ANTIBIOTIC DISK TEST MICs
BIOMIC V3 provides MICs from the continuous agar gradient for disk diffusion tests for most common drug-organism combinations. Reporting MICs from BIOMIC V3 is optional.
A highly predictable logarithmic gradient is formed by diffusion of antibiotic from the disk. At the time of visible growth, the concentration at the edge of the zone is, by definition, the minimum inhibitory concentration (MIC) -- i.e. there is no growth at the higher concentrations within the zone. Shown is a typical zone. The left scale shows the zone diameter and calculated MIC based on regression analysis. The right scale shows the equivalent doubling MIC's, evenly spaced, as diffusion is a logarithmic function. Note the similarity of the scale to an Etest strip (bioMerieux). Acceptable variation of +/- 1 doubling dilution is about 7mm in this example. Diffusion is time dependent, thus organisms with markedly different growth rates require a different gradient, as provided in BIOMIC. Both BIOMIC and Etest scales are calibrated using the reference CLSI broth dilution MIC method. BIOMIC MICs were determined by the US-FDA via 510(k) to be substantially equivalent to reference broth dilution MICs. |
Accuracy and Precision of Disk Versus Dilution Test MIC Results
Correlation coefficients between zone diameters (produced by continuous gradient disk-diffusion method) and MICs (by discontinuous gradient dilution methods) are published for most antimicrobial agents early in development. Coefficients clearly demonstrate the direct correlation between diffusion and a painstakingly executed dilution method performed without large gaps between dilutions. D'Amato and Thornsberry JCM 1985 showed a 96% correlation between BIOMIC and microdilution MICs. However, microdilution panels often make room for more drugs by reducing the number of concentrations tested, creating broad gaps in dilutions; the quality of panel results is difficult or impossible to control in-laboratory. Broth media often does not support fastidious organism growth as well as agar; this may result in trailing end-points that are difficult to read and highly variable. While dilution methods measure susceptibility on a discontinuous scale, interrupted by dilution intervals, the disk method measures susceptibility precisely on a continuous scale to millimeters of growth inhibition.
MICs in BIOMIC are determined by direct regression analysis; this is a correlation or calibration of the concentration of each antimicrobial agent associated with each quantitative millimeter measurement by organism species. These regression lines are very carefully determined by comparative studies correlating dilution MIC results with agar diffusion results. A major factor in the accuracy of these results is, that organisms with relatively common growth rates need to be grouped together in regression analysis, (note CLSI drug-organisms groupings on published interpretive charts) because zone formation in all agar diffusion assays is time dependent. Agar diffusion test results, as with all quantitative methods, must be measured quantitatively, and quality controlled regularly, in order to produce accurate quantitative MIC test results. Breakpoint and MIC equivalents published by CLSI or other groups can not be accurately used for this purpose.
The following example may help to clarify the issues of precision in reading endpoints and subsequent accuracy of continuous agar gradient versus broth dilution MICs.
Acceptable MIC dilution accuracy is +/- 1 dilution. Thus an MIC of 16 mcg/ml can be from 8 - 32 mcg/ml. MICs > 4.1 are read/reported as 8 mcg/ml. Therefore a "True MIC of 16 is acceptably any value from 4.1 to 32 mcg/ml. This is not to say that dilution method MICs are not accurate, however, when continuous agar gradient MICs are well calibrated and segmented by species, they are more precise (more endpoints on the results scale). The bottom line is that both agar gradient and dilution provide accurate MICs, for a biological system. It is also important to consider that MICs estimate "relative resistance" at one inoculum in an artificial medium. Any MIC concentration is difficult to correlate accurately with achievable levels in vivo, except on a relative/gross basis, as adult doses usually apply to body weights of 80 to 320 lb (4-fold), and tissue levels will vary accordingly.
BIOMIC MIC Study References
- Evaluation: BIOMIC Antimicrobial Susceptibility System; R. D'Amato, C. Thornsberry. J. Clin. Micro. 1985, 22,793. MICs of 511 isolates of Enterobacteriaceae, nonfermenters, enterococci & staph, 10,085 organism-drug combinations.
- Early/Overnight BIOMIC MICs. T.Lawrence. D.Amsterdam. Abs.C137, 1991, ASM 162 Enterobacteriaceae and 56 nonfermenters, were tested against 8 antibiotics.
- Comparison: BIOMIC & Panel Systems for... Staph-Testing. R.L.Sautter et al., Abstract 1991, ASM. 93 S.aureus isolates were tested using MICs and S-R interpretation with 28 antibiotics (6831 combinations).
- Correlation: BIOMIC & Microdilution MICs. R.Morfin et al., Abstract C138, 1991,ASM.
- Comparison: BIOMIC & MicroScan MICs. S. Hodowanec et al. 1993. Abs.C183 ASM. 400 gram negative bacilli, 12,000 drug combinations.
- Evaluation: BIOMIC with Staphlococci. P. Rohner et al. 1993. Abs.C112 ASM. 102 S aureus and 63 coag-neg Staph were tested.
- Evaluation: BIOMIC, E-Test & microbroth-dilution. J.Daly et al. 1993. Abstract C97 ASM. 24 hour MICs on 10 antibiotics and 101 beta-hem-streptococci.
- Comparison: BIOMIC and E-Test MICs. J.McLaughlin et al., Abstract ASM 1994. With 852 H.influenzae - antibiotic combinations (6 drugs)
- Comparison: BIOMIC and E-Test MICs: D. DeWeese et al 1995, Abstract ASM 1995. With 166 routine isolates, 7 species, and 4 drugs, 638 drug combinations.
- Comparison: BIOMIC and Vitek: Berke & Tierno 1996. JCM 34:(8) 1980-4. 2948 organism-drug combinations on gram positive and gram negative bacteria.
- Comparison: BIOMIC vs Agar Dilution, Microdilution & E-Test: Jacobs, et al 1997 Unpublished, 183 isolates of strep pneumo vs penicillin; using penicillin, oxacillin & methicillin disks.