13

AntimicrObial suscaptibility tasting

Paula C. Mister and Donald C. Lehman

CHAPTER 0UTLINE
0BJECTIVES
PrOcedures in antimicrObial susceptibility testing, 272
After reading and studying this chapter, yOu shOuld be able tO:
ReasOns and indicatiOns fOr perfOrming antimicrObial susceptibility
tests, 273 1. Explain the ratiOnale behind the perfOrmance Of antimicrObial
SuSceptibility teStS.
FactOrS tO cOnSider when determining whether SuSceptibility teSting
iS warranted, 273 2. DeScribe the methOd fOr SelectiOn Of Specific drugS in teSting
and repOrting.
Selecting antimicrObial agents fOr testing and repOrting, 274
3. Define minimal inhibitOry cOncentratiOn (MIC).
SelectiOn Of teSt batterieS, 274
4. JuStify the methOdS uSed fOr determining MICS.
RepOrting Of SuSceptibility teSt reSultS, 274
5. Explain hOw zOne-interpretive criteria uSed with the diSk diffuSiOn teSt
TraditiOnal antimicrObial susceptibility test methOds, 275
are eStabliShed.
InOculum preparatiOn and uSe Of McFarland StandardS, 275
6. LiSt the variableS that muSt be cOntrOlled when antimicrObial
DilutiOn SuSceptibility teSt methOdS, 276 SuSceptibility teStS are perfOrmed.
DiSk diffuSiOn teSting, 279 7. JuStify teSt mOdificatiOnS fOr antimicrObial SuSceptibility teSting
InterpretatiOn Of in vitrO antimicrObial susceptibility test results, 282 Of HelicObacter/CampylObacter Spp., StreptOcOccus Spp.
MOdified teSt methOdS fOr SlOw-grOwing Or faStidiOuS bacteria, (including StreptOcOccus pneumOniae), HaemOphilus Spp.,
284 AdditiOnal OrganiSm and antimicrObial agent teSting cOncernS, Neisseria gOnOr- rhOeae, Neisseria meningitidis, and Obligate
287 β-LactamaSe teStS, 291 anaerObeS.
EteSt, 292 8. Explain the principleS behind autOmated antimicrObial SuSceptibility
teSt methOdS.
AutOmated antimicrObial susceptibility test methOds, 293
9. DiScuSS Several cOmmercially available antimicrObial SuSceptibility
PrincipleS Of technOlOgieS uSed, 293
teSt SyStemS in current uSe.
AutOmated SyStemS, 293
10. JuStify teSting bacterial iSOlateS fOr β-lactamaSe.
Quality cOntrOl Of antimicrObial susceptibility tests, 295
11. LiSt the OrganiSmS fOr which β-lactamaSe teSting iS indicated.
Selecting an antimicrObial susceptibility test methOd, 298
12. Explain the reliable methOdS fOr detectiOn Of methicillin-reSiStant
Rapid SuSceptibility determinatiOn, 299
StaphylOcOccus aureus, vancOmycin-intermediate S. aureus, and
Special antimicrObial susceptibility tests, 299 vancOmycin-reSiStant S. aureus
Minimum bactericidal cOncentratiOn test, 300 13. DiScuSS uSe Of the D-zOne teSt.
COntrOlling teSt variableS, 14. DeScribe the Significance Of high-level aminOglycOSide reSiStance
301 InterpretatiOn cOncernS, in enterOcOcci.
301 15. DeScribe extended-Spectrum β-lactamaSeS and carbapenemaSeS and
Time-kill assays, 301 aSSayS tO detect them.
Synergy tests, 302 16. JuStify teSting fOr extended-Spectrum β-lactamaSeS and
Serum bactericidal test, 302 carbapenemaSeS.
Measurement Of antimicrObial agents in serum and bOdy fluids, 302 17. DiScuSS quality cOntrOl prOcedureS fOr antimicrObial SuSceptibility teStS.
BiOlOgical aSSayS, 302 18. DeScribe hOw antibiOgramS can be uSed tO help verify the accuracy Of
ImmunOaSSayS, 303 reSultS generated by teSting patient iSOlateS.
ChrOmatOgraphic aSSayS, 303 19. DiScuSS SituatiOnS in which cumulative antibiOgramS may help guide
BibliOgraphy, 305 antimicrObial therapy.

271
272 PART 1 13 AntimicrObial SuSceptibility teSting

20. Explain hOw tO Select a particular SuSceptibility teSt methOd fOr

rOutine uSe. scientist who had reported these results might have
forgot- ten the current method for detecting
21. Explain the antimicrObial SuSceptibility teSt interpretive reSultS Of methicillin-resistant Gtap%ylococcuS aureuS (MRSA) and
nOnsusceptible, susceptible, intermediate, and resistant
reporting rules for β-lac- tam agents and MRSA. She
22. Explain the functiOn Of the InternatiOnal 0rganizatiOn reminded the laboratory scientist of the special testing
fOr StandardizatiOn. method and reporting rules, and the laboratory
23. DeScribe the minimum bactericidal cOncentratiOn (MBC) teSt and the scientist corrected and rereleased the report, which
indicatiOnS fOr perfOrming thiS teSt. was as follows: clindamycin-R, erythromycin-R,
24. JuStify determining the MBC Of bacterial iSOlateS. oxacillin-R, penicillin-R, and vancomycin-S.
25. Define synergism, antagOnism, and indifference aS related tO teSting
cOmbinatiOnS Of antimicrObial agentS. Issues to Consider
26. DeScribe the Serum bactericidal teSt and the indicatiOnS fOr perfOrm- After reading the patient´s case history, consider:
ing thiS teSt. • Use of surrogate antimicrobial agents to test for
27. Explain hOw mOlecular prObeS might be uSed tO detect resistance or susceptibility for the drug reported
antimicrObial reSiStance. • Editing certain susceptible results to resistant, and
28. DiScuSS methOdS uSed fOr meaSuring cOncentratiOnS Of antimicrObial not reporting related drugs, when bacteria with
agentS in Serum and bOdy fluidS and when Such teStS are uSed. specific types of resistance are encountered
• Modifications of standard tests that might be
necessary to detect specific types of emerging
resistance

KEY TERMS

AntagOnism
Minimal inhibitOry cOncentratiOn
AntibiOgram (MIC) PrOcedures in
β-Lactamases
BOrderline Oxacillin-resistant
Minimum bactericidal cOncentra-
tiOn (MBC)
antimicrObial susceptibility
isOlates NOnsusceptible testing
BreakpOint Oxacillin-salt screen plate
Carbapenemase-prOducing Penicillinase-prOducing Neisseria Antimicrobial susceptibility testing is performed on
EnterObacterales (CPE) gonorrhoeae (PPNG) bacte- ria and fungi isolated from clinical specimens
ChrOmOsOmally mediated resis- to determine which antimicrobial agents might be
Penicillinase-resistant
tant Neisseria gonorrhoeae effective in treating infections caused by these
penicillins organisms. Only organisms that are likely to be
(CMRNG)
Persister cells contributing to an infection should be tested. Testing
Clinical and LabOratOry
Resistant of organisms that are not involved in an infection is
Standards Institute (CLSI)
Selective repOrting misleading to the physician and could lead to
Cumulative antibiOgram unnecessary treatment and a more serious infection
Serum bactericidal test (SBT)
D-ZOne test with development of antimicrobial resistance.
Susceptible
Etest A major goal of the clinical microbiology laboratory
Synergism
Extended-spectrum β-lactamase is to identify organisms that are causing infections.
(ESBL) Time-kill assay Often, these organisms must be distinguished from
HeterOresistant TOlerance the normal microbiota that can reside at the site of
High-level aminOglycOside Trailing the infection, although in some situations the
U.S. FOOd and Drug microbiota that resides at the site of the infec- tion
resistance
AdministratiOn (FDA) may be contributing to the infection. Therefore it
Indifference must be determined which organisms from a
Intermediate VancOmycin agar screen
specimen are clinically relevant and will be identified
Kirby-Bauer test plate VancOmycin- and tested for susceptibility to antimicrobials. Most
McFarland turbidity standards intermediate microbiology laboratories have guide- lines for
Staphylococcus aureus (VISA) determining when and on which microorganism
necA gene
VancOmycin-resistant enterOcOcci susceptibility testing will be done. When in doubt
Methicillin-resistant (VRE) about the significance of an organism, it is best to
Staphylococcus aureus
VancOmycin-resistant discuss the situation with the attending physician
(MRSA)
Staphylococcus aureus (VRSA) and microbiology laboratory director.
ZOne Of inhibitiOn In clinical laboratories, susceptibility testing on
Case in pOint bacte- ria can be performed by classic disk diffusion
or dilution (minimal inhibitory concentration [MIC])
methods. More commonly, one of several different
types of commercial automated antimicrobial
susceptibility test systems may be used. Protocols
that describe these methods are published
The microbiology laboratory supervisor was reviewing following disk diffusion testing results: cefazolin-S, cefoxitin-
patient reports that the floating laboratory scientist R, clindamycin-R, erythromycin-R,
had generated earlier in the day. She saw a penicillin-R, and vancomycin-S. However, there was no result
susceptibility report for Staphylococcus aureus for oxacillin. The supervisor realized that the laboratory
isolated from a patient’s wound that included the
and frequently updated by the Clinical and
Laboratory Standards Institute (CLSI) in the
United States. The European counterpart is
The European Committee on Antimicrobial
Susceptibility Testing (EUCAST). Worldwide,
antimicro- bial susceptibility testing is guided
by the International Organization for
Standardization (ISO). This organization is a
worldwide federation of national standards
bodies. The work of preparing international
standards is conducted
 ReaSOnS and indicatiOnS fOr perfOrming antimicrObial SuSceptibility teStS 273
considered when

through ISO technical committees, whose main task

is to prepare international standards. Publication as
an inter- national standard requires approval by at
least 75% of the member bodies casting a vote. In
all cases, it is important to maintain an awareness
of which antimicrobial agents are appropriate to test,
the reliability of various test systems for detecting
antimicrobial resistance, and strategies for effec-
tively communicating results on laboratory reports to
those who must be informed.

ReasOns and indicatiOns fOr perfOrming

antimicrObial susceptibility tests
Antimicrobial susceptibility testing should be
performed on a bacterial isolate from a clinical
specimen if the isolate is determined to be a
probable cause of the patient’s infec- tion and the
susceptibility of the isolate to antimicrobials cannot
be reliably predicted based on previous experience
with the bacteria at a specific health care facility.
Although information can be determined on the
susceptibility of spe- cific bacteria from the
literature, there may be significant differences at a
specific health care facility, based on the patient
population. Susceptibility tests are not performed
on bacteria that have a predictable susceptibility
pattern to antimicrobial agents. For example, group
A β-hemolytic Streptococcus is not routinely tested
because it is univer- sally susceptible to penicillin,
the drug of choice in treating infections caused by
this bacterium. In contrast, the recom- mended
agent for treating Staphylococcus aureus infections
is oxacillin, but not all S. aureus isolates are
susceptible to oxacillin. Thus susceptibility testing is
routinely indi- cated for an S. aureus isolate that is
suspected of causing an infection.
Susceptibility testing of isolates can also provide
infor- mation on decreases in the susceptibility of
bacteria to antimicrobials. A separate set of critical
MICs other than the MICs used to judge a bacterium
susceptible or not susceptible to an antimicrobial are
developed for this purpose. These values are
termed epidemiologic cutoff val- ues (ECVs). ECV
MICs separate bacterial populations into those likely
to be wild type (susceptible) versus those that are
becoming resistant, either by mutation or
acquisition of resistance genes. When the ECV MIC
value is achieved, it is an indication that the
bacterium is likely developing resistance to the
antimicrobial. This information can be used to
adjust the use of that antimicrobial in the health
care facility to delay or avoid those species from
becoming resistant to the antimicrobial. Such
adjustments include discontinuing the use of the
drug for a time, reserving use of the drug for
specific patients, using the drug in com- bination
with another, and generally using an antimicro- bial
of a different class to treat infections caused by that
bacterium.

FactOrs tO cOnsider when determining

whether susceptibility testing is warranted
In addition to the unpredictable susceptibility of a
potential pathogen, other important factors must be
 resistant to these agents, so testing might be
performed.

determining whether antimicrobial susceptibility

testing is warranted, including:

• Body site from which the bacterium was isolated

• Presence of other organisms and quality of the
specimen from which the organism was grown
• Host’s status

BOdy site
Susceptibility tests are not routinely performed on
organisms isolated from an anatomic site for which
they are normal inhabitants. For example,
Escherichia coli is normal microbi- ota in the lower
gastrointestinal tract and therefore would not be
tested when isolated from stool. However, E. coli
from a blood culture would be tested because blood
is normally ster- ile. Similarly, viridans group
streptococci represent normal microbiota in throat
specimens and would not be routinely tested.
Coagulase-negative staphylococci isolated from mul-
tiple blood cultures would be tested; but because
coagulase- negative staphylococci are commonly
found on skin surfaces, these bacteria would not
normally be tested when isolated from superficial
wound specimens. Yeasts isolated in low numbers in
vaginal specimens or in the throat, if other micro-
biota is present, would not be considered significant.
Testing of normal microbiota isolates, or isolates
likely to represent contamination or colonization,
should be avoided because reporting antimicrobial
susceptibility results may encourage a physician to
treat a nonpathogen and prevent further inves-
tigation of the true cause of the patient’s problem.

Presence Of Other bacteria and quality Of the

specimen
The isolation of an organism in pure culture is less
likely to represent contamination than a mixed
culture. The presence of more than two species at
greater than 105 colony-forming units (CFU) per
milliliter isolated from urine suggests con-
tamination, and these organisms may not require
susceptibil- ity testing; however, a pure culture of E.
coli at more than 105 CFU/mL would likely represent
true infection and would be tested. A few colonies of
Klebsiella pneumoniae in the presence of
oropharyngeal microbiota in a sputum culture may
not be significant. However, in the absence of
oropharyngeal micro- biota, a few colonies of this
species, particularly if gram-neg- ative bacilli were
noted on a direct Gram stain of the sputum, can be
significant and warrant susceptibility testing.

HOst status
The host status of the patient often influences
susceptibility testing decisions. In an
immunocompromised patient, species usually
viewed as normal microbiota might be responsible
for an infection and therefore may require
susceptibility test- ing. Also, in patients who are
allergic to penicillin and have a β-hemolytic
streptococcal infection (e.g., group A or group B),
erythromycin or clindamycin are alternative drugs of
choice. Occasionally β-hemolytic streptococci are
274 PART 1 13 AntimicrObial SuSceptibility teSting

Selecting antimicrObial agents Specific rules are developed by laboratories for dealing with spe- cific
susceptibility test results based on institutional data and data from
fOr testing and repOrting outside sources. These include reporting antimicrobial results on
appropriate bacteria (e.g., gram-positive vs. gram-negative) from relevant
clinical sites.
Currently, more than 50 antimicrobial agents are in
use for treating bacterial and fungal infections.
Although hundreds more exist, many of these have
comparable clinical efficacy. Each laboratory must
determine which agents are appropri- ate for routine
testing against various organisms in its setting.
Testing and reporting protocols are formulated with
input from drug prescribers, representatives from the
microbiology laboratory, infectious diseases and
other services, the institu- tion’s pharmacy, and the
therapeutics committee regarding the clinical
usefulness of various agents at an institution.
Antimicrobial package inserts written by the U.S.
Food and Drug Administration (FDA) should be
consulted for infor- mation concerning the dosing
and indications for which the antimicrobial was
approved and the performance of the anti- microbial
agent during initial clinical trials.
It is important that the antimicrobial drugs used
by the laboratory match the institutional formulary
as closely as possible. From the laboratory
perspective, the limiting factor for the number of
drugs tested is usually the number that can be
practically tested with a specific method. For
example, the standard disk diffusion test for bacteria
uses a 150-mm agar plate, which can accommodate
no more than 12 disks. In the case of MIC test
panels, 12 to 15 antimicrobials can be tested on one
panel. Some automated antimicrobial susceptibility
systems can test more drugs at one time.
The patient population must be considered in the
choice of antimicrobial agents to be tested. Some
agents are contraindi- cated in pediatric patients
(e.g., fluoroquinolones, which can impair cartilage
development, and tetracycline, which damages
developing teeth). Emphasis should be placed on
testing oral agents when dealing with outpatient
specimens. Additional guidelines for developing a
testing and reporting strategy are found in the CLSI
documents for disk diffusion or MIC testing. These
documents include tables that list primary and
second- ary agents appropriate for testing against
various organism groups. CLSI guidelines are
updated at least annually.

SelectiOn Of test batteries

Generally, a laboratory will define a battery of 10 to
15 antimi- crobial agents for routine testing against
the Enterobacterales, Pseudomonas spp.,
nonfastidious gram-negative bacilli (e.g.,
Acinetobacter spp., Stenotrophomonas maltophilia,
and Burkholderia cepacia), staphylococci, and
enterococci. Sometimes a separate battery is
performed for urine isolates, representing drugs
appropriate for treating urinary tract infections
(UTIs). A sup- plemental battery that contains
antimicrobial agents with enhanced activity may be
included by laboratories that encoun- ter a significant
number of bacteria resistant to the more com- monly
used antimicrobials.

Case check 13.1

RepOrting Of susceptibility test results serious infections caused by gentamicin-susceptible
Pseudomonas aeruginosa, and tobramycin or
amikacin may be considered for gentamicin-resistant
Because the identity of the bacterial isolate is isolates. Aminoglycosides are not effective for
sometimes unknown when the susceptibility treating meningitis because they do not readily cross
test is performed, some drugs that are the blood-brain barrier.
inappropriate to report may be tested on an
A secondary agent may also be reported if the
isolate. In such cases, a drug should not be
patient has a polymicrobial infection, and a
indiscriminately reported because results can
secondary (but not a primary)
be misleading. Some drugs may appear active
against certain species in vitro but are inap-
propriate for clinical use (e.g., most
cephalosporins against methicillin-resistant
Staphylococcus aureus [MRSA], trimetho- prim-
sulfamethoxazole against enterococci). The
final deci- sion about the antimicrobials to
report is made once the identity of the isolate is
known (sometimes a preliminary identification
is sufficient), along with the overall susceptibil-
ity results and specimen source.
Reporting protocols should be developed
following dis- cussion with infectious disease
clinicians, pharmacists, and others who have
clinical experience with the antimicrobial
therapy practices of the institution. A primary
tenet of anti- microbial therapy is to use the
least toxic, most cost-effective, and most
clinically effective agents and to refrain from
use of costly, broader-spectrum agents.
Achieving physician compli- ance with this
objective can be difficult. If all drugs tested are
reported, inappropriate antimicrobial
prescribing may occur, so most laboratories
refrain from reporting broad-spectrum agents if
narrower-spectrum agents are active in vitro.
CLSI provides guidance for development of a
selective or so called cascade reporting
protocol. In addition, antimicrobial steward- ship
programs in many institutions help educate and
guide cli- nicians in appropriate and responsible
antimicrobial therapy.
For several organism groups, the CLSI
categorizes antimi- crobial agents into four
groups. An example of a listing for
Enterobacterales is seen in Box 13.1. As a
general guideline, it is suggested that within an
antimicrobial class, primary (group A) agents be
reported first and secondary (group B) agents
be reported only if one of the following
conditions exists:

• The isolate is resistant to the primary agents.

• The patient cannot tolerate the primary agents.
• The infection has not responded to the primary
agents.
• A secondary agent would be a better clinical
choice for the infection (e.g., meningitis).
• Bacteria were isolated from another site,
and a secondary agent might be useful for
treating both infections.

