In vitro and in vivo models of infection suggest that the area under the concentration-time curve (AUC): minimum inhibitory concentration (MIC) ratio is the pharmacodynamic parameter that is most predictive of bactericidal activity for the fluoroquinolones. Additionally, this parameter has also been correlated with clinical outcomes in humans with respiratory tract infection. Despite these defined relationships, broad-scale application of these principles in the section of optimal therapy in the clinical arena has been restricted by the imprecise estimates of drug exposures (ie, AUC:MIC ratio) in the infected population. In an effort to best describe the known variability in the pharmacokinetic and susceptibility profile of agents of interest, Monte Carlo simulation has been employed to assess the probability of attaining the desired AUC:MIC ratio. In this report Monte Carlo simulation was used to estimate the probability of attaining various target AUC:MIC ratios using AUC values from patients treated with either gatifloxacin or levofloxacin and the in vitro potency of the compounds as assessed in 881 clinical isolates of Streptococcus pneumoniae isolated from outpatients. Using a simulated patient population of 5,000, the median AUC:MIC ratios were 144 and 50 for gatifloxacin and levofloxacin, respectively. The probability of attaining AUC:MIC ratios of 30, 40, 65, 100, and 125 for gatifloxacin were 99%, 95%, 85%, 68%, and 60%, respectively. For levofloxacin, similar dynamic hit rates were 82%, 61%, 35%, 17%, and 12% over the range of target values. Regardless of the optimal ratio selected, gatifloxacin had a higher probability of achieving the AUC:MIC target than did levofloxacin. Simulations of this type can assist in the decision process surrounding the choice of optimal therapies based on our current understanding of antimicrobial pharmacodynamic principles.
By, Nicolau DP, Ambrose PG