Introduction: As a result of drug sequestration and increased volume of distribution, the extracorporeal membrane oxygenation (ECMO) procedure might lead to a decrease in drug concentrations during a patient’s treatment. The aim of this study was to evaluate sedative, antibiotic and immunosuppressive drug loss in ECMO circuit using ex-vivo and in-vitro experiments. Methods: Blood concentrations of propofol, midazolam, cyclosporine and vancomycin were measured in an ex-vivo ECMO circuit primed with whole human blood, and compared to controls stored in polypropylene tubes. In vitro experiments were also conducted to further explore the role of temperature, oxygen exposure and polyvinylchloride surfaces on propofol loss in the ECMO circuit. Results: Propofol concentration decreased rapidly; 70% of its baseline concentration was lost after only 30 minutes, and only 11% remained after five hours (P <0.001 for the comparison with control polypropylene tube propofol concentration). Further experiments demonstrated that oxygen exposure and contact with polyvinylchloride tubing were respectively responsible for 70% and 85% of propofol loss after 45 minutes. Midazolam concentration also rapidly decreased in the ECMO circuit, with only 54% and 11% of baseline concentration being detected at 30 minutes and 24 hours respectively (P = 0.01 versus control). Alternatively, cyclosporine concentration remained stable for the five first hours, then decreased to 78% and 73% of the baseline value after 24 hours and 48 hours, (P = 0.35 versus control). Lastly, vancomycin concentration remained stable in the ECMO circuit for the 48-hour experimental protocol. Conclusions: We observed important losses of propofol and midazolam, while cyclosporine concentration decreased slowly and moderately, and vancomycin concentration remained unchanged in the ex-vivo ECMO circuit primed with whole human blood. These data might help intensive care unit physicians planning clinical trials with a final objective to better adapt doses of these drugs while treating critically ill ECMO patients.

Background: As a consequence of drug sequestration, increase in volume of distribution, or alteration of elimination, extracorporeal membrane oxygenation (ECMO) might lead to inadequate plasma concentrations of vital drugs. The aim of this experimental study was to develop an ex vivo model to better characterize the impact of ECMO procedure on beta-lactam antibiotics pharmacokinetics. Methods: Plasma concentrations of cefotaxime, ceftazidime, cefepime, piperacillin, oxacillin, amoxicillin, and ceftriaxone were measured in an ex vivo ECMO circuit primed with whole human blood and compared with controls stored in glass tubes and polyvinyl chloride tubing. Serial blood samples were collected over 48 hours, and the concentrations of beta-lactam antibiotics were quantified using a validated high-performance liquid chromatography assay. The concentrations' decay rate over time was compared between the ECMO circuits and controls using nonlinear mixed-effect modeling. Results: Cefotaxime concentrations decreased markedly: 86% of the initial concentration remained after 4 hours and only 21% after 48 hours (P < 0.05 for the comparison in rate of decrease with both glass and polyvinyl chloride controls). There was no difference in the rate of decrease between ECMO circuit and controls for the other beta-lactam antibiotics. The average drug recoveries from the ECMO circuits at 48 hours were as follows: ceftazidime, 73%; cefepime, 67%; piperacillin, 71%; oxacillin, 46%; and amoxicillin, 72%. Concentrations of ceftriaxone remained stable throughout the 48-hour study both in ECMO circuits and in controls. Conclusions: Significant losses of cefotaxime were observed, whereas ceftazidime, cefepime, piperacillin, oxacillin, and amoxicillin decrease was moderate and similar to that of the control group, and ceftriaxone concentrations remained unchanged. These results are reassuring for the use of beta-lactam antibiotics in critically ill patients treated with ECMO.