For example, a primary cephalosporin, such

as cefazolin (a first-generation cephalosporin),
would be a reasonable choice for a susceptible
E. coli, and secondary cephalosporins, such as
cefuroxime (second-generation cephalosporin)
or cefotax- ime (third-generation
cephalosporin), would generally not be
required. An exception would occur with
meningitis because third-generation
cephalosporins cross the blood-brain barrier
much more effectively than their first-
generation counter- parts. Gentamicin is usually
the aminoglycoside of choice for treating
 TraditiOnal antimicrObial SuSceptibility teSt methOdS 275

BOX 13.1 Suggested grOupings Of antimicrObial agents: EnterObacterales

Group A primary test and report Cefiderocol

Ampicillina Ciprofloxacina
Cefazolinb Levofloxacina
Gentamicina Doripenem
Tobramycina Ertapenem
Imipenem
Group B primary test report selectively Meropenem
Amikacina Amoxicillin– Trimethoprim-sulfamethoxazolea
clavulanic acid Ampicillin-
sulbactam Azithromycinc Group C supplemental report selectively
Ceftazidime-avibactam Aztreonam
Ceftolozame-tazobactam Ceftazidime
Imipenem-relebactam Ceftaroline
Meropenem-vaborbactam Chloramphenicola,d
Piperacillin-tazobactam Tetracyclinee
Cefuroxime Group U supplemental for urine only
Cefepime Cefazolin
Cefotetan Fosfomycinf
Cefoxitin Nitrofurantoin
Cefotaximea,b or ceftriaxonea,b Sulfisoxazole
Trimethoprim
Reprinted with permission from Clinical and Laboratory Standards Institute. (2021). Performance standards for antimicrobial susceptibility testing (31st ed.). CLSI supplement M100.
Clinical and Laboratory Standards Institute.
a
When fecal isolates of Salmonella and Shigella spp. are tested, only ampicillin, a fluoroquinolone, and trimethoprim-sulfamethoxazole should be tested and reported
routinely. In addition, a third-generation cephalosporin should be tested and reported (also chloramphenicol and azithromycin if requested) for extraintestinal isolates
of Salmonella spp. Routine testing of nontyphoidal intestinal Salmonella is not indicated, but testing on all Shigella isolates is indicated. For Salmonella and Shigella
isolates, aminoglycosides, first- and second-generation cephalosporins, and cephamycins can appear active in vitro but should not be reported as they are
ineffective clinically. bCefotaxime/ceftriaxone should be reported on CSF isolates instead of cefazolin.
c
Report for S. Typhi or Shigella spp. only.
d
Not routinely reported on isolates from urine.
e
Organisms susceptible to tetracycline are considered susceptible to doxycycline and minocycline; organisms intermediate or resistant to tetracycline may be susceptible
to doxycycline, minocycline, or both.
f
Test and report on urinary isolates for E. coli and E. faecalis only.

agent would more likely be effective against all

pathogens present. Similarly, a secondary agent may and then allowing them to grow to log phase at 35°
be reported if the patient has a disseminated C for 2 to 6 hours, until they reach the McFarland 0.5
infection, and a secondary agent would more likely standard (see later). Four to five colonies, rather
be effective at all sites. Agents with very broad- than a single colony, are selected to minimize the
spectrum activity or increased potency (group C) possibility of testing a colony that might have been
may be tested and reported for the reasons listed for derived from a susceptible mutant. Inocula can also
secondary agents. In addition, group C agents be prepared by suspending colonies grown over-
would be considered for routine testing if a night on an agar plate directly in broth or saline. This
particular institution encounters large numbers of direct inoculum suspension preparation technique is
isolates resistant to group A and group B agents. required for staphylococci and for bacteria that grow
Finally, agents with activity only in the urinary tract unpredictably in broth (e.g., fastidious bacteria) but
should be reported only on isolates from urine; these can be used for any organ- ism. Because it does not
drugs concen- trate in the urine and are clinically rely on growth in an inoculum broth, the use of fresh
ineffective in treating other infections. (16- to 24-hour) colonies is imperative.

McFarland turbidity standard

TraditiOnal antimicrObial susceptibility The inoculum concentration of bacteria to be tested
must be standardized. False-susceptible results may
test methOds occur if too few bacteria are tested, and false-
resistant results may be the out- come of testing too
InOculum preparatiOn and use Of McFarland many bacteria. The most widely used method of
inoculum standardization involves comparing the
standards turbidity of the inoculum preparation with McFarland
turbidity standards. McFarland standards can be
InOculum preparatiOn prepared by adding specific volumes of 1% sulfuric
acid and 1.175% barium chloride to obtain a barium
Inoculum preparation is one of the most critical steps sulfate solution with a specific optical density. The
in sus- ceptibility testing. Inocula are prepared by most used is the McFarland 0.5 standard, which
adding cells from four to five isolated colonies of contains 99.5 mL of 1% sulfuric acid and
similar colony morphology growing on a 0.5 mL of 1.175% barium chloride. This solution is
noninhibitory agar medium to a broth medium dispensed into tubes comparable with those used for
inoculum prepara- tion, which are sealed tightly and
stored in the dark at room
276 PART 1 13 AntimicrObial SuSceptibility teSting

temperature. The McFarland 0.5 standard provides

turbidity comparable with that of a bacterial A convenient and more precise alternative to the
suspension containing approximately 1.5 x 108 tra- ditional, highly subjective visual adjustment to
CFU/mL. Suspensions of latex par- ticles can be used match the McFarland standard is the use of a
as a simpler, more stable alternative to bar- ium nephelometer or spectro- photometer. Several
sulfate to achieve turbidity comparable with that of simple, commercially available benchtop instruments
the McFarland standard. are commonly used for more objective standard-
ization of bacterial inocula (Fig. 13.2).

InOculum standardizatiOn DilutiOn susceptibility test methOds

To standardize the inoculum, the inoculated broth or
direct suspension is vortexed thoroughly. The Principle
traditional method of comparison is to position the
tube side by side with the McFarland 0.5 standard Dilution antimicrobial susceptibility test methods are
against a white card containing sev- eral horizontal used to determine the MIC, or the lowest
black lines (Fig. 13.1) under adequate lighting. The concentration of antimi- crobial agent required to
turbidities are compared by looking at the black lines inhibit the growth of the bacterium. Various
through the suspension. The suspension is too dense concentrations of an antimicrobial agent are added
if it is more difficult to see the lines through the to broth or agar media. Generally, serial twofold-
inoculum suspen- sion than through the McFarland dilution con- centrations are tested (expressed in
0.5 standard. In such a case, the inoculum would be micrograms per milliliter). Concentrations chosen are
diluted with additional sterile broth or saline. If the those that are attainable in vivo following standard
test suspension is too light, more organisms are dosing of the antimicrobial agent. Because the
added, or the suspension is reincubated (depending concentration attainable in vivo differs with different
on the inoculum preparation protocol) until the agents, the ranges of concentrations tested also
turbidity reaches that of the McFarland standard. differ. For example, sustained concentrations greater
Once standardized, the inoculum suspensions should than 8 pg/mL for gentamicin and tobramycin cannot
be used within 15 minutes of preparation. be safely attained in the patient; therefore the
concentrations tested are generally in the range of
0.25 to 8.0 pg/mL. In contrast, much higher levels of
the extended-spectrum penicillin, piperacillin, are
attainable in vivo, and a range of 4 to 128 pg/mL
might be tested.
Once the MIC has been determined, the organism
is inter- preted as nonsusceptible, susceptible,
intermediate, or resistant to each agent with the use
of a table provided in the CLSI dilution testing
document or in the FDA-approved package insert for
the antimicrobial. Susceptible indicates that the
bacteria are inhibited by achievable concentrations
of the agent in vivo when the recommended dose is
administered to

Fig. 13.1 Left, Tha tuba cOntains a McFarland 0.5 turbidity standard. Right, Tha McFarland standard.
tuba cOntains a tast bactarial suspansiOn that has turbidity graatar than that Of
tha McFarland standard; it is mOra difficult tO saa tha black linas thrOugh tha
tast sus- pansiOn than thrOugh tha McFarland standard. Starila salina Or brOth must
ba addad tO tha tast suspansiOn tO diluta it until tha turbidity matchas that Of tha
Fig. 13.2 BanchtOp naphalOmatric-typa davicas can ba usad tO
standardiza tha turbidity Of a tast inOculum suspansiOn tO match
that Of a McFarland 0.5 turbidity (Or Othar) standard.
 TraditiOnal antimicrObial SuSceptibility teSt methOdS 277
Institute. (2021). Performance standards for antimicrobial susceptibility testing
(31st ed.). CLSI supplement M100. Clinical and Laboratory Standards
Institute.

treat an infection. Intermediate implies that

antimicrobial MICs approach the usual blood and
tissue levels, and response rates might be lower than
for susceptible isolates. This applies to agents with
lower toxicity that can be prescribed at high- er-than-
normal doses. Resistant implies that isolates are not
inhibited by the usually achievable concentrations of
the agent with normal dosage. Nonsusceptible is
used for isolates for which only a susceptible
breakpoint is designated because of the absence or
rare occurrence of resistant strains. Both the FDA
and the CLSI are involved in determining
antimicrobial interpretive criteria for MIC test results.
An example is shown in Table 13.1. Note that for
some antimicrobial agents, differ- ent MIC
interpretive criteria exist for different organisms or
organism groups. For each antimicrobial agent, the
MIC break- point separates susceptible from resistant
results. Organisms with MICs at or below the
breakpoint are susceptible, and those with MICs
above that breakpoint are intermediate or resistant.
Broth macrodilution, broth microdilution, and agar
dilution methods are used for determining MIC
values. These methods are briefly summarized in the
following sections.

AntimicrObial stOck sOlutiOns

Antimicrobial stock solutions used in MIC tests must
be prepared from reference standard antimicrobial
powders,

Table 13.1 Minimal inhibitory concentration interpretive

standards for several organism groups (µg/mL)
AntimiCrobial SusCeptible Intermediate Resistant
agent
AmpiCillin
When testing ≤8 16 ≥32
Enterobacterales
When testing ≤8 — ≥16
enterococci
When testing ≤1 2 ≥4
Haemophilus spp.
When testing ≤0.25 0.5–4.0 ≥8
viridans group
streptococci
GentamiCin
When testing ≤4 8 ≥16
Enterobacterales
When testing ≤4 8 ≥16
Pseudomonas
aeruginosa
When testing ≤4 8 ≥16
staphylococci
OxaCillin
When testing ≤2 — ≥4
Staphylococcus
aureus
When testing ≤0.5 — ≥1
coagulase-
negative
staphylococci
Reprinted with permission from Clinical and Laboratory Standards
not from the pharmaceutical preparations
administered to patients. Details of preparation are
found in the CLSI protocols. Stock drug solutions
must be stored frozen in non– frost-free freezers.
Temperature at or below –60° C is opti- mal and
necessary for more temperature-labile drugs such
as imipenem and clavulanic acid; however, –20° C
storage is acceptable for some agents. Antimicrobial
solutions are not to be refrozen after thawing.

BrOth macrOdilutiOn (tube dilutiOn) tests

Broth dilution MIC tests performed in test tubes are
referred to as broth macrodilution MIC or tube
dilution MIC tests. Generally, a twofold serial dilution
series, each containing 1 to 2 mL of antimicrobial
agent, is prepared. Mueller-Hinton broth is the
medium recommended for broth dilution MIC tests of
nonfastidious bacteria. A standardized suspension of
test bacteria is added to each dilution to obtain a
final bac- terial concentration of 5 x 105 CFU/mL. A
growth control tube (broth plus inoculum) and an
uninoculated control tube (broth only) as a sterility
check are used with each assay. After overnight
incubation at 35° C, the MIC is determined visually
as the lowest concentration that inhibits growth, as
demon- strated by the absence of turbidity.
Broth macrodilution is impractical for use as a
routine method when several antimicrobial agents
must be tested on an isolate or if several isolates
must be tested. Some laborato- ries use broth
macrodilution when it is necessary to test drugs not
included in their routine test system or for fastidious
bac- teria that require special growth media. Also,
this method can be used when minimum bactericidal
concentration (MBC) end points are to be
determined subsequently. The MBC test is discussed
later in this chapter.

BrOth micrOdilutiOn tests

The broth macrodilution test has been miniaturized
and adapted to multiwell microdilution trays (Fig.
13.3) for broth microdilution MIC testing. Plastic
trays contain between 80 and 100 (usually 96) wells.
Wells are filled with small vol- umes (usually 0.1 mL)
of twofold dilution concentrations of antimicrobial
agent in broth. Because of the large number of

Fig. 13.3 BrOth micrOdilutiOn minimal inhibitOry cOncantratiOn (MIC) tray

shOwn with an inOculum rasarvOir trOugh. A dilutad inOculum suspansiOn is
placad in tha rasarvOir trOugh. Than tha prOngs ara dippad intO tha suspansiOn,
raisad, and subsaquantly lOwarad intO tha walls Of tha brOth micrOdilutiOn MIC tray
tO inOculata all walls simultanaOusly.
278 PART 1 13 AntimicrObial SuSceptibility teSting

wells, several dilutions of as many as 12 to 15

antimicrobial agents can be contained in a single Growth may be seen as turbidity, a haze, or a pellet
tray that subsequently will be inoculated with one in the bottom of the well. Results for quality control
bacterial isolate. The inoculum suspension is (QC) organisms should be read before reading
prepared and standardized, as described ear- lier. An results for patient isolates to determine whether the
intermediate dilution of this inoculum suspension is test was performed correctly.
prepared in water or saline, and a multipronged Some laboratories use dispensing devices to
inocula- tor or other type of inoculating device is prepare broth microdilution panels, but most
used to inoculate the wells to obtain a final laboratories purchase commer- cially prepared
concentration of approximately 5 x 105 CFU/mL (5 x panels, frozen or freeze-dried. Commercially
104 CFU/0.1-mL well). The actual dilu- tion factor prepared panels must be stored and prepared for
used for preparation of the intermediate dilution use as indicated by the manufacturer. Generally,
depends on the volume of inoculum delivered to frozen panels are thawed at room temperature just
each well by the inoculating device and the before inoculation. For dried panels, the dried or
organism being tested. An example of this lyophilized drugs in the wells are reconstituted when
calculation is illustrated in Table 13.2. A growth the panels are inoculated. In addition to the panels,
control well and uninoculated control well are all manufacturers sell other materials needed for
included on each tray. After overnight incubation at testing (e.g., inoculum broths, inoculum diluents,
35° C, the tray is placed on a tray-reading device to panel inoc- ulators, reading devices). Usually, a
facilitate visual examination of each well (Fig. 13.4). variety of panels contain- ing different drugs are
Provided growth is ade- quate in the growth control available for testing various organism groups (e.g.,
well, the MIC for a particular drug is the lowest gram-positive, gram-negative), and some panels are
concentration showing no obvious growth. designed to include wells containing various
biochemical reagents so organism identification and
antimicrobial suscep- tibility testing can be done
simultaneously in a single panel. Some companies
Table 13.2 Sample calculations of dilution schema for have automated or semi-automated devices to
preparation of inocula for broth microdilution minimal inhibitory facilitate inoculation and reading. These are
concentration testsa
discussed in the following sections.
Step Resulting organism
concentration BreakpOint (cutOff) minimal inhibitOry
cOncentratiOn testing
1. Standardize suspension to McFarland A variation of the standard broth microdilution MIC
1.5 x 108 CFU/mL
0.5 standard panel is the inclusion of breakpoint testing, in which
2. Add 0.75 mL from step 1 to 25 only one or a few concentrations of certain
mL water diluent (1:33 dilution)
(4–5) x 106 CFU/mL antimicrobial agents are tested on a single panel, in
addition to standard MIC testing on other drugs.
3. Use inoculator prong set to inoculate Breakpoint (cutoff) is the term applied to the
wells of MIC tray (each prong (4–5) x 104
CFU/100-µL well concentration of an antimicrobial agent that
delivers coincides with a susceptible or intermediate MIC
0.01 mL, which results in an (4–5) x 105 CFU/mL breakpoint for a particular drug. When two
additional 1:100 dilution)
concentrations are tested and no growth is present
a
in either well, the isolate is susceptible. When there
Calculations shown here are based on use of inoculator prong set (each tha tray-raading stand, and walls ara axaminad by lOOking intO tha magnifying mirrOr. A
prong delivers 0.01 mL). Dilutions in step 2 differ, depending on the number of tray is placad Ovar tha hOla in tha light bOx, and light is allOwad tO shina thrOugh tha tray
organisms in the initial suspension and the volume delivered to each well by the tO facilitata clOsa axaminatiOn Of tha walls.
inoculating device.
CFU, Colony-forming units; MIC, minimal inhibitory concentration.

Fig. 13.4 Tray-raading stand with magnifying mirrOr (left) and light bOx ('ight).
FOllOwing incubatiOn, aithar Of thasa davicas is usad tO axamina grOwth in tha
brOth micrOdilutiOn minimal inhibitOry cOncantratiOn trays. A tray is placad On
is growth in the low concentration but no
growth in the high concentration, the isolate
has intermediate susceptibility, and a resistant
isolate grows in both wells. The qualitative
interpretation—nonsusceptible, susceptible,
intermediate, or resistant—rather than an MIC
value is reported. The primary advantage of
breakpoint over full MIC testing for a drug is
that a greater number of drugs can be tested
on a single panel because a smaller number of
wells is needed. The primary disadvantage of
breakpoint testing is that a precise MIC is not
obtained. Standalone breakpoint panels are
generally not currently used clinically.

Trailing grOwth and skipped wells

Dilution methods sometimes produce an MIC
end point that is not clear-cut, and growth in
the wells may demonstrate trailing or skipped
wells. Trailing involves heavy growth at lower
concentrations followed by one or more wells
that show greatly reduced growth in the form
of a small button or a light haze. This
commonly occurs with sulfonamides, tri-
methoprim, and trimethoprim-
sulfamethoxazole; the mode of action of the
agents allows the bacterial cells to grow
through several generations before inhibition.
In this case the trailing is ignored, and the end
point is read as an 80% reduction in growth
compared with the growth control. Trailing with
most other drugs may represent contamination
and should not be ignored unless it is known
that trailing commonly occurs with the
antimicrobial agent–organism combination.
 TraditiOnal antimicrObial SuSceptibility teSt methOdS 279

Skipped wells involve growth at higher isolates can be simultaneously inoculated onto each
concentrations and no growth at one or more of the 100-mm round Petri dish; 100-mm square plates
lower concentrations. This may occur because of generally accommo- date 36 isolates. After overnight
contamination, improperly inocu- lated wells, incubation, the MIC is read as the lowest
improper concentrations of antimicrobial agent in the concentration of antimicrobial agent that inhib- its
wells, the presence of unusual resistance with the the visible growth of the test bacterium (one or two
test iso- late (e.g., a small resistant subpopulation), colo- nies are ignored).
or a combination of two or more of these factors. As The shelf life of agar dilution plates is only 1 week
with trailing, each skipped well occurrence must be for most antimicrobial agents stored at 2° to 8° C
evaluated individually to determine whether results because many drugs are labile in this temperature
are reportable. If any doubt exists about the validity range. Because plate prepara- tion is laborious and
of the results, they should not be reported, and the this procedure is practical only if large numbers of
test should be repeated. isolates are tested, agar dilution is generally
One shortcoming of the broth microdilution MIC performed in research settings, although it is
method is its inability to produce a penicillin MIC that currently con- sidered the primary method for
is consistently within the resistant range for antimicrobial susceptibility testing of obligate
staphylococci that are low-level, β-lactamase anaerobes and is the only accepted refer- ence
producers. If the penicillin MIC is 0.03 pg/mL or method for Neisseria gonorrhoeae and Helicobacter
lower, the isolate may be reported as penicillin- pylori
susceptible; a result of 0.25 pg/mL or higher is
considered resistant. It is recommended that an Disk diffusiOn testing
induced β-lactamase test be performed on isolates
with penicillin MICs of 0.06 to 0.12 pg/mL. If the β-
lactamase test result is positive, the isolate is Principle
reported as penicillin-resistant; if the result is The disk diffusion test, commonly known as the
negative, the isolate is reported as penicillin- Kirby-Bauer test, has been widely used in clinical
susceptible. laboratories since 1966, when the first standardized
method was described. Briefly, a McFarland 0.5
standardized suspension of bacteria (as described in
Agar dilutiOn tests Inoculum Preparation earlier) in Mueller-Hinton broth
The MIC can also be determined using an agar is swabbed over the surface of a standardized
dilution method. Specific volumes of antimicrobial Mueller- Hinton agar plate, and paper disks containing
solutions are dis- pensed into premeasured volumes specific concen- trations of antimicrobial agent are
of molten agar, which is subsequently poured into placed onto the inoculated surface. After incubation
standard Petri dishes and then cooled. Mueller- of 16 to 18 hours, the diameters of the zones of
Hinton agar is recommended for testing aer- obic inhibition around each disk are measured. The result
isolates; however, this can be supplemented with is interpreted as nonsusceptible, susceptible,
sheep’s blood, to a final concentration of 5% sheep intermediate, or resistant to a particular drug
blood, or other nutrients for testing fastidious according to preset criteria. This method has been
bacteria. A series of plates con- taining various standardized for common, rapidly growing bacteria;
concentrations of each antimicrobial agent and it should not be used for slowly growing bacteria and
growth control plates without antimicrobial agent are obligate anaerobes. An isolate of Klebsiella
prepared. A standard number of test bacteria (10 4 (Enterobacter) aerogenes tested by disk diffusion is
CFU for aerobes) are spot-inoculated onto each plate shown in Fig. 13.6
using a multi- pronged replicating device (Fig. 13.5).
As many as 32 different

Fig. 13.5 Staar's raplicatOr. Tha inOculatOr prOngs ara pOsitiOnad abOva a 36-wall antimicrObial- cOntaining and cOntrOl agar platas (withOut antimicrObial agant) hava
saad trOugh that cOntains 36 diffarant standardizad inOculum suspansiOns. Tha baan inOculatad.
handla On tOp Of tha prOng unit is prassad tO lOwar tha prOngs intO all suspansiOns
simultanaOusly, and whan tha prOngs ara raisad, aach cOntains a standardizad vOluma
Of inOculum. Tha agar plata cOntaining a dafinad cOncantratiOn Of antimicrObial
agant is pOsitiOnad undar tha prOngs (tha staal plata hOlding tha agar plata slidas
back and fOrth), and tha prOngs ara carafully lOwarad tO tha agar, at which pOint
tha inOcula ara dapOsitad On tha agar surfaca. This prOcass is rapaatad until all
Fig. 13.6 Klebsiella (Ente+ObaCte+) ae+Ogenes tastad by tha disk diffusiOn
mathOd. ZOna maasuramants cOnfirm that tha isOlata is suscaptibla tO all
agants tastad, axcapt ampicillin (at tha 1-O'clOck pOsitiOn) and cafazOlin (at
tha 2-O'clOck pOsi- tiOn). NO zOnas ara prasant fOr aithar Of thasa agants.
280 PART 1 13 AntimicrObial SuSceptibility teSting

