Recent trends in diagnosis and treatment of faecal incontinence

Authors


Professor S. S. C. Rao, Neurogastroenterology and Motility, Division of Gastroenterology/Hepatology, University of Iowa, 4612 JCP, 200 Hawkins Drive, Iowa City, IA 5224, USA.
E-mail: satish-rao@uiowa.edu

Summary

The inability to control bowel discharge is not only common but extremely distressing. It has a negative impact on a patient's lifestyle, leads to a loss of self-esteem, social isolation and a diminished quality of life. Faecal incontinence is often due to multiple pathogenic mechanisms and rarely due to a single factor. Normal continence to stool is maintained by the structural and functional integrity of the anorectal unit. Consequently, disruption of the normal anatomy or physiology of the anorectal unit leads to faecal incontinence. Currently, several diagnostic tests are available that can provide an insight regarding the pathophysiology of faecal incontinence and thereby guide management. The treatment of faecal incontinence includes medical, surgical or behavioural approaches. Today, by using logical approach to management, it is possible to improve symptoms and bowel function in many of these patients.

Introduction

Faecal incontinence is defined as the recurrent uncontrolled passage of faecal material for at least 1-month duration.1 The prevalence estimates vary from 2.2 to 18.4% depending on the definition of incontinence, the frequency of occurrence and the clinical setting.2, 3 Although faecal incontinence affects people of all ages, its prevalence is disproportionately higher in the nursing home residents and the elderly.4 Traditionally, faecal incontinence has been felt to be more common in women.5 However, some studies have shown a similar or higher prevalence between men and women.6, 7

Clinically there are three subtypes: (i) passive incontinence – the involuntary discharge of stool or gas without awareness; (ii) urge incontinence – the discharge of faecal matter in spite of active attempts to retain bowel contents; and (iii) faecal seepage – the leakage of small amount of stool without awareness or staining of undergarments following an otherwise normal evacuation.8 The severity of incontinence can range from unintentional elimination of flatus to the seepage of liquid faecal matter to complete evacuation of bowel contents. Because of the embarrassment, patients are often reluctant to report faecal incontinence. This review focuses on the diagnostic evaluation and management of faecal incontinence.

The ability to maintain continence requires structural and functional integrity of the neuromuscular apparatus of the anorectum including the internal and external anal sphincters, pelvic floor musculature and anorectal angle, pudendal nerve function, rectal compliance and rectal sensation. When one or more of these mechanisms of continence are disrupted to an extent that others are unable to compensate incontinence ensues.9

Diagnosis

The first step in the evaluation of faecal incontinence is to establish a rapport with the patient. Patients are often reluctant to admit their symptoms and hence, all patients with diarrhoea, constipation or other anorectal problems should be asked about the occurrence of faecal incontinence. Thereafter, an assessment of its timing and duration, its nature (i.e. incontinence of flatus, liquid or solid stool), and its impact on the quality of life is important. The use of pads or other devices and the ability to discriminate between formed or unformed stool and gas should be documented. A detailed inquiry of obstetric history and co-existing problems such as diabetes mellitus, pelvic radiation, neurological problems, spinal cord injury, dietary history and a history of co-existing urinary incontinence is useful. A prospective stool diary may also be helpful.8

Based on clinical features, several grading systems have been proposed. Recently a modification of the Cleveland Clinic grading system10 has been validated by the St Mark's investigators.11 This system can provide an objective method of quantifying the degree of incontinence and it can also be useful for assessing the efficacy of therapy. Symptom assessment can also provide useful insights regarding the underlying mechanism(s), but may not correlate well with manometric findings. In one study, leakage had sensitivity of 98.9%, a specificity of 11% and a positive predictive value of 51% for detecting low resting anal sphincter pressures.12 The positive predictive value for detecting a low squeeze pressure was 80%.12 Thus, for an individual patient with incontinence, history and clinical features alone are insufficient to define the pathophysiology.

