The surgical approach to treating fecal incontinence is complex. After optimal medical management has failed, surgery remains the best option for restoring function. Patient factors, such as prior surgery, anatomic derangements, and degree of incontinence, help inform the astute surgeon regarding the most appropriate option. Many varied approaches to surgical management are available, ranging from more conservative approaches, such as anal canal bulking agents and neuromodulation, to more aggressive approaches, including sphincter repair, anal cerclage techniques, and muscle transposition. Efficacy and morbidity of these approaches also range widely, and this article presents the data and operative considerations for these approaches.
Key points
- •
Fecal incontinence is a very prevalent but likely underreported disorder.
- •
Appropriate history and physical examination, including obstetric history, is important to determine the cause of incontinence.
- •
Medical management is the appropriate first-line treatment, but when this fails surgery is the next approach.
- •
The most common current surgical approaches include sphincteroplasty, anal canal bulking agents, and sacral nerve stimulation.
- •
More complex approaches include stimulated graciloplasty, artificial bowel sphincter, tibial nerve stimulation, magnetic anal sphincter, radiofrequency ablation, and antegrade continence enema; however, these approaches tend to have a higher morbidity.
Introduction
Fecal incontinence is an embarrassing and socially paralyzing condition that affects up to 18% of the population and as many as 50% of nursing home residents. The maintenance of fecal continence is an extremely intricate process that involves the coordinated efforts and interaction of several different neuronal pathways and the musculature of the pelvic floor and anorectum. Several additional factors, including systemic disease, bowel motility, stool consistency, sphincter integrity, and patient emotion, play important roles in regulating continence. This article outlines the numerous causes of fecal incontinence, details the evaluation of patients complaining of altered fecal continence, briefly describes nonsurgical modalities for treatment, and focuses on the surgical methods of treating this debilitating condition that occurs much more frequently than most realize.
Introduction
Fecal incontinence is an embarrassing and socially paralyzing condition that affects up to 18% of the population and as many as 50% of nursing home residents. The maintenance of fecal continence is an extremely intricate process that involves the coordinated efforts and interaction of several different neuronal pathways and the musculature of the pelvic floor and anorectum. Several additional factors, including systemic disease, bowel motility, stool consistency, sphincter integrity, and patient emotion, play important roles in regulating continence. This article outlines the numerous causes of fecal incontinence, details the evaluation of patients complaining of altered fecal continence, briefly describes nonsurgical modalities for treatment, and focuses on the surgical methods of treating this debilitating condition that occurs much more frequently than most realize.
Etiology
The most common causes of fecal incontinence seen by colorectal surgeons is anal sphincter injury, with obstetric injury being the most frequently cited cause in women. Overt anal sphincter disruption accompanies as many as 10% of vaginal deliveries, whereas sonographic evidence of occult sphincter damage can be seen in as many as 30% of first vaginal deliveries. The obstetric-related disturbance is affected by not just the direct sphincter injury but also other risk factors, such as forceps delivery, occipitoposterior position, and prolonged labor, which can lead to pelvic denervation injury. In fact, 60% of patients with an obstetric injury will also have evidence of pudendal nerve damage. Although a small percentage of women will experience altered continence because of a direct sphincter injury in the immediate postpartum period, most will not present with symptoms until later in life, when the supporting pelvic floor musculature begins to weaken and compensatory mechanisms are lost.
Traumatic injury to the anal sphincter, both accidental and nonaccidental, can also result in altered fecal continence. Perineal impalement injuries, although rare, can be devastating, because they are often associated with not just a direct sphincter injury but also injuries to the rectum and pelvic innervation; they frequently require a temporary diverting stoma, and long-term continence after stoma closure in these patients is often difficult to achieve. Traumatic injuries to the anal sphincter may also occur in conjunction with sexual abuse, voluntary anoreceptive intercourse, or insertion of foreign bodies.
