Optimization of Cefotaxime Treatment in Intensive Care Patients: Impact of Inter- and Intra-patient Variability on Pharmacokinetics and Target Attainment.

BACKGROUND: Cefotaxime is a beta-lactam antibiotic commonly used in critically ill patients. In this population, pathophysiological features induce high inter- and intra-individual pharmacokinetic variability that can result in supra or sub-therapeutic concentrations of cefotaxime. However, factors leading to variability during the treatment course remain poorly described.

OBJECTIVE: The objective of the present study was to characterize the pharmacokinetics of cefotaxime and associated inter- and intra-individual variability in order to optimize individual dosing regimens for adult intensive care patients. Simulations based on the population pharmacokinetic model characterized the impact of these covariates on pharmacokinetic/pharmacodynamic target attainment.

METHODS: Adult intensive care patients were included in a prospective observational study. Cefotaxime concentrations were measured 30 min after the loading dose, during the continuous infusion at days 1, 4, and 7 after drug initiation, and 4 h after the end of infusion. Pharmacokinetic analyses were performed using a parametric population approach. The impact of covariates was determined by combining a baseline covariate model with a time-varying covariate model. Monte Carlo simulations were performed to evaluate the probability of target attainment for different dosing regimens, with a pharmacokinetic/pharmacodynamic target of 100% of the time with cefotaxime free concentration over the minimal inhibitory concentration.

RESULTS: The study included 76 intensive care patients (251 cefotaxime plasma concentrations). A one-compartment model with linear elimination and proportional residual error best described the data. Serum albumin concentration, age, sex, and time-varying serum creatinine concentration were identified as significant covariates. Simulations based on the final model showed that the recommended dosing regimen (2-g loading dose followed by 6 g per day) only achieves pharmacokinetic/pharmacodynamic targets for a minimal inhibitory concentration ≤1 mg/L. As a consequence of serum creatinine concentration intra-individual variability, observed probability of target attainment decreased with time: 59.3, 56.4, and 50.7%, respectively at days 1, 4, and 7, for patients with a baseline serum creatinine concentration between 100 and 150 µmol/L. For minimum inhibitory concentration values above 1 mg/L, an initial dosing regimen of a 2-g loading dose followed by 10 g per day could be appropriate.

CONCLUSIONS: A population pharmacokinetic model of cefotaxime was developed to determine both inter- and intra-individual variability in intensive care patients and was applied to evaluate the impact of various dosing regimens on pharmacokinetic/pharmacodynamic target attainment. Serum creatinine concentration explained a significant part of the observed inter- and intra-individual pharmacokinetic variability and a biomarker model would be helpful to anticipate serum creatinine concentration evolution and improve the prediction.