "cohort studies"[Title/Abstract] OR "case control study"[Title/Abstract] OR "case control
studies"[Title/Abstract] OR "adverse effect"[Title/Abstract] OR "adverse effects"[Title/Abstract] OR "observational study"[Title/Abstract] OR "observational studies"[Title/Abstract] OR
"case series"[Title/Abstract] OR "cross sectional study"[Title/Abstract] OR "cross sectional studies"[Title/Abstract] OR "case report"[Title/Abstract]OR "case reports"[Title/Abstract])
AND (english[Language])
Total hits: 928
Database: Web of Science
Date of literature search: 15 December 2021
Literature search string:
TS=((methylmercury OR MeHg OR methyl-Hg OR CH3Hg OR CH3Hg OR mercur*) AND (epidemiolog* OR "cohort stud*" OR "case control stud*" OR "adverse effect*" OR
"observational stud*" OR "case serie*" OR "case report*" OR "cross sectional stud*")) AND DOP=(2010-01-01/2022-12-31)
Search filters: Including only document types (articles and review articles)
Total hits: 1 834

438
APPENDICES

TABLE A5.2 UPDATED LITERATURE SEARCH STRATEGY FOR THE REVIEW “TOXIC EFFECTS OF MeHg” IN PUBMED AND WEB
OF SCIENCE
Database: PubMed
Date of literature search: 14 November 2022
Literature search string:
((((((((("journal article"[Publication Type]) OR "review"[Publication Type]) OR "scientific integrity review"[Publication Type]) OR "meta analysis"[Publication Type]) OR "systematic
review"[Publication Type]) ) AND ("2010/01/01"[PDat] : "2021/12/15"[PDat])) AND Humans[Mesh]) AND (("methylmercury compounds"[MeSH Terms]) OR (methylmercury[Title/
Abstract] OR MeHg[Title/Abstract] OR methyl-Hg[Title/Abstract] OR CH3Hg[Title/Abstract] OR mercury[MeSH Terms] OR mercury[Title/Abstract]) AND ("2010/01/01"[PDat] :
"2021/12/15"[PDat]))) AND (epidemiolog*[Title/Abstract] OR "cohort study"[Title/Abstract] OR "cohort studies"[Title/Abstract] OR "case control study"[Title/Abstract] OR "case control
studies"[Title/Abstract] OR "adverse effect"[Title/Abstract] OR "adverse effects"[Title/Abstract] OR "observational study"[Title/Abstract] OR "observational studies"[Title/Abstract] OR
"case series"[Title/Abstract] OR "cross sectional study"[Title/Abstract] OR "cross sectional studies"[Title/Abstract] OR "case report"[Title/Abstract] OR "case reports"[Title/Abstract]
OR "systematic review"[Title/Abstract] OR "meta anal*"[Title/Abstract] OR "meta-anal*"[Title/Abstract]) AND (english[Language])
Search filters: Include only article types (reviews, systematic reviews and meta-analysis)
Total hits: 242
Database: Web of Science
Date of literature search: 14 November 2022
Literature search string:
TS=((methylmercury OR MeHg OR methyl-Hg OR CH3Hg OR CH3Hg OR mercur*) AND (epidemiolog* OR "cohort stud*" OR "case control stud*" OR "adverse effect*" OR "observational
stud*" OR "case serie*" OR "case report*" OR "cross sectional stud*" OR "systematic review" OR "meta anal*" OR "meta-anal*")) AND DOP=(2010-01-01/2021-12-15)
Search filters: Including only document types (review articles)
Total hits: 330

439
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

RECORDS EXCLUDED DURING FULL-TEXT SCREENING

TABLE A5.3 SYSTEMATIC REVIEWS EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECTS OF MeHg”
BASED ON INCLUSION AND EXCLUSION CRITERIA
Study (n = 9) Reason for exclusion
Feng, L. Li, P. & Feng, X. 2021. Methylmercury bioaccumulation in rice and health effects: Excluded based on inclusion and exclusion criteria: not a systematic review.
A systematic review. Current Opinion in Environmental Science & Health, 23:100285:
Grandjean, P. & Landrigan, P.J. 2014. Neurobehavioural effects of developmental toxicity. Excluded based on inclusion and exclusion criteria: not a systematic review.
The lancet neurology, 13(3):330-338.
Ke, T. Tinkov, A.A. Skalny, A.V. Bowman, A.B. Rocha, J.B. Santamaria, A. & Aschner, Excluded based on inclusion and exclusion criteria: not a systematic review.
M. 2021. Developmental exposure to methylmercury and ADHD, a literature review of
epigenetic studies. Environmental Epigenetics, 7(1):dvab014.
Perez-Fernandez, C. Flores, P. & Sánchez-Santed, F. 2019. A systematic review on the The systematic review also included animal studies and the results or discussion do not
influences of neurotoxicological xenobiotic compounds on inhibitory control. Frontiers in distinguish between animal and human studies.
Behavioral Neuroscience, 13:139.
Sheehan, M.C. Burke, T.A. Navas-Acien, A. Breysse, P.N. McGready, J. & Fox, M.A. 2014. Excluded based on inclusion and exclusion criteria: no health outcome included.
Global methylmercury exposure from seafood consumption and risk of developmental
neurotoxicity: a systematic review. Bulletin of the World Health Organization, 92:254-
269F.
Spiller, P. Hibbeln, J.R. Myers, G. Vannice, G. Golding, J. Crawford, M.A., Carlson, S.E. et Excluded based on inclusion and exclusion criteria: not a systematic review.
al. 2019. An abundance of seafood consumption studies presents new opportunities
to evaluate effects on neurocognitive development. Prostaglandins, Leukotrienes and
Essential Fatty Acids, 151:8-13.
Vianna, A.D.S. Matos, E.P.D. Jesus, I.M.D. Asmus, C.I.R.F. & Câmara, V.D.M. 2019. Human Excluded based on inclusion and exclusion criteria: The systematic review had included
exposure to mercury and its hematological effects: a systematic review. Cad Saude n = 80 studies, where of n = 48 of these are case-reports (a study type that is an
Publica, 11;35(2) exclusion criteria). Also, of the 80 included studies, 75 were graded as weak in the
quality assessment.
Wang, M.D. Little, J. Gomes, J. Cashman, N.R. & Krewski, D. 2017. Identification of risk Excluded based on inclusion and exclusion criteria: In the section regarding Hg, no
factors associated with onset and progression of amyotrophic lateral sclerosis using specific human studies are included. They have only mentioned Hg in the discussion and
systematic review and meta-analysis. Neurotoxicology, 61:101-130. this refers to case-reports.
Zheng, L. Y. Sanders, A. P. Saland, J. M. Wright, R. O. & Arora, M. 2017. Environmental Excluded based on inclusion and exclusion criteria: Have included n = 5 studies
exposures and pediatric kidney function and disease: A systematic review. Environmental assessing Hg, but all of these compare Hg exposure from amalgam vs. composite use in
research, 158:625-648. RCTs. Not relevant exposure (as per criteria).

440
APPENDICES

TABLE A5.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECTS OF MeHg” BASED ON
INCLUSION AND EXCLUSION CRITERIA
Study (n = 45) Reason for exclusion
Al-Saleh, I. 2021. Health risk assessment of trace metals through breast milk Excluded based on inclusion and exclusion criteria: no health outcome measured
consumption in Saudi Arabia. Biological Trace Element Research, 199(12):4535-4545.
Aranda, N. Valls, R.M. Romeu, M. Sánchez-Martos, V. Albaladejo, R. Fernández-Castillejo, Excluded based on inclusion and exclusion criteria: no Hg measurements in participants
S., Giralt, M. et al. 2017. Consumption of seafood and its estimated heavy metals are
associated with lipid profile and oxidative lipid damage on healthy adults from a Spanish
Mediterranean area: A cross-sectional study. Environmental research, 156:644-651.
Bellinger, D.C. 2012. Comparing the population neurodevelopmental burdens Excluded based on inclusion and exclusion criteria: individual health outcome not
associated with children's exposures to environmental chemicals and other risk factors. measured (only population based).
Neurotoxicology, 33(4):641-643.
Bellinger, D.C. 2012. A strategy for comparing the contributions of environmental Excluded based on inclusion and exclusion criteria: a review, but not a systematic review
chemicals and other risk factors to neurodevelopment of children. Environmental Health or meta-analysis.
Perspectives, 120(4):501-507.
Berky, A.J., Ryde, I.T., Feingold, B., Ortiz, E.J. Wyatt, L.H., Weinhouse, C., Pan, W.K. et al. Excluded based on inclusion and exclusion criteria: no health outcome measured.
2019. Predictors of mitochondrial DNA copy number and damage in a mercury‐exposed
rural Peruvian population near artisanal and small‐scale gold mining: An exploratory
study. Environmental and molecular mutagenesis, 60(2):197-210.
Budtz‐Jørgensen, E. Debes, F., Weihe, P. & Grandjean, P. 2010. Structural equation models Excluded based on inclusion and exclusion criteria: statistical method paper
for meta‐analysis in environmental risk assessment. Environmetrics, 21(5):510-527.
Butler, L.J., Janulewicz, P.A., Carwile, J.L., White, R.F. Winter, M.R. & Aschengrau, A. 2017. Excluded based on inclusion and exclusion criteria: no Hg measurements
Childhood and adolescent fish consumption and adult neuropsychological performance:
An analysis from the Cape Cod Health Study. Neurotoxicology and teratology, 61:47-57.
Butts, C.D., Bloom, M.S., McGough, A., Lenhart, N., Wong, R. Mok-Lin, E., Fujimoto, Excluded based on inclusion and exclusion criteria: no health outcome measured
V.Y. et al. 2020. Seafood consumption is associated with higher follicular fluid arsenic
(As) and mercury (Hg) concentrations in women undergoing in vitro fertilization (IVF).
Environmental Research, 188:109753.
Carneiro, M.F.H., Grotto, D. & Barbosa Jr, F. 2014. Inorganic and methylmercury levels Excluded based on inclusion and exclusion criteria: no health outcome measured (only
in plasma are differentially associated with age, gender, and oxidative stress markers oxidative stress)
in a population exposed to mercury through fish consumption. Journal of Toxicology and
Environmental Health, Part A, 77(1-3):69-79.
Carwile, J.L., Butler, L.J., Janulewicz, P.A., Winter, M.R. & Aschengrau, A. 2016. Childhood Excluded based on inclusion and exclusion criteria: no assessment of Hg exposure
fish consumption and learning and behavioral disorders. International journal of
environmental research and public health, 13(11):1069.
Chevrier, C., Warembourg, C., Gaudreau, E., Monfort, C., Le Blanc, A., Guldner, L. Excluded based on inclusion and exclusion criteria: Hg only included as a co-exposure
& Cordier, S. 2013. Organochlorine pesticides, polychlorinated biphenyls, seafood in paper
consumption, and time-to-pregnancy. Epidemiology, 24(2):251-260.
Davidson, P.W., Myers, G.J. & Shamlaye, C. 2020. Principles of studying low-level Excluded based on inclusion and exclusion criteria: method paper using the Seychelles
neurotoxic exposures in children: using the Seychelles Child Development Study of methyl study as an example.
mercury as a prototype. NeuroToxicology, 81:307-314.
de Marco, K.C., Braga, G.U. & Barbosa Jr, F. 2011. Determination of the effects of eNOS Excluded based on inclusion and exclusion criteria: no health outcome included (only
gene polymorphisms (T-786C and Glu298Asp) on nitric oxide levels in a methylmercury- nitric oxide levels)
exposed population. Journal of Toxicology and Environmental Health, Part A, 74(20):1323-
1333.
Diaz, S.M., Palma, R. M., Muñoz, M. N., Becerra-Arias, C. & Fernández Niño, J. A. 2020. Excluded based on inclusion and exclusion criteria: Hg exposure from gold mining areas
Factors associated with high mercury levels in women and girls from the Mojana region,
Colombia, 2013–2015. International Journal of Environmental Research and Public
Health, 17(6):1827.
Dos Santos Freitas, J., Lacerda, E.M., da Silva Martins, I.C.V., Rodrigues Jr, D., Bonci, Excluded based on inclusion and exclusion criteria: Hg exposure from contaminated
D.M., Cortes, M.I., da Silva Souza, G. et al. 2018. Cross-sectional study to assess the areas
association of color vision with mercury hair concentration in children from Brazilian
Amazonian riverine communities. Neurotoxicology, 65:60-67.
Engström, K.S., Wennberg, M., Strömberg, U., Bergdahl, I. A., Hallmans, G., Jansson, Excluded based on inclusion and exclusion criteria: association between Hg exposure and
J.H., Broberg, K. et al. 2011. Evaluation of the impact of genetic polymorphisms health outcome not measured
in glutathione-related genes on the association between methylmercury or n-3
polyunsaturated long chain fatty acids and risk of myocardial infarction: a case-control
study. Environmental health, 10(1):1-8.
Franken, C., Koppen, G., Lambrechts, N., Govarts, E., Bruckers, L., Den Hond, E., Excluded based on inclusion and exclusion criteria: health outcome not measured (only
Schoeters, G. et al. 2017. Environmental exposure to human carcinogens in teenagers DNA damage)
and the association with DNA damage. Environmental research, 152:165-174.
Ginsberg, G., Sonawane, B., Nath, R. & Lewandowski, P. 2014. Methylmercury-induced Excluded based on inclusion and exclusion criteria: not a human study (only mechanistic)
inhibition of paraoxonase-1 (PON1)—Implications for cardiovascular risk. Journal of
Toxicology and Environmental Health, Part A, 77(17):1004-1023.
Golding, J., Steer, C.D., Hibbeln, J.R., Emmett, P.M., Lowery, T. & Jones, R. 2013. Dietary Excluded based on inclusion and exclusion criteria: health outcome not measured
predictors of maternal prenatal blood mercury levels in the ALSPAC birth cohort study.
Environmental health perspectives, 121(10):1214-1218.

441
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A5.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECTS OF MeHg” BASED ON
INCLUSION AND EXCLUSION CRITERIA (cont.)
Study (n = 45) Reason for exclusion
Gump, B.B., MacKenzie, J.A., Dumas, A.K., Palmer, C., D. Parsons, P.J., Segu, Z.M., Excluded based on inclusion and exclusion criteria: health outcome not measured
Bendinskas, K.G. et al. 2012. Fish consumption, low-level mercury, lipids, and
inflammatory markers in children. Environmental research, 112:204-211.
Hara, N., Saito, H., Takahashi, K. & Takeda, M. 2013. Lower urinary tract symptoms in Excluded based on inclusion and exclusion criteria: Hg exposure not measured
patients with Niigata Minamata disease: A case–control study 50 years after methyl
mercury pollution. International journal of urology, 20(6):610-615.
Inoue, S., Yorifuji, T., Tsuda, T. & Doi, H. 2012. Short-term effect of severe exposure to Excluded based on inclusion and exclusion criteria: Hg exposure not measured
methylmercury on atherosclerotic heart disease and hypertension mortality in Minamata.
Science of the total environment, 417:291-293.
Julvez, J., Davey Smith, G., Ring, S. & Grandjean, P. 2019. A birth cohort study on the Excluded based on inclusion and exclusion criteria: Hg exposure in blood, hair or toenails
genetic modification of the association of prenatal Methylmercury with child cognitive not measured (only cord tissue)
development. American journal of epidemiology, 188(10):1784-1793.
Kim, H., Lee, J., Woo, H.D., Kim, D.W., Oh, J.H., Chang, H.J., Kim, J. et al. 2020. Dietary Excluded based on inclusion and exclusion criteria: Hg exposure not measured directly
mercury intake and colorectal cancer risk: A case-control study. Clinical Nutrition, in participants.
39(7):2106-2113.
Kong, H.K., Gan, C.F., Xiong, M., Kwok, K.W.H., Lui, G.C.S., Li, P., Lo, S.C.L. et al. 2019. Excluded based on inclusion and exclusion criteria: health outcome not measured
Chronic methylmercury exposure induces production of prostaglandins: evidence from
a population study and a rat dosing experiment. Environmental Science & Technology,
53(13):7782-7791.
Kuras, R. Kozlowska, L. Reszka, E. Wieczorek, E. Jablonska, E. Gromadzinska, J., Excluded based on inclusion and exclusion criteria: health outcome not measured (only
Wasowicz, W. et al. 2019. Environmental mercury exposure and selenium-associated biomarker)
biomarkers of antioxidant status at molecular and biochemical level. A short-term
intervention study. Food and Chemical Toxicology, 130:187-198.
Lee, P.H. & Burstyn, I. 2016. Identification of confounder in epidemiologic data Excluded based on inclusion and exclusion criteria: statistical method paper
contaminated by measurement error in covariates. BMC medical research methodology,
16:1-18.
Leung, Y.K., Ouyang, B., Niu, L., Xie, C., Ying, J., Medvedovic, M., Ho, S.M. 2018. Excluded based on inclusion and exclusion criteria: health outcome not measured (only
Identification of sex-specific DNA methylation changes driven by specific chemicals in DNA methylation)
cord blood in a Faroese birth cohort. Epigenetics, 13(3):290-300.
Little, M., Achouba, A., Dumas, P., Ouellet, N., Ayotte, P. & Lemire, M. 2019. Determinants Excluded based on inclusion and exclusion criteria: health outcome not measured (focus
of selenoneine concentration in red blood cells of Inuit from Nunavik (Northern Québec, on selenium)
Canada). Environment international, 127:243-252.
Maeda, E., Murata, K., Kumazawa, Y., Sato, W., Shirasawa, H., Iwasawa, T., Terada, Y. et Excluded based on inclusion and exclusion criteria: health outcome not measured (focus
al. 2019. Associations of environmental exposures to methylmercury and selenium with on selenium)
female infertility: A case–control study. Environmental research, 168:357-363.
Monastero, R.N., Karimi, R., Nyland, J.F., Harrington, J., Levine, K. & Meliker, J.R. 2017. Excluded based on inclusion and exclusion criteria: health outcome not measured
Mercury exposure, serum antinuclear antibodies, and serum cytokine levels in the Long
Island Study of Seafood Consumption: A cross-sectional study in NY, USA. Environmental
research, 156:334-340.
Morris, M.C., Brockman, J., Schneider, J.A., Wang, Y., Bennett, D.A., Tangney, C.C. & van Excluded based on inclusion and exclusion criteria: exposure in blood, hair or toenails not
de Rest, O. 2016. Association of seafood consumption, brain mercury level, and APOE ε4 measured (Hg measured in the brain during autopsy)
status with brain neuropathology in older adults. JAMA, 315(5):489-497.
Nyland, J.F., Wang, S.B., Shirley, D.L., Santos, E.O., Ventura, A.M., de Souza, J.M. & Excluded based on inclusion and exclusion criteria: health outcome not measured (only
Silbergeld, E.K. 2011. Fetal and maternal immune responses to methylmercury exposure: biomarker)
a cross-sectional study. Environmental Research, 111(4):584-589.
Nyland, J.F., Fillion, M., Barbosa Jr, F., Shirley, D.L., Chine, C., Lemire, M., Silbergeld, E.K. Excluded based on inclusion and exclusion criteria: health outcome not measured (only
et al. 2011. Biomarkers of methylmercury exposure immunotoxicity among fish consumers biomarker)
in Amazonian Brazil. Environmental Health Perspectives, 119(12):1733-1738.
Sá, A.B., Simone, C., Oliveira, C.S.B.D., Lima, A.A.D.S., Borges, B.E.S., Santos, G.D.F.S., Excluded based on inclusion and exclusion criteria: health outcome not measured
Pinheiro, M.D.C.N. et al. 2019. Fish consumption frequency and lipid peroxidation in the
riverside population of Lower Tocantins, Pará. Nutrición Clínica y Dietética Hospitalaria,
39(1): 64-68
Schoeman, K., Tanaka, T., Bend, J.R. & Koren, G. 2010. Hair mercury levels of women of Excluded based on inclusion and exclusion criteria: health outcome not measured
reproductive age in Ontario, Canada: implications to fetal safety and fish consumption.
The Journal of Pediatrics, 157(1):127-131.
Sheehan, M.C. Burke, T.A. Breysse, P.N. Navas-Acien, A. McGready, J. & Fox, M.A. Excluded based on inclusion and exclusion criteria: Hg exposure only included as a
2012. Association of markers of chronic viral hepatitis and blood mercury levels confounding variable
in US reproductive-age women from NHANES 2001–2008: a cross-sectional study.
Environmental Health, 11(1):1-11.
Vejrup, K., Brantsæter, A.L., Knutsen, H.K., Magnus, P., Alexander, J., Kvalem, H.E., Excluded based on inclusion and exclusion criteria: Hg exposure only calculated from
Haugen, M. et al. 2014. Prenatal mercury exposure and infant birth weight in the intake, not biomarker exposure
Norwegian Mother and Child Cohort Study. Public health nutrition, 17(9):2071-2080.

442
APPENDICES

TABLE A5.4 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING FOR THE THEME “TOXIC EFFECTS OF MeHg” BASED ON
INCLUSION AND EXCLUSION CRITERIA (cont.)
Study (n = 45) Reason for exclusion
Vejrup, K., Schjølberg, S., Knutsen, H.K., Kvalem, H.E., Brantsæter, A.L., Meltzer, H.M., Excluded based on inclusion and exclusion criteria: Hg exposure only calculated from
Haugen, M. et al. 2016. Prenatal methylmercury exposure and language delay at three intake, not biomarker exposure
years of age in the Norwegian Mother and Child Cohort Study. Environment international,
92, 63-69.
Wang, Y. Goodrich, J. M., Werner, R., Gillespie, B., Basu, N. & Franzblau, A. 2013. Excluded based on inclusion and exclusion criteria: Hg exposure only included as a
Relationship of estimated dietary intake of n-3 polyunsaturated fatty acids from fish confounding variable
with peripheral nerve function after adjusting for mercury exposure. Science of the total
environment, 454, 73-78.
Yeter, D., Portman, M.A., Aschner, M., Farina, M., Chan, W.C., Hsieh, K.S. & Kuo, H.C. Excluded based on inclusion and exclusion criteria: Hg exposure not measured
2016. Ethnic Kawasaki disease risk associated with blood mercury and cadmium in US
children. International journal of environmental research and public health, 13(1):101.
Yorifuji, T., Tsuda, T., Inoue, S., Takao, S., Harada, M. & Kawachi, I. 2013. Critical Excluded based on inclusion and exclusion criteria: Hg exposure not measured
appraisal of the 1977 diagnostic criteria for Minamata disease. Archives of
Environmental & Occupational Health, 68(1):22-29.
Yorifuji, T. & Tsuda, T. 2016. Epidemiological studies of neurological signs and symptoms Excluded based on inclusion and exclusion criteria: Hg exposure not measured
and blood pressure in populations near the industrial methylmercury contamination at
Minamata, Japan. Archives of environmental & occupational health, 71(4):231-236.
Zeilmaker, M.J., Hoekstra, J., van Eijkeren, J.C., de Jong, N., Hart, A., Kennedy, M., Excluded based on inclusion and exclusion criteria: Hg exposure only calculated from
Gunnlaugsdottir, H. 2013. Fish consumption during child bearing age: A quantitative intake, not biomarker exposure
risk–benefit analysis on neurodevelopment. Food and Chemical Toxicology, 54, 30-34.
Zijlmans, W., Wickliffe, J., Hindori-Mohangoo, A., MacDonald-Ottevanger, S., Ouboter, Excluded based on inclusion and exclusion criteria: Description of an ongoing study
P., Landburg, G., Lichtveld, M. et al. 2020. Caribbean Consortium for Research in
Environmental and Occupational Health (CCREOH) Cohort Study: Influences of complex
environmental exposures on maternal and child health in Suriname. BMJ open,
10(9):e034702.

