Methods: Time-concentration curves derived from plasma samples in critically ill patients receiving ciprofloxacin 400mg IV q12h (N=11) or q8h (N=5) or levofloxacin 750mg IV q24h (N=9) were used to calculate individual PK parameters and create population PK models. Monte Carlo simulations were performed to assess the cumulative fraction of response (CFR) to achieve the target pharmacodynamic index (PDI) of AUC:MIC ≥ 125, using gram-negative MIC distributions from the European Committee on Antimicrobial Susceptibility Testing.
Results: The fit of both the ciprofloxacin and levofloxacin population models was improved with the addition of CrCl as a covariate. Despite simulating higher dosing regimens, such as ciprofloxacin 600mg q8h and 800mg q8h and levofloxacin 1125mg q24h and 1500mg q24h, only a single dosing regimen/gram-negative species combination demonstrated a CFR ≥90%. This result was consistent with the finding that the maximum MICs at which individual patients achieved the target PDI were well below the CLSI breakpoints of ciprofloxacin and levofloxacin for Enterobacteriaceae, Pseudomonas, and Acinetobacter of ≤1 mg/mL and ≤2 mg/mL, respectively.
Conclusion: In critically ill trauma, surgical, and burn patients, standard dosing regimens of ciprofloxacin and levofloxacin failed to achieve PDIs sufficient to treat optimally Enterobacteriaceae, Pseudomonas, and Acinetobacter isolates with MICs up to the CLSI breakpoints. When increased doses were simulated, the CFR of all but one dose/species combination remained suboptimal. Individualized dosing guided by therapeutic drug monitoring may be an appropriate next step to improve fluoroquinolone efficacy in these critically ill patients.
T. Schlotman, None