Hepatic Encephalopathy and Age Were Assessed to Ameliorate the Underestimation of the Model for End‐Stage Liver Disease Score in Patients With Acute‐on‐Chronic Liver Failure

Objective To quantify the prognostic inadequacy of the model for end‐stage liver disease (MELD) score in acute‐on‐chronic liver failure (ACLF) and to develop a nomogram incorporating hepatic encephalopathy and age to improve 90‐day mortality prediction. Methods A retrospective cohort of 1364 ACLF patients and 857 non‐ACLF patients was assembled. Propensity score matching (PSM) was applied to create balanced comparison groups. Independent determinants of 90‐day mortality in ACLF were identified through logistic regression, and a nomogram integrating these factors was constructed, followed by comprehensive internal and external validation. Results PSM‐based comparisons revealed that the area under the receiver operating characteristic curve (AUC) of MELD score for 90‐day mortality prediction in ACLF patients was only 0.556 (95% CI: 0.493–0.619), significantly inferior to that of matched non‐ACLF controls (AUC: 0.848, 95% CI: 0.801–0.887; difference: 0.291, p < 0.0001). Multivariate analysis identified age, hepatic encephalopathy (HE), total bilirubin, sodium, creatinine, and the international normalized ratio as independent prognostic variables. A nomogram incorporating age, HE, and MELD score achieved AUCs of 0.760 (95% CI: 0.733–0.785) in the training cohort and 0.837 (95% CI: 0.791–0.877) in the external validation cohort, outperforming all evaluated MELD variants. Conclusion The proposed nomogram, combining age, HE, and MELD score, demonstrates superior predictive performance over conventional MELD‐based models for 90‐day mortality in ACLF patients and may serve as a clinically practical tool to mitigate allocation inequity for ACLF candidates on liver transplantation waiting lists.