Antimicrobial Susceptibility Testing and Resistance Mechanisms

Mueller-Hinton medium is the standard medium because its lots perform consistently and it contains little of the inhibitors that distort several drug classes: thymidine and thymine, which antagonize sulfonamides and trimethoprim, and excess divalent cations, which distort aminoglycoside and tetracycline results. Excess calcium and magnesium make Pseudomonas aeruginosa falsely resistant to aminoglycosides. Low cation content makes it falsely susceptible. The agar must have a final pH of 7.2 to 7.4 and a depth of 4 mm. Agar that is too deep thins the antibiotic gradient and makes the isolate look more resistant. Agar that is too shallow makes it look more susceptible. Disk diffusion uses Mueller-Hinton agar. Broth dilution uses cation-adjusted Mueller-Hinton broth (CAMHB), adjusted to roughly 20 to 25 mg/L calcium and 10 to 12.5 mg/L magnesium. Do not use unadjusted broth, because aminoglycoside and tetracycline MICs shift with cation content.^{4,54 Clinical and Laboratory Standards Institute. M02, Disk Diffusion 14th ed. 2024.5 Centers for Disease Control and Prevention. Laboratory assessment of antibiotic resistance testing capacity.}

Fastidious organisms need supplementation or a different medium: 5% sheep blood on Mueller-Hinton agar for streptococci by disk diffusion, 2.5 to 5% lysed horse blood in CAMHB for streptococci by dilution, Haemophilus test medium (HTM) for Haemophilus influenzae, and GC agar base with 1% defined growth supplement for Neisseria gonorrhoeae. Anaerobes are tested in supplemented Brucella media (see below). Disk diffusion is not a standardized method for anaerobes.

Prepare a 0.5 McFarland suspension, about 1 to 2 × 10⁸ colony-forming units (CFU)/mL, from fresh colonies by direct suspension, or by growth in broth when the organism will not suspend cleanly. For broth microdilution, dilute that suspension to a final inoculum of about 5 × 10⁵ CFU/mL. The 0.5 McFarland standard sets the density of the stock suspension, and the dilution step sets the final concentration in the well. Use the inoculated plate or tray within the time window the method specifies, because a standardized inoculum is part of the method.

Routine incubation is 35 °C ± 2 °C, and the reading time is set by each organism-drug table: disk diffusion for the common rapidly growing organisms is read at 16 to 18 hours, and dilution at 16 to 20 hours. Cefoxitin disk diffusion is read at 16 to 18 hours for Staphylococcus aureus and Staphylococcus lugdunensis, and at 24 hours for Staphylococcus epidermidis and other staphylococcal species for which that method is applicable.^{2323 Clinical and Laboratory Standards Institute. M100, 30th ed., combined corrections. 2020.} Oxacillin and vancomycin MIC tests of staphylococci, and vancomycin testing of enterococci, require a full 24 hours. An MIC endpoint is a concentration. A disk-diffusion endpoint is a zone diameter. Haemophilus zones are read at 16 to 18 hours in 5% CO₂. Streptococcal zones are read at 20 to 24 hours in 5% CO₂. Testing staphylococci above 35 °C can also miss methicillin resistance.^{1,6,71 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.6 Centers for Disease Control and Prevention. Laboratory testing for methicillin (oxacillin)-resistant… 2025.7 Centers for Disease Control and Prevention. Laboratory testing for vancomycin-resistant Staphylococcus… 2024.}

The MIC is the lowest concentration that prevents visible growth. The report requires an interpretive category, assigned by comparing the MIC with the applicable breakpoint. CLSI defines these categories:

European Committee on Antimicrobial Susceptibility Testing (EUCAST) practice differs. EUCAST dropped the intermediate buffer concept and defines I as a drug-exposure target. When the urine-concentration use of I matters clinically, CLSI marks it with a superscript (I^{^}) in the tables.^{88 Clinical and Laboratory Standards Institute. Re-exploring the intermediate interpretive category.}

