Predictors of Mortality in Long-Term Follow-Up of Patients with Terminal Alcoholic Cirrhosis: Is It Time to Accept Remodeled Scores?

OBJECTIVE: To identify the prognostic score that is the best predictor of outcome in patients hospitalized with decompensated liver cirrhosis.
MATERIAL AND METHODS: In this prospective study, 126 patients were enrolled and followed up for 29 months. For each patient, prognostic scores were calculated; these included the Child-Turcotte-Pugh score (CTP score), CTP creatinine-modified I score, CTP creatinine-modified II score, Model for End-Stage Liver Disease (MELD score), MELD model for end-stage liver disease sodium-modified score, Integrated MELD score, updated MELD score, United Kingdom MELD, and the MELD score remodeled by serum sodium index (MESO index). Cox regression analysis was used to assess the ability of each of the scores for predicting mortality in patients with alcoholic cirrhosis. Their discriminatory ability was evaluated using receiver operating characteristic (ROC) curve analysis.
RESULTS: The updated MELD score had the highest predictive value (3.29) among the tested scores (95% CI: 2.26-4.78). ROC curve analysis demonstrated that the MELD score of 22.50 (AUC = 0.914, 95% CI: 0.849-0.978; p < 0.001) had the best discriminative ability for identifying patients with a high risk of mortality; the next best was the MESO index of 16.00 (AUC = 0.912, 95% CI: 0.847-0.978; p < 0.001).
CONCLUSION: The risk of mortality was highest in patients with the highest updated MELD score, and those with MELD scores >22.50 and a MESO index >16.00.