Determining accurate vancomycin MIC values for methicillin-resistant Staphylococcus aureus by the microdilution method.

OBJECTIVES To model the standard broth microdilution method, based on a modified Gompertz function, to obtain accurate vancomycin MIC values for methicillin-resistant Staphylococcus aureus (MRSA). The effect of these MIC values on the vancomycin therapeutic target of AUC(0-24)/MIC ≥ 400 was evaluated. METHODS Three clinical isolates of MRSA with different vancomycin MIC values were used in this model. The optical densities (OD) of each MIC determination were modelled by a non-linear regression method using an F-test. The OD data were adjusted to the Gompertz equation to obtain the MIC values. The mean vancomycin AUC(0-24) obtained with a 30 mg/kg/day dosing schedule was calculated using a Monte Carlo simulation over 5000 subjects, using the pharmacokinetic data obtained in vancomycin-treated patients in our hospital. RESULTS Although the MIC values obtained with this model were lower than those of the diffusion method (Etest) in all three cases, this did not affect the AUC(0-24)/MIC ratio for the strains with MICs of 1 mg/L by Etest. However, in those strains with MIC values >1 mg/L, the confidence intervals obtained for this ratio included values <400. CONCLUSIONS The inherent variability of the broth microdilution method could explain the differences in the clinical outcome in MRSA-infected patients treated with vancomycin, mainly in those due to strains with MIC values of 1.5-2 mg/L by Etest, because the corresponding MIC values would range from 0.84 to 1.52 mg/L by the microdilution method, which could affect the therapeutic target.