Objective  To evaluate the therapeutic effect of lamivudine in the treatment of HBV-related liver failure.   Methods  We collected the literature on lamivudine used to treat liver failure in randomized control trail or clinical control trial studies from January 1989 to July 2007. All these clinical trials were carried out by comparing lamivudine treatment with routine medical treatment for liver failure. The literature must concern about mortality

total bilirubin (TBIL) or prothrombin activity (PTA). Odds ratio (OR) was applied to evaluate the effect of glucocorticoid to improve case fatality. Weighted mean differences (WMD) was applied to evaluate the effect of  lamivudine to improve TBIL and PTA.   Results  There were only 13 literatures obtained

among them all about mortality

and 10 about TBIL and PTA. Effective index was 0.29 (95％CI: 0.21

0.40)

-0.88 (95％CI: -1.27

-0.48) and 0.60 (95％CI: 0.42

0.78) respectively for the 3 indexes.   Conclusion  Combination with lamivudine is more effective in treatment liver failure than only routine medical treatment.