Surgical management of posterior vaginal defects

INTRODUCTION — The surgical approach to management of posterior vaginal wall defects (rectocele) will be discussed here. The clinical manifestations, diagnosis, and nonsurgical management of posterior vaginal defects are reviewed separately. (See "Posterior vaginal defects (eg, rectocele): Clinical manifestations, diagnosis, and nonsurgical management".)

In this topic, when discussing study results, we will use the terms "woman/en" or "patient(s)" as they are used in the studies presented. We encourage the reader to consider the specific counseling and treatment needs of transgender and gender-expansive individuals.

SURGICAL AND FUNCTIONAL ANATOMY

Histology — The apical portion of the posterior vaginal wall consists of mucosa (which includes the epithelium of the posterior wall and the lamina propria), a superficial and deep muscularis layer, and adventitia. This fibromuscularis layer has been called "rectovaginal fascia" and "perirectal fascia."

Histologic examination of the rectovaginal septum reveals that the distal portion near the perineal body contains dense connective tissue; the midportion has an adventitial layer containing fat, fibrous tissue, blood vessels, nerves, and elastic fibers; and the most proximal end is mostly adipose tissue [1]. The adipose tissue between the vaginal tube and rectum allows these two structures to function independently of one another.

Comparisons of the histology of women with and without prolapse have shown that the smooth muscle content of the posterior vaginal wall of women with prolapse is disorganized and significantly reduced compared with women without prolapse [2].

Normal anatomic support — The upper fourth of the posterior vaginal wall is suspended by the cardinal-uterosacral ligament complex. The middle half is attached laterally to the arcus tendineus fascia pelvis proximally and the arcus tendineus rectovaginalis distally. The lower fourth fuses into the perineal body.

Anteriorly, the perineal membrane spans the anterior half of the pelvic outlet and is made up of a dense fibromuscular layer. The perineal body extends cranially in the posterior wall of the vagina to 2 to 3 cm proximal to the hymenal ring. Posteriorly, the perineal body includes the anterior portion of the external anal sphincter and its attachment to the longitudinal fibrous sheath of the internal anal sphincter. In addition, the perineal body is suspended by and attached to the puborectalis muscle.

The levator ani also provides an important support mechanism for the vagina. Posteriorly, the puborectalis acts as a sling that angles the mid-posterior wall approximately 45° from vertical, and closes the potential space of the vagina, as well as the levator hiatus. Proximally, the vagina lies upon, and is supported by, the pubococcygeus and iliococcygeus muscles. With a healthy pelvic floor, there is little stress and strain placed on the connective tissue support system. The levator hiatus is larger in women with prolapse than in women with normal support [3].

Functional anatomy — In a woman with an intact pelvic floor, the puborectalis is in a chronic state of contraction. This contraction closes the vaginal canal such that the anterior and posterior vaginal walls are in direct apposition. During defecation, the anterior rectal wall expands toward the posterior vaginal wall, which is then pushed toward the anterior vaginal wall. The anterior vaginal wall provides a functional backstop to the movement of the posterior vaginal wall/anterior rectal wall complex, and there is minimal stress placed on the endopelvic fascial attachments (figure 1). If there is muscular and/or neurologic damage to the puborectalis, the levator hiatus widens, and the vaginal canal opens. The increased rectal pressure and distension associated with defecation places strain on the endopelvic fascial attachments and the fibromuscularis of the posterior vaginal wall, which can result in rectocele and its frequent companion, excessive perineal descent. One hypothesis for these findings is that the endopelvic fascia undergoes discrete tears [4]. This is the rationale for the site-specific defect repairs described below.

PREOPERATIVE PREPARATION — The surgeon should examine the patient prior to the surgical admission to determine whether there is any evidence of an enterocele, sigmoidocele, or associated apical defects. This knowledge helps in planning the most effective surgical approach and preventing a potential repeat operation to address missed defects. If the patient has significant defecatory dysfunction or her symptoms do not correlate with findings on pelvic examination, then pelvic floor magnetic resonance imaging or defecation cystoproctography may reveal one or more of these other defects, or occult rectal prolapse or intussusception. If such defects are detected, the surgeon should counsel the patient about possible management options. Nonsurgical management of these defects is beyond the scope of this review. (See "Posterior vaginal defects (eg, rectocele): Clinical manifestations, diagnosis, and nonsurgical management".)

