MIC Doubling Dilution Uncertainty Calculator

MICs are powers of two because the test is a doubling dilution series, so a reported MIC is a bracket rather than a value. This turns that bracket into the range of PK/PD ratios it actually permits.

Last updated 8 Oct 2026

PK/PD ratio across the MIC bracket

MIC bracket → ratio range

Exposure (the numerator of your ratio)Whatever sits above the MIC in the ratio you are computing: a 24-hour AUC in mg·h/L, or a peak concentration in mg/L. The arithmetic is the same either way because only the MIC is moving.

Reported MIC (mg/L)Exactly as the report states it. The value will be a power of two (or a power of two divided by a power of two) because that is how the dilution series is laid out: 0.125, 0.25, 0.5, 1, 2, 4, 8, 16. A reported 1 mg/L means growth was inhibited at 1 and not at 0.5.

How much MIC uncertainty to allowOne dilution is the accepted reproducibility: ISO 20776-2 allows a deviation of one dilution from the mode in 95% of cases, and CLSI M07-A9 gives plus or minus one log2 dilution as the acceptable variation for broth microdilution. Two dilutions is a reasonable allowance when the result came from a different laboratory or a different method.

250lowest ratio this MIC permitsExample

exposure 500, reported MIC 1 mg/L, one doubling dilution either way

Formula, and why the MIC is a bracket

lowest ratio = exposure / (MIC × 2ⁿ) · highest ratio = exposure / (MIC / 2ⁿ) · fold range = 4ⁿ

n: the number of doubling dilutions allowed either way. One is the accepted reproducibility; two is reasonable across laboratories or methods
why powers of two: broth microdilution tests a two-fold dilution series, so the only concentrations tested are 0.125, 0.25, 0.5, 1, 2, 4, 8, 16 and so on. EUCAST’s reading guide defines the MIC as the lowest concentration that completely inhibits growth, which means the reported value is a well on that ladder and the true MIC lies between it and the well below
the fold range: moving the MIC n dilutions down multiplies the ratio by 2ⁿ and n dilutions up divides it by 2ⁿ, so the span is 4ⁿ — exactly 4 at one dilution and 16 at two. It does not depend on the exposure or on the MIC, only on n
the standards: ISO 20776-2’s acceptance criterion, as Mouton and colleagues cite it, allows a deviation of one dilution from the mode in 95% of cases, and CLSI M07-A9, as Brusamarello and colleagues cite it, gives plus or minus one log2 dilution as the acceptable variation for broth microdilution. Both are cited rather than reproduced
what it does NOT cover: this is the MEASUREMENT’s resolution only. It says nothing about the error in the exposure — the clearance estimate, the sampling times, the assay — nor about between-patient pharmacokinetic variability, which is usually larger still. The true uncertainty on a PK/PD ratio is wider than this page shows

Worked example

exposure 500, reported MIC 1 mg/L, one doubling dilution either way
One dilution up puts the MIC at 2 mg/L, so the lowest ratio is 500 ÷ 2 = 250
The point estimate is 500 ÷ 1 = 500
One dilution down puts the MIC at 0.5 mg/L, so the highest ratio is 500 ÷ 0.5 = 1,000
The span is 1,000 ÷ 250 = fourfold, and it is always exactly fourfold at one dilution whatever the exposure or the MIC
Set it to two dilutions and the range becomes 125 to 2,000 — sixteenfold
For context: if that exposure were a vancomycin AUC24 in mg·h/L, the 2020 consensus window of 400 to 600 sits inside the one-dilution bracket with room to spare on both sides, so the measurement alone cannot say which side of the window the patient is on

What one and two dilutions do to any ratio

The span is 4 to the power of the number of dilutions, and it is independent of both the exposure and the MIC. That independence is the useful part: whatever ratio you are computing and whatever the organism, the measurement alone puts a fourfold band around the answer.
Dilutions allowed	MIC multiplier range	Ratio multiplier range	Fold span
One	0.5× to 2×	0.5× to 2×	4×
Two	0.25× to 4×	0.25× to 4×	16×
Three	0.125× to 8×	0.125× to 8×	64×

