Effects of extracorporeal membrane oxygenation and/or renal replacement therapy on piperacillin/tazobactam levels in critically ill patients

Background: Although organ replacement therapy serves as a bridge to the recovery of vital organ function, it may lead to significant physiological and pharmacokinetic changes. To identify factors that may affect patient outcomes due to inadequate antimicrobial therapy, we aimed to evaluate the effectiveness of primary dosage determination of piperacillin (PIP) in the intensive care unit (ICU) setting, particularly in patients undergoing renal or lung replacement therapy. Methods: Between January 2020 and December 2021, we retrospectively analyzed all ICU patients with a confirmed SARS-CoV-2 infection who received piperacillin/tazobactam (PIP/TAZ) continuous infusion due to suspected secondary infection. The analysis included a comparison of serum PIP levels, demographic data, and laboratory findings among different patient groups. Results: From 108 included patients, therapeutic drug monitoring (TDM) was performed at the initiation of antibiotic treatment in 91 (84.3%) cases. Of these patients, 15 received extracorporeal membrane oxygenation (ECMO), 12 underwent continuous renal replacement therapy (CRRT), and 43 received both procedures during their ICU stay. The median age was 58 years. The in-hospital mortality rate was significantly higher in patients receiving ECMO therapy (73.3% vs. 23.8%) and in those receiving both ECMO and CRRT (72.1% vs. 23.8%) compared to patients without organ replacement procedures. Dosing calculation adjusted to renal function was correctly performed according to the established standard operating procedure in 96.7% of cases. The PIP level after primary dosage, prior to further adjustments, was within the standard operating procedure-defined concentration range of 32-96 mg/L, encompassing the empirical target range and the higher exposure range relevant in specific clinical scenarios, in 59.0% median target concentration: 44.0 mg/L of patients without organ support, 58.3% of those on ECMO median target concentration: 66.8 mg/L, 80.0% of those undergoing CRRT median target concentration: 56.0 mg/L, and in 27.8% of the patients receiving both ECMO and CRRT median target concentration: 116.0 mg/L. In total, 17.6% of all patients did not reach the minimum PIP target level of 32 mg/L and 27.5% were above the maximum PIP target level of 96 mg/L. Binary regression analysis did not identify significant correlation between plasma bilirubin, plasma albumin, the capillary leak index (CLI) or anthropometric variables and achieving the target plasma concentration of PIP. Conclusion: In total, 45% of first-dose PIP concentrations in the analyzed critically ill patients were outside the recommended concentration corridor. While lowest concentrations of PIP levels were detected in critically ill patients without organ support therapy, levels above the therapeutic range were detected in patients treated with ECMO support. However, based on the data of this study, a predictor to prevent primary drug levels outside the therapeutic range was not identified. These findings underscore the critical role of TDM in daily clinical routine.