Several trials have reported that bowel preparation prior to gynecologic surgery does not improve visualization, yet bowel preparation results in increased gastrointestinal symptoms and patient discomfort [5,6]. For surgical repairs using a transvaginal approach, a bowel preparation only serves to contaminate the operative field [7]. Therefore, for these cases, we suggest that the rectal vault be evacuated, as needed, during the preoperative examination under anesthesia. (See "Antimicrobial prophylaxis for prevention of surgical site infection in adults" and "Overview of preoperative evaluation and preparation for gynecologic surgery", section on 'Bowel preparation'.)

In a small randomized study of women who underwent posterior vaginal wall repair with a transvaginal approach, an anal purse string suture was placed after sterile preparation and removed at the end of the procedure. Gross fecal contamination and positive perianal/perineal cultures were significantly decreased in patients with an anal suture compared with those without an anal suture [8].

SURGICAL APPROACHES — The goal of rectocele repair is to relieve symptoms relevant to the anatomic support defect in the posterior vaginal wall.

Gynecologists typically perform a transvaginal repair with the patient in dorsal lithotomy position, while colorectal surgeons frequently operate through an endoanal approach with the patient in prone or jack-knife position.

There are two main methods of transvaginal rectocele repair: the traditional posterior colporrhaphy and the site-specific repair. Either of these methods may incorporate a biologic or synthetic graft. (See "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)

Posterior colporrhaphy — Traditional rectocele repair has an anatomic cure rate of 76 to 96 percent. By plicating the posterior vaginal muscularis or medial aspect of the levator ani muscles in the midline, the posterior vaginal wall width is decreased, the fibromuscularis in the midline is increased, and the vaginal tube is narrowed. A perineorrhaphy is typically included in this repair.

Initial injection of lidocaine with dilute epinephrine or vasopressin below the vaginal mucosa aids with hydrodissection, as well as hemostasis. If the surgeon plans to perform a perineorrhaphy, a triangular-shaped incision is made into the perineal skin with the base of the triangle at the hymen (figure 2); if the introitus is already narrow, the surgeon may elect to make a vertical incision through the perineal skin and vaginal mucosa. The skin is dissected away from the perineal body. The vaginal epithelium is opened in the midline, extending the incision to an area superior to the defect, which may correspond to the vaginal apex (figure 3). The posterior vaginal epithelium is dissected bilaterally away from the underlying fibromuscularis layer until the levator muscles are reached on the lateral margins. It is important to remain in a plane close to the epithelium to avoid injury to the rectum, especially in the area close to the perineal body.

The posterior vaginal wall, stripped of its epithelium, is plicated in the midline with interrupted vertically or transversely placed sutures that incorporate a generous purchase of the fibromuscularis. Plication begins proximally and progresses toward the hymenal ring (figure 4). The surgeon should take care to ensure that each plication suture is in continuity with the previous one, or ridging of the posterior vaginal wall may occur and cause dyspareunia. A rectal examination should be performed in order to ensure that there are no areas of weakness that require further stitches and to check for any evidence of rectal injury or suture in the rectal mucosa.

The vaginal epithelium is trimmed only if necessary and closed with a running 2-0 absorbable suture. This step also narrows the caliber of the vagina, so the surgeon should not trim too much, especially in women with vaginal atrophy. The caliber of the vagina at the conclusion of the vaginal reconstruction should be at least three finger-breadths in sexually active women. To estimate adequate caliber, Allis clamps are placed on the posterior hymen and, when brought together in the midline, should allow for a three finger-breadth genital hiatus.

The rectocele repair can include a levator myorrhaphy, or a plication of the levator ani muscles. Interrupted sutures are placed laterally into the levator ani muscles, incorporating a portion of the lateral posterior fibromuscularis and tying down in the midline. This step provides a muscular posterior shelf, but may further constrict the vaginal caliber and be a source of postoperative pain and/or dyspareunia. However, it is an effective option for older adult women with a wide levator hiatus who do not expect to be sexually active.