Disk diffusion (the Kirby-Bauer method) applies to rapidly growing aerobic and facultative bacteria and to the fastidious organisms for which CLSI publishes disk criteria. A paper disk carrying a fixed drug amount is placed on a standardized lawn. The drug diffuses outward, and the zone of inhibition correlates inversely with the MIC. A 150-mm plate holds no more than 12 disks, a 100-mm plate no more than 5, and disks sit at least 24 mm apart center to center so zones do not overlap. The method is inexpensive and gives a category with no MIC, which limits its use when a quantitative result matters, as with vancomycin against S. aureus or any polymyxin.^{6,76 Centers for Disease Control and Prevention. Laboratory testing for methicillin (oxacillin)-resistant… 2025.7 Centers for Disease Control and Prevention. Laboratory testing for vancomycin-resistant Staphylococcus… 2024.}

Direct β-lactamase testing is a separate rapid method: a chromogenic cephalosporin (nitrocefin) changes color when hydrolyzed. It is most useful for H. influenzae and, on selected blood or cerebrospinal fluid (CSF) isolates of enterococci, where routine MIC testing can miss β-lactamase. A positive nitrocefin result predicts resistance to ampicillin and amoxicillin. A negative result does not establish ampicillin susceptibility, because the test does not detect β-lactamase-negative, ampicillin-resistant (BLNAR) or penicillin-binding protein (PBP)-mediated resistance (see below). For S. aureus, CLSI prefers the penicillin zone-edge test before reporting penicillin susceptible, because a sharp "cliff" edge reveals penicillinase that a plain nitrocefin test can miss. For other staphylococci, CLSI recommends a nitrocefin-based test.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.}

The MBC is found by subculturing the clear wells of an MIC series to drug-free agar. It is the lowest concentration that kills ≥99.9% of the original inoculum. Persisters, a small surviving subpopulation that remains fully susceptible on retesting, can blunt the endpoint, as can the Eagle (paradoxical) effect, in which very high concentrations kill less than intermediate ones. Tolerance has historically been described as an MBC:MIC ratio of at least 32. Current reviews treat that cutoff as a historical convention with no validated clinical threshold behind it, and consider time-kill methods the better reference.^{1212 Deventer AT, et al. Antibiotic tolerance among clinical isolates. 2024.}

Application: MBC and the persister concept. Continuing the Drug Y case (MIC = 4 µg/mL), the clear wells at 4, 8, 16, and 32 µg/mL are subcultured to drug-free agar. The starting inoculum was about 4 × 10⁵ CFU/mL, so 99.9% killing allows no more than 400 CFU/mL to survive. The 4-µg/mL subculture is too numerous to count, but the 8-µg/mL subculture yields 250 CFU/mL. The MBC is 8 µg/mL and the MBC:MIC ratio is 8:4 = 2, far below the conventional tolerance ratio of 32.

Commercial automated broth microdilution systems predominate in US clinical laboratories. Turbidimetric and fluorometric platforms read inoculated cards or trays at intervals and report an MIC and category within hours. Their advantages are paired identification and AST inoculation, expert-system software, and laboratory information system (LIS) interfacing. Verify a system against a reference method before patient reporting: best practice is at least 90% essential agreement (MIC within one doubling dilution) and 90% categorical agreement, along with a review of very major errors (false susceptible), which carry the most clinical risk. Verification is required when a system is introduced and again when it materially changes, including implementation of new breakpoints, new agents, or new software.^{1313 Clinical and Laboratory Standards Institute. M52, Verification of Commercial Microbiological Systems. 2015.}

