Anal Incontinence and Rectovaginal Fistula
James Unger
DEFINITIONS
Anal incontinence—Involuntary loss through the anus of flatus, liquid stool, or solid stool that is perceived as a social or hygienic problem.
Anal manometry—A diagnostic test that provides information regarding function, sensation, compliance, and the presence of intact reflexes within the anal canal and distal rectum.
Anal ultrasonography—An imaging modality used to define the anatomy of the sphincter complex. It is the simplest, most reliable, least invasive test for definition of anatomic defects in external and internal anal sphincter.
Closing reflex—Exaggerated increase in muscle contraction allowing for anal closure at the end of defecation.
Defecography—A diagnostic test that involves imaging of the rectum with contrast material and observation of the process, rate, and completeness of rectal evacuation using fluoroscopic techniques.
External anal sphincter—Striated muscle sphincter, under voluntary control, that is responsible for much of the squeeze pressure in the anal canal.
Extra-anal incontinence—Involuntary loss through a fistula of flatus, liquid stool, or solid stool.
Fecal incontinence—Involuntary loss through the anus of liquid or solid stool only; excludes flatal incontinence.
Fourth-degree laceration—Complete rupture of the external anal sphincter and laceration of the anorectal mucosa.
Internal anal sphincter—Thickened continuation of the circular smooth muscle layer of the bowel, under autonomic control, that is responsible for much of the resting pressure in the anal canal.
Noble procedure—Transperineal approach to anal sphincteroplasty and perineoplasty in patients with chronic perineal laceration. Since it involves mobilization of the anterior rectal wall outside the anal orifice, it can also be used for analvaginal fistula repair. Major advantage is the lack of sutures in the anorectal canal.
Pelvic floor electromyography—A diagnostic test that evaluates the pelvic floor muscles for evidence of nerve injury. It is performed for three purposes: to identify areas of sphincter injury by mapping the sphincter, to determine whether the muscle contracts or relaxes, and to identify denervation-reinnervation potential indicative of nerve injury.
Puborectalis muscle—Striated muscle medial portion of the levator ani that is responsible for maintaining and increasing the anorectal angle and contributes to the squeeze pressure in the anal canal.
Pudendal nerve terminal motor latency (PNTML)—Operatordependent technique to measure the response of the external anal sphincter to stimulation of the pudendal nerve. Prolongation of response has been associated with anal sphincter weakness due to pudendal nerve damage, although this is very controversial.
Rectovaginal fistula—Epithelial-lined connection between the rectosigmoid or anal canal and the vagina or perineum.
Rectoanal inhibitory reflex (RAIR)—Complex process in which sampling of rectal contents by the anal canal occurs every 8 to 10 minutes in response to rectal filling. This reflex primarily occurs through relaxation of the upper internal anal sphincter with continued pressure maintained by the lower internal anal sphincter. Concurrent contraction of the external anal sphincter and puborectalis muscle force the sampled contents back into the rectum.
Resting pressure—Reflects the tonic activity of both internal anal sphincter and external anal sphincter; approximately 75% to 85% of this pressure is derived from the internal anal sphincter.
Sacral neuromodulation—Stimulation of S2-S4 nerve roots via an implanted electrode introduced through selected sacral foramina, which in turn is connected to an implanted electrostimulator in order to cause contraction of levator muscles and external anal sphincter.
Secca procedure—Transanal delivery of temperature-controlled radiofrequency energy to the sphincter complex, which results in tissue remodeling and improved anal sphincter function.
Third-degree laceration—A partial or complete laceration of the external anal sphincter without involvement of the anorectal mucosa.
INTRODUCTION
Obstetricians and gynecologists have an important role to play in the diagnosis, management, and prevention of anal incontinence in women. Women with anal incontinence may bring this problem to the attention of their obstetrician-gynecologist as a
primary complaint or in many cases discuss it only after close questioning about symptoms. This is because many women are understandably reluctant to mention such problems on their own due to feelings of embarrassment. This discussion may occur with the acute onset of symptoms shortly after a pregnancy or much later in life unrelated to recent childbirth. While in the first case such incontinence may be the direct result of acute childbirth injury, in the latter case, it may be related to chronic injury due to childbirth injury years ago or unrelated to any childbirth event. In either case, the gynecologist is often the first consulted to provide evaluation and management.
primary complaint or in many cases discuss it only after close questioning about symptoms. This is because many women are understandably reluctant to mention such problems on their own due to feelings of embarrassment. This discussion may occur with the acute onset of symptoms shortly after a pregnancy or much later in life unrelated to recent childbirth. While in the first case such incontinence may be the direct result of acute childbirth injury, in the latter case, it may be related to chronic injury due to childbirth injury years ago or unrelated to any childbirth event. In either case, the gynecologist is often the first consulted to provide evaluation and management.
In addition, the prevention of obstetric trauma and its adverse effects on pelvic floor function including anal continence has become a major public health question. It seems clear that sphincter injury at the time of vaginal delivery significantly increases the odds of anal incontinence even years later beyond the postpartum period. Pollack and coworkers reported a 53% anal incontinence rate at 5 years in women who had undergone an anal sphincter tear compared to 32% in women without a tear. They calculated an odds ratio of 2.3 (95% CI 1.1 to 5.0) for anal sphincter tear and later incontinence. More recently, Evers and coworkers also calculated an odds ratio of 2.3 (95% CI 1.27 to 4.26) for anal incontinence women 5 to 10 years following an anal sphincter laceration. They also concluded that vaginal delivery in the absence of an anal sphincter laceration was not a risk factor for anal incontinence.
Surgical repair of chronic anal sphincter injury, the major subject of this chapter, is by no means uniformly successful, and results seem to deteriorate with time. Gutierrez and colleagues reported on 191 patients at 10 years after anal sphincter repair and found that only 6% of the 130 patients who returned questionnaires had full continence. In addition, 58% were incontinent of solid stool. Zutshi and coworkers reported on 31 patients with a median age of 44 years at the time of overlapping sphincter repair with a median follow-up of 129 months. No patients were fully continent at 129 months. Patients who were older at the time of surgery and those who had undergone multiple vaginal births had the worse outcomes. Dudding and coworkers in a systematic review of the literature in 2008 determined that only 20% of patients were continent to liquid and/or solid stool at 10 years after overlapping sphincter repair.
Likewise, obstetric lacerations involving the rectum may result in rectal/anal-vaginal fistulae even when expertly repaired. Although childbirth-related fistulae can usually be successfully repaired with surgery, they are physically and emotionally distressing. Sometimes, more than one surgical attempt is required. It is also evident that certain obstetric practices such as midline episiotomy and operative vaginal delivery may contribute to later adverse effects on fecal continence, most likely by increasing the risks of anal sphincter and rectal injury. This has generated considerable discussion and debate regarding the appropriate use of episiotomy, operative vaginal delivery, and even cesarean section on demand for the sole purpose of preventing pelvic floor injury as noted in a recent commentary by Nygaard.
