OBJECTIVETo investigate the effect of different nucleoside analogues on the long-term survival rate of patients with acute-on-chronic liver failure (ACLF) associated with hepatitis B virus (HBV) infection.

METHODSOne hundred and eighty patients with HBV-related ACLF were enrolled in this prospective cohort study and divided into a basic treatment group (n=30) and an antiviral treatment group, the latter of which was further subdivided into the lamivudine treatment group (n=66), telbivudine treatment group (n=38) and entecavir treatment group (n=46) according to voluntary choice by the patient.All study participants were followed-up for 24 months. The Kaplan-Meier method was applied for survival analysis.

RESULTSThe patients in the four antiviral treatment groups had statistically similar baseline clinical characteristics and 1-month survival rates (Breslow =4.475, P=0.215).However, the basic treatment group had a significantly lower survival rate than the antiviral treatment groups that received lamivudine, telbivudine, or entecavir (all P less than 0.05) at the treatment periods of 2, 3, 6, 12 and 18-months; however, these three treatment groups showed no significant differences in survival rates. At the time point of 24 months of treatment, the basic treatment group retained its lower rate of survival than the three antiviral treated groups (lamivudine:Breslow =5.604, P=0.018; telbivudine:Breslow =5.621, P=0.018; entecavir:Breslow =14.701, P less than 0.001); while the survival rates were similar for the lamivudine treatment group and the telbivudine treatment group at this time point, their survival rates were significantly lower than that of the entecavir treatment group (Breslow =4.010, P=0.045; Breslow =4.307, P=0.038).Stratification analysis showed that when the baseline was 30 less than PTA less than or equal to 40 or MELD less than or equal to 29 or HBV DNA more than or equal to 5 log10 IU/mL, the cumulative survival rates of the basic treatment group and antiviral treatment group were statistically similar even though the patients had completed 1 month of treatment After being treated for 2, 3, 6, 12, 18 and 24 months, the cumulative survival rates of the basic treatment group were consistently below those of the overall antiviral treatment group (P less than 0.05). The cumulative survival rate of the basic treatment group followed-up for 1 to 24 months, with PTA values between 20 and 30, was lower than that of the overall antiviral treatment group (P less than 0.05); two groups of patients with PTA less than or equal to 20 or MELD more than or equal to 30 were followed-up for 1 months to 24 months, and their cumulative survival rates showed no significant difference (P more than 0.05). Among the patients whose baseline was HBV DNA less than 5 log10 IU/mL, the comparison of survival rates between the basic treatment group and the overall antiviral treatment group showed no significant differences after treatment for 1, 2, 3, 6, 12 or 18 months, and the survival rate was lower than that of the overall antiviral treatment group (Breslow =4.055, P=0.044) after 24 months.

CONCLUSIONNucleoside analogues can improve the long-term survival rate of HBV-related ACLF patients.Entecavir is preferred for the long-term treatment of these patients.Patients in the early and middle stages of this disease and HBV DNA-positive patients should adopt antiviral treatment as early as possible.