Fecal Incontinence 

Updated: Mar 05, 2021
Author: Tanaz R Ferzandi, MD, MA; Chief Editor: Kris Strohbehn, MD 


Practice Essentials

Fecal incontinence is a syndrome that involves the unintentional loss of solid or liquid stool. Many definitions of fecal incontinence exist, some of which include flatus (passing gas), while others are confined to stool.[1]

True anal incontinence is the loss of anal sphincter control leading to the unwanted or untimely release of feces or gas. This must be distinguished from other conditions that lead to stool passing through the anus. Stool seepage that produces soilage of undergarments may result from hemorrhoids, enlarged skin tags, poor hygiene, fistula-in-ano, and rectal mucosal prolapse. Other conditions that result in poor bowel control are inflammatory bowel disease, laxative abuse, parasitic infection, and toxins. Fecal urgency also must be differentiated from fecal incontinence because urgency may be related to medical problems other than anal sphincter disruption.


History and physical examination

The history should include a thorough account of the type and duration of fecal incontinence, frequency of incontinent episodes, type of stool lost, impact of the disorder on the patient's life, history of associated trauma or surgery, and associated factors such as protective undergarment use. In addition, an obstetric history should be obtained.

A focused physical examination should be performed. After inspection, sensation of the perianal region should be assessed. Digital examination should be performed to detect obvious anal pathology and provide an initial assessment of the anal resting tone.

Imaging studies

The standard diagnostic imaging study for the anal sphincters is transanal or endoanal ultrasonography. See the image below.

Fecal incontinence. Ultrasound taken with the pati Fecal incontinence. Ultrasound taken with the patient in the low lithotomy position showing a defect in the internal anal sphincter at the level of the puborectalis muscle that can be seen as a U-shaped echogenic mass posteriorly.

Diagnostic procedures

Diagnostic procedures used in the workup of fecal incontinence include the following:

  • Anal manometry
  • Pudendal nerve terminal motor latency
  • Electromyelography
  • Defecography (evacuation proctography)

See Workup for more detail.


Conservative treatment options for fecal incontinence include bulking agents and biofeedback. The goal of medical therapy is to reduce stool frequency and improve stool consistency.

Once medical therapy has been maximized, minimally invasive therapies, such as sacral nerve stimulation, and surgical therapies may be considered. In select patients, injectable materials may improve anal sphincter function.

Several surgical procedures are performed for the treatment of anal incontinence. The standard procedure for incontinence due to anal sphincter disruption is the anterior overlapping sphincteroplasty.

See Treatment for more detail.


Fecal incontinence is one of the most psychologically and socially debilitating conditions in an otherwise healthy individual. It can lead to social isolation, loss of self-esteem and self-confidence, and depression. This article is devoted to the problem of fecal incontinence in the adult female patient.[2]

In 2015, summaries of the epidemiology, pathophysiology, classification, and treatment of fecal incontinence were published from the 2013 National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Workshop.[3, 4]


Bowel function is controlled by multiple factors including anal sphincter pressure, anorectal sensation, rectal compliance, colonic transit time, and stool volume and consistency. In addition, adequate cognitive function with appropriate ability to access bathroom facilities are necessary for continence. If any of these factors are compromised, incontinence can occur.

Fecal continence is maintained by the structural and functional integrity of the anorectal unit. Normal anal sphincter function is a critical part of continence. The anal sphincter is comprised of 2 components: the internal anal sphincter (IAS), which is a 0.3-0.5 cm expansion of the circular smooth muscle layer of the rectum and the external anal sphincter (EAS), a 0.6-1.0 cm expansion of the levator ani muscles. The IAS is chiefly responsible for maintaining continence at rest and contributes approximately 70-80% of the resting sphincter tone. This barrier is reinforced during voluntary squeeze by the EAS, the anal mucosal folds, and the anal endovascular cushions.

These barriers are further augmented by the puborectalis muscle, which forms a sling around the rectum and creates a forward pull to reinforce the anorectal angle. The anorectal angle, which is approximately 90 degrees at rest, is created as the rectum perforates the levator complex. During voluntary squeeze, the angle becomes more acute, whereas during defecation, the angle becomes more obtuse. Innervation of the EAS is from the pudendal nerve, a mixed motor and sensory nerve that arises from the second, third, and fourth sacral nerves (S2, S3, and S4). Innervation of the puborectalis arises more directly from the sacral nerves listed above.[5]

Continence requires the complex integration of signals among the smooth muscle of the colon and rectum, the puborectalis muscle, and the anal sphincters. As colonic contents are presented to the rectum, the rectum distends. The sensation of rectal distension is most likely transmitted along the S2, S3, and S4 parasympathetic nerves. This results in a parasympathetically mediated relaxation of the IAS (rectoanal inhibitory reflex) and a contraction of the EAS (rectoanal contractile reflex).

Rectal contents are allowed to come into contact with the very sensitive epithelial lining of the upper anal canal. The epithelial lining of the upper anal canal has a rich supply of sensory nerve endings, especially in the region of the anal valves. The contents are then sampled as to their nature (ie, gas, liquid, or solid). This sampling is described as an equalization of the rectal and upper anal canal pressures. Miller et al found that sampling occurred spontaneously in 16 of 18 control patients but in only 6 of 18 incontinent patients. Similarly, Miller's study demonstrated that patients with impaired continence had decreased thermal and electrical sensitivity to stimuli.[6] The decreased anorectal sensation and abnormal sampling likely contribute in the pathogenesis of anal incontinence as sampling facilitates the fine tuning of the continence barrier. This process is incompletely understood.[5, 7]

When defecation is desired, the anorectal angle straightens (which is facilitated by squatting or sitting), and abdominal pressure is increased by straining. This results in descent of the pelvic floor, contraction of the rectum, inhibition of the external anal sphincter, and subsequent evacuation of the rectal contents.

If evacuation of the rectum is not socially appropriate, sympathetically mediated inhibition of the smooth muscle of the rectum and voluntary contraction of EAS and puborectalis musculature occur. The anorectal angle becomes more acute and prevents the bolus of stool from descending further. The contents of the rectum are forced back into the compliant rectal reservoir above the levators, which allows the IAS to recover and contract again. A decrease in the compliance of this rectal reservoir has been associated with fecal urgency and anal incontinence. The exact mechanism of this is unclear and debate continues as to whether it is a cause or result of anal incontinence.

In essence, any process that interferes with these mechanisms, including trauma from vaginal delivery or a neurological insult, can result in fecal incontinence.

A 2015 summary of the epidemiology, pathophysiology, and classification of fecal incontinence from the 2013 National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Workshop has been published.[3]


Fecal incontinence has many etiologies. One or a combination of several factors can lead to the inability to control passage of stool or flatus. Congenital abnormalities such as spina bifida and myelomeningocele with resultant spinal cord damage can result in fecal incontinence. Inflammatory bowel disease may lead to decreased compliance of the rectum and may manifest as fecal urgency, frequency, soilage, or incontinence. Anal surgery, such as hemorrhoidectomy and sphincterotomy, has been associated with internal sphincter injury and subsequent urgency and incontinence.[8]

Medical conditions that may result in fecal incontinence include diabetes mellitus, stroke, spinal cord trauma, and degenerative disorders of the nervous system. These conditions may alter normal sensation, feedback, or function of the complex mechanism of anal continence. The patient's ambulatory status is also important when considering the etiology of the disorder. Wheelchair-bound patients and those who are dependent on others for access to bathroom facilities may have episodes of functional incontinence, yet have completely normal anorectal function.

Vaginal delivery is widely accepted as the most common predisposing factor to fecal incontinence in an otherwise young and healthy woman.[9] Vaginal delivery may result in internal or external anal sphincter disruption, or may cause more subtle damage to the pudendal nerve through overstretching and/or prolonged compression and ischemia.