443
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A5.5 PRIMARY STUDIES EXCLUDED AS THEY HAD ALREADY BEEN ASSESSED IN ONE OF THE SYSTEMATIC REVIEWS FROM
THE LITERATURE SEARCH “TOXIC EFFECT OF MeHg”
Primary studies assessed (n = 44) Primary study assessed in systematic review
Barbone, F., Rosolen, V., Mariuz, M., Parpinel, M., Casetta, A., Sammartano, F., Horvat, Excluded, as the primary study has already been assessed in an included systematic
M. et al. 2019. Prenatal mercury exposure and child neurodevelopment outcomes at 18 review (Hibbeln et al., 2019)
months: Results from the Mediterranean PHIME cohort. International journal of hygiene
and environmental health, 222(1):9-21.
Bulka, C.M., Persky, V.W., Daviglus, M.L., Durazo-Arvizu, R.A. & Argos, M. 2019. Multiple Excluded, as the primary study has already been assessed in an included systematic
metal exposures and metabolic syndrome: A cross-sectional analysis of the National review (Xu et al., 2021)
Health and Nutrition Examination Survey 2011–2014. Environmental research, 168:397-
405.
Deroma, L., Parpinel, M., Tognin, V., Channoufi, L., Tratnik, J., Horvat, M., Barbone, F. Excluded, as the primary study has already been assessed in an included systematic
2013. Neuropsychological assessment at school-age and prenatal low-level exposure to review (Hibbeln et al., 2019)
mercury through fish consumption in an Italian birth cohort living near a contaminated
site. International journal of hygiene and environmental health, 216(4):486-493.
Downer, M.K., Martínez-González, M.A., Gea, A., Stampfer, M., Warnberg, J., Ruiz- Excluded, as the primary study has already been assessed in an included systematic
Canela, M., PREDIMED Study Investigators et al. 2017. Mercury exposure and risk of review (Chowdhury et al. 2019 and Hu et al. 2018)
cardiovascular disease: a nested case-control study in the PREDIMED (PREvention with
MEDiterranean Diet) study. BMC Cardiovascular Disorders, 17:1-11.
Ettinger, A.S., Bovet, P., Plange-Rhule, J., Forrester, T.E., Lambert, E.V., Lupoli, N., Luke, Excluded, as the primary study has already been assessed in an included systematic
A. et al. 2014. Distribution of metals exposure and associations with cardiometabolic review (Roy et al., 2017)
risk factors in the “Modeling the Epidemiologic Transition Study”. Environmental Health,
131-15.
Giacoppo, S., Galuppo, M., Calabr✖, R.S., D’Aleo, G., Marra, A., Sessa, E., Mazzon, E. et al. Excluded, as the primary study has already been assessed in an included systematic
2014. Heavy metals and neurodegenerative diseases: an observational study. Biological review (Sarihi et al., 2021)
trace element research, 161:151-160.
Golding, J., Hibbeln, J.R., Gregory, S.M., Iles-Caven, Y., Emond, A. & Taylor, C.M. 2017. Excluded, as the primary study has already been assessed in an included systematic
Maternal prenatal blood mercury is not adversely associated with offspring IQ at 8 years review (Hibbeln et al., 2019)
provided the mother eats fish: a British prebirth cohort study. International journal of
hygiene and environmental health, 220(7):1161-1167.
Golding, J., Rai, D., Gregory, S., Ellis, G., Emond, A., Iles-Caven, Y., Taylor, C. et al. 2018. Excluded, as the primary study has already been assessed in an included systematic
Prenatal mercury exposure and features of autism: a prospective population study. review (Hibbeln et al., 2019)
Molecular autism, 9:1-9.
Govarts, E., Remy, S., Bruckers, L., Den Hond, E., Sioen, I., Nelen, V., Schoeters, G. 2016. Excluded, as the primary study has already been assessed in an included systematic
Combined effects of prenatal exposures to environmental chemicals on birth weight. review (Dack et al., 2021)
International journal of environmental research and public health, 13(5):495.
Gregory, S., Iles-Caven, Y., Hibbeln, J.R., Taylor, C.M. & Golding, J. 2016. Are prenatal Excluded, as the primary study has already been assessed in an included systematic
mercury levels associated with subsequent blood pressure in childhood and adolescence? review (Gallego-Vinas et al., 2019)
The Avon prebirth cohort study. BMJ open, 6(10):e012425.
Gustin, K., Barman, M., Stråvik, M., Levi, M., Englund-Ögge, L., Murray, F., Kippler, M. Excluded, as the primary study has already been assessed in an included systematic
et al. 2020. Low-level maternal exposure to cadmium, lead, and mercury and birth review (Dack et al., 2021)
outcomes in a Swedish prospective birth-cohort. Environmental Pollution, 2651:14986.
Hu, Y., Chen, L., Wang, C., Zhou, Y., Zhang, Y., Wang, Y., Tian, Y. et al. 2016. Prenatal Excluded, as the primary study has already been assessed in an included systematic
low-level mercury exposure and infant neurodevelopment at 12 months in rural northern review (Hibbeln et al., 2019)
China. Environmental Science and Pollution Research, 23:12050-12059.
Hu, X.F., Laird, B.D., & Chan, H.M., 2017. Mercury diminishes the cardiovascular Excluded, as the primary study has already been assessed in an included systematic
protective effect of omega-3 polyunsaturated fatty acids in the modern diet of Inuit in review (Hu et al., 2018 and Hu et al., 2021)
Canada. Environmental research, 152470-477.
Jeppesen, C., Valera, B., Nielsen, N.O., Bjerregaard, P. & Jørgensen, M.E. 2015. Excluded, as the primary study has already been assessed in an included systematic
Association between whole blood mercury and glucose intolerance among adult Inuit in review (Roy et al., 2017 and Xu et al., 2021)
Greenland. Environmental research, 143192-197.
Julvez, J., Méndez, M., Fernandez-Barres, S., Romaguera, D., Vioque, J., Llop, S., Sunyer, Excluded, as the primary study has already been assessed in an included systematic
J. 2016. Maternal consumption of seafood in pregnancy and child neuropsychological review (Hibbeln et al., 2019)
development: a longitudinal study based on a population with high consumption levels.
American journal of epidemiology, 183(3):169-182.
Kalish, B.T., Rifas-Shiman, S.L., Wright, R.O., Amarasiriwardena, C.J., Jayawardene, Excluded, as the primary study has already been assessed in an included systematic
I., Gillman, M.W., Oken, E. 2014. Associations of prenatal maternal blood mercury review (Hu et al., 2018 and Gallego-Vinas et al., 2019)
concentrations with early and mid-childhood blood pressure: a prospective study.
Environmental research, 133:327-333.
Kim, K.N., Bae, S., Park, H.Y., Kwon, H.J. & Honga, Y.C. 2015. Low-level mercury exposure Excluded, as the primary study has already been assessed in an included systematic
and risk of asthma in school-age children. Epidemiology, 26(5):733-739. review (Roy et al., 2017)
Kim, B.M., Chen, M.H., Chen, P.C., Park, H., Ha, M., Kim, Y., Ha, E.H. et al. 2017. Path Excluded, as the primary study has already been assessed in an included systematic
analysis of prenatal mercury levels and birth weights in Korean and Taiwanese birth review (Dack et al., 2021)
cohorts. Science of the Total Environment, 605:1003-1010.
Lee, Y.J. & Hwang, I.C. 2014. Relationship between serum ferritin level and blood mercury Excluded, as the primary study has already been assessed in an included systematic
concentration using data from the Korean national health and nutrition examination review (Roy et al., 2017 and Xu et al., 2021)
survey (2010–2012). Environmental research, 135:271-275.

444
APPENDICES

TABLE A5.5 PRIMARY STUDIES EXCLUDED AS THEY HAD ALREADY BEEN ASSESSED IN ONE OF THE SYSTEMATIC REVIEWS FROM
THE LITERATURE SEARCH “TOXIC EFFECT OF MeHg” (cont.)
Primary studies assessed (n = 44) Primary study assessed in systematic review
Llop, S., Ballester, F., Murcia, M., Forns, J., Tardon, A., Andiarena, A., Lopez-Espinosa, Excluded, as the primary study has already been assessed in an included systematic
M.J. et al. 2017. Prenatal exposure to mercury and neuropsychological development review (Hibbeln et al., 2019)
in young children: the role of fish consumption. International Journal of Epidemiology,
46(3):827-838.
Llop, S., Tran, V., Ballester, F., Barbone, F., Sofianou-Katsoulis, A., Sunyer, J., Broberg, K. Excluded, as the primary study has already been assessed in an included systematic
et al. 2017. CYP3A genes and the association between prenatal methylmercury exposure review (Hibbeln et al., 2019)
and neurodevelopment. Environment international, 105:34-42.
Marques, R.C. Bernardi, J.V. Dórea, J.G. Brandão, K.G. Bueno, L. Leão, R.S. & Malm, O. Excluded, as the primary study has already been assessed in an included systematic
2013. Fish consumption during pregnancy, mercury transfer, and birth weight along the review (Dack et al., 2021)
Madeira River Basin in Amazonia. International journal of environmental research and
public health, 10(6):2150-2163.
McKean, S.J., Bartell, S.M., Hansen, R.L., Barfod, G.H., Green, P.G. & Hertz-Picciotto, Excluded, as the primary study has already been assessed in an included systematic
I. 2015. Prenatal mercury exposure, autism, and developmental delay, using review (Jafari et al., 2017)
pharmacokinetic combination of newborn blood concentrations and questionnaire data: a
case control study. Environmental Health, 14(1):1-12.
Mozaffarian, D., Shi, P., Morris, J.S., Spiegelman, D., Grandjean, P., Siscovick, DS. & Excluded, as the primary study has already been assessed in an included systematic
Rimm, E.B. 2011. Mercury exposure and risk of cardiovascular disease in two US cohorts. review (Chowdhury et al. 2019, Gallego-Vinas et al. 2018 and Hu et al., 2018)
New England Journal of Medicine, 364(12):1116-1125.
Mozaffarian, D., Shi, P., Morris, J.S., Grandjean, P., Siscovick, D.S., Spiegelman, D., Excluded, as the primary study has already been assessed in an included systematic
Forman, J.P. 2012. Mercury exposure and risk of hypertension in US men and women in 2 review (Hu et al., 2018)
prospective cohorts. Hypertension, 60(3):645-652.
Mozaffarian, D., Shi, P., Morris, J.S., Grandjean, P., Siscovick, D.S., Spiegelman, D. & Hu, Excluded, as the primary study has already been assessed in an included systematic
F.B. 2013. Methylmercury exposure and incident diabetes in US men and women in two review (Roy et al., 2017)
prospective cohorts. Diabetes Care, 36(11):3578-3584.
Nielsen, A.B.S., Davidsen, M. & Bjerregaard, P. 2012. The association between blood Excluded, as the primary study has already been assessed in an included systematic
pressure and whole blood methylmercury in a cross-sectional study among Inuit in review (Hu et al., 2018)
Greenland. Environmental Health, 11(1):1-10.
Park, S.K., Lee, S., Basu, N. & Franzblau, A. 2013. Associations of blood and urinary Excluded, as the primary study has already been assessed in an included systematic
mercury with hypertension in US adults: the NHANES 2003–2006. Environmental review (Hu et al. 2018)
research, 123:25-32.
Rajaee, M., Sánchez, B.N., Renne, E.P. & Basu, N. 2015. An investigation of organic Excluded, as the primary study has already been assessed in an included systematic
and inorganic mercury exposure and blood pressure in a small-scale gold mining review (Hu et al., 2018)
community in Ghana. International journal of environmental research and public health,
12(8):10020-10038.
Suzuki, K., Nakai, K., Sugawara, T., Nakamura, T., Ohba, T., Shimada, M., Satoh, H. et Excluded, as the primary study has already been assessed in an included systematic
al. 2010. Neurobehavioral effects of prenatal exposure to methylmercury and PCBs, and review (Hibbeln et al., 2019)
seafood intake: neonatal behavioral assessment scale results of Tohoku study of child
development. Environmental research, 110(7):699-704.
Valent, F., Mariuz, M., Bin, M., Mazej, D., Tognin, V., Tratnik, J., Barbone, F. et al. 2013. Excluded, as the primary study has already been assessed in an included systematic
Associations of prenatal mercury exposure from maternal fish consumption and review (Hibbeln et al., 2019)
polyunsaturated fatty acids with child neurodevelopment: a prospective cohort study in
Italy. Journal of epidemiology, 23(5):360-370.
Valera, B., Dewailly, É., Poirier, P., Counil, E. & Suhas, E. 2011. Influence of mercury Excluded, as the primary study has already been assessed in an included systematic
exposure on blood pressure, resting heart rate and heart rate variability in French review (Gallego-Vinas et al., 2019)
Polynesians: a cross-sectional study. Environmental Health, 10, 1-10.
Valera, B., Dewailly, E. & Poirier, P. 2011. Impact of mercury exposure on blood pressure Excluded, as the primary study has already been assessed in an included systematic
and cardiac autonomic activity among Cree adults (James Bay, Quebec, Canada). review (Hu et al., 2018)
Environmental research, 111(8):1265-1270.
Valera, B., Dewailly, É. & Poirier, P. 2013. Association between methylmercury and Excluded, as the primary study has already been assessed in an included systematic
cardiovascular risk factors in a native population of Quebec (Canada): a retrospective review (Hu et al., 2018)
evaluation. Environmental research, 120, 102-108.
van Wijngaarden, E., Harrington, D., Kobrosly, R., Thurston, S.W., O’Hara, T., McSorley, Excluded, as the primary study has already been assessed in an included systematic
E.M., Davidson, P.W. et al. 2014. Prenatal exposure to methylmercury and LCPUFA in review (Dack et al., 2021)
relation to birth weight. Annals of epidemiology, 24(4):273-278.
Vejrup, K., Brandlistuen, R.E., Brantsæter, A.L., Knutsen, H.K., Caspersen, I.H., Alexander, Excluded, as the primary study has already been assessed in an included systematic
J., Haugen, M. et al. 2018. Prenatal mercury exposure, maternal seafood consumption review (Hibbeln et al., 2019)
and associations with child language at five years. Environment International, 110,
71-79.
Virtanen, J.K., Nyantika, A.N., Kauhanen, J., Voutilainen, S. & Tuomainen, T.P. 2012. Serum Excluded, as the primary study has already been assessed in an included systematic
long-chain n-3 polyunsaturated fatty acids, methylmercury and blood pressure in an review (Hu et al., 2018 and Hu et al., 2021)
older population. Hypertension Research, 35(10):1000-1004.

445
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A5.5 PRIMARY STUDIES EXCLUDED AS THEY HAD ALREADY BEEN ASSESSED IN ONE OF THE SYSTEMATIC REVIEWS FROM
THE LITERATURE SEARCH “TOXIC EFFECT OF MeHg” (cont.)
Primary studies assessed (n = 44) Primary study assessed in systematic review
Wells, E.M., Herbstman, J.B., Lin, Y.H., Jarrett, J., Verdon, C.P., Ward, C., Goldman, L.R. et Excluded, as the primary study has already been assessed in an included systematic
al. 2016. Cord blood methylmercury and fetal growth outcomes in Baltimore newborns: review (Dack et al., 2021)
potential confounding and effect modification by omega-3 fatty acids, selenium, and sex.
Environmental health perspectives, 124(3):373-379.
Wells, E.M., Herbstman, J.B., Lin, Y.H., Hibbeln, J.R., Halden, R.U., Witter, F.R. & Goldman, Excluded, as the primary study has already been assessed in an included systematic
L.R. 2017. Methyl mercury, but not inorganic mercury, associated with higher blood review (Hu et al., 2018)
pressure during pregnancy. Environmental research, 154, 247-252.
Wennberg, M., Bergdahl, I.A., Hallmans, G., Norberg, M., Lundh, T., Skerfving, S., Jansson, Excluded, as the primary study has already been assessed in an included systematic
J.H. et al. 2011. Fish consumption and myocardial infarction: a second prospective review (Chowdhury et al., 2018)
biomarker study from northern Sweden. The American journal of clinical nutrition,
93(1):27-36.
Wennberg, M., Strömberg, U., Bergdahl, I.A., Jansson, J.H., Kauhanen, J., Norberg, M., Excluded, as the primary study has already been assessed in an included systematic
Virtanen, J.K. et al. 2012. Myocardial infarction in relation to mercury and fatty acids review (Hu et al., 2018 and Hu et al., 2021)
from fish: a risk-benefit analysis based on pooled Finnish and Swedish data in men. The
American journal of clinical nutrition, 96(4):706-713.
Yorifuji, T., Tsuda, T., Kashima, S., Takao, S. & Harada, M. 2010. Long-term exposure to Excluded, as the primary study has already been assessed in an included systematic
methylmercury and its effects on hypertension in Minamata. Environmental Research, review (Hu et al., 2018)
110(1):40-46.
Yorifuji, T., Debes, F., Weihe, P. & Grandjean, P. 2011. Prenatal exposure to lead and Excluded, as the primary study has already been assessed in an included systematic
cognitive deficit in 7-and 14-year-old children in the presence of concomitant exposure to review (Karita et al., 2018)
similar molar concentration of methylmercury. Neurotoxicology and teratology, 33(2):205-
211.
Yorifuji, T., Tsuda, T., Inoue, S., Takao, S. & Harada, M. 2011. Long-term exposure to Excluded, as the primary study has already been assessed in an included systematic
methylmercury and psychiatric symptoms in residents of Minamata, Japan. Environment review (Puty et al., 2019)
international, 37(5):907-913.

446
APPENDICES

QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS AND PRIMARY STUDIES

TABLE A5.6 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS FROM FULL-TEXT SCREENING FOR THE THEME
“TOXIC EFFECTS OF MeHg”, USING THE QUALITY ASSESSMENT TOOL AMSTAR 2
Study (n = 34) AMSTAR 2 grading
High (n = 4)
Hibbeln, Joseph R. et al. 2019. Relationships between seafood consumption during pregnancy and childhood and neurocognitive development: Two High
systematic reviews. Prostaglandins, Leukotrienes and Essential Fatty Acids 151:14-36.
Hu, Q., Han, X., Dong, G., Yan, W., Wang, X., Bigambo, F.M., Wang, X. et al. 2021. Association between mercury exposure and thyroid hormones High
levels: A meta-analysis. Environmental Research, 196:110928.
Hu, X.F., Lowe, M. & Chan, H.M. 2021. Mercury exposure, cardiovascular disease, and mortality: A systematic review and dose-response meta- High
analysis. Environmental research, 193:110538.
Sarihi, S., Niknam, M., Mahjour, S., Hosseini-Bensenjan, M., Moazzen, F., Soltanabadi, S. & Akbari, H. 2021. Toxic heavy metal concentrations in High
multiple sclerosis patients: A systematic review and meta-analysis. EXCLI journal, 20:1571.
Moderate (n = 12)
Chowdhury, R., Ramond, A., O’Keeffe, L.M., Shahzad, S., Kunutsor, S.K., Muka, T., Di Angelantonio, E. et al. 2018. Environmental toxic metal Moderate
contaminants and risk of cardiovascular disease: systematic review and meta-analysis. BMJ, 29:362.
Dack, K., Fell, M., Taylor, C.M., Havdahl, A. & Lewis, S.J. 2021. Mercury and prenatal growth: a systematic review. International Journal of Moderate
Environmental Research and Public Health, 18(13):7140.
Gallego-Vinas, G., Ballester, F., & Llop, S. 2019. Chronic mercury exposure and blood pressure in children and adolescents: a systematic review. Moderate
Environmental Science and Pollution Research, 26:2238-2252.
Hu, X.F., Singh, K,. & Chan, H.M. 2018. Mercury exposure, blood pressure, and hypertension: A systematic review and dose–response meta-analysis. Moderate
Environmental health perspectives, 126(07):076002.
Jafari, T., Rostampour, N., Fallah, A.A. & Hesami, A. 2017. The association between mercury levels and autism spectrum disorders: a systematic Moderate
review and meta-analysis. Journal of Trace Elements in Medicine and Biology, 44:289-297.
Puty, B., Leão, L.K.R., Crespo-Lopez, M.E., Carvalho, A.P.C.P.S., Fagundes, N.C.F., Maia, L.C. & Lima, R.R. 2019. Association between methylmercury Moderate
environmental exposure and neurological disorders: a systematic review. Journal of Trace Elements in Medicine and Biology, 52:100-110.
Roy, C., Tremblay, P.Y. & Ayotte, P. 2017. Is mercury exposure causing diabetes, metabolic syndrome and insulin resistance? A systematic review of Moderate
the literature. Environmental research, 156:747-760.
Saghazadeh, A. & Rezaei, N. 2017. Systematic review and meta-analysis links autism and toxic metals and highlights the impact of country Moderate
development status: Higher blood and erythrocyte levels for mercury and lead, and higher hair antimony, cadmium, lead, and mercury. Progress in
Neuro-Psychopharmacology and Biological Psychiatry, 79:340-368.
Sirohi, D., Al Ramadhani, R. & Knibbs, L.D. 2021. Environmental exposures to endocrine disrupting chemicals (EDCs) and their role in Moderate
endometriosis: A systematic literature review. Reviews on Environmental Health, 36(1):101-115.
Xu, P., Liu, A., Li, F., Tinkov, A.A., Liu, L. & Zhou, J.C. 2021. Associations between metabolic syndrome and four heavy metals: a systematic review Moderate
and meta-analysis. Environmental Pollution, 273:116480.
Yoshimasu, K., Kiyohara, C., Takemura, S. & Nakai, K. 2014. A meta-analysis of the evidence on the impact of prenatal and early infancy exposures Moderate
to mercury on autism and attention deficit/hyperactivity disorder in the childhood. Neurotoxicology, 44:121-131.
Zhang, J., Li, X., Shen, L., Khan, N.U., Zhang, X., Chen, L., Luo, P. et al. 2021. Trace elements in children with autism spectrum disorder: a meta- Moderate
analysis based on case-control studies. Journal of Trace Elements in Medicine and Biology, 67:126782.
Low (n = 13)
Amadi, C.N., Orish, C.N., Frazzoli, C. & Orisakwe, O.E. 2022. Association of autism with toxic metals: a systematic review of case-control studies. Low
Pharmacology Biochemistry and Behavior, 212:173313.
Bauer, J.A., Fruh, V., Howe, C.G., White, R.F. & Claus Henn, B. 2020. Associations of metals and neurodevelopment: a review of recent evidence on Low
susceptibility factors. Current epidemiology reports, 7:237-262.
De Palma, G., Catalani, S., Franco, A., Brighenti, M. & Apostoli, P. 2012. Lack of correlation between metallic elements analyzed in hair by ICP-MS Low
and autism. Journal of autism and developmental disorders, 42:342-353.
Asmus, C.I.F., Camara, V.M., Landrigan, P.J. & Claudio, L. 2016. A systematic review of children's environmental health in Brazil. Annals of global Low
health, 82(1):132-148.
Gribble, M.O., Cheng, A., Berger, R.D., Rosman, L. & Guallar, E. 2015. Mercury exposure and heart rate variability: a systematic review. Current Low
environmental health reports, 2:304-314.
Henriques, M.C., Loureiro, S., Fardilha, M. & Herdeiro, M.T. 2019. Exposure to mercury and human reproductive health: A systematic review. Low
Reproductive toxicology, 85:93-103.
Jafari Mohammadabadi, H., Rahmatian, A., Sayehmiri, F. & Rafiei, M. 2020. The relationship between the level of copper, lead, mercury and autism Low
disorders: a meta-analysis. Pediatric Health, Medicine and Therapeutics, 11: 369-378.
Kuo, C.C., Moon, K., Thayer, K.A. & Navas-Acien, A. 2013. Environmental chemicals and type 2 diabetes: an updated systematic review of the Low
epidemiologic evidence. Current diabetes reports, 13:831-849.
Ledda, C., Cannizzaro, E., Lovreglio, P., Vitale, E., Stufano, A., Montana, A., Rapisarda, V. et al. 2019. Exposure to toxic heavy metals can influence Low
homocysteine metabolism?. Antioxidants, 9(1):30.

