Minimum Inhibitory Concentration Explained: A Guide 2026

The minimum inhibitory concentration (MIC) is the lowest concentration of an antimicrobial agent, such as an antibiotic or plant extract, that stops visible growth of a microorganism after overnight incubation in the laboratory. It is reported in micrograms per milliliter (µg/mL), which is the same as milligrams per liter (mg/L), and it tells you how much of a substance that one strain needs before growth stops.

A single MIC number is easy to quote and easy to misread. It describes what happened in a well of broth on a bench, under one set of conditions, with one strain of microbe. It is not a dose, it is not a safety statement, and it is not a verdict on whether a treatment works in a person.

This guide covers what the number means, how laboratories produce it, how to read it alongside its units and breakpoint, and where a laboratory result stops being useful.

What Is Minimum Inhibitory Concentration?

The minimum inhibitory concentration (MIC) is the lowest concentration of an antimicrobial agent that prevents visible growth of a microorganism after overnight incubation, expressed in µg/mL or mg/L. It is the standard way laboratories describe how susceptible one strain is to one substance.

Two details in that definition carry most of the meaning. The first is visible growth: the endpoint is what a trained eye, or a turbidity reading, picks up in the well. The second is lowest: a lower MIC means less of the substance is needed to reach that endpoint.

Inhibition is not killing

An inhibited culture has stopped multiplying. The cells may still be alive, and a subculture onto antibiotic-free medium will often show regrowth. This is why an MIC describes a bacteriostatic effect, not a bactericidal one, and why a second test is needed when the killing question matters.

Where MIC values come from

Three uses cover almost every result you will meet. Clinical laboratories run susceptibility testing to help clinicians choose an antibiotic. Resistance surveillance programmes run the same tests across many isolates to track which strains are drifting. Research groups, including groups screening plant extracts, run them to rank new substances against reference organisms.

In the ethnobotanical setting the value is usually reported in mg/mL of extract rather than µg/mL of a purified compound, because the extract is a mixture. That single difference changes how the number can be compared with anything else.

What Does an MIC Value Tell You?

A reported MIC only means something once you know the units, the organism, the substance, the method, and the control results. Read those first, then read the number.

What you see reported	What it means in practice
MIC of 0.5 µg/mL	The strain grew at 1 µg/mL and not at 0.5 µg/mL. A twofold dilution series can only report values in steps, so a result is always a doubling dilution, never a precise number.
MIC of 16 µg/mL	Growth was present at 32 µg/mL and absent at 16 µg/mL. Compare this against the breakpoint for that organism and drug before drawing any conclusion.
MIC of 256 µg/mL or higher	Growth persisted across the whole range tested. The true value may sit above the highest concentration in the assay, so it is reported as a lower bound rather than an exact figure.
MIC reported in mg/mL	Usual for a crude plant extract or herbal preparation. It measures the weight of extract added, not the weight of the compound responsible.
No MIC, growth in every well	Either the strain is genuinely resistant, or the assay failed. Check the growth control and the quality control strain before concluding anything.
Zone of inhibition in millimetres	A disk diffusion result, not an MIC. It cannot be converted to a concentration without a calibration curve.

Susceptibility categories and the breakpoint

EUCAST and CLSI publish a clinical breakpoint for each organism group and drug. A breakpoint is a concentration threshold set from clinical outcomes, pharmacokinetics, and dose data, and the reported MIC is compared against it to place the isolate in a susceptible, intermediate, or resistant category.

The category describes drug choice, not biology. It also depends on the organism group: a breakpoint set for Escherichia coli does not transfer to Staphylococcus aureus, and a breakpoint for a meningitis isolate is not the same as one for a non-meningeal infection.

EUCAST and CLSI also disagree on some values. For many agents the EUCAST breakpoint sits one doubling dilution below the CLSI breakpoint, so an identical MIC can fall into two different categories depending on which system the laboratory applied. Always record which guideline and which version the lab used.

MIC50 and MIC90, and why they matter

MIC50 is the concentration that inhibits 50 percent of a tested population, and MIC90 is the concentration that inhibits 90 percent. A single strain gives one MIC; a panel of isolates gives a distribution, and MIC50 and MIC90 summarise where that distribution sits.

