Risk factors for the development of cirrhosis within 1-year in non-cirrhotic patients with HBV- related acute-on-chronic liver failure

TL;DR

This study identifies key factors that predict whether non-cirrhotic patients with HBV-related liver failure will develop cirrhosis within a year, helping doctors make better treatment decisions.

Contribution

The study introduces a risk stratification algorithm based on INR normalization, bilirubin resolution, and platelet count to predict cirrhosis progression in HBV-ACLF patients.

Findings

21.1% of non-cirrhotic HBV-ACLF patients developed cirrhosis within one year.

Non-resolution of bilirubin and INR within 28 days, and low baseline platelet count were strong predictors of cirrhosis.

A risk stratification model using INR and platelet count categorized patients into low, medium, and high-risk groups with cirrhosis rates of 5.4%, 29.4%, and 77.8%.

Abstract

This study aimed to provide a preliminary clarification of the predictive factors for the development of cirrhosis within 1-year cirrhosis in non-cirrhotic HBV-associated acute-on-chronic liver failure (HBV-ACLF) patients and develop a risk stratification algorithm.

Non-cirrhotic HBV-ACLF patients who survived for at least 1 year and had complete clinical records were included from January 2016 to December 2023 at Beijing Youan Hospital, Capital Medical University. Multivariate logistic regression was performed to identify independent predictors of cirrhosis progression.

Of the 329 HBV-ACLF patients initially screened, 109 were enrolled in the study, and a 1-year follow-up revealed that 21.1% of the non-cirrhotic patients developed cirrhosis. Multivariate logistic regression identified independent predictors of cirrhosis progression: non-resolution of total bilirubin (TB) levels within 28 days [5.64 (1.39–25.00)], failure to normalize the international normalized ratio (INR) within 28 days [6.34 (1.52–29.50)], and baseline platelet (PLT) count [0.98 (0.97–0.99)]. ROC analysis demonstrated strong predictive accuracy for cirrhosis with INR normalization (AUC = 0.82), TB resolution (AUC = 0.78), and baseline PLT count (AUC = 0.75). A risk stratification pathway incorporating INR normalization and baseline PLT effectively categorized patients into low, medium, and high-risk groups, with corresponding cirrhosis incidence rates of 5.4%, 29.4%, and 77.8%, respectively.

The findings underscore the importance of INR normalization, TB resolution, and baseline PLT count as independent predictors of cirrhosis and provide a useful framework for clinical decision-making and early intervention.

Figures