Site-specific defect repair — The anatomic cure rate of the site-specific posterior repair is 82 to 100 percent. The rationale for the site-specific repair is based on the theory that herniation of the rectum into the vagina results from specific tears in the rectovaginal fascia [9]. These tears in the posterior vaginal wall may occur as an isolated defect in the lateral, distal, midline, or superior portions of the wall, or as a combination of defects (figure 5A-E).

The vaginal epithelium is opened at the perineal body in the manner previously described for the traditional posterior colporrhaphy. The posterior vaginal epithelium is incised in the midline to a level proximal to the rectocele bulge and dissected away from the underlying fibromuscularis. The dissection is extended laterally to the endopelvic fascial attachment of the posterior vaginal wall to the arcus tendineus fasciae pelvis and arcus tendineus fasciae rectovaginalis. The fibromuscularis is carefully inspected to identify breaks.

Irrigation and a rectal examination with the nondominant hand may accentuate the defects to aid identification. Defects are individually isolated and repaired with a delayed-absorbable 0 or 2-0 suture. If a distal defect is present (likely the result of a separation of the fibromuscularis from the perineal body), it should be repaired with absorbable suture to reapproximate the rectovaginal septum to the perineal body and to avoid a perineal pocket.

Repair of perineal body defects are also addressed with interrupted sutures. Repeating the rectal examination should confirm repair of the rectocele. The vaginal epithelium is then closed with a running 2-0 absorbable suture.

Graft augmentation — Use of reconstructive materials (synthetic or biologic) to augment repair of posterior vaginal wall prolapse is not supported by the body of evidence, although some smaller trials have reported conflicting outcomes [10-13]. In the largest trial to date comparing synthetic mesh, graft, and standard repair (ie, placebo) for women undergoing posterior repair, use of mesh or graft were not associated with improved outcomes for efficacy, quality of life, or overall adverse effects [14]. However, more than one in ten women who underwent a mesh repair had mesh-related complications. A different large cohort study reported that mesh-based posterior repair was associated with an increased risk of repeat prolapse surgery and later complications [15].

In observational studies of posterior vaginal prolapse repair, there have been poor results reported for a particular xenograft, a porcine dermal acellular collagen matrix (Pelvicol) [16-20]. Case series have reported high rectocele recurrence rates and poor wound healing. Well-designed studies are needed to further evaluate this graft.

Pre-cut mesh kit procedures have been developed by commercial companies for repair of posterior vaginal wall prolapse and other prolapse sites. However, these procedures have gained popularity without sufficient safety or outcome data supporting their uses, and transvaginal mesh kits for the treatment of prolapse are now off the market in many countries due to a ban on such devices. (See "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)

Endorectal repair — The colorectal surgeon typically operates on the distal rectocele from the endoanal or endorectal approach. The anatomic cure success rate for the transanal approach is 70 to 96 percent [21]. An endorectal approach allows for simultaneous correction of other common anorectal pathology, such as hemorrhoids and mucosal rectal prolapse. As an example, one study reported that 53 percent of women undergoing a rectocele repair also had rectal mucosa prolapse and 41 percent had hemorrhoids [22]. Nevertheless, repair of proximal (high) rectoceles is difficult with the endorectal approach because of inadequate exposure.

This procedure is done in the prone jack-knife position with the buttocks spread and taped. An anal retractor is inserted to expose the anterior rectal wall. A U-shaped or T incision is made in the rectal mucosa at or proximal to the dentate line. A flap is developed to a level proximal to the rectocele (usually approximately 7 cm in length). The rectal mucosal flap is trimmed and the rectal mucosa is closed with a running 5-0 polyglycolic acid suture. If the mucosa is not removed, plication of the rectal mucosa can lead to necrosis and postoperative infection, and some patients also complain of persistent tenesmus. The rectovaginal septum is reapproximated using vertical plication sutures with a 3-0 polyglycolic acid suture. Levator ani plication also can be done with the endorectal approach.

There are several disadvantages to the transrectal or transanal repair. This approach does not allow the surgeon to perform a perineorrhaphy, anal sphincteroplasty, or a high rectocele repair unless a second incision is made. De novo fecal incontinence is also a concern after transanal repair, with rates that can range as high as 38 percent postoperatively [23]. Another rare, but serious, complication is the development of a rectovaginal fistula.

