#66 - In-Hospital Mortality of Patients With Complicated Cirrhosis: Development of a Predictive Model.

This is the authors' abstract. We don't add key points for this paper.

Introduction and Objectives Complications of hepatic cirrhosis are a major cause of hospitalization and mortality among affected patients. To determine clinical predictors of in-hospital mortality in patients with complicated cirrhosis. Materials and Methods An observational, analitycal, retrospective study was conducted in the Gastroenterology Service of a tertiary-level institution between January 2022 and December 2024. The sample included 104 patients who met the selection criteria. Variables collected were Child-Pugh score, clinical stage (D’Amico), cirrhosis-related complications, history of prior decompensation, recent hospitalization (within the last 3 months), and discharge status (alive/deceased). Associations with mortality were evaluated using ² or Fisher’s exact test. A LASSO (L1) logistic regression model was constructed to identify predictors of in-hospital mortality, with a significance level of = 0.05. Results The mean age was 59.3 ± 11.0 years, and 59.6% were men. Toxic alcohol use was reported in 41.3%. Ascites was the leading cause of admission (41.3%) and the most frequent complication (76%), followed by hepatocellular carcinoma (26.9%) and encephalopathy (23.1%). Most patients were classified as D’Amico stage 5 (48.1%) and Child-Pugh class B (50.9%). Overall mortality was 14.4%. The LASSO model identified six predictors of in-hospital mortality: prior decompensation, Child-Pugh class C, D’Amico stage 6, portal hypertension–related gastrointestinal bleeding, hepatorenal syndrome, and ACLF. The model demonstrated good discriminative capacity (p < 0.05) and acceptable calibration despite class imbalanced (survivors = 89; deceased = 15). Conclusions In-hospital mortality in patients with complicated cirrhosis is associated with advanced disease stages and major complications.