Development and Validation of a Prognostic Nomogram for Predicting 3-Month Mortality After Liver Transplantation

Area of Science:

• Transplantation Medicine
• Medical Informatics
• Clinical Prognostics

Background:

• Liver transplantation is a life-saving procedure, but patient mortality remains a concern.
• Early identification of high-risk individuals post-transplant is crucial for effective intervention.
• Developing accurate prognostic tools can significantly improve patient outcomes.

Purpose of the Study:

• To develop and validate a predictive model for mortality after liver transplantation.
• To create a nomogram incorporating key risk factors for post-transplant mortality.
• To aid clinicians in identifying patients requiring immediate post-operative care.

Main Methods:

• A retrospective cohort of 1951 liver transplant recipients was analyzed.
• Lasso regression and Boruta feature selection identified significant predictors.
• Model performance was assessed using ROC curves, calibration plots, and decision curve analysis.

Main Results:

• The nomogram achieved an AUC of 0.824 (training) and 0.774 (validation).
• Calibration plots indicated good predictive accuracy.
• Decision curve analysis showed clinical utility across various risk thresholds.

Conclusions:

• The developed nomogram is a valuable tool for predicting mortality in liver transplant recipients.
• It assists healthcare professionals in identifying high-risk patients needing urgent interventions.
• Further validation in prospective cohorts is warranted.