Perineorrhaphy — A perineorrhaphy, when indicated, completes the vaginal approach to a rectocele repair. Allis clamps are placed on the posterior hymen and brought together in the midline. Preservation of at least three finger-breaths at the genital hiatus is important for comfortable future coital activity. A triangular incision is made medial to the Allis clamps, extending to the midline of the perineal skin with the base of the triangle at the posterior hymen. The bulbocavernosus muscles are plicated in the midline of the perineal body with an interrupted 0 polyglactin suture. The surgeon should avoid creating a ridge from overly plicating the bulbocavernosus muscles or closing the perineal epithelium too high, which can result in a skin bridge. The transverse perinei muscles are plicated. An anal sphincteroplasty may be performed as indicated for anal incontinence and an external/internal anal sphincter defect. The skin is closed with a running suture of 2-0 polyglactin. A perineorrhaphy can slightly increase the functional length of the posterior vaginal wall. However, aggressive perineorrhaphy may constrict the vaginal introitus and cause superficial dyspareunia or apareunia.

There are several retrospective studies suggesting that a postoperative genital hiatus less than 4 cm results in lower recurrence rates [24,25]. Increasing genital hiatus also appears to be predictive of loss of apical support [26]: a genital hiatus >3.5 cm should prompt an evaluation for concomitant apical prolapse and consideration of a concomitant apical support procedure at the time of prolapse repair.

Anal sphincteroplasty — Extensive dissection and repair are required for women with an absent perineal body. Preoperative work-up with endoanal ultrasound may reveal an anal sphincter defect. A difficult vaginal delivery or surgical trauma are the usual causes.

A transverse semi-circular incision is made in the layer separating the posterior vaginal wall and the anterior rectal wall. Dissection is extended laterally and proximally. To facilitate proximal dissection in the rectovaginal space without injury to the rectal mucosa, the surgeon may insert a finger of the non-dominant hand into the rectum. The internal anal sphincter is plicated in the midline with a 3-0 absorbable suture. The rectal mucosa is reapproximated, as needed, with a running absorbable 3-0 suture and extended to the skin overlying the external anal sphincter. Dissection of the anus is performed to identify the retracted ends of the external anal sphincter. Care should be taken with this step to avoid extensive lateral and posterior dissection and injury to the inferior hemorrhoidal nerves and vessels. The ends of the external anal sphincter are identified and reapproximated in an overlapping fashion with vertical mattress sutures of 0 delayed absorbable suture. The scar on the ends of the external anal sphincter are left intact and used for suture placement.

The transverse perinei and bulbospongiosus muscles are plicated in the midline. With midline construction of the perineal body, the transverse portion of the initial incision becomes vertically oriented. The vaginal epithelium is closed with 2-0 absorbable suture. The skin of the perineal body is closed in an inverted Y shape with interrupted absorbable 2-0 sutures. The patient is instructed that superficial wound breakdown on the perineum may occur. In most cases, this superficial wound breakdown will respond to conservative management, rather than require aggressive debridement.

Abdominal sacral colpopexy (colpoperineopexy) — Posterior wall support defects are often secondary to loss of apical support and by correcting the apical support, the posterior wall is restored without a separate rectocele repair. The abdominal sacral colpopexy is a procedure that attaches mesh to the anterior and posterior surfaces of the vagina to suspend the apex of the vagina to the sacral promontory. An abdominal or laparoscopic approach may be performed when the rectocele is accompanied by apical prolapse. These procedures are discussed in detail separately. (See "Pelvic organ prolapse in women: Surgical repair of apical prolapse (uterine or vaginal vault prolapse)".)

Iliococcygeus fascia suspension — This procedure is primarily used for apical suspension, but may be used to correct high lateral rectocele defects. The surgical procedure is described in detail separately. (See "Pelvic organ prolapse in women: Surgical repair of apical prolapse (uterine or vaginal vault prolapse)".)