Establishing zOne diameter interpretive breakpOints

as other than susceptible (nonsusceptible) when only
The disk diffusion test depends on the formation of suscep- tible or nonsusceptible criteria are specified
an antimicrobial concentrations gradient as the for the particular antimicrobial agent–organism
antimicrobial agent radially diffuses into the agar. combination.
The drug concentration decreases at increasing
distances from the disk. At a critical point, the
amount of drug at a specific location in the medium Test perfOrmance
is unable to inhibit the growth of the test organism, Most clinical laboratories follow the protocol specified
and a zone of inhibition is formed. by the CLSI for disk diffusion testing. The CLSI
The zones of inhibition are related to MICs, and document contains explicit details for test
this rela- tionship is used to determine the performance.
breakpoints for interpret- ing a particular zone
measurement indicating that the isolate is Disk stOrage
nonsusceptible, susceptible, intermediate, or
Disks must be stored properly to ensure that the
resistant. To establish breakpoints for a single agent, drugs main- tain their potency. For long-term
the first step is to determine the optimum storage, disks are stored at –20° C or below in a non–
concentration of drug to incorpo- rate into the disk, frost-free freezer. A working sup- ply of disks can be
which considers the drug’s pharmacoki- netic stored in a refrigerator at 2° to 8° C for at least 1
properties and some of its biochemical properties week. Disks should always be stored in a tightly
(e.g., molecular size, solubility, diffusibility in agar). sealed container with desiccant. The container
Next, a sam- ple of 150 to 200 isolates with should be allowed to warm to room temperature
comparable growth rates and differing susceptibility
before it is opened to prevent condensation from
to the agent are tested by the disk dif- fusion test
forming on the disks when warm room air contacts
and a dilution MIC test, and results are plotted on a
the cold disks. Only FDA-cleared disks should be
graph. For each isolate, the observed MIC is
used.
expressed in logarithmic form (log2) and plotted on
the y-axis, and the corresponding zone measurement InOculatiOn and incubatiOn
is plotted on the x-axis on an arithmetic scale
Inoculum suspensions are prepared using a log phase
(scattergram; Fig. 13.7). Zone diame- ter breakpoints
are then selected based on a comparison of the or direct colony suspension to match the turbidity of
susceptible, intermediate, and resistant MIC a McFarland 0.5
breakpoints with the zone diameters, while standard, as described earlier. A sterile swab is
considering the typical in vivo achievable dipped into the suspension, pressed and rotated
concentrations from standard dosing. The best zone firmly against the side of the tube to express excess
diameter breakpoints are chosen to categorize liquid, and then swabbed evenly across the surface
isolates correctly with minimal interpretive errors. of a Mueller-Hinton agar plate. The plate should be
With some of the newer, extremely potent drugs, swabbed two more times, turning the plate 60
resistant isolates may not presently exist, so only a degrees each time, using the same swab (without
susceptible criterion is defined; no intermediate- or going back into the suspension) to ensure an even
resistant-zone interpretive cri- teria are specified. “lawn” of bacteria on the plate. Commercially
Further tests, possibly in a reference labo- ratory, manufactured plate inoculators are available as well
should be performed on any isolate that is (e.g., bioMérieux, Durham, NC). Usually, a plate 150
interpreted mm in diameter is used; it can accommodate testing
of as many as 12 different antimicrobial disks with
most bac- teria; placement of more than 12 disks on
Resistant Intermediate Susceptible the plate can result in overlapping zones, which are
64 difficult to measure and may produce erroneous
results. At the same time, it is recom- mended that a
32
“purity plate” (usually sheep blood agar, or chocolate
agar for more fastidious organisms) be inoculated
from the swab so the next day when results are
16
checked, it can be visually determined with certainty
MIC (mg/µL)

that the isolate was pure on the lawn plate.

8 Within 15 minutes of inoculation, the antimicrobial
disks are applied to the agar individually with
4 sterile forceps or with a multiple-disk dispenser (Fig.
13.8). The disks are pressed firmly to ensure contact
2 with the agar. Within 15 min- utes of disk placement,
the plates are inverted and placed in a 35° C
ambient air incubator for 16 to 18 hours. Although
≤1
incu- bation in an atmosphere of increased carbon
dioxide (CO2) is recommended for testing some
fastidious bacteria, this
11 20 should not be done with most organisms. Incubation
6 10 15 25 30
in CO2 results in a decreased pH, which affects the
Zone diameter (mm) activity of some antimicrobial agents.
Fig. 13.7 Scattargram and ragrassiOn analysis plOt usad tO datarmina disk cOrraspOnding zOnas Of 20 mm Or mOra wOuld ba intarpratad as suscaptibla. IsOlatas
diffusiOn zOna diamatar intarprativa braakpOints fOr hypOthatical drug X. Basad with MICs Of 32.0 µg/mL Or mOra and zOnas Of 11 mm Or lass ara rasistant. Tha
On clinical raspOnsa data, isOlatas with a minimal inhibitOry cOncantratiOn (MIC) intarmadiata dasignatiOn is usad fOr isOlatas whOsa valuas fall batwaan tha suscaptibla
Of 8.0 µg/mL Or lass ara cOnsidarad suscaptibla. As darivad frOm this scattargram, and tha rasistant MIC (16 µg/mL) and zOna intarprativa braakpOint (12 tO 19 mm).
Reading plates
After incubation, the purity plate is visually
inspected to confirm a pure culture was tested.
QC plates are read before reading results of
patient isolates to determine if the test was
performed correctly. The test plate is examined
to ensure that the organism grew satisfactorily.
The lawn of growth must
 TraditiOnal antimicrObial SuSceptibility teSt methOdS 281

Fig. 13.8 Cartridgas cOntaining antimicrObial suscaptibility tast disks ara insartad
intO tha dispansar (left). Tha dispansar (which can hOld up tO 12 diffarant
cartridgas Of disks) is pOsitiOnad Ovar an inOculatad plata, and light prassura is
appliad tO tha handla tO simultanaOusly dapOsit Ona Of aach typa Of disk OntO
tha plata. Tha tight-saaling cOntainar in tha backgrOund cOntains a dasiccant
packat and is usad fOr stOraga (at 2° tO 8° C) Of tha dispansar cOntaining a
wOrking supply Of disks.

Fig. 13.10 ROutina disk diffusiOn tasts ara axaminad by placing tha plata On Or 2
tO 3 inchas abOva a black nOnraflacting surfaca. Raflactad light is usad tO illuminata
tha plata.

swarm of growth into the zone that often occurs with

swarm- ing Proteus spp. are ignored; the obvious zone
is measured.
As with dilution tests, the end point for the
sulfonamides, trimethoprim, and trimethoprim-
sulfamethoxazole is an 80% reduction of growth.
Obvious colonies within a clear zone should not be
ignored. These colonies may occur because of
contamination or testing of a mixed culture;
however, these colonies sometimes represent a
minority resistant subpopula- tion. When such
colonies are noted, the original isolate should be
retested. If repeated testing produces the same
results, the isolate should be reported as resistant.
Transmitted light (plate held up to a light source;
Fig. 13.11) rather than reflected light will increase
the accuracy of tests with the penicillinase-resistant
penicillins linezolid and vancomycin when testing
Fig. 13.P EsChe+iChia COli tastad by tha disk diffusiOn mathOd. Tha lawn Of staphylococci and for vanco- mycin when testing
grOwth fOllOwing Ovarnight incubatiOn shOws individual cOlOnias, enterococci. Automated disk diffusion zone reading
raprasanting unsatis- factOry grOwth. Tha mOst likaly axplanatiOn fOr tha scanty devices, such as BIOMIC (Giles Scientific USA, Santa
grOwth is tha usa Of an inOculum that is tOO light Or cOntains tOO many Barbara, CA), that recognize disk codes and interpret
nOnviabla calls, rasulting in largar than nOrmal zOnas and pOtantially falsa- zones digitally are available. The advantages of
suscaptibla rasults.
these devices include speed of analysis in high-
volume laboratories and the capacity to store QC and
patient isolate images.
be confluent or almost confluent. The appearance of Once zone measurements have been made, the
individ- ual colonies is unacceptable (Fig. 13.9). If millimeter reading for each antimicrobial agent is
growth is satisfac- tory, the diameter of each compared with that specified in the interpretive
inhibition zone is measured using a ruler, or tables of the CLSI documents or FDA drug package
preferably calipers. Plates are placed a few inches insert, and results are interpreted as susceptible,
above a black, nonreflecting surface, and zones are nonsusceptible, intermediate, or resistant. An
examined from the back side (agar side) of the plate excerpt from the chart is shown in Table 13.3. The
illuminated with reflected light (Fig. 13.10). Tiny equivalent MIC breakpoints that are used to define
colonies at the zone edge and resistance and sus- ceptibility are also shown. Note
that as with MIC interpretive criteria, several sets of
interpretive criteria that are specific for various
organisms or organism groups may exist for some
antimicrobial agents.
282 PART 1 13 AntimicrObial SuSceptibility teSting

InterpretatiOn Of in vitrO antimicrObial

susceptibility test results
Several elements are important when performing in
vitro antimicrobial susceptibility tests. A summary of
the variables that must be carefully controlled in the
performance of disk diffusion and broth microdilution
MIC tests is given in Table
13.4. The procedural steps of each method must be
followed explicitly to obtain reproducible results. A
susceptibility test should never be performed using
an inoculum that is not standardized or a mixed
culture. For this reason, direct sus- ceptibility tests
that incorporate the use of a patient’s infected body
fluid cannot be recommended, even to obtain a pre-
sumptive result. Direct detection of drug resistant
genes is a much better option.
The results of a susceptibility test must be
interpreted in the laboratory before a report is
communicated to a patient’s physician. With a disk
diffusion test, the inhibition zone size must be
interpreted using a table of values that relates the
diameter of the zone of inhibition to a category of
suscepti- bility, which is sometimes related to the
identity of the iso- late. The table used for such
Fig. 13.11 Disk diffusiOn tasts fOr staphylOcOcci with Oxacillin and vancOmycin interpretations must represent the most recent
and fOr antarOcOcci with vancOmycin ara axaminad by hOlding tha plata up tO
a light sOurca (transmittad light) fOr zOna axaminatiOn. Any grOwth within tha zOna criteria that have been established by the CLSI. It is
is significant. important to recognize that the CLSI documents are
updated frequently, often annually. Use of old or out-
dated CLSI tables could represent a serious
shortcoming in patient care.
The inhibition zone size and MIC interpretive
criteria published by the CLSI and FDA are
established by careful

Table 13.3 Zone diameter interpretive standards and equivalent minimal inhibitory concentration breakpoints for several organism
groups
Antimicrobial agent Disk content
Zone diameter (nearest whole mm) Equivalent MIC breakpoints
(µg)
(µg/mL)
Resistant Intermediate Susceptible Resistant Susceptible
Ampicillin
When testing 10 ≤13 14–16 ≥17 ≥32 ≤8
Enterobacterales
When testing enterococci 10 ≤16 — ≥17 ≥16 ≤8
When testing Haemophilus 10 ≤18 19–21 ≥22 ≥4 ≤1
spp.
When testing β- 10 — — ≥24 — ≤0.25
hemolytic
streptococci
Gentamicin
When testing 10 ≤12 13–14 ≥15 ≥16 ≤4
Enterobacterales
When testing Pseudomonas 10 ≤12 13–14 ≥15 ≥16 ≤4
aeruginosa
When testing staphylococci 10 ≤12 13–14 ≥15 ≥16 ≤4
Oxacillin
When testing 30 <21 >22 ≥4 (ox) ≤2 (ox)
Staphylococcus aureus Surrogate: >8 <4
cefoxitin (cefoxitin) (cefoxitin)
When testing 1 — ≥20 — ≤0.06
Streptococcus pneumoniae
(nonpneumo- nia
isolates) for penicillin
susceptibility
MIC, Minimal inhibitory
concentration.
Reprinted with permission from Clinical and Laboratory Standards Institute. (2021). Performance standards for antimicrobial susceptibility testing (31st ed.).
CLSI supple- ment M100. Clinical and Laboratory Standards Institute.
 InterpretatiOn Of in vitrO antimicrObial SuSceptibility teSt reSultS 283

Table 13.4 Primary variables that must be controlled in performance of routine disk diffusion and broth microdilution minimal inhibitory
concentration tests
Variable Standard Comments
Inoculum Disk diffusion: 1.5 x 108 CFU/mL Use “adequate” McFarland turbidity standard (0.5 for disk diffusion)
Broth microdilution: 5 x 105 When preparing direct suspensions (without incubation), do not use growth
CFU/mL (final concentration) from plates >1 day old
Media
Formulation Mueller-Hinton Prepare in house or purchase from reliable source
Perform media quality control to verify acceptability before use for patient tests
Ca2+, 25 mg/L Ca2+, 12.5 mg/L Mg2+ Increased concentrations result in decreased activity of aminoglycosides
Mg2+ against Pseudomonas aeruginosa and decreased activity of tetracyclines
content against all organisms (decreased concentrations have the opposite effect)
Minimal or absent Excessive concentrations can result in false resistance to sulfonamides
Thymidine and trimethoprim
content
pH 7.2–7.4 Decreased pH can lead to decreased activity of aminoglycosides, erythromycin,
and clindamycin and increased activity of tetracyclines (increased pH has the
opposite effect)
Agar depth 3–5 mm Possibility for false susceptibility if <3 mm or false resistance if >5 mm
(disk diffusion)
Incubation
Atmosphere Humidified ambient air CO2 incubation decreases pH, which can lead to decreased activity of aminogly-
cosides, erythromycin, and clindamycin and increased activity of tetracyclines
Temperature 35° C Some MRSA may go undetected if >35° C
Length Disk diffusion: 16–18 hours Some MSRA may go undetected if <24 hours
Broth microdilution: 16–20 Some vancomycin-resistant enterococci may go undetected if <2 hours with
hours disk diffusion
Some HLAR (gentamicin) enterococci may go undetected if <24 hours (broth
24 hours for staphylococci with microdilution)
oxacil- lina and vancomycin and for Some HLAR (streptomycin) enterococci may go undetected if <48 hours
entero- cocci with vancomycin and (broth microdilution)
gentamicin HLAR; 48 hours for
enterococci with streptomycin HLAR;
24 hours some- times needed for
fastidious bacteria Check CLSI publication or FDA package insert (accompanying disks) for
Antimicrobial agents specifications.
Disks Use disks containing appropriate FDA-
or CLSI-defined concentration of drug
Proper storage For long-term storage, use non–frost-free freezer at –20° C or less in a
tightly sealed, desiccated container
For short-term storage (at least 1 week), maintain temperature at 2° to 8° C in a
tightly sealed, desiccated container
Allow to warm to room temperature before opening
container Proper placement on agar Place 12 or fewer disks/150-mm plate (no overlapping zones)
Solutions Prepared from reference Pharmacy-grade antimicrobial agents unacceptable (may not show
standard antimicrobial activity in vitro)
powders
Proper storage Store in non–frost-free freezer, optimally at –70° C or
less Never refreeze
End point measurement
Disk diffusion Reflected light (except for staphylococci Lawn must be confluent or almost confluent
with oxacillina and vancomycin, and for
Ignore faint growth of tiny colonies at zone
enterococci with vancomycin) and plate
held against black background edge

Zones measured from back of plate Trimethoprim and sulfonamide end point at 80% or more inhibition
Ignore swarm within obvious zone for swarming Proteus spp
Retest when colonies are within zone (except for staphylococci with oxacillin,
and enterococci with vancomycin)
Transmitted light used for vancomycin Call resistant if
staphylococci with oxacillina and any growth
vancomycin, and for enterococci with
within zone (unless possibly artifactual or contaminated)
Reproducibility of zone measurements is within ±2 mm

Continued
284 PART 1 13 AntimicrObial SuSceptibility teSting

Table 13.4 Primary variables that must be controlled in performance of routine disk diffusion and broth microdilution minimal
inhibitory concentration tests—Cont’d
Variable Standard Comments
Broth
Adequate lighting and reading device MIC is the lowest concentration that inhibits growth (turbidity, haze, or pellet)
microdilution
Sulfonamides and trimethoprim may trail (ignore trailing <2-mm buttons)
Justify “skip wells” or repeat
Staphylococci and penicillin: perform induced β-lactamase test if
MIC = 0.06–0.12 µg/mL
Reproducibility should be within ±1 twofold dilution
a
Includes all penicillinase-resistant penicillins (oxacillin, methicillin, nafcillin, and dicloxacillin).
CFU, Colony-forming units; CLSI, Clinical and Laboratory Standards Institute; FDA, U.S. Food and Drug Administration; HLAR, high-level aminoglycoside resistance; MIC,
minimal inhibitory concentration; MRSA, methicillin-resistant Staphylococcus aureus.
Modified from Hindler, J. A., & Mann, L. M. (1992). Principles and practices for the laboratory guidance of antimicrobial therapy. In R. C. Tilton, et al. (Eds.), Clinical
labora- tory medicine. St. Louis: Mosby.

analysis of three types of data: (1) microbiological results are still nonsusceptible, it may be appropriate to
data (e.g., a comparison of MICs vs. zone sizes on a send the bacte- rium to a reference laboratory. The
large number of bac- terial strains, including those nonsusceptible category commonly occurs with newer
with known resistance mecha- nisms); (2) antimicrobials because during
pharmacokinetic data (e.g., serum, cerebrospinal
fluid [CSF], urine, and other body fluids and tissue
levels of an antimicrobial agent); and (3) results of
clinical studies obtained before FDA approval and
marketing of an antimicro- bial agent. Thus MIC
interpretive criteria are not based simply on a
comparison of serum levels of an antimicrobial agent
and MICs. Zone diameter interpretive criteria are,
however, based largely on direct correlations of MICs
and zone sizes.
Whether based on the determination of an MIC or
on inter- pretation of a disk diffusion zone diameter,
the four categories of susceptibility should be
interpreted in the same manner. If the MIC or zone
size is interpreted as susceptible using the most
recent interpretive criteria, the clinical interpretation
of the result is that the patient’s infecting organism
should respond to therapy with that antimicrobial
agent using the recommended dosage for the site of
infection. Conversely, an MIC or zone size interpreted
as resistant is unlikely inhibit the bacterium by the
usually achievable concentrations of the
antimicrobial agent based on the normal dosages
used with that drug. An intermediate result indicates
that a bac- terium falls into a range of susceptibility
in which the MIC approaches or exceeds the level of
antimicrobial agent that can ordinarily be achieved,
and for which clinical response is likely to be less
than with a susceptible strain. Exceptions can occur
if the antimicrobial agent is highly concentrated in a
body fluid, such as urine, or if a higher-than-normal
dosage of the antimicrobial agent can be safely
administered (e.g., some penicillins and
cephalosporins). At times, the intermediate result
means that certain variables in the susceptibility test
may not have been well controlled, and the values
have fallen into a zone, separating susceptible from
resistant strains.
Another category that may be used for reporting is
non- susceptible. This category is used when there
are no inter- mediate or resistant interpretive
criteria, only a susceptible interpretive criterion, and
the MIC or disk diffusion zone size for classifying the
organism as susceptible is not achieved. In these
cases, the identity of the organism should be con-
firmed and susceptibility testing repeated. If the
drug development and initial clinical trials, no
organisms were found that were resistant to
the agent. The category nonsusceptible does
not necessarily mean that the organism is
resistant to the antimicrobial.
Certain other specific aspects of
susceptibility test report- ing are detailed in the
CLSI tables, such as refraining from reporting
results for antimicrobial agents that do not
enter the CSF on isolates from patients who
have meningitis. Also, results for antimicrobial
agents that are only useful for treat- ing UTIs
must not be reported on isolates from
specimens other than urine.

MOdified test methOds fOr slOw-

grOwing Or fastidiOus bacteria
Mueller-Hinton broth and agar are the standard
media used for routine dilution and disk
diffusion susceptibility tests. These media,
however, do not support the growth of all bac-
teria that require testing; consequently, routine
methods must be modified for testing fastidious
bacteria that require sup- plemental nutrients,
modified incubation conditions, or both.

