Use of Monte Carlo simulation to evaluate the development of vancomycin resistance in meticillin-resistant Staphylococcus aureus

Monte Carlo simulations were performed for various vancomycin dosage regimens to evaluate the potential for development of vancomycin resistance in meticillin-resistant Staphylococcus aureus (MRSA). When the target of free AUC24/MIC ≥ 200 was considered (where AUC24 is the area under the drug concentration-time curve in a 24-h interval and MIC is the minimum inhibitory concentration), a standard dose regimen (1000 mg every 12 h) yielded unacceptable simulated outcomes in patients with normal renal function; in particular, the probability of target attainment (PTA) was only 30.5% at an MIC of 1 mg/L. For the same dosage regimens and the mutant prevention concentration (MPC)-based pharmacokinetic target (total AUC24/MPC > 15), the cumulative fraction of response exceeded 80% for all renal function strata; low values of PTA (<80%) were obtained only for isolates with MPCs of ≥22.4 mg/L, which consisted of all 21 strains of heterogeneous vancomycin-intermediate S. aureus (hVISA) and a handful of non-hVISA strains with MICs of 2 mg/L (32%; 16/50). Based on the current status of vancomycin resistance, we conclude that total AUC24/MPC > 15, derived from in vivo experiments, is more suitable to predict the development of vancomycin resistance. In clinical practice, individualised vancomycin therapy should be considered to minimise selection of resistance mutations.