Unplanned Intraoperative ECMO Support during Double Lung Transplantation: Risk-Factors and Outcome Analysis

Abstract

Purpose The extracorporeal membrane oxygenation (ECMO) is a technical option to support cardiorespiratory function during double lung transplantation (DLT). To date, intraoperative ECMO support during DLT seems to give excellent results when planned and used prophylactically, on the other hand the need to use an emergency unplanned ECMO support seems to increase significantly the postoperative morbidity and mortality. The primary endpoint of our study was to search for any factors that could significantly influence the risk of unplanned intraoperative ECMO support. The perioperative and one-year outcome analysis of patients who underwent unplanned ECMO were the secondary endpoints. Methods Records of 95 patients who underwent DLT at our institution between January 2016 and December 2019 were retrospectively reviewed. Patients who underwent lung transplantation with elective intraoperative cardio-circulatory support were excluded from the study. Patients who underwent DLT with unplanned intraoperative ECMO (50 patients, Group A) were compared with no ECMO patients (45 patients, Group B). Results Group A patients showed more preoperative pulmonary artery hypertension (PAH), NYHA class>3, low cardiac-index, higher lactate and dose of inotropes during anesthesia induction. Group A more frequently received donors with higher PCO2 and with cerebral ischemic or hemorrhagic cause of death. After regression analyses PAH (OR7.42), PaCO2 of donors (OR6.29) were the major risk factors for unplanned ECMO. The postoperative period was more complicated for patients requiring intraoperative ECMO. The failure to wean from ECMO at the end of surgery further increase the risk of post-operative morbidity (OR8.43) and mortality (OR4.21). Group A showed a higher in-hospital mortality (24%vs7%, p Conclusion The PAH and a higher PaCO2 of the donor were risk factors for unplanned ECMO. These results support the hypothesis that resolving critical intraoperative events by an unplanned ECMO does allow the intervention to be completed but then is reflected in a worse outcome, in particular after weaning failure at the end of surgery. Identifying a preoperative risk profile could be useful to use prophylactic ECMO and thus avoid the establishment of critical intraoperative events.