Approximately 1.4% of normal community-dwelling adults are incontinent to liquid or solid stools. Perry and coworkers reported on a total sample of 10,116 adults over age 40 years a major fecal incontinence rate of 1.4%. There was no significant difference between men and women, although incontinence was more prevalent and more severe in older people. Walter and colleagues in a random sample of 2,000 community-dwelling individuals between the ages of 31 and 76 reported a rate of leakage of solid stools of 1.4% in women and 0.4% for men. The prevalence increases with normal aging to 6% to 7% of elderly adults over age 60 years living independently according to a study performed in the Netherlands by Teunissen and colleagues. On the other hand, Tobin and Brocklehurst reported that 10.3% of residents in 30 different residential homes for the elderly had fecal incontinence at least once weekly. Therefore, anal incontinence is a symptom that affects millions of women and has many causes in addition to childbirth trauma (Table 44.1). In fact,
Whitehead and coworkers demonstrated that after adjusting for age, neither vaginal delivery nor gender was predictive of anal incontinence. Like Perry, they reported that increasing age is the major determinant of fecal incontinence. However, many cases in women do seem to be strongly associated with pelvic floor injury, especially injury to the anal sphincter sustained either overtly or covertly during childbirth as previously noted. These injuries may result from direct traumatic disruption to some component of the anal sphincter mechanism or functional deterioration over time as a result of occult nerve injury and aging. Identifying the role of childbirth and risk factors associated with delivery that are linked to later incontinence is important since some of these factors may be modifiable by altering certain childbirth practices and thus lowering the risk of subsequent incontinence. Women with either acute or chronic anatomic injury as a cause of their incontinence are the ones most likely to benefit from care of the well-trained surgical gynecologist. The obstetrician-gynecologist not only will be the first one to be involved in the prevention of such injuries but also will be responsible for the surgical repair required in the delivery room. In addition, the obstetrician and gynecologist may be the first one consulted to evaluate women with chronic fecal incontinence and must be able to carefully assess such women with a thorough history, physical examination, and appropriate imaging or testing, in order to arrive at a precise etiology of this condition. At this point, patients may be referred to other specialists such as those in female pelvic medicine and reconstructive surgery, gastroenterologists, colorectal surgeons, physical therapists, and others if treatment of the underlying condition is outside the expertise of the patient’s obstetrician-gynecologist.
Whitehead and coworkers demonstrated that after adjusting for age, neither vaginal delivery nor gender was predictive of anal incontinence. Like Perry, they reported that increasing age is the major determinant of fecal incontinence. However, many cases in women do seem to be strongly associated with pelvic floor injury, especially injury to the anal sphincter sustained either overtly or covertly during childbirth as previously noted. These injuries may result from direct traumatic disruption to some component of the anal sphincter mechanism or functional deterioration over time as a result of occult nerve injury and aging. Identifying the role of childbirth and risk factors associated with delivery that are linked to later incontinence is important since some of these factors may be modifiable by altering certain childbirth practices and thus lowering the risk of subsequent incontinence. Women with either acute or chronic anatomic injury as a cause of their incontinence are the ones most likely to benefit from care of the well-trained surgical gynecologist. The obstetrician-gynecologist not only will be the first one to be involved in the prevention of such injuries but also will be responsible for the surgical repair required in the delivery room. In addition, the obstetrician and gynecologist may be the first one consulted to evaluate women with chronic fecal incontinence and must be able to carefully assess such women with a thorough history, physical examination, and appropriate imaging or testing, in order to arrive at a precise etiology of this condition. At this point, patients may be referred to other specialists such as those in female pelvic medicine and reconstructive surgery, gastroenterologists, colorectal surgeons, physical therapists, and others if treatment of the underlying condition is outside the expertise of the patient’s obstetrician-gynecologist.
TABLE 44.1 Causes of Anal Incontinence | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
FUNCTIONAL ANATOMY OF ANAL CONTINENCE
The mechanisms and structures involved in defecation and preservation of anal continence are reviewed in detail in the following section. A summary of the structures and processes involved is illustrated in Figures 44.1 and 44.2.
 FIGURE 44.1 Location of the anal sphincter in relation to other structures of the female pelvic floor. |
 FIGURE 44.2 Algorithm for anal continence mechanism. Diagrammatic illustration of the anal canal and sphincters. |
Rectum
Adequate knowledge of the anatomy of the rectum and the anorectal canal and its normal function in defecation is critical to understanding the mechanisms of anal continence (Table 44.2). The rectum lies posterior to the vagina as it descends below the cul-de-sac peritoneum. Here, it accomplishes one of its major functions contributing to continence, that of a temporary appropriate storage unit when defecation needs to be delayed. The puborectalis muscle, which encircles the rectum and pulls
it forward toward its own attachment at the symphysis pubis, marks the distal margin of the rectum and the beginning of the anal canal. In addition to the external and internal anal sphincters, it is an important muscular component of the continence mechanism. In addition to its storage role, the rectum functions as a very important sensory unit responsible for the initiation of the continence mechanism. Finally, when the appropriate time arrives, the rectum also has dynamic propulsive activity allowing complete emptying of its contents until it is empty.
it forward toward its own attachment at the symphysis pubis, marks the distal margin of the rectum and the beginning of the anal canal. In addition to the external and internal anal sphincters, it is an important muscular component of the continence mechanism. In addition to its storage role, the rectum functions as a very important sensory unit responsible for the initiation of the continence mechanism. Finally, when the appropriate time arrives, the rectum also has dynamic propulsive activity allowing complete emptying of its contents until it is empty.
TABLE 44.2 Mechanisms of Anal Continence | ||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||
The process of storage and appropriate evacuation of stool begins when the rectum senses the presence of material within its lumen. Not unlike the bladder, the cerebral cortex is responsible for interpreting this information as regards its precise nature and the problem of what to do about it. Fortunately, also like the bladder, the rectum is a high-volume, low-pressure system when healthy. It can accommodate increasing volumes of material, that is, feces, until a critical volume, and thus ultimately critical pressure is reached. This compliance of the rectum can be compromised due to inflammation of the rectum as in inflammatory bowel disease or with fibrosis as occurs with radiation. Compliance can be measured clinically with anorectal manometry with balloon distention as is discussed later in the section on testing. Patients with urgency, incontinence, and even constipation may have issues with abnormal compliance and/or abnormal sensation. When normal sensation and compliance exists, the sensation of rectal filling allows one to consciously initiate the anal continence mechanism, which is discussed further below. However, without normal rectal compliance, even an otherwise intact continence mechanism will be severely compromised.
 FIGURE 44.3 Rectosigmoid colon and anal canal showing collateral arterial circulation from superior hemorrhoidal (inferior mesenteric), middle hemorrhoidal (hypogastric or internal iliac), and inferior hemorrhoidal (internal pudendal) arteries. |
Rectal expulsion of its contents when appropriate is the final event in defecation. It seems to be under both voluntary and reflex control. The voluntary nature involves the willful relaxation of the muscular components of the continence mechanism followed by the reflex propulsive contractions of the rectum usually assisted by a voluntary increase in intra-abdominal pressure. This continues until the rectum is completely empty as determined by sensory input from the anal canal.
Anal Canal
The actual structures involved in the anal continence mechanism include the anal canal and its surrounding muscular structures. However, not to be forgotten is the important role of the rectum as it interacts with the continence process as described above. For the obstetrician and gynecologist, it is the anal canal and its related structures such as the anal sphincters that will be most directly linked to the problem of fecal incontinence.
