Introduction

Whereas the prevalence of all pelvic organ prolapse has been estimated to exceed 11% in several studies, the prevalence of posterior-compartment prolapse (rectocele) specifically is not known. Posterior vaginal wall relaxation is most simply a herniation of the anterior rectal wall that produces a vaginal bulge. The bulge is purely an anatomic finding and can be associated with a wide spectrum of clinical symptoms, ranging from a relative absence of symptoms to significant pain, constipation, and splinting to evacuate the bowels. One of the challenges of treating rectoceles has been that a direct correlation between anatomic findings and clinical symptoms does not always exist. The authors have found that operative repair symptomatically improved a majority of patients with impaired defecation associated with a large rectocele, but this improvement was likely related at least in part to factors other than the size of the rectocele. Multiple surgical techniques are available for rectocele repair, and the literature is not clear regarding indications for each type of surgical intervention. The object of this article is to review the literature regarding various types of posterior-compartment repair, and draw conclusions regarding their absolute efficacy and relative efficacy in comparison with one another. This review is limited to standard repairs and repairs augmented with biological grafts. Repairs using nonabsorbable synthetic mesh, with or without the addition of commercial, transvaginal prolapse kits, are not covered.

Anatomy of posterior-compartment laxity

DeLancey subdivided the vaginal support of the anterior and posterior compartments into 3 levels. Level I support consists of the cardinal/uterosacral ligament complex, which originates at the cervix and upper vagina and inserts at the sacrum and pelvic side wall. In the posterior compartment, the vagina is separated from the rectum by the trapezoidal rectovaginal septum, with the narrow end located distally. The paravaginal attachments constitute Level II support. While Level II vaginal support in the anterior compartment is provided entirely by the arcus tendineus fasciae pelvis (ATFP), the attachment of the posterior vaginal wall to the pelvic side wall is more complex. The distal one-third to one-half of the posterior vaginal wall fuses with the aponeurosis of the levator ani muscle from the perineal body along a line called arcus tendineus rectovaginalis. This line converges with the ATFP approximately halfway between the pubic symphysis and the ischial spine. The rectovaginal septum is fused distally with the urogenital diaphragm and proximal perineal body (Level III support).

Defects of the posterior compartment mirror those found in the anterior compartment in several ways. Loss of Level II support may be central or lateral, and these types of defects have traditionally been repaired with a plication of the rectovaginal fascia in the midline (posterior colporrhaphy). In addition, a site-specific posterior repair that repairs discrete rents in the rectovaginal fascia may be performed instead of a midline plication. Proximal detachment of the rectovaginal septum from the uterosacral ligaments (Level I support) may be associated with an enterocele and repair of concomitant apical prolapse (abdominal sacral colpopexy and uterosacral, sacrospinous, or iliococcygeus suspensions) may be necessary. Loss of Level III support may result in perineal weakness and may be repaired by reapproximating the perineal body (perineorrhaphy).

Anatomy of posterior-compartment laxity

DeLancey subdivided the vaginal support of the anterior and posterior compartments into 3 levels. Level I support consists of the cardinal/uterosacral ligament complex, which originates at the cervix and upper vagina and inserts at the sacrum and pelvic side wall. In the posterior compartment, the vagina is separated from the rectum by the trapezoidal rectovaginal septum, with the narrow end located distally. The paravaginal attachments constitute Level II support. While Level II vaginal support in the anterior compartment is provided entirely by the arcus tendineus fasciae pelvis (ATFP), the attachment of the posterior vaginal wall to the pelvic side wall is more complex. The distal one-third to one-half of the posterior vaginal wall fuses with the aponeurosis of the levator ani muscle from the perineal body along a line called arcus tendineus rectovaginalis. This line converges with the ATFP approximately halfway between the pubic symphysis and the ischial spine. The rectovaginal septum is fused distally with the urogenital diaphragm and proximal perineal body (Level III support).

Defects of the posterior compartment mirror those found in the anterior compartment in several ways. Loss of Level II support may be central or lateral, and these types of defects have traditionally been repaired with a plication of the rectovaginal fascia in the midline (posterior colporrhaphy). In addition, a site-specific posterior repair that repairs discrete rents in the rectovaginal fascia may be performed instead of a midline plication. Proximal detachment of the rectovaginal septum from the uterosacral ligaments (Level I support) may be associated with an enterocele and repair of concomitant apical prolapse (abdominal sacral colpopexy and uterosacral, sacrospinous, or iliococcygeus suspensions) may be necessary. Loss of Level III support may result in perineal weakness and may be repaired by reapproximating the perineal body (perineorrhaphy).