447
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A5.6 QUALITY ASSESSMENT (RISK OF BIAS) OF SYSTEMATIC REVIEWS FROM FULL-TEXT SCREENING FOR THE THEME
“TOXIC EFFECTS OF MeHg”, USING THE QUALITY ASSESSMENT TOOL AMSTAR 2
Study (n = 34) AMSTAR 2 grading
Miller, S., Pallan, S., Gangji, A.S., Lukic, D. & Clase, C.M. 2013. Mercury-associated nephrotic syndrome: a case report and systematic review of the Low
literature. American Journal of Kidney Diseases, 62(1):135-138.
Rossignol, D.A., Genuis, S.J., & Frye, R.E. 2014. Environmental toxicants and autism spectrum disorders: a systematic review. Translational Low
psychiatry, 4(2):e360-e360.
Saavedra, S., Fernández-Recamales, Á., Sayago, A., Cervera-Barajas, A., González-Domínguez, R. & Gonzalez-Sanz, J.D. 2022. Impact of dietary Low
mercury intake during pregnancy on the health of neonates and children: a systematic review. Nutrition Reviews, 80(2):317-328.
Yu, V., Juhász, M. Chiang, A. & Mesinkovska, N.A. 2018. Alopecia and associated toxic agents: a systematic review. Skin Appendage Disorders, Low
4(4):245-260.
Very low (n = 5)
Bellinger, D.C., O’Leary, K., Rainis, H. & Gibb, H.J. 2016. Country-specific estimates of the incidence of intellectual disability associated with Very low
prenatal exposure to methylmercury. Environmental research, 147:159-163.
Kadawathagedara, M., de Lauzon-Guillain, B. & Botton, J. 2018. Environmental contaminants and child’s growth. Journal of Developmental Origins Very low
of Health and Disease, 9(6):632-641.
Karita, K., Iwata, T., Maeda, E., Sakamoto, M. & Murata, K. 2018. Assessment of cardiac autonomic function in relation to methylmercury Very low
neurotoxicity. Toxics, 6(3):38.
Kern, J.K., Geier, D.A., Homme, K.G., King, P.G., Bjørklund, G., Chirumbolo, S. & Geier, M.R. 2017. Developmental neurotoxicants and the vulnerable Very low
male brain: a systematic review of suspected neurotoxicants that disproportionally affect males. Acta Neurobiol Exp (Wars), 77(4):269-296.
Sánchez-Alarcón, J., Milić, M., Bustamante-Montes, L.P., Isaac-Olivé, K., Valencia-Quintana, R. & Ramírez-Durán, N. 2021. Genotoxicity of Mercury Very low
and Its Derivatives Demonstrated In Vitro and In Vivo in Human Populations Studies. Systematic Review. Toxics, 9(12):326.

448
APPENDICES

TABLE A5.7 QUALITY ASSESSMENT (RISK OF BIAS) OF PRIMARY STUDIES FROM FULL-TEXT SCREENING FOR THE THEME “TOXIC
EFFECTS OF MeHg”, USING THE QUALITY ASSESSMENT TOOL OF THE OFFICE OF HEALTH ASSESSMENT
AND TRANSLATION
Study (n = 10) Risk-of-bias assessment
Andrew, A.S., Chen, C.Y., Caller, T.A., Tandan, R., Henegan, P.L., Jackson, B.P, Stommel, E.W. et al. 2018. Toenail mercury Levels are associated with Tier 2
amyotrophic lateral sclerosis risk. Muscle & nerve, 58(1):36-41.
Basta, P.C.,, Viana, P.V.D.S., Vasconcellos, A.C.S.D., Périssé, A.R.S., Hofer, C.B., Paiva, N.S., Hacon, S.D.S. et al. 2021. Mercury exposure in Tier 2
Munduruku indigenous communities from Brazilian Amazon: Methodological background and an overview of the principal results. International
journal of environmental research and public health, 18(17):9222.
Benefice, E., Luna-Monrroy, S. & Lopez-Rodriguez, R. 2010. Fishing activity, health characteristics and mercury exposure of Amerindian women Tier 2
living alongside the Beni River (Amazonian Bolivia). International Journal of Hygiene and Environmental Health, 213(6):458-464.
Creed, J.H., Peeri, N.C., Anic, G.M., Thompson, R.C., Olson, J.J., LaRocca, R.V., Egan, K.M. et al. 2019. Methylmercury exposure, genetic variation in Tier 2
metabolic enzymes, and the risk of glioma. Scientific reports, 9(1):1-7.
Karatela, S., Paterson, J. & Ward, N.I. 2017. Domain specific effects of postnatal toenail methylmercury exposure on child behaviour. Journal of Tier 2
trace elements in medicine and biology, 41:10-15.
Kishi, R., Araki, A., Minatoya, M., Hanaoka, T., Miyashita, C., Itoh, S., Goudarzi, H. et al. 2017. The Hokkaido birth cohort study on environment and Tier 2
children’s health: cohort profile—updated 2017. Environmental health and preventive medicine, 22(1):1-16.
Oliveira, R.A.A.D., Pinto, B.D., Rebouças, B.H., Ciampi de Andrade, D., Vasconcellos, A.C.S.D. & Basta, P.C. 2021. Neurological impacts of Tier 2
chronic methylmercury exposure in Munduruku indigenous adults: somatosensory, motor, and cognitive abnormalities. International journal of
environmental research and public health, 18(19):10270.
Peplow, D. & Augustine, S. 2014. Neurological abnormalities in a mercury exposed population among indigenous Wayana in Southeast Suriname. Tier 2
Environmental Science: Processes & Impacts, 16(10):2415-2422.
Turunen, A.W., Jula, A., Suominen, A.L., Männistö, S., Marniemi, J., Kiviranta, H., Verkasalo, P.K. et al. 2013. Fish consumption, omega-3 fatty acids, Tier 2
and environmental contaminants in relation to low-grade inflammation and early atherosclerosis. Environmental research, 120:43-54.
Zareba, W., Thurston, S.W., Zareba, G., Couderc, J.P., Evans, K., Xia, J. et al. et al. 2019. Prenatal and recent methylmercury exposure and heart rate Tier 2
variability in young adults: the Seychelles Child Development Study. Neurotoxicology and teratology, 74:106810.

449
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

APPENDIX 6
Se AND MeHg

LITERATURE SEARCH STRATEGY

TABLE A6.1 LITERATURE SEARCH STRATEGY FOR THE REVIEW “Se AND MeHg”
Database: Web of Science
Date of search: 7 January 2022
Literature search string Search field Hits
#1 *selen* OR HgSe TS 125 791
#2 methylmercury OR MeHg or methylHg OR methyl-Hg or CH3Hg TS 13 428
#3 mammal* OR human* OR human OR rat OR rats OR mice OR mouse OR rodent* OR ''guinea pig'' OR ''guinea pigs'' OR dogs TS 16 212 122
OR dog OR monkey* OR men OR man OR women OR woman OR patient* OR child* OR toddler* OR infant* OR fetus
#4 health* OR benefit* OR effect* OR toxi* OR intervention* OR cohor* OR uptake OR accumulation OR excretion OR clearance TS 22 328 589
OR response* OR stress* OR exposure OR cocktail* OR interaction*
#5 #1 AND #2 TS 1 302
#6 #3 AND #4 TS 8 559 078
#7 #5 AND #6 TS 715
#8 selen* OR HgSe OR methylselen* ALL fields 122 719
#9 methylmercury OR MeHg or methylHg OR methyl-Hg or CH3Hg ALL fields 13 442
#10 mammal* OR human* OR human OR rat OR rats OR mice OR mouse OR rodent* OR ''guinea pig'' OR ''guinea pigs'' OR dogs ALL fields 18 794 267
OR dog OR monkey* OR men OR man OR women OR woman OR patient* OR child* OR toddler* OR infant* OR fetus
#11 health* OR benefit* OR effect* OR toxi* OR intervention* OR cohor* OR uptake OR accumulation OR excretion OR clearance ALL fields 24 538 170
OR response* OR stress* OR exposure OR cocktail* OR interaction*
#12 #8 AND #9 ALL fields 1 282
#13 #10 AND #11 ALL fields 11 148 862
#14 #12 AND #13 ALL fields 771
#15 #1 AND #9 AND #13 TS + ALL fields hybrid 792
Database: PubMed
Date of search: 7 January 2022
Literature search string Search field Hits
#1 selen* OR HgSe All fields 52 139
#2 methylmercury OR MeHg or methylHg OR methyl-Hg or CH3Hg All fields 8 400
#3 mammal* OR human* OR human OR rat OR rats OR mice OR mouse OR rodent* OR ''guinea pig'' OR ''guinea pigs'' OR dogs All fields 23 699 237
OR dog OR monkey* OR men OR man OR women OR woman OR patient* OR child* OR toddler* OR infant* OR fetus
#4 health* OR benefit* OR effect* OR toxi* OR intervention* OR cohor* OR uptake OR accumulation OR excretion OR clearance All fields 17 949 060
OR response* OR stress* OR exposure OR cocktail* OR interaction*
#5 #1 AND #2 656
#6 #3 AND #4 13 421 491
#7 #5 AND #6 362

450
APPENDICES

RECORDS EXCLUDED DURING FULL-TEXT SCREENING

TABLE A6.2 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING OF HUMAN STUDIES FOR THE REVIEW “Se AND MeHg”,
BASED ON INCLUSION AND EXCLUSION CRITERIA
Studies (n = 68) Reason for exclusion
Achouba, A., Dumas, P., Ouellet, N., Little, M., Lemire, M. & Ayotte, P. 2019. Selenoneine is a major selenium species in beluga skin and red blood No health outcome measured.
cells of Inuit from Nunavik. Chemosphere, 229: 549-558. https://doi.org/10.1016/j.chemosphere.2019.04.191
Afonso, C., Bernardo, I., Bandarra, N.M., Martins, L.L. & Cardoso, C. 2019. The implications of following dietary advice regarding fish consumption No health outcome measured.
frequency and meal size for the benefit (EPA + DHA and Se) versus risk (MeHg) assessment. Int J Food Sci Nutr, 70(5): 623-637. https://doi.org/10.
1080/09637486.2018.1551334
Alves, A.C., Monteiro, M.S., Machado, A.L., Oliveira, M., Bóia, A., Correia, A., Oliveira, N., Soares, A. & Loureiro, S. 2017. Mercury levels in parturient No health outcome measured.
and newborns from Aveiro region, Portugal. J Toxicol Environ Health A, 80(13-15): 697-709. https://doi.org/10.1080/15287394.2017.1286926
Ask, K., Akesson, A., Berglund, M. & Vahter, M. 2002. Inorganic mercury and methylmercury in placentas of Swedish women. Environmental Health No health outcome measured.
Perspectives, 110(5): 523-526. https://doi.org/10.1289/ehp.02110523
Ballesteros, M.T.L., Barrado, B.G., Serrano, I.N., Alvarez, S.I., Anaya, M.D.G. & Munoz, M.J.G. 2020. Evaluation of blood mercury and serum selenium No health outcome measured.
levels in the pregnant population of the Community of Madrid, Spain. Journal of Trace Elements in Medicine and Biology, 57: 60-67. https://doi.
org/10.1016/j.jtemb.2019.09.008
Barany, E., Bergdahl, I.A., Bratteby, L.E., Lundh, T., Samuelson, G., Skerfving, S. & Oskarsson, A. 2003. Mercury and selenium in whole blood and No health outcome measured.
serum in relation to fish consumption and amalgam fillings in adolescents. Journal of Trace Elements in Medicine and Biology, 17(3): 165-170.
https://doi.org/10.1016/s0946-672x(03)80021-4
Bates, C.J., Prentice, A., Birch, M.C. & Delves, H.T. 2007. Dependence of blood indices of selenium and mercury on estimated fish intake in a No health outcome measured.
national survey of British adults. Public Health Nutrition, 10(5): 508-517. https://doi.org/10.1017/s1368980007246683
Bates, C.J., Prentice, A., Birch, M.C., Delves, H.T. & Sinclair, K.A. 2006. Blood indices of selenium and mercury, and their correlations with fish No health outcome measured.
intake, in young people living in Britain. British Journal of Nutrition, 96(3): 523-531. https://doi.org/10.1079/bjn20061847
Binnington, M.J., Curren, M.S., Chan, H.M. & Wania, F. 2016. Balancing the benefits and costs of traditional food substitution by indigenous Arctic No access to full-text paper.
women of childbearing age: Impacts on persistent organic pollutant, mercury, and nutrient intakes. Environment International, 94: 554-566.
https://doi.org/10.1016/j.envint.2016.06.016
Bjornberg, K.A., Vahter, M., Petersson-Grawe, K., Glynn, A., Cnattingius, S., Darnerud, P.O., Atuma, S., Aune, M., Becker, W. & Berglund, M. 2003. No health outcome measured.
Methyl mercury and inorganic mercury in Swedish pregnant women and in cord blood: Influence of fish consumption. Environmental Health
Perspectives, 111(4): 637-641. https://doi.org/10.1289/ehp.111-1241457
Björnberg, K.A., Vahter, M., Grawé, K.P. & Berglund, M. 2005. Methyl mercury exposure in Swedish women with high fish consumption. Sci Total No health outcome measured.
Environ, 341(1-3): 45-52. https://doi.org/10.1016/j.scitotenv.2004.09.033
Bridges, K.N., Furin, C.G. & Gerlach, R.F. 2020. Subsistence fish consumption in rural Alaska: Using regional monitoring data to evaluate risk and No health outcome measured.
bioavailability of dietary methylmercury. Sci Total Environ, 736: 139676. https://doi.org/10.1016/j.scitotenv.2020.139676
Brumatti, L.V., Rosolen, V., Mariuz, M., Piscianz, E., Valencic, E., Bin, M., Athanasakis, E. et al. 2021. Impact of Methylmercury and Other Heavy Study protocol and does not
Metals Exposure on Neurocognitive Function in Children Aged 7 Years: Study Protocol of the Follow-up. Journal of Epidemiology, 31(2): 157-163. include primary data.
https://doi.org/10.2188/jea.JE20190284
Cardoso, C., Bernardo, I., Bandarra, N.M., Martins, L.L. & Afonso, C. 2018. Portuguese preschool children: Benefit (EPA plus DHA and Se) and risk Health outcome not measured
(MeHg) assessment through the consumption of selected fish species. Food and Chemical Toxicology, 115: 306-314. https://doi.org/10.1016/j.
fct.2018.03.022
Carneiro, M.F., Grotto, D. & Barbosa, F., Jr. 2014. Inorganic and methylmercury levels in plasma are differentially associated with age, gender, and Health outcome not measured
oxidative stress markers in a population exposed to mercury through fish consumption. J Toxicol Environ Health A, 77(1-3): 69-79. https://doi.org/1
0.1080/15287394.2014.865584
Dewailly, E., Suhas, E., Mou, Y., Dallaire, R., Chateau-Degat, L. & Chansin, R. 2008. High fish consumption in French Polynesia and prenatal Health outcome not measured
exposure to metals and nutrients. Asia Pacific Journal of Clinical Nutrition, 17(3): 461-470. <Go to ISI>://WOS:000260195600015
Dewailly, E., Chateau-Degat, L. & Suhas, E. 2008. Fish consumption and health in French Polynesia. Asia Pacific Journal of Clinical Nutrition, 17(1): Health outcome not measured
86-93. <Go to ISI>://WOS:000254880400015
Gilman, C.L., Soon, R., Sauvage, L., Ralston, N.V.C. & Berry, M.J. 2015. Umbilical cord blood and placental mercury, selenium and selenoprotein Health outcome not measured
expression in relation to maternal fish consumption. Journal of Trace Elements in Medicine and Biology, 30: 17-24. https://doi.org/10.1016/j.
jtemb.2015.01.006
Grandjean, P., Weihe, P., Needham, L.L., Burse, V.W., Patterson, D.G., Sampson, E.J., Jorgensen, P.J. & Vahter, M. 1995. Relation of a seafood diet to Health outcome not measured
mercury, selenium, arsenic, and polychlorinated biphenyl and other organochlorine concentrations in human milk. Environmental Research, 71(1):
29-38. https://doi.org/10.1006/enrs.1995.1064
Gundacker, C., Komarnicki, G., Zodl, B., Forster, C., Schuster, E. & Wittmann, K. 2006. Whole blood mercury and selenium concentrations Health outcome not measured
in a selected Austrian population: Does gender matter? Science of the Total Environment, 372(1): 76-86. https://doi.org/10.1016/j.
scitotenv.2006.08.006
Hayat, L. 1996. Cations in malignant and benign brain tumors. Journal of Environmental Science and Health Part a-Environmental Science and Health outcome not measured
Engineering & Toxic and Hazardous Substance Control, 31(8): 1831-1840. https://doi.org/10.1080/10934529609376459
Hoang, V.A.T., Do, H.T.T., Agusa, T., Koriyama, C., Akiba, S., Ishibashi, Y., Sakamoto, M. & Yamamoto, M. 2017. Hair mercury levels in relation to fish Health outcome not measured
consumption among Vietnamese in Hanoi. J Toxicol Sci, 42(5): 651-662. https://doi.org/10.2131/jts.42.651
Iwai-Shimada, M., Kameo, S., Nakai, K., Yaginuma-Sakurai, K., Tatsuta, N., Kurokawa, N., Nakayama, S.F. & Satoh, H. 2019. Exposure profile of Health outcome not measured
mercury, lead, cadmium, arsenic, antimony, copper, selenium and zinc in maternal blood, cord blood and placenta: the Tohoku Study of Child
Development in Japan. Environ Health Prev Med, 24(1): 35. https://doi.org/10.1186/s12199-019-0783-y

451
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A6.2 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING OF HUMAN STUDIES FOR THE REVIEW “Se AND MeHg”,
BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Studies (n = 68) Reason for exclusion
Johansen, P., Mulvad, G., Pedersen, H.S., Hansen, J.C. & Riget, F. 2007. Human accumulation of mercury in Greenland. Sci Total Environ, 377(2-3): No health outcome measured.
173-8. https://doi.org/10.1016/j.scitotenv.2007.02.004
Karimi, R., Fisher, N.S. & Meliker, J.R. 2014. Mercury-nutrient signatures in seafood and in the blood of avid seafood consumers. Sci Total Environ, No health outcome measured.
496: 636-643. https://doi.org/10.1016/j.scitotenv.2014.04.049
Karita, K. & Suzuki, T. 2002. Fish eating and variations in selenium and mercury levels in plasma and erythrocytes in free-living healthy Japanese No health outcome measured.
men. Biological Trace Element Research, 90(1-3): 71-81. https://doi.org/10.1385/bter:90:1-3:71
Kim, B.M., Choi, A.L., Ha, E.H., Pedersen, L., Nielsen, F., Weihe, P., Hong, Y.C., Budtz-Jørgensen, E. & Grandjean, P. 2014. Effect of hemoglobin Methodical but not related
adjustment on the precision of mercury concentrations in maternal and cord blood. Environ Res, 132: 407-12. https://doi.org/10.1016/j. directly to health.
envres.2014.04.030
Korbas, M., O'Donoghue, J.L., Watson, G.E., Pickering, I.J., Singh, S.P., Myers, G.J., Clarkson, T.W. & George, G.N. 2010. The chemical nature of Exposure to MeHg after
mercury in human brain following poisoning or environmental exposure. ACS Chem Neurosci, 1(12): 810-8. https://doi.org/10.1021/cn1000765 poisoning.
Kosatsky, T., Przybysz, R. & Armstrong, B. 2000. Mercury exposure in Montrealers who eat St. Lawrence River sportfish. Environmental Research, No health outcome measured
84(1): 36-43. https://doi.org/10.1006/enrs.2000.4073 and only Se in blood
Lemire, M., Kwan, M., Laouan-Sidi, A.E., Muckle, G., Pirlde, C., Ayotte, P. & Dewailly, E. 2015. Local country food sources of methylmercury, No health outcome measured.
selenium and omega-3 fatty acids in Nunavik, Northern Quebec. Science of the Total Environment, 509: 248-259. https://doi.org/10.1016/j.
scitotenv.2014.07.102
Lemire, M., Philibert, A., Fillion, M., Passos, C.J.S., Guimaraes, J.R.D., Barbosa, F. & Mergler, D. 2012. No evidence of selenosis from a selenium-rich Only selenium measured (not
diet in the Brazilian Amazon. Environment International, 40: 128-136. https://doi.org/10.1016/j.envint.2011.07.005 mercury)
Lemire, M., Mergler, D., Fillion, M., Sousa Passos, C.J., Guimaraes, J.R.D., Davidson, R. & Lucotte, M. 2006. Elevated blood selenium levels in the No health outcome measured.
Brazilian Amazon. Science of the Total Environment, 366(1): 101-111. https://doi.org/10.1016/j.scitotenv.2005.08.057
Li, Y.F., Chen, C.Y., Li, B., Wang, Q., Wang, J.X., Gao, Y.X., Zhao, Y.L. & Chai, Z.F. 2007. Simultaneous speciation of selenium and mercury in human Method development and
urine samples from long-term mercury-exposed populations with supplementation of selenium-enriched yeast by HPLC-ICP-MS. Journal of specification
Analytical Atomic Spectrometry, 22(8): 925-930. https://doi.org/10.1039/b703310a
Lindberg, A., Björnberg, K.A., Vahter, M. & Berglund, M. 2004. Exposure to methylmercury in non-fish-eating people in Sweden. Environ Res, 96(1): No health outcome measured
28-33. https://doi.org/10.1016/j.envres.2003.09.005
Little, M., Achouba, A., Dumas, P., Ouellet, N., Ayotte, P. & Lemire, M. 2019. Determinants of selenoneine concentration in red blood cells of Inuit No health outcome measured
from Nunavik (Northern Quebec, Canada). Environment International, 127: 243-252. https://doi.org/10.1016/j.envint.2018.11.077
Lubick, N. 2010. A balanced diet? Selenium may offset the effects of methylmercury on cataract development. Environ Health Perspect, 118(11): News article
A491. https://doi.org/10.1289/ehp.118-a491b
Marumoto, M., Sakamoto, M., Marumoto, K., Tsuruta, S. & Komohara, Y. 2020. Mercury and Selenium Localization in the Cerebrum, Cerebellum, A single toxicology case
Liver, and Kidney of a Minamata Disease Case. Acta Histochem Cytochem, 53(6): 147-155. https://doi.org/10.1267/ahc.20-00009
Miklavčič, A., Casetta, A., Snoj Tratnik, J., Mazej, D., Krsnik, M., Mariuz, M., Sofianou, K., Spirić, Z., Barbone, F. & Horvat, M. 2013. Mercury, arsenic No health outcome measured
and selenium exposure levels in relation to fish consumption in the Mediterranean area. Environ Res, 120: 7-17. https://doi.org/10.1016/j.
envres.2012.08.010
Muckle, G., Ayotte, P., Dewailly, E., Jacobson, S.W. & Jacobson, J.L. 2001. Determinants of polychlorinated biphenyls and methylmercury exposure in No health outcome measured
inuit women of childbearing age. Environ Health Perspect, 109(9): 957-63. https://doi.org/10.1289/ehp.01109957
Muckle, G., Ayotte, P., Dewailly, E.E., Jacobson, S.W. & Jacobson, J.L. 2001. Prenatal exposure of the northern Québec Inuit infants to environmental No health outcome measured
contaminants. Environ Health Perspect, 109(12): 1291-9. https://doi.org/10.1289/ehp.011091291
Nakayama, S.F., Iwai-Shimada, M., Oguri, T., Isobe, T., Takeuchi, A., Kobayashi, Y., Michikawa, T. et al. 2019. Blood mercury, lead, cadmium, No health outcome measured
manganese and selenium levels in pregnant women and their determinants: the Japan Environment and Children's Study (JECS). Journal of
Exposure Science and Environmental Epidemiology, 29(5): 633-647. https://doi.org/10.1038/s41370-019-0139-0
Niane, B., Guedron, S., Moritz, R., Cosio, C., Ngom, P.M., Deverajan, N., Pfeifer, H.R. & Pote, J. 2015. Human exposure to mercury in artisanal small- No health outcome measured
scale gold mining areas of Kedougou region, Senegal, as a function of occupational activity and fish consumption. Environmental Science and
Pollution Research, 22(9): 7101-7111. https://doi.org/10.1007/s11356-014-3913-5
Parajuli, R.P., Goodrich, J.M., Chan, H.M., Lemire, M., Ayotte, P., Hegele, R.A. & Basu, N. 2021. Variation in biomarker levels of metals, persistent No health outcome measured
organic pollutants, and omega-3 fatty acids in association with genetic polymorphisms among Inuit in Nunavik, Canada. Environmental Research,
200. https://doi.org/10.1016/j.envres.2021.111393
Pinheiro, M.C.N., Muller, R.C.S., Sarkis, J.E., Vieira, J.L.F., Oikawa, T., Gomes, M.S.V., Guimaraes, G.A., do Nascimento, J.L.M. & Silveira, L.C.L. No health outcome measured
2005. Mercury and selenium concentrations in hair samples of women in fertile age from Amazon riverside communities. Science of the Total
Environment, 349(1-3): 284-288. https://doi.org/10.1016/j.scitotenv.2005.06.026
Sakamoto, M., Haraguchi, K., Tatsuta, N., Nakai, K., Nakamura, M. & Murata, K. 2021. Plasma and red blood cells distribution of total mercury, No health outcome measured
inorganic mercury, and selenium in maternal and cord blood from a group of Japanese women. Environ Res, 196: 110896. https://doi.org/10.1016/j.
envres.2021.110896
Sakamoto, M., Chan, H.M., Domingo, J.L., Koriyama, C. & Murata, K. 2018. Placental transfer and levels of mercury, selenium, vitamin E, and No health outcome measured
docosahexaenoic acid in maternal and umbilical cord blood. Environ Int, 111: 309-315. https://doi.org/10.1016/j.envint.2017.11.001
Sakamoto, M., Chan, H.M., Domingo, J.L., Kubota, M. & Murata, K. 2012. Changes in body burden of mercury, lead, arsenic, cadmium and No health outcome measured
selenium in infants during early lactation in comparison with placental transfer. Ecotoxicol Environ Saf, 84: 179-84. https://doi.org/10.1016/j.
ecoenv.2012.07.014

