The aim of this study was to compare the results of two surgical techniques (direct end-to-endvs.overlapping) of delayed repair of a localized anterior defect of external anal sphincter after an obstetric trauma. During a five-year period, 23 patients were randomly assigned to direct end-to-end repair (n = 12) or overlapping sphincter repair (n = 11), using 2-0 PDS™ sutures. Two patients from each group had an internal anal sphincter defect that also was repaired. All patients had a normal pudendal nerve terminal motor latency preoperatively. Evaluations included endoanal ultrasound, anorectal manometry, and neurophysiologic evaluation. Continence was assessed by the Cleveland Clinic Continence Score (0-20; 0, perfect continence; 20, complete incontinence). The two groups were comparable with regard to age (median, 45 years), past history of sphincter repair (n = 2), and posterior vaginal repair. There was no major morbidity. The wound-healing rate was identical between the two groups. However, of the patients undergoing overlapping repair, two had fecal impaction, and one had a urinary retention. Median preoperative continence score was 17 in both the direct-repair group (score, 8-20) and the overlap group (score, 7-20). At a median follow-up of 18 months, the improvement in continence was similar between the two surgical groups, with a median continence score of 3, respectively. In both surgical groups there was a significant and similar improvement in maximum squeeze pressure and in the functional anal canal length postoperatively (P< 0.05), but the mean resting pressure was relatively unchanged. In the overlap group, one patient developed a unilaterally prolonged pudendal nerve terminal motor latency that was persistent 22 months after surgery, and two patients had impaired fecal evacuation postoperatively. This randomized, controlled study suggests that the outcome is similar whether direct end-to-end or overlapping repair of a sphincter defect is performed. Overlapping repair might be associated with more difficulties with fecal evacuation and a prolonged pudendal nerve terminal motor latency postoperatively.