Iatrogenic injury to the anal sphincter may occur in several settings, the most common of which is after surgical intervention for various anorectal disorders. A recent systematic review showed an overall continence disturbance rate of 14% after lateral internal anal sphincterotomy for anal fissure; incontinence to flatus was seen in 9%, soilage/seepage in 6%, incontinence to liquid stool in 0.67%, and incontinence to solid stool in 0.83%. Incontinence after hemorrhoidectomy should occur infrequently, although patients with preexisting altered continence are more likely to show further deterioration in their continence postoperatively, a fact that should be taken into consideration when contemplating excisional hemorrhoidectomy. Fecal incontinence after anal fistulotomy has long been the bane of the procedure; minor alterations in continence can be seen in as many as 18% to 52% of patients, and major changes can occur in 35% to 45% of patients. Long-term fecal soiling can be seen in up to 43% of patients after endoanal advancement flaps for fistula in ano. In a database review of patients with alterations in fecal continence after anorectal surgery, Lindsey and colleagues showed that a common factor was the presence of an internal sphincter injury, resulting in reversal of the normal resting pressure gradient of the anal canal. Another iatrogenic cause of fecal incontinence is pelvic radiotherapy, which can directly affect anal sphincter function and also decrease rectal compliance.
Anorectal disease not managed surgically may also be an independent risk factor for developing altered fecal continence. Rectal prolapse may stent the anus open or chronically dilate the sphincters. Prolapsing internal hemorrhoids may also stent the anal canal open, allowing for fecal seepage. Patients with large external hemorrhoids may have difficulty with perianal hygiene after defecation, leading to chronic soiling, which is often perceived as incontinence by the patient. Chronic rectal inflammation or decreased compliance to drug therapy for inflammatory bowel disease or infectious proctitis may also lead to alterations in continence. Distal rectal cancer can cause proximal impaction and overflow incontinence, and direct sphincter or neural invasion.
Although obstetric trauma is a common cause of pelvic denervation injury, other neurologic causes of altered fecal continence exist. Patients who have multiple sclerosis or other central nervous system–related disorders may experience altered fecal continence as their disease progresses. Patients with longstanding diabetes are prone to autonomic neuropathies, including pudendal neuropathy. Patients with congenital malformations such as spina bifida, meningomyelocele, and imperforate anus often have derangements in pelvic floor and anal sphincter function.
Lastly, simple factors such as the senescent process of aging may affect fecal continence as muscles lose bulk and strength over time. Senile dementia and psychiatric illness may also play a role in altered fecal continence because of patient’s lack of awareness of the call to defecate.
Evaluation
History
Obtaining a detailed history may be difficult, because patients are often embarrassed and hesitant to offer details of their incontinence unless they are approached with a direct line of questioning. It is important to create as comfortable an environment as possible during the history and physical examination. A thorough obstetric history in women in essential, including number of pregnancies and deliveries, means of delivery (eg, vaginal, cesarean, forceps-assisted, vacuum-assisted), obstetric injury from perineal laceration or episiotomy, and episodes of prolonged labor. Timing of onset and progression of symptoms can provide meaningful information regarding the cause and potential efficacy of therapy. Neurologic causes should be considered and investigated in appropriate patients.
Questioning should clarify whether the patient exhibits incontinence to solid stool, liquid stool, or flatus, or a combination thereof. A determination should also be made regarding whether the patient has active or passive incontinence. Active incontinence, or urge incontinence, describes the loss of stool despite efforts by the patient to control it. This condition typically suggests an intact sensory mechanism with the loss of function or integrity of the external anal sphincter. Passive incontinence is the loss of stool without the patient sensing it, and typically suggests a neurologic origin of the patient’s complaints.
Quantifying the degree of incontinence is important, especially in establishing a baseline score that can then be compared with posttreatment scores and used to measure the effect imparted by a particular management scheme. Numerous scoring systems are available for use. The Wexner (Cleveland Clinic Florida) Fecal Incontinence Score (CCF-FIS) is an independently validated tool that is easy to use for the patient and the physician, and incorporates a quality of life component. The Fecal Incontinence Quality of Life questionnaire is another useful tool for quantifying a patient’s degree of incontinence and its impact on quality of life.