Molecular AST detects resistance genes directly by nucleic acid amplification test (NAAT), without waiting for growth. It can return results within hours from a positive blood-culture bottle or a rectal screening swab. Interpret these results with three limits in mind. Resistance to one drug can arise from many unrelated genes. A detected gene does not prove an expressed phenotype. A negative panel excludes only the mechanisms it targets. Molecular detection is most reliable where a single dominant gene accounts for the mechanism, such as mecA in staphylococci, vanA/vanB in enterococci, or rpoB mutations in Mycobacterium tuberculosis complex. CLSI also standardizes direct disk diffusion from positive blood-culture bottles for Enterobacterales, P. aeruginosa, and Acinetobacter, which shortens turnaround without molecular testing. Whole-genome sequencing can infer a resistance genotype without a predefined target. It misses determinants that are not yet catalogued, and it is used mainly as a reference and epidemiologic tool.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.}

The Clinical Laboratory Improvement Amendments (CLIA) require the laboratory to check each batch of medium and each lot or shipment of antimicrobial agent before or concurrent with initial use, and to run appropriate control organisms each day tests are performed, with all results in range before patient results are reported. The default is daily quality control (QC). A laboratory may reduce frequency only under a written individualized quality control plan (IQCP), which the Centers for Medicare & Medicaid Services (CMS) treats as voluntary and which must document the risk assessment supporting the reduction. A routine weekly QC schedule adopted without an IQCP does not meet the CLIA requirement.^{14,1514 Code of Federal Regulations. 42 CFR §493.1261.15 Centers for Medicare and Medicaid Services. FAQs for Individualized Quality Control Plan (IQCP).}

Standard control strains for disk diffusion and dilution QC include Escherichia coli ATCC 25922, P. aeruginosa ATCC 27853, S. aureus ATCC 29213 (MIC) or ATCC 25923 (disk), Enterococcus faecalis ATCC 29212, H. influenzae ATCC 49247 and 49766, N. gonorrhoeae ATCC 49226, Streptococcus pneumoniae ATCC 49619, and the resistance-mechanism controls needed for specific tests (E. coli ATCC 35218, a β-lactamase producer, for inhibitor combinations; E. faecalis ATCC 51299 for vancomycin resistance). Acceptable zone and MIC limits are published in M100 for each strain-agent-method combination. Apply those published ranges. For the nitrocefin test, S. aureus ATCC 29213 is the positive control and H. influenzae ATCC 10211 the negative control. When a QC result falls out of range, stop testing, find the cause, review the patient results reported since the last in-range control, and document corrective action before reporting resumes.^{1,14,151 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.14 Code of Federal Regulations. 42 CFR §493.1261.15 Centers for Medicare and Medicaid Services. FAQs for Individualized Quality Control Plan (IQCP).}

Test when an organism's susceptibility is unpredictable. Streptococcus pyogenes pharyngitis needs no penicillin testing, because resistance has not been documented in the United States. Anaerobic susceptibility testing is indicated for isolates from brain abscess, endocarditis, osteomyelitis, joint infection, prosthetic or vascular-device infection, and bacteremia. CLSI M11 names agar dilution on supplemented Brucella blood agar as the reference method for anaerobes. Broth microdilution is acceptable for Bacteroides and Parabacteroides with selected agents. CLSI does not provide a routine anaerobic disk-diffusion method. EUCAST publishes a separate standardized method for selected rapidly growing anaerobes whose media, breakpoints, and categories are not interchangeable with CLSI's. Most Bacteroides are presumed penicillin-resistant. A positive β-lactamase test predicts penicillin and ampicillin resistance. A negative test does not establish susceptibility, because non-enzyme mechanisms also occur. β-Lactamase results do not predict piperacillin-tazobactam or carbapenem results.^{1616 Clinical and Laboratory Standards Institute. M11, Methods for Antimicrobial Susceptibility Testing of… 2018.}

CLSI's M100 tables group suggested drugs by pathogen category and tier: Tier 1 for routine primary testing and reporting, Tier 2 for testing routinely with selective or cascade reporting, Tier 3 for institutions with endemic multidrug-resistant organisms or patients allergic to primary agents, and Tier 4 for testing or reporting on request or in special circumstances. Agents with a U designation apply to urinary isolates only. When two agents appear in the same box separated by "or", near-complete cross-susceptibility and cross-resistance mean one result predicts the other. When they share a box without "or", clinical efficacy is similar but the results do not predict each other.