Worked example. Test 100 isolates and record the MIC for each. If half the isolates are inhibited by 0.5 µg/mL, that is the MIC50. If 90 of them are inhibited by 4 µg/mL, that is the MIC90. A narrow gap between the two numbers, 0.5 against 1 µg/mL, tells you the population is uniformly susceptible. A wide gap, 0.5 against 32 µg/mL, tells you a subgroup behaves very differently and is worth investigating.

How Is Minimum Inhibitory Concentration Tested?

The workflow is the same shape whichever method is used: prepare the organism, expose it to a graded range of concentrations, incubate, and read for the boundary between growth and no growth. What changes between methods is where the gradient lives.

1. Prepare a pure culture. Grow the strain overnight, then confirm it is a single organism before testing. A contaminated culture produces a falsely low MIC because something in the mixture is sensitive to the agent.
2. Standardise the inoculum. Adjust the suspension to a defined cell density, commonly around 5 × 10⁵ CFU/mL for broth work. Inoculum density is a common source of variation, because a heavier inoculum raises the measured MIC.
3. Prepare the antimicrobial series. Make twofold dilutions, so each step is half the previous one: 64, 32, 16, 8, 4, 2, 1, 0.5 µg/mL. Poorly soluble drugs need specific handling, and some agents such as polymyxins are prepared fresh in glass because they bind to lab plastics.
4. Combine culture and antimicrobial. In broth microdilution this happens in a 96-well microplate. In agar dilution the concentrations are poured into separate plates. With a gradient strip the antibiotic is already laid down the strip.
5. Include controls. A growth control with no antimicrobial must show growth, otherwise the inoculum or the medium is at fault. A sterility control must stay clear. A quality control strain such as E. coli ATCC 25922 is run alongside the test strains to confirm the assay itself was valid.
6. Incubate under defined conditions. Temperature, atmosphere, and incubation time all belong in the report. Overnight, roughly 16 to 24 hours, is the usual window for bacteria read by eye.
7. Read and record the endpoint. Find the lowest concentration with no visible growth and report it in µg/mL or mg/L, together with the method, medium, and quality control result.

How Does the Broth Microdilution Test Work?

Broth microdilution puts the whole dilution series into a single microplate, usually 100 µL per well, and it is the reference method in most standards because the result can be traced to a twofold dilution. Agar dilution is the older alternative, and gradient strips are the quickest to run day to day.

The read is straightforward once the plate is there. A typical row in one column looks like this, with the antimicrobial concentration falling from left to right:

Well	A	B	C	D	E	F	G	H
Antimicrobial µg/mL	64	32	16	8	4	2	1	0.5
Appearance	Clear	Clear	Clear	Clear	Turbid	Turbid	Turbid	Turbid

Growth is present from 4 µg/mL downwards, so the lowest concentration with no visible growth is 8 µg/mL. That is the reported MIC: 8 µg/mL. The same test replicated three times gives the same value, which is why results are usually run in duplicate or triplicate rather than trusted from a single row.

Two read problems come up constantly. Faint growth that sits between clear and turbid is the top complaint among people running these assays, and the usual fix is to read the plate against a black background and to have a second reader confirm the boundary well. Trailing growth, where a thin film clings to the well wall above the liquid line, is treated differently from a button of growth at the bottom, and gradient strips show a similar artefact that readers are trained to ignore.

Automated readers measuring turbidity at OD₆₀₀ help resolve faint wells. EUCAST reading rules treat a well as no growth only when it looks clear to the eye, so an instrument reading is accepted only where that automated method has been validated for the test in question.

Why Are MIC Units and Test Conditions Important?

MIC is reported in micrograms per milliliter, and 1 µg/mL equals 1 mg/L, so the two notations are interchangeable rather than different measurements. What does not convert is a mass of extract: 4 mg/mL of a crude extract is 4,000 µg/mL of extract, and it says nothing about how much of the active compound is in it.

Every value also carries its conditions: the medium, commonly Mueller-Hinton broth or agar, the inoculum density, the incubation temperature and time, and the solvent used to dissolve the antimicrobial. For polymyxins the cation content of the medium changes the answer, which is why cation-adjusted Mueller-Hinton is specified for those agents.

Compare values only when the conditions match. A broth result and an agar result for the same strain are close, usually within one doubling dilution, but they are not interchangeable numbers, and a value reported without its medium and incubation time is an incomplete result.

How Is MIC Different From MBC and Zone Diameter?

