Functional Disorders: Rectocele

Abstract

Rectoceles are a very common finding in patients, and symptoms most commonly include pelvic pain, pressure, or difficulty with passing stool. However, there are often other associated pelvic floor disorders that accompany rectoceles, making the clinical significance of it in an individual patient often hard to determine. When evaluating a patient with a rectocele, a thorough history and physical exam must be conducted to help delineate other causes of these symptoms. Treatment consists of addressing other defecatory disorders through various methods, with surgery reserved for select cases in which obstructed defecation is well documented.

A rectocele is an outpouching of the anterior wall of the rectum through a weakened rectovaginal septum, creating a bulge into the posterior vaginal wall. In and of itself, a rectocele is a purely anatomic disorder, the functional significance of which is difficult to determine and seems to have little correlation with the size or location of the defect.¹ Symptoms attributed to rectocele include pelvic pain or pressure, ulceration of prolapsing vaginal mucosa, or difficulty evacuating stool, sometimes requiring manual support of the posterior vaginal wall to evacuate. However, because of the high rate of coexisting disturbances of pelvic floor anatomy and function in patients with a rectocele, caution must be used before attributing symptoms to just an obvious bulge. A careful and detailed history, accompanied by a thorough physical examination, dynamic imaging studies, and physiologic testing, will often reveal a constellation of symptoms and derangements that are unlikely to resolve with simple reinforcement of the rectovaginal septum, regardless of the surgical approach.

Incidence

The true incidence of rectoceles is unknown, but asymptomatic posterior compartment prolapse has been reported in approximately 40% of parous women.² Performing defecography in young, healthy volunteers, Shorvon et al identified small rectoceles in 17 of 21 women (81%), with 10 of these (48% of total) measuring >1 cm.³ Rectovaginal septal defects and posterior compartment prolapse have been reported in more than 10% of nulliparous women aged 18 to 24 years.⁴ Olsen et al, in a retrospective analysis of 149,554 women in the Kaiser Permanente Northwest system, estimated that the lifetime risk of having surgery for pelvic organ prolapse by age 80 years was 11.1%, with 46% of those who undergo repair having defects in the posterior compartment.⁵ Risk factors for development of a rectocele include age, multiparity, vaginal delivery, surgery (including episiotomy, hysterectomy, or hemorrhoidectomy), and conditions which chronically increase intra-abdominal pressure such as chronic obstructive pulmonary disease (COPD), obesity, and constipation.

Symptomatology

Small rectoceles may be completely asymptomatic and may be a normal finding. Patients with larger rectoceles may complain of a variety of symptoms including pelvic pain or pressure, sexual dysfunction, and defecatory dysfunction including obstructed defecation, the need to self-digitate or “splint” to evacuate the rectum, incomplete emptying, or fecal incontinence. A patient's primary symptoms will likely dictate the type of surgeon to whom she is referred. Women complaining of pelvic pressure, dyspareunia, general pelvic organ prolapse, or urinary incontinence will likely be treated by gynecologists or urogynecologists, while those presenting with fecal incontinence, chronic constipation, rectal prolapse, or rectal bleeding are often referred to a colorectal surgeon. This phenomenon accounts for some of the difficulty in comparing retrospective series across disciplines. While both groups of surgeons are generally aimed at restoring anatomy and avoiding recurrence, the symptoms from which their patients are seeking relief are likely very different.

Traditionally, both gynecologists and proctologists have repaired rectoceles by sutured plication of the tissue between the vagina and rectum, approached through an incision in the organ each knows best. More recently, abdominal operations and transanal resectional procedures have been applied to the problem, recognizing that reinforcement of the weakened septum alone may not address the entire problem, or may create new ones. There remains no consensus in either specialty as to the indications for rectocele repair or the definition of a successful outcome. The two important endpoints of restoring normal anatomy and alleviating symptoms are poorly correlated in most studies.⁶ Increasingly, pelvic floor surgeons and researchers recognize the complex interplay of anatomy, physiology, and psychology in the dysfunctional pelvic floor. Most experts advocate a thorough evaluation process followed by a tailored, multimodal treatment plan aimed at identifying and correcting functional outcomes, with surgery offered selectively for specific patterns of dysfunction.^{7 8 9}

The invited topic of this review is that of rectoceles as a functional disorder. With that in mind, and considering the target audience, we will focus on rectoceles as they relate to defecatory dysfunction, primarily obstructed defecation. As previously stated, this is a challenging subject because it is still unclear whether rectoceles are the cause or the result of obstructed defecation. There does appear to be a subset of patients who experience difficulty evacuating purely as a result of propulsive forces being transmitted anteriorly into a rectocele rather than down through the anal canal. This leads to stool trapping in the rectocele and the feeling that “pushing doesn't push it out.” These patients often report the need to apply digital pressure to the perineum or posterior vaginal wall to defecate. On the other hand, studies have shown that the majority of patients with a rectocele and constipation have some combination of slow-transit constipation, internal intussusception, paradoxical contraction of the puborectalis, and abnormal perineal descent in addition to a rectocele.⁹ In the following paragraphs, we will review the normal anatomy of the pelvic floor and the alterations that contribute to rectocele formation. We will discuss the evaluation of a patient with a rectocele and constipation, focusing on findings which point toward or away from the rectocele as the primary cause. We will discuss the role of nonsurgical management, including biofeedback, and will review the general types of surgical repair, potential advantages and disadvantages of each, and scenarios where one may be preferable to another.

