Insufficient plasma concentrations of empiric anti-pseudomonal beta-lactam antibiotics in critically ill patients with suspected sepsis

Abstract

Aim: To determine if empirical anti-pseudomonal beta-lactam antibiotic (BLA) dosing achieves adequate drug exposure in septic patients. Method: A single-centre, prospective pharmacokinetic study was conducted in Intensive Care Unit (ICU) patients with sepsis, receiving empirical therapy with piperacillin/tazobactam or meropenem. The pharmacokinetic/pharmacodynamic (PK/PD) targets were free BLA concentrations above the MIC at the mid-point (Concentration A, 50%f T > MIC) and end of the dosing interval (Concentration B, 100% fT > MIC). Sub-therapeutic concentrations were defined as concentration A < MIC, and sub-optimal as concentration B < MIC. The MIC breakpoint for Pseudomonas aeruginosa, as defined by The European Committee on Antimicrobial Susceptibility Testing (EUCAST) (available from ) was used. Results: Of the 37 eligible patients, 20 were receiving piperacillin/tazobactam (TZP), and 17 meropenem (MEM). PK data were available for 36 patients (concentration A) and 34 patients (concentration B). The median measured plasma concentrations (mg/L) were: piperacillin for doses 4 g 8-hourly and 4 g 6-hourly – concentration A 53.9 [14.38–123.71] and 36.44 [13.38–107.45], concentration B 8.01 [2.57–71.08] and 27.31 [3.32–59.76], MEM for doses 1 g 8-hourly and 2 g 8-hourly – concentration A 12.49 [7.35–23.63] and 30.5, concentration B 7.47 [2.97–11.33] and 9.31. 27.8% (10 of 36) of patients had sub-therapeutic concentrations (concentration A) and 50% (17 of 34) had sub-optimal concentrations (concentration B). Conclusion: Our study confirms that sub-therapeutic unbound plasma anti-pseudomonal BLA concentrations are common in critically ill septic patients and underscores the urgent need for future trials exploring the efficacy of BLA therapeutic drug monitoring in these patients.