Population pharmacokinetic modelling of total and unbound cefazolin plasma concentrations as a guide for dosing in preterm and term neonates

Cefazolin is frequently administered for antimicrobial prophylaxis and treatment of infections. In neonates, pharmacokinetic observations are limited and dosing regimens variable. The aim of this study was to describe the pharmacokinetics of cefazolin in neonates based on total and unbound concentrations to optimize cefazolin dosing. Thirty-six neonates [median birth body weight 2720 (range 5404200) g, current body weight (cBW) 2755 (8304200) g and postnatal age (PNA) 9 (130) days] receiving intravenous cefazolin (50 mg/kg/8 h) were included. Based on 119 total and unbound plasma concentrations, a population pharmacokinetic analysis with a covariate analysis was performed. Monte Carlo simulations were performed aiming for unbound concentrations above an MIC of 8 mg/L (60 of the time) in all patients. A one-compartment pharmacokinetic model was developed in which total and unbound concentrations were linked by maximum protein binding (B-max) of 136 mg/L and a dissociation constant (K-D) for cefazolin protein binding of 46.5 mg/L. cBW was identified as covariate for volume of distribution (V), bBW and PNA for clearance and albumin plasma concentration for B-max, explaining 50, 58 and 41 of inter-individual variability in V, clearance and B-max, respectively. Based on Monte Carlo simulations, a body weight- and PNA-adapted dosing regimen that resulted in similar exposure across different weight and age groups was proposed. A neonatal pharmacokinetic model taking into account total and unbound cefazolin concentrations with saturable plasma protein binding was identified. As cBW and PNA were the most important covariates, these may be used for individualized dosing in neonates.