The anal canal in the female is approximately 2.5 to 4 cm in length and normally remains completely collapsed because of the tonic contraction of the sphincters. The lining of the anal canal is nonuniform. The proximal 1 cm is lined by rectal-type columnar mucosa, followed by modified or stratified columnar epithelium for about 1.5 cm. The distal half of the anal canal is lined by squamous epithelium, which is richly supplied by branches of the inferior hemorrhoidal nerves and is exquisitely sensitive. The anal mucosa, like the rectum, is also surrounded by the inner circular layer of smooth muscle, which is in turn surrounded by the outer longitudinal layer. The main blood
supply to the rectum and the anal canal is from branches of the superior and inferior hemorrhoidal arteries (Fig. 44.3).
supply to the rectum and the anal canal is from branches of the superior and inferior hemorrhoidal arteries (Fig. 44.3).
 FIGURE 44.4 Anal manometry demonstrating pressures in the anal canal. Rectal pressure reflects intra-abdominal baseline pressure recorded in the rectal reservoir. Resting pressure essentially reflects the effect of the involuntary, smooth muscle internal anal sphincter. Squeeze pressure largely reflects the transient contractile effort of the voluntary, striated external anal sphincter. |
The muscular portions of the anal continence mechanism or anal sphincter consist of three separate muscular structures surrounding the anal canal: the internal anal sphincter, the external anal sphincter, and the puborectalis muscle. The internal anal sphincter provides most of the involuntary resting tone to the anal canal, while the external anal sphincter and the puborectalis muscle are under voluntary control and contribute the most when we call upon the need to delay defecation.
The Internal Anal Sphincter
The circular smooth muscle layer of the rectal wall, which is under autonomic control, thickens at the proximal anal canal to form the internal anal sphincter. It triples in thickness and is approximately 3 cm in length and 3 mm wide. It is longer in men and thickens with age. It provides about 85% of the resting anal tone with the remainder supplied by the hemorrhoidal vessels and external sphincter. These pressures can be shown by anal manometry demonstrated in Figure 44.4. The internal sphincter terminates just short of the most caudal portion of the external anal sphincter and is separated from it by the longitudinal smooth muscle layer of the bowel. The internal anal sphincter seems to have two major functions: prevention of passive leakage of anal contents and initiation of the rectoanal inhibitory reflex (RAIR) or anorectal sampling as demonstrated in the anal manometry study illustrated in Figure 44.5.
Injury to the internal anal sphincter may result in passive leakage of anal contents, which may be continuous or intermittent. It should also be suspected when patients complain of loss of stool completely without warning or awareness. Injury to the internal anal sphincter can be diagnosed by endoanal sonography as is discussed in the section on childbirth injury.
 FIGURE 44.5 Rectoanal inhibitory reflex. Arrival of stool in the rectum triggers reflex relaxation of the internal anal sphincter, allowing the bolus to enter the proximal anal canal, where discrimination of the solid, liquid, or gaseous nature of the bolus by the sensory-rich upper anal canal occurs. |
Anorectal sampling or the RAIR occurs every 8 to 10 minutes in response to rectal filling. The upper internal anal sphincter relaxes in response to rectal filling to allow the sensory epithelium of the anal canal to determine the nature of the rectal contents. The amount of relaxation of the internal sphincter is directly related to the degree of rectal distention. Incontinence does not occur because the lower internal sphincter maintains higher pressure all the while this is occurring. Concurrent contraction of the external anal sphincter and the puborectalis muscle forces the sampled contents back into the rectum. Clearly, damage to the internal anal sphincter, which allows for a larger volume of sampling contents to enter an anal canal that is further compromised by a lack of opposing pressure, may result in leakage. This compromise will be further worsened if there is a lack of voluntary action on the part of the external anal sphincter and puborectalis to force the contents back into the rectum.
The External Anal Sphincter
The external anal sphincter extends approximately 1 cm beyond the internal sphincter. It is generally but somewhat arbitrarily divided into three components: deep, superficial, and subcutaneous (Fig. 44.1). The deep part blends with the fibers of the puborectalis as its fibers encircle the undersurface of the rectum on its way to the symphysis pubis. The subcutaneous portion is attached to the perianal skin, and the superficial portion forms the bulk of the muscle. Classically, the female anal sphincter complex has been thought of as forming a broad band of tissue posteriorly but narrowing to a small tubular bundle of tissue anteriorly within the perineal body. In the past, the internal sphincter has been most often portrayed as a minor structure. These concepts came from many centuries of anatomic study of cadavers augmented by observations in the delivery room, at the time of third- and fourth-degree laceration repairs, of the transected ends of the anterior portion retracted into round holes in the perineal body. In the recent past, it has become possible to view this anatomy undisturbed in living, continent, nulliparous subjects with magnetic resonance imaging (MRI). In an early MRI study, Aronson and colleagues found the shape of the combined internal and external anal sphincter complex to be nearly cylindrical as it encircles the anal canal. Measured in the midline, the anterior portion of the anal sphincter complex appeared as a broad band of tissue, not a narrow tube. In this study, 54% of the anterior thickness was smooth muscle of the internal anal sphincter. Subsequent studies, including the cadaver studies of DeLancey et al., found a similar shape with a substantial contribution from the internal anal sphincter. This substantial
anterior anal sphincter length and thickness, as well as the contribution from the internal sphincter, is important to keep in mind during primary repair of obstetric lacerations as well as in surgical correction of anal incontinence secondary to chronic perineal laceration. The external anal sphincter is a striated muscle that is under voluntary control via the pudendal nerve. It provides a certain degree of constant tone to the anal canal due to its slow-twitch muscle fibers. However, it can undergo muscle fatigue due to its “fast-twitch” fibers as well. Its primary function is to generate adequate pressure in the anal canal when needed to allow for postponement of defecation. It accomplishes this by allowing the rise in anal pressure to exceed that in the rectum, until rectal accommodation occurs through normal rectal compliance, and rectal pressure falls (Fig. 44.6).
anterior anal sphincter length and thickness, as well as the contribution from the internal sphincter, is important to keep in mind during primary repair of obstetric lacerations as well as in surgical correction of anal incontinence secondary to chronic perineal laceration. The external anal sphincter is a striated muscle that is under voluntary control via the pudendal nerve. It provides a certain degree of constant tone to the anal canal due to its slow-twitch muscle fibers. However, it can undergo muscle fatigue due to its “fast-twitch” fibers as well. Its primary function is to generate adequate pressure in the anal canal when needed to allow for postponement of defecation. It accomplishes this by allowing the rise in anal pressure to exceed that in the rectum, until rectal accommodation occurs through normal rectal compliance, and rectal pressure falls (Fig. 44.6).
 FIGURE 44.6 A: Lateral defecography of normal continent patient deferring defecation by contracting puborectalis sling, increasing anorectal angle, and exerting squeeze pressure in the anal canal. Note how barium bolus is supported over the levator plate and away from the fatigable anal sphincter complex. B: Lateral defecography of normal continent patient willfully defecating by relaxing puborectalis sling, decreasing anorectal angle, and decreasing pressure in the anal canal. Note how barium bolus becomes lined up over the anal canal. |
As previously mentioned, the external anal sphincter also plays a vital role in the RAIR by helping to generate opposing pressure when the internal anal sphincter relaxes and along with the puborectalis returns the sampled contents into the rectum. This response of the external anal sphincter is critical in preventing passive fecal loss.