Some results must never be reported as susceptible regardless of the in vitro reading, because the drug is not clinically effective for that organism or site. The recurring combinations are:

• For Salmonella and Shigella, aminoglycosides, first- and second-generation cephalosporins, and cephamycins may look active in vitro but are not clinically effective and must not be reported as susceptible. Intestinal isolates of nontyphoidal Salmonella need no routine AST. Shigella and typhoidal Salmonella always need routine AST. For extraintestinal Salmonella, the preferred fluoroquinolone test is a ciprofloxacin MIC, because plasmid-mediated low-level resistance can test susceptible by nalidixic acid or disk screen.
• For Enterococcus, cephalosporins, clindamycin, and TMP-SMX may look active in vitro but are not clinically effective and must not be reported as susceptible. Aminoglycosides are reported only through high-level synergy testing (below).
• For isolates from cerebrospinal fluid, do not report agents given by mouth only, first- and second-generation cephalosporins and cephamycins, ertapenem, doripenem, imipenem, clindamycin, macrolides, lefamulin, and tetracyclines. These are not drugs of choice for central nervous system (CNS) infection and may not reach it.

Intrinsic resistance is a property of the species: an isolate of that species is resistant regardless of what the in vitro result shows, and CLSI's intrinsic-resistance appendix lists the specific combinations that qualify. The entries seen most often include ampicillin for Klebsiella; nitrofurantoin, tetracyclines, and colistin for Proteus, Providencia, and Morganella; colistin for Serratia; carbapenems and aminoglycosides for Stenotrophomonas maltophilia, whose L1 metallo-β-lactamase and efflux make it intrinsically carbapenem-resistant. In a recent change, CLSI no longer lists intrinsic-resistance entries for Burkholderia cepacia complex because the previous entries could not be confirmed. CLSI removed the complex's disk diffusion breakpoints in 2024 (34th edition) and its MIC breakpoints in 2025 (35th edition), so the current M100 publishes no breakpoints for it and recommends against routine AST. If testing is performed on request, frozen reference broth microdilution is the only reproducible method, and the result is reported with the caution that correlation of MIC values with clinical outcome is not known.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.},^{2424 Huse H, et al. A clinical case of Burkholderia cepacia complex and changes to… 2025.}

Selective and cascade reporting withhold second-line agents until first-line ones are reported resistant, which preserves active drugs for genuinely resistant infections and supports stewardship. Formulary and menu decisions are made jointly with the antimicrobial stewardship team and the Pharmacy and Therapeutics Committee, since commercial panels constrain which drugs can be tested together. Stewardship programs also restrict specific agents so that their use needs approval. The notable example is fluoroquinolone restriction, which curbed outbreaks of fluoroquinolone-resistant ribotype 027 Clostridioides difficile as one part of a bundle of control measures.^{1717 Kallen AJ, et al. Complete restriction of fluoroquinolone use to control an… 2009.}

An antibiogram compiles a facility's cumulative susceptibility results as percentage susceptible by species and drug. The Clinical and Laboratory Standards Institute (CLSI) M39 5th-edition guidance recommends at least annual analysis of final, verified diagnostic results, with only the first isolate of each species per patient in the analysis period, regardless of specimen source or susceptibility pattern. Screening and surveillance records are excluded. The routine report includes species with at least 30 isolates tested and counts only susceptible results in its susceptible numerator. Intermediate and susceptible-dose-dependent categories remain distinct. Smaller groups give less precise estimates.^{1818 Simner PJ, et al. What's new in antibiograms? Updating CLSI M39 guidance with… 2022.}

Selecting isolates for an annual antibiogram

A fictional hospital reviews diagnostic cultures collected from January 1 through December 31, 2025, across its inpatient and outpatient services. Its export contains 50 Escherichia coli records. The scientist first checks collection dates, patient identifiers, species, culture purpose, and final result status. The counts below are teaching data and do not represent local resistance rates.

