Risk factors and predictive model for the progression of HBV-related acute-on-chronic liver failure

Abstract

Objective To investigate the risk factors associated with the development of hepatitis B virus related acute-on-chronic pre-liver failure (pre-ACHBLF) to ACHBLF, and establish a new predictive model. Methods A total of 97 patients with pre-ACHBLF were divided into progress group and improvement group according to their clinical outcomes. Clinical data were collected. Multivariate logistic regression analysis was performed to evaluate the independent risk factors associated with the occurrence of ACHBLF, and establish a new predictive model. Results Total bilirubin, prothrombin time, serum creatinine and the ratio of platelet-white blood cell were independent risk factors. The risk model for predicting the occurrence of ACHBLE was established, which was Logit(P)=-1.494+1.530×TBIL(mg/dl)+3.111×PT(s)-1.711×PWR-1.786×Cr(mg/dl). Goodness of fit test Nagelkerke R2 coefficient was 0.558. Hosmer-Lemeshow test showed that the predicted performance was 76.3% (P=0.490). The area under receiver operating characteristic carve (ROC) curve was 0.83 and the optimal threshold was 13.88, suggesting that the patients with scores higher than 13.88 had an increased risk of developing ACLF. Conclusion The predictive model established in this study has a high value in predicting the risk of disease progression from pre-ACHBLF to ACHBLF.