University of Maryland School of Nursing

NRSG 790: Methods for Research and Evidence Based Practice

TOOL
RESEARCH APPRAISAL TOOL
Citation: Atia, H., Badawie, A., Elsaid, O., Kashef, M., Alhaddad, N., & Gomaa, M. (2022). The
hematological impact of umbilical cord milking versus delayed cord clamping in premature
neonates: a randomized controlled trial. BMC pregnancy and childbirth, 22(1), 714.
https://doi.org/10.1186/s12884-022-05046-7
Background
The article by Atia et al. (2022) discusses hematological outcomes of neonatal morbidities
among immature babies based on Umbilical Cord Milking (UCM) versus Delayed Cord Clamping
(DCC). This study was motivated by the need to assess other cord management methods that
would allow for accelerated placental transfusion, especially in preterm infants. Placental
transfusion offers extra blood and iron stores to newborns, which is very important in stabilizing
their postnatal hemodynamics and oxygenation.
The investigators aimed to establish whether UCM would be a quicker yet equally
effective process of placental transfusion enhancement as compared to DCC while investigating
whether UCM was associated with increased risks of neonatal morbidities such as intraventricular
hemorrhage, necrotizing enterocolitis, and other adverse outcomes. The current study also focused
on premature neonates born between 24 and 34+6 weeks of gestation category, in which
immediate resuscitation after birth is often critical.
Methodology
This study was an open-label randomized controlled trial (RCT). RCTs are considered the
best evidence on the effectiveness of an intervention, given that randomization and controllable
groups allow for much less bias (Atia et al., 2022). The targeted population of the study were pre-
term neonates delivered between 24- and 34+6-weeks gestation. Such infants were chosen because
their predisposition to postnatal complications makes them an ideal group in assessing benefits
from methods of placental transfusion.
The sampling strategy was based on simple randomization; an online randomizer tool-
generated was created. This was continued in a manner that would keep the balance in the
allocation of UCM versus DCC, with strict inclusion and exclusion criteria to control the
confounding variables. This ensures a balance in the distribution of participants, 1:1, allocating the
participant to either UCM or DCC (Atia et al., 2022). The analysis was limited to singleton
pregnancies and also excluded congenital anomalies, multifetal pregnancy, or significant fetal
distress to focus on those pregnancies where interventions of cord management would more than
likely have an impact.
The sample size was an estimation to achieve a power of 80%, providing the researchers
with an opportunity to detect a 3% difference in hematocrit values and a 1g alteration in the levels
of hemoglobin between the UCM and DCC groups. Therefore, the sample of 200 preterm
neonates was split into two equal intervention groups.
Sample, Inclusion/Exclusion Criteria, and Adequacy of Sample Size (Power Analysis)
Atia et al. (2022) carried out their research with 200 preterm neonates who were randomly
divided into two groups: Umbilical Cord Milking, UCM, (n=100), and Delayed Cord Clamping,
DCC, (n=100).
Inclusion/Exclusion Criteria
The inclusion criterion for this study was only singleton pregnancies between 24+0 and
34+6 weeks gestation admitted for preterm labor. The selection of only singleton pregnancies was
to minimize confounding variables as the pregnancy with multifetal gestation adds complexity to
the variable that will tend to have a biased result on the study. Pregnancy with multifetal gestation,
congenital anomalies, fetal anemia, significant antepartum hemorrhage, and category III
cardiotocography tracings were excluded from the study. These exclusions ensured that the study
focused on cases where Umbilical Cord Milking (UCM) or Delayed Cord Clamping (DCC) would
have a direct and measurable impact on neonatal outcomes without interference from severe
pregnancy-related conditions (Atia et al., 2022). This controlled for variability and allowed a more
precise comparison between the two interventions.
Adequacy of Sample Size (Power Analysis)
The power analysis was conducted to provide the study with 80% power at a significance
level of 0.05 (alpha). These calculations were based on observed differences of 3% in hematocrit
values and 1g in hemoglobin levels between the UCM and DCC groups. Thus, to attain such
results, a sample size of 196 neonates would be required, while in this study, the sample size is
200 participants and, therefore, adequate for ascertaining clinically and statistically significant
differences in the primary outcomes. The robust sample size used in the study presents confidence
in the detection of clinically meaningful differences in hematological outcomes between the two
interventions. Atia et al., 2022.
Conclusions from Study Participant Flowchart (Melnyk, pg. 130)
The flowchart of the study participants (Appendix A) further supports these findings by
demonstrating that all 200 eligible participants completed the intervention, with no significant
withdrawals or post-randomization exclusions. Hence, the statistical power of this study was
maintained, and the final analysis included a sufficient number of participants to detect clinically
meaningful differences between the two groups.
This flowchart reflects the effective retention of participants and thus supports internal
validity, as there is no significant loss of participants whereby results would have been biased. As
further shown from this flowchart, participants were equally distributed between the UCM and
DCC groups, ensuring that only the intervention and not external factors or bias can be confidently
considered to account for any difference in outcomes.
Table 1 Review
In Table 1, the maternal demographic data and delivery outcome reveal that the two
groups, UCM versus DCC, were generally similar at baseline, though there are some notable
differences. Homogeneity between the groups is evident in parity, the number of antenatal care
visits, and medical disorders such as diabetes mellitus and hypertensive disorders, which did not
attain any difference at a statistically significant level between the two groups. This serves as an
assurance that the randomization process is generally effective in generating comparable groups.
However, there are some key differences that introduce heterogeneity between the groups:
 Maternal age was higher, p = 0.045, and BMI was significantly higher, p = 0.001, in the
UCM group than in the DCC group, which may influence the results regarding delivery and
neonatal health.
 The UCM group had a relatively lower mean gestational age compared with the general
group, 31.4 weeks versus 32.1 weeks (p = 0.022). In the UCM group, more neonates were
extremely preterm than in the general group, 10% and 5%, respectively (p = 0.037). This could
introduce some confounding variables since prematurity is one of the strongest predictors of
neonatal morbidity and mortality.
 Birth weight was similarly significantly lower in the UCM group, 1573g versus 1781g,
p=0.004, further emphasizing that neonates in the UCM group were more premature and at greater
risk for complications than those of the DCC group.
 The mode of delivery significantly differed: 36% of cesarean deliveries before the onset of
labor in the UCM group, as compared with 13% (p < 0.0001), while in the DCC group, there were
more vaginal deliveries-56% as compared to 21%, p < 0.0001. This difference in the mode of
delivery might also influence neonatal outcomes.