 FIGURE 44.7 Function of the anal sphincter and puborectalis muscle. A: At rest, the constant tone of the puborectalis muscle pulls the anorectal junction anteriorly to create an approximately 90-degree angle between the rectum and anal canal, closing the rectal inlet and maintaining continence of solid stool. At moments of need, this muscle can be contracted further, increasing the angle and supporting the stool bolus over the levator plate. B: During defecation, the puborectalis muscle relaxes, opening the rectal inlet, while intestinal peristalsis and the voluntary increase in intraabdominal pressure move stool into the anal canal. The anorectal angle is decreased, and the stool bolus is lined up over the anal canal. |
Finally, the external anal sphincter is important in protecting against postdefecation incontinence. This is demonstrated by the so-called closing reflex. When traction is released on the external anal sphincter as fecal material is evacuated, there is an exaggerated increase in muscle contraction. This allows for anal closure at the end of defecation.
The Puborectalis Muscle
The puborectalis muscle is the most medial portion of the levator ani. The puborectalis muscle encircles the anorectal junction pulling it toward its insertion on the inner surface of the pubis. It thus forms a sling that creates the anorectal angle. The anorectal angle is an approximately 90-degree angle that closes the rectal inlet and maintains the rectum over the levator plate rather than directly over the anal canal (Fig. 44.7A). It does this by tonic contraction utilizing its slow-twitch muscle fibers.
When there is a need to delay defecation, it can contract further under voluntary control and further increase the anorectal angle. When defecation is permitted, the puborectalis relaxes, decreasing the anorectal angle, allowing the rectum to line up directly over the anal canal (Fig. 44.7B). The role of puborectalis injury in anal incontinence is unclear.
When there is a need to delay defecation, it can contract further under voluntary control and further increase the anorectal angle. When defecation is permitted, the puborectalis relaxes, decreasing the anorectal angle, allowing the rectum to line up directly over the anal canal (Fig. 44.7B). The role of puborectalis injury in anal incontinence is unclear.
THE EFFECT OF PREGNANCY AND CHILDBIRTH ON THE PELVIC FLOOR
In this section, we review some of the effects of pregnancy and childbirth on the function of the pelvic floor and also the direct effects of childbirth trauma on the anal sphincter mechanism. This is an area of much conflicting data that are difficult to interpret in terms of developing a clear understanding of the natural history of pregnancy and childbirth and how they relate to anal incontinence. It is hard to tease out the effects of pregnancy itself, vaginal delivery with or without sphincter injury, occult sphincter injury, neurologic injury, efficacy of sphincter repair, and other issues contributing to incontinence. However, it is an area of intense current investigation, and hopefully, many of these questions will be answered in the future. For obstetricians-gynecologists and the women they care for, childbirth-related injuries will be the most common cause of anal incontinence they will have to deal with.
General Effects on the Pelvic Floor and Continence
Signs of pelvic floor dysfunction often develop during pregnancy itself. This adds to the difficulty of trying to differentiate pregnancy-related pelvic floor dysfunction as to its timing— antepartum versus intrapartum. These problems include both urinary incontinence and fecal incontinence. King and coworkers reported that 65% of their nulliparous patients reported anal (flatus and/or fecal) incontinence during the third trimester. Fourteen percent complained of fecal incontinence. By 2 weeks postpartum, the anal incontinence rate had dropped to 47%, while the fecal incontinence rate had risen to 17%, including 10 women who had new-onset fecal incontinence by the 2-week visit. Of these 10 subjects, 3 had spontaneous vaginal delivery, 3 delivered by forceps, and 4 had delivered by cesarean delivery after labor. At 6 months postpartum, 49% still complained of incontinence of flatus and 11% of fecal incontinence. Solans-Domenech reporting for the Catalan Agency for Health Technology Assessment and Research found that 39.1% of previously continent pregnant women developed urinary incontinence at some time during their pregnancy. Ten percent of women in this study developed anal incontinence at some point during the pregnancy. This was incontinence of flatus only in 90%, although during the postpartum period nearly one third of these women noted some loss of liquid or solid stool. For both urinary incontinence and fecal incontinence during pregnancy, risk factors were related to increased weight and age over 35 years. The major risk factor for the persistence of incontinence at the 7-week postpartum visit was vaginal delivery. However, in the case of anal incontinence, this was true only if anal incontinence was also present during pregnancy. Instrumental delivery and episiotomy were not significant risk factors for postpartum incontinence. Overall, 7.3% of women continued to complain of fecal incontinence at their postpartum visit. This is very similar to the 11.2% prevalence of anal incontinence noted in the vaginal delivery group and the 10.3% prevalence in the cesarean delivery group reported by Borello-France for the Pelvic Floor Disorders Network. At 6 months after delivery, the anal incontinence rate had dropped to 8.2% and 7.6% in the vaginal and cesarean groups, respectively. However, women who had sustained a known sphincter injury reported much higher prevalence of fecal incontinence at both the 6-week postpartum visit and at 6 months, 26.6% and 17%, respectively.
Stretch and compression injury to the pudendal nerve, which innervates the external anal sphincter, and to the S2, S3, and S4 nerve fibers, which directly innervate the levator muscles, may be responsible for incontinence not associated with direct muscle trauma. This may be due to the normal descent of the fetal head or by trauma from forceps. Snooks and coworkers have identified such other risk factors as length of the second stage of labor, multiparity, and high birth weight. Allen and colleagues studied 96 nulliparous women and found electromyography (EMG) evidence of reinnervation in the pelvic floor muscles after vaginal delivery in 80% of those studied. Women with a prolonged second stage and larger babies had the most evidence of nerve damage. Surprisingly, forceps deliveries and perineal tears did not affect nerve damage. None of these studies included neurophysiologic testing during pregnancy itself, and therefore, it is not possible to tease out the role pregnancy itself apart from delivery may have played. However, even though it appears these denervation injuries are common, occurring in up to 80% of women, they seem to be minor and to resolve in most. However, Snooks in a follow-up study in 1990 found evidence of persistent neuropathy in some women even at 5 years.
Another area that may contribute to pelvic floor dysfunction due to childbirth is a direct effect on pelvic muscle strength. This may or may not be due to direct injury to the levator. A recent study by Friedman and coworkers in parous women revealed decreased muscle strength in women with anal incontinence and pelvic organ prolapse 6 to 11 years following vaginal delivery. Vaginal delivery overall was associated with a significant reduction in pelvic muscle strength compared to cesarean delivery. The greatest reduction in strength was in women who had forceps delivery. However, in the study noted above by King, which demonstrated a very high prevalence of anal incontinence in pregnant nulliparous women, they were not able to demonstrate a difference in pelvic muscle strength between those with and without symptoms during the antepartum or postpartum period. Finally, DeLancey and coworkers have reported an association between defects in the levator ani muscle demonstrated by MRI and pelvic organ prolapse. However, the exact role these injuries may play in the etiology of anal incontinence is unknown.
