Continuous infusion of piperacillin/tazobactam optimizes intraoperative antibiotic exposure in patients undergoing elective pelvic exenteration surgery

Abstract

Patients undergoing elective pelvic exenteration surgery who receive piperacillin/tazobactam as surgical prophylaxis are at risk of suboptimal intraoperative antibiotic exposure. With this work, we aimed to study the plasma pharmacokinetics of piperacillin and tazobactam in this population to provide dosing recommendations that optimize antibiotic exposure. We developed a prospective, observational, pharmacokinetic study of piperacillin/tazobactam in patients undergoing pelvic exenteration. Population pharmacokinetic analysis and Monte Carlo simulations were performed with Monolix and Simulx software. Probabilities of target attainment of different dosing regimens against the minimum inhibitory concentration (MIC) breakpoints (8 and 16 mg/L) were calculated. Twelve patients were included in the study, with a median age of 50.0 years [interquartile interval (45.3-57.5)] and a median weight of 79.0 kg (61.3-88.3). Median surgical time was 10.5 h (9.8-11.7). A two-compartment linear model best fitted piperacillin and tazobactam data (190 plasma samples). Monte Carlo simulations showed that a lower dose of 2 g/0.25 g loading dose followed by 4 g/0.5 g q8h by continuous infusion provided ≥90% probability of target attainment for MIC = 16 mg/L for most of the patients. For non-continuous infusion regimens, only the 2-hourly bolus re-dosing achieved intraoperative concentrations of piperacillin ≥16 mg/L. Patients with weights ≥ 100 kg and glomerular filtration rates ≥ 120 mL/min required 4 g/0.5 g q6h by continuous infusion after a loading dose. In conclusion, continuous infusion of lower doses of piperacillin/tazobactam is as adequate as the 2-hourly re-dosing recommended by the current guidelines for surgical prophylaxis in pelvic exenteration. Patients with higher weights and glomerular filtration rates are at greater risk of inadequate exposure.

Keywords: continuous infusion; pelvic exenteration; pharmacokinetics; piperacillin/tazobactam; surgical prophylaxis.

Fig 1

Observed concentrations over time of piperacillin and tazobactam. Concentrations are in mg/L, and time is in h. Lower graphs are depicted in a semilogarithmic scale.

Fig 2

Goodness-of-fit plots for the final population pharmacokinetic model of piperacillin. Upper panels: (A) observed concentrations versus population predicted concentrations and (B) observed concentrations versus individual predicted concentrations. Lower panels: (C) NPDE versus time and (D) NPDE versus population predicted concentrations. Black dots represent observations. Black solid lines represent line of unity. Light gray lines represent spline lines. All concentrations are in mg/L, and time is in h.

Fig 3

Goodness-of-fit plots for the final population pharmacokinetic model of tazobactam. Upper panels: (A) observed concentrations versus population predicted concentrations and (B) observed concentrations versus individual predicted concentrations. Lower panels: (C) NPDE versus time and (D) NPDE versus population predicted concentrations. Black dots represent observations. Black solid lines represent line of unity. Light gray lines represent spline lines. All concentrations are in mg/L, and time is in h.