Anatomy of the Pelvic Floor and the “Rectovaginal Septum”

The supportive function of the pelvic floor is derived from the interactions of the pelvic musculature and connective tissue, with the levator ani complex and the endopelvic fascia as the primary components. The levator ani complex consists of the U-shaped sling of the puborectalis and pubococcygeus muscles, and the broad, shelf-like iliococcygeus muscle. Arising from the pelvic sidewall along the tendinous arch of the levator ani, the iliococcygeus forms a wide, flat surface which supports the rectum posteriorly. The puborectalis and pubococcygeus arise from the pubic bone on either side of the midline, forming a muscular sling around the rectum and the urogenital hiatus. At rest, the levator ani muscles are in a state of constant contraction, pulling the rectum ventrally, closing the hiatus, and providing a muscular shelf on which the pelvic organs are supported.^{10 11}

The endopelvic fascia is a fibromuscular layer of connective tissue which invests the vaginal walls and apex, anchoring them to the boney pelvis. The term fascia is a misnomer as the tissue consists of a loosely arranged layer of collagen, elastin, fibroblasts, and smooth muscle cells, rather than the distinct parietal fascia which envelopes skeletal muscles.¹² The endopelvic fascia extends in a continuous sheet from the perineal body below to the vaginal apex where it coalesces into more dense aggregations of collagen referred to as the uterosacral and cardinal ligaments.¹¹ Delancey describes three levels of support provided by the endopelvic fascia along the length of the vagina (Fig. 1).¹³ The apex of the vagina (Level I) is suspended by a sheet-like mesentery of vertical fibers which are a continuation of the cardinal ligament. In the midvagina (Level II), the fibers attach to the levator ani muscles on the pelvic sidewall along a line known as the arcus tendineus fascia rectovaginalis.¹⁴ The fibers are more robust at the posterior lateral vaginal wall, where they pull in a dorsal and cranial direction, creating a vaginal sulcus on either side of the rectum and resisting the ventral movement of the posterior vaginal wall.¹⁵ In the distal vagina (Level III), the connective tissue of the endopelvic fascia becomes confluent with the fibers of the perineal body. This fusion creates a dense layer which extends cephalad from the perineal body for a distance of 2 to 3 cm above the hymenal ring and resists anterior displacement of the rectum when downward force is applied.¹⁵ The perineal body is further supported by its ventral and lateral attachments to the pubic rami through the perineal membrane.

Fig. 1.

Illustration of the endopelvic fascia as it fans out to cover the pelvic floor and provide support of the surrounding organs. The levels 1, 2, and 3 depict the vaginal support. (From The ASCRS Textbook of Colon and Rectal Surgery, 2nd edition, Ch. 19, page 325.)

The structure of the posterior vaginal wall and the existence of a true rectovaginal septum have been debated without clear consensus, though the argument may be more a matter of terminology than substance. A recent examination of the discrepancies between authors is provided by Kleeman et al.¹⁶ The authors conducted histologic analysis of cadaveric specimens and confirmed the earlier work of Goff and Ricci and Lisa,^{17 18} finding no evidence of a fascial layer between the posterior vaginal wall and the anterior rectal wall. Instead the authors demonstrate vaginal mucosa (epithelium and lamina propria) overlying the fibromuscular wall of the vagina, followed by an “adventitia” of the vagina which immediately abuts the longitudinal muscle of the anterior rectum. This adventitia, they argue, does not constitute a fascia or septum; it contains little collagen and is composed mainly of adipose tissue admixed with bundles of fibrous tissue, blood vessels, elastic fibers, and nerves. Both the studies of Goff and Ricci and Lisa described similar findings, referring to this adventitia as “areolar fascia” and “areolar mesh,” respectively. Both recognized that the layer became thicker at the lateral edges and attached to the pelvic sidewall, corresponding to Delancey's description of the Level II endopelvic fascia.¹⁵ On the contrary, studies by Uhlenhuth and Wolfe, Miley and Nichols, and more recently Richardson describe the identification of a dense layer behind the vagina composed of smooth muscle and elastic fibers, amenable to dissection and discreet repair.^{19 20 21 22} This discrepancy between histologic and surgical layers is difficult to reconcile but has led many authors to conclude that the “rectovaginal septum” that is plicated during transvaginal rectocele repair is merely a surgical artifact, created by dissecting the lamina propria of the vaginal mucosa off of the fibromuscular wall of the vagina.^{15 16 23 24} In the interest of clarity, for the remainder of this review, we will refer to all of the fibromuscular and areolar tissues between the vaginal mucosa and the rectal wall as the rectovaginal septum.

Development of a Rectocele

Richardson described discreet defects in the rectovaginal septum as the etiology of rectocele formation.²¹ He listed the most common site as a transverse break just above the perineal body. Other defects were low vertical breaks in the midline or lateral separations from the iliococcygeus fascia. With the exception of the vertical midline, these “breaks” in the septum occur essentially at the fixation points of the posterior vaginal wall to the boney pelvis, laterally to the levator complex and distally to the perineum. This loss of fixation impairs the ability of the posterior wall to resist pressures from behind. Denervation of the pelvic floor and widening of the genital hiatus worsen the condition by disrupting the normal balancing force of the opposing anterior vaginal wall.²⁵ For these reasons, advocates of transvaginal colporrhaphy recommend addressing not only the integrity of the rectovaginal septum but also its fixation to the perineum and the levator fascia.²² Not all posterior vaginal wall bulges are rectoceles. The presence of a deep cul-de-sac or vault prolapse may permit the peritoneum and its contents to come into direct contact with the posterior vaginal wall as an enterocele or sigmoidocele. The distinction between these entities and rectocele can typically be made by rectovaginal exam, as bowel is felt filling the space between the vaginal and rectal fingers during straining.