Success rates after standard posterior-compartment repair

Although there are no long-term prospective studies, anatomic cure rates after posterior-compartment repair appear to be high. Maher and colleagues performed a prospective evaluation of 38 consecutive women with symptomatic rectoceles (≥stage II) and obstructed defecation who underwent midline fascial plication of the posterior vaginal wall. The median follow-up was 12.5 months and the subjective success rates were 97% (95% confidence interval [CI] 0.83%–1.00%) at 12 months and 89% (95% CI 0.55%–0.98%) at 24 months. The objective success rates were 87% (95% CI 0.64%–0.96%) at 12 months and 79% (95% CI 0.51%–0.92%) at 24 months. The average points, Ap and Bp, were significantly reduced from preoperative values and the depth of rectocele was also reduced significantly on defecography. The correction of the anatomic defect was associated with improved functional outcome, with 33 women (87%) no longer experiencing obstructed defecation, and there was a significant reduction in postoperative straining to defecate, hard stools, and dyspareunia. The improved anatomic and functional outcomes were reflected in the fact that 97% of the women reported very high satisfaction with their surgery.

The data also support high success after site-specific repairs. Cundiff and colleagues performed site-specific repair without levator plication or perineorrhaphy in 69 women with a median preoperative posterior Pelvic Organ Prolapse Quantitation (POP-Q) stage II. The POP-Q stage had improved for all but 2 women at 6 weeks, and 18% had recurrent rectoceles at 12 months. Mean values for the points describing the posterior vaginal wall improved by greater than 2 cm ( P <.0001) and functional results mirrored anatomic results, with statistically significant improvement in constipation, splinting, tenesmus, and fecal incontinence. Kenton and colleagues reported on 66 patients with abnormal fluoroscopic results and objective rectocele formation, 46 of whom were objectively assessed at 12 months. Resolution of postoperative symptoms was as follows: protrusion, 90% (35 of 39; P <.0005); difficult defecation, 54% (14 of 24; P <.0005); constipation, 43% (9 of 21; P = .02); dyspareunia, 92% (11 of 12; P = .01); and manual evacuation, 36% (4 of 11; P = .125). Vaginal topography at 12 months was improved, with a mean Ap point value of −2 cm (range, −3 to 2 cm).

Porter and colleagues performed a retrospective observational study that included 125 women who had undergone site-specific posterior colporrhaphy, either alone or in conjunction with other pelvic procedures. At follow-up examination, surgical correction was achieved in 82% of eligible patients (73 of 89). All daily aspects of living improved significantly ( P <.05), including ability to do housework (56% improvement or cure), travel (58% improvement or cure), and social activities (60% improvement or cure). Emotional well-being also significantly improved after the operation, as measured by thoughts of embarrassment (57% improvement or cure) or frustration (71% improvement or cure). Sexual function was not affected; however, reports of dyspareunia significantly improved or were cured after the operation in 73% of women (19 of 26), worsened in 19% of women (5 of 26), and arose de novo in 3 women. Results showed no other significant differences in vaginal dryness, orgasm ability, sexual desire, sexual frequency, or sexual satisfaction. Bowel symptoms were assessed subjectively and were noted to improve significantly postoperatively ( P <.008). The following improvement or cure rates were obtained: defecation difficulties, 55%; pelvic pain or pressure, 73%; vaginal mass, 74%; and splinting, 65%. Singh and colleagues reported on 42 women with symptomatic rectocele (≥stage II) who underwent site-specific repair. Forty of 42 women (95%) were assessed at 6 weeks and 78.5% (33 of 42) followed up at 18 months. At 6 weeks, vaginal protrusion resolved in 87.5% and bowel symptoms in 87%, while at 18 months there was anatomic cure in 92%, improvement in defecation in 81%, and improvement of sexual dysfunction in 35%.

Conversely, Sardeli and colleagues recently reported a case series of 51 women who underwent site-specific rectocele repair under local anesthesia. The mean follow-up period was 26.7 months. Whereas pelvic examination revealed recurrence of posterior prolapse in 31% (16 of 51), improvement in rectal emptying was achieved in only 23% (7 of 30), and 23% (7 of 30) of women experienced relief from constipation. One patient developed de novo dyspareunia. Overall, 92% of the patients (47 of 51) would recommend local anesthesia.

All of these studies are limited by their retrospective nature, relatively brief follow-up periods, variations in patient populations and surgical technique, definitions of success and failure, and indications for repair. These results should thus be interpreted with caution.