Accidental needlestick injury during cesarean delivery closure exposes surgeons to percutaneous injury and risk of human immunodeficiency virus (HIV), viral hepatitis, and other blood-borne diseases. In a number of surgical procedures, use of blunt or rounded-tip needles has been associated with a marked decrease in glove perforations (a proxy for percutaneous injuries). However, no study to date has investigated their use in cesarean delivery. The primary aim of this randomized unblinded trial was to compare rates of glove perforation in surgeons using blunt and sharp needles during cesarean-delivery closure. A secondary aim was to assess surgeons' acceptance and satisfaction of blunt needles. A total of 194 patients requiring nonemergent cesarean delivery were randomized to receive either blunt (study group, n = 97) or sharp needles (control group, n = 97) for closure. Surgeons and their assistants completed a survey providing information on their experience level, estimated blood loss during the procedure, prevalence of glove perforations or percutaneous injuries, their satisfaction with the assigned needles and other clinical variables. After surgery, glove perforation was determined using the water test method by an investigator who was blinded to which needle type had been assigned and to any reported glove perforations. No statistical differences between the groups in patient demographics or obstetric and clinical variables were observed. Use of the blunt needles was associated with a statistically significant reduction in the total glove perforation rate in comparison with the sharp needles (7.2% [7/97] vs. 17.5% [17/97]; the relative risk [RR] was 0.66, with a 95% confidence interval [CI] of 0.49-0.89). The risk reduction was significant only for assistant surgeons (RR, 0.54; 95% CI, 0.41-0.71). There was no significant reduction in risk of perforation among the primary surgeons (RR, 0.8; 95% CI, 0.53-1.2). Only 16 of the 24 perforations found with the water test were reported in the physician survey. Thus there was poor overall correlation between perforations detected by the water test and those reported by the surgeons (R 2 = 0.3) and their assistants (R 2 = 0.37). Both surgical groups reported less overall satisfaction with the use of blunt needles compared with sharp needles (P < 0.001 for each group) but over 90% of each group rated the blunt needles as at least acceptable. The findings suggest that use of blunt needles is an effective procedure to minimize the risk of accidental needlesticks during cesarean delivery closure. It must be noted, however, that the risk reduction was observed only for assistant surgeons.