A detailed physical and neurological examination should be performed to rule out a systemic or neurological disorder. The perineal inspection and digital rectal examination is best performed with the patient lying in the left lateral position and with good illumination. Upon inspection, the presence of faecal matter, prolapsed haemorrhoids, dermatitis, scars, skin excoriation, the absence of perianal creases or a gaping anus may be noted. These features suggest either sphincter weakness or chronic skin irritation and provide clues regarding the underlying aetiology. Excessive perineal descent or rectal prolapse can be demonstrated by asking the patient to attempt defecation. An outward bulge that exceeds 3 cm is usually defined as excessive perineal descent.13 The perianal sensation should also be checked. The anocutaneous reflex examines the integrity of the connection between the sensory nerve and the skin, the intermediate neurones in the spinal cord segments S2, S3, and S4 and the motor innervation of external anal sphincter. Impaired or absent anocutaneous reflex suggests either afferent or efferent neuronal injury.14 A digital rectal examination should assess the resting sphincter tone, length of anal canal, the integrity of the puborectalis sling, the acuteness of the anorectal angle, and the strength of the anal muscle and the elevation of the perineum during voluntary squeeze. The sensitivity, specificity and positive predictive value of digital rectal examination – as an objective test for evaluating anal sphincter function is very low.15

Investigations of faecal incontinence

Several specific tests are available for defining the underlying mechanisms of faecal incontinence.16, 17 The selection of diagnostic tests will depend on probable aetiologic factors, symptom severity, impact on quality of life and the patient's age. These tests are often complementary. Flexible sigmoidoscopy or colonoscopy is usually desirable in most patients to exclude mucosal disease or colon cancer. However, in patients with long-standing faecal incontinence without diarrhoea, these tests may not be necessary.

Anorectal manometry

Anorectal manometry provides an objective assessment of anal sphincter pressures and rectoanal reflexes. Currently, several types of probes and pressure recording devices are available to measure anorectal pressures. Each system has distinct advantages and drawbacks.18 Typically, a manometry probe with multiple pressure sensors and a balloon is placed in the rectum and anal canal and the resting and squeeze sphincter pressures are assessed.8, 16, 18

When performed meticulously, anorectal manometry can provide useful information regarding anorectal function.15–17 A reduced resting pressure on anorectal manometry correlates with internal sphincter abnormality and reduced squeeze pressures correlates with external sphincter defects.19 Two large studies have reported that maximum squeeze pressure has the greatest sensitivity and specificity in discriminating faecal incontinence from continent and healthy controls.15, 20 The ability of the external anal sphincter (EAS) to contract reflexly can also be assessed during abrupt increases of intra-abdominal pressure such as when coughing.18, 20 This reflex response causes the anal sphincter pressure to rise above that of the intra-rectal pressure to preserve continence. This reflex response is absent in patients with lesions of the cauda equina or sacral plexus.21, 22

Anorectal manometry is also useful in evaluating the responses to biofeedback training, as well as assessing objective improvement following drug therapy or surgery.17, 23, 24 Normal values of anorectal manometry are influenced by technique, age and gender.25, 26 Women and older patients exhibit lower sphincter pressures. Unfortunately, there is no uniform technique or equipment for performing anorectal manometry.27 Also, there is a dearth of normative date and uniform methods of interpreting test results. Understandably anal canal pressures do not always correlate with continence and other factors should be considered in incontinent patients with normal sphincter pressures.28 The American and European Motility Societies have initiated a collaborative effort to develop standards of manometry testing and recently a consensus document has been published.29

Sensory testing

Rectal sensation and compliance are commonly measured by incremental balloon distension. Often, three types of threshold for rectal sensation can be detected: (i) first detectable sensation; (ii) urge to defecate; and (iii) maximum tolerable volume.17 A higher threshold for rectal sensory perception is associated with autonomic neuropathy, congenital neurogenic anorectal malformation (spina bifida, Hirschsprung disease, myelomeningocele), and functional and somatic alterations of the rectal reservoir, such as megarectum and descending perineum syndrome.30, 31 In patients with faecal incontinence both hyper and hyposensitivity can be seen. In some patients, rectal sensory thresholds may be altered because of changes in the compliance of the rectal wall32 and hence alterations in sensory data should be interpreted along with measurement of rectal compliance.