Anal Sphincter Injury and Incontinence
Clinically recognizable anal sphincter injury occurs in from 3.3% to 19% of vaginal deliveries. Handa and coworkers in a large population-based study of over 2 million vaginal deliveries in California reported a frequency of anal sphincter lacerations of 5.85%. Faltin and colleagues reported occult injuries detectable by ultrasound in up to 28% of women without a clinically suspected injury. Direct disruption injury to the external anal sphincter during childbirth has been commonly associated with anal incontinence in women. Farrell and coworkers in a prospective observational study found that incontinence of flatus at 6 months postpartum was higher in women who had sustained a third- or fourth-degree laceration compared to those women without anal sphincter injury. Likewise, earlier, Sultan and coworkers reported higher rates of anal incontinence in women with demonstrable external anal sphincter defects postpartum. Pollack and colleagues reported
a 5-year prospective study where 53% of women with a sphincter injury reported anal incontinence compared to 32% without a sphincter injury. Likewise, as reported above by Borello-France for the Pelvic Floor Disorders Network, there was an 11.2% prevalence of anal incontinence noted in the vaginal delivery group and the 10.3% prevalence in the cesarean delivery group. At 6 months after delivery, the anal incontinence rate had dropped to 8.2% and 7.6% in the vaginal and cesarean groups, respectively, in this study. However, women who had sustained a known sphincter injury reported much higher prevalence of fecal incontinence at both the 6-week postpartum visit and at 6 months, 26.6% and 17%, respectively.
a 5-year prospective study where 53% of women with a sphincter injury reported anal incontinence compared to 32% without a sphincter injury. Likewise, as reported above by Borello-France for the Pelvic Floor Disorders Network, there was an 11.2% prevalence of anal incontinence noted in the vaginal delivery group and the 10.3% prevalence in the cesarean delivery group. At 6 months after delivery, the anal incontinence rate had dropped to 8.2% and 7.6% in the vaginal and cesarean groups, respectively, in this study. However, women who had sustained a known sphincter injury reported much higher prevalence of fecal incontinence at both the 6-week postpartum visit and at 6 months, 26.6% and 17%, respectively.
More recently, Evers and coworkers, using anal incontinence and quality-of-life scores to define incontinence, reported on 937 women 5 to 10 years after first delivery, 90 of whom had sustained a clinically diagnosed sphincter injury at the time of delivery. Nineteen percent of women with a sphincter injury had anal incontinence as defined by the score, compared to 10% in the vaginal delivery without injury group and 9% in the cesarean group. Sultan also raised the issue of the effectiveness of primary end-to-end repair of the external anal sphincter performed at the time of delivery by reporting that 82% of women still had a sphincter defect present by ultrasound 7 weeks later. Of note, 33% of women who did not have a recognized anal sphincter injury at delivery were noted to have a defect present at 7 weeks. Subsequently, Sultan and his team reported retrospective case series of 32 women who underwent overlapping sphincter repair in the operating room under anesthesia, of whom only 7% had incontinence of flatus and none had fecal incontinence at 5 months postpartum. However, three subsequent randomized studies (Fitzpatrick et al., 2000; Williams et al., 2006; Fernando et al., 2006) as well as a Cochrane review in 2006 all failed to demonstrate a difference in fecal incontinence between repair types.
In 2010, Farrel and coworkers reported a prospective randomized study on nulliparous women assigned to either end-to-end or overlapping repair after sustaining third- or fourth-degree lacerations. By confining their study to nulliparous women, they hoped to eliminate bias generated by those cases of multiparous women who may have a preexisting injury. They reported a significantly increased rate of incontinence of flatus at 6 months in women undergoing overlapping repair compared to those undergoing end-to-end repair (61% vs. 39%). However, there was not a significant difference in fecal incontinence between the two groups (15% vs. 8%). Ultrasound findings of sphincter defects did not correlate with flatal or fecal incontinence except for those women who demonstrated both internal and external anal sphincter defects. Combined disruption noted on ultrasound was significantly associated with fecal incontinence with an odds ratio of 6.5. There was also no significant difference in rates of isolated internal, external, or combined sphincter defects in either group. Internal anal sphincter defects were present in 46% of the end-to-end repair group versus 38% in the overlapping repair; external anal sphincter defects in 53% versus 62%, respectively; and both sphincters in 32% for each. Rates for incontinence following repair were not statistically different whether there was an intact sphincter on ultrasound or not except as noted above for fecal incontinence in the women with persistent combined defects versus no defect (incontinence of flatus 52% vs. 54%, respectively; fecal incontinence 30% vs. 6%, respectively). Women with a persistent internal anal sphincter defect had a 60% rate of incontinence of flatus versus 50% of those with an intact internal anal sphincter, while fecal incontinence was present in 23% versus 7%, respectively. Finally, women with a persistent external anal sphincter defect had a 51% rate of incontinence of flatus versus 57% with an intact sphincter, while fecal incontinence occurred in 20% versus 7%, respectively. The authors concluded that the primary cause of incontinence following a third- or fourth-degree laceration was denervation injury rather than persistent structural defects following repair. Likewise, it is interesting to note that Frudinger and coworkers failed to demonstrate a link between anal sphincter disruption diagnosed in asymptomatic postpartum women by ultrasound only and anal incontinence 10 years later. The issue of the true relationship between endoscopically defined anal sphincter defects, denervation injury, and symptoms of incontinence later remains unclear.
There are known risk factors for anal sphincter trauma at vaginal delivery. A recent review by Dudding and colleagues in which 451 articles and abstracts were studied concluded that major risk factors for injury included instrumental delivery, prolonged second stage of labor, birth weight greater than 4 kg, fetal occipitoposterior presentation, and episiotomy. The population-based study of over 2 million vaginal deliveries by Handa and coworkers noted earlier identified primiparity, macrosomia, and operative vaginal delivery as major risk factors. In addition, episiotomy, especially midline episiotomy, is an important risk factor for sphincter lacerations as demonstrated by Angioli and colleagues as well as by FitzGerald and coworkers.
Finally, obstetric lacerations can extend into the rectum resulting in a so-called fourth-degree laceration. Although these usually also involve a complete tear through the anal sphincter, this is not always the case. At times, there may be an isolated laceration through the rectum or anal canal with an intact external anal sphincter. Therefore, these isolated injuries must be looked for and are usually diagnosed by digital rectal exam at the time of delivery. On occasion, the rectal mucosal portion of a repair will break down with or without concomitant disruption of the sphincter and perineum, which can result in rectovaginal fistula. In Hibbard’s series of 24 rectovaginal fistulae and 27 chronic perineal lacerations, 47 (92%) were caused by obstetric trauma. Even meticulously repaired lacerations will occasionally completely break down. This leaves the patient with an open cloacal deformity or an isolated rectal/anal-vaginal fistula with an intact perineum and sphincter. Anal incontinence usually is a problem for these patients until the tissues are healed sufficiently to allow secondary repair. Surgical repair will be discussed in detail in the section on fistula repair.
Obviously, the issue of anal incontinence during and following pregnancy and delivery is complex as to its etiology and treatment. Especially confusing is the role of neurologic injury versus structural damage to the anatomic sphincter mechanism. The best surgical method to repair sphincter injury and the effect of the adequacy of the repair on overall continence later is equally unclear. Finally, the overall contribution of childbirth injury to the ultimate development of long-term anal incontinence is uncertain as well, although the significance of intrapartum injury appears to decrease with age. This subject is particularly important when it comes to counseling older patients with anal incontinence and who are also found to have clinical or imaging evidence of sphincter disruption remote from delivery.
