Anal continence after rectocele repair.

Abstract

Rectocele may be associated with both chronic constipation and anal incontinence. Several different surgical procedures have been advocated for rectocele repair. The aim of the present study was to evaluate anorectal function and clinical outcome in a consecutive series of patients who underwent selected endorectal or transperineal surgery for rectocele for whom operative treatment was determined by clinical and proctographic features. Attention was paid to the cohort of rectocele patients presenting with incontinence as a leading symptom. Sixty consecutive patients with symptomatic rectocele underwent surgical treatment at our institution. Fifty-eight of the patients were female (mean age 56; range, 21-70 years). Incontinence was graded according to a previously reported scoring system that accounts for the type and frequency of incontinence episodes. Preoperative anorectal manometry was performed using an open perfused polyethylene probe. Rectal sensation was recorded by balloon distention. Endoanal ultrasonography was performed with a 7.5-MHz probe. Preoperative defecography was performed at rest and on maximal squeeze and straining. Patients with obstructed defecation as their principal symptom, with associated mucosal rectal prolapse, underwent an endorectal procedure. For patients with associated anal incontinence (Grade B2 or greater), and without a rectal mucosal prolapse, a transperineal approach was performed with either an anterior external overlapping sphincteroplasty or levatorplasty. The median follow-up was 48 (range, 9-122) months. There was no operative mortality. Postoperative complications occurred in 18 patients (30 percent). Of 43 patients with incontinence, 34 (79 percent) were available for postoperative evaluation. None were fully continent. However, in 25 patients (73.5 percent), continence improved after surgery; half had only mucus soiling or loss of gas. Incontinence scores decreased (i.e., improved) from 4.8 +/- 0.9 to 3.9 +/- 0.9 (P = 0.002). A significant improvement was found both after transanal and perineal procedures. Only ten initially continent patients were available for postoperative assessment. All patients stated that they had clinical improvement in constipation. Their preoperative mean anal resting pressure was 62.5 +/- 3.9 (standard error of the mean) mmHg, with a postoperative mean of 75.5 +/- 7 mmHg. The preoperative mean squeeze pressure was 83.1 +/- 8.5 mmHg, with a mean postoperative squeeze pressure of 88.5 +/- 7.9 mmHg (P = not significant). The maximal tolerable volumes were all within normal limits, confirming the proctographic evidence that there were no cases of megarectum in our patient series. The pudendal nerve terminal motor latency was abnormal in all but two patients with incontinence (mean pudendal nerve terminal motor latency = 3.1; range, 1.2-4 milliseconds). Rectoceles recurred in six patients (10 percent): five after a Block procedure and one after a Sarles-type operation. The postoperative endosonographic appearance varied according to the nature of the procedure performed. There are few data concerning patients with rectocele who have associated anal incontinence, however, surgical decision analysis resulted in improvement in both constipation and incontinence in the majority of our patients with rectocele. Nevertheless, because none of the patients gained full continence postoperatively, pelvic floor rehabilitation might be also needed to achieve better sphincter function in patients with incontinence.