Many studies support the theory that mechanical sphincter disruption contributes to fecal incontinence.[10] Abramowitz et al found a de novo sphincter defect rate of 16.7% after vaginal delivery (14% with external, 1.7% internal, and 1% both). The overall anal incontinence rate was 9%, and 45% of these women had identifiable sphincter defects on anal endosonography.[11]

In addition, inadequate repairs of obstetric sphincter injuries may contribute to delayed symptoms of fecal incontinence. In a study of 34 women who sustained third-degree obstetric anal sphincter tears and 88 matched controls, Sultan et al found that approximately half the women (47%) who sustained a third-degree tear experienced some impairment of anal continence despite a primary repair. Anal pressures were significantly lower, but pudendal nerve terminal motor latency measurements were not different. They relate the cause as a persistent mechanical sphincter disruption rather than pudendal nerve damage.[12]

Similarly, Nielsen et al found that 13 of 24 (54%) patients who underwent primary repair for sphincter disruption at the time of delivery demonstrated external sphincter defects at 3- and 12-month follow-up. Tetzschner et al found that 18 of 94 (19%) of women who sustained obstetric anal sphincter rupture had fecal incontinence shortly after delivery.[13] A follow-up study by the same group demonstrated that 30 of 72 (42%) women who had sustained an obstetric anal sphincter rupture had fecal incontinence at 2-4 years postpartum.[14] Factors that are significantly associated with an increased risk of third-degree obstetric sphincter tears are primiparity, occiput posterior presentation, use of forceps, fetal weight greater than 4000 g, perineal tears, episiotomy, and prolonged second stage of labor.[12, 15]

Pudendal nerve injury may also be a mechanism in fecal incontinence. The pudendal nerve innervates the external anal sphincter muscle, anal canal skin, and coordinates reflex pathways. Evidence of neurologic damage from the forces of labor is illustrated in a study by Fynes et al, who evaluated anorectal physiology in women undergoing cesarean delivery at different stages of labor. They found that anorectal function was unaltered in women undergoing cesarean delivery early in labor. In women with advanced cervical dilation (>8 cm) at the time of cesarean delivery, they found delayed pudendal nerve terminal motor latency (PNTML) and reduced anal squeeze pressures.[16]

In the Tetzschner study of women who sustained obstetric anal sphincter rupture during vaginal delivery, postpartum fecal incontinence was also associated with delayed PNTML of greater than 2.0 milliseconds.[13] This type of neurologic injury was confirmed in a recent large study. In 923 patients with fecal incontinence (745 women), 56% demonstrated pudendal neuropathy (38% were unilateral). Neuropathy was associated with decreased anal resting tone and squeeze pressures.[17] Similarly, Allen et al found that 80% of patients had electromyelogram (EMG) evidence of nerve damage after their first vaginal delivery, particularly if they had a long second stage of labor or delivered larger infants.[18]

Successive vaginal deliveries increase the risk of developing fecal incontinence. Sultan et al demonstrated that 35% of primiparous women and 44% of multiparous women had sphincter disruption after delivery but were asymptomatic. In this same study, 13% of primiparous patients and 23% of multiparous patients undergoing vaginal delivery complained of fecal incontinence or fecal urgency at 6 weeks postpartum. Symptoms are worse if a patient has had damage to the sphincter in a prior delivery or was symptomatic with anal incontinence previously. Ryhammer et al found a significant long-term association between the number of vaginal deliveries and several anorectal parameters in perimenopausal women. They noted lower perineal position at rest, increased perineal descent with maximal straining, decreased anal sensibility to electrical stimulus, and prolonged latency in pudendal nerves to be associated with increasing parity.[19]

In recent years, national attention has been focused on the risks and benefits of cesarean delivery prior to labor and its impact on pelvic floor health. However, the literature is divided as to whether cesarean delivery is protective against fecal incontinence.

In a population-based Oregon study from 2009 of 5491 primiparous women, nearly half (45.2%) experienced a new onset of at least 1 symptom of fecal incontinence in the survey period of 3-6 months after delivery. Among these women with fecal incontinence, 46% reported incontinence of flatus only, 22.8% were incontinent of liquid stool, and 18% were incontinent of solid stool. Vaginal delivery was associated with a greater risk of fecal incontinence compared with cesarean section (OR=1.45). However, vaginal delivery without laceration of instrument assistance did not increase the risk of fecal incontinence over cesarean. Interestingly, cesarean was not completely protective against fecal incontinence, given that 38% of women who delivered by cesarean without labor or pushing reported new-onset symptoms of fecal incontinence.[20]

Age may also play a role in the development of fecal incontinence. Haadem et al evaluated 49 continent, healthy women with a mean age of 51 years (range, 20-79 y). Anal manometry studies showed a decrease in maximum anal resting pressure and maximum squeeze pressure that was associated with age, irrespective of parity. These findings were more pronounced after menopause, suggesting a role for estrogen in the mechanism of this disorder.[21] Similar findings of decreased anal pressures on manometry results were reported by Bannister et al, who also found that aging resulted in lower rectal volumes needed to inhibit anal sphincter tone.[22]

When evaluating the effects of aging on anorectal sphincters, Laurberg and Swash found no significant changes in resting anal pressure. However, in women older than 50 years, they found that anal canal squeeze pressure was significantly lower compared with that found in younger women.[23] Johanson and Lafferty found that the prevalence rate of fecal incontinence in individuals younger than 30 years was 12.3%, compared with 19.4% in those older than 70 years. Approximately 56% of psychogeriatric patients and 32% of geriatric patients in long-term care facilities are anally incontinent on a regular basis. Increasing age is also associated with slowed pudendal nerve conduction, perineal descent at rest, and decreased anorectal sensory function.[24]


United States statistics

Exact data on the prevalence of fecal incontinence are difficult to obtain. This is partly because patients are reluctant to volunteer information about their condition to their health care provider. Fewer than 30% of women with fecal incontinence seek medical attention.[25] Conversely, heath care providers are not diligent about soliciting symptoms during maintenance visits. In addition, the definition (incontinence of solid stool, diarrhea or flatus) and frequency of fecal incontinence vary greatly in each study population.

The reported prevalence of fecal incontinence in the general population is approximately 2-3%.[26] Nelson et al found a prevalence rate of 2.2% among community-dwelling persons in the state of Wisconsin. In their study, 6959 individuals were surveyed and 30% of the individuals were older than 65 years and 63% were women.[26] In another study, Johanson and Lafferty observed an overall prevalence rate of 18.4% among individuals at the time of visit to their primary care physician or gastroenterologist.[27] Only one third of the group had ever discussed the problem with a physician. However, in this study, any involuntary loss of stool or soiling was considered to be incontinence; therefore, patients with soilage for reasons other than true incontinence may have been included. In the nursing home setting, the prevalence of fecal incontinence approaches 50% and can be a primary cause for admission.[28]

Rates of fecal incontinence in women have been investigated, both in the immediate and remote postpartum period. At 3-6 months after vaginal or cesarean delivery, as many as 13-25% of women report fecal incontinence.[29] In a population-based, postnatal survey of 6000 women (age 30-90 y), the prevalence of fecal incontinence (defined as at least monthly loss of liquid or solid stool) was 7.2%. Older age, major depression, urinary incontinence, medical comorbidity, and operative vaginal delivery were significantly associated with increased odds of fecal incontinence.[30, 31]

Nygaard et al evaluated the prevalence of anal incontinence 30 years postpartum in 3 groups of women. One group had disruption of the anal sphincter with vaginal delivery, the second group had an episiotomy without disruption of the anal sphincter, and the third group delivered by cesarean. They found no significant difference in the frequency of flatus incontinence among the 3 groups, at a rate of 30-40%. However, the number of women with bothersome flatus incontinence was higher in the sphincter disruption group (58%) compared with 30% of the episiotomy group and 15% of the cesarean group. The percentage of bothersome fecal incontinence was similar among the 3 groups at about 20%.[32]

The financial cost of fecal incontinence is significant. More than $400 million is spent each year for adult diapers that control urinary and fecal incontinence. It is the second leading cause of admission to long-term care facilities in the United States. In younger patients who desire treatment and correction, the costs are surprisingly high. In a 1996 study of 63 patients treated for fecal incontinence secondary to obstetric injuries, the average cost of treatment per patient was $17,166.[33] Approximately 21,000 women underwent inpatient surgery for fecal incontinence between 1998 and 2003 (approximately 3500 women per year). Total charges gradually increased from $34 million in 1998 to $57.5 million in 2003. The mean cost per surgical admission in 2003 was $16,847.[34]


Patients who present for evaluation of fecal incontinence usually have had to overcome extreme embarrassment over their condition prior to their office visit. Care should be given to the manner in which the topic is approached in order to promote an open and comfortable discussion.

The history should include a thorough account of the type and duration of fecal incontinence, frequency of incontinent episodes, type of stool lost, impact of the disorder on the patient's life, history of associated trauma or surgery, and associated factors such as protective undergarment use. True incontinence must be differentiated from pseudoincontinence, and patients may perceive drainage of mucous, pus, or blood from the perianal region as incontinence. Medications and dietary habits should be reviewed to determine if an easily remedied cause exists. A review of systems for systemic medical conditions that contribute to fecal incontinence is important and should be elicited.

An obstetric history should be taken carefully. Information about the number of vaginal deliveries and the presence of any risk factors for fecal incontinence pertaining to those deliveries should be obtained. As previously mentioned, prolonged second stage of labor, forceps delivery, significant tears, and episiotomy, among other causes, are associated with increased risk for anal sphincter disruption and pudendal nerve injury.