452
APPENDICES

TABLE A6.2 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING OF HUMAN STUDIES FOR THE REVIEW “Se AND MeHg”,
BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Studies (n = 68) Reason for exclusion
Sakamotoa, M., Kubota, M., Murata, K., Nakai, K., Sonoda, I. & Satoh, H. 2008. Changes in mercury concentrations of segmental maternal hair No health outcome measured
during gestation and their correlations with other biomarkers of fetal exposure to methylmercury in the Japanese population. Environmental
Research, 106(2): 270-276. https://doi.org/10.1016/j.envres.2007.10.002
Sekovanic, A., Piasek, M., Orct, T., Grgec, A.S., Saric, M.M., Stasenko, S. & Jurasovic, J. 2020. Mercury Exposure Assessment in Mother-Infant Pairs No health outcome measured
from Continental and Coastal Croatia. Biomolecules, 10(6). https://doi.org/10.3390/biom10060821
Ser, P.H., Omi, S., Shimizu-Furusawa, H., Yasutake, A., Sakamoto, M., Hachiya, N., Konishi, S., Nakamura, M. & Watanabe, C. 2017. Differences in No health outcome measured
the responses of three plasma selenium-containing proteins in relation to methylmercury-exposure through consumption of fish/whales. Toxicol
Lett, 267: 53-58. https://doi.org/10.1016/j.toxlet.2016.12.001
Suzuki, T., Hongo, T., Yoshinaga, J., Imai, H., Nakazawa, M., Matsuo, N. & Akagi, H. 1993. The hair-organ relationship in mercury concentration in No health outcome measured
contemporary Japanese. Arch Environ Health, 48(4): 221-9. https://doi.org/10.1080/00039896.1993.9940363
Svensson, B.G., Nilsson, A., Jonsson, E., Schütz, A., Akesson, B. & Hagmar, L. 1995. Fish consumption and exposure to persistent organochlorine No health outcome measured
compounds, mercury, selenium and methylamines among Swedish fishermen. Scand J Work Environ Health, 21(2): 96-105. https://doi.org/10.5271/
sjweh.16
Svensson, B.G., Schütz, A., Nilsson, A., Akesson, I., Akesson, B. & Skerfving, S. 1992. Fish as a source of exposure to mercury and selenium. Sci No health outcome measured
Total Environ, 126(1-2): 61-74. https://doi.org/10.1016/0048-9697(92)90484-a
Trdin, A., Snoj Tratnik, J., Stajnko, A., Marc, J., Mazej, D., Sešek Briški, A., Kastelec, D. et al. 2020. Trace elements and APOE polymorphisms in No use of Se data
pregnant women and their new-borns. Environ Int, 143: 105626. https://doi.org/10.1016/j.envint.2020.105626
Trdin, A., Snoj Tratnik, J., Mazej, D., Fajon, V., Krsnik, M., Osredkar, J., Prpić, I. et al. 2019. Mercury speciation in prenatal exposure in Slovenian and No health outcome measured
Croatian population - PHIME study. Environ Res, 177: 108627. https://doi.org/10.1016/j.envres.2019.108627
Valent, F., Pisa, F., Mariuz, M., Horvat, M., Gibicar, D., Fajon, V., Mazej, D., Daris, F. & Barbone, F. 2011. Fetal and perinatal exposure to mercury and Article in Italian
selenium: baseline evaluation of a cohort of children in Friuli Venezia Giulia, Italy. Epidemiol Prev, 35(1): 33-42.
Valera, B., Muckle, G., Poirier, P., Jacobson, S.W., Jacobson, J.L. & Dewailly, E. 2012. Cardiac autonomic activity and blood pressure among Inuit Selenium not measured
children exposed to mercury. Neurotoxicology, 33(5): 1067-1074. https://doi.org/10.1016/j.neuro.2012.05.005
Vecchi Brumatti, L., Rosolen, V., Mariuz, M., Piscianz, E., Valencic, E., Bin, M., Athanasakis, E. et al. 2021. Impact of Methylmercury and Other Study protocol only
Heavy Metals Exposure on Neurocognitive Function in Children Aged 7 Years: Study Protocol of the Follow-up. J Epidemiol, 31(2): 157-163. https://
doi.org/10.2188/jea.JE20190284
Walker, J.B., Houseman, J., Seddon, L., McMullen, E., Tofflemire, K., Mills, C., Corriveau, A. et al. 2006. Maternal and umbilical cord blood levels No health outcome measured
of mercury, lead, cadmium, and essential trace elements in Arctic Canada. Environmental Research, 100(3): 295-318. https://doi.org/10.1016/j.
envres.2005.05.006
Wells, E.M., Herbstman, J.B., Lin, Y.H., Hibbeln, J.R., Halden, R.U., Witter, F.R. & Goldman, L.R. 2017. Methyl mercury, but not inorganic mercury, Selenium as a covariate in a
associated with higher blood pressure during pregnancy. Environ Res, 154: 247-252. https://doi.org/10.1016/j.envres.2017.01.013 multi model
Wilhelm, M., Wittsiepe, J., Schrey, P., Lajoie-Junge, L. & Busch, V. 2003. Dietary intake of arsenic, mercury and selenium by children from a German No health outcome measured
North Sea island using duplicate portion sampling. Journal of Trace Elements in Medicine and Biology, 17(2): 123-132. https://doi.org/10.1016/
s0946-672x(03)80008-1
Yalcin, S.S., Yurdakok, K., Yalcin, S., Engur-Karasimav, D. & Coskun, T. 2010. Maternal and environmental determinants of breast-milk mercury No health outcome measured
concentrations. Turkish Journal of Pediatrics, 52(1): 1-9. <Go to ISI>://WOS:000276572900001
Yamashita, M., Yamashita, Y., Ando, T., Wakamiya, J. & Akiba, S. 2013. Identification and determination of selenoneine, 2-selenyl-N α , N α , N α No health outcome measured
-trimethyl-L-histidine, as the major organic selenium in blood cells in a fish-eating population on remote Japanese Islands. Biol Trace Elem Res,
156(1-3): 36-44. https://doi.org/10.1007/s12011-013-9846-x
Yoo, Y.C., Lee, S.K., Yang, J.Y., Kim, K.W., Lee, S.Y., Oh, S.M. & Chung, K.H. 2002. Interrelationship between the concentration of toxic and essential No health outcome measured
elements in Korean tissues. Journal of Health Science, 48(2): 195-200. <Go to ISI>://WOS:000174717400015
Yoshinaga, J., Matsuo, N., Imai, H., Nakazawa, M., Suzuki, T., Morita, M. & Akagi, H. 1990. Interrelationship between the concentrations of some No health outcome measured
elements in the organs of Japanese with special reference to selenium-heavy metal relationships. Sci Total Environ, 91: 127-40. https://doi.
org/10.1016/0048-9697(90)90294-5
Zhang, M.Y., Buckley, J.P., Liang, L.M., Hong, X.M., Wang, G.Y., Wang, M.C., Wills-Karp, M., Wang, X.B. & Mueller, N.T. 2022. A metabolome-wide Effect of selenium on Hg
association study of in utero metal and trace element exposures with cord blood metabolome profile: Findings from the Boston Birth Cohort. toxicity not measured
Environment International, 158. https://doi.org/10.1016/j.envint.2021.106976
Zhang, M.Y., Liu, T.G., Wang, G.Y., Buckley, J.P., Guallar, E., Hong, X.M., Wang, M.C., Wills-Karp, M., Wang, X.B. & Mueller, N.T. 2021. In Utero No health outcome measured
Exposure to Heavy Metals and Trace Elements and Childhood Blood Pressure in a US Urban, Low-Income, Minority Birth Cohort. Environmental
Health Perspectives, 129(6). https://doi.org/10.1289/ehp8325
Studies (n = 68) Reason for exclusion
Gilman, C. L., Soon, R., Sauvage, L., Ralston, N. V., & Berry, M. J. 2015. Umbilical cord blood and placental mercury, selenium and selenoprotein No health outcome measured
expression in relation to maternal fish consumption. Journal of Trace Elements in Medicine and Biology, 30, 17-24.
Achouba, A., Dumas, P., Ouellet, N., Little, M., Lemire, M. & Ayotte, P. 2019. Selenoneine is a major selenium species in beluga skin and red blood No health outcome measured
cells of Inuit from Nunavik. Chemosphere, 229: 549-558. https://doi.org/10.1016/j.chemosphere.2019.04.191
Afonso, C., Bernardo, I., Bandarra, N.M., Martins, L.L. & Cardoso, C. 2019. The implications of following dietary advice regarding fish consumption No health outcome measured
frequency and meal size for the benefit (EPA + DHA and Se) versus risk (MeHg) assessment. Int J Food Sci Nutr, 70(5): 623-637. https://doi.org/10.
1080/09637486.2018.1551334
Alves, A.C., Monteiro, M.S., Machado, A.L., Oliveira, M., Bóia, A., Correia, A., Oliveira, N., Soares, A. & Loureiro, S. 2017. Mercury levels in parturient No health outcome measured
and newborns from Aveiro region, Portugal. J Toxicol Environ Health A, 80(13-15): 697-709. https://doi.org/10.1080/15287394.2017.1286926

453
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A6.2 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING OF HUMAN STUDIES FOR THE REVIEW “Se AND MeHg”,
BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Studies (n = 68) Reason for exclusion
Ask, K., Akesson, A., Berglund, M. & Vahter, M. 2002. Inorganic mercury and methylmercury in placentas of Swedish women. Environmental Health No health outcome measured
Perspectives, 110(5): 523-526. https://doi.org/10.1289/ehp.02110523
Ballesteros, M.T.L., Barrado, B.G., Serrano, I.N., Alvarez, S.I., Anaya, M.D.G. & Munoz, M.J.G. 2020. Evaluation of blood mercury and serum selenium No health outcome measured
levels in the pregnant population of the Community of Madrid, Spain. Journal of Trace Elements in Medicine and Biology, 57: 60-67. https://doi.
org/10.1016/j.jtemb.2019.09.008
Barany, E., Bergdahl, I.A., Bratteby, L.E., Lundh, T., Samuelson, G., Skerfving, S. & Oskarsson, A. 2003. Mercury and selenium in whole blood and No health outcome measured
serum in relation to fish consumption and amalgam fillings in adolescents. Journal of Trace Elements in Medicine and Biology, 17(3): 165-170.
https://doi.org/10.1016/s0946-672x(03)80021-4
Bates, C.J., Prentice, A., Birch, M.C. & Delves, H.T. 2007. Dependence of blood indices of selenium and mercury on estimated fish intake in a No health outcome measured.
national survey of British adults. Public Health Nutrition, 10(5): 508-517. https://doi.org/10.1017/s1368980007246683
Bates, C.J., Prentice, A., Birch, M.C., Delves, H.T. & Sinclair, K.A. 2006. Blood indices of selenium and mercury, and their correlations with fish No health outcome measured.
intake, in young people living in Britain. British Journal of Nutrition, 96(3): 523-531. https://doi.org/10.1079/bjn20061847
Binnington, M.J., Curren, M.S., Chan, H.M. & Wania, F. 2016. Balancing the benefits and costs of traditional food substitution by indigenous Arctic No access to full-text paper.
women of childbearing age: Impacts on persistent organic pollutant, mercury, and nutrient intakes. Environment International, 94: 554-566.
https://doi.org/10.1016/j.envint.2016.06.016
Bjornberg, K.A., Vahter, M., Petersson-Grawe, K., Glynn, A., Cnattingius, S., Darnerud, P.O., Atuma, S., Aune, M., Becker, W. & Berglund, M. 2003. No health outcome measured.
Methyl mercury and inorganic mercury in Swedish pregnant women and in cord blood: Influence of fish consumption. Environmental Health
Perspectives, 111(4): 637-641. https://doi.org/10.1289/ehp.111-1241457
Björnberg, K.A., Vahter, M., Grawé, K.P. & Berglund, M. 2005. Methyl mercury exposure in Swedish women with high fish consumption. Sci Total No health outcome measured.
Environ, 341(1-3): 45-52. https://doi.org/10.1016/j.scitotenv.2004.09.033
Bridges, K.N., Furin, C.G. & Gerlach, R.F. 2020. Subsistence fish consumption in rural Alaska: Using regional monitoring data to evaluate risk and No health outcome measured.
bioavailability of dietary methylmercury. Sci Total Environ, 736: 139676. https://doi.org/10.1016/j.scitotenv.2020.139676
Brumatti, L.V., Rosolen, V., Mariuz, M., Piscianz, E., Valencic, E., Bin, M., Athanasakis, E. et al. 2021. Impact of Methylmercury and Other Heavy Study protocol and does not
Metals Exposure on Neurocognitive Function in Children Aged 7 Years: Study Protocol of the Follow-up. Journal of Epidemiology, 31(2): 157-163. include primary data.
https://doi.org/10.2188/jea.JE20190284
Walker, J.B., Houseman, J., Seddon, L., McMullen, E., Tofflemire, K., Mills, C., Corriveau, A. et al. 2006. Maternal and umbilical cord blood levels Health outcome not
of mercury, lead, cadmium, and essential trace elements in Arctic Canada. Environmental Research, 100(3): 295-318. https://doi.org/10.1016/j. measured.
envres.2005.05.006
Cardoso, C., Bernardo, I., Bandarra, N.M., Martins, L.L. & Afonso, C. 2018. Portuguese preschool children: Benefit (EPA plus DHA and Se) and risk Health outcome not
(MeHg) assessment through the consumption of selected fish species. Food and Chemical Toxicology, 115: 306-314. https://doi.org/10.1016/j. measured.
fct.2018.03.022
Carneiro, M.F., Grotto, D. & Barbosa, F., Jr. 2014. Inorganic and methylmercury levels in plasma are differentially associated with age, gender, and Health outcome not
oxidative stress markers in a population exposed to mercury through fish consumption. J Toxicol Environ Health A, 77(1-3): 69-79. https://doi.org/1 measured.
0.1080/15287394.2014.865584
Dewailly, E., Chateau-Degat, L. & Suhas, E. 2008. Fish consumption and health in French Polynesia. Asia Pacific Journal of Clinical Nutrition, 17(1): Health outcome not
86-93. measured.
Dewailly, E., Suhas, E., Mou, Y., Dallaire, R., Chateau-Degat, L. & Chansin, R. 2008. High fish consumption in French Polynesia and prenatal Health outcome not
exposure to metals and nutrients. Asia Pacific Journal of Clinical Nutrition, 17(3): 461-470. measured.
Grandjean, P., Weihe, P., Needham, L.L., Burse, V.W., Patterson, D.G., Sampson, E.J., Jorgensen, P.J. & Vahter, M. 1995. Relation of a seafood diet to Health outcome not
mercury, selenium, arsenic, and polychlorinated biphenyl and other organochlorine concentrations in human milk. Environmental Research, 71(1): measured.
29-38. https://doi.org/10.1006/enrs.1995.1064
Gundacker, C., Komarnicki, G., Zodl, B., Forster, C., Schuster, E. & Wittmann, K. 2006. Whole blood mercury and selenium concentrations Health outcome not
in a selected Austrian population: Does gender matter? Science of the Total Environment, 372(1): 76-86. https://doi.org/10.1016/j. measured.
scitotenv.2006.08.006
Hayat, L. 1996. Cations in malignant and benign brain tumors. Journal of Environmental Science and Health Part a-Environmental Science and Health outcome not
Engineering & Toxic and Hazardous Substance Control, 31(8): 1831-1840. https://doi.org/10.1080/10934529609376459 measured.
Hoang, V.A.T., Do, H.T.T., Agusa, T., Koriyama, C., Akiba, S., Ishibashi, Y., Sakamoto, M. & Yamamoto, M. 2017. Hair mercury levels in relation to fish No health outcome measured
consumption among Vietnamese in Hanoi. J Toxicol Sci, 42(5): 651-662. https://doi.org/10.2131/jts.42.651 in relation to toxicity of MeHg.
Iwai-Shimada, M., Kameo, S., Nakai, K., Yaginuma-Sakurai, K., Tatsuta, N., Kurokawa, N., Nakayama, S.F. & Satoh, H. 2019. Exposure profile of No health outcome measured.
mercury, lead, cadmium, arsenic, antimony, copper, selenium and zinc in maternal blood, cord blood and placenta: the Tohoku Study of Child
Development in Japan. Environ Health Prev Med, 24(1): 35. https://doi.org/10.1186/s12199-019-0783-y
Johansen, P., Mulvad, G., Pedersen, H.S., Hansen, J.C. & Riget, F. 2007. Human accumulation of mercury in Greenland. Sci Total Environ, 377(2-3): No health outcome measured.
173-8. https://doi.org/10.1016/j.scitotenv.2007.02.004
Karimi, R., Fisher, N.S. & Meliker, J.R. 2014. Mercury-nutrient signatures in seafood and in the blood of avid seafood consumers. Sci Total Environ, No health outcome measured.
496: 636-643. https://doi.org/10.1016/j.scitotenv.2014.04.049
Karita, K. & Suzuki, T. 2002. Fish eating and variations in selenium and mercury levels in plasma and erythrocytes in free-living healthy Japanese No health outcome measured.
men. Biological Trace Element Research, 90(1-3): 71-81. https://doi.org/10.1385/bter:90:1-3:71
Kim, B.M., Choi, A.L., Ha, E.H., Pedersen, L., Nielsen, F., Weihe, P., Hong, Y.C., Budtz-Jørgensen, E. & Grandjean, P. 2014. Effect of hemoglobin Methodical, but not related
adjustment on the precision of mercury concentrations in maternal and cord blood. Environ Res, 132: 407-12. https://doi.org/10.1016/j. directly to health.
envres.2014.04.030