POSTOPERATIVE INSTRUCTIONS — There are no data from randomized trials on which to base recommendations for postoperative care. Thus, postoperative management is based on the surgeon's clinical experience and varies significantly among surgeons. We place a vaginal pack at the end of the procedure. This pack and the Foley catheter are removed the next morning. We suggest that patients avoid lifting more than 10 pounds and take stool softeners to avoid straining for three months. We also ask them to maintain pelvic rest (avoid the use of tampons, douching, or sexual intercourse) for at least six weeks.

COMPLICATIONS — Pain, temporary urinary retention, and constipation are common in the perioperative period. Serious complications are uncommon complications, and include development of a hematoma, infection, inclusion cyst formation, fecal impaction, and injury to the rectum with the development of a rectovaginal or a rectoperineal fistula. Bowel and defecatory dysfunction may continue long-term, and prolapse may recur. De novo dyspareunia or fecal incontinence may also occur. Utilization of mesh is associated with additional complications, such as mesh erosion, infection of the graft, and persistence of granulation tissue. (See "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)

CHOICE OF PROCEDURE — As discussed above, the repair of posterior vaginal prolapse may involve a vaginal, transanal, or laparoscopic approach. The choice of approach depends upon the surgeon and patient preference, surgeon's skill level (especially in the case of laparoscopic repair), type of defect, symptomatology, and whether there is a need for concomitant surgical procedures.

Vaginal versus transanal approach — Posterior colporrhaphy, a vaginal approach to posterior prolapse repair, is more effective than a transanal approach. This was illustrated in a meta-analysis of three randomized trials that found a significantly lower rate of objective failure for posterior colporrhaphy (10 versus 42 percent, risk ratio 0.24, 95% CI 0.09-0.64) [10,27-29]. The trials also demonstrated that the transvaginal repair led to better anatomic results (by mean Ap points on the follow-up POP-Q examination). The risk of de novo dyspareunia was similar in both groups.

Posterior colporrhaphy versus site-specific versus graft augmentation — A prospective trial randomly assigned 106 patients to one of three surgical techniques of rectocele repair: traditional colporrhaphy, site-specific repair, or site-specific rectocele repair augmented with a porcine-derived, acellular collagen matrix graft (Fortagen) [12]. This trial used condition-specific validated quality-of-life questionnaires at baseline and follow-up: the Pelvic Floor Distress Inventory short form (PFDI-20); the Pelvic Floor Impact Questionnaire short form 7 (PFIQ-7); and a sexual function questionnaire, the Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire short form (PISQ-12). Major findings were:

●At one year after surgery, anatomic cure of prolapse (defined as ≤Stage II) was comparable in the posterior colporrhaphy group (86 percent) and the site-specific group (78 percent). The cure rates of both traditional and site-specific rectocele repair groups were statistically significantly better than the graft-augmented site-specific repair (54 percent).

●Posterior wall prolapse to or beyond the hymen developed in 20 percent in those who received graft augmentation, compared with 7.1 percent in the posterior colporrhaphy group and 7.4 percent in the site-specific repair group. Time to development of posterior vaginal wall prolapse was significantly earlier in the graft augmentation group compared with the traditional rectocele repair group. The site-specific group also developed recurrent prolapse earlier than the traditional posterior colporrhaphy group, but this was not statistically significant.

●Functional failure (worsening of prolapse or colorectal symptoms one year after surgery) occurred in 16 percent of the posterior colporrhaphy group, 12 percent of the site-specific repair group, and 21 percent of the graft augmentation group. These differences were not statistically significant.

●Defecatory dysfunction decreased significantly postoperatively, with no significant differences among treatment groups. At one-year follow-up, the development of new onset bowel dysfunction was uncommon (11 percent), with the cure of posterior vaginal prolapse at Stage 0 or 1 associated with the significant reduction in bothersome straining and incomplete emptying [30].

●Women who underwent perineorrhaphy (where the rectovaginal septum was reattached to the perineal body) were significantly less likely to report bothersome splinting one year after surgery after controlling for treatment group and baseline symptoms (odds ratio 0.08, 95% CI 0.004-0.71).

●The preoperative dyspareunia rate was 51 percent, and the postoperative dyspareunia rate was 36 percent, with no significant differences among the groups (20 percent in the posterior repair group, 14 percent in the site-specific group, and 6 percent in the graft augmentation group). The high preoperative dyspareunia rate may be due to the use of a validated sexual function questionnaire. There was improvement in sexual function after rectocele repair, regardless of the technique used.