Anorectal sensation can also be evaluated by using thermal and electric probes. A recent study investigated the application of rectal heat threshold in 31 healthy subjects and found a strong correlation between thermal sensitivity and balloon distension to desire to defecate and maximum tolerable volume.33 Bipolar ring electrodes have also been used to test rectal mucosal electrosensitivity.34 A marked variation in the sensitivity in different regions of section has been demonstrated, and the validity of electromucosal sensitivity as a test of anorectal function has been questioned.35 Many patients with faecal incontinence have significant impairment of electric and thermal anal canal sensation.36, 37 At present, it is unclear whether assessment of anal canal sensation plays any role in diagnosis or treatment of faecal incontinence.17

Rectal compliance reflects both the distensibility and the ability of the rectum to accommodate. It is measured by assessing the changes to rectal pressure during a balloon distension.17 Patients with incontinence often have lower rectal compliance. Rectal compliance is also reduced in patients with colitis,32 low spinal cord lesions and diabetics.38, 39 In contrast, compliance is increased in high spinal cord lesions.22, 40

Imaging the anal canal

Anal endosonography.  Anal endosonography is performed using either a 7 mHz rotating transuducer or 10–15 mHz transducer.41 It provides an assessment of the thickness and structural integrity of the external and internal anal sphincter muscle and can detect the presence of scarring, thinning of sphincter, loss of muscle tissue and other local pathology.42, 43 Anal endosonography has confirmed finding seen histologically,44 physiologically,45 and during surgery.46 Visualization of EAS can be problematic as echogenecity of this is similar to ischioanal fat. Although endosonography can distinguish between internal and external sphincter injury, the findings of a sphincter defect does not necessarily mean that it is the cause of faecal incontinence. The identification of abnormalities is dependent on the training and experience of the operator and the test is subject to an inter-observer variability. In one study interobserver agreement for diagnosis of sphincter defect was reported to be very good,47 whereas another study reported lack of reproducibility of anal sphincter diameter by endoanal ultrasound in healthy volunteers.48 Anal endosonography is a simple and inexpensive method of imaging the anal sphincters and is currently the preferred technique for examining the morphology of anal sphincter.

Defecography.  In this radiographic test, approximately 150 mL of contrast material is placed into the rectum and the subject is asked to squeeze, cough or expel the contrast.49 It is used to assess several parameters; such as the anorectal angle, pelvic floor descent, length of anal canal, presence of rectocele, rectal prolapse or mucosal intussusception. There is poor agreement between observers when measuring the anorectal angle.50 Many investigators have also questioned the rationale for performing defecography in patients with incontinence as it adds very little additional information to that obtained from manometry.17, 51, 52

Magnetic resonance imaging.  Magnetic resonance imaging (MRI) is the only imaging modality that can visualize the anal sphincters and the global pelvic floor motion in real time without radiation exposure. Endoanal MRI has been shown to provide superior imaging with better spatial resolution of EAS. The addition of dynamic pelvic MRI, using fast imaging sequences or MRI colpocystography that involves filling the rectum with ultrasound gel as a contact agent and having the patient evacuate this while lying inside the magnet may define the anorectal structures more precisely.53–55 With the availability of open-magnet units, dynamic MR imaging can be performed in a more physiologic position of patient sitting up.56

Magnetic resonance imaging and endosonography have been compared for the evaluation of anal sphincters. Internal anal sphincter (IAS) is seen more clearly on anal endosonography, whereas the EAS is seen more clearly on MRI. Endoanal MRI, ultrasound and surgical findings were compared in 22 women. MR made the correct diagnosis in 95% of patients compared with 77% with endoanal ultrasound.53 Disadvantages of MRI defecography include limited availability and lack of data comparing symptomatic with normal volunteers. MRI defecography can detect a number of abnormalities in otherwise asymptomatic individuals but their presence correlates poorly with impaired rectal evacuation.17 The functional significance of identifying morphological defects has been questioned17, 57 Inter-observer variability of endoanal MR may be higher than endoanal ultrasound.58

Pudendal nerve terminal latency

The pudendal nerve terminal motor latency (PNTML) measures the functional integrity of the terminal portion of pudendal nerve. Measurement of the nerve latency can help to distinguish whether a weak sphincter muscle is because of muscle or nerve injury. A prolonged nerve latency time suggests pudendal neuropathy and this may occur following obstetric or surgical trauma, excessive perineal descent or idiopathic faecal incontinence.59 A normal PNTML does not exclude pudendal neuropathy, because the presence of a few intact nerve fibres can give a normal result, whereas an abnormal latency time is more significant. The American Gastroenterology Association technical review did not recommend PNTML for the evaluation of patients with faecal incontinence because it correlated poorly with clinical symptoms and histology findings, it did not discriminate muscle weakness caused by nerve or muscle injury, it had poor sensitivity and specificity, was operator dependent and it did not predict surgical outcome.17 However, two recent reviews of eight uncontrolled studies59, 60 suggest that patients with pudendal neuropathy generally have a poor surgical outcome when compared with those without neuropathy.

