The data are limited on the long-term clincial outcome of monotherapy versus combinations of nucleos(t)ide analog (NUCs) for hepatitis B related decompensated cirrhotic patients. This study was to evaluate the efficacy in treatment-naive patients using NUCs monotherapy or combinations. Three hundred and six patients with decompensated hepatitis B cirrhosis were selected from cirrhosis cohort and divided into treatment-naive (n = 260) and control groups (n = 46). Antiviral therapies included monotherapy of lamivudine (LAM, n = 39), adefovir (ADV, n = 73), telbivudine (LDT, n = 36), entecavir (ETV, n = 48), and combinations of LAM+ADV (n = 39) or LDT+ADV (n = 25). Of these patients, 193 in antiviral therapy and 39 in control group were included for analysis over two years. The cumulative drug-resistance rate at two year was higher in the LAM (37.9%), ADV (21.2%), LDT (23.3%) than in the ETV monotherapy (2.6%), and with combinations of LAM+ADV (8.7%) or LDT+ADV (6.3%), respectively, P < 0.001. Serum hepatitis B virus (HBV) DNA undetectability in the ETV and the LDT+ADV group was higher than in the LAM, ADV and LAM+ADV group at over two years (P < 0.05). The child-pugh score (CPs) in the antiviral therapy group was decreased at two years (P < 0.05). In the control group and drug-resistant patients, however, CPs was increased. The two years cumulative incidence of liver failure in the antiviral therapy group was significantly less than the control group (OR 24.9, 95%; CI 6.5 to 94.7, P = 0.001). The total cumulative survival rate in the antiviral therapy group was higher than in control group (OR 4.2, 95%; CI 1.4 to 12.9, P = 0.017). The combinations of NUCs therapy and ETV momotherapy are optimum management for hepatitis B related decompensated cirrhotic patients.