Population Pharmacokinetic Modeling of Cefepime, Meropenem, and Piperacillin-Tazobactam in Patients with Cystic Fibrosis.

Abstract

Patients with cystic fibrosis (CF) experience recurrent bacterial pulmonary exacerbations. Management of these infections is increasingly challenging due to decreased antimicrobial susceptibility to beta-lactam antibiotics. The pharmacokinetics of these agents are inadequately characterized in patients with CF. One hundred fifty-five pediatric and adult participants with CF receiving cefepime (n=82), meropenem (n=42), or piperacillin-tazobactam (n=31) were enrolled. Opportunistic blood samples were obtained during hospitalization. Population PK analysis was conducted using nonlinear mixed-effects modeling. Clinical and demographic characteristics were evaluated as potential covariates. Monte Carlo simulations were performed to evaluate probability of target attainment (PTA) for different dosing regimens. Estimated creatinine clearance, and total or lean body weight, affected the pharmacokinetics of cefepime and meropenem. No covariates were identified for piperacillin and tazobactam. In the cefepime group, a 3-h infusion achieved higher PTA than a 0.5-h infusion for all participants. Estimated breakpoints (the respective minimum inhibitory concentration (MIC) up to which ≥90% of patients are predicted to reach a PK/PD target) were two- to four-fold higher in pediatric participants receiving a 3-h vs. 0.5-h infusion. In the meropenem group, increased creatinine clearance led to reduced PTA. In the piperacillin-tazobactam group, total daily dose and mode of administration were principal drivers of PTA. Standard dosing regimens fail to achieve specific MIC targets in patients with CF. Therefore, clinicians should incorporate local antibiograms and PK models to determine optimal dosing. Further PK optimization to account for interindividual differences could be achieved by real-time beta-lactam therapeutic drug monitoring.