Effects of ex vivo Extracorporeal Membrane Oxygenation Circuits on Sequestration of Antimicrobial Agents

Objectives: Our ex vivo study was designed to determine the sequestration of teicoplanin, tigecycline, micafungin, meropenem, polymyxin B, caspofungin, cefoperazone sulbactam, and voriconazole in extracorporeal membrane oxygenation (ECMO) circuits. Methods: Simulated closed-loop ECMO circuits were prepared using 2 types of blood-primed ECMO. After the circulation was stabilized, the study drugs were injected into the circuit. Blood samples were collected at 2 min, 5 min, 15 min, 30 min, 1 h, 3 h, 6 h, 12 h, and 24 h after injection. Drug concentrations were measured by high-performance liquid chromatography-tandem mass spectrometry. Control groups were stored at 4 °C after 3h, 6h, 12h, and 24h immersing in a water bath at 37 °C to observe spontaneous drug degradation. Results: Twenty-six samples were analyzed. The average drug recoveries from the ECMO circuits and control groups at 24 h relative to baseline were 67% and 89% for teicoplanin, 100% and 145% for tigecycline, 67% and 99% for micafungin, 45% and 75% for meropenem, 62% and 60% for polymyxin B, 83% and 85% for caspofungin, 79% and 98% for cefoperazone, 75% and 87% for sulbactam, and 60% and101% for voriconazole, respectively. Simple linear regression showed no significant correlation between lipophilicity (r2 = 0.008, P = 0.225) or the protein binding rate (r2 = 0.168, P = 0.479) of drugs and the extent of drug loss in the ECMO circuits. Conclusions: In the two ECMO circuits, meropenem and voriconazole were significantly lost, cefoperazone was slightly lost, while tigecycline and caspofungin were not lost. Drugs with high lipophilicity were lost more in the Maquet circuit than in the Sorin circuit. This study needs more in vivo studies with larger samples for further confirmation, and it suggests that therapeutic drug concentration monitoring should be strongly considered during ECMO.