Population pharmacokinetics of continuous infusion ceftazidime

Objective: The purpose of this study was to develop and validate a model that predicts clearance and steady-state ceftazidime concentrations during continuous infusion. Design: This was a prospective clinical observational trial. Two models describing drug clearance during the continuous infusion of ceftazidime to infected patients were developed. The first model included inter- and intraindividual variability (IIV) while the second extended the first model by including interoccasional variability (IOV). Setting: This was a study of patients in a US hospital between January and June 1996. Patients and participants: The analysis included 39 patients aged >18 years with infections at various sites. Interventions: Patients received ceftazidime as either a 1000 or 2000mg loading dose followed by a continuous infusion of 1000 to 4000 mg/day. Serum samples were collected under approximate steady-state conditions and ceftazidime concentrations were analysed using high performance liquid chromatography. The models were fitted to the data using a nonlinear mixed effects model as implemented in the NONMEM program. Results: 75 serum concentration measurements were included in the analysis. The routinely available clinical variables bodyweight, age, gender and serum creatinine were found to be statistically independent predictors of ceftazidime clearance. The IIV model was cross validated yielding a mean prediction error (with a 95% confidence interval) of -0.51 mg/L (-2.5 to 1.4 mg/L) and a mean absolute prediction error of 6.5 mg/L (5.3 to 7.8 mg/L). Conclusion: We have developed and validated a model to estimate ceftazidime concentrations during continuous infusion using commonly available clinical information. Additional work is needed to compare outcomes of patients receiving continuous and intermittently administered ceftazidime, and to define the optimal target steady-state ceftazidime concentrations during continuous infusion.