Several fecal incontinence surveys attempt to quantify and qualify the severity of fecal incontinence. Two examples are the Fecal Incontinence Quality of Life Scale produced by Rockwood et al and the Fecal Incontinence Questionnaire by Reilly et al.[35, 36] The forms are largely designed as outcome measures and are most useful in a research setting. The benefit to the practitioner not in a research environment is that a great deal of information about patient symptoms and the impact of those symptoms can be obtained in a short period of time.

Once a complete history has been obtained, a focused physical examination should be performed. Practitioners who routinely perform physical examinations on female patients should be familiar with normal perineal, vaginal, and anal sphincter anatomy. The examination should consist of a careful visual inspection of the perineal body and anus. The vagina should be inspected with the aid of a speculum. Examination of the anal region may reveal skin tags, hemorrhoids, anal fissures, scars, or chemical dermatitis. Rectal prolapse or fistulae may also be present.

After obstetric injuries, a discrete sphincter defect will almost always be anterior in location and be associated with loss of the perineal body or attenuation of the rectovaginal septum. Sometimes, however, the connective tissue of the perineal body is intact and only the sphincter has failed to heal. With an intact anal sphincter, the creases in the anal skin are arranged radially around the anal canal. When the sphincter is disrupted and the perineal skin is intact, there is a classic dovetail appearance to the anus, in which the radial distribution of the anal skin creases is absent anteriorly.[37]

After inspection, sensation of the perianal region should be assessed. This is performed by gently stroking the skin surrounding the anus and observing a reflex contraction of the external anal sphincter. The absence of this anocutaneous reflex suggests interruption of the spinal arc.

Digital examination should be performed to detect obvious anal pathology and provide an initial assessment of the anal resting tone. As the provider is beginning the rectal examination, resistance is met at the anal verge. If the examining finger meets little resistance and the anus feels patulous, significant sphincter dysfunction may be present. Deformity may be due to fistulotomy, fissurectomy, or hemorrhoidectomy.

A bimanual digital rectal examination is performed using 1 digit in the rectum and 1 digit in the vagina. This examination is mandatory to palpate the anal sphincters, determine perineal body thickness, evaluate anal sphincter and levator tone, and test for perineal body hypermobility. The external anal sphincter is palpable as a 1.5-2 cm, moderately firm, doughnut-shaped mass within the perineal body. If, upon palpation of the perineal body, the examining fingers are not separated by much tissue or the muscle mass is not palpated, damage to the anal sphincter is likely. Palpation from a midline position laterally helps to define where normal muscle exists. These findings can be subtle and may require some degree of experience to elicit.

The patient is asked to tighten the sphincter around the examining finger. A circumferential increase in the pressure should be felt. Patients who have a disrupted anal sphincter may have adequate innervation to the existing muscle bellies, and equal pressure may be felt on the examining finger. Evaluating the nature and location of increases in muscle tone is important. Pressure may be placed on the finger laterally from existing functional muscle, but no or very little pressure may be felt anteriorly. Disruption of the muscle sphincter allows the muscle bellies to retract laterally and posteriorly, similar to the ends of a rubber band that is cut after being placed on gentle stretch. Scar tissue may develop anteriorly between the 2 ends of muscle and, with contraction of the lateral muscle bellies, gives the impression of an intact anal sphincter. Palpation of the muscle can assist in this situation.

During rectal examination, the tip of the examining finger is gently flexed and traction is placed on the perineal body. The perineal body is normally supported by the perineal membrane (urogenital diaphragm), which originates from the ischiopubic rami bilaterally. The perineal membrane, along with other fascial supports, inhibits the caudad displacement of the perineal body. Birth trauma sufficient enough to result in detachment or tearing of this support also may have resulted in damage to the anal sphincters, and this history finding may prompt further questioning and evaluation.

Although the physical examination is a standard part of any evaluation for fecal incontinence, there is no exact correlation between digital assessment of sphincter tone and contractility. Physical examination results have also been noted to conflict with anal manometry.


Indications for surgical repair for fecal incontinence are largely based on physical findings and the degree to which the patient is symptomatic. Surgical repair of the anal sphincter is not without potential complications and discomfort. Importantly, the physician should counsel the patient about the risks of the procedure and the potential for recurrence.

Examination may reveal a sphincter defect that is discovered in association with other defects in pelvic support. The patient may be asymptomatic or only mildly symptomatic for anal incontinence. Whether surgical repair should be performed in this situation is controversial and must be a decision reached between the patient and the physician. The evidence presented suggesting that age alone is a risk factor for the development of anal incontinence must be factored into the equation if a known anatomic defect is present. Clearly, this type of patient is at increased risk of anal incontinence in the future. Additionally, surgical outcomes of sphincter repair have been shown to be poorer in older women.

Relevant Anatomy

A clear understanding of anatomy is mandatory for any surgical procedure, including surgical procedures of the perineum and anal sphincter. The perineal body is composed of a fusion of several structures. The Denonvilliers fascia is fused, which serves to suspend the perineal body from the sacrum via the endopelvic fascia. The bulbospongiosus and superficial transverse perineum muscles insert onto the perineal body. The bulbospongiosus originates from the pubic rami and inserts into the anterior aspect of the central tendon of the perineum. The superficial transverse perineum muscle originates on the ischial tuberosities and approaches the perineal body from a lateral position. The perineum is innervated by the labial nerve and inferior rectal nerve, which are branches of the pudendal nerve. The external anal sphincter is fused to the perineal body more posteriorly.

See the image below.

An attenuated perineal body results from trauma to An attenuated perineal body results from trauma to the perineum and damage to the structures that normally compose the perineal body.

The anatomy of the external anal sphincter continues to be debated. Some suggest that it composed of 3 parts that are fused, while others maintain that it is composed of 2 more distinct parts. A subcutaneous portion (just deeper than the perineal skin), a superficial portion, and a deep portion exist. Some combine the subcutaneous and superficial components into a single component. The separation of these components is indistinct in a surgical repair. The deep portion of the external anal sphincter is fused to the puborectalis muscle of the levator complex, and some even suggest that it is a continuation of it. It receives innervation from the pudendal nerve.

The internal anal sphincter is a continuation of the inner circular smooth muscle of the bowel wall. This musculature thickens over the last 2.5-4 cm of the rectum. The internal anal sphincter lies deep to the external anal sphincter, and an identifiable and surgically important plane exists between them. The internal anal sphincter begins at the level of the levator complex, and its distal extent is just proximal to the subcutaneous portion of the external sphincter. The internal anal sphincter is supplied by the autonomic nervous system. It exists in a state of continuous maximal contraction and provides a barrier to the involuntary loss of stool.

The levator complex is composed of the pubococcygeus, the iliococcygeus, and the coccygeus muscles. The most medial fibers of the pubococcygeus make up the puborectalis. These fibers loop around the posterior aspect of the rectum and create an anterior displacement of the rectum known as the anorectal angle. The pelvic surface of the levator complex is innervated by sacral efferents from S2 through S4. The inferior surface is supplied by the perineal and inferior rectal branches of the pudendal nerve. Consequently, pudendal block does not abolish voluntary contraction of the pelvic floor but completely abolishes external anal sphincter function.[5]


Contraindications to surgery for anal incontinence occur when the etiology for incontinence is uncertain; when a nonsurgical therapy, such as medication or diet alteration, would provide relief of symptoms; a woman is pregnant; or when surgical repair would place the patient at risk for complications from other medical conditions.



Laboratory Studies

Surgical repair for fecal incontinence is usually an elective procedure. Evaluation and treatment of anemia, infection, or other abnormalities indicated by findings from a complete blood cell count should be initiated prior to surgical repair.

A pregnancy test should be performed at or near the time of surgery on every female patient who may be pregnant.

Imaging Studies


The standard diagnostic imaging study for the anal sphincters is transanal or endoanal ultrasonography. Much of the early work on endoanal ultrasonography was performed at St. Marks Hospital in the United Kingdom by Law, Burnett, and Sultan et al. Multiple reports have qualified ultrasonography as the criterion standard in the diagnostic evaluation of the anal sphincters.[38, 39, 40, 41]

Ultrasonography allows the provider to perform a real-time, 360-degree evaluation of both the internal and external anal sphincters. Sensitivity and specificity of ultrasonography findings are 98-100% for the external anal sphincter and 95.5% for the internal anal sphincter. Burnett et al have shown that 90% of women presenting with anal incontinence and a history of a vaginal delivery had ultrasonographic evidence of damage to the external anal sphincter, internal anal sphincter, or both.[42]

Endoanal ultrasonography is performed in the left lateral or lithotomy position. The lithotomy position in females allows for the evaluation of other pelvic support defects.

Traditionally, 3 regions of the rectum are evaluated ultrasonographically. The areas of focus are distal, where only the external anal sphincter is observed (first image below); at the level where both the internal and external anal sphincters can be observed in the mid anal canal (second image below); and proximal at the level of the levators (pubococcygeus) (third image below).