Vaginal delivery, obesity, and advanced age are the biggest risk factors for rectocele development.²⁶ The forces of labor can stress or tear the muscles, connective tissue, and nerves of the pelvic floor and rectovaginal septum. The perineal body, external anal sphincter, or levator ani complex can be damaged by traumatic obstetric lacerations or episiotomy.^{27 28} Prolonged pushing during labor can stretch or compress the peripheral nerves that innervate the levator ani complex, causing a transient denervation of the pelvic floor. While function is recovered by most women within 2 months, lasting electrophysiologic evidence of denervation injury has been documented, and the effects of parity may be cumulative.^{28 29} As muscular support is weakened, the connective tissue of the endopelvic fascia is subjected to greater stretch and may weaken under chronic tension.³⁰ Age-related degeneration of this connective tissue coupled with chronic increases in intra-abdominal pressure from obesity or frequent straining compound the damage. Many of these risk factors can be interconnected and are additive in their effects.

Luo et al recently characterized the structural deformations of the posterior vaginal wall in women with rectocele by three-dimensional MRI.³¹ With Valsalva, the authors demonstrated a consistent downward displacement of the upper two-thirds of the vagina as well as a complex folding of the vaginal wall likened to a person kneeling. Other findings, such as perineal descent, vaginal widening, or forward protrusion, were less consistent. As prior imaging of rectoceles has focused almost entirely on the anterior wall of the rectum, better characterization of the movements of the vagina itself may help explain some of the divergent symptoms among patients with rectocele.

The Role of Rectocele in Obstructed Defecation

The obstructed defecation syndrome (ODS) is discussed in detail in a separate section in this issue but is reviewed briefly here. Chronic constipation is an extremely common problem affecting an estimated 63 million people in North America or roughly 15% of the population, depending on defining criteria.^{32 33} The true incidence is difficult to determine, however, as defining constipation by frequency of bowel movements alone, typically less than three times per week, may exclude several patients who report difficulty defecating, which they would describe as constipation. The Rome Foundation, an international working group on functional gastrointestinal disorders, has published expanded diagnostic criteria for functional constipation (Table 1).^{34 35} The most recent criteria emphasize difficulty evacuating stool as being equally important to the absolute number of bowel movements achieved. The American Gastroenterological Association (AGA) further subcategorizes functional constipation into three subgroups: normal transit constipation, slow-transit constipation, and defecatory disorders.³⁶ The last of these includes conditions in which the evacuation of stool from the rectum is impaired by either lack of propulsive force or excessive resistance to evacuation. Slow-transit constipation may be present in up to 50% of patients with defecatory disorders.³⁶ Even in isolated defecatory disorder constipation, multiple derangements of structure and function may exist simultaneously, leading many experts to refer to this type of constipation as ODS. Contributing causes may include nonrelaxation or paradoxical contraction of the puborectalis or external anal sphincter (also called anismus), abnormal perineal descent, rectocele, enterocele, or rectal intussusception. Altomare et al published the first scoring system for ODS severity in 2006 (Table 2).³⁷

Table 1. Rome III diagnostic criteria for functional constipation.

1. Must include two or more of the following

a. Straining during at least 25% of defecations  b. Lumpy or hard stools in at least 25% of defecations  c. Sensation of incomplete evacuation for at least 25% of defecations  d. Sensation of anorectal obstruction/blockage for at least 25% of defecations  e. Manual maneuvers to facilitate at least 25% of defecations (e.g., digital evacuation, support of the pelvic floor)  f. Fewer than three defecations per week

2. Loose stools are rarely present without the use of laxatives

3. There are insufficient criteria for irritable bowel syndrome

Criteria fulfilled for the last 3 mo with symptom onset at least 6 mo prior to diagnosis

Source: Longstreth et al.³⁵

Table 2. The obstructed defecation syndrome score questionnaire.

Variables	Score
	0	1	2	3	4
Mean time spent at the toilet	≤5 min	6–10 min	11–20 min	21–30 min	>30 min
N attempts to defecate per day	One	Two	Three to four	Five to six	>Six
Anal/vaginal digitation	Never	>1/mo, <1/wk	Once a week	Two to three per week	Every defecation
Use of laxatives	Never	>1/mo, <1/wk	Once a week	Two to three per week	Every day
Use of enemas	Never	>1/mo, <1/wk	Once a week	Two to three per week	Every day
Incomplete/fragmented defecation	Never	>1/mo, <1/wk	Once a week	Two to three per week	Every day
Straining at defecation	Never	<25% of the time	<50% of the time	<75% of the time	Every defecation
Stool consistency	Soft	Hard	Hard and few	Fecaloma formation

