A Nomogram Incorporating Sarcopenia and Nutritional Indicators for Mortality Prediction in HBV-Related Acute-Chronic Liver Failure

TL;DR

This study creates a tool to predict 12-week mortality in patients with HBV-related liver failure by combining nutritional and muscle indicators with traditional factors.

Contribution

A novel nomogram integrating sarcopenia and nutritional indicators for mortality prediction in HBV-related ACLF patients.

Findings

The nomogram achieved high discrimination with an AUC of 0.95 for predicting 12-week mortality.

Incorporating nutritional indicators like ceruloplasmin and skeletal muscle index improved risk stratification.

The model showed excellent calibration and clinical utility through validation methods.

Abstract

Background: The prognosis of acute-on-chronic liver failure (ACLF) is impaired by etiology heterogeneity across regions. Currently, prognostic models incorporating nutrient anabolism–related indicators for patients with hepatitis B virus (HBV)–associated ACLF are lacking. Objectives: This study aimed to construct a nomogram that incorporates nutrition-related indexes alongside traditional predictors to estimate 12-week mortality in HBV-ACLF. Methods: We retrospectively analyzed adult patients with HBV-ACLF treated at our department between May 2020 and December 2021. A total of 242 HBV-ACLF patients were enrolled and categorized into survivor (n = 174) and progression (n = 68) groups. Independent prognostic factors were identified using logistic regression analysis and incorporated into a nomogram. Nomogram performance was evaluated in terms of discrimination, calibration, and clinical utility, with internal validation using bootstrap resampling. Results: Patients in the progression group were older, more prone to hepatorenal syndrome and spontaneous peritonitis, and had lower levels of prothrombin activity, L3 skeletal muscle index and ceruloplasmin (all p < 0.05). These six independent predictors were incorporated into the nomogram, which demonstrated superior discrimination ability, with an area under the receiver operating characteristic curve of 0.95, enabling accurate identification of patients at high risk of short-term mortality. The Hosmer–Lemeshow test confirmed excellent calibration, decision curve analysis confirmed the clinical benefit, and bootstrap validation confirmed the robustness. Conclusions: The developed nomogram, incorporating nutritional status, may provide complementary information to support short-term risk stratification and clinical decision-making in patients with HBV-ACLF awaiting liver transplantation.

Figures