The use of Monte Carlo simulation to examine pharmacodynamic variance of drugs: fluoroquinolone pharmacodynamics against Streptococcus pneumoniae

Background: For fluoroquinolones, AUC:MIC ratios correlate with maximal bac terial eradication in in vitro models of infection and favorable cure rates in humans with respiratory tract infection. Inter-subject pharmacokinetic and MIC variability may impact the probability of attaining optimal AUC:MIC ratios and hence favorable clinical outcome. Methods. Monte Carlo simulati on was utilized to estimate the probability of attaining AUC:MIC ratios of 30, 40, 50, 60, 70, 80, 90, 100, 110 and 120 using AUC values from patients treated with either gatifloxacin or levofloxacin and microbiologic activit y against S. pneumoniae observed in 1997 SENTRY Antimicrobial Surveillance Program. Results: The probability curves for 5000 patient simulations were plotted. The median AUC:MIC ratios were 120 for gatifloxacin and 50.5 for l evofloxacin. The probability of attaining AUC:MIC ratios of 30, 50, 70 and 100 for gatifloxacin were 94%, 86%, 78% and 62%, and for levofloxacin were 80%, 51%, 31% and 17%, respectively. Conclusion: Gatifloxacin has a higher probability of achieving target AUC:MIC ratios than levofloxacin. Monte Car lo simulation, using patient-based AUC and MIC distributions, may have impl ications for selection of optimal antibiotics for the empiric treatment of infections. Moreover, Monte Carlo simulation may have utility in the determ ination of MIC breakpoints. (C) 2000 Elsevier Science Inc. All rights reser ved.