Antimicrobial Susceptibility Testing (AST): Methods, Interpretation, and How to Choose

Antimicrobial susceptibility testing (AST) answers one question a clinician cannot afford to guess at: will this drug work against this patient's organism? Everything in the microbiology lab up to this point identifies what is causing the infection. AST decides what to do about it.

The result looks deceptively simple. A report comes back reading Susceptible, Intermediate, or Resistant for each drug tested. Behind each of those letters sits a chosen method, a standardized inoculum, a defined incubation, a measured endpoint, and an interpretive breakpoint set by an expert committee. Get any of those wrong and the letter is wrong, and a wrong letter sends the wrong drug to a real patient.

This article is the map of that whole process. It covers the four families of AST methods and when each one earns its place, how a raw measurement becomes an S/I/R category, why CLSI and EUCAST can read the same MIC differently, and where the common interpretive traps lie. Each method and resistance mechanism has its own detailed article; this page is the overview that ties them together and points you to the right one.

Why AST exists: the gap between "identified" and "treated"

Identifying an organism narrows the drug choices but rarely settles them. Two Escherichia coli isolates from two patients can have opposite susceptibility profiles because resistance is acquired, mobile, and local. Empirical therapy (best guess before results) buys time, but it is a bet placed against local resistance patterns. AST replaces the bet with a measurement specific to this isolate, and lets therapy be de-escalated from a broad-spectrum guess to a targeted, narrower drug. That de-escalation is the single most important thing AST enables, and it is why the discipline now sits at the center of antimicrobial stewardship.

The four families of AST methods

Every routine method belongs to one of four families. The choice among them is a trade-off between cost, speed, and how much information you need.

Diffusion methods. Antibiotic diffuses out of a disk into agar, creating a concentration gradient; the zone of inhibition around the disk is measured and read against a chart. This is the workhorse of most laboratories: cheap, flexible, and well-standardized. The dominant version is the modified Kirby-Bauer disk diffusion method. Where a paired susceptible control strain shares the plate, the Stokes disk diffusion method is used. Diffusion gives a category (S/I/R), not a number.

Dilution methods. The organism is exposed to doubling dilutions of the drug, and the lowest concentration that stops visible growth is the minimum inhibitory concentration (MIC), an actual number in µg/mL. This is done in tubes or microplates (broth dilution method for MIC) or in agar (agar dilution). Extending the test to subculture the clear tubes gives the minimum bactericidal concentration (MBC). Dilution is the reference standard and the only routine way to get a true MIC, which some clinical situations require.

Gradient methods. A plastic strip carrying a pre-formed antibiotic gradient is laid on the agar; where the growth ellipse meets the strip reads off the MIC directly. The E-test (Epsilometer test) combines the simplicity of a diffusion plate with the numeric MIC of a dilution method, at higher cost per test.

Automated methods. Commercial systems (for example, VITEK 2, BD Phoenix, MicroScan) read growth in drug-containing wells photometrically and derive MICs by algorithm, coupling identification and AST with a fast turnaround. They dominate high-volume clinical labs but depend on the manufacturer's panels and still need manual methods as backup for unusual organisms and mechanisms.

Alongside these sit the standardizing tools every method depends on: the inoculum is matched to a McFarland turbidity standard (read by eye or by a McFarland densitometer), and the disks themselves must be potent, which is why antimicrobial disk storage and stock/disk preparation are treated as their own procedures.

Choosing a method: what the situation demands

The method families are not interchangeable. The right one depends on what the clinical question needs.

A routine urine or wound isolate needs a category, not a number, so disk diffusion is sufficient and economical. Endocarditis, osteomyelitis, meningitis, or a bloodstream infection in a neutropenic patient needs a precise MIC, because dosing must clear a specific threshold, so a dilution or gradient method is used. When a bactericidal drug is essential and tolerance is suspected, the MBC and the MBC/MIC ratio are added.

A high-volume lab that must also identify the organism fast leans on automated systems, keeping manual methods for the isolates the machine cannot call. This selection logic, matching method to clinical stakes, is the judgment an AI summary of "what is AST" leaves out entirely.

From measurement to meaning: breakpoints and S/I/R

A zone diameter in millimeters or an MIC in µg/mL is only a measurement. What turns it into a treatment decision is a breakpoint: the cutoff, set for each organism-drug pair, that divides Susceptible from Intermediate from Resistant. Breakpoints are not arbitrary lines. They fold together the wild-type MIC distribution of the species, the drug's pharmacokinetics and pharmacodynamics (whether killing is concentration-dependent or time-dependent), and the concentrations achievable at the site of infection.

This is why the same drug carries different breakpoints for different organisms, and why "Intermediate" is not laboratory hedging but a real pharmacological zone (often meaning the drug may work at maximum dosing or where it concentrates, such as in urine). The full logic of what Susceptible, Intermediate, and Resistant actually mean is its own topic, and it is the single most misread part of an AST report.