Physical Examination
The physical examination should start with the patient in a comfortable position, typically the Sims (left lateral decubitus) position. The perineum should be inspected, looking particularly for scars from prior surgeries or trauma (obstetric or otherwise). Other findings may include fistulae in ano, skin excoriation from chronic soiling, rectal prolapse, and prolapsing hemorrhoids. The bulk of the perineal body should be assessed. At rest, the anal verge should be closed and not patulous. The patient should be asked to perform a Valsalva maneuver to evaluate for prolapse. Perianal sensation to pinprick should be evaluated, as should the anocutaneous “wink” reflex to light tough, which can be thought of as a “poor man’s assessment” of neurologic function.
Digital rectal examination should be performed to subjectively determine the patient’s resting sphincter tone and voluntary squeeze pressures. Palpation for rectal masses or fecal impaction should be performed. The experienced examiner may appreciate a large anterior sphincter defect or motion abnormality if present. Visualization of the anorectal mucosa should also be performed, either through anoscopy or rigid proctosigmoidoscopy to evaluate for mucosal abnormalities, which might suggest inflammatory bowel disease, infectious proctitis, or a neoplastic process.
Diagnostic Studies
Endoanal ultrasound
The primary means of imaging the anatomy of the anal canal, including the anal sphincters, has become endoanal ultrasonography. The test is easily performed in an office-based setting, without sedation, and is very well tolerated by most patients. Ultrasonographically, the external anal sphincter appears hyperechoic, the internal anal sphincter appears hypoechoic ( Fig. 1 ), and scar tissue shows mixed echogenicity. Anatomic sphincter defects can be easily identified and the degree of the defect can be measured ( Fig. 2 ). The thickness of the perineal body can also be measured.
Anorectal manometry
Anorectal manometry can provide valuable information regarding the functional status of the anal sphincters and the distal rectum in the evaluation of the patient with fecal incontinence. This technique also is a well-tolerated, office-based procedure that is easy to perform, but requires dedicated equipment and trained personnel. Several commercially available systems are available; some utilize multi-channel water-perfused catheters, whereas others use direct microtransducers to record resistance.
Several measurements can be obtained from manometry studies. The mean resting pressure of the anal sphincters is normally 40 to 70 mm Hg, most of which is generated by the internal anal sphincter, which is in a continuous state of maximal contraction. Low resting pressures may be seen in patients with altered fecal continence from disturbances in the function of the internal anal sphincter. The maximal squeeze pressure is typically 2 to 3 times the baseline resting pressure; it is generated mostly by voluntary contraction of the external anal sphincter. Defects of the external anal sphincter muscle from trauma (eg, surgical, obstetric) may result in decreased maximal squeeze pressures. The high-pressure zone is the length of the internal anal sphincter through which pressures are greater than half of the maximal resting pressure; it measures approximately 2.0 to 3.0 cm in women and 2.5 to 3.5 cm in men.
The rectoanal inhibitory reflex (RAIR) is represented by initial contraction of the external anal sphincter followed by a pronounced internal anal sphincter relaxation in response to rectal distension. This reflex enables the sensory mucosa of the anal canal to “sample” the contents of the distal rectum and help distinguish between gas, liquid, and solid stool. The RAIR is altered in patients with Hirschsprung disease, scleroderma, and Chagas disease. Rectal sensation can be measured by incrementally instilling a balloon placed in the rectum with small aliquots of air; normal sensation usually occurs with 40 mL of insufflation. Decreased rectal sensation can lead to fecal impaction with subsequent overflow incontinence. Lastly, rectal compliance can be measured as the change in pressure associated with a change in volume. A poorly compliant rectum is not able to readily accommodate the stool it receives and may contribute to fecal incontinence. Inflammatory conditions or external beam radiation may lead to decreased rectal compliance.