Clinical utility of tests for faecal incontinence

Few studies have evaluated the utility of anorectal physiologic tests in faecal incontinence. In the study by Wexner and Jorge, history and physical examination alone could detect an underlying cause in only nine of 80 patients (11%) with faecal incontinence, whereas anorectal physiologic tests revealed an abnormality in 66% of patients.61 In another prospective study, anorectal manometry with sensory testing confirmed clinical impression of faecal incontinence and management was altered in 76% of patients.16 A recent prospective study showed that a clinical diagnosis was confirmed in 51% of patients and a new diagnosis was established in the remaining patients by combining anorectal physiological testing with imaging.62

A large study of 350 patients compared continent and incontinent patients. Incontinent patients had lower resting and squeeze pressures, a smaller rectal capacity and leaked earlier following saline infusion in the rectum. Maximum squeeze pressure showed best discrimination. However there was complete overlap between continent and incontinent patients for all the tests and the authors concluded that physiological tests were not useful in predicting faecal incontinence.15 These findings have been further confirmed by another study, which showed that no single test is confirmatory in faecal incontinence, but combination of tests with clinical evaluation is helpful in evaluation of patient with faecal incontinence.63 These studies emphasize the wide range of normal values and the ability of the body to compensate for the loss of any one mechanism.

Newer techniques in diagnosis of faecal incontinence

Dynamic transperineal ultrasound.  This is non-invasive technique designed to investigate the anatomy of the female pelvic floor and perineum both at rest and during straining. In a pilot study, anal sphincter images were comparable with those obtained using anal endosonography. Saggital images permitted the measurement of puborectalis contraction, anorectal angle, rectocele, enterocele and rectoanal intussusception and these were comparable with those obtained during defecography.64

Magnetic pudendal neurostimulation.  This is a novel technique for measuring PNTML using magnetic stimulation to activate the neuromuscular tissue of EAS. In one study magnetic PNTML was comparable in accuracy to that measured by conventional method but was better tolerated.65

Magnetic and electrical stimulation of sacroanal motor pathways.  The combined PNTML and sacral root terminal latency measurement may facilitate an assessment of both the proximal and distal portions of the pudendal nerve in patients with faecal incontinence. However more work is needed in this area.66

Management of patients with faecal incontinence  The goal of treatment for patients with faecal incontinence is to restore continence and to improve the quality of life. Several strategies that include supportive and specific measures may be useful (Table 1).

Table 1.  Treatment of faecal incontinence
Treat underlying cause
Supportive therapy
 Education/counselling/habit training
 Diet (fibre, lactose, fructose)
 Reduce caffeine intake
 Anal hygiene/skin care
Specific therapy
 Pharmacological
  Loperamide
  Diphenoxylate/atropine (Lomotil)
  Codeine
  Cholestyramine/colestipol
  Oestrogen
  Phenylephrine
  Sodium valproate
 Biofeedback therapy (neuromuscular conditioning)
  Anal sphincter muscle strengthening
  Rectal sensory conditioning
  Recto-anal coordination training
 Others
  Anal plugs
  Sphincter bulking (collagen, GAX, silicone)
  Anal electrical stimulation
Surgery
 Sphincteroplasty
 Anterior repair
 Gracilis/gluteus muscle transposition ± stimulation
 Artificial bowel sphincter
 Sacral nerve stimulation
 Colostomy

Supportive measures.  The underlying predisposing condition(s), such as faecal impaction, dementia, neurological problems, inflammatory bowel disease or dietary factors (carbohydrates intolerance) should be treated. In the institutionalized patient with faecal incontinence, the availability of personnel experienced in the treatment of faecal incontinence, timely recognition of soiling and immediate cleansing of the perianal skin is of paramount importance.67 Hygienic measures such as changing undergarments, cleaning the perianal skin immediately following a soiling episode, the use of moist tissue paper (baby wipes) rather than dry toilet paper and barrier creams such as zinc oxide and calamine lotion may be useful in preventing the skin excoriation.67, 68 More significantly, scheduled toileting with a commode at the bedside or bedpan and supportive measures to improve the general well-being and nutrition of the patient may all prove effective. Stool deodorants can be useful for disguising the smell of faeces. In the institutionalized patient, ritualizing bowel habit and instituting cognitive training may prove beneficial. These measures are important, failing which these patients have been shown to have a higher mortality compared with those without incontinence.69 Patients who have constipation, faecal impaction and overflow incontinence may benefit from a regular bowel evacuation programme, incorporating timed evacuation by digital stimulation and/or bisacodyl/glycerol suppositories, fibre supplementation and use of laxatives. Currently, there is no evidence based prospective study to support these approaches.