For example, patient P07 has a screening rectal isolate on January 1, a diagnostic blood isolate on January 3, and a diagnostic urine isolate on February 6. The January 3 isolate is retained. The later urine isolate is a repeat for this species even if its susceptibility differs. A different species from P07 would be considered separately. Thus, 50 − 8 − 8 = 34 eligible E. coli isolates from 34 patients.

Both ceftriaxone and cefepime were routinely tested on all 34 retained isolates. The extract includes verified results withheld from individual patient reports by selective reporting rules. Neither drug was tested only on resistant isolates. The laboratory's recorded breakpoint version was consistent throughout this fictional dataset. The example begins with final categories and makes no new breakpoint assignments. Using only selectively released results could distort the percentage.^{1818 Simner PJ, et al. What's new in antibiograms? Updating CLSI M39 guidance with… 2022.}

Susceptible (S), intermediate (I), susceptible-dose dependent (SDD), and resistant (R) results contribute to the tested total. Only S contributes to percent susceptible.

Percent susceptible = (susceptible eligible isolates ÷ eligible isolates with a valid result for that drug) × 100. The denominators happen to match here because both drugs have results for every retained isolate. An untested isolate is excluded from that drug's denominator.

• Ceftriaxone: 24 ÷ (24 + 2 + 8) × 100 = 70.6% susceptible.
• Cefepime: 26 ÷ (26 + 4 + 4) × 100 = 76.5% susceptible.

Counting the two ceftriaxone I results as S would give 26 ÷ 34 × 100 = 76.5%. Counting the four cefepime SDD results as S would give 30 ÷ 34 × 100 = 88.2%. Both calculations overstate percent susceptible. The I and SDD isolates remain in their denominators without being relabeled R. This is the routine M39 5th-edition percent-S calculation. It does not combine categories for a separate exposure- or urinary-site analysis.^{1818 Simner PJ, et al. What's new in antibiograms? Updating CLSI M39 guidance with… 2022.}

All eight excluded diagnostic repeats were ceftriaxone R in this dataset. Including them would give 24 ÷ 42 × 100 = 57.1%, overrepresenting those repeatedly cultured patients. The exclusion rule would apply equally if the repeats were susceptible.

The laboratory publishes the two E. coli percentages with their tested counts, period, population, and selection rules after review. A separate Klebsiella pneumoniae group has only 29 eligible isolates, so its percentages are withheld from this routine report under the stated minimum rule. The antibiogram describes the selected population. It cannot change an individual isolate's reported category or establish which antimicrobial will work for a particular patient.^{1818 Simner PJ, et al. What's new in antibiograms? Updating CLSI M39 guidance with… 2022.}

Resistance genes move vertically with clonal expansion under antibiotic pressure or horizontally by transformation (uptake of free DNA), transduction (phage-mediated transfer), or conjugation (cell-to-cell plasmid transfer, which crosses species boundaries). The transfer of vanA from enterococci to S. aureus, producing vancomycin-resistant S. aureus (VRSA), is the best-known example.

A detected gene is conclusive evidence when an MIC is near a breakpoint. Molecular detection can also bypass culture for speed or safety. A negative molecular result excludes only the mechanisms the assay targets. A resistant phenotype can arise from an untested gene, a novel mutation, or a permeability or efflux change the panel does not target.

Methicillin resistance in staphylococci comes from PBP2a, an altered penicillin-binding protein with low β-lactam affinity, encoded by mecA (or the rarer mecC) on the SCCmec cassette. It confers resistance to nearly all β-lactams except the cephalosporins active against methicillin-resistant S. aureus (MRSA), such as ceftaroline. Resistance can be homogeneous, which is easy to detect, or heterogeneous, where a small resistant subpopulation makes the isolate look falsely susceptible. Detection depends on the species:

