Extubation of patients undergoing extracorporeal life support. A retrospective study.

The use of extracorporeal life support (ECLS) is increasing worldwide, in particular for the management of refractory cardiac arrest, cardiogenic shock and post cardiopulmonary bypass ventricular failure. Extubation of patients under extracorporeal membrane oxygenation (ECMO) for respiratory failure is a growing practice for adult and pediatric patients, especially for lung transplantation candidates. Because of potential complications and, specifically, accidental arterial decannulation, extubation of patients under ECLS is not standard practice. Our goal was to evaluate the interest in patient extubation under ECLS. We performed a monocentric, retrospective study of all ECLS cases between January 2014 and January 2016. We excluded patients who died within the first 48 hours of ECLS. We analyzed 57 of the initial 109 patients included in the study. The initial SOFA score was higher in the non-extubated group under ECLS, without significant difference (8.6 ± 2.8 vs 7.2 ± 2.1, p=0.065). Patients who were not extubated had a higher rate of acquired ventilator pneumonia (61.9% vs 26.7%, p=0.03). Moreover, patients who were extubated under ECLS had better 30-day survival rates (73.3% vs 40.5%, p=0.04). In multivariate analyses, the independent factors associated with mortality were age, duration of ECLS and the lack of extubation under ECLS. Extubation of patients under ECLS is safe and feasible. Furthermore, in extubated patients, we observed fewer cases of ventilator-associated pneumonia and better 30-day survival rates.