Pudendal nerve terminal motor latency
Determination of pudendal nerve terminal motor latency (PNTML) is an important component in the evaluation of patients with fecal incontinence, because pudendal neuropathy can often lead to alterations in fecal control. A disposable electrode is attached to the examiner’s finger, which is directed toward the Alcock canal within the rectum. An impulse is delivered to the pudendal nerve, and a more distal sensory electrode determines how long it takes for the external anal sphincter to contract. Normal values are less than 2.0 ± 0.2 ms. Prolonged PNTML may be seen as a result of obstetric trauma to the pudendal nerves or with systemic diseases, such as diabetes.
Electromyography
Electromyography uses needles placed into the sphincter muscles to record electrical activity; it can be used to map the external anal sphincter and identify neuromuscular integrity. Alternatively, an intra-anal sponge or surface electrode can be used. Because of the discomfort and expertise required, single-fiber electromyography never gained universal acceptance.
Defecography
Defecography, or a defecating proctography, involves instillation of contrast material into the rectum, after which the patient evacuates the material under direct fluoroscopic visualization. Although the test is often uncomfortable and embarrassing for patients, it provides extremely useful dynamic images of the rectum and pelvic floor during the act of defecation. In the evaluation of the patient with fecal incontinence, defecography may show incomplete evacuation of the rectum, leading to overflow incontinence. It may also show lack of rectal distention, indicating decreased rectal compliance. Dynamic magnetic resonance imaging defecography is a more recently described alternative to fluoroscopic defecography.
Colonoscopy
Endoscopic evaluation of the colonic mucosa should be considered essential in patients with fecal incontinence. A stricture or mass may cause partial obstruction, allowing for overflow incontinence. Various inflammatory conditions of the colon and rectum may also lead to looser stools or loss of rectal compliance, predisposing to incontinence. A large villous lesion may cause excessive mucous production, often leading to soilage.
Medical management
The most important initial step in managing fecal incontinence is determining the underlying cause. Certain causes, such as chronic diarrhea or constipation, neurologic conditions, and systemic illnesses, may be best managed by medical means. Unfortunately, published data are lacking regarding the true efficacy of medical management, as evidenced by a Cochrane review evaluating this topic. Initial management should center on a conservative approach that aims to change the consistency and frequency of stools by avoiding laxatives, starting or increasing the use of stool bulking agents, changing dietary habits, and starting antimotility drugs. The goal should be bulkier, more solid stools that are evacuated completely and the establishment of a regular, predictable bowel pattern.
Fiber supplements serve to add bulk to stool and absorb fluid, creating a more solid stool. Both natural and synthetic products are available, and no one product has been shown to have a benefit over the others. Commercially available products include Metamucil, Benefiber, Citrucel, Konsyl, and Fibercon. Patients should be educated regarding a high-fiber diet, aiming for 25 to 30 g of fiber per day.
Up to 50% of patients with chronic diarrhea also have fecal incontinence. Therefore, constipating agents may play a key role in managing these patients’ symptoms. Constipating agents, such as loperamide, diphenoxylate-atropine, codeine, and amitriptyline, can be particularly helpful in this group of patients. Loperamide is a synthetic opioid that inhibits bowel motility via mu receptors in the gut; it also has been shown to increase resting internal anal sphincter pressures.
Regular use of enemas to maintain an empty rectum may also be of benefit to several patients. In a study of patients with spina bifida, fecal continence was able to be achieved in 76% of children and 60% of adults with the use of a scheduled regimen of enemas. In elderly patients with constipation and overflow incontinence, a combination of 30 g of lactulose along with daily glycerin suppositories and weekly tap water enemas was shown to reduce fecal incontinence episodes by 35%.