StreptOcOccuS species
Streptococcus spp., including Streptococcus
pneumoniae, require a more nutritious medium
for growth; they will not grow satisfactorily on
unsupplemented Mueller-Hinton medium. Broth
dilution tests are performed in Mueller-Hinton
broth that has been supplemented with 2% to
5% lysed horse blood. The standard disk
diffusion procedure uses Mueller-Hinton agar
supplemented with 5% sheep blood incubated
in 5% CO2 for 20 to 24 hours. Tests performed
on media containing blood are examined from
the top of the plate with the lid removed. For
plates containing blood, it is important to read
the zone of growth inhibition and not the zone
of hemolysis.
Penicillins and cephalosporins remain the
drugs of choice for treating pneumococcal
infections; however, resistance to penicillin
and to other potentially useful agents is wide-
spread. Penicillin resistance in S. pneumoniae is
caused by the presence of altered penicillin-
binding proteins (drug targets in the cell wall).
It is possible to screen for penicillin suscep-
tibility in S. pneumoniae using an oxacillin disk
(1 pg), except on isolates from blood or CSF, in
which case penicillin MIC
 InterpretatiOn Of in vitrO antimicrObial SuSceptibility teSt 285
reSultS

Accurate penicillin susceptibility results are

needed for viridans streptococci isolated from
serious infections such as bacteremia or
endocarditis. If the isolate has a penicillin MIC of 0.12
pg/mL or less, penicillin alone is often prescribed;
however, higher penicillin MICs (0.25 to 2.0 pg/mL)
suggest the need for concomitant therapy with an
aminoglycoside. Isolates with penicillin MICs greater
than 2.0 pg/mL are highly resistant, and for these
Fig. 13.12 An Oxacillin (1-µg) disk is usad tO scraan fOr panicillin suscaptibility in isolates, vancomycin
performed, rather later
as described than inpen-
this icillin
chapteris for
St#eptOCOCCus pneumOniae isOlatad frOm spacimans Othar than carabrOspinal fluid generally prescribed.
staphylococci. Because of the critical nature
and blOOd. If tha Oxacillin zOna Of inhibitiOn is 20 mm Or largar frOm a nOn maningitis
isOlata, it is rapOrtad as panicillin suscaptibla. If tha Oxacillin zOna is 19 mm Or
smallar, a panicillin minimal inhibitOry cOncantratiOn (MIC) tast must ba
parfOrmad. Tha isOlata picturad has an Oxacillin zOna Of apprOximataly 15 mm,
sO a panicillin MIC tast must ba parfOrmad.

testing must be performed. A penicillin disk should

not be used for screening purposes because it is less
accurate. If the oxacillin zone of inhibition is 20 mm
or larger, the isolate can be safely reported as
penicillin (not oxacillin) susceptible (penicillin MIC ≤
0.06 pg/mL). If the oxacillin zone is 19 mm or
smaller, a penicillin (not oxacillin) MIC test must be
per- formed to determine the degree of resistance
(Fig. 13.12).
S. pneumoniae isolates from non-CSF sites with
penicillin MICs of 0.06 pg/mL or less are interpreted
as susceptible, with penicillin MICs of 0.12 to 1.0
pg/mL as intermediate, and with penicillin MICs of 2.0
pg/mL or more as resistant when orally administered
penicillin V is to be used for treatment. When
parenterally administered penicillin is to be used for
treatment of isolates from non-CSF sources (e.g.,
from spu- tum), the isolate is considered susceptible
when the MIC is 2 pg/mL or less, intermediate when
the MIC is 4 pg/mL, and resistant when the MIC is 8
pg/mL or more. If the S. pneu- moniae isolate is from
CSF, the isolate is considered suscep- tible to
penicillin when the MIC is 0.06 pg/mL or less and
resistant when the MIC is 0.12 pg/mL or more.
Determining the degree of penicillin resistance is
important because the recommended therapy may
be different for penicillin-inter- mediate and
penicillin-resistant strains. It is important to test
penicillin-resistant isolates with other clinically
relevant antimicrobial agents, such as cefotaxime or
ceftriaxone (by an MIC method, not disk diffusion),
erythromycin, a flu- oroquinolone, vancomycin, and
perhaps clindamycin or a tetracycline.
The methods for testing nonpneumococcal
streptococci are the same as those for S.
pneumoniae; however, separate breakpoints for this
group of bacteria have been defined. β-hemolytic
streptococci remain universally susceptible to
penicillin, the drug of choice for treating infections
caused by these organisms; therefore routine
susceptibility testing on β-hemolytic streptococci is
generally not necessary. For peni- cillin-allergic
patients, alternative agents for therapy include a
macrolide (e.g., erythromycin) or clindamycin. In
contrast with penicillin, some β-hemolytic
streptococci are resistant to erythromycin only or to
erythromycin and clindamycin. Isolates that appear
erythromycin-resistant and clindamy- cin-susceptible
may have inducible or constitutive resistance to
clindamycin. Before reporting clindamycin as
susceptible in these cases, a D-zone test must be
recommended; instead, MIC tests should be
performed. As with S. pneumoniae, penicillin
resistance in viridans strepto- cocci is caused by
altered penicillin-binding proteins.

HaemOphiluS influenzae and HaemOphiluS

parainfluenzae
Haemop'ilus test medium (HTM), which consists of
Mueller-Hinton base supplemented with X
(hematin) and V (nicotinamide adenine dinucleotide
[NAD]) factors, has been standardized for testing
Haemophilus in}uenzae and Haemophilus
parain}uenzae. HTM broth is used for broth dilution
tests, and HTM agar is used for disk diffusion tests.
The procedures for Haemophilus spp. are identical to
those described for nonfastidious bacteria, with the
exception that the disk diffusion test with HTM is
incubated in an atmo- sphere of 5% to 7% CO2. Zone
diameter and MIC interpre- tive criteria unique for
this genus are available. For some agents, such as
cefotaxime, only a susceptible range is defined
because cefotaxime-resistant Haemophilus spp. have
not been identified. In cases in which the determined
MIC is greater than what would make the organism
cefotaxime-susceptible, the organism is reported as
nonsusceptible. In these cases, the organism should
be reidentified and the MIC redetermined. If the
same results are obtained after retesting, the
organism should be submitted to a reference
laboratory, such as a state health laboratory.
Ampicillin or amoxicillin is often effective in
treating localized, less serious H. in}uenzae
infections; however, 25% to 50% of H. in}uenzae
isolates produce a β-lactamase that inactivates these
agents. β-Lactamase–producing isolates can be
quickly identified by using a rapid β-lactamase bench
test.
H. in}uenzae also may be resistant to ampicillin and
amoxicil- lin because of altered penicillin-binding
proteins. These iso- lates are referred to as β-
lactamase–negative, ampicillin-resistant (BLNAR).
This resistance occurs in less than 1.0% of clinical
isolates in the United States but is higher in other
countries, such as Japan. Very small proportions of
isolates possess resistance caused by altered
penicillin-binding proteins and production of β-
lactamase and are referred to as β-lactamase–
positive, amoxicillin-clavulanate–resistant (BLPACR)
isolates. The BLNAR and BLPACR isolates are both
detectable only by in vitro susceptibility testing.
There are no rapid detection methods for these types
of resistance. Because most ampi- cillin-resistant
(amoxicillin-resistant) H. in}uenzae isolates produce
β-lactamase, and H. in}uenzae is often susceptible to
alternative agents currently recommended, some
laboratories only perform a β-lactamase test or test
only ampicillin and trimethoprim-sulfamethoxazole
by the disk diffusion or MIC method. However, they
will test additional agents (e.g., fluo- roquinolones) if
the isolate is from nonrespiratory sources or if
requested by clinicians.
286 PART 1 13 AntimicrObial SuSceptibility teSting

NeiSSeria gOnOrrhOeae and NeiSSeria a fluoroquinolone (e.g., ciprofloxacin) is usually

meningitidiS administered prophylactically to individuals in close
contact with patients with meningococcal meningitis. The
Neisseria gonorrhoeae and Neisseria meningitidis are CLSI has described MIC and disk diffusion methods for
of public health significance. Therapy for testing meningococci that are the same as those used for
disseminated meningococcal infections and various testing the streptococci, except that broth microdilution
types of gonococcal infections is gen- erally empiric, tests with meningococci require CO2
based on recommendations from the Centers for
Disease Control and Prevention (CDC) and various
pro- fessional groups. Usually, clinical microbiology
laboratories are not required to perform
antimicrobial susceptibility test- ing of these two
species. Public health laboratories, however, may
receive requests to test them on a periodic basis.
The CLSI has described MIC and disk diffusion
methods for these species.
Penicillin was the drug of choice for treating
uncom- plicated gonorrhea. However, penicillin-
resistant isolates emerged in the 1970s, and
fluoroquinolone-resistant isolates emerged in the
1990s. This led to the current use of ceftriax- one or
cefixime as first-line therapy. Penicillin resistance in
N. gonorrhoeae can be caused by the production of a
β-lact- amase like that produced by ampicillin-
resistant H. in}uenzae, and this resistance can be
readily detected with a rapid β-lac- tamase test. β-
Lactamase–producing isolates are referred to as
penicillinase-producing Neisseria gonorr+oeae
(PPNG). Some N. gonorrhoeae isolates are penicillin
resistant because of an altered penicillin-binding
protein; this resistance can be detected only with
conventional dilution or disk diffusion tests. The
production of altered penicillin-binding proteins is
chromosomally mediated, and N. gonorrhoeae with
altered penicillin-binding proteins is often referred to
as chromosom- ally mediated resistant Neisseria
gonorr+oeae (CMRNG).
Because of continuing changes in the
susceptibility of N. gonorrhoeae, the need for testing
the susceptibility of isolates in hospital laboratories
may be necessary again after a hiatus for several
years. Since the implementation and widespread use
of rapid nucleic acid testing methods for N.
gonorrhoeae, however, fewer isolates are grown in
culture, making it diffi- cult, unfortunately, to monitor
these changes.
Gonococcal agar base is supplemented with
various nutri- ents for testing N. gonorrhoeae.
Dilution tests are performed using agar dilution
because this species tends to lyse in broth media,
resulting in false-susceptible results. Disk diffusion
tests are performed on the same agar, and all tests
are incu- bated in an atmosphere containing 5% to
7% CO2. The CLSI and FDA have specified
interpretive criteria unique for this species. For
several agents, resistant isolates have not yet been
encountered, so only susceptible criteria are
available.
Except for very rare isolates that produce β-
lactamase, N. meningitidis is generally susceptible to
penicillin; however, in the United States, ceftriaxone
or cefotaxime is usually the drug of choice for
treating invasive meningococcal infec- tions. Isolates
with elevated penicillin MICs (penicillin-in-
termediate) have been reported, although their
significance is minimized through the extensive use
of third-generation cephalosporins for therapy.
Meningococci are often resistant to sulfonamides
(including trimethoprim-sulfamethoxazole) and are
occasionally resistant to rifampin. For these reasons,
 for use in the routine clinical laboratory; a broth
incubation. Breakpoints for determining microdilu- tion method is used more often. The CLSI
susceptibility or resistance in meningococci for broth microdilution procedure resembles that used
several agents can be found in the latest CLSI for testing aerobes, except that Brucella broth with
tables. lysed horse blood is used for testing the

HelicObacter pylOri and CampylObacter species

The susceptibility of H. pylori to antimicrobials
is best deter- mined by agar dilution testing
using Mueller-Hinton agar containing 5% aged
(≥2 weeks old) sheep blood. Susceptibility test
plates must be incubated in a microaerobic
environment and read at 3 days of incubation.
Genotypic tests in develop- ment for
identification of the organism and resistance
genes using polymerase chain reaction (PCR)
assays and sequenc- ing techniques directly
from the specimen show promising results.
Campylobacter spp. can also be tested using
agar dilu- tion, or broth microdilution and disk
diffusion methods. CLSI guidelines detail
specific procedures for testing for both gen-
era, as well as the QC strains to use and
interpretive criteria for specific antimicrobials.

Susceptibility testing 0f agents 0f bi0terr0rism

With the occurrence of the terrorist attacks in
the United States on September 11, 2001, the
more imminent threat of the use of bacteria
and other pathogens as agents of bioterrorism
was given heightened attention. Before this
event, there were no standardized methods for
testing the susceptibility of Bacillus anthracis,
Yersinia pestis, Burkholderia mallei,
Burkholderia pseu- domallei, Francisella
tularensis, and Brucella spp. to antimicrobi- als.
In addition, knowledge was limited regarding
the use of antimicrobials to prevent or treat
infections caused by these bacteria. Since then,
the CLSI has standardized the methods for
susceptibility testing of these bacteria against a
variety of antimicrobials and has established
interpretive criteria for the results of these tests
(publication M45). In addition, a great deal
more information is now available on the use of
anti- microbials that can be used to prevent
infections with these organisms and to treat
infections if they do occur. However, these
organisms are an imminent danger when being
tested in the laboratory, and work with these
organisms should be conducted only in a
biosafety level 2 (BSL2) or higher facility by
trained individuals. This limitation applies even
to routine susceptibility tests. As soon as it is
suspected that a specimen may contain one of
these agents or that one of these bacteria has
been isolated, the specimen or isolate must be
quaran- tined and sent to a state health
laboratory after state health officials are
alerted. Clinical laboratories should establish
contingency plans to deal with situations in
which bioter- rorism agents have appeared, or
potentially appeared, in the laboratory,
especially if a staff member has been exposed
and might need prophylaxis.

Obligate anaer0bic bacteria

The reference method described by the CLSI for
testing anaerobic bacteria is an agar dilution
method, and the rec- ommended medium is
supplemented Brucella laked sheep blood agar.
As noted, however, agar dilution is not practical
 InterpretatiOn Of in vitrO antimicrObial SuSceptibility teSt reSultS
287
reliable results than testing any of the other agents.
When an isolate shows resistance to one of the
penicillinase-resis- tant penicillins, it must be
B. fragilis group of anaerobes. Also, the number of considered resistant to the entire group.
organisms in the test inoculum is 0.5 log10 higher (106 Staphylococcal resistance to the penicillinase-
CFU/mL) than that for testing aerobes, and panels resistant
are incubated anaerobically at 35° C for 48 hours. As
with tests for aerobic bacteria, the CLSI has defined
susceptible, intermediate, and resistant criteria for
the interpretation of MICs for anaerobes. Several
commer- cial companies produce broth microdilution
MIC panels for testing anaerobes. The Etest (AB
Biodisk, Solna, Sweden), discussed in detail later, has
been shown to perform satis- factorily for
susceptibility testing of anaerobes and is used in
many clinical laboratories.

Infrequently encOuntered Or fastidiOus bacteria

Besides anaerobes, the CLSI publication M45 focuses
on infre- quently encountered or fastidious bacteria
not addressed in the standard MIC and disk diffusion
documents. When clin- ically indicated, isolates of
corynebacteria (and some other gram-positive
bacilli), Aeromonas, Vibrio, Pasteurella, Moraxella
catarrhalis, some members of the HACEK
(Haemophilus spp., Aggregatibacter spp.,
Cardiobacterium spp., Eikenella corrodens, and
Kingella spp.) group of gram-negative bacteria, and
Abiotrophia and Granulicatella can be tested using
various stan- dard CLSI methods; the results are
interpreted using the CLSI tables specific for each
organism. CLSI publication M24-A2 (2011) offers
guidelines for antimicrobial testing of acid-fast bacilli,
including Mycobacterium tuberculosis,
nontuberculous mycobacteria (NTM), and Nocardia.
Yeasts and filamentous fungi can be tested using
agar dilution, standardized disk dif- fusion and Etest
tests, broth macrodilution, and microdilution
techniques. Some commercial automated systems
can per- form antifungal susceptibility testing of
Candida spp. as well.

AdditiOnal Organism and antimicrObial agent

testing cOncerns
Special procedures must be used to detect clinically
signifi- cant resistance in some nonfastidious
bacteria.

Case check 13.2

Laboratories' procedures include ways of detecting drug resis-

tance in all clinically significant microorganisms. The method must
be determined to provide an accurate and reproducible result. It
is important to know which assay to use and if modifications are
necessary for unusual or fastidious isolates.

DetectiOn Of Oxacillin (methicillin) resistance in

staphylOcOcci
Oxacillin and other penicillinase-resistant penicillins,
such as methicillin, nafcillin, cloxacillin, and
dicloxacillin, consti- tute the drug class of choice for
treating staphylococcal infec- tions. Oxacillin is the
class representative generally used to detect
resistance in staphylococci and has produced more
 staphylococci.
Testing a surrogate marker of resistance (i.e.,
cefoxitin) may provide a more accurate indication of
penicillins is most often caused by the presence of a oxacillin resistance
unique penicillin-binding protein (PBP2a or PBP2u) in
the cell wall. The mecA gene encodes an altered
penicillin-binding protein, which has a low affinity for
binding all β-lactam drugs. In addition, oxacillin
resistance in staphylococci can be caused by the
mecC gene, which has been recovered from isolates
from livestock and humans with animal exposure,
although the prevalence is currently low. The mecC
gene has about 70% homology to mecA and codes
for a PBP2a protein with only about 63% similarity to
the mecA product. Therefore the PCR assay for mecA
and the antigen tests for PBP2a will not detect mecC
oxacillin resistance.
Although methicillin is no longer available, isolates
of oxa- cillin-resistant S. aureus are commonly
referred to as MRSA for historic reasons. Because
some staphylococci exhibit heterore- sistance or
heterogeneous expression of resistance to oxacillin,
detecting oxacillin resistance in isolates that possess
the mecA gene has sometimes proven difficult under
standard suscep- tibility testing conditions. In
heteroresistant isolates, all cells in the population
have the genetic element (mecA gene) for oxacillin
resistance, but not all the cells express this
resistance by virtue of PBP2a production.
Consequently, in the oxacillin susceptibility test,
some cells may appear resistant, and some can
appear susceptible. If too few cells appear resistant,
an oxacillin-resistant strain might not be detected. In
addition, a rapid (5-minute) PBP2a colorimetric
bench test, such as the Clearview PBP2a SACulture
Colony Test (Abbott Laboratories, Chicago, IL), may be
performed on the isolate. However, this is not 100%
reliable in detecting oxacillin resistance.
In vitro testing conditions can be modified to
enhance the expression of oxacillin resistance, as
follows:

• Preparation of inocula using the direct inoculum

suspen- sion procedure
• Incubation of tests at temperatures no higher than
35° C
• Making final test readings after a full 24
hours of incubation
• Supplementation of Mueller-Hinton broth or
agar with 2% NaCl for dilution tests

The extended incubation allows the more slowly

growing resistant subpopulation sufficient time to
grow to detectable numbers. In addition, test plates
should always be examined very closely. For
oxacillin disk diffusion tests, zones of inhi- bition
must be examined by using transmitted light (hold-
ing the plate up to the light source; see Fig. 13.11),
and any growth is considered significant. A haze of
growth within the inhibition zone for oxacillin-
resistant isolates is sometimes observed (Fig.
13.13).
An oxacillin-salt screen plate that contains
Mueller-Hinton agar supplemented with 4% NaCl
and 6 pg/mL oxacillin has been used to detect MRSA.
The high salt concentration makes the medium
selective for staphylococci. To perform the oxa- cillin
screen test, a McFarland 0.5 suspension is prepared.
A swab is dipped into this suspension and streaked
over an area of approximately 2 x 5 cm or
deposited as a spot on the agar surface. After
overnight incubation at 35° C, growth (more than
one colony) is an indication that the isolate is
oxacillin resistant. This method does not reliably
detect oxacillin-resis- tant, coagulase-negative
288 PART 1 13 AntimicrObial SuSceptibility teSting

susceptible or oxacillin-resistant. MRSA is a significant

Fig. 13.13 Tha Oxacillin zOna fOr hatarOrasistant Oxacillin-rasistant

StaphylOCOCCus au-eus Oftan shOws a haza Of grOwth within tha zOna Of
inhibitiOn. This haza is significant, and tha isOlata hara is Oxacillin rasistant.

than testing oxacillin itself. This is because cefoxitin

induces greater expression of PBP2a in mecA-
containing strains of staphylococci and functions as a
test agent to detect resis- tance. Studies have shown
that performing a disk diffusion test with cefoxitin
and using the specific cefoxitin breakpoints
recommended by the CLSI provides sensitivity and
specific- ity equivalent to testing oxacillin (by disk
diffusion and MIC methods) with S. aureus and
greater sensitivity and specificity of cefoxitin when
testing coagulase-negative staphylococci. With S.
aureus, cefoxitin disk diffusion test results are more
easily interpreted than the results of disk diffusion
tests using oxacillin. Thus cefoxitin disk diffusion
testing is now recom- mended by the CLSI as the
preferred method for detection of oxacillin resistance
in S. aureus and coagulase-negative staph- ylococci.
Even though it is coagulase-negative,
Staphylococcus lugdunensis has characteristics that
resemble those of S. aureus, and the results of the
cefoxitin screen test should be inter- preted using S.
aureus interpretive criteria. It is important to report
the findings from the cefoxitin disk diffusion test as
indicative of oxacillin susceptibility or resistance;
cefoxitin and cefazolin results should not be
reported.

Case check 13.3

Just because a bacterium is susceptible in vitro to an antimicro-

bial, it does not mean that it can be used to treat an infection
caused by the bacterium. In some cases, additional testing is
required, as in the case of clindamycin-inducible resistance.

Sometimes, oxacillin-resistant staphylococci can

appear susceptible in vitro to other β-lactam agents,
such as the cepha- losporins; however, these are
clinically ineffective. Regardless of the in vitro test
results, all oxacillin-resistant staphylococci must be
reported as resistant to all β-lactam agents (includ-
ing cephalosporins, β-lactam–β-lactamase inhibitor
com- binations, and carbapenems) if those agents
are tested. For practicality, it is better simply to
perform the cefoxitin disk diffusion test and deduce
susceptibility to other β-lactam agents based on
whether a staphylococcal isolate is oxacil- lin-
 cause of community-associated and concern because vancomycin is one of the most
healthcare-associated infections and requires proven agents for treating serious MRSA infections.
that all clinical microbiology laborato- ries VRSA is defined as isolates with a vancomycin MIC
accurately detect drug resistance in S. aureus. ≥16 pg/mL.
Previously, healthcare-associated strains of The CLSI recommends use of the broth
MRSA were often resistant to several other microdilution test or vancomycin agar screen for
drug classes in addition to the β-lactams. enterococci as the best methods
Although recent community-associated MRSA
isolates often have been resistant only to the
macrolides (e.g., erythromy- cin) in addition to
various β-lactams, an increasing number of
MRSA isolates from the community have the
same antimicro- bial resistance profile as
healthcare-associated MRSA isolates. Rare S.
aureus isolates have a subtler and less
common type of oxacillin resistance unrelated
to the presence of the mecA or mecC genes.
The resistance mechanism in these iso- lates is
caused by hyperproduction of β-lactamase, or
some- times the presence of a point mutation
in the gene coding for penicillin-binding protein
but not PBP2a. These isolates generally have
MICs just above (typically 1 to 8 pg/mL) or
zones of inhibition just below the breakpoint for
oxacillin sus- ceptibility, and they are
sometimes referred to as borderline oxacillin-
resistant G. aureuS (BORSA). Currently, the CLSI
breakpoint for S. aureus <2 pg/mL is sensitive
and >4 pg/mL is resistant; there is no
intermediate result. Isolates with bor- derline
oxacillin resistance generally do not grow on
oxacillin screening plates. The clinical response
of BORSA to penicilli- nase-resistant penicillins
and to other β-lactam agents has not been
clearly defined. The exact prevalence of BORSA
strains is difficult to establish but is estimated
to be 5% of S. aureus isolates. One study found
that approximately 5% of asymp- tomatic
children carried BORSA in the nares.
Antimicrobial administration is likely the main
risk factor for developing
the BORSA phenotype.