Source: Altomare et al.³⁷

While both the failure of surrounding tissues to support propulsion and the failure of the pelvic outlet to open can mechanistically be understood to make evacuation difficult, the former may be a consequence of the later rather than a primary cause of dysfunction. Patients who chronically strain due to an inability to relax their puborectalis or anal sphincter, as well as those with slow-transit constipation, subject their pelvic floor and posterior compartment to damaging stretch and strain. With time, these and other causes of ODS can lead to rectocele formation through the mechanisms outlined earlier. Herein lies the difficulty in defining rectocele as a functional disorder, as well as the danger in offering rectocele repair to treat ODS symptoms. Rotholtz et al found coexisting causes of ODS, including abnormal perineal descent, intussusception, or anismus, in 90% of patients with “significant” rectoceles as defined by size larger than 4 cm, prolonged symptoms despite increased fiber, or the need for self-digitation.³⁸ Johansson et al identified paradoxical puborectalis and external sphincter contraction by EMG in 70% of patients with rectocele and ODS.³⁹ Not only is the incidence of coexisting defecatory disorders high in patients with a rectocele, it seems identical to those in patients with ODS who do not have a rectocele. Hicks et al found similar rates of paradoxical puborectalis contraction (64 and 57%) and internal intussusception (32 and 25%) in ODS patients with and without a rectocele, respectively. With no significant manometric or other radiographic differences identified between patients with and without rectocele, the authors conclude that the presence of a rectocele is likely an incidental side effect of ODS rather than a significant cause.⁴⁰

Furthering the confusion about the relationship of a rectocele with ODS is the lack of correlation between objective measurements of rectoceles and severity of symptoms or success of repair. Multiple studies from the gynecologic literature have failed to show correlation between the degree of pelvic organ prolapse and defecatory dysfunction.^{41 42 43 44} Traditionally, size larger than 3 cm and barium trapping on defecography have been considered indications for rectocele repair, but multiple studies have failed to show correlation between these factors and either preoperative symptoms or outcomes from surgery. Halligan and Bartram found no difference in the time to evacuation, the need to splint, or the maximal rectal pressure in rectocele patients with or without contrast retention on defecography.⁴⁵ Ting et al found no correlation between the volume of contrast retained on defecography and symptoms of incomplete evacuation.⁴⁶ Carter and Gabel, reporting retrospectively on more than 400 patients, found no correlation between rectocele size and either severity of symptoms or rectal hyposensitivity. Larger size was associated with more contrast retention, but consistent with other studies did not predict symptoms of ODS.⁴⁷

Rectocele may be the most obvious physical exam finding in a patient with ODS, but it is unlikely to be the only abnormality present and even less likely to be the only cause of the patient's constipation. Pescatori et al liken the presence of a rectocele in ODS to an iceberg, in which the rectocele is only the tip visible above the water, while larger “underwater rocks” may be present and affect the outcome of surgery.⁴⁸ Failure to recognize the associated complex pelvic floor dysfunction in patients with a rectocele will lead to unrealistic expectations from the surgeon and the patient regarding the role of surgery in alleviating symptoms.

Evaluation

The presentation and chief complaint of a patient with a rectocele can be highly variable, but will typically fall into one of two categories: vaginal symptoms (dyspareunia, looseness with intercourse, perineal pressure, mucosal ulceration from prolapse) or defecatory dysfunction (ODS or fecal incontinence). A rectocele may be identified during the workup of any of these presenting symptoms or it may be the primary reason for referral in a patient noted to have a rectocele by another clinician. In the later scenario, it is probably best to ignore the presence of the rectocele altogether and start from scratch with a thorough history aimed at elucidating the primary symptoms from which the patient seeks relief. As previously stated, some rectoceles are asymptomatic and may be an incidental finding unrelated to the patient's complaints. Even symptoms which are easily attributable to a rectocele should be scrutinized because multiple diagnoses frequently exist in these patients. The surgeon should thoroughly work up the presenting symptoms and try to elicit unstated ones that may suggest coexisting disorders, resisting the urge to settle on a diagnosis after spotting just the tip of the iceberg.

Again, in keeping with the overall theme of this issue, the following section will focus on patients presenting with defecatory disorders and a rectocele, as opposed to those seeking treatment for isolated vaginal symptoms.

History

The initial interview should center on the chief complaint and the reason for seeking medical attention. The importance of understanding which symptoms are most distressing to the patient and the goals for treatment cannot be overstated. A 70-year old widow with mild constipation who has worsening perineal pressure when standing for long periods will have a different definition of a successful outcome than a 30-year-old woman who is 6 months postpartum and has new onset fecal incontinence. Both may have a rectocele, but they likely have different etiologies, different risk factors, different lifestyles, and different expectations, all of which must factor into their treatment plans.

In addition to the primary complaint, it is important to inquire about the patient's overall pelvic health. Women may be reluctant to volunteer information about their bowel habits or sexual function unless specifically asked. Patients referred to a colorectal surgeon for constipation or fecal incontinence will likely undergo a thorough inquiry into these symptoms, but may not be asked about coexisting urinary incontinence or vaginal looseness with intercourse. Likewise patients with obvious pelvic organ prolapse and predominantly vaginal complaints may view their chronic constipation and straining as a separate, unrelated issue. Understanding the entire picture may influence the recommended course of therapy or affect the type of operation offered. The use of standard surveys for sexual function, incontinence, and constipation can help standardize patient evaluations, spare patients having to answer uncomfortable questions face to face, and save time in the clinic when filled out beforehand.

