RESUMO
Objective: To validate deep learning models' ability to predict post-transplantation major adverse cardiovascular events (MACE) in patients undergoing liver transplantation (LT). Patients and Methods: We used data from Optum's de-identified Clinformatics Data Mart Database to identify liver transplant recipients between January 2007 and March 2020. To predict post-transplantation MACE risk, we considered patients' demographics characteristics, diagnoses, medications, and procedural data recorded back to 3 years before the LT procedure date (index date). MACE is predicted using the bidirectional gated recurrent units (BiGRU) deep learning model in different prediction interval lengths up to 5 years after the index date. In total, 18,304 liver transplant recipients (mean age, 57.4 years [SD, 12.76]; 7158 [39.1%] women) were used to develop and test the deep learning model's performance against other baseline machine learning models. Models were optimized using 5-fold cross-validation on 80% of the cohort, and model performance was evaluated on the remaining 20% using the area under the receiver operating characteristic curve (AUC-ROC) and the area under the precision-recall curve (AUC-PR). Results: Using different prediction intervals after the index date, the top-performing model was the deep learning model, BiGRU, and achieved an AUC-ROC of 0.841 (95% CI, 0.822-0.862) and AUC-PR of 0.578 (95% CI, 0.537-0.621) for a 30-day prediction interval after LT. Conclusion: Using longitudinal claims data, deep learning models can efficiently predict MACE after LT, assisting clinicians in identifying high-risk candidates for further risk stratification or other management strategies to improve transplant outcomes based on important features identified by the model.
RESUMO
With donation after circulatory death (DCD) liver transplantation (LT), the goal of the recipient implantation procedure is to minimize surgical complexity to avoid a tenuous environment for an already marginal graft. The presence of portal vein thrombosis (PVT) at the time of LT adds surgical complexity, yet' to date, no studies have investigated the utilization of DCD liver grafts for patients with PVT. Methods: All DCD LT performed at Mayo Clinic-Florida, Mayo Clinic-Arizona, and Mayo Clinic-Rochester from 2006 to 2020 were reviewed (N = 771). Patients with PVT at the time of transplant were graded using Yerdel classification. A 1:3 propensity match between patients with PVT and those without PVT was performed. Results: A total of 91 (11.8%) patients with PVT undergoing DCD LT were identified. Grade I PVT was present in 62.6% of patients, grade II PVT in 27.5%, grade III in 8.8%, and grade 4 in 1.1%. At the time of LT, thromboendovenectomy was performed in 89 cases (97.8%). There was no difference in the rates of early allograft dysfunction (43.2% versus 52.4%; P = 0.13) or primary nonfunction (1.1% versus 1.1%; P = 0.41) between the DCD PVT and DCD without PVT groups, respectively. The rate of ischemic cholangiopathy was not significantly different between the DCD PVT (11.0%) and DCD without PVT groups (10.6%; P = 0.92). Graft (P = 0.58) and patient survival (P = 0.08) were similar between the 2 groups. Graft survival at 1-, 3-, and 5-y was 89.9%, 84.5%, and 79.3% in the DCD PVT group. Conclusions: In appropriately selected recipients with grades I-II PVT, DCD liver grafts can be utilized safely with excellent outcomes.
