Loading dose of ceftazidime need to be increased in critically ill patients: a retrospective study to evaluate recommended loading dose with pharmacokinetic modeling

Abstract

Abstract Background In order to rapidly achieve target concentrations and bactericidal efficacy, the administration of a loading dose (LD) is recommended before starting ceftazidime continuous infusion. However, the adequacy of the 2g-LD usually administered should be investigated considering the special pharmacokinetic characteristics of critically ill patients. Materials PK dataset for model development and external validation included patients hospitalized in 6 intensive care units (ICU) in the Saint-Etienne region (France) and in Paris, with ceftazidime continuous infusion and at least one measurement of plasma concentration [IRBN992021/CHUSTE]. Data were analysed with MONOLIX and R softwares. A review of the literature was performed to search for PK models developed in ICU patients, to compare our results with existing models. A simulation of the LD needed to reach a target concentration of 60mg/L was performed with all models. Results Ceftazidime was well described by a one-compartment model with allometrically scaled lognormalized e-glomerular filtration rate as a covariate of clearance, using a dataset of 86 patients/223 samples. Typical ceftazidime clearance and volume of distribution were 4.45L/h and 88L, respectively. The predicted individual ceftazidime concentrations were significantly lower at 24 ± 4hours than at 48 ± 4hours. Of the 8 publications of pharmacokinetics models developed in ICU populations, median volume of distribution was 37.2L. The simulated LD to achieve 60mg/L in 80% of the patients from the models found in the literature was higher than 2g in all but one study. Median LD was 4.9g. Conclusions Standard LD results in delay in achieving target ceftazidime concentration in ICU patients.