454
APPENDICES

TABLE A6.2 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING OF HUMAN STUDIES FOR THE REVIEW “Se AND MeHg”,
BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Studies (n = 68) Reason for exclusion
Korbas, M., O'Donoghue, J.L., Watson, G.E., Pickering, I.J., Singh, S.P., Myers, G.J., Clarkson, T.W. & George, G.N. 2010. The chemical nature of Exposure to MeHg after
mercury in human brain following poisoning or environmental exposure. ACS Chem Neurosci, 1(12): 810-8. https://doi.org/10.1021/cn1000765 poisoning.
Kosatsky, T., Przybysz, R. & Armstrong, B. 2000. Mercury exposure in Montrealers who eat St. Lawrence River sportfish. Environmental Research, No health outcome measured
84(1): 36-43. https://doi.org/10.1006/enrs.2000.4073 and only Se in blood
Lemire, M., Kwan, M., Laouan-Sidi, A.E., Muckle, G., Pirlde, C., Ayotte, P. & Dewailly, E. 2015. Local country food sources of methylmercury, No health outcome measured.
selenium and omega-3 fatty acids in Nunavik, Northern Quebec. Science of the Total Environment, 509: 248-259. https://doi.org/10.1016/j.
scitotenv.2014.07.102
Lemire, M., Mergler, D., Fillion, M., Sousa Passos, C.J., Guimaraes, J.R.D., Davidson, R. & Lucotte, M. 2006. Elevated blood selenium levels in the No health outcome measured.
Brazilian Amazon. Science of the Total Environment, 366(1): 101-111. https://doi.org/10.1016/j.scitotenv.2005.08.057
Lemire, M., Philibert, A., Fillion, M., Passos, C.J.S., Guimaraes, J.R.D., Barbosa, F. & Mergler, D. 2012. No evidence of selenosis from a selenium-rich Only selenium measured (not
diet in the Brazilian Amazon. Environment International, 40: 128-136. https://doi.org/10.1016/j.envint.2011.07.005 mercury)
Li, Y.F., Chen, C.Y., Li, B., Wang, Q., Wang, J.X., Gao, Y.X., Zhao, Y.L. & Chai, Z.F. 2007. Simultaneous speciation of selenium and mercury in human Method development and
urine samples from long-term mercury-exposed populations with supplementation of selenium-enriched yeast by HPLC-ICP-MS. Journal of specification
Analytical Atomic Spectrometry, 22(8): 925-930. https://doi.org/10.1039/b703310a
Lindberg, A., Björnberg, K.A., Vahter, M. & Berglund, M. 2004. Exposure to methylmercury in non-fish-eating people in Sweden. Environ Res, 96(1): No health outcome measured
28-33. https://doi.org/10.1016/j.envres.2003.09.005
Little, M., Achouba, A., Dumas, P., Ouellet, N., Ayotte, P. & Lemire, M. 2019. Determinants of selenoneine concentration in red blood cells of Inuit No health outcome measured
from Nunavik (Northern Quebec, Canada). Environment International, 127: 243-252. https://doi.org/10.1016/j.envint.2018.11.077
Lubick, N. 2010. A balanced diet? Selenium may offset the effects of methylmercury on cataract development. Environ Health Perspect, 118(11): News article
A491. https://doi.org/10.1289/ehp.118-a491b
Marumoto, M., Sakamoto, M., Marumoto, K., Tsuruta, S. & Komohara, Y. 2020. Mercury and Selenium Localization in the Cerebrum, Cerebellum, A single toxicology case
Liver, and Kidney of a Minamata Disease Case. Acta Histochem Cytochem, 53(6): 147-155. https://doi.org/10.1267/ahc.20-00009
Miklavčič, A., Casetta, A., Snoj Tratnik, J., Mazej, D., Krsnik, M., Mariuz, M., Sofianou, K., Spirić, Z., Barbone, F. & Horvat, M. 2013. Mercury, arsenic No health outcome measured
and selenium exposure levels in relation to fish consumption in the Mediterranean area. Environ Res, 120: 7-17. https://doi.org/10.1016/j.
envres.2012.08.010
Muckle, G., Ayotte, P., Dewailly, E., Jacobson, S.W. & Jacobson, J.L. 2001. Determinants of polychlorinated biphenyls and methylmercury exposure in No health outcome measured
inuit women of childbearing age. Environ Health Perspect, 109(9): 957-63. https://doi.org/10.1289/ehp.01109957
Muckle, G., Ayotte, P., Dewailly, E.E., Jacobson, S.W. & Jacobson, J.L. 2001. Prenatal exposure of the northern Québec Inuit infants to environmental No health outcome measured
contaminants. Environ Health Perspect, 109(12): 1291-9. https://doi.org/10.1289/ehp.011091291
Nakayama, S.F., Iwai-Shimada, M., Oguri, T., Isobe, T., Takeuchi, A., Kobayashi, Y., Michikawa, T. et al. 2019. Blood mercury, lead, cadmium, No health outcome measured
manganese and selenium levels in pregnant women and their determinants: the Japan Environment and Children's Study (JECS). Journal of
Exposure Science and Environmental Epidemiology, 29(5): 633-647. https://doi.org/10.1038/s41370-019-0139-0
Niane, B., Guedron, S., Moritz, R., Cosio, C., Ngom, P.M., Deverajan, N., Pfeifer, H.R. & Pote, J. 2015. Human exposure to mercury in artisanal small- No health outcome measured
scale gold mining areas of Kedougou region, Senegal, as a function of occupational activity and fish consumption. Environmental Science and
Pollution Research, 22(9): 7101-7111. https://doi.org/10.1007/s11356-014-3913-5
Parajuli, R.P., Goodrich, J.M., Chan, H.M., Lemire, M., Ayotte, P., Hegele, R.A. & Basu, N. 2021. Variation in biomarker levels of metals, persistent No health outcome measured
organic pollutants, and omega-3 fatty acids in association with genetic polymorphisms among Inuit in Nunavik, Canada. Environmental Research,
200. https://doi.org/10.1016/j.envres.2021.111393
Pinheiro, M.C.N., Muller, R.C.S., Sarkis, J.E., Vieira, J.L.F., Oikawa, T., Gomes, M.S.V., Guimaraes, G.A., do Nascimento, J.L.M. & Silveira, L.C.L. No health outcome measured
2005. Mercury and selenium concentrations in hair samples of women in fertile age from Amazon riverside communities. Science of the Total
Environment, 349(1-3): 284-288. https://doi.org/10.1016/j.scitotenv.2005.06.026
Sakamoto, M., Chan, H.M., Domingo, J.L., Koriyama, C. & Murata, K. 2018. Placental transfer and levels of mercury, selenium, vitamin E, and No health outcome measured
docosahexaenoic acid in maternal and umbilical cord blood. Environ Int, 111: 309-315. https://doi.org/10.1016/j.envint.2017.11.001
Sakamoto, M., Chan, H.M., Domingo, J.L., Kubota, M. & Murata, K. 2012. Changes in body burden of mercury, lead, arsenic, cadmium and No health outcome measured
selenium in infants during early lactation in comparison with placental transfer. Ecotoxicol Environ Saf, 84: 179-84. https://doi.org/10.1016/j.
ecoenv.2012.07.014
Sakamoto, M., Haraguchi, K., Tatsuta, N., Nakai, K., Nakamura, M. & Murata, K. 2021. Plasma and red blood cells distribution of total mercury, No health outcome measured
inorganic mercury, and selenium in maternal and cord blood from a group of Japanese women. Environ Res, 196: 110896. https://doi.org/10.1016/j.
envres.2021.110896
Sakamoto, M., Murata, K., Kubota, M., Nakai, K. & Satoh, H. 2010. Mercury and heavy metal profiles of maternal and umbilical cord RBCs in No health outcome measured
Japanese population. Ecotoxicol Environ Saf, 73(1): 1-6. https://doi.org/10.1016/j.ecoenv.2009.09.010
Sekovanic, A., Piasek, M., Orct, T., Grgec, A.S., Saric, M.M., Stasenko, S. & Jurasovic, J. 2020. Mercury Exposure Assessment in Mother-Infant Pairs No health outcome measured
from Continental and Coastal Croatia. Biomolecules, 10(6). https://doi.org/10.3390/biom10060821
Ser, P.H., Omi, S., Shimizu-Furusawa, H., Yasutake, A., Sakamoto, M., Hachiya, N., Konishi, S., Nakamura, M. & Watanabe, C. 2017. Differences in No health outcome measured
the responses of three plasma selenium-containing proteins in relation to methylmercury-exposure through consumption of fish/whales. Toxicol
Lett, 267: 53-58. https://doi.org/10.1016/j.toxlet.2016.12.001
Suzuki, T., Hongo, T., Yoshinaga, J., Imai, H., Nakazawa, M., Matsuo, N. & Akagi, H. 1993. The hair-organ relationship in mercury concentration in No health outcome measured
contemporary Japanese. Arch Environ Health, 48(4): 221-9. https://doi.org/10.1080/00039896.1993.9940363

455
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A6.2 PRIMARY STUDIES EXCLUDED DURING FULL-TEXT SCREENING OF HUMAN STUDIES FOR THE REVIEW “Se AND MeHg”,
BASED ON INCLUSION AND EXCLUSION CRITERIA (cont.)
Studies (n = 68) Reason for exclusion
Svensson, B.G., Nilsson, A., Jonsson, E., Schütz, A., Akesson, B. & Hagmar, L. 1995. Fish consumption and exposure to persistent organochlorine No health outcome measured
compounds, mercury, selenium and methylamines among Swedish fishermen. Scand J Work Environ Health, 21(2): 96-105. https://doi.org/10.5271/
sjweh.16
Svensson, B.G., Schütz, A., Nilsson, A., Akesson, I., Akesson, B. & Skerfving, S. 1992. Fish as a source of exposure to mercury and selenium. Sci No health outcome measured
Total Environ, 126(1-2): 61-74. https://doi.org/10.1016/0048-9697(92)90484-a
Trdin, A., Snoj Tratnik, J., Mazej, D., Fajon, V., Krsnik, M., Osredkar, J., Prpić, I. et al. 2019. Mercury speciation in prenatal exposure in Slovenian and No use of Se data
Croatian population - PHIME study. Environ Res, 177: 108627. https://doi.org/10.1016/j.envres.2019.108627
Trdin, A., Snoj Tratnik, J., Stajnko, A., Marc, J., Mazej, D., Sešek Briški, A., Kastelec, D. et al. 2020. Trace elements and APOE polymorphisms in No health outcome measured
pregnant women and their new-borns. Environ Int, 143: 105626. https://doi.org/10.1016/j.envint.2020.105626
Valent, F., Pisa, F., Mariuz, M., Horvat, M., Gibicar, D., Fajon, V., Mazej, D., Daris, F. & Barbone, F. 2011. Fetal and perinatal exposure to mercury and Article in Italian
selenium: baseline evaluation of a cohort of children in Friuli Venezia Giulia, Italy. Epidemiol Prev, 35(1): 33-42.
Valera, B., Muckle, G., Poirier, P., Jacobson, S.W., Jacobson, J.L. & Dewailly, E. 2012. Cardiac autonomic activity and blood pressure among Inuit Selenium not measured
children exposed to mercury. Neurotoxicology, 33(5): 1067-1074. https://doi.org/10.1016/j.neuro.2012.05.005
Vecchi Brumatti, L., Rosolen, V., Mariuz, M., Piscianz, E., Valencic, E., Bin, M., Athanasakis, E. et al. 2021. Impact of Methylmercury and Other Study protocol only
Heavy Metals Exposure on Neurocognitive Function in Children Aged 7 Years: Study Protocol of the Follow-up. J Epidemiol, 31(2): 157-163. https://
doi.org/10.2188/jea.JE20190284
Walker, J.B., Houseman, J., Seddon, L., McMullen, E., Tofflemire, K., Mills, C., Corriveau, A. et al. 2006. Maternal and umbilical cord blood levels No health outcome measured
of mercury, lead, cadmium, and essential trace elements in Arctic Canada. Environmental Research, 100(3): 295-318. https://doi.org/10.1016/j.
envres.2005.05.006
Wells, E.M., Herbstman, J.B., Lin, Y.H., Hibbeln, J.R., Halden, R.U., Witter, F.R. & Goldman, L.R. 2017. Methyl mercury, but not inorganic mercury, Selenium as a covariate in a
associated with higher blood pressure during pregnancy. Environ Res, 154: 247-252. https://doi.org/10.1016/j.envres.2017.01.013 multi model
Wilhelm, M., Wittsiepe, J., Schrey, P., Lajoie-Junge, L. & Busch, V. 2003. Dietary intake of arsenic, mercury and selenium by children from a German No health outcome measured
North Sea island using duplicate portion sampling. Journal of Trace Elements in Medicine and Biology, 17(2): 123-132. https://doi.org/10.1016/
s0946-672x(03)80008-1
Yalcin, S.S., Yurdakok, K., Yalcin, S., Engur-Karasimav, D. & Coskun, T. 2010. Maternal and environmental determinants of breast-milk mercury No health outcome measured
concentrations. Turkish Journal of Pediatrics, 52(1): 1-9. <Go to ISI>://WOS:000276572900001
Yamashita, M., Yamashita, Y., Ando, T., Wakamiya, J. & Akiba, S. 2013. Identification and determination of selenoneine, 2-selenyl-N α , N α , N α No health outcome measured
-trimethyl-L-histidine, as the major organic selenium in blood cells in a fish-eating population on remote Japanese Islands. Biol Trace Elem Res,
156(1-3): 36-44. https://doi.org/10.1007/s12011-013-9846-x
Yoo, Y.C., Lee, S.K., Yang, J.Y., Kim, K.W., Lee, S.Y., Oh, S.M. & Chung, K.H. 2002. Interrelationship between the concentration of toxic and essential No health outcome measured
elements in Korean tissues. Journal of Health Science, 48(2): 195-200. <Go to ISI>://WOS:000174717400015
Yoshinaga, J., Matsuo, N., Imai, H., Nakazawa, M., Suzuki, T., Morita, M. & Akagi, H. 1990. Interrelationship between the concentrations of some No health outcome measured
elements in the organs of Japanese with special reference to selenium-heavy metal relationships. Sci Total Environ, 91: 127-40. https://doi.
org/10.1016/0048-9697(90)90294-5
Zhang, M.Y., Buckley, J.P., Liang, L.M., Hong, X.M., Wang, G.Y., Wang, M.C., Wills-Karp, M., Wang, X.B. & Mueller, N.T. 2022. A metabolome-wide No health outcome measured
association study of in utero metal and trace element exposures with cord blood metabolome profile: Findings from the Boston Birth Cohort.
Environment International, 158. https://doi.org/10.1016/j.envint.2021.106976
Zhang, M.Y., Liu, T.G., Wang, G.Y., Buckley, J.P., Guallar, E., Hong, X.M., Wang, M.C., Wills-Karp, M., Wang, X.B. & Mueller, N.T. 2021. In Utero Effect selenium has on Hg
Exposure to Heavy Metals and Trace Elements and Childhood Blood Pressure in a US Urban, Low-Income, Minority Birth Cohort. Environmental toxicity not measured
Health Perspectives, 129(6). https://doi.org/10.1289/ehp8325

456
 QUALITY ASSESSMENT (RISK OF BIAS) OF HUMAN PRIMARY STUDIES
TABLE A6.3 QUALITY ASSESSMENT (RISK OF BIAS) OF HUMAN PRIMARY STUDIES
APPENDICES

Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (n = 45) (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Ai, C.E., Li, C.J., Tsou, M.C., Chen, J.L., Hsi, H.C. & Chien, L.C. 2019. Blood and seminal plasma mercury levels and predatory fish intake in relation to
+ + + ++ - ++ - Tier 2
low semen quality. Environmental Science and Pollution Research, 26(19):19425-19433.
Ayotte, P., Carrier, A., Ouellet, N., Boiteau, V., Abdous, B., Sidi, E.A.L., Dewailly, É. et al. 2011. Relation between methylmercury exposure and
+ + + ++ ++ ++ + Tier 1
plasma paraoxonase activity in inuit adults from Nunavik. Environmental health perspectives, 119(8):1077-1083.
Bélanger, M.C., Dewailly, É., Berthiaume, L., Noël, M., Bergeron, J., Mirault, M.É. & Julien, P. 2006. Dietary contaminants and oxidative stress in
+ + ++ + ++ ++ ++ Tier 1
Inuit of Nunavik. Metabolism, 55(8):989-995.
Bélanger, M.C., Mirault, M.É., Dewailly, E., Berthiaume, L. & Julien, P. 2008. Environmental contaminants and redox status of coenzyme Q10
+ + ++ + ++ ++ ++ Tier 1
and vitamin E in Inuit from Nunavik. Metabolism, 57(7):927-933.
Bélanger, M.C., Mirault, M.E., Dewailly, E., Plante, M., Berthiaume, L., Noël, M. & Julien, P. 2008. Seasonal mercury exposure and oxidant-
- - ++ + ++ ++ + Tier 2
antioxidant status of James Bay sport fishermen. Metabolism, 57(5):630-636.
Boucher, O., Muckle, G., Jacobson, J.L., Carter, R.C., Kaplan-Estrin, M., Ayotte, P., Jacobson, S.W. 2014. Domain-specific effects of prenatal
exposure to PCBs, mercury, and lead on infant cognition: results from the Environmental Contaminants and Child Development Study in ++ + ++ + + + ++ Tier 1
Nunavik. Environmental health perspectives, 122(3):310-316.
Chen, C., Xun, P., McClure, L.A., Brockman, J., MacDonald, L., Cushman, M., He, K. et al. 2018. Serum mercury concentration and the risk of
+ + ++ + + ++ + Tier 1
ischemic stroke: the reasons for geographic and racial differences in stroke trace element study. Environment international, 117:125-131.
Chen, Z., Myers, R., Wei, T., Bind, E., Kassim, P., Wang, G. Wang, X. et al. 2014. Placental transfer and concentrations of cadmium, mercury,
-- + - + - - ++ Tier 2
lead, and selenium in mothers, newborns, and young children. Journal of exposure science & environmental epidemiology, 24(5):537-544.
Choi, A.L., Budtz-Jørgensen, E., Jørgensen, P.J., Steuerwald, U., Debes, F., Weihe, P. & Grandjean, P. 2008. Selenium as a potential protective
- + ++ + ++ ++ ++ Tier 2
factor against mercury developmental neurotoxicity. Environmental Research, 107(1):45-52.
Emanuele, E. & Meliker, J. 2017. Seafood intake, polyunsaturated fatty acids, blood mercury, and serum C-reactive protein in US National
+ - ++ + ++ ++ + Tier 2
Health and Nutrition Examination Survey (2005–2006). International Journal of Environmental Health Research, 27(2):136-143.
Engström, K.S., Wennberg, M., Strömberg, U., Bergdahl, I.A., Hallmans, G., Jansson, J.H., Broberg, K. et al. 2011. Evaluation of the impact of
genetic polymorphisms in glutathione-related genes on the association between methylmercury or n-3 polyunsaturated long chain fatty acids ++ ++ ++ ++ + ++ ++ Tier 1
and risk of myocardial infarction: a case-control study. Environmental health, 10(1):1-8.
Fillion, M., Lemire, M., Philibert, A., Frenette, B., Weiler, H.A., Deguire, J.R., et al. Mergler, D. 2011. Visual acuity in fish consumers of the
++ + + ++ - ++ + Tier 1
Brazilian Amazon: risks and benefits from local diet. Public health nutrition, 14(12):2236-2244.
Fillion, M., Lemire, M., Philibert, A., Frenette, B., Weiler, H.A., Deguire, J.R., Mergler, D. et al. 2013. Toxic risks and nutritional benefits of
++ + + ++ - ++ + Tier 1
traditional diet on near visual contrast sensitivity and color vision in the Brazilian Amazon. Neurotoxicology, 37:173-181.
Golding, J., Gregory, S., Emond, A., Iles-Caven, Y., Hibbeln, J., Taylor, C.M. 2016. Prenatal mercury exposure and offspring behaviour in
++ + ++ ++ ++ + ++ Tier 1
childhood and adolescence. Neurotoxicology, 57:87-94.
Golding, J., Gregory, S., Iles-Caven, Y., Hibbeln, J., Emond, A., Taylor, C.M. 2016. Associations between prenatal mercury exposure and early
++ + ++ ++ ++ + ++ Tier 1
child development in the ALSPAC study. Neurotoxicology, 2016;53:215-22.
Golding, J., Hibbeln, J.R., Gregory, S.M., Iles-Caven, Y., Emond, A. & Taylor, C.M. 2017. Maternal prenatal blood mercury is not adversely
associated with offspring IQ at 8 years provided the mother eats fish: a British prebirth cohort study. International journal of hygiene and ++ + ++ -- + + - Tier 2
environmental health, 220(7):1161-1167.
Gregory, S., Iles-Caven, Y., Hibbeln, J.R., Taylor, C.M. & Golding, J. 2016. Are prenatal mercury levels associated with subsequent blood
++ + ++ - ++ + ++ Tier 1
pressure in childhood and adolescence? The Avon prebirth cohort study. BMJ open, 6(10):e012425.

457
458
TABLE A6.3 QUALITY ASSESSMENT (RISK OF BIAS) OF HUMAN PRIMARY STUDIES (cont.)
Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
Study (n = 45) (Selective Tier
(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Gustin, K., Barman, M., Skröder, H., Jacobsson, B., Sandin, A., Sandberg, A.S., Kippler, M. et al. 2021. Thyroid hormones in relation to toxic metal
++ + ++ ++ ++ ++ ++ Tier 1
exposure in pregnancy, and potential interactions with iodine and selenium. Environment International, 157:106869.
Hu, X.F., Eccles, K.M. & Chan, H.M. 2017. High selenium exposure lowers the odds ratios for hypertension, stroke, and myocardial infarction
+ + - ++ + - + Tier 2
associated with mercury exposure among Inuit in Canada. Environment International, 102:200-206.
Hui, L.L., Chan, M.H.M., Lam, H.S., Chan, P.H.Y., Kwok, K.M., Chan, I.H.S, Fok, T.F. et al. 2016. Impact of fetal and childhood mercury exposure
++ - + + - + + Tier 2
on immune status in children. Environmental research, 144:66-72.
Karimi, R., Vacchi-Suzzi, C. & Meliker, J.R. 2016. Mercury exposure and a shift toward oxidative stress in avid seafood consumers.
+ + + + + + + Tier 1
Environmental research, 146:100-107.
Kobayashi, S., Kishi, R., Saijo, Y., Ito, Y., Oba, K., Araki, A., Children's Study Group et al.. 2019. Association of blood mercury levels during pregnancy with infant
++ ++ + + + ++ - Tier 1
birth size by blood selenium levels in the Japan Environment and Children's Study: A prospective birth cohort. Environment international, 125:418-429.
Kuras, R., Kozlowska, L., Reszka, E., Wieczorek, E., Jablonska, E., Gromadzinska, J., Wasowicz, W. et al. 2019. Environmental mercury exposure and selenium-
+ ++ + ++ + + + Tier 1
associated biomarkers of antioxidant status at molecular and biochemical level. A short-term intervention study. Food and Chemical Toxicology, 130:187-198.
Kuras, R., Reszka, E., Wieczorek, E., Jablonska, E., Gromadzinska, J., Malachowska, B., Wasowicz, W. et al. 2018. Biomarkers of selenium status
++ + - ++ + ++ + Tier 2
and antioxidant effect in workers occupationally exposed to mercury. Journal of Trace Elements in Medicine and Biology, 49:43-50.
Lei, H.L., Wei, H.J., Chen, P.H., Hsi, H.C. & Chien, L.C. 2015. Preliminary study of blood methylmercury effects on reproductive hormones and
-- ++ + -- + + - Tier 2
relevant factors among infertile and pregnant women in Taiwan. Chemosphere, 135:411-417.
Lemire, M., Fillion, M., Frenette, B., Mayer, A., Philibert, A., Passos, C.J.S., Mergler, D. et al. 2010. Selenium and mercury in the Brazilian
+ ++ ++ + + + + Tier 1
Amazon: opposing influences on age-related cataracts. Environmental Health Perspectives, 118(11):1584-1589.
Lemire, M., Fillion, M., Frenette, B., Passos, C.J.S., Guimarães, J.R.D., Barbosa Jr, F. & Mergler, D. 2011. Selenium from dietary sources and
+ ++ + + + - + Tier 1
motor functions in the Brazilian Amazon. Neurotoxicology, 32(6):944-953.
Maeda, E., Murata, K., Kumazawa, Y., Sato, W., Shirasawa, H., Iwasawa, T., Terada, Y. 2019. Associations of environmental exposures to
- + ++ -- + -- -- Tier 2
methylmercury and selenium with female infertility: A case–control study. Environmental research, 168:357-363.
Mao, X., Chen, C., Xun, P., Daviglus, M., Steffen, L.M., Jacobs Jr, D.R., He, K. 2019. Effects of seafood consumption and toenail mercury and
+ - ++ + + + + Tier 1
selenium levels on cognitive function among American adults: 25 y of follow up. Nutrition, 61:77-83.
Monastero, R.N., Karimi, R., Nyland, J.F., Harrington, J., Levine, K. & Meliker, J.R. 2017. Mercury exposure, serum antinuclear antibodies, and serum
+ + ++ + + ++ + Tier 1
cytokine levels in the Long Island Study of Seafood Consumption: A cross-sectional study in NY, USA. Environmental research, 156:334-340.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Mozaffarian, D., Shi, P., Morris, J.S., Grandjean, P., Siscovick, D.S., Spiegelman, D., Forman, J.P. 2012. Mercury exposure and risk of
++ - ++ + + + + Tier 2
hypertension in US men and women in 2 prospective cohorts. Hypertension, 60(3):645-652.
Mozaffarian, D., Shi, P., Morris, J.S., Spiegelman, D., Grandjean, P., Siscovick, D.S., Rimm, E.B. 2011. Mercury exposure and risk of
++ - + + + + + Tier 2
cardiovascular disease in two US cohorts. New England Journal of Medicine, 364(12):1116-1125.
Nakamura, M., Hachiya, N., Murata, K.Y., Nakanishi, I., Kondo, T., Yasutake, A., Sakamoto, M. et al. 2014. Methylmercury exposure and
+ - ++ - - + - Tier 2
neurological outcomes in Taiji residents accustomed to consuming whale meat. Environment international, 68:25-32.
Nyland, J.F., Fillion, M., Barbosa Jr, F., Shirley, D.L., Chine, C., Lemire, M., Silbergeld, E.K. et al. 2011. Biomarkers of methylmercury exposure
+ ++ + + - ++ + Tier 2
immunotoxicity among fish consumers in Amazonian Brazil. Environmental health perspectives, 119(12):1733-1738.
Oken, E., Rifas-Shiman, S.L., Amarasiriwardena, C., Jayawardene, I., Bellinger, D.C., Hibbeln, J.R., Gillman, M.W. et al. 2016. Maternal prenatal
+ + + + - ++ + Tier 1
fish consumption and cognition in mid childhood: mercury, fatty acids, and selenium. Neurotoxicology and teratology, 57:71-78.
Park, K. & Seo, E. 2016. Association between toenail mercury and metabolic syndrome is modified by selenium. Nutrients, 8(7):424.
+ + + + + ++ + Tier 1
 TABLE A6.3 QUALITY ASSESSMENT (RISK OF BIAS) OF HUMAN PRIMARY STUDIES (cont.)
Q6
KQ1 KQ2 KQ3 Q4 Q5 Q7
(Selective
APPENDICES

