Population pharmacokinetics and dosing simulations of cefepime in septic shock patients receiving continuous renal replacement therapy

• We investigated the influence of dialysis settings on cefepime concentrations.
• We found important between-patient variability in pharmacokinetic parameters.
• The optimal dose is dependent on the ultrafiltration flow rate (UFR).
• A dose of 2 g q8h is needed for high UFRs (≥1500 mL/h).
• One gram q8h is optimal for low UFRs (≤1000 mL/h).

The aim of this study was to describe the population pharmacokinetics of cefepime in septic shock patients requiring continuous renal replacement therapy and to determine whether current or alternative dosing regimens can achieve PK/PD targets. In this observational PK study, 62 samples from 13 patients were analysed using non-linear mixed-effects modelling. Different dosing regimens were evaluated using Monte Carlo simulations with ultrafiltration flow rates (UFRs) of 1000, 1500 and 2000 mL/h. The probability of target attainment was calculated against a conservative (60% T>MIC) and a higher PK/PD target (100% T>MIC) against an MIC of 8 mg/L, the clinical susceptibility breakpoint for Pseudomonas aeruginosa. A one-compartment model with between-subject variability (BSV) on clearance and volume of distribution (Vd) described the data adequately. UFR was supported as a covariate on both parameters. Typical values for clearance and Vd were 4.4 L/h (BSV 37%) and 40.9 L (BSV 20%), respectively. Dosing simulations showed failure to achieve both a conservative and a higher PK/PD target using a dose of 1 g q12h for patients treated with a high UFR (≥1500 mL/h). The dose of 2 g q8h or 1 g q6h leads to optimal target attainment for high UFR. One gram q8h is optimal for low UFR (≤1000 mL/h). We found important variability in PK parameters. Dosing simulations show that a dose of 2 g q8h or 1 g q6h is needed to ensure rapid achievement of adequate levels if the UFR is ≥1500 mL/h and 1 g q8h for low UFR (≤1000 mL/h).