Each of these measures a different thing on a different scale. Mixing them up is the most common source of confusion when reading a paper or a report.

Test	What it measures	How it is read	When it is used
MIC	The lowest concentration that stops visible growth	A concentration in µg/mL or mg/L	Routine susceptibility testing, screening, resistance monitoring
Minimum bactericidal concentration, MBC	The lowest concentration that kills rather than only inhibits, found by subculturing clear wells onto antibiotic-free medium	A concentration in µg/mL or mg/L, usually the same as or a few dilutions above the MIC	When killing matters, such as testing for viable persistence after exposure
Disk diffusion zone diameter	The clear zone around an antimicrobial disc, which reflects diffusion as much as potency	A diameter in millimetres, converted to a category through the disc’s own calibration	Routine testing where a microdilution setup is not available
Inhibition percentage	Growth relative to an untreated control, often from a resazurin or turbidity readout	A percentage, sometimes with a dose-response curve	Screening panels where a full series is impractical

The practical rule: MIC and MBC are concentrations, zone diameter is a distance, and inhibition percentage is a ratio. A 22 mm zone cannot be called 22 µg/mL, and a percentage cannot be compared with a breakpoint without the reference used.

What Are the Limitations of MIC Testing?

An MIC is a controlled laboratory measurement with a narrow question: how much agent stops visible growth in this medium, at this inoculum, in this strain. Everything outside that frame is inference.

• Technical variability. Two laboratories testing the same strain often differ by one to two doubling dilutions. That is normal and within essential agreement, so a small gap between your result and a published one is not automatically an error.
• Visual endpoints. Reading no growth versus faint growth is a judgement call, which is why controls, replicate wells, and a second reader are used.
• Laboratory conditions. pH, temperature, media, cation content, inoculum size, and growth phase all move the value. Change one and the number changes.
• Strain differences. Resistance genes, plasmids, and species variation mean two strains of the same species can give very different MICs. A result belongs to a strain, not to a species label.
• Biofilms. Cells in a biofilm grow more slowly and are shielded by the matrix, so a planktonic MIC routinely overstates activity against an attached population.
• Inhibition versus killing. A low MIC can describe an agent that halts growth without removing it.
• The in vitro to in vivo gap. A well contains broth. A patient contains tissue penetration, drug distribution, protein binding, a dosing schedule, and an immune system, none of which the assay models. A low MIC in a plate does not by itself establish that a substance is effective or safe in people, and it says nothing about toxicity, which has to be measured separately.

Reported results are also easy to detach from their context. A value without the guideline version, medium, or quality control result is a number from an unknown experiment, and papers that quote MIC values without them should be read with that in mind.

How Is MIC Interpreted in African Ethnomedicine Research?

Testing a plant extract follows the same logic as testing an antibiotic, with two differences that matter for how the result can be described. The extract is a mixture, so a low MIC in mg/mL is a promising signal rather than a potency figure, and the result applies to the extract in that solvent, at that pH, against that organism.

Three habits make these results more useful. Report the extract concentration in mg/mL and, where possible, the identified compound concentrations separately. Run a quality control strain with each batch so a reader can tell a real inhibition from a failed assay. And test more than one organism, because activity against a single strain rarely generalises to a family of them.

Researchers can also express the result as a population summary. If a study tests 20 isolates and 10 are inhibited at 1 mg/mL and 19 at 8 mg/mL, that gives an MIC50 of 1 mg/mL and an MIC90 of 8 mg/mL, which describes the spread far better than a single figure from one strain.

Where this gets risky is the jump from a plate to a person. A laboratory MIC says nothing about a safe dose, absorption, metabolism, interactions with medicines, or effects during pregnancy, and crude extracts vary between batches in a way a purified drug does not. For any question about using a plant preparation medicinally, talk to a doctor or pharmacist rather than treating a published MIC as a dosing guide. This guide is for research and education, and clinical testing belongs in an accredited laboratory.

Frequently Asked Questions

Conclusion

Start by reading a reported minimum inhibitory concentration as a set rather than a number. The units, the organism and strain, the substance and its concentration basis, the method and medium, and the quality control result all belong to the value, and without them the number cannot be interpreted or compared.

When you run the test yourself, record the conditions as you go and treat a mismatch with a published result of one to two doubling dilutions as normal variation. Anything larger usually points to inoculum density, medium, solubility, or endpoint differences worth checking before the assay is repeated.