Study (n = 45) Tier

(Confounding) (Exposure) (Outcome) (Selection) (Attrition) (Other bias)
reporting)

Park, K. & Seo, E. 2017. Toenail mercury and dyslipidemia: Interaction with selenium. Journal of Trace Elements in Medicine and Biology,
+ + + + + + + Tier 1
39:43-49.
Park, S.K., Lee, S., Basu, N. & Franzblau, A. 2013. Associations of blood and urinary mercury with hypertension in US adults: the NHANES
++ + + + + + + Tier 1
2003–2006. Environmental research, 123:25-32.
Rocha, A.V., Cardoso, B.R., Zavarize, B., Almondes, K., Bordon, I., Hare, D.J., Cozzolino, S.M.F. et al. 2016. GPX1 Pro198Leu polymorphism and GSTM1 deletion
+ + + - + + - Tier 2
do not affect selenium and mercury status in mildly exposed Amazonian women in an urban population. Science of the Total Environment, 571:801-808.
Saint-Amour, D., Roy, M.S., Bastien, C., Ayotte, P., Dewailly, E., Després, C., Muckle, G. et al. 2006. Alterations of visual evoked potentials in
+ ++ + - + ++ + Tier 1
preschool Inuit children exposed to methylmercury and polychlorinated biphenyls from a marine diet. Neurotoxicology, 27(4):567-578.
Steuerwald, U., Weihe, P., Jørgensen, P.J., Bjerve, K., Brock, J., Heinzow, B. Grandjean, P. 2000. Maternal seafood diet, methylmercury exposure,
+ ++ + + + + - Tier 1
and neonatal neurologic function. The Journal of pediatrics, 136(5):599-605.
Tatsuta, N., Murata, K., Iwai-Shimada, M., Yaginuma-Sakurai, K., Satoh, H. & Nakai, K. 2017. Psychomotor ability in children prenatally exposed to
+ + + + + + - Tier 1
methylmercury: the 18-month follow-up of Tohoku study of child development. The Tohoku Journal of Experimental Medicine, 242(1):1-8.
Tratnik, J.S., Falnoga, I., Trdin, A., Mazej, D., Fajon, V., Miklavčič, A., Horvat, M. et al. 2017. Prenatal mercury exposure, neurodevelopment and
+ ++ + + - + + Tier 1
apolipoprotein E genetic polymorphism. Environmental research, 152:375-385.
Walker, E.V., Girgis, S., Yuan, Y. & Goodman, K.J. 2021. Community-driven research in the Canadian arctic: dietary exposure to methylmercury
- -- - -- + - - Tier 3
and gastric health outcomes. International Journal of Circumpolar Health, 80(1):1889879.
Wells, E.M., Herbstman, J.B., Lin, Y.H., Jarrett, J., Verdon, C.P., Ward, C., Goldman, L.R. et al. 2016. Cord blood methylmercury and fetal growth outcomes in Baltimore
+ - + - + + + Tier 2
newborns: potential confounding and effect modification by omega-3 fatty acids, selenium, and sex. Environmental health perspectives, 124(3):373-379.
Wennberg, M., Bergdahl, I.A., Hallmans, G., Norberg, M., Lundh, T., Skerfving, S., Jansson, J.H. et al. 2011. Fish consumption and myocardial
+ + + + - - + Tier 1
infarction: a second prospective biomarker study from northern Sweden. The American journal of clinical nutrition, 93(1):27-36.
Zhang, J., Wang, J., Hu, J., Zhao, J., Li, J. & Cai, X. 2021. Associations of total blood mercury and blood methylmercury concentrations with
- -- - + + - - Tier 3
diabetes in adults: an exposure-response analysis of 2005-2018 NHANES. Journal of Trace Elements in Medicine and Biology, 68:126845.

459
460
EXTRACTION OF DATA FROM HUMAN PRIMARY STUDIES
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg”
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Ai et al. 2019. Cross-sectional N = 84 ICP-MS ICP-MS Sperm quality according to WHO Dose-dependent correlation Tier 2 No effect of Se on Hg effect
Taiwan Multiple linear 37 years (average) Mean ±SD, range Mean ±SD, range reference values was used to between blood Hg and normal on semen quality. Study
regression 26-52 (range) divide into high sperm count. group small.
May 2012–February Hg, µg/L Se, µg/L (N = 27) and low (N = 57)
2013 100% No significant difference
Blood (N = 84) Blood (N = 84) semen quality groups. between high and low quality
Participants from Center 9 ±5.9 205 ±58 Measurement of Hg and Se levels
for Reproductive Medicine semen for Se. Weekly fresh fish
0.3-31.3 112-39 in blood and sperm plasma. Fish intake in low-semen-quality
and Science, Taipei Medical consumption by questionnaire.
University, Taiwan, without Low quality semen group Low quality semen group group 1.9 ±0.5 meals per
sub- or infertility diagnosis, 9.3 ±5.9 206±60.9 Categorization of Hg levels into week not significantly different
had regular sex for one year High quality semen group High quality semen group high, medium and low. from high-semen-quality
and their female partners did 8.9 ±5.9 205±57.8 group, 2.3±1.2 meals/week.
not get pregnant High predatory fish intake had
Seminal plasma (N = 39) Seminal plasma (N = 39) lower percentage of normal
1.12±0.56 82.3±24.0 morphology in sperm, highest
0.2-2.48 27.6-152 percentage with no predatory
Low quality semen group Low quality semen group fish intake. Same trend for
1.26 ±0.61 73.5±22.3 sperm plasma Hg levels, but
not significant.
High quality semen group High quality semen group
1.05 ±0.52 87.3±23.9
Ayotte et al. 2011. Cross-sectional, N = 896 Blood Hg Blood Se Coronary heart disease (CHD) Associated with PON1 Tier 1 MeHg exposure seems to
None community-stratified Mean 36.5 years ICP MS ICP-MS protective marker paraoxonase 1 activities was inverse for exert an inhibitory effect on
random sampling of Geometric mean: 53.2 nmol/L Geometric mean: 3.8 µmol/L (PON1) activity, has been shown blood Hg concentrations, PON1 activity, which seems
Inuit with high households. N= 405 male and N= 896 to be inhibited by MeHg whereas positive for blood Se to be offset by Se intake.
seafood diet of female (0.4-720 nmol/L) (1.5-23 µmol/L)
Multivariate concentrations.
Nunavik (Québec,
Canada). analyses/ multiple
regression
2004 adjusted for age,
HDL cholesterol
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

levels, omega-3
fatty acid content
of erythrocyte
membranes and
PON1 variants.
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Bélanger et al. 2006. Cross-sectional N = 99 Blood Hg Blood Se Effects on molecules sensitive to Concentration of plasma Tier 1 In adult Inuit, MeHg from
None Randomly selected Mean ±sem Cold vapor atomic absorption ICP MS oxidative stress: plasma oxidized homocysteine (Hcy) was traditional diet seems to
from the municipal 43 ±1 year technique Mean ±SEM low-density lipoproteins (OxLDLs), negatively predicted by have no direct oxidative
Canadian Northern plasma homocysteine (Hcy), Se, suggesting a possible effects on molecules
Inuit village of list. 71f/28m Mean ±SEM 635.5 ±38.7 µg/L
106.2 ±9.8 nmol/L blood glutathione peroxidase beneficial effect of Se. Dietary involved in oxidative stress
Salluit (Nunavik, Pairwise correlations 28% calculated (GPx), glutathione reductase (GR) MeHg showed no association. known to be involved
Northern Quebec) with log transformed and glutathione (GSH). in the pathogenesis
No information data combined with of atherosclerosis.
multivariate analysis Concentration of plasma
using stepwise homocysteine was
models for each negatively predicted
dependent variable by Se, suggesting a
Time point possible beneficial effect
of Se as it can activate
pathogenic mechanisms
leading to oxidative stress
including a decrease in
vascular reactivity and
cardiovascular morbidity
and mortality.
Bélanger et al. 2008. Cohort N = 99 Blood Hg Blood Se Assess oxidative stress Ubiquinol-10 (beta= .23, Tier 1 Low susceptibility of Inuit to
None Randomly selected Mean ±SD Cold vapor atomic absorption ICP MS by measuring plasma P=.007) and Co10 total (beta= LDL oxidation may be partly
from the municipal 45 ±13 years technique Mean ±SEM concentrations and redox states .27, P=.009) were predicted by explained by high blood Se
Canadian Northern of alfa-tocopherol and coenzyme blood Se. No evidence for MeHg status that might reduce
Inuit village of list. 71/28 Mean ±SEM 635.5 ±38.7 µg/L
106.2 ±9.8 nmol/L Q10 (CoQ10), two sensitive associated oxidative stress. the deleterious effects of
Salluit (Nunavik, Multivariate 28% calculated biomarkers of oxidative stress MeHg on cardiovascular
Northern Quebec) analyses. health.
No information Time point
Bélanger et al. 2008. Healthy sport N = 31 Cold vapor atomic absorption Blood Assess effects of seasonal Homocysteine did not change. Tier 2 Seasonal fishing activity/
None fishermen 22-61 years technique ICP-MS exposure to Hg through No association between Hg fish consumption had
Matched pair tests Mean ±SEM Hair Hg Blood Se lipoprotein cholesterol and fatty and any of the biomarkers beneficial effects on
Healthy sport acid profiles, LDL oxidation, investigated. Strong predictors cardiovascular health, and
fisherman working at Six months 46.7 ±1.3 Mean ±SEM Mean ±SEM and blood oxidant-antioxidant of cardiovascular risk such as may suppress detrimental
James Bay (Northern 100% 1.4 ±0.3 µg/g 242.9 ±6.2 µg/L balance. HDL cholesterol, oxidized LDL effects of moderate MeHg
Québec) as white- 2.8 ±0.4 µg/g 247.7 ±5.6 µg/L and glutathione peroxidase exposure.
collar employees of Within subject longitudinal
Blood Hg seasonal variations in hair and improved during fishing
Hydro-Québec. Mean ±SEM season.
blood mercury, plasma oxidized
No information 21.9 ±3.7 nmol/L LDL, lipophilic antioxidants,
35.6 ±5.2 nmol/l homocysteine, blood Se,
glutathione peroxidase and
reductase activities before and
after fishing season.

461
462
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Boucher et al. 2014. Repeated measures, 94 infants Umbilical cord blood Umbilical cord blood Hg, Se, Fagan Test of Infant Hg moderately correlated with Tier 1 No association of Se
None multiple regression 6.5 months (Method in Muckle et al., 2001) (Method in Muckle et al 2001) Intelligence (FTII), A-not-B test Se. Small correlation (A not with most intelligence
November 1995 11 months Hg Se and Bayley Scales of Infant B test – negative correlation parameters negatively
Pregnant women Development – 2nd Edition with perseverative errors with associated with Hg: A-not
from Inuit villages in to March 2001 63.8% Mean ±SD Mean ±SD
(follow-up time: 4.5 (BSID-II) Se) Pearson coefficient=0.21, B, which depends on
Hudson Bay coast 22.5 ±16.6 µg/L 296.4 ±122.8 µg/L p<0.1. Otherwise, no working memory and is
months) 2.4-97.3 67.9-915.9 significant association. believed to be a precursor
of executive function,
with the exception of
perseverative errors.
Chen et al. Case-cohort study, 662 stroke incidents randomly Serum Hg Serum Se Association of serum Hg levels No statistically significant Tier 1 This study does not support
2018. multivariate- selected in 2 494 participants Nippon MA-3000 direct Instrumental neutron with ischemic stroke, Se modified association between serum Hg an association between
Reasons for adjusted hazard sub-cohort mercury analyser activation analysis with by demographic and geographic and the incidence of ischemic Hg and the incidence of
Geographic and ratios (HRs) ≥45 standard reference material. factors stroke. stroke within a population
and confidence Median with low-to-moderate Hg
Racial Differences in 45% 0.03 µg/dl (inter-quartile Median Serum Se associated with
Stroke cohort intervals using the serum Hg. Higher serum Hg exposure. Neither does
Barlow-weighting range= 0.02-0.06 µg/dl) 13.1 µg/dl serum Se modify this,
Residents of the US associated with less likely to
method for the Cox have history of myocardial presumably due to the
“stroke belt” and proportional hazards relatively high level of Se in
blacks oversampled. infarction. Inverse association
regression model of serum Hg with ischemic this population.
2003-2007 Time point stroke in women, not men,
Mean ±SD which was not modified by Se.
6 years ±2.4
Chen et al. 2014. Paired maternal 50 mother-infant pairs ICP-MS ICP-MS Premature birth and birthweight. Red blood cells are better Tier 2 Mother–child transfer
Boston Birth cohort umbilical cord (N = Red blood cell Red blood cell Se than plasma or whole of MeHg and Se are best
(BBC) 17) and postnatal Hg µg/L blood in reflecting the free measured in red blood
blood samples 24-72 µg/L transplacental transfer of Hg cells.
African-American geometric mean, range:
hour (N = 39) after geometric mean, range: and Se. No effect of Se on
delivery, infant blood Maternal blood: 2.5 (1.9-3.25)
Maternal blood: 277.81 No strong correlation between premature birth or
at 12.5 months Umbilical cord venous blood: (255.51-302.05) Se and Hg. No effect of Se on birthweight.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

(mean) 6-39 months 3.83 (2.93-5.04) premature birth or birthweight.

(range) Umbilical cord venous blood:
Infant venous blood: 2.17 311.10 (286.48-337.84)
(1.74-2.70)
Infant venous blood: 286.82
(272.93-301.45)
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Choi et al. 2008. Singleton birthed 1 022+182= Cord whole Blood Hg Cord whole blood Se Neurodevelopment within No consistent effects of Se or Tier 2 No evidence that Se was
National Hospital 1986-1987, high 1 104 recruited Cold vapor atomic absorption Electrothermal atomic lowest, middle and highest Se significant interaction between an important protective
Torshavn, Faroe pilot whale diet 880+142 = 1022 measured spectrometry absorption distribution (25%, 50%, 25%) Se and MeHg. factor against MeHg
Islands 7 years for Se Motor function: Finger tapping neurotoxicity. Increased Se
µg/L µg/L levels were not associated
Faroe islands 917 completed evaluation nml/l nml/l and hand eye coordination,
at age 7 attention, visuospatial with decreased mercury-
1994-1995 geometric mean geometric mean related neuropsychological
IQR IQR performance, language
dysfunctions.
Cohort 1 Cohort 1 Hg/Se ratio
22.9 111.6 Cohort 1
13.4-41.3 100.8-123.1 0.08
114.3 1412.5 1276-1558 0.05-0.15
67-206.5 Cohort 2 Cohort 2
Cohort 2 102.5 0.08
20.9 93.2-112.0 0.05-0.14
12.5-40 1297.2
102.7 1180-1418
62.4-200
Emanuele 2017. Cross-sectional, 1 217 ICP-MS Mcg from dietary interview Seafood consumption, Hg, Se, No associations of Hg, Se Tier 2 Suggested confounding
US National Health sex-stratified sample 16-49 years Hg, µg/L Se µg/L polyunsaturated fatty acids or fish intake with serum in models that do not
and Nutrition weighted multiple Mean 34 and vitamin D concentrations C-reactive protein in sex- mutually adjust for
linear regression Blood 143 in relation to C-reactive stratified sample weighted seafood contaminants and
Examination Survey 706 male 511 non-pregnant within range: % male; 18-578 (male)
(NHANES) protein (mg/dl), an indicator of multiple linear regression. nutrients.
female, fasting % female 97 inflammation and predictor of When all variables included
2005-2006 3-406 (female) Effect of Se on Hg toxicity
58% <1: cardiovascular disease. in one model, fish intake was not investigated.
48.7%; 46.1% associated with lower levels
1-5.8: of CRP in females. When
47.6%; 50.5% controlling for seafood intake,
no association for CRP with Se
>5.8: was found.
3.7%; 3.4%

463
464
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Engström et al. Logistic regression 1 027 (458 cases of myocardial Erythrocyte Hg Erythrocyte-Se To elucidate whether No associations of No associations of
2011. in prospective health infarction and 569 matched In acid-digested samples ICPMS genetic polymorphisms in polymorphisms with myocardial polymorphisms with
surveillance cohort controls) using cold vapor atomic µg/L glutathione-related genes infarction, but study not large myocardial infarction, but
The Västerbotn modify the association enough. Also no effect when study not large enough.
Intervention Program Time point Age and sex distribution not fluorescence spectrometry Mean ±SD
given between eicosapentaenoic and Ery-Se taken into account. No effect when Ery-Se taken
(VIP), the WHO’s µg/L 130 ±37/ docosahexaenoic acid or MeHg
Multinational Ery-Hg associated with into account.
Mean ±SD 130 ±22 and risk of first myocardial decreasing MI risk, most likely
Monitoring of Trends 4.9 ±5.1/ infarction. Ery-Hg associated with
and Determinants (controls/carriers of GCLM TT) because Ery-Hg is a marker for decreasing MI risk, most
5.4 ±5.7 (controls/carriers of fish intake.
in Cardiovascular GCLM TT) 72-710 likely because Ery-Hg is a
Disease (MONICA) marker for fish intake.
Study in northern 0.01-87
Sweden, and the
Mammography
Screening Project
(MSP)
Västerbotn, Northern
Sweden
1987-1999 (medical
history)
Fillion et al. 2011. Cross-sectional 243 adults Hair Hg ICP MS Associations between near Near visual acuity was Tier 1 Hg may affect visual acuity
None study ≥15 years Cold vapor atomic absorption Total blood Se and distal visual acuity and negatively associated with hair in older persons. Blood
Time point spectrometry Mean ±SD biomarkers of Hg and Se from the Hg, ≥40 years of age. Age did Se was not significantly
Lower Tapajós River Mean ±SD local diet. not influence blood Se status. associated with this
Basin (State of Pará, Median Mean ±SD Median
Median Range Blood Se was significantly outcome.
Brazil Range lower in the group excluded
35.9 ±12.3 years Range 313.4 ±215.5 µg/L
May to July 2006 14.4 ±15.5 µg/g 250.6 for age-related cataracts.
35 Log-blood Se did not show
15-66 11.5 103.3-1500.2
1.0-57.9 any association with the
48.1% Plasma Se outcome of acuity. Few people
Total Blood Hg Mean ±SD with Se deficiency were in
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

ICP-MS Median the investigated population

Mean ±SD Range and normal Se levels may be
Median 178.8 ±120.1 µg/L sufficient to maintain visual
Range 141.3 function.
51.6 ±37 53.6-913.2
41.8
1.7-189.3 µg/L
Plasma Hg
Mean ±SD
Median (Range)
8.0 ±6.5 µg/L
6.9 (0.2-30.9)
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Fillion et al. 2013. Cross-sectional 228 participants Hair Hg ICP MS Near visual contrast sensitivity Reduced contrast sensitivity Tier 1 Acquired colour vision loss
None study ≥15 years Cold vapor atomic absorption Plasma Se and colour vision. associated with hair Hg. No increased with hair Hg and
Time point spectrometry Mean ±SD association was observed for decreased with plasma
Lower Tapajós River Mean ±SD near visual contrast sensitivity Se. Associated effect of
Basin (State of Pará, Median Mean ±SD Median
Median Range for plasma Se. Mean colour less colour confusion and
Brazil Range confusion index tended to near visual contrast might
35.3 ±12.5 years Range 182.3 ±124.9 µg/L
May to July 2006 14.4 ±10.6 µg/g 144.5 be negatively associated counteract colour vision
33 with plasma Se. A positive loss associated with Hg.
15-66 11.5 53.6-913.2
1-57.9 association was observed
49.6% between plasma Se and near
visual contrast at 12 cpd.
Intermediate spatial frequency
of contrast sensitivity (12
cycles/degree) positively
associated with plasma
Se. Acquired colour vision
loss increased with hair Hg
and decreased with plasma
Se. Hair Hg was positively
correlated to plasma Se.
Golding et al. 2016. Cohort Prenatal Hg measurements Maternal blood Hg (measured Maternal blood Se (measured Offspring behavioural No significant differences Tier 1 There were no adverse
Avon Longitudinal At 4 years and 16-17 and offspring behaviour results in first half of pregnancy). in first half of pregnancy). assessment: Strengths and between the associations with effects of maternal
Study of Parents and years were available for 2 776 (at 47 Whole blood Hg analysed with Whole blood Se analysed with Difficulties Questionnaire, Hg found among the offspring prenatal mercury levels
Children (ALSPAC) months) and 1 599 (at 16-17 ICP-DRC-MS ICP-DRC-MS. compared with Hg and Se levels of women who ate fish in on the behaviour of the
years). and fish/no fish consumption of pregnancy and those who did offspring, also not when
(Avon, UK) Maternal Hg (µg/L): Maternal Se (µg/L): mothers not, nor did adjustment for Se analyses were confined
Not specified in this article. Median: 1.86
1991-1992 Data available at <http://www. Median: 108 make a difference. to those offspring whose
Range: 0.17-12.76 mothers had eaten fish
bris.ac.uk/alspac/researchers/ Range: 17-324
data-access/data-dictionary/ in pregnancy and no
http://www.bris.ac.uk/alspac/ consistent differences were
researchers/data-access/data- found between fish and
dictionary/>. non-fish eaters.
Their findings were
0% not influenced by the
association between Hg
and blood Se.
Golding et al. 2016. Cohort N = 2 875–3 264 (depending Maternal blood Hg levels Maternal blood Se levels Child neurodevelopment Positive association Tier 1 Se did not
ALSPAC study Pregnancy cohort on outcome) (measured in first half of (measured in first half of measured from questions in between Hg levels and child ameliorate any deleterious
(first half of Mean 28 years pregnancy). pregnancy). the Denver Developmental developmental outcomes. Se effects of Hg on child
(Avon, UK) Screening Test. Included a concentrations made little
pregnancy) and 0% Whole blood Hg analysed with Whole blood Se analysed with development between 6 and
1991-1992 follow-up of children ICP-DRC-MS. ICP-DRC-MS. total developmental score, difference to the size of the 42 months of age.
at ages 6, 18, 30 and different subcategories positive associations between
Maternal Hg (µg/L): Maternal Se (µg/L): (social, fine motor, gross motor, Hg and the developmental
and 42 months. Median: 1.86 Median: 108 language) outcomes.
Range: 0.17-12.76 Range: 17-324

465
466
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Golding et al. 2017. Cohort 2 062 offspring Maternal blood Hg levels Maternal blood Se levels Verbal, performance and total Positive association of Hg Tier 2 Se did not influence positive
ALSPAC study Multiple and logistic Mean age not provided; data (measured in first half of (measured in first half of intelligence quotient (IQ) with intelligence. Level of effect of fish consumption
regression analyses. available in database. pregnancy). pregnancy). the mother’s blood Se did not on intelligence, despite
(Avon, UK) change the effect sizes. correlation of fish
Stratification 0% (of mothers) Whole blood Hg analysed with Whole blood Se analysed with
1991-1992 considered children ICP-DRC-MS. ICP-DRC-MS. consumption with Hg.
of fish eaters Maternal Hg (µg/L): Maternal Se (µg/L):
separately. Median: 1.86 Median: 108
Pregnancy cohort Range: 0.17-12.76 Range: 17-324
Follow-up at age 8 Children: /L
Range
<1.28 - >3.39 µg/L
Gregory et al. 2016. Cohort N = 4 484 for blood sampling Maternal (whole) blood Hg Maternal (whole) blood Se Child blood pressure measured Little evidence that blood Tier 1 Since there was little or
ALSPAC study Pregnancy cohort and then from 2 207 down to levels (measured in first half of levels (measured in first half of (systolic and diastolic) pressure in children was no effect from maternal
1991-1992. 1 268 for child measurements pregnancy). pregnancy). affected by the mother’s blood prenatal blood Hg on the
Avon, the United after 7 to 17 years mercury blood levels or fish resting blood pressure and
Kingdom with follow up after Whole blood Hg analysed with Whole blood Se analysed with
7, 9, 11, 13,15 & ICP-DRC-MS. ICP-DRC-MS. intake. heart rates, there was also
1991-1992, 17 years little effect of Se. Hg and Se
Maternal Hg (µg/L): Maternal Se (µg/L): blood values were closely
Median: 1.86 Median: 108 correlated.
Range: 0.17-12.76 Range: 17-324
Gustin et al. 2021. Cohort N = 655 included but after Blood levels measured in RBC Blood levels measured in Free and total T3 and T4 as Erythrocyte Hg was non-linearly Tier 1 Authors conclude that Hg
NICE cohort Pregnancy cohort, exclusions 542 were analysed (µg/L): was 0.29 to 4.2 with a plasma (µg/L): health outcomes, together with associated with fT3, TT3 and can interfere with thyroid
(Sweeden) only mothers 30 years, mean median of 1.5. 46–93 TSH. fT3:fT4 ratio. No or very slight hormones, but that Se does
median 67 effect of Se was reported. not affect this interference.
2015-2018
Hu et al. 2017. Cohort N = 2 595 agreed to Blood levels of Hg ranged from Blood levels of Se ranged from The health outcome was “self- High Se and low Hg had Tier 2 This study indicates that
The International Inuits aged 18 + participate and 2 169 0.3 to 70 with a median of 150 to 1500 with a median of reported physician-diagnosed the lowest prevalence of exposure to Se in the
Polar Year Inuit years completed individual 7.8 µg/L 280 µg/L condition of myocardial cardiovascular outcomes absence of Hg is associated
Health Study (IHS) questionnaire and had valid The authors divided the infarction, stroke, and high blood (except stroke) and thereby with cardiovascular risk,
blood samples pressure measured.” No time for provide evidence that Se may while high exposure to
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Nunavut ++ participants in 4 groups

depending on values higher the reporting was given. exhibit a protective effect Hg in absence of Se is
Northern Canada against Hg on vascular associated with increased
or lower of median value of
2007–2008 Hg and Se disease. cardiovascular risk.