Other Causes
A promoter of anal incontinence over time is the fact that the normal process of aging results in decreased efficiency of the anal sphincter complex. Haadem and associates showed a natural decline with aging in resting and squeeze pressures in the anal canal of continent subjects. With increasing age, the
internal sphincter generates a lower resting pressure, and the proportion of fibrous tissue increases. The squeeze pressure able to be exerted by the external sphincter suffers a natural decline as well. Vaccaro and coworkers found pudendal neuropathy to be an age-related phenomenon as well in patients with anal incontinence and constipation. Because of this, many patients with damaged continence mechanisms may be able to compensate for a period of time, but they go on to decompensate secondary to these age-related changes and become incontinent.
internal sphincter generates a lower resting pressure, and the proportion of fibrous tissue increases. The squeeze pressure able to be exerted by the external sphincter suffers a natural decline as well. Vaccaro and coworkers found pudendal neuropathy to be an age-related phenomenon as well in patients with anal incontinence and constipation. Because of this, many patients with damaged continence mechanisms may be able to compensate for a period of time, but they go on to decompensate secondary to these age-related changes and become incontinent.
Functional problems of the bowel, such as constipation or diarrhea, can result in incontinence. Chronic constipation with its repeated straining at stool can cause stretch-induced injury to pudendal enervation. Overall, the most common cause of anal incontinence in the elderly, particularly those who are institutionalized, is fecal impaction with overflow incontinence. A diarrheal state can result in anal incontinence even in a patient with normal anorectal function, owing to the presence of large quantities of liquid stool that may overwhelm a normally functional mechanism. Neurologic disorders affecting sphincter control also may result in anal incontinence. Usually, the neurologic defect is widespread, and anal incontinence is but one manifestation.
Traumatic injury, more often a side-straddle injury in young girls, may result in simple or extensive laceration of the perineum. The extent of the injury may be difficult to determine because of pain, fear, edema, hemorrhage, and hematoma formation. Examination under anesthesia is advisable so that appropriate repair can be made, looking carefully for lacerations of the anal sphincter and rectum as well as other structures. Hematoma dissection above the levator muscles must be ruled out with pelvic ultrasound and with careful rectovaginal examination under anesthesia. Above the levator muscles, there is nothing to impede the progression of a hematoma until the diaphragm is reached.
Rectal trauma from operative procedures can affect rectal capacity and compromise anal continence or perhaps lead to extra-anal incontinence through a rectovaginal fistula. Entry into the rectum may occur during posterior colpoperineorrhaphy, especially when the anterior rectal wall and posterior vaginal skin are closely adherent with little, if any, intervening connective tissue. Such an enterotomy should be repaired transversely if longitudinal closure would compromise rectal capacity. A difficult hysterectomy, either abdominal or vaginal, may result in injury to the rectum, especially when dissection behind the cervix is difficult because of dense adhesions, indurated tissue from infection, or involvement of the cul-de-sac and anterior rectal wall with endometriosis. If the rectal defect does not heal properly or is not closed properly, a high rectovaginal fistula may develop through the newly closed vaginal apex. Partial excision of anal sphincter and other muscles involved in maintaining anal continence may be required in extensive vulvectomy for cancer resulting in anal incontinence, as reported by Berek and others. Injury to the anterior rectal wall may occur during hemorrhoidectomy, excision of a Bartholin gland, or colpotomy for pelvic abscess drainage.
Forty to sixty percent of patients with rectal or rectal mucosal prolapse also have some degree of anal incontinence. Although this was originally thought to be secondary to a dilatation effect on the internal anal sphincter, there is often evidence of associated neurogenic damage to the external sphincter muscles as well. These individuals may present with a patulous anal sphincter, passive stretching of the puborectalis muscles, and a long history of straining with constipation. The tissue prolapse may be related to prolapsing internal hemorrhoidal tissue or rectal mucosa or complete (full-thickness) rectal prolapse. Questions such as the frequency of occurrence of the prolapse, association with activity or defecation, estimated distance of the prolapsed tissue from the anus, spontaneous or manual reducibility, and history of incarceration may be helpful in better defining the type of problem. Patients may have relatively occult prolapse of tissue, which can manifest as mucous or brownish staining in the undergarments with occasional blood.
On physical examination, it is important to note the following: a patulous anus, which may indicate that chronic tissue prolapse has been occurring. The presence of enlarged external hemorrhoidal tissue may suggest the presence of enlarged internal hemorrhoidal complexes; however, they may be found in relative isolation and may be mistaken as prolapsed tissue. Patients should be asked to bear down as if defecating to manifest any prolapsed tissue. Concentric circular folds and a large protruding mass suggest the presence of complete prolapse, which often extends well beyond the anus. Noncircumferential folds may be either mucosal prolapse or hemorrhoidal tissue. If the patient’s history suggests the presence of prolapse and it is not visualized with the patient in the left lateral or lithotomy position, it is often helpful to have the patient sit on a commode with subsequent examination in this position. Correction of prolapsing tissue such as hemorrhoids may lead to improvement in anal continence, if this is the primary cause of the problem. In patients with complete rectal prolapse undergoing surgical repair, the rate of some degree of residual incontinence ranges from 40% to 90%, depending on the type of repair used.
Crohn disease is the most important of the variety of inflammatory bowel diseases that may cause extra-anal incontinence through a rectovaginal fistula. Among 138 patients with rectovaginal fistulae seen at Duke University Medical Center, Bandy and associates reported that 15 (11%) were caused by Crohn disease. The diagnosis, perioperative management, and surgical technique chosen for these patients constitute a special challenge for the gynecologic surgeon. Crohn disease must be considered as a possible cause of rectovaginal fistula in any patient in whom other causes are not clear, particularly if the fistula orifice is tender to palpation. Crohn disease also should be considered in the patient who has failed multiple attempts at fistula repair because the fistula tends to recur at the operative site in these individuals.
The role of diverticulitis as a cause of sigmoidovaginal fistulae has been highlighted by Tancer and Veridiano, who report on 130 such cases. These fistulae usually present with a malodorous vaginal discharge in women older than 50 years of age, some years after a hysterectomy. Such fistulae commonly develop between the inflamed bowel segment and the apical vaginal scar from the previous hysterectomy, although they may infrequently occur through a retained cervix. Often, the diverticular disease is silent and is only discovered with further investigations of the fistula tract. A diverticular abscess may have formed and found drainage though the path of least resistance, which in this case is the thinned out scar of the vaginal cuff.
Malignant tumors may erode through the tissues between the vagina and the rectum. When a patient with tumor involvement of the rectovaginal wall receives radiation, sloughing of the tumor may result in a rectovaginal fistula. Radiation also may cause a rectovaginal fistula without tumor erosion.
Rarely, anal incontinence may be congenital, as seen in one of 5,000 newborn girls who have an imperforate anus with associated fecal incontinence through a congenital rectovaginal or rectoperineal fistula. Total rectal agenesis is rare. Paul and Lloyd reported hindgut duplication with a congenital rectovaginal fistula; however, most anal incontinence is acquired.
Finally, although anal incontinence may be termed idiopathic in about 10% of women, the mechanism of incontinence in these women is usually secondary to pelvic floor denervation.
Finally, although anal incontinence may be termed idiopathic in about 10% of women, the mechanism of incontinence in these women is usually secondary to pelvic floor denervation.
In summary, although an intact and functional anal sphincter complex and puborectalis sling are important in the maintenance of anal continence, it must be remembered that a variety of other factors are involved, including intact anorectal sensation and reflexes, rectal capacity and distensibility, reasonable colonic transit time, appropriate stool volume and consistency, and adequate patient mental function and mobility.
