Anal incontinence and obstetric sphincter tears

PURPOSE: To evaluate the prevalence of anal incontinence (AI) in primiparous women five years after their first delivery; to investigate if an 8-12 hour time delay of primary repair affects AI at 1-year follow-up; to study the association between endoanal ultrasonography and anorectal manometry in relation to AI in primiparous women; to evaluate whether an endoanal ultrasonographic derived sphincter length-index correlates to subjective AI. METHODS: Primiparous women were prospectively followed with questionnaires before the pregnancy, at 5 and 9 months, and 5 years after the delivery. 242/349 women completed all questionnaires. Women with sphincter tear at their first delivery were compared to women without such injury. Risk factors for development of AI were analyzed. 165 women diagnosed with obstetrical sphincter tear at delivery were randomised to immediate or delayed (8-12 hour delay) repair. 78 were allocated to immediate operation and 87 to a delayed repair and followed for 12 months. A nested case-control study of 108/165 women from the randomized controlled trial was performed to compare findings at anorectal manometry and endoanal ultrasound 1 year after their repair. Cases were defined as having a greater score than 2 on the Pescatori incontinence scale, controls as having a score of 2 or less. 116 primiparous women who had experienced a 3rd or 4th degree obstetric tear were examined using three-dimensional endoanal ultrasonography and compared to a control group of nineteen nulliparous women to evaluate a sphincter length-index. RESULTS: AI increased significantly during the 5-year follow-up. Among women with sphincter tears, 44% reported AI at nine months and 53% at five years (p=0.002). Risk factors for AI at five years were age (OR 2.2, 95% CI 1.0; 4.6), sphincter tear (OR 2.3, 1.1; 5.0) and subsequent childbirth (OR 2.4, 1.1; 5.6). In the randomized study of delayed vs. immediate sphincter repair there was no significant difference in AI between the groups. Of the manometric variables, volume of first sensation and volume of first urge proved to be associated with AI at univariate statistical analysis. A distal scar in the external anal sphincter, age, degree of tear, and duration of second stage of labor were independently associated with AI. Patients with a previous grade 3-4 obstetric injury had a shorter external anal sphincter (p