Biofeedback
Biofeedback is a “retraining” program that aims to provide strength, sensory, and coordination training for patients with pelvic floor disorders, including fecal incontinence. The goal is to help patients relearn how to defecate completely, regularly, and effectively. Patients are educated regarding pelvic floor coordination and recognition of sensory thresholds.
This specialized physical therapy can be performed in several different ways, but it most often uses the placement of a pressure-sensitive probe into the anal canal to monitor the strength and coordination of the anal sphincter and pelvic floor musculature. Through transmitting data from the probe to a monitor, patients are able to visualize the effect of their efforts. Through a series of exercises and visual feedback, improvements can be made in pelvic muscle control, rectal sensory threshold, and overall defecatory control.
Several studies have shown the effectiveness of biofeedback in improving fecal incontinence, with success rates ranging from 50% to 90%. In a study comparing biofeedback with pelvic floor exercises alone, 44% of patients in the biofeedback group were able to achieve complete continence, versus 21% in the pelvic floor exercise group ( P = .008).
Unfortunately, many trials lack adequate control groups, and several factors that may have an impact on success, such as the motivation of the patient and the effect of the individual therapist, are difficult to control for. Biofeedback is time-consuming and labor-intensive, and requires a motivated patient. Several methods are described using varying schedules, exercises, and means of feedback. A Cochrane review evaluating the efficacy of biofeedback concluded that no one method has shown superiority over the others; it also concluded that the limited number of trials together with methodological weaknesses do not allow for a definitive assessment of the role of biofeedback in the management of fecal incontinence, although some suggestion was seen that the addition of electrical stimulation may enhance the outcome over biofeedback alone. Regardless, it is an entirely safe technique that offers potential benefit and should be considered first-line therapy in the highly motivated patient for whom traditional medical management has failed.
Surgical management
When medical management of fecal incontinence is inadequate, the next step in management requires surgical intervention, which is the thrust of this article. The next section provides a thorough, but certainly not exhaustive, overview of the most common means of surgical management of fecal incontinence.
Sphincteroplasty
Overlapping sphincteroplasty has traditionally been the most common operation performed for the direct repair of an anatomic sphincter defect associated with fecal incontinence. Through a curvilinear incision anterior to the anus ( Fig. 3 ), the sphincter and associated scar are mobilized ( Fig. 4 ) and reapproximated in an overlapping fashion ( Fig. 5 ), preserving the scar to provide extra bulk and strength to the repair ( Fig. 6 ). Care should be taken not to extend the mobilization too far posterolaterally, to avoid inadvertent pudendal nerve injury. An anterior levatorplasty is often performed in conjunction with the sphincter repair. Outcomes after a direct end-to-end repair are typically inferior to those after an overlapping repair.
Short-term outcomes suggest good to excellent results with overlapping sphincteroplasty in 31% to 83% of patients. However, several studies have shown deterioration in the benefits of sphincteroplasty with regard to fecal continence over time. A recent systematic review evaluating the long-term outcomes of anal sphincter repair reviewed studies with a minimum follow-up of 60 months and found that the initially “good” subjective outcomes with sphincteroplasty deteriorated over time. Despite this, however, most patients remained satisfied with their surgical outcomes and quality of life.
The effect of preexisting pudendal neuropathy on the outcome of overlapping sphincteroplasty is somewhat controversial. Gilliland and colleagues found that bilateral normal pudendal nerve terminal motor latencies were predictive of long-term success after overlapping sphincteroplasty; 62% of 59 patients with bilaterally normal pudendal nerve terminal motor latencies had successful outcomes, compared with only 16.7% of 12 patients with unilateral or bilateral prolonged pudendal nerve terminal motor latencies ( P <.01). Conversely, Chen and colleagues reported significant improvements in continence in patients with unilateral or bilateral prolonged pudendal nerve terminal motor latency.
