Development and validation of a model for early survival prediction following liver transplantation based on donor and recipient characteristics

Background: Circulating cytokine levels not only correlate with the progression of liver disease but also serve as indicators for the infection status of the body. Growing evidence points to the connection between donor cytokines and graft function following transplantation. This study set out to explore the clinical significance of donor cytokines in predicting liver transplantation prognosis. Methods: Data from 172 deceased donor liver transplantations conducted between 2017 and 2022, with available donor serum cytokine information, were collected. The subjects were randomly divided into estimation (n = 120) and validation (n = 52) groups to establish and validate the model. The newly developed SA10 score was compared against established models EAD, MEAF, L-GrAFT7, and L-GrAFT10. Results: Donor IL-10, along with donor age and recipient AST peak value within the first 7 days post-operation, was identified as an independent factor associated with recipient survival and was incorporated into the SA10 score. SA10 exhibited robust predictive capability, particularly for 1-month survival (AUC = 0.90, 95% CI = 0.84–0.96), outperforming EAD (AUC = 0.75, 95% CI = 0.60–0.90, p = 0.04) and L-GrAFT7 (AUC = 0.65, 95% CI = 0.49–0.81, p < 0.01). Comparable performance was observed between SA10, MEAF, and L-GrAFT10. Conclusion: Donor IL-10 independently influences recipient survival, with the SA10 score demonstrating comparable and even superior predictive ability compared to existing models.