The same point, drug by drug, at the page’s defaults

Each middle column crosses or straddles its published target. The fT>MIC row behaves differently from the other three, because the relationship is logarithmic rather than linear in the MIC: there, one dilution costs exactly one half-life of time above the MIC, computed on the time above MIC calculator.
Ratio	Point estimate	One dilution either way	Published target
Vancomycin AUC24/MIC, AUC24 = 500, MIC 1 mg/L	500	250 to 1,000	400-600 (2020 consensus guideline, MIC by broth microdilution)
Gentamicin Cmax/MIC, peak 20 mg/L, MIC 2 mg/L	10.0	5.0 to 20.0	8-10 (Moore 1987, Kashuba 1999, via Bland 2018)
Ciprofloxacin fAUC/MIC, fAUC 30, MIC 0.5 mg/L	60	30 to 120	67.4 for a 1-log10 kill against Enterobacterales (EUCAST v2.0)
Beta-lactam fT>MIC, half-life 1 h, interval 8 h, peak/MIC 10	41.5%	29.0% to 54.0%	40-70% for a 1-log10 kill (Berry and Kuti 2022)

A bracket, not a value, and what that costs

Broth microdilution tests a two-fold dilution series, so the only concentrations ever examined are 0.125, 0.25, 0.5, 1, 2, 4, 8, 16 mg/L and so on up the ladder. EUCAST’s reading guide defines the MIC as the lowest concentration that completely inhibits visible growth, which means a reported MIC of 1 mg/L is a statement about two wells: growth was inhibited at 1 and was not inhibited at 0.5. The true minimum inhibitory concentration lies somewhere in that bracket, and the test has no way to say where. That is not a defect of the method — a twofold series brackets an endpoint closely enough to be useful while remaining practical across a microtitre plate, which is why the same convention is used for antibody titres. It is a limit on what the number can mean.

On top of the series resolution sits measurement variation. Mouton and colleagues quantified it in 2018: a log2 standard deviation of about 0.3 to 0.5 of a doubling dilution on repeat testing within a single laboratory, rising to about 0.5 to 1 dilution or more between laboratories, against an ISO 20776-2 acceptance criterion that allows a deviation of one dilution from the mode in 95% of cases. CLSI M07-A9, as cited by Brusamarello and colleagues, likewise gives plus or minus one log2 dilution as the acceptable variation for the method. So a repeat test of the same isolate in a different laboratory returning a result one dilution away is not a discrepancy; it is the method working as specified.

The arithmetic consequence is unavoidable and is what this page computes. Any ratio with the MIC in its denominator doubles when the MIC halves and halves when the MIC doubles, so one dilution either way puts a fourfold band around the answer and two dilutions a sixteenfold band. The span is exactly 4 to the power of the number of dilutions and does not depend on the exposure or the MIC at all. At a vancomycin AUC of 500 mg·h/L and a reported MIC of 1 mg/L the ratio is anywhere from 250 to 1,000 — a band that contains the whole of the 400 to 600 window the 2020 consensus guideline recommends, with room on both sides. The same guideline’s own practice concedes the point: IDSA guidance, as Brusamarello and colleagues report it, treats an AUC/MIC of 200 to 400 against a gradient-strip MIC as roughly equivalent to 400 to 600 against a broth microdilution MIC, which is one doubling dilution written into a recommendation because the two methods systematically differ by about that much.

Mouton and his co-authors — six people who between them set the European breakpoints — concluded from this that using an individual MIC value to modify a dosing regimen is not justified. That is a strong claim and it sits awkwardly beside guidelines asking for individualised exposure ratios, which is precisely why a reader of either deserves to see the other. The constructive reading is that a PK/PD ratio is worth computing as an order of magnitude rather than to three significant figures, that a change of one dilution on repeat testing should not drive a change of regimen, and that the uncertainty shown here is a floor: it covers only the MIC, not the clearance estimate, the sampling times, the assay or the between-patient variability that population pharmacokinetic models exist to describe. The pipetting side of a dilution series, with its own compounding error and carryover, is covered by the serial dilution calculator. Almost every PK/PD target in use was derived in animal infection models and in-vitro systems — most often the neutropenic murine thigh model — and the human outcome data behind them are thin: the targets are mostly from animal models and in-vitro systems with limited human outcome validation. This page does not hold a breakpoint table. EUCAST and CLSI publish them, they are revised two or three times a year, and the laboratory that issued your report has already applied its own version. Take the MIC and the interpretive category from the report; this page does the arithmetic that sits on top of them.