VancOmycin resistance Or diminished

susceptibility in StaphylOcOccuS aureuS
Between 2002 and 2005, five different isolates
of MRSA with vancomycin resistance were
detected for the first time. These isolates had
apparently either acquired a plasmid contain-
ing the vanA vancomycin resistance gene from
vancomycin- resistant enterococci (VRE) or had
developed mutations resulting in their cell walls
being less susceptible to vanco- mycin. The
MRSA isolates demonstrated various levels of
resistance to vancomycin. Some were
obviously resistant by routine susceptibility
testing methods, and others were initially
missed by routine testing methods. This
important resistance followed recognition in
1996 of the first MRSA iso- late, with subtle
diminished susceptibility to vancomycin.
Shortly thereafter, several similar isolates were
encountered that all had vancomycin MICs of 8
pg/mL. Isolates of S. aureus with reduced
susceptibility to vancomycin (MIC 4 to 8pg/mL)
are called vancomycin-intermediate
Gtap+ylococcuS aureuS (VISA). Resistance is not
caused by the vanA gene but by changes in the
cell wall, rendering them less susceptible to
vancomycin. Although still uncommon, VISA
and vancomy- cin-resistant Gtap+ylococcuS aureuS
(VRSA) isolates (as of 2019, only 14 strains of
VRSA have been documented in the United
States between 2002 and 2015) are of great
 InterpretatiOn Of in vitrO antimicrObial SuSceptibility teSt reSultS 289

for detection of VISA and VRSA. The vancomycin agar

screen plate contains brain-heart infusion (BHI) agar EnterOcOcci
supplemented with 6 pg/mL vancomycin and is The enterococci express low-level resistance to
useful in screening for van- comycin resistance in S. several anti- microbial agents including the
aureus and enterococci. The vanco- mycin disk aminoglycosides and β-lact- ams. In addition, they
diffusion test has not uniformly detected VISA and can also express high-level resistance to several
VRSA isolates, and some of the commercial agents. Penicillin resistance among the Enterococcus
susceptibil- ity testing instruments have similarly not spp. is generally caused by overexpression of PBP
always provided reliable detection of these strains. In with low affinity to the penicillins. Occasionally,
addition, heteroresistant vancomycin-intermediate S. Enterococcus iso- lates express a β-lactamase,
aureus (hVISA) are isolates that normally test as notably Enterococcus faecalis and
susceptible by standard methods but contain a Enterococcus faecium. β-lactamases inactivate β-
lactam mole- cules by disrupting the β-lactam ring
subpopulation of cells that test as nonsusceptible for component of the mole-
vanco- mycin. The use of the macro Etest method, a cule (Fig. 13.15). Ampicillin or penicillin is normally
modification of the standard Etest discussed later, effective in treating uncomplicated enterococcal
has proven to be valuable in detecting hVISA. infections (e.g., UTIs) by most Enterococcus spp.
Because the test uses a higher concentra- tion of These cell wall active agents are only bacteriostatic
bacteria (approximately 1 x 108/mL) than that used against enterococci, and when adminis- tered alone
routinely for susceptibility testing, the probability of are inadequate for treating serious infections, such as
detect- ing the small number of these organisms in endocarditis, that require bactericidal therapy. To
the overall pop- ulation of cells is enhanced. The obtain a bactericidal effect, ampicillin or penicillin (or
Etest method can be used for detecting vancomycin in patients with penicillin allergies or
heteroresistance in other organisms to other with E. faecium infection) must be given in
antimicrobials. combination with an aminoglycoside, usu- ally
gentamicin or sometimes streptomycin.
Inducible clindamycin resistance in staphylOcOcci
Two different mechanisms confer macrolide (e.g.,
erythromy- cin) resistance in staphylococci. The erm
gene codes for meth- ylation of the 23 S ribosomal
ribonucleic acid (rRNA), which results in resistance to
erythromycin and inducible or consti- tutive
resistance to clindamycin. The msrA gene codes for
an efflux mechanism, which results in resistance to
erythromy- cin but susceptibility to clindamycin.
Consequently, when an erythromycin-resistant and
clindamycin-susceptible staph- ylococcal isolate is
encountered, a D-zone test for inducible clindamycin
resistance must be performed before clindamy- cin is
reported as susceptible. For the D-zone test, an
eryth- romycin disk is placed adjacent to a
clindamycin disk (15 to 26 mm, edge to edge) as
part of a standard disk diffusion test. After overnight
incubation, flattening of the clindamy- cin zone
between the two disks (Fig. 13.14) indicates that the
isolate has inducible clindamycin resistance because
of erm. No flattening indicates that the isolate is
erythromycin-resis- tant only (because of msrA). Fig. 13.14 D-zOna tasting Of StaphylOCOCCus au-eus that has e-m-madiatad
When an isolate demonstrates inducible resistance, inducibla clindamycin rasistanca. Tha pOsitiva raactiOn is nOtad by a flattaning Of
tha zOna arOund tha clindamycin disk in tha araa in which thara has baan diffusiOn
clindamycin is reported as resistant. The Of arythrOmycin and clindamycin mOlaculas Ovarlap.
phenomenon of clindamycin-inducible resistance also
exists in S. pyogenes and S. pneumoniae and can be
detected in a sim- ilar manner. Some automated
susceptibility instruments can detect this inducible
resistance.

β-Lactam ring

CH3 CH3
S CH3
R-CONH S CH3 R-CONH

β-Lactamase

N
N
O CO2 O
H CO2
O

Penicillin Penicilloic acid

Fig. 13.15 β-Lactamasa hydrOlyzas tha β-lactam ring pOrtiOn Of tha panicillin mOlacula. Tha hydrOlysis rasults in tha fOrmatiOn Of panicillOic acid, which dOas nOt
hava antibactarial activity.
290 PART 1 13 AntimicrObial SuSceptibility teSting

Enterococci with the blaZ gene constitutively resistance to the other aminoglycosides listed; enzymes that modify gentamicin also
express β-lac- tamase at a much lower level than S. modify the other agents.
aureus, resulting in a clinically significant inoculum b
An isolate that has high-level resistance to streptomycin has high-level resistance to
effect. Inoculating bacteria at concentrations for this agent only, unless it also has high-level resistance to gentami- cin as indicated by
testing high concentrations of gentamicin.
routine susceptibility testing (generally 1 x 105
organisms/mL), enterococci produce so little
enzyme
that they test susceptible to β-lactam agents.
However, at a
high inoculum, such as during an infection, the
presence of more enzyme leads to resistance. Once
the presence of a β-lac- tamase is identified, the
addition of the β-lactamase inhibitor sulbactam (i.e.,
ampicillin-sulbactam) is sufficient to restore the
antibiotic efficacy.

DetectiOn Of high-level aminOglycOside resistance in

enterOcOcci
Enterococci are inherently resistant to low
concentrations of aminoglycosides, precluding their
use as single agents for treatment of enterococcal
infections. This low-level resistance is caused by poor
drug uptake by enterococcal cells. For isolates with
low-level aminoglycoside resistance, a synergistic
interac- tion occurs when an aminoglycoside is
administered together with a cell wall active agent
such as ampicillin, penicillin, or vancomycin.
Sometimes, however, enterococci develop high- level
aminoglycoside resistance in which the aminoglyco-
side does not demonstrate synergism with the cell
wall active agent. High-level aminoglycoside
resistance in enterococci is usually caused by
enzymatic inactivation of the drugs, and the enzymes
that destroy gentamicin also destroy tobramycin,
amikacin, and kanamycin (Table 13.5).
Consequently, none of these agents is used in
treating serious infections caused by enterococci with
high-level gentamicin resistance. If the isolate does
not have concomitant high-level streptomycin
resistance, however, streptomycin could be used,
although some isolates have high-level resistance to
gentamicin and streptomycin.
In vitro tests for the detection of high-level
aminoglycoside resistance include broth, agar, or
disk diffusion methods. For screening, gentamicin is
tested at concentrations of 500 pg/ mL and
streptomycin at 2000 pg/mL (agar) or 1000 pg/mL
(broth). The tests are performed as described for
routine dilu- tion tests, and growth at the high
concentration indicates that the isolate has high-
level resistance to the agent tested. Disk diffusion
tests have also been described that use special
potency disks containing high amounts of gentamicin
(120 pg) or streptomycin (300 pg).

Table 13.5 Aminoglycosides represented when high

concentrations of gentamicin and streptomycin are tested to
determine high-level aminoglycoside resistance in enterococci
Agent tested Aminoglycoside(s) represented
a
Gentamicin Gentamicin
Amikacin
Kanamycin
Tobramycin
b
Streptomycin Streptomycin
a
An isolate that has high-level resistance to gentamicin also has high-level
DetectiOn Of vancOmycin resistance in enterOcOcci of sensitivity for detection, cefpodoxime,
The incidence of vancomycin resistance in ceftazidime, cefotaxime, ceftri- axone, and
enterococci increased sharply in the 1990s and aztreonam and the use of special screening zone or
the early part of the first decade of this century. MIC breakpoints to facilitate recognition of ESBL
Among clinical isolates, E. faecium is the most produc- tion. Indicator drugs have been selected
common species demonstrating vancomycin based on the likeli- hood of their being readily
resis- tance followed by E. faecalis. The most hydrolyzed by one of the many types of ESBLs. The
common mechanisms are the vanA and vanB use of more than one antimicrobial agent increases
genes, but other resistance genes have been the sensitivity of detection.
described. Isolates that are highly resistant to
vanco- mycin can be readily detected when
tested by conventional antimicrobial
susceptibility test methods. Some isolates, how-
ever, may have a more subtle type of
vancomycin resistance, whereby MICs or
inhibition zone measurements are just slightly
above or below, respectively, the breakpoints.
Hence dilution tests must be viewed closely,
and inhibition zones in disk diffusion testing
must be examined using transmit- ted (rather
than reflected) light; any growth within the
zone should be considered significant.
Vancomycin screen agar can be used to detect
vancomycin resistant enterococci as well.
Low-level vancomycin resistance is intrinsic
in the motile enterococcal species Enterococcus
gallinarum and Enterococcus casseli}avus.
These differ from true VRE for infection control
purposes. In addition, vancomycin-variable
enterococci have been described. These
isolates initially appear susceptible to
vancomycin but possess the vanA gene
(detected by molec- ular methods) and can
develop into vancomycin-resistant strains in
vivo; therefore they should be reported as
vanco- mycin resistant. Leuconostoc,
Pediococcus, and Lactobacillus spp.
demonstrate intrinsic, high-level vancomycin
resistance; these bacteria must be carefully
separated from the morpho- logically similar
enterococci and streptococci.

Extended-spectrum β-lactamases
Most K. pneumoniae, Klebsiella oxytoca, and
many E. coli isolates are resistant to ampicillin
because of the production of a plas- mid-
mediated β-lactamase: TEM-1 or SHV-1. Most
isolates are susceptible to later-generation
cephalosporins and aztreonam; however,
spontaneous mutations occur that can result in
novel β-lactamases that can inactivate
extended-spectrum cephalo- sporins, penicillins,
and aztreonam. These β-lactamases are known
as extended-spectrum β-lactamases (ESBLs).
The enzymes are characterized and numbered
based on their rela- tionship to the parent
enzymes. There are now more than 130 TEM-
derived ESBLs and more than 60 derived from
SHV-1. In addition, other novel β-lactamases
have been designated as CTX-M or OXA that are
not a result of mutations of the TEM or SHV
parent enzymes. Some of these β-lactamases
give rise to subtle or difficult to detect
resistance among cephalospo- rins, penicillins,
or aztreonam. Recently these enzymes have
been found in other genera and species,
including Proteus mirabilis, Salmonella spp.,
and Enterobacter spp. most notably.
Strategies for laboratory detection of ESBL-
producing E. coli, Klebsiella spp., and P.
mirabilis include testing of drugs that are most
likely to indicate the presence of an ESBL (indi-
cator drugs). These include, in decreasing order
 InterpretatiOn Of in vitrO antimicrObial SuSceptibility teSt reSultS 291
Klebsiella pneunoniae carbapenemase
K. pneumoniae carbapenemases (KPCs) are the
most com- mon type of CPE in the United States.
Initially identified

Fig. 13.16 Extandad spactrum β-lactamasa (ESBL) phanOtypic cOnfirmatOry

tasting Of Klebsiella pneumOniae. Tha standard disk diffusiOn tast is parfOrmad
with cafO- taxima and caftazidima, with and withOut clavulanic acid. Tha diamatar
Of tha zOna arOund cafOtaxima–clavulanic acid (4-O'clOck pOsitiOn) is mOra
than 5 mm largar than tha zOna arOund cafOtaxima (5-O'clOck pOsitiOn), which
indicatas a pOsitiva raactiOn as clavulanic acid rastOras tha activity Of cafOtaxima.
AlthOugh tha zOnas fOr caftazidima and caftazidima–clavulanic acid ara
cOmparabla, Only Ona sat Of drugs must ba pOsitiva tO cOnfirm an isOlata as an
ESBL prOducar.

Once an ESBL-producing isolate is presumptively

detected, confirmatory testing is optional per CLSI,
because the break- points for the third- and fourth-
generation cephalosporins have been lowered. ESBL
activity is inhibited by β-lactamase inhibitor agents
such as clavulanic acid; therefore this prop- erty
forms the basis of the confirmatory tests. If the
activity of cefotaxime, ceftazidime, or both is
restored when tested in combination with clavulanic
acid by disk diffusion or an MIC test, the resistance is
caused by ESBL production. Fig. 13.16 shows an
ESBL confirmatory test performed using the disk
diffusion method.
Resistance in ESBL-producing strains to various β-
lactam agents is not always predicted by in vitro
tests performed using standard inoculum density. For
example, a confirmed ESBL-producing K.
pneumoniae isolate may appear suscepti- ble to
cefotaxime by a disk diffusion or MIC test; however,
cefotaxime is hydrolyzed at a high inoculum
density and is ineffective in treating infections
caused by these strains. Consequently, when an
ESBL-producing isolate is identified and confirmed, it
should be reported as clinically resistant to all
cephalosporins, penicillins, and aztreonam, despite
the in vitro test results. The carbapenems
(imipenem, meropenem, ertapenem) are active
against ESBL-producing strains as are the
cephamycins (cefoxitin and cefotetan). ESBL-
producing isolates show variable susceptibility to
aminoglycosides, fluo- roquinolones, and
trimethoprim-sulfamethoxazole, although many
isolates show multiple drug resistance to those
agents.

Carbapenemases
Carbapenemase-producing Enterobacterales (CPE)
are increasingly seen in the clinical setting. These
organisms carry resistance genes, often on plasmids,
to carbapenem antibiot- ics as well as penicillins,
monobactams, and cephalosporins to different
degrees. It is vitally important to detect CPE for
effective patient treatment and infection control
purposes.
 cephalosporin nitrocefin. Cefinase disks BD BBL
(Thermo Fisher Scientific, Waltham, MA) are a
commercial
in K. pneumoniae, they have been recognized in a
variety of other members of the Enterobacterales, as
well as some nonfermenters, such as Pseudomonas
aeruginosa. Organisms possessing these enzymes
are often resistant to one or more of the
carbapenems. If an isolate possesses an ESBL and
the MIC for a carbapenem is 2 or 4 pg/mL, there is a
possibility that it may produce a KPC-type or some
other carbapenemase.
Molecular methods can detect carbapenemase
genes, but these tests are limited in terms of
required expertise, time, and sample number. A
simple, objective, inexpensive phenotypic method,
the carbapenem inactivation assay, was endorsed by
the CLSI in 2017. In this test, a carbapenem disk is
incubated in water or trypticase soy broth (TSB)
containing a suspen- sion of a suspected
carbapenemase-producing organism for 4 hours.
Inactivation of the disk is analyzed by placing that
carbapenem disk on a lawn plate of a carbapenem-
suscepti- ble control strain, resulting in no zone of
inhibition after over- night incubation. Lateral flow
assays, such as the NG-Test Carba 5 (Hardy
Diagnostics, Santa Maria, CA), have been developed
that will detect up to 5 different carbapenemases. In
vitro susceptibility testing of members of the
Enterobacterales containing KPCs may indicate that
the isolate is susceptible to carbapenems, but
therapeutically the carbapenem antimicro- bial
might not be effective. Infections often are treated
with colistin or polymyxin with mixed results.

β-Lactamase tests
As mentioned previously, the β-lactamases inactivate
some β-lactam antibiotics. Production of β-lactamase
is a signifi- cant mechanism contributing to
antimicrobial resistance in organisms, such as
Staphylococcus spp., H. in}uenzae, N. gonor- rhoeae,
M. catarrhalis, several Bacteroides spp., and
occasionally Enterococcus spp. and Pseudomonas
spp. β-Lactamases are the primary mode of
resistance to β-lactam antibiotics. Another method
discussed earlier is altered PBPs with lower affinity to
the β-lactams.
Several tests can be performed in the clinical
laboratory to identify β-lactamase production, and
the rapidity of β-lac- tamase tests makes them
attractive as a means for direct detection of one
important resistance determinant. A posi- tive result
means that the β-lactam agents commonly used
(ampicillin, amoxicillin, and penicillin) to treat
infections caused by the organism would be
ineffective. Some β-lac- tams (e.g., oxacillin),
however, are resistant to β-lactamase. A negative
reaction indicates that the test organisms do not
produce β-lactamase, although they may be resistant
to these agents through another mechanism.
Resistance related to other mechanisms can be
detected only with conventional dilution or disk
diffusion tests. Some organisms, such as members of
the order Enterobacterales and the Pseudomonas
spp., produce a variety of different types of β-
lactamases; the currently available direct β-
lactamase tests cannot reliably predict resistance to
the newer β-lactam agents that might be used for
these organisms, so β-lactamase testing should not
be performed on them.
The most accurate means to determine β-
lactamase pro- duction is by detecting the blaZ gene
with real-time PCR. The most used direct detection
method incorporates the chro- mogenic
292 PART 1 13 AntimicrObial SuSceptibility teSting

product consisting of filter paper disks impregnated

edge are considered positive for β-lactamase
with nitrocefin approved for use on isolated colonies
production, while those with a feathered edge are
of N. gon- orrhoeae, Staphylococcus spp., H.
regarded as nega- tive. Reading the results is
in}uenzae, enterococci, and anaerobic bacteria. A
highly subjective; zones must be examined
disk is moistened with water or saline, and a
loopful of organisms is applied directly onto the disk. carefully with reflected light. The European
A β-lactamase–producing organism will turn red Committee on Antimicrobial Susceptibility Testing
within 10 minutes (or within 60 minutes for recom- mends using a 1-unit penicillin disk and a
staphylococci). No color change indicates a β- breakpoint of
<26 mm; studies indicate it to be more sensitive in
lactamase–negative organism (Fig. 13.17). detect- ing β-lactamase production in
Other less commonly used β-lactamase tests staphylococci. The CLSI
utilize peni- does not recommend the zone edge test for
cillin-based acidimetric and iodometric tests that coagulase-neg- ative Staphylococcus spp.
detect the production of penicilloic acid from the
hydrolysis of peni- cillin by a β-lactamase. The Etest
acidimetric method uses citrate- buffered penicillin
and phenol red as a pH indicator. When β-lactamase– The Etest uses the principle of an antimicrobial
producing bacteria are added to the solution, the density gra- dient in agar medium as a means of
penicilloic acid results in a drop in pH, causing a determining antimi- crobial susceptibility. The Etest
color change from red to yellow. In the iodometric uses thin plastic test strips impregnated on the
method, a solution of phosphate-buffered penicillin undersurface with an antimicrobial con- centration
and starch-iodine complex is used. With β- gradient and marked on the upper surface with a
lactamase–positive organisms, penicilloic acid concentration index or scale. The strips may be
reduces iodine and prevents it from combining with placed in a radial fashion on the surface of an agar
starch. A positive reaction is colorless, and a plate that has been inoculated in a manner like that
negative reaction is purple. for a disk diffusion test. After overnight incubation,
Most bacteria, except staphylococci, produce β- the test results are read by viewing the plates from
lactamase the top, with the lids removed. The antimicrobial
constitutively, meaning that the same amount of gradient that forms in the agar around the Etest
enzyme is produced regardless of exposure to an strips gives rise to an elliptic inhibitory area around
inducing agent. Production of β-lactamase in each strip. The MIC is determined by where the
staphylococci is inducible, and exposure to an growth ellipse intersects the Etest strip (Fig. 13.18).
inducing β-lactam agent is often required to obtain a Like the disk diffusion test, the Etest has the
high enough concentration of the enzyme for detec-
flexibility of drug selection and testing because
tion with conventional β-lactamase tests. Studies
have indi- cated that the nitrocefin test has low selected strips are applied to the surface of test
sensitivity for detecting β-lactamase activity in S. plates. The cost of Etest strips is much more than
aureus. Testing organisms that were exposed to an that of antimicrobial disks, however, and is the main
inducing agent can be accomplished by using growth limitation of this product. Generally, published
from the periphery of a zone surrounding a β-lactam studies have indicated that MICs determined using
disk. Taking the inoculum directly from the zone the Etest compare favorably (within one twofold
edge of a cefoxitin disk (30 pg) from a standard disk dilution interval) with those determined by
diffusion test can increase the sensitivity. This is conventional broth or agar dilution methods.
particularly true for the coagu- lase-negative The Etest can be especially useful for testing
Staphylococcus spp. fastidious organisms such as S. pneumoniae, other
Studies showed that the penicillin disk zone streptococci, H. in}u- enzae, and anaerobic bacteria,
edge test is the most sensitive method, and the in part because the strips can be placed on enriched
CLSI recommends this test for determining β- media or in a special incubation atmo- sphere (e.g.,
lactamase production in S. aureus. For the CLSI increased CO2 or anaerobic) and because rela- tively
protocol, a 10-unit penicillin disk is used in a few antimicrobial agents may need to be tested
standard disk diffusion test. The breakpoint for against fastidious organisms. Consequently, the
resistance is a zone <29 mm. Isolates with a relatively high cost of the Etest strips could be
sharp zone minimized.