Several useful instruments exist to grade severity of constipation and the impact on a patient's quality of life. The Cleveland Clinic Constipation Scoring System⁴⁹ (CCCSS) is perhaps the most widely used of these, but other validated systems include the Constipation Severity Instrument⁵⁰ and the Constipation-Related Quality of Life measure.⁵¹ Documenting the frequency of bowel movements or urge to defecate as well as the need to strain or manually splint to evacuate may help distinguish between slow-transit constipation and ODS. The duration of symptoms, current bowel regimen and defecation routine, previous interventions, and stool consistency are all important to determine. A complete list of prescription and over-the-counter medications should be recorded. Other medical conditions which predispose to constipation (hypothyroidism, diabetes) or pelvic floor dysfunction (COPD, connective tissue disorders) should be noted. A careful obstetric history including episiotomies, lacerations, or assisted deliveries may suggest injury to the perineal body or anal sphincter. It is important to ask specifically about symptoms suggesting rectal prolapse (blood or mucous staining, incontinence, excess tissue with straining) which, if present, may direct the surgeon to examine the patient on the commode after an enema, before returning to the exam room for complete pelvic exam and proctoscopy.

Physical Exam

A rectocele should be easily diagnosed by physical exam, but its presence does not obviate the need for a thorough examination of the remainder of the pelvic floor. The first step in any evaluation of the anus, rectum, or pelvic floor is careful inspection. Note any skin lesions or evidence of prior anorectal surgery. Skin irritation or excoriation may suggest chronic fecal soiling or excessive cleaning; hemorrhoids may be present in patients who chronically strain. Note the size of the genital hiatus and its contents, as well as the length and width of the perineal body. The patient should push to simulate evacuation, while the surgeon notes the size and location of the rectocele, the degree of perineal descent, and the presence or absence of mucosal prolapse.

Digital rectal exam should be performed to evaluate sphincter tone at rest and with squeeze, and to feel for obvious sphincter defects. The level, length, and width of a rectocele; the thickness of the perineal body; and the presence of an enterocele can be gauged by a simultaneous rectovaginal exam. Careful attention should be paid to the puborectalis muscle during the digital exam. With a finger hooked posteriorly in the rectum, the puborectalis should be felt contracting forward, narrowing the anorectal angle, when the patient is asked to squeeze, and relaxing when asked to expel the examining finger. Paradoxical contraction or nonrelaxation can be detected by this exam, while excessive levator tone or localized tenderness over the muscle may suggest levator ani syndrome. The vaginal mucosa can be well visualized with a rigid proctoscope which can then be inserted into the rectum to look for solitary rectal ulcer or other pathology that may be causing constipation.

Patients with a rectocele and obstructed defecation should also have some assessment of the anterior compartment, either by the treating surgeon or by a consulting gynecologist, to define the extent of other pelvic organ prolapse. This examination is best performed in the dorsal lithotomy position and standardized grading systems exist to quantify the degree of prolapse. Surgeons should be familiar with the two most popular scoring systems, the Baden-Walker and Pelvic Organ Prolapse Quantification (POP-Q), if for no other reason than to facilitate communication with our gynecology colleagues.^{52 53}

Ancillary Testing

The AGA position statement on constipation provides a detailed approach to the evaluation and management of patients with functional constipation, the complete details of which are beyond the scope of this review.⁵⁴ However, their step-wise and logical approach to the use of ancillary testing to distinguish subtypes of constipation and guide therapy is worth reviewing (Fig. 2). According to their recommendations, anorectal manometry and a balloon expulsion test should be the first ancillary tests performed after an initial trial of medication adjustment and increased fiber. A normal balloon expulsion test and manometry essentially eliminate the diagnosis of ODS and direct the investigation toward an assessment of colonic motility (Fig. 2). Patients with confirmed ODS are further investigated with defecography (with fluoroscopy or MRI) to assess pelvic floor motion (perineal descent and puborectalis relaxation) and structural abnormalities such as intussusception and rectocele. Importantly, surgery is only considered in this algorithm for patients with confirmed ODS who have a significant structural abnormality and normal pelvic floor relaxation (Fig. 3). This emphasizes the results of numerous studies showing that pelvic floor dysfunction is a poor prognostic finding for repair of structural defects associated with ODS.^{55 56}

Fig. 2.

Treatment algorithm for chronic constipation. (Adapted from the Medical position statement on constipation. From American Gastroenterological A, Bharucha AE, Dorn SD, Lembo A, Pressman A. American Gastroenterological Association. Gastroenterology. 2013;144(1):211–217).

Fig. 3.

Treatment algorithm for defecating disorders. MR, magnetic resonance; p.r.n., as needed. (Adapted from the Medical position statement on constipation. From American Gastroenterological A, Bharucha AE, Dorn SD, Lembo A, Pressman A. American Gastroenterological Association. Gastroenterology. 2013;144(1):211–217.)

Colonoscopy is not routinely indicated in the evaluation of constipation or ODS except as needed for cancer screening purposes or to evaluate a specific symptom such as rectal bleeding. We will typically order a Sitzmarks study for colonic transit early in the evaluation, along with balloon expulsion and anorectal manometry, because of the high rate of concurrent slow-transit constipation in patients with ODS. While it is true that addressing the outlet dysfunction should take precedence over motility, it is valuable for both the clinician and the patient to be aware of the dual diagnosis before initiating therapy.

The accuracy and cost-effectiveness of using MRI versus conventional fluoroscopy for dynamic defecography has been contested. Critics point out that conventional MR defecography is performed in a supine or lateral position rather than sitting; however, studies have shown no significant differences in the measurements from supine MRI versus upright colpocystoproctography.⁵⁷ Furthermore, the sensitivity and specificity of dynamic MR defecography has been shown to be equal or superior to conventional defecography, especially in the case of enterocele, rectocele, or middle compartment prolapse.^{58 59 60 61} This, combined with the lack of radiation exposure with MRI, makes it an attractive option for the identification of structural and functional causes of ODS (Fig. 4).