Injectable Bulking Agents
One of the newest modalities to be applied to the surgical management of fecal incontinence is injectable bulking agents. These materials are injected into the anal canal to physically augment the area and provide an enhanced occlusive mechanical barrier to fecal loss. Several materials have been brought to market, including autologous fat, collagen, hydrogel cross-linked with polyacrylamide, polydimethylsiloxane elastomer, bioabsorbable materials such as stabilized hyaluronic acid (NASHA Dx), injectable synthetic calcium hydroxyapatite ceramic microspheres, silicone biospheres (PTQ, Uroplasty Inc, Minnetonka, MN, USA), and carbon-coated beads (Durasphere, Carbon Medical Technologies, St Paul, MN, USA). Despite the myriad of materials and attempts, debate remains as to the most appropriate site of injection (submucosal vs intramuscular) and efficacy of the procedure, which remains modest, with approximately 50% of patients showing mild to moderate improvement of symptoms. Long-term durability has yet to be proven.
NASHA Dx (stabilized hyaluronic acid) is one of the best-studied materials studied. A double-blinded, randomized trial reported that 52% of those treated with NASHA Dx had 50% or more reduction in the number of incontinence episodes compared with 31% who received sham treatment. Adverse events included 1 rectal abscess and 1 prostatic abscess. In a systematic review of a large group of injectable implants, Hussain and colleagues found that short-term success was associated with the use of PTQ or Coaptite (Boston Scientific, Natick, MA, USA). The use of local anesthetic and failure to use laxatives in the postoperative period was associated with a lower likelihood of success. Watson and colleagues performed a review of a large group of agents and similarly found that most published studies showed improvement in at least 50% of subjects with little reported morbidity, and concluded that anal bulking agents may play a role in alleviating symptoms of fecal seepage or soilage rather than treating more severe incontinence to solid stool. From a prospectively maintained database evaluating the use of collagen injection into the internal anal sphincter, Maslekar and colleagues conducted a retrospective cohort study aimed at looking at medium-term efficacy with a minimum of 36 months of follow-up. They found that in a cohort of 100 patients, 56% reported an improvement in fecal incontinence scores from a mean of 14 to a mean of 8. Among patients overall, 68% reported improvement; however, 38% required a repeat injection, whereas another 15% required a third injection. The median interval between first and final injection was 12 months. No morbidity was reported. They concluded that injection of collagen into the internal anal sphincter is safe and effective in patients with passive fecal incontinence, but repeat treatment may be needed for optimal efficacy in approximately 50% of patients.
Finally, Maeda and colleagues conducted a systematic review of all controlled trials of injectable bulking agents. Of only 5 eligible trials, 4 were at high risk for bias based on methodology. Only the NASHA Dx trial structure was truly methodologically sound. As stated earlier, the NASHA Dx trial showed a significant reduction in the number of incontinent episodes, whereas a trial comparing PTQ versus sham was too small to reliably show a clinical benefit. In a trial of PTQ versus carbon-coated beads (Durasphere), PTQ showed some short-term advantages and was safer. Ultrasound-guided delivery was shown to be safer than digital guidance. No study incorporated patient evaluation of outcomes, so objective continence scores were difficult to compare with practical improvements in daily life.
Sacral Nerve Stimulation
Sacral nerve stimulation (SNS) has emerged as one of the most promising modalities for treating severe fecal incontinence. Initially used to treat urinary incontinence, it was also anecdotally found to have efficacy in improving symptoms associated with fecal incontinence. Initial studies showed very promising results. The therapy involves implantation of an electrode adjacent to the S3 nerve root through the sacral foramen, and subsequent delivery of regular electrical impulses ( Fig. 7 ). The implantation of the device involves an initial test phase to determine efficacy, followed by implant of the permanent stimulator in those with improvement. Tjandra and colleagues performed the earliest prospective trial, comparing SNS to best medical therapy in patients with and without external sphincter defects of up to 120°. In this study of 120 patients (60 in SNS group, 60 in the control group), 90% of patients in the SNS group reported success (defined as at least a 50% improvement in the number of incontinent episodes); perfect continence was achieved in 47.2%. The control group showed no appreciable improvement.