Frequently asked questions

Why are MICs always powers of two?

Because susceptibility testing uses a two-fold dilution series, so the only concentrations tested are 0.125, 0.25, 0.5, 1, 2, 4, 8, 16 mg/L and upwards. EUCAST’s reading guide defines the MIC as the lowest concentration that completely inhibits growth, so the reported value is one well on that ladder rather than a continuous measurement.

How reproducible is an MIC?

Mouton and colleagues report a log2 standard deviation of about 0.3 to 0.5 of a doubling dilution on repeat testing within one laboratory, rising to about 0.5 to 1 dilution or more between laboratories. ISO 20776-2’s acceptance criterion allows a deviation of one dilution from the mode in 95% of cases, and CLSI M07-A9 gives plus or minus one log2 dilution as the acceptable variation for broth microdilution.

How much does one doubling dilution change a PK/PD ratio?

By a factor of two in each direction, so the ratio spans a fourfold range. At an exposure of 500 and a reported MIC of 1 mg/L the ratio is anywhere from 250 to 1,000. Two dilutions span sixteenfold. The span is 4 to the power of the number of dilutions and does not depend on the exposure or the MIC.

Does that mean PK/PD ratios are useless?

It means they are order-of-magnitude quantities rather than precise ones. Mouton and colleagues go further and conclude that using an individual MIC value to modify a dosing regimen is not justified, which is a stronger position than most guidelines take. The practical reading: a ratio quoted to three significant figures is false precision, and a one-dilution change on repeat testing is not a change in the patient.

Why is the headline the lowest ratio rather than the point estimate?

Because the point estimate is the number a reader already has, and the lower bound is the one they have not seen. Both are shown. Leading with the lower bound is a presentational choice about which figure deserves attention, not a recommendation to assume the worst case.

Is this all the uncertainty there is?

No, it is a floor. This page covers the MIC’s measurement resolution only. The exposure in the numerator carries the error of whatever clearance estimate, sampling times and assay produced it, and between-patient pharmacokinetic variability is usually larger than either. Published target-attainment work exists precisely because the combined uncertainty is wide.

Related calculators

References

Mouton JW, Muller AE, Canton R, Giske CG, Kahlmeter G, Turnidge J. MIC-based dose adjustment: facts and fables. J Antimicrob Chemother. 2018;73(3):564-568. Source for the log2 variability figures, the ISO 20776-2 criterion and the stated conclusion that individual MIC-based dose adjustment is not justified.
European Committee on Antimicrobial Susceptibility Testing. EUCAST reading guide for broth microdilution, version 4.0, January 2022. Source for the definition of the MIC used here.
Brusamarello C, Daley AJ, Zhu X, Landersdorfer C, Gwee A. How important are MIC determination methods when targeting vancomycin levels in patients with Staphylococcus aureus infections? J Antimicrob Chemother. 2021;76(6):1641-1643. Source for the CLSI M07-A9 one-dilution criterion and the IDSA method equivalence quoted here.
Rybak MJ, Le J, Lodise TP, et al. Therapeutic monitoring of vancomycin for serious methicillin-resistant Staphylococcus aureus infections: a revised consensus guideline and review of ASHP, IDSA, PIDS and SIDP. Am J Health-Syst Pharm. 2020;77(11):835-864.
European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters, version 16.1, 2026, stated valid 24 June to 31 December 2026. https://www.eucast.org. Cited rather than reproduced.
Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 35th ed. CLSI supplement M100-Ed35. Wayne, PA: CLSI; January 2025. A commercial standard whose front matter requires express written consent for any reproduction; cited here, never reproduced or paraphrased.

Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/