Population pharmacokinetics and pharmacodynamics of fosfomycin in non-critically ill patients with bacteremic urinary infection caused by multidrug-resistant Escherichia coli.

OBJECTIVES: To describe the population pharmacokinetics of fosfomycin for patients with bacteraemic urinary tract infection (BUTI). The analysis identified optimal regimens on the basis of pharmacodynamic targets and assessed the adequacy of Clinical and Laboratory Standards Institute (CLSI) and European Committee on Antimicrobial Susceptibility Testing (EUCAST) susceptibility breakpoints for Escherichia coli.

METHODS: Data of 16 patients with BUTI caused by multidrug-resistant E. coli (FOREST clinical trial) received intravenous fosfomycin (4 g every 6 hours) were analysed. A population pharmacokinetic analysis was performed, and Monte Carlo simulations were undertaken using 4 g every 6 hours and 8 g every 8 hours. The probability of pharmacodynamic target attainment was assessed using pharmacodynamic targets for E. coli for static effect, 1-log drop in bacterial burden and resistance suppression.

RESULTS: Sixty-four plasma samples were collected over a single dosing interval (day 2 or 3 after starting fosfomycin treatment). Fosfomycin concentrations were highly variable. Pharmacodynamic target attainment analysis showed mild improvement by increasing fosfomycin dosing (4 g every 6 hours vs. every 8 hours). These dosages showed success for decreasing 1-log bacterial burden in 89% to 96% (EUCAST breakpoints) and 33% to 54% (CLSI breakpoints) of patients, but they were unable to reach bacterial resistance suppression targets.

CONCLUSIONS: Fosfomycin concentrations are highly variable-a fact partially explained by renal impairment. The present work supports the use of 4 g every 6 hours as an effective regimen for the treatment of non-critically ill patients with BUTI caused by multidrug-resistant E. coli, as higher dosages might increase toxicity but may not significantly increase efficacy. The current information may suggest that fosfomycin susceptibility breakpoints need to be reappraised.