Hui et al. 2016. Cross-sectional N = 1 057 in the original Cord blood Blood follow-up Se: Cytokines and TNF-alpha as Authors conclude that there Tier 2 In this study, Se seems
The Chinese mercury Follow-up after 6–9 birth cohort. N = 608 met for nmol/L, median (inter- nmol/L, median (inter-quantile) measurements of immune system is a small but significant to alleviate Hg toxicity.
birth cohort years follow-up examination, 407 got quantile) : 1.17 (1.02-1.26) function. association between mercury Although the clinical
all parameters measured. 46.1 (33.1-65.1) This figure is obviously wrong and IL-10 concentration. significance is unclear.
Children born July The association was more The study has some
2000 to December 0, mean age of follow up was Blood follow-up 13.0 and can be used only as
8.3 years. relative values. On molar basis pronounced when selenium shortcomings, which the
2001 (8.63-18.69) and cord blood mercury were review process should have
54% Se is always higher that Hg.
Follow-up after 6–9 Probably was µml/L low. The clinical significance detected.
years is unclear.
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Karimi et al. 2016. Cross-sectional N = 268 for blood samples. Hg whole blood Se whole blood GSH and GSH:GSSG as health Association of high Hg from fish Tier 1 Association between elevated
Long Island, NY, the Recruited avid seafood 290 were recruited. 7.71 ±8,12 (IQR: 2.46, 10.50) 293.54 ±101.99(IQR: 65.79; outcomes and indicative for oxidative consumption and GSH:GSSG. This Hg from seafood and a shift in
United States. eaters from Long 47.88 (mean) ±18.24 (SD) 277.95) stress. was slightly less pronounced in redox-status towards oxidative
Island Also reported omega-3 index. those with highest Se in blood. stress. This is, however,
(2010-2014) Not 42% shifting in a much quicker
clearly provided time frame than the mercury.
Authors rather vague on the
effect of Se.
Kobayashi et al. 2019 Cohort study N that gave blood samples were Blood Hg 4.21 mean Blood Se at 171.4 (mean; Birth weight and “small for No effect on birth weight and Tier 1 Interestingly high levels of
Japanese Environment These mothers were 96 095 (IQR 2.59; 5.18) (IQR158.0; 183.0) Range 99.9 gestational age” as outcomes. “small for gestational age” as blood Hg in the Japanese
and Children Study only followed to Then 20 000 were selected Range 0.334 to 30.1 and median to 371.0) Also birth head circumference was outcomes, but a weak correlation pregnant women. The only
delivery in this study for blood measurement. After 3.66. measured. with head circumference and possible negative effect was
Recruitment January withdrawals etc. 17 998 gave blood Hg. on head circumference, where
2011 to March 2014 blood samples. authors claim that the higher
Further exclusions reduced the Se did not give a high enough
data set to 15 444 samples with protective effect.
the right meta data included.
Mean age of mothers 30.9 ±
4.9 years
Child sex 51.7m / 48.3f

Kuras et al. 2018. Comparative case- N = 131 Hg exposed Median (IQR) Median (IQR) Antioxidant systems activity Authors conclude that there Tier 2 This work shows several
Polish men from control cohort study. 41 (31-51) years Median (IQR) Exposed group Exposed group: and gene expression of central is upregulation of central interesting correlations
Lodz Comparing work- N = 67 non-exposed Hg blood 6.06 (4.01-8.94) Se-Plasma: 82.85 (72.03- proteins in antioxidant system. genes in antioxidant defence between Hg and Se. High
exposed men with 37 (30-54) y Hg Urine 11.84 (7.55-21.04) 90.28) in exposed workers. Also, Hg in exposed workers can
Both groups Se-U: 13.44 (11.53-16.65) some correlation of Hg draw functional Se out of
sampled from May to non-exposed men. Median (IQR) (The latter group Non-exposed
same as next paper.) Hg blood: 0.52 (0.35-0.73) Non-exposed: with selenoprotein-P and the system.
June 2015 antioxidants in blood was
Hg Urine 0.16 (0.12-0.24) Se-Plasma: 72.74 (66.25-
80.14) detected.
Se-U: 11.89 (9.89-14.94)
Kuras et al. 2019. Cohort study N = 67 Blood levels before exposure: Plasma levels of Se: Biomarkers of pro- and Authors claim that their study Tier 1 The paper shows that both
Polish men from Men with different Healthy, non-exposed to Hg W=0 0.623 ±0.412 W=0; 73.3 ±11.6 antioxidant effects TBARS, TAA, confirms the existence of a Hg and Se increase in an
Lodz nutritional status W=1(E): 0.902 ±0.463 W=2(E): W=1 (E); 81.6 ±10.2 GSH-Px biological relationship between intervention with high fish
Mean W=2 (E); 80.8 ±12.3 Hg and Se. intake. There is interaction
Cohort with eating fish daily 41.0 ± 12.7 years 1.282 ±0.487 Gene expression
during two weeks W=6; 0.782 ±0.599 W=6; 79.6 ±11.7 They report an increase in between Hg and Se but
intervention from no direct evidence for a
June 2015 to August and then back to Hair levels before exposure: …........................... TBARS as an effect of fish
normal diet for 4 intake but there is little protective effect of Se.
2015 0.237 ±0.164 Urine levels of Se:
weeks W=6 0.288 ±0.150 W=0; 13.1 ±3.6 evidence for direct action of Se
W=2 (E); 17.0 ±4.1 into GSH-Px.
Urine levels of Hg :
W=0; 0.23 ±0.22 W=6; 10.4 ±2.8
W=1; 0.23 ±0.19
W=2; 0.16 ±0.17

467
468
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Lei et al. 2015. Cohort study N = 310 infertile women Infertile women – Blood MeHg: Blood Se Measured outcomes: Fertility and Authors found slightly lower Tier 2 No effect of Se on Hg found,
Infertile women from Comparing compared with 57 fertile Median ±SD In infertile women with Hg reproductive hormones correlated prolactin in the highest as also very small effect of
Taiwan enrolled infertile women (pregnant) women ±5.21 > 5.8; to mercury mercury group and higher Hg was found. The authors
August 2008 to with pregnant Mean ±SD Pregnant – 242 ±42.2 MeHg in infertile group discuss a possible link
March 2010 women including Age infertile 35.2 ±3.9 years 5.13 ±2.75 compared with pregnant with fish consumption, but
In infertile women with Hg women. this is not really found.
food frequency Age fertile 34.8 ±4.1 years < 5.8;
questionnaire In the infertile group women 240 ±36.9
were divided into two groups
based on Hg Blood level of
5.8 µg/L
Lemire et al. 2010. Cross-sectional. 448 agreed to participate, Blood Hg: Median (range)- Blood Se: Median (range)- The measured outcome was Authors found that cataracts Tier 1 As this study is performed
Adults from rural Have followed this but due to different problems (µg/L) (µg/L) cataract formation graded in 4 were most prevalent in persons in a population with some
area in Tapajos River population since only about 220 were sampled 44 (4.3-298) 222 (124-1500) levels of seriousness with high B-Hg and Low P–Se. of the world’s highest Hg
Basin, State of Para, mid-1990s. and further exclusions criteria Plasma Hg Median (range)- Plasma Se Median (range)- This was mostly so in lower exposures, it is interesting
Brazil reduced the final number (µg/L) (µg/L) range of P-Se. to observe the correlation
to 211. 6.4 (0.2-40) 133 (57-913) of blood Hg and cataracts.
Human samples May Authors are not sure if Se seems to have an effect,
to July 2006. 96 women and 115 men all this weak effect is caused
over 40 years divided into but mechanism is not
by the effect of Se as an clear. This may be caused
groups age 40 to 65 (171) and antioxidant or as an effect by
65 to 87 (40). by the effect of Se as an
counteracting the negative antioxidant.
Mean age 50 years (40–64) effect of Hg on cataract
in first group and 73 years formation.
(65–87) in second group.
Lemire et al. 2011. Cross- sectional. 448 agreed to participate, and Blood Hg Mean/Median Blood Mean/Median (range): The outcome measurements Authors’ conclusion is mainly Tier 1 Authors indicate that there
Adults from rural Have followed this a total 407 were included in (range): 288/ 228 were different motor functions that there is a positive is some correlation between
area in Tapajos River population since tests and analyses 51.0 / 42.5 (103-1500) (dexterity) correlation between High high P-Se and motor
Basin, State of Para, mid-1990s. Several more were excluded (1.7-289) µg/L Tests: P-Se and motor outcomes, function outcomes.
Brazil because of problems with µg/L Plasma Se Mean/Med. (range): BAMT more when including Hg as a Extremely high Hg
creatinine analyses. As such, 163 / 135 Santa Ana counteracting confounder. concentration in this
Human samples May
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

to July 2006. for urine samples only 319 (54-913) Dynamometer This population has extremely population.
participated. Hair Se Grooved Pegboard high Hg exposure.
Mean age was 41.5 years, Mean/Med.(range):
(range 15–87) 0.84 / 0.69
50/50 female/male (n = 204/ (0.38-3.81)
n= 203) Urine Se (µg/g cr)
57.0 /33.6 (2.3-1375)
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Maeda et al. 2019 Case-control study The case included 98 female Blood Hg in case group Blood Se in case group µg/L The outcome measured in Authors indicate that low Se Tier 2 The difference in Se blood
Japan 2016-2017 Case is females patients aged 35.2 ±4.1 (mean µg/L (mean ± SD) ; this study is infertility and levels in blood were lower in levels is not very strong and
attended at Akita ±SD) (median, (5th-95th percentiles) ; 189± 25 reproductive hormones the infertile group. They also not existing for mercury,
hospital because The control group were 5.3 (2.2-14.5) In control group: claim that mercury is higher in based on unadjusted
lack of pregnancy. recruited from 169 200 ± 25 the infertile group adjusted to figures.
In control group: age and Se and indicate a toxic
Control is other participants, 49 consented and 5.0 (1.8-10.0) (Significant difference) The authors indicate a toxic
women from same in further 6 were excluded. effect of these rather low levels effect of these rather low
their 30s (Not significant difference) of Hg and that Se can mitigate levels of Hg and that Se
Control then included 43 the effect
women with mean age 33.6 ± can mitigate the effect.
2.5 years, meaning that the
control group was significantly
younger than the case group
Mao et al. 2019. Cross- sectional N = 5 115 were recruited in Toenail Hg (µg/g - ppm) Toenail Se (µg/g - ppm) Three cognitive tests: Authors find that higher Tier 1 Unfortunately, the exposure
Analysing data from Followed since 1985 1985 to 1986, Then at 18–30 0.32 ± 0.37 0.86 ± 0.15 RAVLT (words) omega-3 fatty acid intake and (Rated data in terms of toenail
the CARDIA study. and still ongoing. years. I quintiles based on omega-3 I quintiles based on omega-3 DSST (symbols) non-fried fish improves outcome too Hg and toenail Se at Y
Sampling year 2, 5, 7, 10, 15, intake intake Strop test of the DSST test. positi- 2 might not be relevant,
the United States of 0.22-0.26-0.30-0.37-0.43 vely?) since mercury follows the
America 20, 25, 30 and ongoing. 0.88-0.87-0.86-0.85-0.84 All after 25 Y Neither Hg nor Se interfered
Measured Y2 with the main conclusion, but omega-3 data there is no
Sampling from 1985 43.6 Measured Y2 really negative effect to
these statistics are based on
and still ongoing. very old Se and Hg data. improve.

Monastero et al. Cohort study 996 were interested, 746 Blood Hg measured by ICP-MS Blood Se (ng/mL) Cytokines and ANA titres used as Authors did not find any effect Tier 1 This experiment, which is
2017 Food frequency were eligible (based on food and the cohort divided into Mean (SD) measurements of immune system of high Hg on the cytokines (Rated the same as reported by
Long Island, the questionnaire and frequency survey) and 290 lower than 4.58 (µg/L) and 294.3 (101.5) function measured nor at ANA titres. too Karimi et al., showed no
United States of health question chose to enrol. 287 had blood higher than 4.58 µg/L. 295.3 (99.3) for the low Hg Se did not vary between high positi- difference in outcome with
America; avid and then one single samples drawn. Upper decile (19.8 – 51.0) group Hg and low Hg groups. vely?) mercury and no effects on
seafood consumers sampling. Mean (SD); or difference in blood Se
291.4 (106.9) for the high concentration.
2011–2012 48 years (18.2) Hg group
41.1% males

469
470
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Mozaffarian et al. Nested case-control HPFS enrolled 51 529 males Toenail Hg measured by Toenail Se measured by CVD First and fifth quintile Tier 2 No evidence of clinically
2011 (based on two in 1986 aged 40 to 75 years neutron activation analysis. neutron activation analysis. comparison relative risks: relevant adverse effects
(Cardiovascular cohorts) mean ±SD 61.1 ±9.0 Median, Males/Female Coronary heart disease of mercury exposure
disease) Conditional logistic NHS enrolled 120 700 females Interdecile range Mean ±SD 0.85 (CI 0.69-1.04; P=0.1 on coronary heart
regression in 1976 aged 30 to 55 years Case: 0.23 µg/g, 0.06-0.94 Case: 0.92 ±0.61 µg/g/ 0.78 for trend); stroke 0.84 (CI disease, stroke or total
Health Professional mean ±SD 53.8 ±6.1 cardiovascular disease in
Follow Up Study HPFS was only men, Control: 0.25 µg/g, 0.07-0.97 ±0.22 µg/g 0.62-1.14; P=0.27 for trend);
toenail clippings Of these cardiovascular Control: 0.92 ±0.6 µg/g, /0.78 total cardiovascular disease US adults at the exposure
(HPFS) Males/Female levels investigated. This
from 68% of disease (coronary heart Median, ±0.25 µg/g 0.85 (CI 0.72-1.01; P 0.006
(Male US health participants in 1987, disease and stroke) was for trend). Similar in analysis was also true when a
professionals) and Interdecile range subgroup with low Se
and 52% of NHS in prospectively identified in 3 Case: 0.30 µg/g, 0.07-1.26/ of participants with low Se
Nurses Health Study 1982-83 427 participants and matched concentration or low overall concentrations (<0.7 µg/g)
(NHS) 0.21 µg/g 0.06-0.77 was analysed or in analysis
to risk-set-sampled controls; Control: 0.31 µg/g, 0.07- fish consumption.
1986 and 1976 dentists were excluded. stratified according to fish
1.31/0.23 µg/g 0.07-0.76 consumption.
As the authors did not find
any negative effects of
the Hg exposure through
fish, they also could not
conclude on beneficial
effects of seafood intake
or Se.
Mozaffarian et al. Nested case-control HPFS enrolled 51 529 males in Toenail Hg µg/g measured by Toenail Se measured by Outcome was physician- Authors did not find clinically Tier 2 As the authors did not find
2012. (Hypertension (based on two 1986 aged 40 to 75 years neutron activation analysis. neutron activation analysis. diagnosed hypertension, apparent adverse effects any negative effects of the
Health Professional cohorts) NHS enrolled 120 700 females Mean (5th, 95th percentile) Mean ±SD biannually. of chronic methylmercury mercury exposure through
Follow Up Study HPFS was only men, in 1976 aged 30 to 55 years 0.92 ±0.63 in men and 0.79 The validity was confirmed in exposure at usual exposure fish, they also could not
(HPFS) 0.30 (0.07, 1.31) in men ±0.20 in women levels seen in these men and conclude on beneficial
toenail clippings 6 045 were included for 0.21 (0.07, 0.76) in women validation studies based on
US Men from 68% of analyses after exclusion of review of medical charts and women on hypertension. effects of omega-3-fatty
Nurses Health Study participants in 1987, 3 263 with hypertension at direct BP measurements. acids or Se.
(NHS) and 52% of NHS in baseline.
US Women 1982-83
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Nakamura et al. Multivariate analysis 194 Taiji (traditional whaling Measured in 23 subjects Measured in 23 subjects Neurological and dietary Hair mercury levels Tier 2 No significant correlations
2014. and another non- town) residents (117 males, 77 Oxygen combustion-gold ICP-MS surveys. Audiometry, MRI, significantly correlated with between hair mercury levels
Taiji, coastal Japan linear model females, 20–85 years) amalgamation method electromyography: cranial nerve whale meat intake. and neurological outcomes
Only correlation found, not affection, muscular weakness, in multivariate analysis.
2010–2011 60.3% Hair: Male/female/ values. In another nonlinear model,
tremor, rigidity, coordinated significant increase was Positive correlation between
total Geometric mean movements, one foot standing, whole blood Hg and Se,
17.2/12.1/14.9 µg/g observed in the odds ratio for
gait, touch sensation, pain sensorineural hearing loss which might be one cause
Median 18.7, 15.1/17.8 min sensation, position sense, of absence of adverse
max 1.1-101.9, 2.1-73.1/1.1- and simple gait disturbance
vibratory sensation, two-point with mercury levels >50 µg/g, effects of MeHg in this
101.9; 25/75 percentile 11.1, discrimination, graphesthesia, study.
32.7/5.9-24.3/7.9, 28.7 of but were assigned to other
stereognosis neurological causes than Hg
which 10/2/12 above 50 µg/g
(NOAEL by WHO) exposure at closer examination.
Blood
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Nyland et al. 2011. Cross-sectional 232 of 448 in parent study. Hg ICP-MS for fluid, cold Se ICP-MS To test the hypothesis that MeHg Elevated titers of ANA were Tier 2 No association of Se with
Part of study epidemiological 15–87 years vapor atomic absorption exposures affect levels of serum positively associated with Hg any changes in ANA and
described in Lemire study 112 males, 120 females spectrometry for hair biomarkers and to examine exposure for blood and plasma, did not modify associations
et al. 2006. Median (range) interactions between Hg and Se unadjusted and adjusted for between Hg and ANA titers.
48.28% in terms of these responses. sex and age. Proinflammatory
Adults living along Geometric mean Positive correlation between
Antinuclear (ANA) and ([interleukin (IL)-6 and blood Hg and Se.
the Tapajós River Hair interferon (IFN)-γ], anti-
(affected by MeHg 14.1 µg/g (1.1-62.4) antinucleolar (ANoA)
autoantibody levels and eight inflammatory (IL-4), and IL-17
– gold extraction), Blood cytokine levels were increased
Brazil cytokines in serum samples
53.5 µg/L (4.3-288.9) with MeHg exposure. Although
2006 42.5 µg/L Se status was associated
Plasma with MeHg level (correlation
8.8 µg/L (0.2-40) coefficient = 0.86; 95% CI:
0.29, 1.43), Se status was not
Urine associated with any changes
3.0 µg/L (0.2-16.1) in ANA and did not modify
associations between Hg and
ANA titers. Food frequency
questionnaire.
Oken et al. 2016. Prospective 1 068 pairs (872 with blood) Erythrocyte Hg Erythrocyte Se Child neurodevelopment cognitive Child verbal and nonverbal Tier 1 In this population with an
Project Viva cohort longitudinal study 7.7 years for cognitive tests. Mean ±SD Mean ±SD tests: Kauffman Brief Intelligence scores and tests of memory (Half average fish consumption
Multivariable linear Maternal 32.2 years ±5.3 Min max Min max Test (KBIT) and visual motor abilities were the of 1.5 servings per week,
Women attending not related to any exposures. partic- authors did not find
prenatal care at regression adjusted (mean ±SD), 15.3–44.9 4.0 ng/g ±3.6 205.6 ng/ml ±34.6 Food questionnaires at median
for characteristics 0-38.2 44.3-380.3 27.9 weeks gestation Mutual adjustments of each ipan- evidence for an association
Atrius Harvard Children: 50% of the exposure measures ts of maternal prenatal fish
Vanguard Medical including home
environment did not substantially change with- intake or Hg or Se status
Associates, estimates. drew) with verbal- or non-verbal
Massachusetts, the and maternal
intelligence intelligence, visual motor
United States of function, or visual memory
America Mid-childhood at median 7.7 years of age.
1999-2002 (median 7.7 years)
follow up
Park, K. & Seo, E. Prospective cohort 232 male +269 female = 501 Toenail mercury Toenail selenium Metabolic syndrome (MetS) Positive association between Tier 1 Selenium modifies
2016. Multivariable linear >35 years, 44.8 ±0.24 (mean Neutron activation analysis Neutron activation analysis (combination of diabetes, high toenail mercury and metabolic association between toenail
The Trace Element regression analysis ± se) Mean Median blood pressure [hypertension] syndrome, which was stronger mercury and metabolic
Study of Korean 0.4 µg/g 0.985 µg/g and obesity) at lower selenium and weaker syndrome. Association
46.31% at higher selenium. Half the non-significant for above-
Adults in the Mean
Yeungnam area participants were overweight. median levels of toenail
0.69 µg/g selenium.
(SELEN)
Korean area
Yeungnam area
2012–2013

471
472
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Park, K. & Seo, E. Prospective cohort 232 male +269 female = 501 Toenail mercury Toenail selenium Hypercholesterolemia, Participants in the highest Tier 1 High toenail mercury
2017. Multivariable linear >35 years, 44.8 ±0.24 (mean Neutron activation analysis Neutron activation analysis hyper-LDL-cholesterolemia, tertile of toenail mercury levels associated with higher
The Trace Element regression analysis ± se) Mean Median hypo-HDL-cholesterolemia, had 4.08 (95% CI 1.09–15.32, risk of hyper-LDL-
Study of Korean 0.4 µg/g 0.985 µg/g hypertriglyceridemia, and p for trend = 0.02) times cholesterolemia, and
46.31% dyslipidemia higher risk of hyper-LDL- dyslipidemia. These
Adults in the Men: 0.47 µg/g Mean
Yeungnam area cholesterolemia, and 2.24 associations were non-
Women: 0.34 µg/g 0.69 µg/g (95% CI 1.15–4.37, p for trend significant, at toenail
(SELEN)
= 0.004) times higher risk selenium levels >0.685
Korean area of dyslipidaemia than those µg/g.
Yeungnam area in the lowest tertile. These
2012-2013 associations became weak and
non-significant, showing OR
0.98 and 95% CI 0.25-3.80 for
hypercholesterolemia and OR
1.99 and 95% CI 0.73–5.45
for dyslipidemia at toenail
selenium levels >0.685 µg/g.
Park et al. 2013. Cross-sectional 2 117 for selenium data ICP MS Serum The weighted prevalence of Tier 1 Association between blood
National Health and associations >40 years for selenium data geometric means (95% ICP-DRC-MS hypertension was 32.2%. Hg (both MeHg and inorganic
Nutrition Examination confidence intervals) 137.2 ±1.39 µg/L The geometric means (95% Hg) and urinary Hg (mainly
(weighted mean ± se) confidence intervals) of blood inorganic) was not modified
Survey (NHANES) Blood Hg
46.6 ±0.5 ICP MS total and urinary mercury were by Se. No association of
the United States of 1.03 (0.95, 1.11) mg/L and 0.51 hypertension with blood Hg,
America 48.6 1.03 (0.95, 1.11) mg/L
(0.47, 0.54) mg/L, respectively. but a suggestive inverse
2003-2006 Urine Hg The adjusted odds ratios for a association with urinary Hg.
Cold vapor atomic absorption or doubling increase in blood Hg
ICP-DRC-MS and urinary Hg were 0.94 (0.87
0.51 (0.47, 0.54) mg/L to 1.01) and 0.87 (0.78 to 0.99),
respectively, after adjusting
for potential confounders. The
associations remained similar,
even after adjusting for either
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

omega-3 fatty acids or Se or both.