Other supportive measures can include dietary modifications such as reducing caffeine or fibre intake. Caffeine containing coffee enhances the gastro-colonic response and increases colonic motility70 and induces fluid secretion in the small intestine.71 A food and symptom diary may identify appropriate factors that cause diarrhoeal stools and incontinence; avoiding lactose and fructose may help patients intolerant to these products.72 Fibre supplements such as psyllium are often advocated in an attempt to increase stool bulk and reduce watery stools. However, there has been no published study to justify this approach. It is worth noting that fibre supplements can potentially worsen diarrhoea by increasing colonic fermentation of unabsorbable fibre.

Specific treatment

Specific treatment of faecal incontinence may be considered under the following categories: (i) pharmacologic therapy; (ii) biofeedback therapy; (iii) plugs, sphincter bulkers and ancillary therapy; and (iv) surgery.

Pharmacologic therapy.  Several drugs, each with a different mechanism of action, have been proposed to improve faecal incontinence. Anti-diarrhoeal drugs, such as loperamide hydrochloride or diphenoxylate/atropine sulphate, are commonly used. A placebo-controlled study showed that loperamide 4 mg t.i.d. reduced the stool frequency and urgency, increased colonic transit time,24 reduced stool weight73 and increased anal resting sphincter pressure.74 Clinical improvement was also reported with diphenoxylate/atropine (Lomotil), but objective improvement was lacking.75 Codeine phosphate may show similar benefit but is addictive and may cause drowsiness, whereas diphenoxylate/atropine may cause dryness of mouth. Although most patients benefit temporarily, after a few days, many report crampy lower abdominal pain, or difficulty with evacuation on anti-diarrhoeals. Hence, careful titration is required. Patients with diarrhoea and faecal incontinence secondary to bile salt malabsorption may benefit with an ion exchange resins such as cholestyramine or colestipol.18, 76 Postmenopausal women with faecal incontinence may benefit from oestrogen replacement therapy.77 A serotonin (5-HT3) antagonist, alosetron is currently available under a restricted use programme, and may be useful in patients with severe diarrhoea, especially when symptoms do not respond to other agents, but this has not been tested. Topical phenylephrine, an alpha-1 adrenergic agonist applied to the anal canal in healthy controls increased anal resting pressure by 33% and in incontinent patients.78 However, phenylephrine did not significantly improve incontinence scores or resting anal sphincter pressure in a randomized placebo-controlled crossover study.79 A further controlled trial from the same centre, has demonstrated significant improvement in faecal incontinence by topical phenylephrine in patients with ileoanal pouch.80

Biofeedback therapy.  Behavioural therapy using ‘operant conditioning’, techniques has been shown to improve bowel function and incontinence.81 The goals of biofeedback therapy in a patient with faecal incontinence are (i) to improve the strength of the anal sphincter muscles; (ii) to improve the coordination between the abdominal, gluteal and anal sphincter muscles during voluntary squeeze and following rectal perception; and (iii) to enhance the anorectal sensory perception. Because each goal requires a specific method of training, the treatment protocol should be customized for each patient based upon the underlying pathophysiologic mechanism(s). Biofeedback training is often performed using either visual, auditory or verbal feedback techniques.81 The instrument used to provide feedback can either be a manometry probe or an electromyographic (EMG) electrode that is inserted into the anorectum. The pressure or EMG signals are displayed on a monitor or chart recorder.17, 81 Patients are taught how to selectively squeeze their anal muscles without increasing the intra-abdominal pressure or inappropriately contracting their thigh muscles.82 Other approaches include an augmented biofeedback programme that consists of electrical stimulation of the anal sphincter with EMG feedback.