Fig. 13.17 Cafinasa β-lactamasa disk tast. Calls frOm savaral cOlOnias Of rasults aftar tasting Of twO diffarant isOlatas. Within 10 minutas, tha disk On tha left turnad
HaemOphilus in.uenzae wara appliad tO a mOistanad disk. This phOtO shOws brOwn-rad (pOsitiva), and that On tha 2ight maintainad a light yallOw cOlOr (nagativa).
Fig. 13.18 EsChe2iChia COli tastad with an Etast gantamicin strip. Tha
gantamicin minimal inhibitOry cOncantratiOn (whara tha allipsa crOssas
tha gradiant) is
0.75 µg/mL.
 AutOmated antimicrObial SuSceptibility teSt methOdS293
reader that allows the labora- tory scientist to record
the results using customizable light- ing options to
produce an easy-to-read digital plate image.

AutOmated antimicrObial susceptibility

test methOds
Principles Of technOlOgies used
The automated susceptibility testing instruments
currently available represent a choice of several
different levels of automation. One instrument
interprets growth end points of broth microdilution
panels only when they are placed into an automated
reader device, whereas other instruments provide
hands-off incubation and reading functions for
microdilu- tion trays or special cards in an incubator-
reader device. The instruments that offer the highest
level of automation accom- plish these tasks using
robotics to move the panels or cards in the
instrument during the incubation-reading sequences
or to add reagents to certain test wells for
biochemical tests.
Current instruments use one of two optical
approaches for examining the test wells of the
antimicrobial-containing panels or cards. Most
instruments use the principle of tur- bidimetric
detection of bacterial growth in a broth medium in
test wells by a photometer. The determination of
antimi- crobial susceptibility is based on the lack
of development of turbidity (growth suppression) or,
conversely, an indi- cation of resistance based on an
increase in turbidity in the presence of an
antimicrobial agent. The second means of
recognizing growth is detecting hydrolysis of a
fluorogenic growth substrate incorporated in a
special test medium. With this method, growth is
detected by a fluorometer as emission of a
fluorescent signal when a microorganism consumes
fluorophore-labeled substrate in the test medium
during growth.
Instruments for antimicrobial susceptibility testing
can provide test results after a conventional
overnight incuba- tion period or in a period of 5 to 15
hours. Instrumentation can interpret antimicrobial
susceptibility test end points sooner than manual
readings because of the greater sensi- tivity of the
instruments’ optical systems in detecting subtle
increases in microbial growth. All of the instruments
utilize computers to provide reports and to store and
retrieve data on antimicrobial susceptibility. Most of
the instruments can also be used to identify bacteria
and to merge and print iden- tification and
antimicrobial susceptibility results into a single
report and transfer the information to a laboratory
informa- tion system (LIS).

AutOmated systems
Reader devices fOr brOth micrOdilutiOn
susceptibility tests
Commercial manufacturers of broth microdilution
antimi- crobial susceptibility testing panels offer a
view box or other device to facilitate the manual
reading of results after incu- bation. They offer
freeze-dried antimicrobial panels with a mechanized
device to simplify hydration and inoculation of
freeze-dried panels. Some manufacturers provide
frozen panels. The Sensititre Vizion (Thermo Fisher
Scientific) offers an instrument-assisted manual
A similar instrument is the MicroScan TouchScan
(Beckman Coulter Pasadena, CA). The MicroScan
AutoSCAN-4 is a semi-automated system. After trays
are incubated offline, the laboratory scientist inserts
a tray into the reader that inter- prets growth
patterns in panels by turbidimetric analysis.

AutOmated instrument systems

Currently, four instruments are available that can
generate rapid (5 to 15 hours) or overnight (16 to 24
hours) susceptibil- ity test results.
BD PhOenix system
The BD Phoenix automated microbiology system (BD
Diagnostic Systems, Sparks, MD) is marketed in
Europe, Japan, Canada, and the United States. The
original Phoenix system consists of an instrument
with a built-in keyboard and bar-coding station that
can accommodate 100 test pan- els simultaneously,
along with a separate printer (Fig. 13.19). The
specially designed test cartridges contain 136 small
wells (Fig. 13.20) to test as many as 25 different
antimicro- bial agents, alone or in combination with
wells containing biochemical substrates for
simultaneous identification of common gram-
positive or gram-negative bacteria (combo panels).
The test panels are inoculated manually by a simple,
gravity-fed transfer of inoculated medium
throughout the disposable cartridge after pouring
inoculated broth through an opening in the test
device. After inoculation, the panels are placed into
the instrument’s incubator. The Phoenix system

Fig. 13.19 BD PhOanix autOmatad micrObiOlOgy systam. (COu%tesy BD DiagnOstiC

Systems, Spa%ks, MD.)
294 PART 1 13 AntimicrObial SuSceptibility teSting

checks the panels for growth in the test wells every

20 min- utes using a redox indicator system, then placed in the incubator module. The type of test
resembling resazurin, and turbidity. The indicator is to be performed is indicated on an instrument-
added to the broth at the time of organism readable bar-code label located on the end of each
inoculation. This approach allows susceptibility panel. The instrument incu- bates the panels for the
determinations after approximately 6 to 8 hours of appropriate period (depending on the type of panel
incuba- tion. The Phoenix instrument includes a and organism), robotically positions the trays to add
rules-based expert system (BDXpert) and can be reagents for biochemical tests, and moves them
purchased with a data man- agement personal under the photometer to perform growth readings.
computer and software package called BD EpiCenter, Standard panels are read turbidimetrically after
which allows networking with other BD microbiol- ogy 16 to 18 hours of incubation with 24-hour holds for
instruments and a bidirectional interface for specific bac- teria (e.g., oxacillin for Staphylococcus
connection with an LIS. and vancomycin for Enterococcus). The instrument
offers the possibility of early, read-when-ready
MicrOScan WalkAway system interpretations of the standard panels in 4.5 to 6.5
hours if an isolate is unequivocally susceptible or
The first MicroScan WalkAway system (Beckman highly resistant. Organisms with inducible or slow to
Coulter) was developed in the 1980s. The current be expressed resistance are automatically extended
fourth- generation system, the WalkAway plus, to 16- or 18-hour incu- bation. MIC or breakpoint
consists of a self- contained incubator-reader unit panels are available for testing of gram-positive and
with capacity for 40 or 96, 96-well panels, LabPro gram-negative bacteria as are special com- bination
information management software, a printer, and an panels that allow simultaneous susceptibility and
LIS interface (Fig. 13.21). The WalkAway uses organism identification in the same panel. With the
microdilution panels that are hydrated and standard panels, it is possible to read the panels
inoculated with a hand-operated inoculator device visually in case of instrument malfunction. Panels
(MicroScan Renok, Siemens Healthcare Diagnostics, containing lysed horse blood (MICroStrep, Beckman
Tarrytown, NY; Fig. 13.22). Panels are Coulter) are also provided for testing

Fig. 13.22 MicrOScan RanOk inOculatOr davica, which is usad tO racOnstituta

Fig. 13.20 BD PhOanix tast cartridgas, which cOntain 136 walls Of and inOculata driad antimicrObial agants Or biOchamical tasts in tha tast panals.
antimicrObial agants Or biOchamical tasts. A tast panal is alsO shOwn. (COu%tesy Siemens HealthCa%e DiagnOstiCs, Ta%
%ytOwn, NY.)

Fig. 13.21 MicrOScan WalkAway SI. (COu%tesy Siemens HealthCa%e DiagnOstiCs, Ta%%ytOwn, NY.)
 Quality cOntrOl Of antimicrObial SuSceptibility teStS 295

streptococci that can be read manually or by an the nonfastidious, aerobic gram-positive and gram-
automated system. The instrument includes a rules- negative bacteria. The VITEK 2 offers one of the
based expert sys- tem (LabPro Alert EX) and can be highest levels of auto- mation currently available in
purchased with a software package called LabPro microbiology instrumentation.
that allows networking with other MicroScan VITEK hardware consists of a filling module for
analyzers and workstations and a bidirectional inocula- tion of the cards, an incubator-reader
module that incorpo- rates a carousel to hold the test
interface with an LIS.
cards, a robotic system to manipulate the cards, a
Sensititre ARIS 2X photometer for turbidimetric mea- surement of
growth every 15 minutes, and a computer mod- ule
The Sensititre ARIS 2X (Thermo Fisher Scientific) with a video display terminal and printer for viewing
automated incubator reader was the first system to and printing results. VITEK also offers a data
be marketed in the United States (in 1992) that used management sys- tem (Observa) for storing and
a fluorometric detection system for detecting growth retrieving test data for a variety of epidemiologic
end points of common, rapidly growing bacteria. It reports.
was found that the fluorogenic substrate system The VITEK 2 Compact is a less automated version
could not accurately detect some resistance mecha- of the VITEK 2 that uses the same cards but without
nisms in 5 hours. Therefore the system has FDA the smart carrier and programmed cassette for
approval for only 16- to 24-hour incubation for specimen handling automation available in the VITEK
susceptibility testing. 2. It includes a compact, self-contained vacuum
The Sensititre ARIS 2X (Fig. 13.23) is a fully chamber and card sealer. The instrument is a less
automated, benchtop incubating and reading expensive option for laboratories that do not require
system. The system utilizes up to 64, 96-well trays. the level of automation provided by the smart carrier
The AutoReader uses an internal bar-code scanner of the VITEK 2. All VITEK instruments use kinetic
to identify each plate type and assign the measurements of growth in the presence of
appropriate incubation time; when the assigned time antimicrobial agents to provide analysis of growth
has elapsed, the plate is then transported to the curves, leading to computer algorithm–derived MICs.
OptiReader for fluorescence measurement, with no All instruments include a rules-based expert system,
manual intervention. Panels for M. tuberculosis are Advanced Expert System, to assist in controlling
also available. common technical errors and facilitate detection of
unusual resistance mechanisms before results are
VITEK systems reported.
The VITEK systems (bioMérieux) were originally
designed for use in the U.S. space exploration efforts
of the 1970s as an onboard test system for
spacecraft exploring other planets for life. Because of
Quality cOntrOl Of antimicrObial
its original design intention, it was highly automated susceptibility tests
and relatively compact. Small plastic reagent cards,
similar in size to a credit card, contain microliter
QC of antimicrobial susceptibility tests involves
quantities of various concentrations of antimicrobial
testing refer- ence strains with defined antimicrobial
agents in 64 wells for susceptibility and identification
susceptibility or resis- tance to the drugs tested. It is
testing. The VITEK 2 (Fig. 13.24) can accommodate
important to use QC strains that represent the types
up to 60 cards, while the VITEK 2XL holds up to 120
of patient isolates tested in the respec- tive
cards. The newest VITEK 2 Compact for small
laboratory. Also, the QC strains should represent
laboratories is available in three sizes: 15, 30, or
various degrees of susceptibility or resistance. Ideal
60 cards. The susceptibility cards allow quantita- tive
QC strains for MIC tests have what are termed on-
MIC results accompanied by susceptible,
scale MIC end points for the drugs tested. An on-
intermediate, or resistant results interpretations for
scale end point falls within the range of
most rapidly growing, gram-positive and gram-
concentrations tested compared with off-scale end
negative aerobic bacteria in 4 to 18 hours. The
points, in which the MIC is less than the lowest or
VITEK 2 can test S. pneumoniae in addition to
greater than the highest concentration tested.
The CLSI has identified American Type Culture
Collection (ATCC) strains useful for QC testing (Table
13.6). The CLSI documents also include tables that
define acceptable results
Fig. 13.23 Sansititra ARIS 2X. (COu%tesy TREK DiagnOstiC Systems, Cleveland, 0H.) Fig. 13.24 VITEK 2 Systam. (COu%tesy biOMé%ieux Vitek, Du%ham, NC.)
296 PART 1 13 AntimicrObial SuSceptibility teSting

Table 13.6 Commonly used strains for quality control of routine antimicrobial susceptibility tests
Test QC strains used Comments
Antimicrobial susceptibility of gram-positive
Staphylococcus aureus ATCC 25923 β-Lactamase negative for disk diffusion tests
organisms
S. aureus ATCC 29213 β-Lactamase positive for MIC tests
Oxacillin salt agar screen for S. aureus S. aureus ATCC 43300 Oxacillin-resistant
Vancomycin BHI screen, synergy screen Enterococcus faecalis ATCC 29212 Susceptible to vancomycin and to high levels of
for enterococci and S. aureus
gentamicin and streptomycin (synergy screen
tests negative)
E. faecalis ATCC 51299 Resistant to vancomycin and to high levels
of gentamicin and streptomycin (synergy
screen tests positive)
Antimicrobial susceptibility of gram-negative Escherichia coli ATCC 25922
organisms
Pseudomonas aeruginosa ATCC 27853
E. coli ATCC 35218 β-Lactamase positive for testing
β-Lactam–β-lactamase inhibitor
combination agents only
ESBL test Klebsiella pneumoniae ATCC 700603 ESBL screen test and ESBL confirmatory
test positive
Antimicrobial susceptibility of Haemophilus Haemophilus influenzae ATCC 49247 Ampicillin-resistant, non–β-lactamase producing
spp.
H. influenzae ATCC 49766 Ampicillin-susceptible
H. influenzae ATCC 10211 Used by media manufacturers to assess
growth-supporting capabilities of the
Antimicrobial susceptibility of Neisseria medium
gonorrhoeae N. gonorrhoeae ATCC 49226 β-Lactamase negative
Antimicrobial susceptibility of S.
pneumoniae
Streptococcus pneumoniae ATCC 49619 Penicillin-intermediate
and other streptococci
Antimicrobial susceptibility of anaerobes Bacteroides fragilis ATCC 25285
Bacteroides thetaiotaomicron ATCC
29741
Eubacterium lentum ATCC 43055
Assessment of acceptability of
E. faecalis ATCC 29212
medium (low thymine and thymidine
content) for testing sulfonamides,
trimethoprim, and trimethoprim-
sulfamethoxazole
ATCC, American Type Culture Collection; BHI, brain-heart infusion; ESBL, extended-spectrum β-lactamase; MIC, minimal inhibitory concentration; QC, quality control.

(zone measurements for disk diffusion tests) for antimicrobial agent–organism combinations that may be
these strains; examples are shown in Table 13.7. The only modestly controlled with routine reference strains.
procedure followed in testing QC reference strains For example, a
must be identical to that used for testing patient
isolates. If results with QC strains do not fall within
the defined acceptable limits, corrective action must
be taken to determine the reason for the out-of-
control observa- tion before reporting any patient
results. QC testing is recom- mended each day that
patient tests are performed; however, the frequency
of QC testing can be reduced to weekly if a lab-
oratory can demonstrate acceptable performance
with the QC strains. This consists of obtaining results
within acceptable limits for each antimicrobial agent–
QC strain combination for 20 or 30 consecutive test
days. QC procedures must be per- formed when new
lots of materials are put into use, and test materials
must never be used beyond their stated expiration
dates.
Supplemental QC strains may be periodically
tested to validate acceptable performance of specific
MRSA strain could be included to ensure that
the test system can detect heteroresistant
strains. Similarly, ampicillin-resis- tant
Enterobacter cloacae might be included to
ensure that the system can detect ampicillin
resistance in gram-negative bac- teria.
Supplemental QC strains are sometimes used
for trou- bleshooting specific problems or
training new employees. Another component of
a QC program involves the inclusion of
mechanisms to ensure that those performing
the testing are proficient in their tasks. Self-
assessment checklists and supervisory reviews
of reported results are examples of such
mechanisms. Satisfactory performance on
proficiency survey specimens and the use of
relevant testing strategies are also QC
parameters.
The most widely used supplemental QC
measure is antibiograms to verify results
generated on patient iso- lates. An
antibiogram is the antimicrobial susceptibility
profile of a bacterial isolate to a battery of
antimicrobial agents. Some species have
typical antibiograms, which can be used to
verify the identification and suscepti- bility
results on the isolate (Table 13.8). For
example,
 Quality cOntrOl Of antimicrObial SuSceptibility teStS 297

Table 13.7 Acceptable limits for quality control strains used to monitor accuracy of disk diffusion testing of nonfastidious organismsa
Antimicrobial Disk content Escherichia coli Staphylococcus aureus Pseudomonas aerugi- Escherichia coli
agent (µg) ATCC 25922 (mm) ATCC 25923 (mm) nosa ATCC 27853 (mm) ATCC 35218 (mm)
Ampicillin 10 15–22 27–35 — 6
Amoxicillin– 20–10 18–24 28–36 — 17–22
clavulanic acid
Cefazolin 30 21–27 29–35 — —
Gentamicin 10 19–26 19–27 17–23 —
a
Using Mueller-Hinton medium without blood or other supplements.
ATCC, American Type Culture Collection.
Reprinted with permission from Clinical and Laboratory Standards Institute. (2021). Performance standards for antimicrobial susceptibility testing (31st ed.).
CLSI supplement M100. Clinical and Laboratory Standards Institute.

Table 13.8 Typical antibiograms for several gram-negative speciesa

Antimicrobial agent Escherichia coli Enterobacter Proteus Pseudomonas Stenotrophomona
cloacae mirabilis aeruginosa s
maltophilia
Amikacin S S S S R
Ampicillin S R S R R
Ampicillin-sulbactam S R S R R
Cefazolin S R S R R
Cefoxitin S S-R S R R
Cefotaxime S S-R S S-R R
Ceftazidime S S S S S-R
Ciprofloxacin S S S S R
Gentamicin S S S S R
Imipenem S S S S R
Piperacillin S S S S R
Nitrofurantoin S S R R R
Tobramycin S S S S R
Trimethoprim- S S S R S
sulfamethoxaz
ole
a
Results indicated represent the typical response found in most clinical isolates; however, these can vary significantly.
R, Resistant; S, susceptible; S-R, variable result.

P. aeruginosa is typically resistant to ampicillin,

cefazolin and other first- and second-generation • Reexamination of the disk diffusion plate, MIC
cephalosporins, and trimethoprim-sulfamethoxazole; tray, and other components to ensure that results
however, it is often susceptible to gentamicin and were properly interpreted and that the materials
other aminoglycosides, extended-spectrum were not overtly defec- tive (e.g., empty well in
penicillins (e.g., piperacillin), and ciprofloxacin. In the tray)
contrast, E. coli is generally susceptible to all of • Checking earlier reports to see whether the
the antimicrobial agents mentioned. Table 13.9 particular patient previously had an isolate with
shows several atypical antibiograms suggesting that an atypical antibio- gram that was verified
the result highlighted is erroneous. The CLSI • Repeating the test, if necessary. Sometimes it is
provides a list- ing of results that should be verified necessary to repeat the identification and
because they (1) have never been encountered, (2) antimicrobial susceptibil- ity tests to verify the
are uncommon, or (3) repre- sent results that could atypical results; sometimes testing with an
easily occur because of technical errors and might alternative method is useful.
have significant clinical consequences. One list
includes susceptibility test results for certain gen- With emerging resistance and healthcare-
era or species that should be verified by all associated trans- mission of resistant organisms,
laboratories, and another includes results that need there is now more variability in susceptibility profiles
not be verified in facilities in which the results are among clinical isolates than previ- ously seen. In
common. Some of these listings are shown in Table contrast with what was seen a decade ago, it would
13.10 not be uncommon now for a facility to see a high
When atypical antibiograms are seen, the results per- centage of isolates resistant to multiple
must be verified. Verification procedures include: antimicrobial agents.
 Cumulative antibiograms are generated by the
analysis of individual susceptibility results obtained
on isolates from a
298 PART 1 13 AntimicrObial SuSceptibility teSting

Table 13.9 “Problem” antibiograms suggestive of technical errors or patient issue

Antimicrobial agent Escherichia colia Enterobacter cloacaeb Pseudomonas aeruginosac Stenotrophomonas
maltophiliad
Amikacin R S S R
Ampicillin S R S R
Cefazolin S S R R
Cefoxitin S R R R
Cefotaxime S R S-R R
Gentamicin S S S R
Tobramycin S S S R
Trimethoprim- S S R R
sulfamethoxazole
a
lt is very unusual for an isolate to be resistant to amikacin and susceptible to gentamicin and tobramycin because amikacin is typically the most active of these three
aminoglycosides.
b
Third-generation cephalosporins (e.g., cefotaxime) are usually more active than second-generation cephalosporins (e.g., cefoxitin), which in turn are more active than
first-generation cephalosporins (e.g., cefazolin) against the Enterobacterales. ln addition, E. cloacae is typically resistant to cefazolin. Consequently, this antibiogram
is unusual.
c
Ampicillin does not produce activity against P. aeruginosa, and this antibiogram is unusual.
d
S. maltophilia is usually susceptible to trimethoprim-sulfamethoxazole, which is the drug of choice for infections caused by this very resistant species. This antibiogram is
unusual.
R, Resistant; S, susceptible; S-R, variable result.

