Population pharmacokinetics and pharmacodynamics of piperacillin/tazobactam in patients with complicated intra-abdominal infection.

OBJECTIVES We investigated the population pharmacokinetics and pharmacodynamics of piperacillin and tazobactam in hospitalized patients. PATIENTS AND METHODS A multicentre, randomized clinical trial was conducted in hospitalized patients with complicated intra-abdominal infection. Patients received piperacillin/tazobactam administered by either continuous infusion (13.5 g over 24 h, n = 130) or intermittent infusion (3.375 g every 6 h, n = 132). NONMEM was used to perform population pharmacokinetic analysis in a subset of patients (n = 56) who had serum samples obtained at steady-state for drug concentration analyses. Classification and regression tree analysis was used to identify the breakpoints of piperacillin PK-PD indexes in 94 patients with causative pathogen's MIC. RESULTS A one-compartment model was applied to fit the data. Creatinine clearance and body weight were the most significant variables to explain patient variability in piperacillin and tazobactam clearance and volume of distribution. The infusion method had no influence on PK parameters. For patients (n = 30) receiving intermittent infusion in the pharmacokinetic study, mean Cmax and half-life were 122.22 mg/L and 1.17 h for piperacillin, and 15.74 mg/L and 1.81 h for tazobactam. For patients (n = 26) receiving continuous infusion in the pharmacokinetic study, mean steady-state concentration was 35.31 +/- 12.15 mg/L for piperacillin and 7.29 +/- 3.28 mg/L for tazobactam. As a result of a low rate of failures (<11%) observed in the trial and the low MICs for infecting pathogens, no association could be established between clinical/microbiological outcome and drug exposure. CONCLUSIONS Intermittent infusion and continuous infusion of piperacillin and tazobactam provided sufficient drug exposure to treat those pathogens commonly implicated in intra-abdominal infections.