Fecal incontinence. External anal sphincter showin Fecal incontinence. External anal sphincter showing normal narrowing anteriorly.
Fecal incontinence. Ultrasound of mid anal canal t Fecal incontinence. Ultrasound of mid anal canal taken with the patient in the low lithotomy position showing intact internal and external anal sphincters.
Fecal incontinence. Internal anal sphincter at the Fecal incontinence. Internal anal sphincter at the level of the levators. The levators are demonstrated by the U-shaped echogenic band posteriorly.

Normal thickness for the external anal sphincter is consistently described as 8.3 mm (95% confidence interval [CI], 7.6-9 mm). Mean internal anal sphincter thickness is described as 6.5 mm (95% CI, 5.8-7.2 mm), although reports vary from 1-3 mm to 10 mm based on modality. Inflammatory disorders can increase the thickness of these sphincters.[43]

The external anal sphincter appears as a hyperechoic ring that is circumferential. It is striated muscle, which is echogenic in nature. The internal anal sphincter is composed of smooth muscle and appears as a hypoechoic or sonolucent ring that is medial to the external anal sphincter. It can be viewed from the level of the levators to just inside the anal verge. The puborectalis muscle appears as a hyperechoic U-shaped muscle approximately 4 cm into the anal canal (third image above).

Tjandra and colleagues evaluated patients with ultrasonography of the anal sphincters and EMG. They found that ultrasonography helped identify both internal and external sphincter defects with 100% accuracy compared to EMG findings, with considerably less discomfort. They state that ultrasonography is preferable to EMG in mapping anal sphincter defects, especially for evaluation of the internal anal sphincter.[44, 45]

Diagnostic Procedures

Anal manometry

Anal manometry is used to evaluate both the resting and squeeze pressures of the rectum. It can also be used to evaluate the rectoanal inhibitory reflex, rectal capacity, and rectal compliance.

Normal values for manometry vary among institutions. Currently, no uniformly accepted standard exists for performing manometry, and what is considered typical pressure is also quite variable. Normal resting pressures are 40-70 mm Hg (55-95 cm water) from Wexner and 60 cm water (standard deviation [SD], 20 cm water) from St. Marks Physiology Unit in London. Normal squeeze pressures are 100-180 mm Hg (136-244 cm water) from Wexner and 100 cm water (SD, 30 cm water) from St. Marks Physiology Unit in London.

Mean resting and mean squeeze pressures are lower in women who have had a vaginal delivery, regardless of sphincter disruption. Low manometric pressures are not predictive of anal sphincter defects. Women with known sphincter defects tend to have lower mean resting pressures; however, mean squeeze pressures may be unaffected. Squeeze pressures are not significantly related to the presence of an external anal sphincter defect. Tetzschner et al found decreased pressures in continent and incontinent women after vaginal delivery. Both groups had low pressures compared to controls.[13]

The majority of studies that evaluate fecal incontinence in women use manometry as part of their assessment. The results of these studies are variable, and no patients have forgone surgical repair based on a finding of low manometric pressures.

Several studies are available that demonstrate that manometry results are not able to help predict surgical outcomes for anal sphincter repairs. Rasmussen and colleagues found that patients with incontinence persisting after surgical repair had lower anal maximum squeeze pressures compared to continent patients. In patients who are continent after surgery, no difference in anal pressures existed before and after surgery. Felt-Bersma et al found that manometry is not valuable when comparing an individual patient's pressure to the reference range because of the large variability in the reference range. They suggest that evaluation of an individual's pressure changes preoperatively and postoperatively makes more sense.[46] Several investigators have found improvement in manometry pressures postoperatively.

Pudendal nerve terminal motor latency

PNTML helps evaluate the length of time required for a fixed electrical stimulus to travel along the pudendal nerve from the ischial spine to the anal verge. Findings reflect the myelin function of the peripheral nerve, and the test allows for the evaluation of pelvic floor neuromuscular integrity. The pudendal nerve is stimulated at the ischial spine transanally. The latency period between stimulation of the nerve and evoked response of the muscle is measured. Any damage to the neuromuscular unit results in the prolongation of the latency. Normal latency has been described as 2 milliseconds (SD, 0.2 ms).

Several studies have found that prolongation of PNTML occurs after uncomplicated vaginal delivery. PNTML approaches the reference range at three months postpartum; however, Tetzschner et al found a significant and persistent prolongation of PNTML between incontinent and continent women three months postpartum.[13] In addition, the only predictor for the development of anal incontinence at 2-4 years postpartum in women who had rupture of the anal sphincter was abnormal PNTML.

Ryhammer and colleagues evaluated the long-term effects of vaginal delivery on anorectal function and found that PNTML increased with parity.[19] They found the greatest change in PNTML between the second and third deliveries; however, the range of values within groups was large. In a separate study, Tetzschner et al also found a wide range of values within incontinent and continent groups. The incontinent group values ranged from 1.35-3 milliseconds, and the continent group values ranged from 1.44-2.47 milliseconds.[14]

Many investigators have suggested that PNTML is the most significant predictor of functional outcome of a sphincteroplasty. Gilliland et al found that prolonged latency was correlated with outcome. In their series, only 16% of patients with pudendal latency greater than 2.2 milliseconds subjectively described their results as successful (excellent or good, based on telephone interview).[47] In a study of patients with unilaterally prolonged latencies (2 ms; SD, 0.2 ms), Sangwan et al found that only 1 of 7 patients who underwent sphincteroplasty had improvement in their incontinence score. Six patients had either (1) no improvement in incontinence score but less frequent incontinent episodes or (2) no improvement at all.[48]

The difficulty with interpreting findings in the literature is that no consensus has been reached for agreed-upon standard and nonstandard values. Also, variability exists in how PNTML correlates with other physiologic findings. All of the studies previously mentioned used the St. Marks electrode produced by Dantec.

Fynes et al considered latency to be prolonged if longer than 2.4 milliseconds in duration.[49] While evaluating the neurophysiologic changes in fecal incontinence, Osterberg et al used a latency period of longer than 2.5 milliseconds duration (based on a mean of 2 ms and 2 SD to reduce their false-positive rate) as an abnormal value.[50] In the same study, they found no differences in resting pressures, incremental pressures, or rectal sensibility in patients with unilaterally or bilaterally prolonged PNTML compared to patients with normal PNTML. They question the routine use of PNTML in the investigation of patients with fecal incontinence.

As with anal manometry, prolonged PNTML values have not been used as a reason to deny surgical repair. Chen et al found that in patients with prolonged latencies, surgical repair of sphincter defects resulted in improvement of incontinence scores.[51] Felt-Bersma et al, when evaluating anorectal function and ultrasonographic images after sphincter repair, found no relationship between preoperative neurologic function (based on EMG and PNTML findings) and clinical outcome. Acknowledging the fact that PNTML has contributed greatly to the understanding of anorectal pathophysiology, Felt-Bersma's group suggests that PNTML has no place in the routine preoperative assessment of patients. They go one step further by suggesting that withholding a sphincter repair from a patient with a prolonged PNTML may be unethical.[46]

Although prolonged PNTML indicates pudendal neuropathy, normal latency does not exclude nerve injury because only the fastest remaining conducting fibers are recorded. Also, overlap has been shown to exist in the innervation of the external sphincter.


EMG helps evaluate the electrical activity generated by muscle fibers during voluntary muscle contraction, rest, and Valsalva-type activities. The motor unit includes the anterior horn cell, its axon with axonal branches, the motor end plates, and the muscle fibers supplied. Information on the innervation and functional status of motor unit potentials is obtained. Results can be used to map normal muscle fibers. Several methods are available. Electrodes that are available for use are the surface anal plug, concentric needle electrodes, single-fiber electrodes, and monopolar wire electrodes. Injured or damaged muscle results in a lack of electrical activity or a polyphasic pattern. Incomplete damage may allow for reinnervation from adjacent undamaged nerves or regrowth of damaged axons. The result is that a greater number of muscle fibers are innervated by a single nerve, leading to an increase in the amplitude and a more prolonged motor unit action.

Abnormal findings after EMG evaluation are present in more than 90% of patients with fecal incontinence. In a study of 72 patients with fecal incontinence, Osterberg and colleagues found that fiber density observed with single-fiber electrodes correlated with clinical and manometric variables; PNTML did not.[50]

Although EMG is able to help quantitate denervation, findings do not alter clinical management. Similar to PNTML, Felt-Bersma et al suggest that EMG as a means to locate the sphincter defect has also lost its place in the preoperative assessment of anal sphincter defects.[46] EMG studies can be uncomfortable, especially the single-fiber technique, and they may result in incomplete studies due to patient refusal to continue.


