Diagnosis of anal sphincter tears by postpartum endosonography to predict fecal incontinence

Objective: To determine whether anal endosonography immediately after vagin al delivery can predict subsequent fecal incontinence. Methods: We studied nulliparas who delivered vaginally and had no anal sphi ncter tears (third- or fourth-degree perineal tears) diagnosed clinically b y endosonography before any suture of the perineum. The sonographer was una ware of delivery details and the obstetrician and the women were not inform ed of endosonography results. Therefore, the suture of the perineum and the outcomes were not influenced by sonographer's diagnoses. Three months afte r delivery, we assessed fecal incontinence by self-administered questionnai res. Results: Clinically undetected tears of the anal sphincter were diagnosed b y anal endosonography in 42 of 150 women (28%). The external anal sphincter alone was involved in 30 women (20%), the internal anal sphincter alone in two (1.3%), and both in ten (7%). The postal questionnaire was returned by 144 women. Incontinence was reported by 22 (15%, 95% confidence interval [ CI] 10%, 22%), consisting mainly of incontinence to flatus only (16 of 22, 73%, 95% CI 50%, 89%). Clinically undetected anal sphincter tears diagnosed by endosonography were associated with incontinence 3 months after deliver y (odds ratio [OR] 8.8; 95% CI 2.9, 26.5). The sensitivity of anal endosono graphy was 68% (95% CI 49%, 88%) and the positive predictive value 37% (95% CI 22%, 51%). Conclusion: Anal endosonography immediately after vaginal delivery allows d iagnosis of clinically undetected anal sphincter tears that might be associ ated with subsequent fecal incontinence. (Obstet Gynecol 2000;95:643-7. (C) 2000 by The American College of Obstetricians and Gynecologists.).