In summary, anatomic and functional outcomes were not significantly different for the standard posterior colporrhaphy and site-specific repair groups, but were overall superior to the graft augmentation group. Bowel and sexual function and quality of life significantly improved in all three groups.

OUTCOMES

Summary — Anatomic outcomes for traditional versus site-specific rectocele repair do not appear to be significantly different. Levator ani plication likely contributes to de novo dyspareunia and should not be performed in sexually active women. The use of adjunctive graft material still requires more study, but the evidence thus far does not support the use of porcine dermis (Pelvicol) or small intestine (Fortagen). Mesh erosion is a concern with the use of polypropylene, and use of transvaginal prolapse repair with synthetic mesh is avoided when possible. Nevertheless, given the variety of grafts available, there are insufficient data to either recommend or advise against the use of other graft materials at this time.

●(See "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)

●(See "Reconstructive materials used in surgery: Classification and host response", section on 'Types of materials'.)

Site-specific repair — Several case series have reported de novo dyspareunia rates of 2 to 11 percent [31-34]. Bowel symptoms (constipation, splinting, fecal incontinence) also improved, with anatomic cure rates of 82 to 100 percent.

A review comparing 124 patients who underwent site-specific rectocele repair with 183 patients who had a standard posterior colporrhaphy reported that at one year, subjective and objective recurrence rates were significantly higher after the site-specific rectocele repair [35]. However, postoperative dyspareunia, constipation, and fecal incontinence were not significantly different for the two groups. Thirty-three percent in each group reported improvement in dyspareunia, and 11 percent in each group had de novo dyspareunia.

Graft augmentation — As with the other surgical approaches, bowel symptoms improve after surgical repair with biologic graft material. One trial that added porcine subintestinal submucosal (SIS) graft augmentation to rectocele repair (either traditional or site-specific) reported similar subjective or anatomic outcomes between groups, with significant symptomatic improvement and low anatomic failure rates in each group (graft 12 percent versus no graft 8.6 percent) [36]. The failure rates in this trial were lower than those reported in a trial of porcine-derived, acellular collagen matrix graft (commercial name, Fortagen) [12]. One possible explanation is that SIS graft is not crosslinked but the porcine-derived, acellular collagen matrix graft is, and thus may incite a greater inflammatory response. A systematic review by the Society of Gynecologic Surgeons Systematic Review Group on the use of grafts in transvaginal pelvic organ prolapse repair concluded that there is insufficient evidence to guide decisions whether to use graft materials [37]; associated clinical guidelines issued by the same group suggest that native repair remains preferable to graft use [38]. (See "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pelvic organ prolapse".)

SUMMARY AND RECOMMENDATIONS

●Surgical outcomes – Rectocele repair outcomes are generally good. Success rates for traditional colporrhaphy and site-specific repairs range from 76 to 96 percent and 56 to 100 percent, respectively. (See 'Outcomes' above.)

●Surgical technique – In patients undergoing repair of posterior vaginal wall prolapse, we recommend standard posterior colporrhaphy or site-specific repair over crosslinked porcine small intestine graft augmented repair (Grade 1B). Suture-only repair results in a greater reduction of recurrent posterior prolapse than porcine graft. Bowel and sexual function and quality of life significantly improve with any of the three procedures. (See 'Choice of procedure' above and "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)

•Selective use of levator ani plication – We suggest not performing levator plication in patients who are sexually active, given this procedure is associated with a high risk of de novo dyspareunia (Grade 2C). (See 'Summary' above.)

•Concomitant apical suspension – Apical suspension procedures, such as sacral colpopexy or iliococcygeus fascia suspension, may be required in addition to traditional colporrhaphy and site-specific repair to correct the rectocele. (See 'Surgical approaches' above.)

●Concern for porcine xenograft use – In patients undergoing repair of posterior vaginal wall prolapse with xenograft augmentation, we suggest against using Pelvicol porcine dermal graft. (Grade 2C). Use of this material is associated with a high rate of recurrent rectocele and poor wound healing. (See 'Graft augmentation' above and "Transvaginal synthetic mesh: Use in pelvic organ prolapse".)