Analysis of prognostic risk factors and establishment of prognosis model in patients with hepatitis B virus-related acute-on-chronic liver failure

Objective To explore the risk factors for prognosis in patients with hepatitis B virus (HBV)-related acute-on-chronic liver failure (ACLF), and to establish a prognostic model. Methods A total of 193 patients diagnosed with HBV-ACLF who were admitted to the Department of Infectious Diseases of the Third Hospital of Hebei Medical University were collected from 1st January 2013 to 1st November 2018 as a derivation cohort. Thirty-five patients diagnosed with HBV-ACLF who were admitted to the Fifth Hospital of Shijiazhuang during the period from 1st July 2017 to 1st November 2018 were collected as a validation cohort. The survival condition of all patients at week 12 of admission was observed. The risk factors associated with short-term prognosis were analyzed by using multivariate logistic regression analysis, and a logistic regression equation prediction model was established and verified. The diagnostic performance of the prognostic model was evaluated using the receiver operating characteristic (ROC) curve, and was compared with model for end-stage liver disease (MELD) scoring system, Child-Turcotte-Pugh (CTP) scoring system, sequential organ failure assessment (SOFA) scoring system and chronic liver failure (CLIF)-SOFA scoring system. Results Multivariate logistic regression analysis showed that age (odds ratio(OR)=2.133, 95% confidence interval(CI)1.033-4.405), total bilirubin (OR=3.371, 95%CI 1.610-7.060), serum creatinine (OR=4.448, 95%CI 1.697-11.661), hepatic encephalopathy (OR=5.313, 95%CI 2.463-11.461), and ascites (OR=2.959, 95%CI 1.410-6.210) were independent risk factors for predicting the short-term prognosis of patients with HBV-ACLF. The newly established logistic regression model (LRM)=-1.726+ 0.757×age+ 1.215×total bilirubin+ 1.049 2×serum creatinine+ 1.670×hepatic encephalopathy (with=1, without=0) + 1.085×ascites (with=1, without=0). The area under the ROC curve of the LRM for predicting the short-term prognosis of patients with HBV-ACLF was 0.82 (95%CI 0.76-0.88). Furthermore, the areas under the ROC curve of the models of MELD, CTP, SOFA, CLIF-SOFA were 0.67 (95%CI 0.60-0.75), 0.73 (95%CI 0.66-0.80), 0.77 (95%CI 0.70-0.83) and 0.72 (95%CI 0.65-0.80), respectively. The ROC-area under curve of the validation cohort was 0.81 (95%CI 0.65-0.97). Conclusions Age, total bilirubin, serum creatinine, hepatic encephalopathy, and ascites are independent risk factors for the prognosis of HBV-ACLF. The prognostic model established based on these factors can accurately predict the patients′ short-term prognosis, which is superior to MELD, CTP, SOFA and CLIF-SOFA. Key words: Hepatitis B; Acute-on-chronic liver failure; Regression analysis; Prognostic model