Fig. 4.

MR defecography of patient with obstructed defecation syndrome. This 55-year-old woman with a lifelong history of functional constipation was examined by MR defecography. Results are shown at rest (a) and during attempted evacuation (b). The patient was unable to expel the gel during the exam. A rectocele is present (thick arrow), but there is also significant perineal descent (thin arrow), and the anorectal angle remains constant, indicating nonrelaxing puborectalis. B, bladder; PCL, pubococcygeal line; PS, pubic symphysis; R, rectum..

Gynecologists frequently repair pelvic organ prolapse with no workup beyond a physical exam. Only 6% of gynecologists order defecography in their routine preoperative evaluation of rectocele.⁶² This likely reflects a different set of symptoms and different goals for surgery between rectocele patients seen by gynecologists as opposed to those with ODS referred to colorectal surgeons. Patients with isolated vaginal symptoms or simultaneous anterior compartment prolapse or urinary incontinence with no history of constipation may be appropriately treated by transvaginal rectocele repair in conjunction with repair of other vaginal abnormalities. However, failure to inquire about constipation or the need for splinting may underestimate a patient's true symptomatology and predispose the patient to early recurrence due to chronic straining. On the other hand, patients referred to a colorectal surgeon for dysfunctional defecation and/or rectocele warrant a thorough workup before recommending correction of the structural defect. While it is true that identification of all coexisting derangements in a patient's defecatory mechanics does not clarify which treatment will best correct the problem, to simply ignore them in favor of repairing the vaginal bulge is overly simplistic and should be discouraged.

Treatment

A rectocele is a structural problem which is correctable only by surgery, though the symptoms may be treated by medical management or lifestyle alterations. Several surgical techniques have been described for repair of a rectocele with the primary technical goal of restoring normal anatomy. Comparison of retrospective trials is difficult due to the highly variable presentation, evaluation, and follow-up of patients undergoing repair. Most studies report as their primary outcomes the objective correction of prolapse and/or the patients' subjective impression of the repair as successful or unsuccessful. Improvement in specific functional symptoms is variably reported and does not appear to correlate well with restoration of anatomy. In general, the technical goal of preventing the rectum from bulging into the posterior vaginal wall can be accomplished by either strengthening the posterior wall of the vagina (by plicating it in the midline, repairing defects in it, or resuspending it from the sacrum) or reinforcing the front wall of the rectum (also by plication or ventral suspension). Another approach is the interposition of tissue or a graft between the two structures. Recently, partial resection of the rectal wall has been advocated as another method of reinforcing the anterior rectal wall and decreasing redundancy. When faced with the structural disorder of rectocele in the setting of functional constipation from ODS, nonsurgical treatments may prove as effective as operative repair.

Medical

Initial medical management of constipation includes discontinuation of any contributing medications, gradually increasing fiber intake to 35 g/day, increasing water consumption, and adding an osmotic or stimulant laxative. Patients who report significant improvement in symptoms may need no further treatment, regardless of the presence of a rectocele. Suboptimal response and a workup revealing ODS as the cause of constipation is an indication for referral to a pelvic floor physical therapist for biofeedback. As a significant number of ODS patients have severe anxiety, depression, or a history of childhood sexual trauma, referral to a psychotherapist should be considered as well.⁶³

Biofeedback

Biofeedback has been shown to be highly effective in the treatment of ODS, even in the presence of structural defects which may impair evacuation.^{9 64 65 66 67} Lau et al looked specifically at response to treatment among patients with anismus to determine the influence of structural defects on outcomes. The authors found no difference in the incidence of rectocele between patients who did or did not respond to biofeedback.⁶⁷ Mimura et al prospectively investigated the role of biofeedback in patients with ODS specifically with large rectoceles (>2 cm).⁶⁶ They reported modest improvement in multiple outcome measures after biofeedback and significant improvement in bowel frequency. Importantly, they found no predictive influence on response to therapy with pretreatment anismus, need for digitation, rectocele size, or contrast trapping. Recently, Hicks et al reported an overall 71% rate of symptom improvement among rectocele-positive patients treated with medical management and biofeedback, with even higher rates among patients with concomitant intussusception. After biofeedback, ODS scores in patients with rectocele were significantly reduced from 15 to 10.5 (p < 0.001), a similar response to that noted in patients who underwent surgery (13.5–10.5 [p < 0.001]).⁹

Pessary

Pessaries have been used for centuries to treat pelvic organ prolapse, but little objective data on their effectiveness are available. A 3-year prospective study from the United Kingdom found that 76% of women presenting with pelvic organ prolapse and opting for vaginal pessary were able to successfully retain their pessary at 4 weeks. Of these, 85% had success and maintained pessary use for 5 years.⁶⁸ Yamada and Matsubara found that overall successful fitting of vaginal ring pessaries was high, but rectoceles were more common in the failure group than in the successful group (56 vs. 9%) and proved to be a predictor of pessary failure. They did not recommend against pessary use in patients with a rectocele but hypothesized that unlike anterior defects where the pubic bone still provides support to the pessary, posterior defects may render the ring pessary unsuccessful. The authors did not study space-occupying pessaries and did not record functional outcomes related to defecation.⁶⁹ In appropriately selected patients, a trial of pessary use should be made an option.

