BACKGROUND: Current prognostic models for liver transplantation (LT) recipients are not applicable for long-term survival prediction, especially for pediatric living donor LT recipients. In this retrospective prognostic study, we aimed to develop and validate a survival predictive model for pediatric LT recipients, the Survival Predictive Model After Pediatric Liver Transplantation (SPPLT) model. METHODS: A total of 2964 pediatric recipients from Renji Hospital were enrolled and randomly assigned to the training and validation cohort by a ratio of 1:1. Logistic analysis was performed to identify risk factors for posttransplant mortality and least absolute shrinkage and selection operator Cox regression was used to establish the SPPLT model. RESULTS: Univariate analysis identified 19 risk factors that were associated with post-LT mortality, including graft-recipient compatibility, pre-LT hepatorenal function, blood loss and transfusion during the operation, post-LT infection and surgical complications. Using the least absolute shrinkage and selection operator multivariable Cox regression, we established the prognostic SPPLT model which included 7 clinical easily approached factors (pre-LT creatinine level, blood loss during operation, days of intensive care unit and hospital stay after transplantation, sputum culture positivity, post-LT abdominal hemorrhage, and intestinal perforation). The accuracy of predicting 1-, 3-, and 5-y post-LT survival rates in the training and validation cohort reached 0.802, 0.828, 0.811, and 0.803, 0.784, 0.778, respectively. Furthermore, the SPPLT model efficiently stratified all pediatric LT recipients into high-risk and low-risk groups, which were associated with totally different long-term outcomes (hazard ratio = 0.27, 95% confidence interval (0.120-0.620), P <
0.01). CONCLUSIONS: SPPLT is the first prognostic model to accurately predict long-term survival rate after pediatric LT. Recipients with high mortality risk should receive intense monitoring after transplantation.