Pneumococcal β-lactam resistance comes from mosaic pbp2x, pbp2b, and pbp1a genes. Pneumococci do not produce β-lactamase, so β-lactamase testing is not performed. The 1-µg oxacillin disk screens non-meningitis isolates: a zone of at least 20 mm predicts penicillin susceptibility and requires no further penicillin testing. A zone of 19 mm or less requires a penicillin MIC, because the screen cannot separate susceptible from intermediate or resistant within that range. CSF isolates skip the screen and go directly to MIC testing of penicillin and the third-generation cephalosporins with meningitis breakpoints. Meningitis, non-meningitis, and oral penicillin breakpoints are three separate sets, and an amoxicillin or ceftriaxone MIC cannot substitute for the penicillin result.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.}

Inducible clindamycin resistance appears when an isolate tests erythromycin-resistant and clindamycin-susceptible or intermediate. The usual cause is an erm gene (macrolide-lincosamide-streptogramin B, MLS_B) whose methylase is weakly induced by clindamycin but strongly induced by erythromycin. Therapy with clindamycin can then select for expression and fail. The D-zone test places a 15-µg erythromycin disk 15 to 26 mm edge to edge from a 2-µg clindamycin disk (12 mm apart for streptococci and pneumococci on blood agar). Flattening of the clindamycin zone toward the erythromycin disk, or hazy growth within it, is positive. Report the isolate clindamycin-resistant. A negative D-zone in an erythromycin-resistant, clindamycin-susceptible isolate means inducible clindamycin resistance was not detected. When testing is valid, report the clindamycin-susceptible result. A negative D-zone does not identify the cause of erythromycin resistance or guarantee clinical efficacy. The msr family belongs to the antibiotic-resistance ABC-F proteins associated with ribosomal protection. Membrane efflux, its older description, has not been demonstrated. msr(A) in staphylococci is one of these ABC-F proteins. mef(A) encodes a true major-facilitator-superfamily macrolide efflux pump.^{2121 Sharkey LKR, et al. ABC-F proteins mediate antibiotic resistance through ribosomal… 2016.},^{2525 Feßler AT, et al. Mobile macrolide resistance genes in staphylococci. 2019.} A broth microdilution alternative exists for automated panels. When a group B Streptococcus isolate from a pregnant patient is sent for susceptibility testing because of a high-risk penicillin allergy, the laboratory tests clindamycin and erythromycin and performs the D-zone test on any erythromycin-resistant, clindamycin-susceptible (ERY-R/CLI-S) isolate.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.}

Group A streptococci remain penicillin-susceptible. Macrolide resistance occurs, so the macrolide result is needed when a penicillin allergy requires an alternative agent.

• Penicillin and ampicillin. In a β-lactamase-negative enterococcus, penicillin susceptibility predicts ampicillin susceptibility. Ampicillin susceptibility does not predict penicillin susceptibility, so test penicillin when its result is needed. Most resistance, chiefly in E. faecium, comes from low-affinity PBP5. β-lactamase production is rare, inoculum-dependent, and missed by routine MIC methods, so CLSI recommends direct nitrocefin testing for selected blood or CSF isolates only.
• High-level aminoglycoside resistance (HLAR). Aminoglycosides alone are clinically inactive against enterococci. They are tested only for synergy with a cell-wall-active agent. Screening uses 500 µg/mL gentamicin or 1,000 µg/mL streptomycin in broth (2,000 µg/mL streptomycin on agar), or 120-µg gentamicin and 300-µg streptomycin disks. Disk zones of 7 to 9 mm are inconclusive and require dilution confirmation. Gentamicin results do not predict streptomycin, so each is tested separately. A resistant result means that aminoglycoside will not give synergistic bactericidal activity with a cell-wall-active agent. Incubate streptomycin broth tests a full 48 hours. A susceptible result at 24 hours must be reread after continued incubation.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.}
• Vancomycin resistance. Intrinsic low-level resistance from vanC (D-Ala-D-Ser) occurs in E. gallinarum and E. casseliflavus/ E. flavescens. Infection-prevention measures for isolates with intrinsic vanC resistance follow institutional policy.^{2222 Britt NS, et al. Clinical epidemiology of vancomycin-resistant Enterococcus… 2016.} Acquired high-level resistance comes chiefly from vanA (D-Ala-D-Lac; resistant to vancomycin and teicoplanin, though teicoplanin is not marketed in the United States) or vanB (variable vancomycin resistance). Substitution of D-Ala-D-Lac lowers vancomycin binding affinity roughly 1,000-fold. Screening uses brain heart infusion (BHI) agar with 6 µg/mL vancomycin, read at a full 24 hours. Disk diffusion zones for vancomycin are also read at 24 hours with transmitted light, because a faint haze inside the zone indicates resistance and cannot be seen with direct light. Motility and pigment support species identification. They do not detect vanC or establish whether additional acquired resistance genes are present. Validated vanA/vanB assays detect those specific genes and may be used for direct rectal screening.^{11 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.}