Study Variables, Protocols, and Measurement

State the independent variable(s) and primary dependent variable. Include the operational
definition if provided.

Discuss the intervention and control protocols (procedures – what was done, how, when, where,
by whom). Where the groups treated equally, and how do you know? How was treatment fidelity
maintained and why is this important?

State the measure of the primary dependent variable (instrument used to measure the dependent
variable), the measurement protocol (the data collection procedures), and the level of
measurement for the primary dependent variable. Why is knowledge of the level of measurement
important?

Discuss the reliability and validity of the instrument used to measure the primary dependent
variable. What role does instrument reliability and validity play in determining statistical
conclusion validity?

Study Results and Findings

Data Analysis: Discuss the type of statistical procedure used to test the hypothesis and
appropriateness.

Statistical Significance: Provide the study results for the Primary Outcome (Table 2) and
conclusions based on the results. Are the results generalizable to the target population? Why or
why not?

Conclusions: Describe the conclusions reached and possible alternative explanations based on
Table 3.

Clinical Significance: Was there a treatment effect and state implications for clinical practice
 Interpretation
Question Response/Explanation

Validity
Are the study results valid?
(Melnyk pgs 132 – 136)

Reliability
What are the results?
(Melnyk pgs 136 – 141)

Applicability
Will the results help me in caring
for my patients?
(Melnyk pg 148)

Level and Quality rating and rationale (use JHNEBP Model):

 Appendix
Figure 1: Melnyk’s Chart (p. 130)
969 accessd
eligibility

89 ineligible

880 elligible

45 declines to
participate

835 randomized

416 assigned to 419 assigned to

test intervention control

3 did not receive 5 did not receive

intervention control Reasons:
reason

413 received test 414 received

intervention control

17 did not 9 did not

complete study complete study
Reasons: Reasons:

396 analyzed 405 analyzed