Surgery

Posterior Colporrhaphy

Transvaginal plication of the rectovaginal septum is the preferred approach to rectocele repair for most gynecologists and some colorectal surgeons. An incision is made in the vaginal mucosa at the level of the perineal body and extended vertically toward the apex of the vagina. The mucosa is separated from the underlying fibromuscular layer of the vaginal wall by sharp and blunt dissection to a point above the rectocele and laterally to the vaginal sulcus at the medial edge of the puborectalis. Midline plication of the fibromuscular tissue is then performed with absorbable suture, typically in an interrupted fashion. Distal plication of the levators may be performed to normalize the vaginal hiatus. Several authors emphasize the importance of anchoring the repair to the perineum and performing a concomitant perineoplasty when the muscles of the perineal body have been disrupted or separated, to avoid perineal recotocele.^{15 21} Redundant vaginal mucosa is excised and the edges are reapproximated with absorbable suture.

The identification and repair of discreet breaks or tears in the rectovaginal septum has been advocated by some as a more anatomic approach to repair of a rectocele. The technique is performed similar to traditional posterior colporrhaphy except that instead of plicating the fascia and levators to the midline, a finger is inserted in the rectum to aid in identifying specific sites of weakness which are repaired, usually in a transverse fashion. This technique is dependent on the ability of the operator to locate fascial defects, identify their margins, and repair them. It also assumes that the posterior vaginal wall and its attachments are strong at all other points. Identification of discreet defects becomes more difficult at the apex of the vagina, where the fibromuscular wall of the vagina is thin and the rectovaginal septum is composed mainly of thin endopelvic fascia. In the lower vagina, the most common location of a discreet break is actually a transverse separation of the perineal body from the rectovaginal septum, the reattachment and stabilization of which is essential to the repair.²²

The addition of a prosthetic or biologic graft to transvaginal rectocele repairs has been proposed as a potential modification to increase durability. Three studies have randomized patients to transvaginal repairs with or without graft augmentation, using prosthetic or biologic mesh. In a prospective randomized trial, Sand et al found no difference in the rate of recurrent rectocele 12 months after transvaginal repair with or without the addition of a polyglactin mesh (7/67 vs. 6/65, p = 0.71).⁷⁰ Sung et al performed a prospective, randomized, double-blinded study of transvaginal rectocele repair with or without reinforcement with a porcine subintestinal submucosal (SIS) graft. Among 137 women at a median follow-up of 12.2 months, the authors found no difference in the rate of anatomic failure between patients with or without graft reinforcement (11.9 vs. 8.6%, p = 0.5). Furthermore, ODS symptoms (straining, splinting, or incomplete evacuation) remained common in both groups (43.8 vs. 44.8%, p = 0.9).⁷¹ Paraiso et al compared conventional posterior colporrhaphy to site-specific repair with or without SIS graft. At 1 year, there was no difference in the recurrence rate between posterior colporrhaphy (4/28, 14%) and site-specific tissue repair (6/27, 22%), but patients with graft augmentation had a significantly higher recurrence rate (12/26, 46%), p = 0.2.⁷² Based on these data, the use of graft reinforcement cannot be recommended.

Transanal Plication

Longstanding rectoceles can lead to a thinning of the anterior rectal wall and the development of redundant rectal mucosa. This observation has been suggested as a potential explanation for the persistence of defecatory symptoms in many women undergoing traditional posterior colporrhaphy.⁷³ With the aim of decreasing the size of the rectal vault, resecting redundant mucosa, and reinforcing the anterior rectal wall, several transanal approaches to rectocele repair have been described. The transanal repair, the preferred approach of many colorectal surgeons, is usually performed under general anesthesia in the prone jackknife position after mechanical bowel prep. A bi-valved retractor is placed in the rectum and the submucosal plane is infiltrated with an epinephrine:saline solution (1:200,000). A “T-” or “I”-shaped incision is made in the center of the mucosa longitudinally and flaps are elevated. The muscularis of the rectum and the deeper fibromuscular tissue are plicated transversely with interrupted absorbable sutures, with care taken to avoid the vaginal mucosa. Redundant mucosa is excised before closure of the incision with absorbable sutures. Variations on this technique include transverse or elliptical incisions, vertically oriented plication, or plication without mucosal incision.

Transanal Resection

Out of the notion that excess tissue in the anterior rectal wall complicates defecation has evolved the concept of transanal rectal resection as a means of treating ODS associated with rectocele and internal intussusception. Originally described using the circular PPH-01 stapler, the stapled transanal rectal resection (STARR procedure) has been met with great enthusiasm in some circles, especially among European surgeons. Numerous modifications of the technique have been described including using a single PPH-01 stapler, two PPH-01 staplers, the semi-circular Contour 30 stapler, a Contour and a linear stapler, a Contour with two linear staplers, and a Contour with a PPH-01.⁷⁴ Overall, the early results seem promising, but some serious complications have been reported including rectal obliteration, rectal perforation, pelvic sepsis, staple-line hematoma, chronic proctalgia, and rectovaginal fistula.^{75 76 77 78} Furthermore, there is no consensus on which technique should be applied to which patient or what factors are predictive of success.

