Abstract:Background and aims:<bold> </bold>Dynamic evaluation of critically ill patients is the key to predicting their outcomes. Most scores based on the Model for End-stage Liver Disease (MELD) and acute-on-chronic liver failure (ACLF) utilize point-in-time assessment. This study mainly aimed to investigate the impact of dynamic clinical course change on post-liver transplantation (LT) survival.Methods:<bold> </bold>This study included 637 adults (overall cohort) with benign end-stage liver diseases. The authors compared the MELD scores and our ACLF-based dynamic evaluation scores. Patients enrolled or transplanted with ACLF-3 were defined as the ACLF-3 cohort ( n =158). The primary outcome was 1-year mortality. Delta MELD and Delta CLIF-OF (Chronic Liver Failure-Organ Failure) represented the respective dynamic changes in liver transplant function. Discrimination was assessed using the area under the curve. A Cox regression analysis identified independent risk factors for specific organ failure and 1-year mortality.Results:<bold> </bold>Patients were grouped into three groups: the deterioration group (D), the stable group (S), and the improvement group (I). The deterioration group (Delta CLIF-OF >= 2) was more likely to receive national liver allocation ( P =0.012) but experienced longer cold ischemia time ( P =0.006) than other groups. The area under the curves for Delta CLIF-OF were 0.752 for the entire cohort and 0.767 for ACLF-3 cohorts, both superior to Delta MELD ( P <0.001 for both). Compared to the improvement group, the 1-year mortality hazard ratios (HR) of the deterioration group were 12.57 (6.72-23.48) for the overall cohort and 7.00 (3.73-13.09) for the ACLF-3 cohort. Extrahepatic organs subscore change (HR=1.783 (1.266-2.512) for neurologic; 1.653 (1.205-2.269) for circulation; 1.906 (1.324-2.743) for respiration; 1.473 (1.097-1.976) for renal) were key to transplantation outcomes in the ACLF-3 cohort. CLIF-OF at LT (HR=1.193), Delta CLIF-OF (HR=1.354), and cold ischemia time (HR=1.077) were independent risk factors of mortality for the overall cohort, while Delta CLIF-OF (HR=1.384) was the only independent risk factor for the ACLF-3 cohort. Non-ACLF-3 patients showed a higher survival rate than patients with ACLF-3 in all groups ( P =0.002 for I, P =0.005 for S, and P =0.001 for D).Conclusion:<bold> </bold>This was the first ACLF-based dynamic evaluation study. Delta CLIF-OF was a more powerful predictor of post-LT mortality than Delta MELD. Extrahepatic organ failures were core risk factors for ACLF-3 patients. CLIF-OF at LT, Delta CLIF-OF, and cold ischemia time were independent risk factors for post-LT mortality. Patients with a worse baseline condition and a deteriorating clinical course had the worst prognosis. Dynamic evaluation was important in risk stratification and recipient selection.

DynaMELD: A Dynamic Model of End-Stage Liver Disease for Equitable Prioritization

A Transformer-Based Deep Learning Approach for Fairly Predicting Post-Liver Transplant Risk Factors

Evidence-based Incorporation of Key Parameters into MELD Score for Acute-on-chronic Liver Failure

Development of a Novel Prognostic Nomogram for High Model for End-Stage Liver Disease Score Recipients Following Deceased Donor Liver Transplantation

Machine Learning for Post-Liver Transplant Survival: Bridging the Gap for Long-term Outcomes through Temporal Variation Features

A confounder debiasing method for RCT-like comparability enables Machine Learning-based personalization of survival benefit in living donor liver transplantation

Online interpretable dynamic prediction models for clinically significant posthepatectomy liver failure based on machine learning algorithms: A retrospective cohort study

Dynamic evaluation based on acute-on-chronic liver failure predicts survival of patients after liver transplantation: a cohort study

Fairly Predicting Graft Failure in Liver Transplant for Organ Assigning

Predicting Short-term Survival after Liver Transplantation using Machine Learning

Predicting Long-Term Allograft Survival in Liver Transplant Recipients

The MELD-Plus: A Generalizable Prediction Risk Score in Cirrhosis

S1708 Machine Learning Prognostic Models Compared With the MELD and MELD-Na Scores for Prediction of 90-Day Mortality in Liver Diseases: A Systematic Review and Meta-Analysis

Application of model for end-stage liver disease as disease classification in cardiac valve surgery: a retrospective study based on the INSPIRE database

Accurate prediction of all-cause mortality in patients with metabolic dysfunction-associated steatotic liver disease using electronic health records

A Machine Learning Model to Predict Risk for Hepatocellular Carcinoma in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease

The Evolution of the MELD Score and Its Implications in Liver Transplant Allocation: A Beginner's Guide for Trainees

Deep Learning-Adjusted Monitoring of In-Hospital Mortality after Liver Transplantation

Waitlisting and liver transplantation for MetALD in the United States: An analysis of the UNOS national registry

Relationship between updated MELD and prognosis in alcohol-associated hepatitis: Opportunities for more efficient trial design