Pharmacodynamic modeling of ciprofloxacin resistance in Staphylococcus aureus.

摘要

A pharmacodynamic model that describes and predicts the evolution of resistance in Staphylococcus aureus populations exposed to ciprofloxacin was developed using pharmacokinetic, subpopulation, and resistance mechanism data obtained from hollow-fiber in vitro system experiments. The model was designed for two populations of bacteria with different susceptibility to ciprofloxacin and incorporated unique growth and killing rate constants for the ciprofloxacin-susceptible and -resistant populations. Modeling was performed using viable count data obtained from in vitro system experiments versus two ciprofloxacin-susceptible (MIC = 0.5 μg/ml) Staphylococcus aureus strains (MRSA 8043 and MRSA 8282) and the most resistant grlA mutants of MRSA 8043 (MRSA 8043C0–1; MIC = 4 μg/ml) and MRSA 8282 (MRSA 8282C0–1; MIC = 2 μg/ml) detected in the starting cultures. Log-phase cultures (7.08 ± 0.23 log ₁₀ CFU/ml) were exposed to a series of clinical and experimental ciprofloxacin pharmacokinetic profiles (t_1/2 = 4.22 ± 0.11 h) for 96 hours. Nonlinear regression was used to simultaneously fit pharmacodynamic model equations to viable count-time profiles for each parent strain and corresponding grlA mutant. A good model fit was evident by the strong correlation (r = 0.928 to 0.998) between model-predicted and observed viable counts, and the high precision (coefficients of variation 11.9%) of all parameter estimates except R₀, the size of the resistant subpopulations in the starting cultures. The model successfully predicted that grlA mutants of MRSA 8043 and MRSA 8282 would not appear when a low inoculum culture (∼5 log₁₀ CFU/ml) was exposed to clinically-relevant ciprofloxacin concentrations or when a high inoculum culture (∼7 log₁₀ CFU/ml) was exposed to transient high initial concentrations from a simulated experimental dosage regimen. The model, however, did not predict or explain the origin of highly resistant variants of MRSA 8043 and MRSA 8282 that appeared and became the predominant subpopulation in some experiment. Additional refinement and validation of the pharmacodynamic model is warranted.