Abstract: Objective To evaluate the efficacy and safety of glucocorticoids on severe viral hepatitis. Methods We searched Cochrane Central Register of Controlled Trials（The Cochrane Library，Issue 11，2010）， PubMed，OVID，CNKI and Wanfang Data（2000 to 31 October 2010）to collect case-control trials of glucocorticoids for severe viral hepatitis. The language was confined to English and Chinese. We screened the retrieved studies according to the predefined inclusion and exclusion criteria，evaluated the quality of included studies，and performed Meta-analyses using the Cochrane Collaboration’s RevMan 5.0 software. Results Thirteen case-control trials involving 770 patients with severe viral hepatitis were included. All included trials were with average 1.6 of modified Jadad scores. Meta-analysis showed that the mortality of the patients in glucocorticoids group was significantly lower than in control group[RR=0.39，95%CI（0.30 to 0.49），P<0.00001]. Except for some controllable special adverse events found in some patients in glucocorticoids group，one trial showed that the incidence of major complications，such as infections，hemorrhage，hepatic encephalopathy and hepatorenal syndrome，was significantly lower in glucocorticoids group than in control group，which was not performed Meta-analysis because of too small sample size. Conclusion The patients treated with glucocorticoids in a timely and reasonable manner in the early stage of severe hepatitis could obtain better efficacy，lower mortality and no more incidence of complications. However，the accuracy of the results is weakened by the poor quality and quantity of the included trials. Thus，more randomized controlled trials of high quality are required to verify these conclusions.

Key words: Severe hepatitis, Glucocorticoids, Treatment, Meta-analysis