CLSI vs. EUCAST: two committees, sometimes two answers

Two bodies publish the breakpoints the world uses: the Clinical and Laboratory Standards Institute (CLSI) in the United States and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) in Europe. They read the same evidence but can set different breakpoints, different quality-control ranges, and, for EUCAST, a distinct handling of the "I" category as "Susceptible, increased exposure." A laboratory must pick one system and apply it consistently, because a zone diameter that reads Susceptible under one may read Intermediate under the other. Knowing which guideline a report follows is part of reading it correctly.

Detecting resistance mechanisms: when routine AST is not enough

Some clinically critical resistance is invisible to a standard S/I/R read and needs a dedicated test. These sit downstream of the hub as their own procedures:

Methicillin resistance in Staphylococcus aureus (MRSA) is detected with a cefoxitin surrogate, not by testing methicillin directly. Inducible clindamycin resistance hides behind an erythromycin-clindamycin discrepancy and is exposed by the D-test. In Gram-negatives, extended-spectrum beta-lactamases (ESBLs), AmpC enzymes, and carbapenemases each need specific phenotypic detection, because the routine panel can read misleadingly susceptible while a transferable enzyme lurks. The classification behind these enzymes (Ambler and Bush-Jacoby) is covered in the beta-lactamase classification article, and the confirmatory tests (Modified Hodge, Carba NP, phenotypic carbapenemase methods) each have their own page.

AST as the engine of stewardship

An AST result is not the end of the workflow; it is the input to a decision. Restricted (selective) reporting, releasing only the narrowest effective drugs to the clinician, steers prescribing away from broad-spectrum agents. Local AST data aggregated over time builds the antibiogram that guides empirical therapy for the next patient. And the WHO AWaRe classification frames which drugs to protect. This is where the laboratory stops being a service and becomes a partner in antimicrobial stewardship.

AMR/AST Directory

Standardizing the test

• Modified Kirby-Bauer disk diffusion method
• Stokes disk diffusion method
• McFarland turbidity standards
• McFarland densitometer
• Antibiotic stock solutions and disk preparation
• Storing antimicrobial disks
• Antimicrobial Susceptibility Testing (AST): Guidelines and Best Practices

Measuring the MIC

• Broth dilution method for MIC
• MIC vs. MBC
• E-test (Epsilometer test)

Interpreting the result

• Susceptible, Intermediate, Resistant: what they mean

Detecting specific resistance mechanisms

• MRSA: emergence, types, detection
• D-test for inducible clindamycin resistance
• Beta-lactamase and ESBL classification
• Phenotypic methods for carbapenemase detection
• Carba NP test
• Modified Hodge test

Stewardship

• WHO AWaRe classification of antibiotics

How to Remember

The one line to hold onto: identification tells you who the pathogen is; susceptibility testing tells you what will beat it, and the letter on the report (S/I/R) is only as trustworthy as the method, the inoculum, and the breakpoint behind it.

Method-selection anchor (a single question): "Do I need a number or a category?" A category (S/I/R) means disk diffusion is enough. A number (MIC) means a dilution or gradient method, and you reach for it precisely when the drug has to clear a threshold at a hard-to-reach site: endocarditis, osteomyelitis, meningitis. Say the sentence "deep site needs a number" and the method choice follows.

Key exam facts in one table

Where Students Get Confused

"Susceptible means the drug is strong / Resistant means the drug is weak." No. S/I/R is about this organism versus achievable drug levels, not the drug's inherent power. A "weak" old drug can read Susceptible for the right bug; a powerful drug can read Resistant.

"Intermediate means the lab wasn't sure." No. Intermediate is a defined pharmacological zone, often meaning the drug may still work at maximum dose or where it concentrates (e.g., urine). It is a real category, not a hedge.

"MIC and zone diameter are just two ways of getting the same S/I/R." They are read differently and are not interchangeable numbers: a small zone corresponds to a high MIC (more resistant). The inverse relationship trips students constantly.

"Every isolate needs an MIC." Most routine isolates need only a category, which disk diffusion supplies cheaply. The MIC is reserved for situations where the number changes management.

"CLSI and EUCAST are basically the same." They frequently differ on breakpoints and QC ranges, and EUCAST reframes "I" as "Susceptible, increased exposure." A report is only interpretable if you know which system it follows.

References

• Clinical and Laboratory Standards Institute (CLSI). Performance Standards for Antimicrobial Susceptibility Testing. M100, current edition. Wayne, PA: CLSI.
• European Committee on Antimicrobial Susceptibility Testing (EUCAST). Breakpoint tables for interpretation of MICs and zone diameters, current version. https://www.eucast.org
• Tille PM. Bailey & Scott's Diagnostic Microbiology. 15th ed. St. Louis: Elsevier; 2022.
• Khan ZA, Siddiqui MF, Park S. Current and Emerging Methods of Antibiotic Susceptibility Testing. Diagnostics (Basel). 2019;9(2):49.
• World Health Organization. Basic laboratory procedures in clinical bacteriology. 2nd ed. Geneva: WHO.