Table 13.10 Suggestions for verification of antimicrobial suscep- among the patients in that facility. An investigation
tibility test results and confirmation of organism identification by infec- tion control personnel into the reason
behind such observa- tions would be warranted.
Organism or Category I: verify at all Category II:
verify; group laboratories institution
specific
Klebsiella spp. Ampicillin, S
Automatically reported
ESBL confirmed
positive
Selecting an antimicrObial
R regardless of result
Some instruments susceptibility test methOd
in some laboratories
can reliably do this
testing (i.e., Phoenix) Clinical microbiology laboratories can choose from several
Streptococcus Fluoroquinolone, R Penicillin, R manual or instrument-based methods to perform
pneumoniae their rou- tine antimicrobial susceptibility testing—
Linezolid, NS Third-generation the disk diffusion (or Kirby-Bauer) test, broth
cephalosporin, microdilution (with or without use of an instrument
Vancomycin, NS
R for panel readings), and rapid auto-
ESBL, Extended-spectrum β-lactamase; NS, not susceptible; R, resistant; S, infection control purposes. An increase in the incidence of
susceptible. MRSA from 25% in the first quar- ter of the year to 60%
Reprinted with permission from Clinical and Laboratory Standards lnstitute. (2021). in the second quarter might suggest a problem with
Performance standards for antimicrobial susceptibility testing (31st
ed.). CLSl supplement M100. Clinical and Laboratory Standards healthcare-associated transmission of MRSA
lnstitute.

particular institution in a defined period; this

represents the percentage of isolates of a given
species susceptible to the antimicrobial agents
commonly tested against that species (e.g., the
percentage of E. coli isolates that are susceptible to
ampicillin). Cumulative antibiograms are generally
compiled annually to guide physicians in empiric
therapy decisions. If a physician were treating a
patient with a suspected infection caused by P.
aeruginosa and the culture and susceptibility test
results were not yet available, the physician could
review the cumulative antibiogram to see the
percentage of P. aeruginosa isolates in that facility
susceptible to various antipseudomo- nal agents.
This information could assist the physician in
designing the empiric therapy regimen pending
completion of culture and susceptibility testing.
Cumulative antibiogram data also might be used for
mated instrument methods. The Etest also can
be useful for certain fastidious or obligate
anaerobic bacteria.
The disk diffusion test provides the greatest
flexibility and cost-effectiveness. A problem
mentioned frequently with commercial
microdilution or automated systems is the
inflex- ibility of the standard antimicrobial
agent batteries or test panels. With the current
availability of more than 50 antimi- crobial
agents in the United States and the diversity
among antimicrobial agent formularies in
different hospitals, it is impossible for
manufacturers to provide standard test pan- els
that fit every hospital’s needs. Thus the
inherent flexibility of the disk diffusion test
allows a laboratory to test any of 12
antimicrobial agents considered appropriate on
a 150-mm Mueller-Hinton agar plate.
Other assets of the disk diffusion procedure
are that it is one of the longest practiced
standardized methods, and its performance is
continually updated by the CLSI. The inter-
pretive category results—nonsusceptible,
susceptible, inter- mediate, resistant—of the
disk diffusion test should be readily
understandable by all physicians, which is
frequently not the case with MIC results.
Therefore it is useful to provide the interpretive
category results along with the MIC.
As noted, commercial microdilution
susceptibility test products are widespread in
U.S. clinical laboratories.
 Special antimicrObial SuSceptibility teStS 299
been more successful than the rapid determina- tion
of the bacterium’s antimicrobial susceptibility.
However, the detection of drug-resistant genes has
Advantages of this method include its quantitative demonstrated prom- ise to increase the speed of
nature (an MIC rather than a category result), ability determining a bacterium’s resis- tance to
to determine MICs with some organisms for which antimicrobials. These rapid tests allow the treatment
the disk test may not be stan- dardized, and the
availability of automated panel readers. In addition,
the computerized data management systems that
accompany some of the instruments can be helpful
to some laboratories for the storage and calculation
of cumulative antibiograms and other susceptibility
statistics.
A laboratory can perform automated antimicrobial
sus- ceptibility testing to reduce the amount of labor
required for susceptibility tests and to generate test
results more rapidly than can be accomplished by
manual methods. Providing antimicrobial
susceptibility tests a day or even hours sooner is a
great advantage in patient care. Studies have docu-
mented a decrease in patient morbidity or mortality,
as well as demonstrated substantial cost savings. If
physicians are aware of critical susceptibility data on
patients’ isolates, an opportunity exists for improving
the quality of care, quicker discharge of patients, or
removal of patients from isolation or unnecessary
and expensive empiric treatment (i.e., vancomy-
cin), thus realizing cost savings.
Some time savings can be achieved with
automated sus- ceptibility instruments by using
combination panels that allow antimicrobial
susceptibility testing and organism identification on
the same panel. In addition, some instru- ments
provide mechanized or simplified panel inoculation
and reading to achieve labor savings. The magnitude
of the labor savings realized in clinical chemistry or
hematology through automation has yet to be fully
achieved in clinical microbiology. If a serious
mechanical failure occurs, only tests from
instruments that use conventional microdilution
trays normally interpreted after overnight incubation
can be com- pleted by manual incubation and
interpretation. The results obtained with instrument
methods that incorporate rapid test interpretation or
use fluorogenic substrate analysis cannot be
manually interpreted. Therefore laboratories must
have a backup testing procedure, such as disk
diffusion or manual overnight broth microdilution
testing, to avoid delays in gen- erating patients’
results.
A previous shortcoming of rapid susceptibility
test- ing methods was the inability to detect
certain inducible or otherwise subtle antimicrobial
resistance mechanisms. Manufacturers have made
significant strides toward solving these problems. In
some cases, this has meant reformulating the test
devices or improving the software used to interpret
the susceptibility testing results, which can be aided
in part by computer expert systems that detect
common technical errors and explain some newer
resistance mechanisms. Nevertheless, it is important
for microbiologists to scrutinize carefully all
susceptibility results before issuing final patient
reports.

Rapid susceptibility determinatiOn

The successful treatment of infectious diseases
requires not only the identification of the causative
bacterium, but also determination of the bacterium’s
susceptibility to antimicro- bials. Over the past
several years, the rapid identification of bacteria has
 blood culture bottles submitted from a patient with
suspected bacterial endocar- ditis. The isolate was
susceptible to penicillin (MIC ≤0.03 pg/mL), but the
of bacterial infections much sooner. Molecular physician was concerned about the bactericidal activity
methods for the detection of antimicrobial of penicillin alone. The laboratory performed MBC
resistance genes are limited to those organism- testing and reported an MBC of 32 pg/mL. The physician
antimicrobial combinations in which only a few considered this
genes are associated with the resistance, and the
resis- tance has a high degree of clinical
significance.
Currently, about eight systems have received FDA
approval for nucleic acid amplification testing. They
are based on two methodologies—PCR and solid-
phase microar- rays—and can be performed directly
on clinical specimens or culture isolates. For
example, the Verigene BC-GP (Luminex Corporation,
Austin, TX) can identify several species of gram-
positive bacteria and detect mecA resistance gene
in
S. aureus and Staphylococcus epidermidis and vanA/B
genes in
E. faecalis and E. faecium in a few hours directly
from positive blood culture bottles. This assay is
based on hybridization of DNA target sequences to
nanoparticles and is classified as moderate
complexity by CLIA. The BioFire FilmArray
(bioMérieux) and GeneXpert (Cepheid, Sunnyvale,
CA) are PCR-based instruments also classified as
moderate complex- ity. Other systems are classified
as high complexity. ESBL production by gram-
negative bacteria is one of the greatest threats to
antimicrobial therapy today. Several commercial
assays for the molecular detection of ESBL exist.
Some, like the FilmArray, have a single target (KPC),
while others detect a variety of β-lactamases (e.g.,
Verigene). Probes directed toward genes responsible
for other resistance mechanisms (e.g.,
aminoglycoside-modifying enzymes and tetracycline
resistance factors) have also been described.

Special antimicrObial susceptibility

tests
Clinical microbiology laboratories that perform
bacterial cul- tures generally perform antimicrobial
susceptibility tests by a disk diffusion or MIC
method. These tests assess the ability of
antimicrobial agents to inhibit the growth of bacteria
and are sufficient for guiding antimicrobial therapy in
most situations. In select cases, a physician may
wish to determine how well specific antimicrobial
agents kill bacteria or how the activity of a
combination of antimicrobial agents compares with
that of a single agent. Unlike disk diffusion and MIC
tests, how- ever, tests for bactericidal activity (e.g.,
MBC tests and serum bactericidal tests [SBTs]) and
for antimicrobial synergism are not highly
standardized. When patients are prescribed anti-
microbial agents that have a low toxic-to-
therapeutic ratio, it may be necessary to monitor the
concentrations of drug in the patient’s serum to
avoid excessive concentrations that can be harmful.
Various methods are used for these assays. Some
antimicrobial susceptibility tests are generally
performed only in specialized laboratories and are
used in only a few defined clinical settings or for
research purposes (Table 13.11).

Case in pOint
Viridans group Streptococcus was isolated from all six
300 PART 1 13 AntimicrObial SuSceptibility teSting

information in conjunction with clinical observations of the

patient and documentation in the literature of previous infection site is important for achieving a cure. For
experiences with patients with similar conditions. This many types of infections, the bactericidal capacity of
patient continued with a therapeutic regimen of a specific anti- microbial regimen can be predicted
penicillin and gentamicin. based on previous expe- rience. For example, most
β-lactam antimicrobial agents are bactericidal for E.
coli, provided their MIC is in the susceptible range.
Issues to consider On the other hand, the bactericidal activity of β-
After reading t,e patient’s case ,istory, consider: lactams and other cell wall–active agents (e.g.,
• Circumstances when it might be appropriate to do vancomycin) against
more than a disk diffusion or MIC test on a clinical S. aureus is less predictable. In circumstances in
isolate which a seri- ous S. aureus infection occurs in a
• Factors that might contribute to the MBC being five patient with poor immune defense mechanisms, an
twofold dilutions greater than the MIC in vitro determination of the amount of antimicrobial
• Rationale for adding a second agent (gentamicin) to agent required to kill the isolate may be help- ful.
treat the infection The MBC test is used for this purpose. The MBC is
defined as the lowest concentration of antimicrobial
agent that kills 99.9% of the test bacteria. The CLSI
has described procedures for assessing bactericidal
activity; however, unlike disk diffu- sion and MIC
Minimum bactericidal tests, a standardized MBC test has not been in use
cOncentratiOn test for a sustained period. In the past, numerous
procedural variations compromised the use of the
test results.
MIC tests identify the amount of antimicrobial agent The MBC test is performed in conjunction with a
required to inhibit the growth or multiplication of a broth mac- rodilution or broth microdilution MIC test.
bacterial isolate. If a concentration of antimicrobial The antimicrobial agent concentrations that show
agent that is the same as or preferably exceeds the inhibition (equal to or greater than the MIC) may or
MIC is attained at the infection site for an may not have killed the bacteria in the test inoculum
appropriate length of time, the drug generally (Fig. 13.25). After the MIC determination, a 0.01-mL
inhibits multiplication of the bacteria such that the aliquot from each clear tube or well is subcultured
patient’s immune defense mechanisms can onto an agar medium to determine the MBC. The
successfully eliminate the infecting bacteria. The numbers of colonies that grow on subculture are
immune defense mechanisms (e.g., phagocytic cells, compared with the actual number of organisms
antibody) work together with antimicrobial agents to inoculated into the MIC test tubes or wells to deter-
eradicate infecting bacteria. For this reason, mine the extent of bactericidal activity at each
inhibitory con- centrations of the drug at the antimicrobial con- centration. If the numbers of
infection site are generally suffi- cient for treating colonies on a subculture plate total less than 0.1% of
most infections. the initial inoculum (indicating 99.9% or more killing),
In immunosuppressed patients and patients with by definition, a bactericidal effect has been achieved.
serious infections such as endocarditis, meningitis, The number of bacteria in each tube (or well)
and osteomyeli- tis, immune defense mechanisms immediately after inoculation of the MIC test tube or
are suboptimal. Inhibitory concentrations of the drug well is approximately
may not be sufficient, and obtain- ing bactericidal
concentrations of antimicrobial agents at the

Table 13.11 Special antimicrobial susceptibility tests

Test Purpose
Antimicrobial concen-
Measure of amount of antimicrobial
tration test (assay)
agent in serum or body fluid
Minimum bactericidal
Measure of lowest concentration of
concentration test
antimicrobial agent that kills 99.9% of a
bacterial isolate population
Serum bactericidal
Measure of highest dilution or titer of a
test
patient's serum that is inhibitory to the
patient's own infecting bacterium and
highest dilution or titer that is
bactericidal
Synergy test Measure of susceptibility of a Reprinted with permission from Clinical and Laboratory Standards Institute. (2021).
bacterial isolate to a combination of Performance standards for antimicrobial susceptibility testing (31st ed.). CLSI
supplement M100. Clinical and Laboratory Standards Institute.
two or more antimicrobial agents
Time-kill assay Measure of rate of killing of bacteria by
an antimicrobial agent (as determined by
examining the number of viable bacteria
remaining at various intervals after expo-
sure to the agent)
 cOlOny cOunt plata, immadiataly aftar
Fig. 13.25 BrOth macrOdilutiOn tast shOwing vancOmycin and inOculatiOn, tha grOwth cOntrOl tuba was dilutad 1:1000, and 0.1 mL was platad.
StaphylOCOCCus au-eus. Tha minimal inhibitOry cOncantratiOn NOw 0.01 mL will ba platad frOm aach claar tuba (tubas cOntaining 1.0–128
µg/mL vancOmycin) fOr tha minimum bactaricidal cOncantratiOn datarminatiOn.
(MIC) is 1.0 µg/mL. Tha purity plata
Bacausa it was shOwn that tha actual cOlOny cOunt in tha MIC tast was 5.3 x 105
shOws a pura cultura. Tha cOlOny cOunt plata shOws 53 cOlOnias, which maans that CFU/mL, grOwth Of fiva Or fawar cOlOnias wOuld indicata a 3 lOg10 dacraasa, Or
5.3 x 105 cOlOny-fOrming units (CFU) par millilitar wara in tha tast
tubas immadiataly aftar inOculatiOn Of tha MIC tast tubas. FOr tha 99.9% killing. By dafini-
tiOn, tha cOncantratiOn Of drug in tha raspactiva tuba wOuld ba cOnsidarad bactaricidal.
 Time-kill aSSayS 301
vancOmycin and StaphylOCOCCus au-eus shOwn in Fig. 13.25. Subculturas frOm
tubas cOntaining 8.0 tO 128 µg/mL vancOmycin shOw nO grOwth. Subculturas
frOm tha tubas cOntaining 2.0 and 4.0 µg/mL shOw Ona and twO cOlOnias,
raspactivaly, indicating a graatar than 99.9% killing. MOra than fiva cOlOnias hava
5 x 105 CFU/mL. For the MBC test, however, an grOwn frOm tha
actual col- ony count must be performed on the test 1.0 µg/mL tuba, sO tha minimum bactaricidal cOncantratiOn is 2.0 µg/mL.
inoculum when the MIC test tube or well is
inoculated. A small aliquot from the growth control
tube or well is diluted in saline or broth to obtain a
countable number of colonies; the final dilution is
plated on an agar medium. Generally, a 1:1000
dilution is performed (0.01 mL from the growth
control tube is diluted in 10 mL), and 0.1 mL is
spread over the surface of an agar plate, resulting in
an overall dilution of 1:10,000. After over-
night incubation, the number of colonies that grew
on the col- ony count plate are counted. Because this
count represents a 1:10,000 dilution, the count is
multiplied by 104 to determine the number of
bacteria in the original growth control and the
antimicrobial agent solutions. The number of
colonies repre- senting a 99.9% reduction in the
original inoculum is deter- mined. This is the MBC
end point.
In the example shown in Fig. 13.25, the actual
count on the colony count plate is 53; therefore the
number of bacteria in each tube of the MIC test
immediately after inoculation was
5.3 x 105 CFU/mL (53 times the dilution factor 104). A
99.9%
killing (or 0.1% survival) would be accomplished if
five or fewer colonies grow on subculture of each
clear well or tube after reading of the MIC test (Fig.
13.26). In this example, sub- cultures from all tubes
containing 2.0 pg/mL vancomycin or more grew
fewer than five colonies. Consequently, the MBC is
2.0 pg/mL. The 99.9% end point (or 3 log 10 reduction
in growth of the original inoculum) is an arbitrary
value, with 95% confidence limits, although its
clinical relevance has not been rigorously confirmed.
Measurement of MBC is thus a rough prediction of
bacterial elimination and can be used for research
purposes to evaluate new antimicrobial drugs.

Case check 13.4

The MIC of an antimicrobial agent determines the

concentration required to inhibit the growth of a bacterium.
This inhibition can be bacteriostatic or bactericidal. The
concentration that must be achieved in certain clinical
situations (e.g., endocarditis, menin- gitis) must be bactericidal
to affect a successful outcome for the
patient. In this case, an MBC test could improve patient outcomes.

Fig. 13.26 Subcultura platas frOm tha minimal inhibitOry cOncantratiOn tast Of
 standardized procedure for these assays.
COntrOlling test variables
MBC tests are subject to more technical pitfalls
than MIC tests, and several variables must be rigidly
controlled during MBC testing. The first involves
inoculum. Because many anti- microbial agents exert
a bactericidal effect only on growing cells, bacteria in
the midlogarithmic phase of growth must be used as
the inoculum to prevent falsely elevated MBCs. The
inoculum preparation methods described for MIC tests
that use stationary phase growth are unacceptable for
MBC tests. Second, during inoculation for MIC tests,
care must be taken to ensure that all bacteria in the
test inoculum are depos- ited directly into the
antimicrobial solution. If this is not done, bacteria may
stick to the wall of the tube or well above the
meniscus of the antimicrobial solution and can remain
viable during incubation of the MIC portion of the test.
These cells (which have not been exposed to
antimicrobial agent) might then be inadvertently
transferred during the subculture step,
ultimately resulting in falsely elevated MBCs.
Third, the volume subcultured after reading of the
MIC test must be large enough to contain sufficient
inoculum but small enough to prevent carryover of
large amounts of anti- microbial agent that can
continue to exert an antibacterial effect. Usually, 10
pL (0.01 mL) is recommended.

InterpretatiOn cOncerns
Several interpretive problems, most common with β-
lactam agents, have been associated with MBC
tests, and they relate to technical or biological
issues. Sometimes more colonies are growing on
subcultures at higher drug concentrations than at
lower concentrations. This decreased bactericidal
activity at higher concentrations is referred to as a
paradoxical (Eagle) effect. Sometimes small
numbers, but slightly greater than 0.1% of the test
inoculum, of bacteria grow on several subcul-
ture plates. This may occur if some bacteria are
metabolically inactive or dormant (persister cells) at
the times of testing. However, when the persisting
colonies are retested, their MICs are comparable
with those originally obtained. Finally, tolerance to
the intrinsic bactericidal effect of an antimicrobial
agent is demonstrated when the number of colonies
growing on subculture plates exceeds the 0.1%
cutoff for several suc- cessive drug concentrations
above the MIC. Tolerance is gen- erally defined as an
MBC/MIC ratio of 32 or higher. Tolerance has been
associated with a defect in bacterial cellular auto-
lytic enzymes.

Time-kill assays
Bactericidal activity of antimicrobial agents also can
be assessed by performance of in vitro time-kill
assays. These assays examine the dynamics of
synergism or antagonism of antimicrobial agents in
combination, by determining the number of viable
bacteria that remain over time after expo- sure to
each individual drug and drug combinations.
Because they are labor intensive, these assays are
rarely performed in the clinical microbiology
laboratory. They are performed rarely in the case of
patients with meningitis or endocarditis with
treatment failure and in research to study novel
antimi- crobial drugs. CLSI does not have a
302 PART 1 13 AntimicrObial SuSceptibility teSting

Case check 13.5 the agent is therapeutic.