In the literature, the terms ‘improvement’, ‘success’, or ‘cure’ have been used interchangeably, and the definition for each term has been inconsistent. In uncontrolled studies, subjective improvement has been reported in 40–85% of patients.23, 83, 84 The technique of biofeedback therapy is also not standardized. A few studies have relied upon prospective symptom diaries23, 82, 83, 85–87 but many others have used telephone inquires or one time surveys.84, 88–90 Objective improvement in anorectal function has been less commonly reported.23, 82, 85, 91 A few authors have argued that therapeutic efficacy of biofeedback training cannot be predicted on the basis of manometric results.82, 85 In most of these studies, there was minimal objective improvement. However, one prospective study showed that 1 year after starting therapy, there was significant increase in voluntary squeeze pressure, recto anal coordination, rectal sensation and capacity to retain saline infusion.23 Similar results of improved anal sphincter function have been reported in other studies.92, 93 In a recent study of 100 patients with faecal incontinence, two-thirds improved at the end of treatment, and those with urge incontinence alone faired better than those with passive incontinence (55% vs. 23%).93 The same centre recently reported that change in manometric findings after biofeedback does not correlate with success or failure of biofeedback treatment.87

The mechanism of improvement from biofeedback therapy is not clearly known. The beneficial results of biofeedback therapy may be due to the motivation of the patient and the enthusiasm of the therapist rater than to the technical aspects of therapy.81, 83, 93 A recent randomized controlled trial from a tertiary care centre supports this concept. In this study, biofeedback was compared with conservative standard care or standard care with Kegal exercises. Both the biofeedback and specialist nurse care that included advice on diet, fluids, technique to improve evacuation and bowel training produced 50% or more improvement in patients with faecal incontinence and there was no difference between treatments.87 Anal ultrasound or manometric findings did not predict outcome. Why patients underwent repeat testing and more than 25% were lost for follow up.

In spite of a lack of uniform approach, most techniques of biofeedback therapy seem to confer benefit. Conservative treatment and biofeedback therapy should be offered to all patients before reconstructive surgery for faecal incontinence and to those who have failed supportive measures, especially older patients, those with comorbid illnesses, or pudendal neuropathy.

Surgery.  Surgical treatment for faecal incontinence should be considered in patients who do not respond to medical treatment or have well documented sphincter defect.

In 80% of patients with obstetrical damage, anterior overlap repair of the external anal sphincter resolves symptoms.94 In patients with incontinence due to a weak, but intact anal sphincter, postanal repair has been tried.95 Over time success of sphincter repair seems to wear off and less than one-third of patients are continent to liquid or solid stool after 5 years.96, 97

If the anal sphincter is irreparably damaged, reconstruction of the sphincter may be required. This repair can be made either by reinforcing the existing sphincter with skeletal muscles or by artificially creating a sphincter (neo-sphincter construction). Striated muscles, gracilis or gluteus can be surgically wrapped around the anal canal to increase resting pressure.60, 98 The technique of stimulated gracilis muscle transposition (dynamic gracioplasty) has been most frequently tried.99 Continuous electrical stimulation with an implantable pulse generator helps to maintain muscle contraction.100 This technique uses the principle that a fast twitch, fatigable skeletal muscle when stimulated over a long term can be transformed into a slow twitch, nonfatigable muscle that can provide a sustained, sphincter-like muscle response. Clinical improvement (success) rates have ranged between 38 and 90%.60, 101 A recent systemic review showed that dynamic gracioplasty is associated with 2% mortality and a significant risk of re-operation.102 Another surgical approach has been to implant an artificial bowel sphincter. A multicentre, prospective study evaluated the safety and efficacy of the artificial bowl sphincter; 115 patients were implanted, and a successful outcome was achieved in 85% of patients with functioning device, and with an intention to treat analysis, the success rate was 53%. However, 384 device-related adverse events were reported in 99 patients, and 46% of patients required revisonal operation, 37% had their device completely explanted.103

Plugs, procon incontinence device, sphincter bulkers.  Innovative disposable anal plugs have been designed to temporarily occlude the anal canal.104These plugs may be useful for patients with impaired anal canal sensation, those with neurological disease and those who are institutionalized or immobilized. In some patients with faecal seepage, insertion of an anal plug made up of cotton wool may prove beneficial.

Procon incontinence devices, which consist of a rubber catheter with an infrared photo-interrupter sensor and flatus vent holes, have been used for patients with severe faecal incontinence. This technique requires patients who can participate in their care.105 Bulking the anal sphincter in order to augment its surface area and thereby provide better seal for the anal canal has been attempted using a variety of agents, including autologous fat,106 Glutaraldehyde-treated collagen (GAX),107 or synthetic macromolecules,108 or silicon.109 These materials are usually injected submucosally either at the site where the sphincter is deficient or circumferentially, if the whole muscle is degenerated or fragmented. Studies have shown some improvement in the short-term in patients with passive faecal incontinence.107, 108 However, the experience with these techniques is limited, and there is no controlled or long-term outcome study.

