Pharmacokinetic/Pharmacodynamic Based Breakpoints of Polymyxin B for Bloodstream Infections Caused by Multidrug-Resistant Gram-Negative Pathogens

Abstract

The latest PK/PD findings have demonstrated negligible efficacy of intravenous polymyxins against pulmonary infections. We investigated pharmacokinetic/pharmacodynamic (PK/PD)-based breakpoints of polymyxin B for bloodstream infections and the rationality of the recent withdrawal of polymyxin susceptibility breakpoints by the CLSI. Polymyxin B pharmacokinetic data were obtained from a phase I clinical trial in healthy Chinese subjects and population pharmacokinetic parameters were employed to determine the exposure of polymyxin B at steady state. MICs of 1,431 recent clinical isolates of Pseudomonas aeruginosa, Acinetobacter baumannii and Klebsiella pneumoniae collected from across China were determined. Monte-Carlo simulations were performed for various dosing regimens (0.42–1.5 mg/kg/12 h via 1 or 2-h infusion). The probability of target attainment, PK/PD breakpoints and cumulative fraction of response were determined for each bacterial species. MIC90 of polymyxin B was 1 mg/L for P. aeruginosa and 0.5 mg/L for A. baumannii and K. pneumoniae. With the recommended polymyxin B dose of 1.5–2.5 mg/kg/day, the PK/PD susceptible breakpoints for P. aeruginosa, A. baumannii and K. pneumoniae were 2, 1 and 1 mg/L respectively for bloodstream infection. For Chinese patients, polymyxin B dosing regimens of 0.75–1.5 mg/kg/12 h for P. aeruginosa and 1–1.5 mg/kg/12 h for A. baumannii and K. pneumoniae were appropriate. Breakpoint determination should consider the antimicrobial PK/PD at infection site and delivery route. The recent withdrawal of polymyxin susceptible breakpoint by CLSI primarily based on poor efficacy against lung infections needs to be reconsidered for bloodstream infections.