Establishment of a Risk Prediction Model for Prolonged Mechanical Ventilation After Liver Transplantation: A Multicenter Retrospective Study

PURPOSE: To identify risk factors for prolonged mechanical ventilation (PMV) after liver transplantation (LT) and to develop and internally validate a predictive nomogram. METHODS: A multicenter retrospective cohort of 746 LT recipients (January 2015-June 2023) was randomly split 7:3 into training (n = 522) and validation (n = 224) sets. We conducted univariate and multivariate logistic regression to identify factors linked to PMV. A risk nomogram was created based on the multivariate analysis, and its performance was evaluated for calibration, discrimination, and clinical utility, with internal validation using the Validation Set data. RESULTS: The incidence of PMV was 21.6%. Multivariate analysis identified five independent predictors of PMV: ASA scores ≥ 3, mechanical ventilation before LT, MELD score, albumin levels, and anhepatic phase. The nomogram developed from these predictors showed a concordance index (C-index) of 0.780 (95% CI, 0.734-0.827) and demonstrated good calibration. After internal validation, the model maintained a C-index of 0.731 (95% CI, 0.653-0.810). Decision curve analysis and clinical impact curve confirmed the nomogram's effectiveness. CONCLUSION: The nomogram enables individualized PMV risk estimation in LT recipients and may guide early intervention and resource allocation.