Evacuation proctography (defecography) involves imaging the rectum with contrast material and observation of the process, rate, and completeness of rectal evacuation using fluoroscopic techniques. A variety of contrast materials have been used, including esophageal contrast barium and barium mixed with oatmeal or other viscous materials. Barium paste may also be inserted into the rectum. Images are obtained by lateral fluoroscopy and taken during attempted defecation. Defecography helps to assess qualitatively the function of the anorectum and the adequacy of rectal emptying; static images measure the length of the anal sphincter, the anorectal angle, and perineal descent. Diagnosis of rectoceles, enteroceles, and internal or occult rectal prolapse is possible. The true benefit in evaluation of patients with fecal incontinence is unclear, because most incontinent patients cannot tolerate the procedure.[52]



Medical Therapy

Conservative treatment options for fecal incontinence include bulking agents and biofeedback. After history and physical examination findings have helped exclude systemic disease and local anal pathology as the source of the patient's problem, the provider can initiate treatment.

The goal of medical therapy is to reduce stool frequency and improve stool consistency. A regular bowel regimen including daily laxatives should be established. If impacted, manual disimpaction and a daily tap-water enema may help prevent reaccumulation. The etiology of diarrhea should be diagnosed and treatment initiated. Mild incontinence can often be improved by initiating simple conservative measures. For patients with infrequent, low volume stools, bulking agents are helpful, as formed stools are easier to control than liquid stools. Methylcellulose (Citrucel) or psyllium (Metamucil, Fiberall, Hydrocil) can be taken daily. Additional firming of the stool can be obtained by restricting fluid with intake of the bulking agent. This may be helpful therapy in the patient who has incontinence of soft stool or liquid stool.

In patients with diarrhea due to noninfectious etiologies or with reduced rectal compliance due to radiation proctitis or inflammatory bowel disease, agents that slow the motility of the gut may be helpful. Loperamide hydrochloride increases gut transit time, allowing for increased absorption of water from the volume of stool. This results in a firmer, more easily controlled stool. The maximum daily dosage is 16 mg. The usual dose regimen is 2-4 mg twice or three times daily to control symptoms. An additional benefit of the opiate derivative loperamide (Imodium) is that it increases internal anal sphincter tone and may improve rectal compliance.[52] Diphenoxylate hydrochloride/atropine (Lomotil) has also been used; however, diphenoxylate hydrochloride can cause dependence and is a schedule V medication under the Controlled Substance Act.

Biofeedback is a safe, minimally invasive behavioral technique that uses auditory or visual feedback to reeducate the pelvic floor musculature. Although many different therapies have been used, several studies that have demonstrated a significant improvement in fecal incontinence by treatment with biofeedback.[53, 54] Other data, including a recent Cochrane review, does not provide clear evidence of therapeutic benefit.[55] The most commonly used techniques are rectal sensitivity training and anal sphincter strength training.

During rectal sensitivity training, a rectal balloon is gradually distended with air or water and the patient is asked to report first sensation of rectal filling. Once this threshold volume is determined, repeated reinflations of the balloon are performed with the objective being to teach the patient to feel the distension at progressively lower volumes. The rationale is that some patients are found to have high threshold volumes and if the patient detects stool arriving sooner, there is more possibility to either find a toilet or use an anal squeeze, or both. Conversely, the same technique has also be used to teach the patient to tolerate progressively larger volumes in those with urgency and a hypersensitive rectum.[55]

Biofeedback techniques have also been used to demonstrate anal sphincter pressures or activity to the patient, thereby enabling teaching of anal sphincter exercises and giving feedback on performance and progress. This can be achieved by using EMG skin electrodes, manometric pressures, intra-anal EMG, or anal ultrasonography. The patient is encouraged, by seeing or hearing the signal, to enhance squeeze strength and endurance. There is no consensus on an optimum exercise regimen for use at home between sessions, nor on the number of squeezes, frequency of exercises, or treatment duration. Different authors may describe very different programs.[55]

Biofeedback requires some rectal sensation and the ability to voluntarily contract the sphincter. It appears to be effective for neurogenic and idiopathic anal incontinence and for incontinence related to disruption of anal sphincters[54] , but a recent Cochrane review did not demonstrate conclusive therapeutic benefits.[55] Biofeedback's success seems to depend on improving rectal sensation, because manometric studies have not shown consistent improved sphincter pressure. Personal units are now available (without prescription) for home use after initial clinical instruction. They use a vaginally placed, air-filled sensor that provides information on force and duration of contractions. Results from biofeedback can diminish over time, but home devices provide the patient the opportunity for a prolonged course of therapy and intermittent reeducation in a private setting.

In a 2012 randomized, double-blind study, 3 months of treatment with transcutaneous electrical tibial nerve stimulation (TENS) was not significantly better than sham treatment in improving the number of incontinence and urgency episodes in 144 patients with fecal incontinence.[56]

A 2015 summary of the treatment of fecal incontinence from the 2013 National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Workshop has been published.[4]

Surgical Therapy

Once medical therapy has been maximized, minimally invasive and surgical therapies may be considered. In select patients, injectable materials may provide improvement in anal sphincter function.

Several reports have described injection of various materials to augment the function of the internal anal sphincter. Injectable silicone has been shown to be effective. In a study of 82 patients with severe fecal incontinence and a low anal resting pressure caused by internal anal sphincter dysfunction, patients were randomized to silicone injection into the intersphincteric space and internal anal sphincter with (Group A, n = 42) or without (Group B, n = 40) guidance by endoanal ultrasonography. Results show that fecal incontinence improved significantly in both groups with up to 12 months of follow-up but to a greater extent in the group in whom injection was ultrasonographically guided. No serious complications occurred.[57]

Results with carbon-coated microbeads have been less promising. In a pilot study of 33 patients, the submucosal injection of carbon-coated microbeads improved minor fecal incontinence by increasing anal pressure but did not significantly improve quality of life.[58] Experience with collagen injections has been more limited.

Several surgical procedures are performed for the treatment of anal incontinence. The type of procedure used is based on the patient history, physical examination findings, and results of diagnostic evaluation. The current philosophy in pelvic reconstructive surgery is restoration of normal anatomy. Usually, sphincter complex defects are secondary to obstetric injury, fistula repair, or lateral internal sphincterotomy. The standard procedure for anal incontinence due to anal sphincter disruption is the anterior overlapping sphincteroplasty. This procedure was first proposed by Parks et al in 1971 and modified by Slade et al in 1977.

Anterior sphincteroplasty consists of dissecting out the external anal sphincter, dividing the scar tissue in the midline, and then overlapping the scar so that muscle is approximated to muscle as closely as possible. The surgery can be performed with the patient in the prone jackknife position or the dorsal lithotomy position. Controversy exists as to the need to identify and plicate the internal anal sphincter. Its value in the continence mechanism has been discussed. Several postoperative studies have demonstrated improvement in resting and squeeze pressures, which suggest that either the internal sphincter was plicated intentionally or that the internal anal sphincter was also unintentionally plicated in the process of overlapping the scar mass. In an older study, Fang and colleagues suggest not separating the internal from the external anal sphincters; however, they do not discuss the reasoning for this.[59]

Several other studies have performed ultrasonography of the sphincters postoperatively, yet few specifically mention the condition of the internal sphincter. Briel et al compared 2 groups of patients who underwent surgical repair. One group had surgical repair of only the external sphincter between 1973 and 1989. The second group underwent surgical repair consisting of restoration of the rectovaginal septum, perineal body, and repair of the external and internal sphincters. These patients had surgical repair between 1989 and 1994. They found that the more complex repair conferred no advantage. The measured outcome focused on anal continence, which was restored or improved in 63% and 68%, respectively.[60]

For the internal anal sphincter repair, the surgical approach requires dissection along the intersphincteric plane and identification of the internal anal sphincter. The sphincter is then dissected free from the rectal mucosa and mobilized. The surgical technique varies depending on the bulk of scar tissue. The scar tissue is either divided or left intact as the sphincter is plicated.

Abou-Zeid performed isolated internal sphincter repair in 8 patients with ultrasonographically proven defects of the internal anal sphincter. All patients had anal incontinence of varying degrees. All had undergone a prior surgical procedure, such as hemorrhoidectomy or sphincterotomy, and had developed symptoms subsequently. All patients underwent preoperative ultrasonography, and 6 patients underwent postoperative endoanal ultrasonography to document internal sphincter anatomy. Continence scores improved in all patients, and 2 patients achieved complete continence.[8] The small numbers of this cases series are encouraging, but do not allow for definitive conclusions.

