Abstract
Background and aim Acute-on-chronic liver failure (ACLF) is a dynamic condition with very high short-term mortality. Although several mortality predictors have been studied, no single factor can reliably predict its course and outcome. This study aimed to evaluate various biochemical parameters and severity scores at presentation to assess their ability to predict mortality in ACLF patients. Methods An observational cross-sectional study was conducted from March 2024 to February 2025 at a tertiary care center in Nepal. A total of 51 patients were enrolled over one year. ACLF was diagnosed based on the Asia-Pacific Association for the Study of the Liver (APASL) criteria and/or the European Association for the Study of the Liver-Chronic Liver Failure Consortium (EASL-CLIF) criteria. Clinical features, biochemical parameters, and severity scores at admission were analyzed. Results The overall mortality rate was high (56.9%), consistent with previous studies. About 28% of patients who met the APASL definition of ACLF did not fulfill the EASL-CLIF criteria. On univariate analysis, factors such as oliguria, hepatic encephalopathy (HE) grade, ACLF grade, total serum protein, urea, ascitic fluid total leukocyte count, serum lactate, and APASL-ACLF Research Consortium (AARC) score were associated with mortality. However, multivariate analysis did not identify a single independent predictor of mortality. Different scoring systems were useful in predicting mortality at admission. Among them, the CLIF-C ACLF score had the highest predictive accuracy (area under the curve (AUC): 0.990), followed by the AARC score (AUC: 0.79), Sequential Organ Failure Assessment (SOFA) score, and Model for End-Stage Liver Disease (MELD) score. However, the Child-Turcotte-Pugh (CTP) score at admission was not effective in predicting mortality. Conclusion ACLF represents an acute flare of inflammation on a background of chronic liver disease. Using multiple mortality predictors at the time of admission can help guide treatment decisions and provide better prognostication of patient outcomes.