Calibration of disk diffusion antibiotic susceptibility testing: species-related trovafloxacin interpretive zone breakpoints and selection of disk potency.

Abstract

International comparisons of antibiotic susceptibility require the use of common minimum inhibitory concentration (MIC) limits. Disk diffusion test results are not directly suitable for such comparisons, since different standards are often used and zone breakpoints issued might reflect different MIC limits. We have used single strain regression analysis (SRA) for the calibration of the disk test, both according to species and individual laboratory, and for quality control of trovafloxacin disk diffusion tests in 5 laboratories in Sweden. Preliminary controls using histogram analysis including subtraction histograms of reference strains revealed marked differences between different laboratories. SRA was performed on 4 reference strains, S. aureus, E. faecalis, E. coli and P. aeruginosa, using disks containing 1, 3, 10, 30 and 100 microg trovafloxacin. The results using SRA showed a difference between laboratories using Biodisk PDM medium, which produced smaller zones, and those using Oxoid IsoSensitest. Species-related regression lines for laboratories using either medium were calculated and corresponding interpretive zone breakpoints determined for MIC limits. Rational criteria for the selection of a suitable disk content of an antibiotic were also defined and applied to trovafloxacin. The 10 microg disk selected by NCCLS (National Committee for Clinical Laboratory Standards) proved optimal.