Vancomycin-intermediate S. aureus (VISA; MIC 4 to 8 µg/mL) arises from progressive cell-wall thickening through mutations in several genes, including the vraTSR, graSR, and walKR regulators and rpoB.^{2727 Hu Q, et al. Molecular events for promotion of vancomycin resistance in… 2016.} It is missed by methods that only report categories. VRSA (MIC ≥16 µg/mL) comes from acquisition of the enterococcal vanA operon. All US VRSA isolates have also been MRSA. Vancomycin disk diffusion is not a reliable screen for either. The BHI 6-µg/mL vancomycin screen agar reliably detects MICs of 8 and misses some MICs of 4, so a negative screen does not exclude VISA. Save any S. aureus with a vancomycin MIC of 8 µg/mL or higher and refer it to the state public health laboratory and the Centers for Disease Control and Prevention (CDC) after purity checks and a repeat MIC. Save and refer a confirmed VISA as well. Repeat and confirm a vancomycin MIC of 4 µg/mL. A vancomycin MIC that creeps upward within the susceptible range can reflect heterogeneous intermediate strains, which routine testing does not detect.^{6,76 Centers for Disease Control and Prevention. Laboratory testing for methicillin (oxacillin)-resistant… 2025.7 Centers for Disease Control and Prevention. Laboratory testing for vancomycin-resistant Staphylococcus… 2024.}

The Ambler scheme sorts β-lactamases structurally into classes A-D. The Bush-Jacoby scheme sorts them functionally by substrate and inhibitor profile.

A positive nitrocefin test in H. influenzae predicts ampicillin and amoxicillin resistance. A negative test does not establish ampicillin susceptibility. BLNAR strains have altered PBPs (ftsI mutations) and test ampicillin-resistant on AST. BLNAR isolates are reported resistant to amoxicillin-clavulanate, ampicillin-sulbactam, cefaclor, cefamandole, cefetamet, cefonicid, cefprozil, cefuroxime, loracarbef, and piperacillin-tazobactam despite apparent in vitro susceptibility, because those agents depend on the PBPs the mutations alter. β-lactamase-positive, amoxicillin-clavulanate-resistant (BLPACR) strains have both mechanisms.

Neisseria meningitidis penicillin nonsusceptibility is chiefly PBP-mediated. β-Lactamase is rare. Perform all manipulation of N. meningitidis cultures, including AST, in a biological safety cabinet under enhanced biosafety level 2 (BSL-2) conditions, because laboratory-acquired meningococcal disease has a case fatality rate near 50% and nearly every reported case involved open-bench work. CLSI provides MIC and selected disk methods on blood-supplemented media. For N. gonorrhoeae, the reference method is agar dilution on GC agar base with 1% supplement. Broth dilution is not used. Ceftriaxone and cefixime disk breakpoints have been removed pending review, so MIC testing is required for cephalosporin results. Culture with susceptibility testing is indicated after apparent treatment failure, because NAAT cannot provide AST results.^{1,201 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 2026.20 Centers for Disease Control and Prevention. Fatal meningococcal disease in a laboratory worker. 2014.}