Rocha et al. 2016. Cohort study Because of the genetic The food survey showed Plasma Se Outcome measured was activity No effect was found. Tier 1 This study had very low
Young women from Young women, dimension, they enrolled 200 surprisingly low fish intake. µg/L 49.76 ±19.34 (5.85- of Se containing enzymes (GSH- Further, there was no difference exposure of mercury and
the relatively large premenopausal participants but 51 were Hair Hg 109.37) Mean ±SD (range) Px) and biomarkers of oxidative in outcomes between the no effect of mercury or
city of Porto Velho excluded due to incomplete µg/g (Mean ±SD) which shows Se deficiency. stress (ORAC & MDA) different genetic isoforms of modification of that effect
Mean sampling, and 149 women of Se was found.
Western Amazonas, 26.6 years (range 0.60 ±0.74, which is rather low Estimated mean (SD) selenium Differentiated due to genetic antioxidant defence.
Brazil were in the study. for this area. intake was 48.84 ±20.37 profiling.
18-48)
The aim was to differentiate µg/day
the response to mercury
and selenium exposure due
to genotyping selenium
containing enzymes known to
affect mercury and selenium
metabolism.
 TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type
APPENDICES

study (n) Risk

Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Saint-Amour et al. Cohort study of There were 483 newborns in Cord blood Hg (=µg/L Cord blood Visual test named. Visual evoked Authors conclude that there Tier 1 The study design focused
2006. children from the original Nunavik Cord calculated from nmol/L) 16.5 Se (=µg/L calculated from potentials (VEPs). Designated was a negative impact on on toxicological effect,
Nunavik Arctic Blood Monitoring Program. (13.4-20.3) Mean (95% CI) nmol/L) 319 (278-367) Mean N75 and P100, N150. these outcomes caused by and not very specific on Se
Canada Of these, 110 children were Arithmetic mean 23.9; SD 20.3, (95% CI) Arithmetic mean 351; the PCB and/or MeHg. They effect on MeHg toxicity.
Mothers recruited in recruited 5-6 years after birth. range 1.8-104. SD 164; range 164-774. did not find any significant
1993 to 1996 and (mean 5.4 years). Child blood Hg (=µg/L Child blood Se (=µg/L interaction with nutrients. They
children recruited 38.5% calculated from nmol/L) 5.9 calculated from nmol/L) 331 therefore conclude that they
5 to 6 years after (4.5-7.7) Mean (95% CI) (288-382) Mean (95% CI) contradict notions that Se can
delivery. Arithmetic mean 9.9; SD 9.1, Arithmetic mean 429; SD 427, protect against environmental
range 0.2-38.2. range 158-2566. neurotoxins.

Steuerwald et al. Cohort study of The cohort was established of Maternal whale meat, whale Serum µg/L Outcomes were neurological The authors found a significant Tier 1 Very high levels of Hg
2000. pregnant women 182 singleton term births born blubber and seafood intake was Mean ±SD examination of newborn infants at decrease in the neonatal NOS in all mercury-exposure
The women were from Faroe Islands in Torshavn. recorded. 103.4 ±14.2 two weeks of age. connected to methylmercury parameters.
residents of the Neurologic 51.1% Cord blood Hg µg/L 20.4 (11.8- Collected in a Neurological from seafood (but not from PCB Very important part of Faroe
northwestern region examination of 40.0; 1.90-102) Optimality Score (NOS). which was also high). Island study but not really
and not in the capital newborn (only) within Cord serum Hg µg/L 2.54 (1.65- They also conclude that there designed/protocolled to test
area. 14 days of delivery. 3.66; 0.70-8.74) was no evidence that Se caused Se effects.
Recruitment in important protection against Hg
Hair Hg µg/g 4.08 (2.45-7.35; associated decrease in NOS.
1994–1995 0.36-16.3)
geometric mean (interquartile
range; total range)
Tatsuta et al. 2017. Prospective cohort of There were 879 pregnant Maternal seafood intake was Cord plasma selenium Outcomes measured using Bayley Main finding was that prenatal Tier 1 Good correlation between
pregnant women in women who gave written estimated to provide Hg at 0.9 Ng/g was 66.3 ±10.2 and 67.0 Scales of Infant Development Hg negatively affected cord blood Hg and hair
Japan, giving birth informed consent, 749 mother- (0.1-3.8) and 0.9 (0.2-3.0) ±9.6 in mothers of boys and (BSID-II) and Kyoto Scale of psychomotor performance as Hg. Small but significant
in 2003-2006 and child pairs registered and µg/kg body weight/week girls respectively. Psychological Development measured in I BSID-II. negative effect of Hg on
children tested in completed data was collected respectively in mothers of boys (KSPD). Measured at 18 months the psychomotor part of
2004-2008. for 566. The children were 285 and girls. of age. BSID-II test. No effect of Se
boys and 281 girls. Hair mercury 2.5 µg/g (range corrected as confounder.
0.3-11.0).
Cord blood Hg 16.5 (5.7-36.9)
- 15.0 (4.8-39.3) for mothers
of boys and girls respectively
(median (5-95th percentile)

473
474
TABLE A6.4 OVERVIEW OF HUMAN PRIMARY STUDIES (N = 45) FOR THE REVIEW “HEALTH EFFECTS OF MeHg” (cont.)
Author, year Number of participants in the
Study type study (n) Risk
Trial or study name Study duration and Measurement and levels of Measurement and levels of
Age (years) at exposure Measurement of outcome Overall results of bias Overall conclusion
Geography follow-up time exposure (Hg) exposure (Se)
assessment (OHAT)
Year of sampling Statistical approach Sex (% males)
Tratnik et al. 2017. Two comparative 601 pregnant women were Consumption of sea fish was Selenium in cord blood serum Outcome measured as child Hg-related decrease in Tier 1 Weak effect of Se for a
Women giving birth cohorts from recruited in Slovenia and 243 different in the cohorts In was 40.1 ng/g (range 15-70 neurodevelopmental outcomes cognitive score only in children genetic subgroup.
in Slovenia and Ljubljana, Slovenia from Croatia. But only 361 of Slovenia 25% ate fish more ng/g) using Bayley-III (cognitive, carrying at least one Apoeε4
Croatia (large city) and the children had Hg in cord than once per week, in Croatia language and motor scales) allele, and general decrease
Southern Europe Rijeka, Croatia, blood, Bayley-III assessment 68.5%. in fine motor scores. Positive
(coastal city) and Apoc genotyping, leaving Mean cord blood was 1.58 ng/g association between Se and
Prospective cohort in total 361 mother/child pairs in Slovenia and 3.40 ng/g in the language score, but not
in the study, 237 from Slovenia Croatia. Maternal hair mean in the subgroup of children
Project begun in and 124 from Croatia. carrying the Apoε4 allele.
2006, completed in Hg was 273 ng/g in Slovenia
Children were also stratified and 576 ng/g in Croatia Unclear if effect of Se on Hg
2011. toxicity or just separate effects.
genetically through
Apolipoprotein E genotyping.
Wells et al. 2016. Cross-sectional There were 597 births, of which 341 Me-Hg in umbilical cord blood Umbilical cord serum Outcomes measured were different Ponderal index decreased with Tier 2 The negative association of
Women giving birth in included collection of umbilical cord Inorganic Hg in blood. Se geometric mean 70 µg/L (95% birth parameters such as birth weight, increasing MeHg. This was not MeHg with birth weight and
Baltimore, the United blood. 300 of these had enough for CI 68.5; 71.4) birth length, gestational age, head influenced by Se. ponderal index was influenced
initial lab analysis and is called the Hg method had rather high LOD (or circumference and Ponderal Index. by Se.
States of America, Nov. more correct LOQ) which does have Hg and Se did not correlate.
2004 to March 2005 Baltimore THREE study. 29 more
were excluded due to lack of total a negative impact on the data set.
data sets. Geometric mean cord MeHg was
55.4% 0.94 µg/L with 95% CI 0.8 (1.07)

Wennberg et al. Case-control study The total base of screened Meals of fish/week Meals of fish /week; 1.00 in Outcome measured was Authors report no effect of Hg Tier 1 Since there were no
2011 based on three subjects was about 73 000 by Erythrocyte Hg was 3.47 cases and 1.00 in controls for myocardial infarction, of which on myocardial infarction. negative effects of raised
Northern Sweden studies. 31 Dec. 1999 (0.01-45.9) in males in cases men, 1.05 and 1.05 for women some resulted in sudden cardiac Hg concentrations caused
Health and Disease Number of cases with and 3.95 (0.15-81.4) in male Erythrocyte mean (range) death (SCD) by fish intake, it was not
Study (NSHDS) with myocardial infarction (MI) controls and for women 3.04 Se was 120 (72.5-302) µg/L These are compared with a possible to detect protective
three underlying was 7 337, of which Hg/Se/ (0.43-23.4) in cases and 3.68 in males in cases and 123 reference group. effects of Se. There was a
studies (VIP, Fatty acid data was available (0.19-16.7) in controls. (75.5-250) in male controls small increase in risk of
MONICA, MSP) for 431/431/374. These and for women 126 (86-207) SCD with increasing Ery-Se,
were matched with controls in cases and 132 (81.0 – 400) but the authors indicate it
1987–1999 could be a random effect.
(499/497/434). in controls.
FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

Mean age at event 58.7 years

(34.1 – 77.1), average time
between baseline and event 7
years, 11 months.
APPENDICES

APPENDIX 7
OCCURRENCE DATA OF MeHg
AND DIOXINS AND dl-PCBs

LITERATURE SEARCH STRATEGY

TABLE A7.1 LITERATURE SEARCH STRATEGY FOR THE REVIEW “OCCURRENCE DATA” FOR DATA ON HG AND MeHg
Database: Web of Science
Date of literature search: 22 November 2021
# Literature search string Total hits
1 TI=(Hg OR mercury OR MeHg OR Me-Hg OR “methyl Hg” OR “methyl-Hg” OR "methylmercury" OR "methyl-mercury") 22 384
2 AB=(Hg OR mercury OR MeHg OR Me-Hg OR “methyl Hg” OR “methyl-Hg” OR "methylmercury" OR "methyl-mercury") 75 521
3 AK=(Hg OR mercury OR MeHg OR Me-Hg OR “methyl Hg” OR “methyl-Hg” OR "methylmercury" OR "methyl-mercury") 16 453
4 #1 OR #2 OR #3 79 534
5 TI=(Fish* OR finfish* OR crayfish* OR crawfish* or cuttlefish* OR inkfish* or milkfish* or catfish* or Pisces or seafood* or “sea food*” or sea-food* 329 805
or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin*
or mackerel* or pilchard* or crab* or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod or
“Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad* or Crassostrea or tagelus or clam* or cockle*
or urchin* OR echinoderm* OR echinoid* or “sea cucumber*” OR holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark*
or skate* or crustacea* or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* OR whelk* OR gastropod* or abalone* or snail* or
limpet* or conch* or periwinkle* or nautilus*)
6 AB=(Fish* OR finfish* OR crayfish* OR crawfish* or cuttlefish* OR inkfish* or milkfish* or catfish* or Pisces or seafood* or “sea food*” or sea-food* 725 705
or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin*
or mackerel* or pilchard* or crab* or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod or
“Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad* or Crassostrea or tagelus or clam* or cockle*
or urchin* OR echinoderm* OR echinoid* or “sea cucumber*” OR holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark*
or skate* or crustacea* or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* OR whelk* OR gastropod* or abalone* or snail* or
limpet* or conch* or periwinkle* or nautilus*)
7 AK=(Fish* OR finfish* OR crayfish* OR crawfish* or cuttlefish* OR inkfish* or milkfish* or catfish* or Pisces or seafood* or “sea food*” or sea-food* 208 439
or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin*
or mackerel* or pilchard* or crab* or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod or
“Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad* or Crassostrea or tagelus or clam* or cockle*
or urchin* OR echinoderm* OR echinoid* or “sea cucumber*” OR holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark*
or skate* or crustacea* or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* OR whelk* OR gastropod* or abalone* or snail* or
limpet* or conch* or periwinkle* or nautilus*)
8 #5 OR #6 OR #7 803 288
9 TI=(concentration* OR level* OR measure* OR amount* OR value* OR content* OR determin*) 1 190 004
10 AB=(concentration* OR level* OR measure* OR amount* OR value* OR content* OR determin*) 10 070 218
11 AK=(concentration* OR level* OR measure* OR amount* OR value* OR content* OR determin*) 478 607
12 #9 OR #10 OR #11 10 410 632
13 #4 AND #8 AND #12 6 202
14 #4 AND #8 AND #12 and Review Articles or Proceedings Papers or Meeting Abstracts or Corrections or Editorial Materials or Letters or Retracted 5 884
Publications or News Items (Exclude - Document Types)

475
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

TABLE A7.2 LITERATURE SEARCH STRATEGY FOR THE REVIEW “OCCURRENCE DATA” FOR DATA ON DIOXINS AND dl-PCBs
Database: Web of Science
Date of literature search: 22 November 2021
# Literature search string Total hits
1 TI=(dioxin* OR furan* OR PCDD* OR PCDF* OR “polychlorinated dibenzodioxin*” OR “polychlorinated dibenzofuran*” OR TEQ) 9 089
2 AB=(dioxin* OR furan* OR PCDD* OR PCDF* OR “polychlorinated dibenzodioxin*” OR “polychlorinated dibenzofuran*” OR TEQ) 22 298
3 AK=(dioxin* OR furan* OR PCDD* OR PCDF* OR “polychlorinated dibenzodioxin*” OR “polychlorinated dibenzofuran*” OR TEQ) 6 734
4 #1 OR #2 OR #3 24 262
5 TI=(Fish* OR finfish* OR crayfish* OR crawfish* or cuttlefish* OR inkfish* or milkfish* or catfish* or Pisces or seafood* or “sea food*” or sea-food* 329 805
or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin*
or mackerel* or pilchard* or crab* or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod or
“Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad* or Crassostrea or tagelus or clam* or cockle*
or urchin* OR echinoderm* OR echinoid* or “sea cucumber*” OR holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark*
or skate* or crustacea* or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* OR whelk* OR gastropod* or abalone* or snail* or
limpet* or conch* or periwinkle* or nautilus*)
6 AB=(Fish* OR finfish* OR crayfish* OR crawfish* or cuttlefish* OR inkfish* or milkfish* or catfish* or Pisces or seafood* or “sea food*” or sea-food* 725 705
or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin*
or mackerel* or pilchard* or crab* or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod or
“Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad* or Crassostrea or tagelus or clam* or cockle*
or urchin* OR echinoderm* OR echinoid* or “sea cucumber*” OR holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark*
or skate* or crustacea* or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* OR whelk* OR gastropod* or abalone* or snail* or
limpet* or conch* or periwinkle* or nautilus*)
7 AK=(Fish* OR finfish* OR crayfish* OR crawfish* or cuttlefish* OR inkfish* or milkfish* or catfish* or Pisces or seafood* or “sea food*” or sea-food* 208 439
or "blue food*" or blue-food* or "aquatic food*" or aquatic-food* or anchov* or Engraulis or pollock* or tuna* or herring* or whiting* or sardin*
or mackerel* or pilchard* or crab* or shrimp* or prawn* or squid* or octopus or cephalopod* or mollusc* or mussel* or shell* or scallop* or cod or
“Largehead hairtail” or carp* or tilapia* or salmon* or pangasius or krill or decapod* or oyster* or scad* or Crassostrea or tagelus or clam* or cockle*
or urchin* OR echinoderm* OR echinoid* or “sea cucumber*” OR holothuroid* or catla or Carassius or labeo or trout* or bream* or Cyprinid* or shark*
or skate* or crustacea* or “aquatic invertebrate*” or “marine invertebrate*” or lobster* or bivalv* OR whelk* OR gastropod* or abalone* or snail* or
limpet* or conch* or periwinkle* or nautilus*)
8 #5 OR #6 OR #7 803 288
9 TI=(concentration* OR level* OR measurement* OR amount* OR value* OR content* OR determinat*) 1 190 004
10 AB=(concentration* OR level* OR measurement* OR amount* OR value* OR content* OR determinat*) 10 070 218
11 AK=(concentration* OR level* OR measurement* OR amount* OR value* OR content* OR determinat*) 478 607
12 #9 OR #10 OR #11 10 410 632
13 #4 AND #8 AND #12 1 031
14 #4 AND #8 AND #12 and Review Articles or Proceedings Papers or Editorial Materials or Letters or Meeting Abstracts or Retracted Publications (Exclude 967
– Document Types)

476
APPENDICES

QUALITY ASSESSMENT OF ARTICLES FOR OCCURRENCE DATA

Insert Table A7.3 Quality assessment of articles for Occurrence data

OCCURRENCE DATA
Insert Table A7.4 Occurrence data from the literature
Insert Table A7.5 Occurrence data from EFSA
Insert Table A7.6 Occurrence data of Hg and MeHg from WHO-GEMS

477
 FAO/WHO BACKGROUND DOCUMENT ON THE RISKS AND BENEFITS OF FISH CONSUMPTION

FOOD SAFETY AND

QUALITY SERIES
11. FAO. 2016. Risk based imported food control, manual.
2. FAO and WHO. 2016. Risk communication applied to food safety, handbook.
3. FAO. 2016. Enhancing early warning capabilities and capacities for food safety.
4. FAO. 2017. Food safety risk management: evidence-informed policies and
decisions, considering multiple factors.
5. FAO and WHO. 2018. Technical guidance for the development of the growing
area aspects of bivalve mollusc sanitation programmes.
5a. FAO and WHO. 2021. Technical guidance for the development of the growing
area aspects of bivalve mollusc sanitation programmes, second edition.
6. FAO. 2019. Technical guidance principles of risk-based meat inspection and
their application.
7. FAO and WHO. 2019. Food control system assessment tool.
> Introduction and glossary
> Dimension A – inputs and resources
> Dimension B - control functions
> Dimension C - interaction with stakeholders
> Dimension D - science/knowledge base and continuous improvement
8. FAO. 2020. Climate change: unpacking the burden on food safety.
9. FAO and WHO. 2020. Report of the expert meeting on ciguatera poisoning.
10. FAO. 2020. FAO guide to ranking food safety risks at the national level.
11. FAO and WHO. 2020. Joint FAO/WHO expert meeting on tropane alkaloids.
12. FAO. 2022. Technical guidance for the implementation of e-notification systems
for food control.
13. FAO and WHO. 2022. Report of the expert meeting on food safety for seaweed
- Current status and future perspectives.
14. FAO and WHO. 2022. Risk assessment of food allergens. Part 1: Review and
validation of Codex Alimentarius priority allergen list through risk assessment.
15. FAO and WHO. 2022. Risk assessment of food allergens. Part 2: Review and
establish threshold levels in foods for the priority allergens.

478
16. FAO and WHO. 2023. Risk assessment of food allergens – Part 3: Review and
establish precautionary labelling in foods of the priority allergens.
17. FAO and WHO. 2024. Risk assessment of food allergens – Part 4: Establishing
exemptions from mandatory declaration for priority food allergens.
18. FAO. 2022. Microplastics in food commodities – A food safety review on human
exposure through dietary sources.
19. FAO. 2023. The impact of pesticide residues on the gut microbiome and human
health – A food safety perspective.
20. FAO. 2023. The impact of veterinary drug residues on the gut microbiome and
human health – A food safety perspective.
21. FAO. 2023. The impact of microplastics on the gut microbiome and health
– A food safety perspective.
22. FAO and WHO. The impact of food additives on the gut microbiome and
health - A food safety perspective. In progress.
23. FAO and WHO. 2023. Risk assessment of food allergens. Part 5: Review and
establish threshold levels for specific tree nuts (Brazil nut, macadamia nut or
Queensland nut, pine nut), soy, celery, lupin, mustard, buckwheat and oats.
24. FAO. 2023. Food safety implications from the use of environmental inhibitors
in agrifood systems.
25. FAO. 2024. Risk assessment of 3-monochloropropane-1,2-diol, glycidol,
and their fatty acid esters in lipid-based nutrient supplements and ready-to-use
therapeutic food.
26. FAO. 2024. Report of the FAO Technical Meeting on the gut microbiome in
food safety chemical risk assessment. In progress.
27.FAO & WHO. 2024. FAO/WHO background document on the risks and
benefits of fish consumption.
28. Joint FAO/WHO Expert Consultation on risks and benefits of fish consumption.
Meeting report. In progress.
29. FAO. 2024. Food Safety in a Circular Economy. In progress.

479
© FAO/Luis Tato © FAO/Giulio Napolitano
RISK ASSESSMENT OF 3-MONOCHLOROPROPANE-1,2-DIOL, GLYCIDOL,
AND THEIR FATTY ACID ESTERS IN LIPID-BASED NUTRIENT
SUPPLEMENTS AND READY-TO-USE THERAPEUTIC FOOD
FOOD SAFETY IN THE CONTEXT OF LIMITED FOOD AVAILABILITY

New literature became available following the last Joint FAO/WHO Expert Consultation on the Risks and Benefits of
Fish Consumption held in 2010. FAO and WHO decided to generate a background report consisting of a comprehensive
literature review, followed by an expert consultation, to update the report with new scientific evidence. This background
document aims to provide scientific evidence about the risks and benefits of fish consumption that were the basis for
the Expert Consultation and contains the results and methodology followed to carry out five extensive literature reviews
focused on: evidence of health benefits from fish consumption; toxic effects of dioxins and dioxin-like polychlorinated
biphenyls (dl-PCBs); toxic effects of methylmercury (MeHg); the role of selenium (Se) with regard to the health effects of
MeHg; and occurrence data for MeHg, dioxins and dl-PCBs in fishery and aquaculture products.

FOOD SYSTEMS AND FOOD SAFETY - ECONOMIC AND SOCIAL DEVELOPMENT

WEBSITE: WWW.FAO.ORG/FOOD-SAFETY
FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS
ROME, ITALY

ISBN 978-92-5-138944-7 ISSN 2415-1173

9 789251 389447
CD1548EN/1/07.24