Most reconstructive surgeons perform colporrhaphy and perineorrhaphy as part of the overall repair. Numerous articles describe the physiologic and supportive role of the rectovaginal septum and perineal body. In a study of 143 women, Weber and colleagues found that defects in the posterior compartment often coexist with bowel symptoms; however, they were unable to show a direct correlation between stage of prolapse and severity of bowel symptoms. This study did not proceed with surgical repair and evaluate for symptoms postoperatively.[61] Wexner and Olivera support the philosophy of repair of all defects and suggest that failure to recognize and repair concomitant injuries of the anal sphincter mechanism is usually accompanied by continued anal incontinence.[62]

Pudendal nerve neuropathy is associated with a higher failure rate after sphincteroplasty. As discussed previously, Gilliland found that the only factor predictive of successful outcome of overlapping sphincteroplasty was bilaterally intact PNTML.[47] Sangwan et al found that patients with unilaterally intact pudendal nerves had a less favorable outcome than those with bilaterally normal latencies.[48] Chen et al evaluated patients who had normal (1), unilateral (7), or bilateral (4) prolongation of PNTML with preoperative and postoperative incontinence scores. They found significant improvement in incontinence scores among all groups. These results were sustained at 20 to 72 month follow-up.

Different studies have demonstrated that anterior overlapping sphincter repair can improve continence scores, resting pressures, and squeeze pressures regardless of pudendal nerve latencies; therefore, if sphincter disruption is present, repair should be offered. Counseling patients regarding surgical outcomes and risks is imperative.

Some researchers perform postanal repair in patients with anal incontinence from a neurogenic or idiopathic cause. This type of anal incontinence is often associated with denervation of the pelvic floor. Patients have a decreased ability to sense impending defecation and may initially become aware of the need for a bowel movement only after they have passed stool and notice the odor or sensation of fecal material around the anus. The internal and external anal sphincters are usually intact. Evaluation often demonstrates an increase in the anorectal angle, which has also been described as a "flattening" of the anorectal angle.

The original theory behind postanal repair was restoration of the anorectal angle and lengthening of the anal canal. The incision is made posterior to the anal canal and carried to the levator plate. The dissection is carried in the intersphincteric plane between the internal and external anal sphincters. Once identified, the ileococcygeus, puborectalis, and pubococcygeus muscles are plicated posterior to the rectum. The internal anal sphincter, external anal sphincter, or both can also be plicated during the procedure. This approach was first described by Sir Alan Parks, whose initial series had an 83% success rate. Many subsequent series by different investigators have been unable to match this original success rate.

In an attempt to better understand the mechanism of postanal repair, several investigators have performed preoperative and postoperative physiologic evaluation of patients. In a review of 30 patients who had undergone postanal repair, Setti Carraro et al found that 19 of the 30 patients had breaks in the internal anal sphincter (6), the external anal sphincter (4), or both (9) on endoanal ultrasonography.[63] Matsuoka and colleagues performed preoperative PNTML, EMG, and manometry and found no predictive factor in the outcome of postanal repair.[64] Scott et al performed manometry prior to postanal repair and were unable to show any correlation between manometry and successful outcome.

Laurberg et al performed extensive preoperative and postoperative evaluation including manometry, perineal descent, PNTML, and single-fiber EMG. They found minimal correlation with preoperative physiologic testing and successful surgical outcome. Of interest in this study is the finding of increased fiber density and prolonged PNTML in those patients who improved with surgery, suggesting damage as a result of postanal surgical repair.[65] Setti Carraro et al also found postoperative prolongation of PNTML, from 2.2 milliseconds to 2.85 milliseconds, in patients with a successful surgical outcome. They found that the only preoperative factor predicting successful outcome was intact PNTML.[63] Others also found that intact PNTML is associated with better surgical outcomes. These findings seem to indicate that the patient with idiopathic fecal incontinence and normal PNTML has a better chance of success compared to a patient with a neurogenic cause of incontinence.

Both Laurberg et al and Setti Carraro et al have found that postanal repair may cause trauma to the pudendal nerves that did not exist preoperatively. The potential consequences of this are not known. Jameson et al evaluated patients 2 years following postanal repair and found a decrease in the number of patients who had some benefit (83% to 53%), with only 28% who were markedly better at 2 years.[66]

The effect of postanal repair on the anorectal angle has also been evaluated. Bartolo et al have shown that restoration of the anorectal angle does not correlate with the success of postanal repair. Failure of postanal repair to restore the anorectal angle has been confirmed by other investigators. Womack et al had improvement in 70% of 16 patients and found no significant change in the anorectal angle when measured radiographically. They recommend that the procedure not be limited to those patients with widening of the anorectal angle.[67]

In a comparison of anterior sphincteroplasty and postanal repair, Orrom et al found no postoperative advantage for either procedure.[68] They found that patients who had undergone postanal repair had no change in anorectal angle, whereas those who had undergone anterior sphincteroplasty with anterior plication of the levators had a more obtuse anorectal angle. They concluded that restoration of the anorectal angle was not important, and their group has abandoned the postanal approach. Matsuoka et al concluded that although the success rate for postanal repair was low (~35%), it is a valid therapeutic approach because of low morbidity and the absence of mortality.[64] The alternative for these patients, in whom other modalities have failed, is to live with this very disabling condition, undergo diverting colostomy, or undergo muscle transfer.

Muscle-wrap techniques have been developed in which striated muscles from the gracilis or gluteus muscles are transposed and wrapped around the anal canal to increase tone. The use of the gluteus muscle has largely been replaced by gracilis transfer. These techniques create a neosphincter when there is not enough muscle present to repair. The procedure may be indicated in patients who have congenital absence of the anal sphincter or in those who have lost the anal sphincter as a result of disease, although some have undertaken this procedure after attempts at sphincter repair have failed.

To perform gracilis transfer, the muscle is mobilized while maintaining the proximal attachments and the neurovascular bundle. Care must be taken to prevent devitalization of the muscle. The muscle is tunneled and then wrapped around the anus. The distal end is sutured to the contralateral ischial tuberosity. Implantation of a nerve stimulator to the transposed muscle to aid in long-term contraction is described. Results have been good, with a success rate of 66% in a series of 139 patients.[69] However, the procedure has a high complication rate and is usually performed in the research setting.[70] In patients with severe anal incontinence or those in whom other procedures have failed, muscle transposition may offer an improvement in quality of life.[71]

The artificial bowel sphincter (Acticon Neosphincter) was designed to act as a patient's own anal sphincter in cases of severe fecal incontinence. This implantable device is produced by American Medical Systems and is available in the United States. The inflatable cuff is placed around the anus, and an inflation reservoir is placed in the space of Retzius. As the patient feels the need to have a bowel movement, a control pump is squeezed and forces water out of the cuff and into the reservoir. This allows the patient to have a bowel movement. The cuff slowly refills over several minutes. Its use in patients with fecal incontinence has been somewhat limited and recent long-term data has been disappointing. In a study of 25 patients who were followed for a median of 50 months, only 3 had good functional results with the system. Complications included infection, erosion, chronic pain, and obstructed defecation; the device removal rate was approximately 50% in this series.[72]

When fecal incontinence persists after medical and surgical therapies have failed, a colostomy may be considered. This converts a perineal stoma into a manageable abdominal stoma and removes the constant fear of public humiliation.

For more recent information, the reader is encouraged to review the 2015 summary of the treatment of fecal incontinence from the 2013 National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Workshop.[4]

Sacral nerve stimulation

Sacral nerve stimulation (SNS) is an established, FDA-approved, minimally invasive procedure for the treatment of fecal incontinence. The neurostimulator device is known as the InterStim® System, manufactured by Medtronics. SNS has been shown to benefit patients with fecal incontinence caused by minor anal sphincter defects or due to neurologic disorders resulting in rectal hyposensitivity, as well as those with intact sphincters who have failed conservative therapies, including low-residue diet, medications, and biofeedback physiotherapy.

Interstim® SNS is a two-step process that begins with the placement of a temporary external electrode into the sacral foramen to provide low-amplitude electrical stimulation of the S3 sacral nerve roots. It likely works at the pelvic afferent and/or central level rather than primarily peripheral motor neurostimulation.[73] SNS then reduces symptoms of fecal incontinence by improving the resting and squeeze pressures of the anal sphincter, rectal sensation, and colonic motility.[73, 74, 75, 76] Patients who respond to a 2-week trial subsequently undergo placement of a permanent electrode connected transcutaneously to an embedded neurostimulator. Both steps are performed in the operating room and require general or local anesthesia. The practice of employing a temporary, percutaneous SNS for a 2- to 3-week period allows practitioners to identify patients who are most likely to respond positively to a permanent implant.[77]  Percutaneous nerve evaluation (PNE) is an alternative, office-based procedure that allows the temporary neuromodulator lead to be placed under local anesthesia. If this initial, temporary phase is successful, the InterStim device is implanted in the operating room. Studies have shown superiority of the 2-staged trial as compared to PNE.[78, 79]

Several clinical studies have shown that SNS achieves greater rate of continence among patients with fecal incontinence compared to placebo or to optimal medical management. Good outcomes were first reported in 1995, and since then, numerous trials in the United States and in Europe have yielded evidence suggesting that SNS can improve continence in a proportion of patients with fecal incontinence.[77, 74]