In some cases, a combination of antimicrobial agents (combina-

tion therapy) is required to effect killing of the bacterium to result
in treatment success. An example is the use of a β-lactam drug
and an aminoglycoside to treat serious enterococcal infections. A
synergy test can be performed in the laboratory to confirm that this
therapy would be effective.

Synergy tests
Some types of infections require therapy with a
combination of two or more antimicrobial agents.
Enterococcal endocardi- tis, for example, requires
use of a penicillin (or vancomycin) and an
aminoglycoside for reliable killing of the organism. A
broad-spectrum cephalosporin and aminoglycoside
are often prescribed for gram-negative sepsis in
neutropenic patients. The goals of combination
therapy are to obtain broad-spec- trum coverage,
enhance antibacterial activity through synergistic
interactions, and minimize drug resistance devel-
opment. For most infections requiring combination
therapy, single-agent MIC results and previous
experience in treating similar types of infections are
sufficient to guide the proper selection of an
antimicrobial agent.
Synergy tests are not routinely performed in the
clinical laboratory but may be useful when a patient
is not respond- ing to what would appear to be an
adequate regimen, unusual organisms or resistance
properties are encountered, or host factors preclude
the use of certain agents. Synergy testing for
multidrug-resistant gram-negative bacilli and
extremely drug-resistant mycobacteria is currently
under study.
In vitro synergy tests may be performed using a
broth microdilution checkboard assay or broth
macrodilution method, as described by CLSI. The
checkerboard assay is a type of two-dimensional
test. All steps are performed as for single agents, but
two agents are tested in each well or tube.
Checkerboard synergy tests are usually performed in
broth microdilution MIC trays. A wide variety of
combinations of concentrations are tested by
dispensing drugs in a two-di- mensional
checkerboard format, and each drug tested in the
combination is also tested alone. A combination is
said to show synergism if its antibacterial activity is
significantly greater than that of a single agent; that
is, when the MIC for each drug in the combination is
less than or equal to 25% of the single-agent MIC.
Conversely, antagonism is defined as the activity of
the combination less than (and MICs are greater
than) that of the single agents. For indifference, the
activity of the combination is equal to that of the
single agents (Fig. 13.27).

Case check 13.6

Determining that the concentration of an antimicrobial agent in the

blood (serum, plasma) is sufficient to kill the bacterium causing
the infection is sometimes needed to confirm that the dosing of
Serum bactericidal test Bi0l0gical assays
In the late 1940s, Schlichter and MacLean Antimicrobial assays were initially performed by a
described a test that measured the biological assay (bioassay) method; bioassays are
effectiveness with which penicillin in serum still sometimes used
killed bacteria associated with subacute
bacterial endocar- ditis (SBE). This test was
subsequently referred to as the serum
bactericidal test (SBT). There is no single
standardized SBT method, though the CLSI has
described methods in an approved guideline,
including broth macrodilution and broth
microdilution methods.
The SBT is like the MIC-MBC test in that the
serum inhibitory test (SIT) and SBT parameters
are evaluated. The patient’s serum during drug
therapy and the bacterial isolate responsible for
the patient’s infection are required. Serial two-
fold dilutions of the patient’s serum are
prepared, and then a standardized inoculum of
the patient’s bacterial isolate is added to each
dilution. After overnight incubation, the tubes
or wells are examined to determine the greatest
dilution of the patient’s serum that inhibits the
bacteria (SIT). Subsequently, as with the MBC
test, the contents of all tubes or wells show- ing
inhibition are subcultured on an agar medium
to deter- mine the highest dilution that kills
>99.9% of the bacteria (SBT). The SBT result
relates to the amount of antimicrobial agent
and other antibacterial factors (e.g., antibody,
comple- ment) present in the patient’s serum.
Timing the collection of serum is critical, and
generally
trough (taken just before drug is administered)
and peak (taken soon after dose is
administered) titers are tested. A trough
bactericidal titer of 1:32 or greater and a peak
bacte- ricidal titer of 1:64 or greater has been
historically correlated with bacteriologic cure in
patients with endocarditis. The described
methods are controversial, and currently there
are few situations in which this test is of clinical
value. Current drugs used for the treatment of
SBE generally achieve bacte- ricidal levels when
given for therapy, and it is very difficult to
achieve and monitor serum levels of the drug in
the patient accurately.

Measurement 0f antimicr0bial
agents in serum and b0dy fluids
The amount of antimicrobial agent in serum or
other body fluid can be measured by a variety
of antimicrobial assay procedures.
Antimicrobial assays are performed for antimi-
crobial agents in which the therapeutic
concentration is close to the toxic
concentration. Assay results often lead to
modifi- cation of subsequent doses to prevent
accumulation of exces- sive drug
concentrations that might be harmful to the
patient. A patient’s renal status and hepatic
status greatly influence in vivo levels of some
antimicrobial agents. The antimicrobial agents
with the greatest toxic risks and those most
commonly monitored are the aminoglycosides,
vancomycin, and chlor- amphenicol. For
evaluation of antimicrobial levels, trough and
peak samples should be assayed as for the
SBT.
 MeaSurement 0f antimicr0bial agentS in Serum and b0dy fluidS 303

0.5

0.25

0.12

0.06

A 0 0.06 0.12 0.25 0.5 1 2 B

C
Fig. 13.27 Assassmant 0f antimicr0bial c0mbinati0ns with tha chackarb0ard math0d. All thraa diagrams dapict tha rasults 0f tasting c0mbinati0ns 0f tw0 drugs
(dilutad in ga0matric tw0f0ld incramants al0ng tha x- and y-axas, drug A al0ng tha x-axis and drug B al0ng tha y-axis). Shading indicatas visibla gr0wth, and
c0ncantrati0ns ara axprassad as multiplas 0f tha minimal inhibit0ry c0ncantrati0n. A, Indiffaranca. B, Synargism. C, Antag0nism. (MOdified f2Om Pillai, S. K., et al.
(2005). AntimiC2Obial COmbinatiOns. In V. LO2ian (Ed.), AntibiOtiCs in labO2atO2y mediCine: making a diffe2enCe (5th ed.). Philadelphia: LippinCOtt Williams & Wilkins.)

today when the focus is on the amount of biologically vancomycin, and chloramphenicol assays.
active drug present, rather than the amount of
“chemical” present. The bioassays use a specific
strain of bacteria (indicator organ- ism) that is
susceptible to the drug to be assayed. The test can
be performed in broth or on agar. The antibacterial
activity of the patient’s specimen against the
bacterium is compared with that of solutions
containing defined concentrations of the antimicrobial
being assayed to determine the concentration in the
patient’s specimen using standard dose-response
curves.

ImmunOassays
Radioimmunoassay, fluorescent immunoassay,
fluorescent polarization immunoassay, and enzyme
immunoassay pro- cedures have all been used to
measure antimicrobial agents in serum and other
body fluids. The basic principles of these assays are
similar in that they all use antibodies directed
against the specific antimicrobial agents to be
assayed. Because of the nature of these tests, they
may be performed in the chemistry or immunology
sections of the laboratory. Several commercial
manufacturers offer various immunoas- say kits for
quantifying gentamicin, tobramycin, amikacin,
ChrOmatOgraphic assays
Various chromatographic methods, including gas-
liquid, thin-layer, paper chromatography, and high-
performance liq- uid chromatography, have been
used on occasion for antimi- crobial assays,
primarily to measure levels of antimicrobial agents
for which commercial immunoassay kits are not
avail- able. These tests are performed in research
settings only.

P0INTS T0 REMEMBER

• Antimicrobial susceptibility testing is performed

only on bacteria likely causing an infection.
• The CLSI determines standards for antimicrobial
susceptibil- ity testing and reporting in clinical
laboratories in the United States.
• The FDA has the regulatory authority for setting
susceptibil- ity test interpretive criteria and QC
parameters.
• Antimicrobial susceptibility testing protocols
describing what, when, and how to test and report
should be developed with input from clinicians,
pharmacists, and others who have
304 PART 1 13 AntimicrObial SuSceptibility teSting

clinical experience with the antimicrobial therapy order Enterobacterales.

practices of the institution. • Most automated instruments for antimicrobial
• Selective or cascading reporting refers to susceptibil- ity testing are based on turbidimetric
reporting broad- er-spectrum agents only in select detection of bacterial growth or detection of hydrolysis
situations, such as when of a fluorogenic growth substrate.
the patient’s isolate is resistant to narrower-spectrum • QC of antimicrobial susceptibility tests is performed
agents. with standard reference strains that have defined
• Several variables must be controlled when antimicrobial susceptibility or resistance to the drugs
performing any type of antimicrobial susceptibility tested.
test; inoculum standard- ization is one of the most
important of these. Testing too few or too many
bacteria can yield erroneous results.
• The disk diffusion, Etest, broth microdilution MIC
test, and automated MIC tests are the most
common methods cur- rently used for antimicrobial
susceptibility testing in clinical laboratories.
• The disk diffusion (or Kirby-Bauer) test is a
qualitative method; results are reported as
susceptible, intermediate, nonsusceptible, or
resistant.
• The MIC test is a semi-quantitative method, and the
concentration (pg/mL) of a drug required to inhibit the
growth of bacterial isolate is reported together with a
susceptible, intermediate, nonsusceptible, or
resistant interpretation (e.g., ampicillin MIC = 8
pg/mL; susceptible).
• Routine antimicrobial susceptibility test methods
can be modified for testing fastidious bacteria that
require supple- mental nutrients, modified
incubation conditions, or both.
• An oxacillin disk (1 pg) can be used to screen for
penicillin susceptibility in S. pneumoniae on isolates
not recovered from blood or CSF.
• β-hemolytic streptococci remain universally susceptible
to pen-
icillin, the drug of choice for treating infections caused
by these organisms, and susceptibility testing is
usually not required.
• The CLSI reference method described for testing
obligate anaerobic bacteria is agar dilution; however,
a broth microdi- lution method can be used for less
fastidious species of anaerobes, such as Bacteroides
spp.
• The gene most responsible for oxacillin resistance in
staphy- lococci is mecA.
• The cefoxitin disk performs better than the oxacillin
disk in detecting oxacillin-resistant staphylococci by
the disk diffu- sion method.
• Oxacillin-resistant staphylococci should be
considered resistant to all β-lactam agents
(including cephalosporins,
β-lactam–β-lactamase inhibitor combinations, and
carbapen-
ems), despite any susceptible result for these in vitro.
• The D-zone test uses erythromycin to detect
inducible clindamycin resistance in staphylococci
and β- hemolytic streptococci.
• The vancomycin broth microdilution MIC test or
vancomy- cin agar screen is recommended to
detect VISA or VRSA.
• VRE are important healthcare-associated pathogens;
however, low-level vancomycin resistance is intrinsic
in E. gallinarum and E. casseli}avus. These differ
from true VRE for infection control purposes.
• ESBL-producing bacteria should be considered
resistant
to all cephalosporins, penicillins, and aztreonam,
despite a susceptible result for these in vitro.
ESBLs are prevalent in bacteria belonging to the
• Certain species have typical antibiograms that resistance to the aminoglycoside gentamicin.
can be used in empiric therapy and to verify Which of the following statements is true?
the identification and suscepti- bility results a. Ampicillin and gentamicin will be synergistic.
generated on the isolate. b. Ampicillin and gentamicin will not be synergistic.
• Cumulative antibiograms represent the
percentage of isolates of a given species
susceptible to the antimicrobial agents
commonly tested against the species.
• Nonsusceptible does not necessarily mean that
the organism is resistant to the antimicrobial.
• MBC testing or serum bactericidal testing
may be useful in select situations, such as
for patients who are immuno- suppressed
with serious infections or in whom infection is
at a site where immune mechanisms are not
optimal, e.g., endocarditis.
• MBC testing is performed after completion of
a broth dilu- tion MIC test, and the end point
is 99.9% killing of the test bacteria.
• Serum bactericidal testing requires use of the
patient’s serum obtained at appropriate times
surrounding dosing of the antimicrobial
agent(s), that is, peak and trough, and the bac-
terium causing the patient’s infection.
• Synergism testing of drug combinations can
be performed using a time-kill or
checkerboard assay described by the CLSI in
certain select situations.
• Many clinical laboratories use a mecA assay
to determine oxacillin susceptibility or
resistance in staphylococci and vanA and
vanB assays to detect vancomycin
resistance in enterococci.

LEARNING ASSESSMENT QUESTI0NS

1. Why would it be inappropriate to perform

antimicrobial susceptibility tests on viridans
streptococci isolated from a throat culture?
2. The turbidity of a McFarland 0.5 standard
corresponds to approximately —— colony-
forming units per milliliter.
3. The —— disk cannot be used to screen for penicillin
suscep-
tibility in Streptococcus pneumoniae from sputum.
4. Which method does the Clinical and
Laboratory Standards Institute (CLSI) suggest
as the most reliable for detecting oxacillin
resistance in staphylococci?
a. Oxacillin disk diffusion test
b. Cefoxitin disk diffusion test
C. Penicillin minimal inhibitory concentration (MIC)
test
d. Oxacillin agar screen
5. What does the mecA gene code for in staphylococci?
a. β-Lactamase and penicillin resistance
b. Altered penicillin-binding protein and
vancomycin resistance
C. Penicillin-binding protein 2a and oxacillin
resistance
d. β-Lactamase and oxacillin resistance
6. With which of the following profiles should S. aureus
or
β-hemolytic streptococci be subjected to in D-zone
testing?
a. Oxacillin-resistant
b. Penicillin-resistant
C. Erythromycin-resistant and clindamycin-resistant
d. Erythromycin-resistant and clindamycin-
susceptible
7. An ampicillin-susceptible Enterococcus
faecalis from a blood culture has high-level
 MeaSurement 0f antimicr0bial agentS in Serum and b0dy fluidS 305
describes synergism?
a. The activity of the drug combination is greater
than that of the individual agents.
C. Penicillin and gentamicin will be synergistic. b. The activity of the drug combination is less than
d. Cefazolin and gentamicin will be synergistic. that of the individual agents.
8. Extended-spectrum β-lactamase–producing isolates C. The activity of the drug combination is equal to
should be considered resistant to which of the that of the individual agents.
following agents? d. The test organism is resistant to both drugs in the
a. Cephalosporins, penicillins, and aztreonam combination.
b. Cephalosporins, penicillins, and aminoglycosides
C. Cephalosporins, penicillins, and β-lactamase
inhibitors
d. Cefotaxime–clavulanic acid and ceftazidime–
clavulanic acid
9. Which of the following organisms is commonly
tested for
β-lactamase production?
a. Neisseria meningitidis
b. Klebsiella pneumoniae
C. Streptococcus pneumoniae
d. Haemophilus in}uenzae
10. Which of the following statements is true about
quality control (QC) testing of antimicrobial
susceptibility tests?
a. A laboratory must perform QC every day that
patient isolates are tested.
b. A laboratory can perform QC weekly if it performs
fewer than 10 tests daily.
C. A laboratory can perform QC weekly once
accurate performance of 20 to 30 days of daily
QC has been documented.
d. A laboratory does not have to perform QC if an
automated test antimicrobial susceptibility test
system is used.
11. Identify each of the following statements as true or
false.
a. Ampicillin susceptibility in Pseudomonas
aeruginosa is unusual. ——
b. Many Staphylococcus aureus strains are
vancomycin
resistant. ——
C. Penicillin is the drug of choice for treating
gonorrhea. ——
d. Stenotrophomonas maltophilia strains are usually
susceptible to trimethoprim-sulfamethoxazole.
——
e. Escherichia coli strains are expected to be
resistant to all
aminoglycosides. ——
12. Which of the following statements is true?
a. An organism that is reported as nonsusceptible
to an antimicrobial is definitely resistant to the
antimicrobial.
b. The term nonsusceptible is used when the
interpretive categories of intermediate and
resistant do not apply.
C. The organization that oversees preparing
technical doc- uments for international use is
the European Medicines Agency.
d. The vanA gene codes for chloramphenicol
resistance.
13. Why is a bactericidal drug regimen necessary for
treating patients with bacterial endocarditis?
14. The MBC end point is the lowest concentration of
antimi- crobial agent that kills --- of the test
bacteria.
15. It is important to test bacteria in the --- phase of
growth when performing tests to assess
bactericidal activity.
16. Which of the following definitions best
17. Why is the use of molecular assays to detect
antimicrobial resistance genes limited?
a. Genes are not responsible for most types of
antimicrobial resistance.
b. Genes may be present but may not be
expressed; there- fore the presence of the gene
does not always correlate with resistance.
C. Researchers have been unable to identify genes
for anti- microbial resistance.
d. Testing is too expensive.
18. Which of the following classes of antimicrobial
agents poses the greatest toxicity risks and
therefore is frequently moni- tored using
antimicrobial assays?
a. Cephalosporins
b. Sulfonamides
C. Aminoglycosides
d. Tetracyclines
19. The presence of the blaZ gene is expected to
confer resist- ance to which of the following?
a. Cephalosporins only
b. Erythromycin
C. Oxacillin
d. Several β-lactam antibiotics

B I B L I 0 G RA P H Y

Clinical and Laboratory Standards Institute. (2008).

Development of in vitro susceptibility testing criteria and
quality control parameters: approved guideline M2S-MAS.
Wayne, PA: Clinical and Laboratory Standards Institute.
Clinical and Laboratory Standards Institute. (2011).
Susceptibility testing of mycobacteria, nocardiae, and
other aerobic actinomycetes: approved standard, 2nd.
M24-A2. Wayne, PA: Clinical and Laboratory Standards
Institute.
Clinical and Laboratory Standards Institute. (2014). Analysis
and presentation of cumulative antimicrobial susceptibility
test data: approved guideline MS9-A4. Wayne, PA: Clinical
and Laboratory Standards Institute.
Clinical and Laboratory Standards Institute. (2015). Methods
for antimicrobial susceptibility testing of anaerobic
bacteria: approved standard M11 (9th ed.). Wayne, PA:
Clinical and Laboratory Standards Institute.
Clinical and Laboratory Standards Institute. (2016). Methods
for antimicrobial dilution and disk susceptibility testing of
infrequently isolated or fastidious bacteria, approved
guideline M45 (3rd ed.). Wayne, PA: Clinical and
Laboratory Standards Institute.
Clinical and Laboratory Standards Institute. (2018). Methods
of dilution antimicrobial susceptibility testing for bacteria
that grow aerobically: approved standard M7 (11th ed.).
Wayne, PA: Clinical and Laboratory Standards Institute.
Clinical and Laboratory Standards Institute. (2018).
Performance standards for antimicrobial disk
susceptibility tests: approved standard M2 (13th ed.).
Wayne, PA: Clinical and Laboratory Standards Institute.
Clinical and Laboratory Standards Institute. (2022).
Performance standards for antimicrobial susceptibility
testing. CLSI publication M100 (32nd ed.). Wayne, PA:
Clinical and Laboratory Standards Institute.
Ford, B. (2017). mecC-Harboring methicillin-resistant
Staphylococcus aureus: hiding in plain sight. Commentary.
Journal of Clinical Microbiology, 56, e01549-17.
Hryniewicz, M. M., & Garbacz, K. (2017). Borderline oxacillin-
resistant Staphylococcus aureus (BORSA) – a more
common problem than expected? Journal of Medical
Microbiology, 66, 1367.
Karlowsky, J. A., & Richter, S. S. (2019). Antimicrobial testing
systems. In K. C. Carroll, et al. (Eds.), Manual of clinical
microbiology (12th ed., p. 1300). Washington, DC: ASM
Press.
306 PART 1 13 AntimicrObial SuSceptibility teSting

Kest, H., & Kaushik, A. (2019). Vancomycin-resistant

Staphylococcus aureus: formidable threat or silence before Simner, P. J., & Humphries, R. (2019). Special phenotypic
the storm? Journal of Infectious Diseases and Epidemiology, methods for detecting antibacterial resistance. In K. C.
5. https://doi.org/10.23937/2474- 3658/1510093. Carroll, et al. (Eds.), Manual of clinical microbiology (12th
Koeth, L. M., & Miller, L. A. (2019). Antimicrobial ed., p. 1316). Washington, DC: ASM Press.
susceptibility test methods: dilution and disk diffusion Singh, N., et al. (2019). Infections in solid organ transplant recipients. In
methods. In K. C. Carroll, et al. (Eds.), Manual of clinical J.
microbiology (12th ed., p. 1284). E. Bennett, et al. (Eds.), Mandell, Douglas, and Bennett’s
Washington, DC: ASM Press. principles and practice of infectious diseases (9th ed., p.
Marsik, F. J., & Nambiar, S. (2012). Review of carbapenemases 3672). Philadelphia: Elsevier.
and AmpC-beta lactamases. Pediatric Infectious Disease U.S. Food and Drug Administration. (2020). Postmarket drug
Journal, S0, 1094. safety information for patients and providers. Available at:
Munro, S. D. (2016). Antimicrobial susceptibility testing. In A. https://www.fda.
Leber (Ed.), Clinical microbiology procedures handbook gov/drugs/drug-safety-and-availability/postmarket-drug-
(4th ed.). Washington, DC: ASM Press. 5.0.1. safety- information-patients-and-providers. (Accessed 10
May 2022).
van der Zwaluw, K., et al. (2015). The carbapenem inactivation
method (CIM), a simple and low-cost alternative for the
Carba NP test to assess phenotypic carbapenemase activity
in gram-negative rods. PLoS ONE, 10, e0123690.