Early Prediction of Acute-on-Chronic Liver Failure Development in patients with diverse chronic liver diseases

Abstract

Abstract Background & aims: Acute-on-chronic liver failure (ACLF) is a syndrome characterized by the acute decompensation of chronic liver disease, leading to organ failures and high short-term mortality. The course of ACLF is dynamic and reversible in a considerable proportion of patients during hospital admission. Early detection and accurate assessment of ACLF are crucial, yet ideal methods remain lacking. Therefore, this study is aimed to develop a new score for predicting the onset of ACLF in patients with diverse chronic liver diseases. Methods: A total of 6188 patients with diverse chronic liver diseases were included in the study. Clinical and laboratory data were collected, and the occurrence of ACLF within 28 days was recorded. Lasso-cox regression was utilized to establish prediction models for the development of ACLF at 7, 14, and 28 days. Findings: Among 5221 patients without ACLF, 477 patients progressed to ACLF within 28 days. Seven predictors were found to be significantly associated with the occurrence of ACLF at 7, 14, and 28 days. The new score had the best discrimination with the c-index of 0.958, 0.944, and 0.938 at 7, 14, and 28 days, respectively, outperforming those of four other scores(CLIF-C-ACLF-Ds, MELD, MELD-Na, and CLIF-C-ADs score, all P<0 .001). The new score also showed improvements in predictive accuracy, time-dependent receiver operating characteristics, probability density function evaluation, and calibration curves, making it highly predictive for the onset of ACLF at all time points. The optimal cut-off value (9.6) differentiated high and low-risk patients of ACLF onset. These findings were further validated in a separate group of patients. Conclusion: A new progressive score, based on seven predictors, has been developed to accurately predict the occurrence of ACLF within 7, 14, and 28 days in patients with diverse chronic liver diseases and might be used to identify high-risk patients, customize follow-up management, and guide escalation of care, prognostication, and transplant evaluation.