Several authors have also shown good outcomes among patients with fecal incontinence caused by anterior resection and chemoradiation for rectal cancer,[80, 81] systemic sclerosis,[82, 83] and Crohn’s disease.[84] Studies suggest that after implantation, 41 to 75% of patients achieve complete fecal continence and 75 to 100% experience improvement in incontinence episodes.[85] Importantly, the therapeutic effect and improved quality of life for fecal incontinence is maintained for at least 5 years after SNS implantation, with 89 to 92% having > 50% improvement and 36 to 48% having complete continence.[86] However, 10 to 35% of patients with fecal incontinence do not respond after permanent implantation, either immediately or later, despite initially satisfactory test stimulation for reasons yet unknown. Several cohort studies have not been able to identify specific predictive factors during pre-operative evaluation for successful outcomes with the permanent SNS implant.[87, 88, 89]

A 120-patient prospective multicenter cohort study with an FDA-approved investigational protocol revealed that 83% and 85% of subjects with a permanent neurostimulator achieved therapeutic success (defined as ≥ 50% reductions of incontinent episodes per week) at 12 and 24 months, respectively, and 41% of patients achieved 100% continence at 24 months. Incontinent episodes decreased from a mean of 9.4 per week to 1.9 at 12 months and 2.9 at 24 months.[90]

In a randomized study by Tjandra et al, 120 patients with severe fecal incontinence were randomized to have SNS or best supportive therapy, which involved pelvic floor exercises, bulking agent, and dietary manipulation. During a 12-month follow-up period, full assessment included endo-anal ultrasound, ano-rectal physiology, a 2-week bowel diary, and fecal incontinence quality of life index. Compared to the control group, the SNS group experienced a significant decrease in mean incontinent episodes per week from 9.5 to 3.1 and in mean incontinent days per week from 3.3 to 1, as well as a significant improvement in fecal incontinence quality of life index. Forty-seven patients (39%) achieved perfect continence with SNS.[91]

In a multicenter study comparing SNS to placebo, Leroi et al randomized 27 patients with SNS in a double-blind crossover design to stimulation ON or OFF for 1-month periods for up to 8 months. Patients who received stimulation reported a significant reduction in frequency of fecal incontinence episodes and symptom severity, preference for ON phase, and an improvement in the ability to postpone defecation, the quality of life, anal sphincter function. Continence was fully restored in 5 of the 19 (26%) patients who actually received stimulation.[92]

A 5-year prospective study by Boyle et al was first to report data with intention to treat in an attempt to assess the true efficacy of SNS for fecal incontinence. Among 50 patients with fecal incontinence, 13 patients (26%) did not respond during the first, temporary implantation stage or were dissatisfied with the result. Ten additional patients (20%) did not achieve a > 50% reduction in symptoms following permanent implantation. Nevertheless, 27 patients (54%) experienced > 50% reduction in symptoms, including 13 (26%) who achieved apparent continence. SNS resulted in a significant reduction of fecal incontinence episodes per night from 14 to 2, and in an improvement in the ability to defer defecation. Therefore, while good outcomes were comparable to aforementioned studies, this study revealed that symptoms of fecal incontinence continue in the majority (74%) of patients when analyzed by intention to treat.[93]

Among the abovementioned studies, analyzed in a Cochrane Review, adverse events occurred among 12-25% and included pain, hematoma, and/or seroma at the implant site, lead migration, paresthesia, change in the sensation of stimulation, and infection. There were no septic events.[77] Faucheron et al also investigated reasons for neurostimulator implant revision, thus elucidating SNS-associated morbidity. Among 87 patients who received the transcutaneous neurostimulator implant, 36 (41%) required surgical revision of the device due to site infection, pain or adverse stimulation, electrode displacement, electrode breakage, total or partial loss of clinical efficacy, device dysfunction, and battery depletion.[94]

Dueland-Jakobsen et al explored alternative neurostimulator settings in a randomized double-blind cross-over study with 15 patients who have sustained loss of efficacy. By altering pulse frequencies and pulse widths, the authors found a preferred setting for each individual that ultimately achieved significant improvement in the Fecal Incontinence Quality of Life Scale. They noted a trend toward highest patient satisfaction and improved treatment outcome with high-frequency stimulation, which was preferred by 8 of the patients, which was sustained at 3 month follow-up.[95]

Sacral transcutaneous electrical nerve stimulation (S-TENS) is a non-invasive, cheap, simple and promising alternative to the transcutaneous SNS with Interstim® System. A small prospective study by Chew et al showed optimistic results among 17 patients with idiopathic fecal incontinence who used TENS. Treatment included 2 hours of daily S3 nerve stimulation for 3 months. Sixty-nine percent of the subjects showed improvement in the fecal incontinence severity index, with a decrease in the number of gas and/or stool incontinence episodes per week. All patients reported subjective impression of improvement, and 88% scored > 6/10 for bowel control. The mean rectal volume of first sensation and first urge fell, and the maximum tolerable volume rose. The satisfaction score was > 6/10 in all patients.[96] However, more studies are necessary to evaluate its efficacy.

The InterStim® System neurostimulator manufactured The InterStim® System neurostimulator manufactured by Medtronic shown in place. Courtesy of Medtronic Inc.
The InterStim® System neurostimulator manufactured The InterStim® System neurostimulator manufactured by Medtronic. Courtesy of Medtronic Inc.

Injectable anal bulking agent

Solesta® (Oceana Therapeutics, Inc.) is a bulking agent consisting of a dextranomer stabilized in hyaluronic acid that was approved by the FDA in 2011 for the treatment of passive fecal incontinence in patients who have failed other conservative therapies. In Europe and Canada it is known as NASHA Dx or Zuidex®. Dextranomer/hyaluronic acid is a biocompatible bulking agent injected under the anal submucosa. Four 1 mL injections are administered into the deep submucosa in the proximal part of the high pressure zone of the anal canal, approximately 5 mm above the dentate line. If the response is inadequate after a minimum of 4 weeks, treatment can be repeated a second time. By expanding the anal tissue, the proximal anal canal narrows, thus preventing fecal leakage. Dextranomer/hyaluronic acid injection can be administered in an outpatient setting without anesthesia.[97]

Recent studies suggest that just over one half of patients achieve >= 50% reduction in number of fecal incontinence episodes at 6 months, and this effect is sustained for up to 3 years. Furthermore, patients report significant improvement from baseline in most of the domains of the Fecal Incontinence Quality of Life scale with dextranomer/hyaluronic acid injections.[98] A 3-year cost-effectiveness model comparing dextranomer/hyaluronic acid injection and sacral nerve stimulation following failed conservative management revealed that dextranomer/hyaluronic acid injection is cost-effective and results in more efficient use of resources for the treatment of fecal incontinence than sacral nerve stimulation.[99]

One of the first, and most cited, studies to support the use of dextranomer/hyaluronic acid for the treatment of fecal incontinence was a randomized, double-blind, sham-controlled trial with 206 patients by Graf et al. Of the 136 patients receiving dextranomer/hyaluronic acid injections, 71 (52%) reported > 50% reduction in the number of incontinence episodes at 6-month follow-up, compared to 22 (32%) of the 70 patients receiving sham injections. There was no blinding from 6 to 12 months, and all remaining participants received dextranomer/hyaluronic acid injection, with 57% achieving > 50% reduction in symptoms. The number of incontinence episodes decreased from 15 at baseline to 6.2 at 12 months, and the mean number of incontinence-free days increased from 4.4 at baseline to 7.9 at 12 months. The mean Fecal Incontinence Quality of Life scores for all four items improved significantly between baseline and month 12.[100]

Interestingly, the percentage of subjects achieving at least 50% reduction in fecal incontinence symptoms at 6 and 36 months remained stable (52%), while the percentage of subjects achieving 100% reduction rose from 6% at 6 months to 13% at 36 months. Of note, the majority of patients available for analysis at 36 months had received a second injection 1 month after the initial treatment. Therefore, it appears that the achieved effect of dextranomer/hyaluronic acid injection is sustained for up to 3 years.[101]

An open-label study by La Torre and de la Portilla produced comparable results for efficacy of dextranomer/hyaluronic acid at 24 months. Of the 83 individuals who completed 24-month follow-up, 63% experienced a >= 50% reduction in the total number of episodes of fecal incontinence, with the median number of episodes declining by 69%. The number of incontinence-free days increased from 15 at baseline to 22 at 24 months.[102]

Among these and other studies, most treatment-related adverse events were mild to moderate, self-limited and resolved within 1 month of injection. These include proctalgia, rectal hemorrhage, constipation, injection site bleeding, rectal discharge, anal pruritus, proctitis, painful defecation, and fever. Three serious adverse events were reported, representing a total of 1.3% of all adverse events related to dextranomer/hyaluronic acid injection: Escherichia coli bacteremia and 2 minor abscesses.[103]

Vaginal bowel control device

A new device received FDA approval in 2015.  The Eclipse System offers a conservative, safe, and effective option for the management of fecal incontinence with no reported serious adverse outcomes. It is a vaginal insert that is intended to treat fecal incontinence in women 18 to 75 years old who experience at least 4 incontinence episodes in a 2-week period. The device includes an inflatable balloon, which is placed in the vagina. Upon inflation, the balloon exerts pressure through the vaginal wall onto the rectal area, thereby reducing the number of fecal incontinence episodes. The device is initially fitted and inflated by a clinician (with the use of a pump), and after proper fitting, the patient can inflate and deflate the device at home as needed. The device should be removed periodically for cleaning. 