EVALUATION OF ANAL INCONTINENCE
The most common cause of anal incontinence that the obstetrician-gynecologist will encounter is that related to obstetric injury. As noted previously, these patients may present shortly after childbirth or many years later. In the latter case, this may be because of failure of compensatory mechanism with time and aging. It is these later cases that require the most extensive diagnostic workup and which are the most resistant to surgical repair of the structural defect.
History and Physical Examination
Most often, history alone provides strong clues as to the etiology of the incontinence. The temporal relationship between a recent traumatic vaginal delivery and the postpartum onset of incontinence, especially urgency, will quickly lead one to suspect an undiagnosed sphincter injury or a failed sphincter repair. This will be quickly confirmed by pelvic examination without the need for additional testing. Likewise, flatal or fecal incontinence without urgency in this context should lead one to carefully search for a rectovaginal or anovaginal fistula. These are usually located in the lower one third of the vagina and are often identified as a red area of what appears to be granulation tissue. This usually represents an isolated small disruption of the previously repaired anal or rectal suture line. Fecal material is usually present near the fistula opening within the vagina. These fresh injuries are usually easy to identify as opposed to some small older fistulae for which patients seek care years later.
Patients who present remote from childbirth with anal incontinence require a more extensive workup both due to the possibility of a cause unrelated to childbirth and to develop a rational plan for repair, if indicated, with reasonable patient expectations for success. It may be helpful in these cases especially to attempt to quantify the degree of incontinence using one of a number of scoring systems. Scoring systems should include the effect of incontinence on the patient’s quality of life and allow for monitoring the response to treatment. There are a variety of these validated questionnaires available. Avery and coworkers have published a very helpful review of these with graded recommendations. Although no questionnaires received an “A” grade for anal incontinence, three were recommended with grades of “B.” These were the Fecal Incontinence Quality of Life Scale (Rockwood et al., 2000), Manchester Health Questionnaire (Bug et al., 2001), and Birmingham Bowel and Urinary Symptoms Questionnaire (Hiller et al., 2002).
Pelvic examination should include a careful inspection of the posterior vaginal wall, perineum, anal sphincter, anal canal, and rectum, including assessment of the patient’s function as well as anatomy. A patulous anus indicates a major loss of sphincter function and can be associated with rectal prolapse. Most obstetric injuries are associated with an anterior segmental defect in the external anal sphincter and may appear as the loss of the perineal body, loss of the corrugated appearance surrounding the anus, or attenuation of the rectovaginal septum in some instances. In more subtle cases in which the perineal body appears intact but the external sphincter is actually separated, only the dimples of the laterally retracted ends of the anal sphincter muscles may be apparent. This produces a “dovetail” appearance, as described by Toglia and DeLancey, in which the normal radial distribution of the anal creases is absent anteriorly but is present laterally and inferiorly. If there is a question as to the presence of a segmental defect, endoanal or transperineal ultrasound can be useful in delineating the anatomy. Next, a screening evaluation of the perineal reflexes to assess the integrity of the S2-S4 dermatomes should be performed. Perception of pinpoint and light touch over the perineal skin and buttocks can be tested easily with the broken end of a wooden cotton swab or a safety pin. Light stroking of the inferolateral margin of the labia majora should cause a reflex contraction of the bulbocavernosus muscle within the labia. The anal wink reflex can be elicited by lightly stroking the perianal skin or touching it with a pin to cause a reflex contraction of the external anal sphincter. Asking the patient to cough also should elicit the reflex contraction of the external anal sphincter. With both of these maneuvers, the anal canal should constrict concentrically owing to contraction of the external sphincter, and the anus should be pulled inward secondary to the contraction of the puborectalis muscle. In women with separation of the external anal sphincter, voluntary contraction of the pelvic floor muscles can cause an accentuation in the lateral perineal dimpling of the retracted ends. In patients with a denervated sphincter, there is no retraction of the anal skin during voluntary contraction. Patients who demonstrate abnormalities of these pelvic floor reflexes may require more in-depth neurologic evaluation.
Extra-anal incontinence may result from a fistulous tract. A rectovaginal fistula is usually easily diagnosed by careful inspection of the posterior vaginal wall. By spreading the labia, a low fistula can be revealed, usually involving the area of a previous episiotomy or obstetric laceration. A high fistula can be seen using a bivalve speculum and often appears at the vaginal cuff scar. A straight-handled speculum can be useful as it can be rotated to allow full visualization of both the anterior and posterior vaginal walls. The vaginal opening of a fistula may be localized by the presence of feces in the vagina or by the dark red rectal mucosa seen protruding at the fistulous opening contrasting with the lighter pink vaginal mucosa. Colposcopic examination of the vagina sometimes assists in the identification of a small fistula orifice. When the fistula is small, it may be difficult to locate both the vaginal and rectal ends of the fistula, but both orifices must be located for complete care to be given. A small probe can be pushed gently from the vaginal side of the fistula and the tip felt on a rectally placed finger. Instillation of methylene blue through the vaginal orifice of the fistula may aid in the proctoscopic visualization of the rectal orifice. Carey describes the following examination technique to identify a suspected rectovaginal fistula. A Foley catheter with a 10-mL balloon is inserted into the anus, while the posterior vaginal wall is painted with a concentrated solution of soap and water, or, alternatively, the vagina can be filled with water. As the rectum is distended with air by a syringe attached to the Foley catheter, the vaginal orifice of the fistula may be localized by the presence of bubbles forming at the fistula site. A small probe may then be passed along the fistula tract.
Alternatively, when a rectovaginal fistula is suspected but cannot be identified, radiologic studies such as a vaginogram or fistulogram may identify a fistulous tract. These studies are superior to barium enema for identifying a fistula because they
use a thin, water-soluble radiopaque medium rather than a thick barium solution. As such, although fistulae occasionally may be identified by barium enema, the intraluminal pressure of the bowel often is inadequate to force the barium solution through a small fistula opening. In addition, the presence of barium in the lower bowel may obscure a fistula tract. Likewise, a tampon test may be performed in which a tampon is placed in the vagina and the patient is given a small enema with methylene blue-colored water, which she is asked to retain for 20 minutes. Blue discoloration of the tampon is noted in the presence of a fistula.
use a thin, water-soluble radiopaque medium rather than a thick barium solution. As such, although fistulae occasionally may be identified by barium enema, the intraluminal pressure of the bowel often is inadequate to force the barium solution through a small fistula opening. In addition, the presence of barium in the lower bowel may obscure a fistula tract. Likewise, a tampon test may be performed in which a tampon is placed in the vagina and the patient is given a small enema with methylene blue-colored water, which she is asked to retain for 20 minutes. Blue discoloration of the tampon is noted in the presence of a fistula.
When a rectovaginal fistula is diagnosed, it is important to complete a thorough assessment of the anal sphincter and pelvic floor as well because these patients may have multiple defects. If not properly evaluated preoperatively, the patient may undergo successful repair of her fistula only to become anally incontinent postoperatively. Her compromised sphincter function may have been adequate preoperatively when excess pressure was bypassing the sphincter through the fistula. After her fistula is repaired and the full force of rectal contents is delivered to the sphincter complex, there may not be adequate function to maintain continence. A full treatment of rectovaginal fistula management appears later in this chapter.
