Post-transplant predictors of mortality and primary graft dysfunction in lung transplant recipients bridged with veno-venous extracorporeal membrane oxygenation

PURPOSE: Veno-venous extracorporeal membrane oxygenation (VV-ECMO) is increasingly used to bridge critically ill candidates to lung transplantation, with reported outcomes approaching those of non-ECMO recipients. However, factors associated with early mortality in this high-risk subpopulation remain incompletely defined, particularly when postoperative complications occur after transplantation and may introduce time-dependent bias. We aimed to explore factors associated with 1-year survival among VV-ECMO-bridged lung transplant recipients. METHODS: We performed a single-center retrospective cohort study of adult lung transplant recipients bridged with preoperative VV-ECMO between January 2018 and May 2024. The primary endpoint was 1-year overall survival. Associations with primary graft dysfunction grade 3 were assessed using logistic regression. Because postoperative complications occurred after transplantation, landmark Kaplan-Meier analyses were performed to explore their associations with subsequent survival while reducing immortal time bias. RESULTS: Forty recipients were included, and 1-year mortality was 20% (8/40). Primary graft dysfunction grade 3 occurred in 16 recipients (40%). Higher pretransplant creatinine and lower albumin showed suggestive associations with primary graft dysfunction grade 3, although these associations did not reach statistical significance in multivariable analysis. In conventional Cox analyses, postoperative pulmonary embolism, dialysis/renal replacement therapy, and antibody-mediated rejection were associated with mortality; however, these findings should be interpreted cautiously because these postoperative complications were time-dependent and the number of deaths was limited. Landmark analyses showed that dialysis/renal replacement therapy by POD30 was associated with worse subsequent survival, whereas the associations for pulmonary embolism by POD90 and antibody-mediated rejection by POD90 were limited by sparse events. CONCLUSIONS: In this single-center cohort of VV-ECMO-bridged lung transplant recipients, postoperative complications including pulmonary embolism, dialysis/renal replacement therapy, and antibody-mediated rejection were associated with mortality in exploratory analyses, with dialysis showing the most consistent association in landmark analysis. Given the small sample size and limited number of events, these findings should be interpreted as hypothesis-generating and require validation in larger multicenter cohorts.