Perhaps more concerning than the uncertainty regarding indications and outcomes for the STARR procedure is its potentially harmful effect on the general understanding of ODS. As outlined throughout this review, numerous derangements of structure and function are typically found in patients with an inability to defecate. The STARR procedure was developed specifically for patients with ODS attributable to internal intussusception or rectocele. In the recent STARR literature, there is a trend toward omitting this distinction, referring to the STARR procedure simply as a “treatment for ODS,”^{79 80 81 82 83} with ODS regarded as a single entity caused by mechanical obstruction of stool by internally prolapsing tissue.⁸⁴ Even if the studies themselves have more stringent inclusion criteria, the description of STARR as an “operation for ODS” is problematic. Generalizing ODS as a single entity that can be cured with a stapler is a dramatic reversal of everything we have learned about the complex physiologic and psychological mechanisms that contribute to a patient's inability to evacuate stool.

Abdominal Suspension

An alternative approach to direct reinforcement of the rectovaginal septum for rectocele repair is to resuspend the vagina, rectum, and/or perineal body from the sacral promontory. Cundiff et al described a constellation of defects addressed by rectocele repair via sacral colpoperineopexy, which they proposed would correct the common finding of perineal descent, improve constipation symptoms, and avoid damaging stretch on the pudendal nerves.⁸⁵ As described earlier, a posterior mesh anchored to the perineal body and posterior vaginal wall was combined with an anterior mesh on the vaginal apex and suspended from the sacral promontory. Unfortunately, the long-term outcomes were extremely poor with regard to defecatory problems, with 89% of patients reporting persistent symptoms of splinting, straining, or incomplete evacuation after surgery.⁸⁶ The authors cite unrecognized “non-prolapse etiologies of obstructed defecation” as a possible explanation for the lack of improvement despite restoration of anatomy.

Like the posterior vaginal wall, the anterior rectum may be resuspended from the sacrum as a treatment for rectocele. Originally described by D'Hoore in 2004 as a procedure for rectal prolapse,⁸⁷ the laparoscopic ventral rectopexy has recently been applied to the treatment of symptomatic rectocele with ODS. Wong et al examined ODS scores in a consecutive series of 41 patients undergoing rectocele repair by minimally invasive ventral rectopexy.⁸⁸ While overall ODS scores did not change significantly from preoperatively to postoperatively (median: 9.5 vs. 6, p = 0.74), a subset of analysis of patients with preoperative ODS scores more than 6 did reveal a significant improvement (median: 14.5 vs. 9.5, p = 0.030). In a series of 157 patients with symptomatic rectocele, Jonkers et al used the CCCSS to analyze the impact of laparoscopic ventral rectopexy on ODS symptoms.⁸⁹ The authors found a decrease in the percentage of patients with a CCCSS score more than 15 from 51 to 17% after surgery (p < 0.001).

Choice of Operation

The small number of prospective studies, inconsistent inclusion criteria, and variable outcome measures make it difficult to know the optimal approach for rectocele repair. Functional outcomes, specifically those related to ODS symptoms, are poorly documented, especially in the gynecologic literature, where these symptoms often are not the primary indication for repair. In one of the largest published series on transvaginal repair, Kahn and Stanton assessed outcomes in 244 women undergoing posterior colporrhaphy with long-term follow-up (mean: 42.5 months) in 140 patients. They reported an anatomic cure in 76% and an overall improvement in prolapse symptoms, but the reported rates of constipation, incomplete emptying, and fecal incontinence were all higher after surgery.⁹⁰ Studies focusing specifically on patients with ODS symptoms often include patients with mucosal prolapse, rectal prolapse, internal intussusception, enterocele, or any other number of coexisting defects that might influence the success of a particular repair. Functional derangements such as paradoxical contraction of the puborectalis or abnormal perineal descent are likewise inconsistently reported.

Few meaningful comparisons of surgical techniques exist in the literature. Regarding traditional posterior colporrhaphy versus site-specific repair, numerous retrospective series and a single randomized trial have found similar rates of anatomic cure and similar rates of dyspareunia and ODS symptoms, with a trend toward higher recurrence with site-specific repair.^{25 72 91 92} Data comparing transvaginal and transanal rectocele repair are also weak. A recent Cochrane review on the surgical management of pelvic organ prolapse in women identified only two randomized studies with a total of 87 patients on which to base its analysis of transvaginal versus transanal rectocele repair. The report found no significant difference in the recurrence rates between the two approaches (2 out of 39 transvaginal vs. 7 out of 48 transanal; relative risk: 0.32, 95% confidence interval: 0.07–1.34).⁹³ No reliable comparison of functional outcomes between the 2 approaches is available.

Summary

Ultimately, the decision to offer surgical repair of rectoceles should be rooted in the surgeon's belief that repair of the defect, by whatever method they choose, will significantly improve the symptoms with which the patient presented. Restoration of anatomy and avoidance of recurrence are not sufficient endpoints to define the success of an operation for rectocele. Unlike inguinal or incisional hernia repair in which relief of symptoms and restoration of function are secondary goals to the primary purpose of reducing the risk of incarceration, in rectocele repairs, relieving symptoms is the only goal. It is therefore incumbent on the surgeon to determine what symptoms are present, the degree to which they are attributable to the rectocele, and the likelihood that they can be positively impacted by surgery. A patient with multicompartment pelvic organ prolapse and no history of constipation who wants relief from pelvic heaviness and mucosal irritation may be very happy with a transvaginal colporrhaphy as part of a more involved vaginal prolapse repair. In patients seeking relief from ODS who are noted to have a rectocele, a more thorough investigation is warranted before simply offering to fix the bulge. Even if the primary reason for repair is vaginal symptoms, some examination of bowel function is indicated because chronic constipation, even if unrelated to the rectocele, may predispose to recurrence of the prolapse.