Abstract Background: Extracorporeal membrane oxygenation (ECMO) is commonly used during double-lung transplantation. ECMO can be planned or unplanned, and used only during the procedure or extended postoperatively (intraoperative or extended). Our practice is to limit its use and duration as much as possible. We conducted this retrospective single-center study to assess prognoses of patients undergoing unplanned-intraoperative ECMO. Methods: From among 436 patients who underwent double-lung transplantation from 2012 to 2018, we excluded those who underwent bridge-to-transplantation, multiorgan transplantation, repeated transplantation during the study period, and cardiopulmonary bypass. Unplanned-intraoperative ECMO group was compared with no-ECMO and planned-intraoperative ECMO groups. Results: In our sample, 209 patients did not require ECMO, 77 underwent unplanned-intraoperative ECMO, and 14 underwent planned-intraoperative ECMO. One-year and 3-year survival were lower in unplanned-intraoperative ECMO group than in the no-ECMO group (P = .043 and P = .032, respectively). The only independent protective factor related to 1-year mortality was history of cystic fibrosis (P = .013). Lung allocation score (P = .001), grade 3 pulmonary graft dysfunction at end-surgery status (P = .014), and estimated intraoperative blood loss (P = .031) were risk factors. Conclusions: Patients who underwent unplanned-intraoperative ECMO showed poorer prognoses than patients who did not require ECMO. This finding may be explained by differences in initial condition severity, by long-term consequences of the intraoperative complications leading to ECMO pump implantation, or by flaws in our weaning protocol.