Optimizing ciprofloxacin dosing in intensive care unit patients through the use of population pharmacokinetic-pharmacodynamic analysis and Monte Carlo simulations.

OBJECTIVES: To explore different ciprofloxacin dosage regimens for the treatment of intensive care unit (ICU) patients with respect to clinical outcome and the development of bacterial resistance for the major Gram-negative pathogens.
METHODS: A population pharmacokinetic model was first developed on ciprofloxacin serum concentrations obtained in 102 ICU patients. Then, based on this model, pharmacokinetic-pharmacodynamic Monte Carlo simulations (MCSs) were carried out to explore the appropriateness of different ciprofloxacin dosage regimens in ICU patients. The defined targets were free AUC(24)/MIC ≥90 h (as a predictor of clinical outcome) and T(MSW) ≤20% (as a predictor of selecting resistance), where T(MSW) is the time spent within the mutant selection window over 24 h. Two simulation trials were conducted: Trial 1 took into account the whole MIC distribution for each causative pathogen in line with empirical antibiotherapy; Trial 2 used MIC breakpoints given by the Antibiogram Committee of the French Microbiology Society in order to treat the 'worst-case' scenario.
RESULTS: Trial 1 showed that for Pseudomonas aeruginosa and Acinetobacter baumannii, the common dosage regimens of 400 mg twice or three times a day did not achieve the desired target attainment rates (TARs) with respect to T(MSW), while suboptimal TARs were found for AUC(24)/MIC. Trial 2 showed that ≤ 18% of patients reached the target of T(MSW) ≤ 20% for MIC breakpoints of 0.5 and 1 mg/L, regardless of the administered dose.
CONCLUSIONS: Based on the mutant selection window concept, our simulations truly question the use of ciprofloxacin for the treatment of P. aeruginosa and A. baumannii infections in ICU patients due to the potential for developing resistance.