Abstract: Objective The aim of this study was to explore the risk factors of death and the construct a prognosis prediction model in patients with hepatitis B virus-related acute on chronic liver failure (HBV-ACLF). Methods 126 patients with HBV-ACLF were admitted to our hospital between January 2016 and September 2019,and their clinical materials were recorded. The multivariate Logistic regression was applied to set up a prediction model. The model fit was evaluated by Hosmer-Lemeshow test, and the predictive value of the model was verified by ROC curves. Results At the end of 90 days, 49 patients(38.9%) in our series died; the multivariate Logistic analysis showed that total serum bilirubin levels, red blood cell distribution width (RDW),INR) and with complications were the independent risk factors impacting the prognosis; the Hosmer-Lemeshow test demonstrated the formula, P=1/[1+e^{(1.441×(TBIL)+1.385×(RDW)+0.983×(INR)+1.639×complications-4.449)}] we established, goodness-of-fit (x²=9.055, P=0.338), and the AUC of the model was 0.822, with the 95%CI of 0.749-0.894.Conclusion We successfully set up a formula to predict the prognosis of patients with HBV-ACLF, which might guide the clinicians to make up a correct decision early.

Key words: Acute on chronic liver failure, Hepatitis B, Risk factors, Prognosis, Prognostic model