Objective To evaluate the effects of intraoperative mild hypothermia on the prognosis of patients undergoing craniotomy.Methods A literature search was performed in CNKI, CBM, Wanfang Data, VIP, PubMed, and Cochrane Library to identify the randomized controlled trials (RCTs) about intraoperative mild hypothermia in craniotomy published up to August 2016. The obtained data were subjected to meta-analysis using RevMan5.3 software.Results A total of 6 RCTs involving 1306 patients were enrolled in the meta-analysis. The patients were divided into two groups: mild hypothermia group (n=657) and normothermia group (n=649). There were no significant differences between the two groups in Glasgow Coma Scale score [minor or no disability (OR=1.24, 95% CI: 0.97-1.58, P=0.08), moderate disability (OR=0.97, 95% CI: 0.73-1.30, P=0.86), severe disability (OR=0.77, 95% CI: 0.50-1.18, P=0.23), vegetative state (OR=1.35, 95%CI: 0.32-5.65, P=0.68), and death (OR=0.71, 95% CI: 0.47-1.09, P=0.12)] and postoperative complications (P>0.05). However, the mild hypothermia group had a significantly smaller arterio-venous difference of oxygen than the normothermia group (P<0.05).Conclusions Intraoperative mild hypothermia cannot improve the prognosis of patients undergoing craniotomy and reduce the incidence of postoperative complications.