A study looked at 61 of 110 (56%) participants from 6 clinical sites that were successfully fit and entered treatment. At 1 month, intention-to-treat success was 79% (48/61); per protocol success, 86% (48/56) considered bowel symptoms “very much better” or “much better.” There was significant improvement in all Fecal Incontinence Quality of Life and Modified Manchester subscales. Success rate at 3 months was 86%.[104] Similarly, another multi-center, open-label, prospective trial showed that 62% of the 91 intent-to-treat subjects achieved a > 50% reduction in incontinence frequency. Mean fecal incontinence severity scores improved by 32%, and 78% of completers were “very” or “extremely” satisfied with the device.[105]

Eclipse System (vaginal insert) from Pelvalon. Cou Eclipse System (vaginal insert) from Pelvalon. Courtesy of Pelvalon Inc.

Preoperative Details

Once the decision has been made to proceed with surgical repair, mechanical bowel cleansing is performed. This can be performed with a variety of agents. Full mechanical bowel preparation with large volumes of solution is probably unnecessary, although it has some benefits for postoperative management. Smaller volumes of solution are better tolerated by patients, with some surgeons using only Fleet enemas prior to repair. The value of a more thorough preparation, such as GoLYTELY, is in the postoperative period when most surgeons are concerned about delaying mechanical stretch to the newly repaired sphincter.

Antimicrobial prophylaxis for colorectal operations can consist of an oral antimicrobial bowel preparation, preoperative parenteral antimicrobials, or a combination of both. Oral prophylaxis consists of neomycin plus erythromycin, or neomycin plus metronidazole, started no more than 18-24 hours before surgery along with a mechanical bowel preparation. Current recommendations for parenteral antibiotic prophylaxis include a third generation cephalosporin with metronidazole. A single preoperative parenteral dose of antibiotic is sufficient and should be administered within 1 hour prior to incision.[106]

Intraoperative Details

To minimize the risk of surgical site infections, care should be taken to keep the operative field clean from contamination. Preoperative antibiotics should be administered prior to beginning the procedure.

During overlapping sphincteroplasty, maintaining the scar tissue on the muscle belly and using this to decrease the likelihood of suture pull-through has been one of the most important factors improving surgical outcomes. The adequate mobilization of the existing scar tissue and separation of the scar in the midline to allow for overlap is also important. The judicious use of cautery to prevent devascularization is also wise.

Postoperative Details

Postoperative management after surgical treatment for fecal incontinence varies significantly according to provider. Agreement exists about the use of postoperative stool softeners and dietary fiber, and several products are available (see Medical therapy). Mineral oil may be used for a short period of time postoperatively, but long-term use is associated with absorption problems.

Specific dietary restrictions are commonly used postoperatively. Many surgeons delay feeding and keep patients on clear liquid diets or soft foods for several days. Others allow a more liberal diet and use stool softeners and mineral oil to decrease stool firmness. If a mechanical bowel preparation is used, the time to first bowel movement will be delayed, especially if the patient is kept NPO for the immediate postoperative period. Many encourage patients to use sitz baths as a means of decreasing perianal edema and to assist with cleanliness. However, some believe that osmotic gradients actually allow for more swelling and tissue maceration, and they promote the use of peribottles to wash the perineum and rectum. In these cases, showering is acceptable as the perineum is not submerged in water.

The use of postoperative antibiotics is controversial. Due to the location and nature of the tissues, infection of the surgical site is a risk; however, the blood supply to this area is rich and, unless compromised with extensive electrocautery, the need for postoperative antibiotics is questioned.


Owing to the nature and location of the surgical repair, pain control is an important issue. Patients should have adequate access to their surgeon for additional medications as necessary. Because the rate of separation of superficial tissues is as high as 25% and because of the potential for infection, patient concerns and symptoms must be taken seriously. Evaluation by trained providers, if only to reassure the patient, allows the surgeon to detect serious complications early.

Postoperative evaluation should be scheduled for 4-6 weeks after the procedure. At this time, most postoperative swelling and tissue distortion is usually resolved. A history of the patient's bowel habits should be taken and problems addressed. If modification of the stool softener regimen is required, it can be done at this time. Mineral oil regimens should be stopped if the patient has continued these medications postoperatively. Additional follow up can be scheduled depending on the individual needs of the patient and practice of the surgeon.


The complication rate from surgical repair of the anal sphincter varies based on the series reviewed. The patient population requiring surgical repair may have significant comorbidities that predispose them to postoperative complications. A review of several series for both postanal repair and anterior sphincteroplasty reveals that most patients are in the fifth through seventh decade of life. Underlying medical conditions, such as obesity, diabetes, or heart disease, can increase the postoperative risk of myocardial infarction or deep vein thrombosis.

The most common complication is superficial separation of skin and subcutaneous tissues, and the frequency rate is as high as 25% in some series. Care must be taken to place as little tension as possible on these tissues. Conversion of the curvilinear incision into a Y-shaped incision at the time of closure can place a great deal of tension on the perineal skin and predispose it to separation. Plication of the bulbospongiosus and superficial transverse perineum muscles can remove tension from the overlying skin and serve to restore anatomy. Devascularization of the vaginal or rectal mucosa can result in necrosis of these tissues.

Risk of infection is 3-5%. Opening the wound to allow for drainage and treatment with antibiotics may allow the physician to salvage the surgical repair. Fistula formation occurs in fewer than 1% of the series reviewed, but it is more common in those cases in which infection develops.

Bleeding and hematoma formation are also possible complications. Bleeding can usually be controlled with pressure achieved with packing. Hematoma formation into the perirectal space can go unnoticed and result in the sequestration of large amounts of blood. Treatment requires evacuation of the hematoma and surgical hemostasis.

Other complications include anal stricture, fecal impaction, and pain. Pain may be associated with bowel movements and intercourse, leading to a great deal of frustration for both the provider and patient. Many of these problems improved with time. Most complications that arise do not affect the sphincter repair. Although patients may be distressed, they should be reassured that the risk of failure of the procedure is not increased.

Outcome and Prognosis

Initial outcome after sphincteroplasty is 64-90% with short-term follow-up.[68, 107] The success rate starts to fall after the first few years and continues to fall with longer follow-up. Two long-term studies demonstrate that only half of the patients have satisfactory continence at 69-80 months postoperatively.[108, 109]

A secondary analysis of data from a multicenter study evaluating adaptive behaviors among women with fecal incontinence in the Pelvic Floor Disorders Network revealed significant improvements in symptom severity and condition-specific quality of life at 3 and 12 months following treatment.[110] Treatment was nonsurgical in 78% of the women; 22% underwent anal sphincter repair.

Future and Controversies

The disappointing results of the artificial anal sphincter have led investigators to explore other therapies for severe fecal incontinence.

Sacral spinal nerve stimulation has been used successfully for the treatment of female patients with urinary incontinence associated with overactive bladder and nonobstructive urinary retention. The Medtronic InterStim device (Medtronic, Inc, Minneapolis, MN) is FDA approved in the United States for this indication. In Europe, sacral spinal nerve stimulation has given patients with fecal incontinence an option that is minimally invasive and appears to offer significant improvement. A recent randomized control trial of 120 patients with severe fecal incontinence demonstrated a significant improvement of incontinence symptoms, decreasing from 9.5 to 3.1 mean incontinence episodes per week when compared with optimal medical therapy that comprised bulking agents, pelvic floor exercises, and dietary management. These results are even more encouraging given that half of the patients had evidence of a sphincter defect and two thirds had evidence of pudendal neuropathy.

Interestingly, neither the maximal resting pressure, squeeze anal pressures, or pudendal nerve terminal latency had any association with improvement. In addition to a sustained functional improvement, quality of life was significantly enhanced as measured by fecal incontinence quality of life (FIQL) scores. Medtronic InterStim has been the modality of choice of investigators in this and other studies treating fecal incontinence.[111] Long-term data of greater than 5 years since implantation are reassuring. In 52 patients who had undergone implantation, at least 50% improvement occurred in three quarters of the patients.[112] Successful results have been demonstrated even with significant anal sphincter disruption.[113]