A thorough digital rectal examination should be performed. Any rectal mass must be noted and the stool consistency assessed. A gross assessment of the patient’s resting and squeeze pressures within the anal canal and her ability to contract her levator ani muscles should be included. Anal sphincter tone should be evaluated with the patient at rest and during sphincter contraction. An anterior sphincter defect may be easily detectable as the loss of the palpable muscular ring within the perineal body. Even in the absence of external anal sphincter muscle anteriorly, a scarified band of tissue can remain that completes the contractile ring and helps the patient maintain continence. Next, the anorectal axis can be assessed. On rectal examination, the puborectalis muscle is palpable posteriorly at the junction between the rectum and the anal canal. By directing the examining finger posteriorly, the angle between the anus and rectum can be estimated and should approximate 90 degrees in a normal woman. More important, when the patient is asked to squeeze the sphincter, the puborectalis muscle should pull the examiner’s finger anteriorly toward the pubic bone.
Finally, some patients may complain of passive incontinence, or the loss of stool without their awareness. This may indicate damage to the internal anal sphincter. It cannot be assessed adequately by physical examination alone, but can be determined by radiologic and physiologic tests only. A defect of the internal sphincter can be assumed, however, if there is significant thinning of the rectovaginal septum. In many of these cases, both the internal and external anal sphincters are injured. Certainly, the patient with a chronic perineal laceration or cloacal deformity by definition has a segmental defect in both sphincters. From a practical and technical viewpoint, the internal anal sphincter cannot be repaired unless the external sphincter also is repaired. Surgery is rarely effective for an isolated defect of the internal sphincter. When surgical repair of the external anal sphincter is indicated, however, one should consider repairing defects in the internal sphincter as well.
Testing
Clearly, many patients will have a diagnosis and be ready to proceed to treatment after their history and physical examination. For others, the picture will be less clear. There may be questions about a particular patient that thoughtful use of testing can answer. Is a segmental defect present in the internal or external anal sphincter? What is the functional status within the anal canal? Is rectal sensation normal? Is the innervation to the striated musculature of the continent mechanism intact? How does the patient actually defecate? Judicious use of testing based on the information needed to arrive at an accurate diagnosis and to plan successful treatment can be important. The American Gastroenterological Association guidelines (March, 1999) for anorectal testing techniques are available through their Web site at www.gastro.org/practice/medical-position-statements. Procedures of value in addition to symptom diary and rectal examination include (a) anal ultrasonography, (b) anorectal manometry, (c) rectal and anal sensory testing, and (d) rectal compliance testing. These are now covered in detail. Endoanal, transvaginal, and transperineal ultrasound techniques have made it simple and relatively inexpensive to identify defects in both the internal and external anal sphincters (Fig. 44.8). These defects can go clinically unrecognized, but may be amenable to surgical repair if the patient is symptomatic. Anal endosonography is one radiologic technique for assessing posttraumatic defects of the internal and external anal sphincters. High-resolution images of the separate sphincter muscles are obtained using a rotating Endoprobe, and anatomic defects can be identified as a loss of continuity of the muscle rings. Several studies have found that anal endosonography correlates well with needle EMG mapping of sphincter defects, manometric mapping of sphincter defects, and intraoperative findings. Ultrasound is less timeconsuming than EMG or manometry and much more comfortable for the patient. Other studies have established that transvaginal ultrasound is equally efficacious to endoanal ultrasound in identifying sphincter defects. Peschers and colleagues described normal sphincter and puborectalis anatomy as well as defects in both the internal and external sphincters using exoanal ultrasonography: a conventional 5-MHz convex transducer placed on the perineum. Ultrasound currently is the study of choice for establishing the presence or absence of a segmental anal sphincter defect. The approach chosen— endoanal, transvaginal, or transperineal—may depend on the equipment available and operator expertise.
 FIGURE 44.8 Endoanal ultrasound image of an anally incontinent patient with a segmental external anal sphincter defect. Probe, endoanal ultrasound probe; M, anal mucosa; IAS, internal anal sphincter; EAS, external anal sphincter; defect, segmental defect of external anal sphincter. (Photograph courtesy of Justin A. Maykel, MD.) |
Endoanal ultrasound identifies anatomic defects or thinning of the internal anal sphincter, but Heyer and associates showed that interpretation of external sphincter images is much more subjective and confounded by normal anatomic variations in the external anal sphincter. Indeed, the external anal sphincter and perirectal fat are both echogenic and frequently indistinguishable; the external sphincter may be asymmetrical in the upper anal canal, particularly in women. A new imaging modality with better soft tissue definition may be desirable. MRI may prove to be superior to endoanal ultrasound because of high tissue contrast between the external anal sphincter and the perirectal fat. DeSouza and associates reported 100% concordance between MRI performed with an endoanal coil and surgical findings for presence, size, and location of anal sphincter tears in seven patients with obstetric trauma. Additionally, Lienemann and colleagues and Healy and associates have demonstrated that MRI also can define dynamic pelvic floor motion during defecation and squeeze. Rapid image acquisition is crucial for optimal visualization of dynamic motion, particularly because patients cannot maintain rectal expulsion or puborectalis contraction for more than 15 to 30 seconds.
Anal manometry provides information regarding function, sensation, compliance, and the presence of intact reflexes within the anal canal and distal rectum. It is most useful in defining functional sphincter weakness. Here, it is complementary to endoanal ultrasonography in determining whether this weakness is secondary to an anatomic defect. The first part of this test is essentially a pressure profile of the anal canal, providing information on the functional status of the internal and external anal sphincters. Some computer-based, multichannel manometry equipment can provide graphic cross-sectional analysis of the anal canal to help detect the presence of segmental sphincter defects.
The test usually is performed with the patient in the left lateral decubitus position without any special bowel preparation. There are many different protocols for performing anal manometry. Most commonly, a fluid-filled pressure catheter with radial side ports located 90 degrees apart circumferentially is connected to pressure transducers, and a recording device is used to measure the anal canal pressures during rest and during voluntary contraction of the anal sphincter (Fig. 44.9). These pressures may be recorded either as the pressure catheter is slowly pulled through the anal canal—a station pull-through technique—or at static points along the anal canal as the pressure catheter is pulled out in certain increments, usually every 0.5 cm. The average pressure measured at rest in the anal canal is the resting pressure, and the highest pressure recorded along the anal canal with the patient at rest is the maximum resting pressure. The increase in pressure over the basal canal pressure initiated by voluntary contraction of the anal sphincter is the squeeze pressure, and the highest such increment is the maximum squeeze pressure. The resting anal canal length is measured from the point at which the anal sphincter pressure continuously exceeds the average intrarectal pressure by 4 mm Hg. The resting pressure is largely a reflection of the internal anal sphincter function, whereas the squeeze pressure reflects the strength of the external anal sphincter voluntary contraction (Fig. 44.4).
 FIGURE 44.9 Drawing of an anorectal manometry catheter. (Courtesy of Sandhill Scientific.) |
“Normal” range for maximum resting pressure is 40 to 80 cm H2O and for maximum squeeze pressure is 100 to 200 cm H2O; however, there is tremendous overlap in values between patients who are continent and those who are incontinent. As a group, incontinent patients have lower values on anal manometry testing than continent patients, although there is no discriminatory level that can be used to predict incontinence. It has been suggested that patients with a maximum resting pressure of less than 20 cm H2O and a maximum squeeze pressure of less than 40 cm H2
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
