Objective To assess the efficacy of finasteride in treating perioperative bleeding in patients undergoing transurethral resection of the prostate (TURP). Methods We searched MEDLINE (1966 to 2005), EMBase (1984 to 2004), CBM (1980 to 2005), The Cochrane Library (Issue 4, 2005) and relevant journals to identify cl inical trials involving finasteride in patients undergoing TURP. We also checked the references in the reports of each included trial. The qual ity of randomized controlled trials (RCTs) was assessed according to the methods recommended by The Cochrane Collaboration, and the qual ity of non-RCTs was assessed based on the methods recommended by Jiang-ping Liu, Stroup and Hailey. Two reviewers extracted data independently and data analyses were conducted with The Cochrane Collaboration’ s RevMan 4.2. Result We included 4 RCTs and 1 non-RCT. The qual ity of 3 RCTs was graded C and the other one was graded B. The quality of the non-RCT was relatively high. Meta-analyses showed that with comparable age, international prostate symptom score, prostate specific antigen, preoperative volume of prostate and excision volume between the two groups (Pgt;0.05), the perioperative bleeding volume (WMD –85.44, 95%CI –117.31 to –53.58), the bleeding volume per gram of resected prostate tissue (WMD –3.5, 95%CI –6.34 to –0.58) and hemoglobin reduction (WMD –1.61, 95%CI –1.96 to –1.26) of the finasteride group were significantly smaller than those of the control group. Conclusion The evidence currently available indicates that preoperative use of finasteride may reduce bleeding in patients undergoing TURP.