Outcomes of donation after circulatory death (DCD) and ex-vivo lung perfusion (EVLP) lung transplantation

Background: Donation after circulatory death (DCD) and ex-vivo lung perfusion (EVLP) have been adopted to expand the donor pool in lung transplantation, but outcomes data have been conflicting. This study explores outcomes of DCD and EVLP lung transplantation in the modern era.

Methods: The United Network for Organ Sharing database was queried for adult lung transplants from January 1, 2015 to March 1, 2023. Loss to follow-up, multiorgan, and prior lung transplants were excluded. DCD versus donation after brain death (DBD) lung transplants were compared with subgroup analysis +/- EVLP. Outcomes were survival and postoperative complications.

Results: The study included 1,103 DCD (221 with EVLP and 882 without) and 17,973 DBD lung transplants (524 with EVLP and 17,449 without). Median follow-up was 3 years. DCD donors were less likely to be CDC high risk (19.3% vs 24.1%, p < 0.001), have purulence on bronchoscopy (13.3% vs 18.3%, p < 0.001), or infiltrates on chest X-ray (66.7% vs 67.8%, p = 0.013). EVLP was more likely to be used for DCD transplants (20.0% vs 2.9%, p < 0.001). After transplant, DCD recipients were more likely to be reintubated (24.3% vs 18.5%, p < 0.001) and require ECMO within 72 hours (14.9% vs 7.8%, p < 0.001), and DCD donation was an independent risk factor for these complications on multivariable logistic regression. Overall survival did not differ significantly between DCD and DBD transplants on adjusted survival analysis in the early or modern era (p = 0.774 and p = 0.468, respectively). On subgroup analysis, the DCD+EVLP cohort had significantly worse survival in the modern era, which remained significant after adjusting for donor and recipient factors (p = 0.005). EVLP was an independent risk factor for graft failure in the DCD cohort (hazard ratio [HR] 1.33, 95% confidence interval [CI] 1.00-1.77, p = 0.047) but did not significantly affect DBD graft survival (p = 0.870). Risk factors for graft failure and mortality in the DCD+EVLP cohort included pulmonary hypertension (HR 77.5, 95% CI 6.15-979, p < 0.001), transfusion before transplant (HR 2.60, 95% CI 1.07-6.31, p = 0.035), elevated creatinine (HR 2.82, 95% CI 1.34-5.90, p = 0.006), and higher allocation score (HR 1.02, 95% CI 1.00-1.04, p = 0.017) CONCLUSIONS: Study findings suggest increased risks of mortality and perioperative complications following transplantation with DCD lungs that have undergone EVLP. DCD lung transplantation without EVLP confers equivalent survival but with some increase in perioperative complications. Further investigation and careful recipient selection are warranted to optimize the use of these extended criteria donors in the modern era.