A Novel Nomogram Integrating Early Postoperative Hemodynamics with Perioperative Clinical Variables for Predicting Liver Transplantation Overall Survival.

RATIONALE AND

OBJECTIVES

Post-liver transplantation (LT) mortality remains significant despite surgical advances. Current prediction models show limited predictive performance for post-LT outcomes. We aimed to develop and validate a novel LT overall survival (LTOS)-nomogram integrating early postoperative hemodynamic parameters with perioperative clinical variables for predicting survival after LT.

METHODS

This single-center retrospective study included 204 recipients who underwent LT between April 2016 and December 2023. Patients were randomly divided into training (n=104) and testing (n=100) sets. The Boruta algorithm combined with least absolute shrinkage and selection operator regression identified key predictors for survival. A Cox regression-based nomogram was constructed and validated using time-dependent receiver operating characteristic curves and area under the curves (AUCs), calibration, and decision curve analysis curves.

RESULTS

Five key predictors were identified by both algorithms: portal vein flow and hepatic artery resistive index measured by Doppler ultrasound within 24 h post-LT, preoperative white blood cell count, portal vein thrombosis, and abdominal adhesion. Kaplan-Meier analyses demonstrated significant survival differences for patients categorized by these factors (P<0.001). The LTOS-nomogram demonstrated good predictive performance for 30-day, 90-day, and 1-year survival in both training (AUCs=0.907, 0.888, and 0.845, respectively) and testing sets (AUCs=0.847, 0.900, and 0.816, respectively), with good calibration and higher net benefit in decision curve analysis, outperforming the model for end-stage liver disease score and Child-Pugh classification. An online calculator was developed to facilitate clinical application (https://ltosproject.netlify.app).

CONCLUSION

The LTOS-nomogram effectively predicts post-LT survival by integrating early postoperative hemodynamic parameters with perioperative clinical variables, providing a valuable tool for risk stratification and individualized management of LT recipients.