Background Bladder cancer is widely known as the most common malignant tumor in the urinary tract,with 75％-85％ of patients suffering from nonmuscle invasive bladder cancer (NMIBC).However,the optimal dose of Bacillus Calmette-Guérin (BCG) remains controversial.The aim of this study was to compare the therapeutic efficacy of full dose (FD) with the reduced dose (RD) of BCG.Methods Randomized controlled trials (RCTs) were selected through the Cochrane Library,PubMed and Embase and were supplemented by hand searching of bibliographies.The end points include overall survival rate,recurrence rate,progression rate and side effects.Results Five RCTs that included a total of 1 473 patients (727 in the reduced dose group vs 746 in the full dose group),with a median follow-up period from 33.5 month to 7.1 year.Disease in 80 of 687 (11.6％) patients assigned to the RD group progress to the muscular layer or distant metastasis,compared with 81 of 698 (11.6％) patients assigned to the FD group (RR=1.02; 95％ CI,0.77-1.36; P=0.89).The incidence of recurrence at three year was reported in all five studies to be 41.1％ (299 of 727) and 36.1％ (269 of 746) in the RD and FD groups,respectively (RR=1.13; 95％ CI,1.00-1.29; P=0.05).The 5-year survival rate was 75.9％ (502 of 662) in the RD group,and 75.8％ (510 of 673) in the FD group.In the RD group 41 of 655 (6.3％) patients and 56 of 663 (8.7％) patients in the FD group did not complete the treatment due to systemic or local side effects (RR=0.75; 95％ CI,0.51-1.10; P=0.14) Conclusions In general,the results of our study demonstrate a trend towards a reduction of the toxicity in reduced dose group without affecting the efficacy of treatment when compared with full dose.More trials with large sample size are still necessary to explore the prognosis of the patients with high risk of tumor in different dose group.