Electrical stimulation.  With this technique, transanal electric stimulation is intermittently applied to the anal canal by means of electrodes attached to a portable stimulator in an effort to stimulate muscle contraction. The value of electrostimulation in the treatment of incontinence is controversial; some studies have shown improvement in faecal incontinence,110, 111 although one study showed no effect.112 A meta-analysis reported that there was insufficient data to draw meaningful conclusions regarding the efficacy of this treatment.113

Encouraged by the therapeutic effect of radio-frequency energy delivered to the lower oesophageal sphincter for the treatment of gastro-oesophageal reflux disease, ‘secca procedure’ was developed to deliver energy to all quadrants of the anal canal muscles in patients with faecal incontinence. A multicentre study of 50 patients demonstrated that radiofrequency energy can be safely delivered to the lower rectum and anal canal and it significantly improved faecal incontinence and overall quality of life.114 In a single centre study of 10 patients, this procedure resulted in a significant improvement in symptoms and quality of life, which persisted for 2 years after treatment.115 The proposed mechanism of action is heat-induced tissue contraction and remodelling of the anal canal and distal rectum.

Sacral nerve stimulation.  This new therapeutic approach is less invasive and has the advantage that a temporary procedure can be carried out prior to the final operation to make a reliable estimate of its outcome. This technique is well established in the treatment of urinary incontinence. Temporary electrodes are placed percutaneously through the sacral foramina. If the test period of 2–3 weeks shows satisfactory continence, the permanent electrode is placed and a neurostimulator is implanted. In one study that assessed the short-term effects,116 continence was restored in eight of nine patients. Another study analysed the medium term results of permanent sacral nerve stimulation in 15 patients and report improvement in continence in all patients up to 5 years. 117

Other procedures.  The Malone or ante-grade continent enema procedure118 consists of fashioning a caceostomy button or appedicostomy,119, 120 which allows ante-grade wash out of the colon and may be suitable for children and patients with neurological lesions.121 In the long term, fibrosis of the stoma site may lead to a loss of response but overall success rate of 61% has been reported at a mean follow up of 3.25 years.122

If none of these techniques are suitable or have failed, a colostomy is a safe option, although aesthetically less preferable.123 It is particularly suitable for patients with spinal cord injury, those immobilized with skin problems or other complications.123, 124 A colostomy should not be regarded as a failure of medical or surgical treatment.123 In many, the restoration of a normal quality of life and ameliorization of symptoms can be very rewarding. The use of laparoscopic-assisted approach, Trephine colostomy may help to fashion a stoma with minimal morbidity for the patient.125

There are no controlled studies that have compared surgical management with pharmacological therapy or biofeedback therapy. Similarly, there are no controlled studies that have compared the different surgical approaches. Because the outcome of most procedures range from significant improvement initially to a less satisfactory results in the long-term, no single procedure is universally accepted. It is likely that through a better understanding of the underlying pathophysiology and the development of safer and better techniques combined with prospective controlled trials, it may become possible in the near future to select younger patients with well-defined sphincter defects for appropriate surgery.

Conclusion

Faecal incontinence is a subjective complaint with complex aetiology and pathogenesis and scant evidence-based data for management. A detailed history and examination including digital rectal examination, and where appropriate colorectal mucosal biopsy may help in the diagnosis of common disorders. All patients should be offered conservative management, and if this fails, further investigations should be undertaken. Anorectal physiological tests help in the evaluation of functional abnormalities and anal endosonography in the assessment of sphincter defects. The results of these tests may guide further management. However, abnormal findings demonstrated by these procedures do not predict severity of incontinence or response to treatment. Behavioural therapy is successful in most patients and should be offered first. If a well-defined sphincter defect with no pudendal neuropathy is identified, sphincteroplasty may be effective. If the sphincter is badly damaged, dynamic graciloplasty or artificial sphincter reconstructive surgery may be considered. However, these procedures are only performed in a few specialist centres, and if unavailable colostomy should be regarded as viable option. Several experimental approaches, including bulking of the anal sphincter, sacral nerve stimulation and the delivery of radiofrequency energy to the anal canal are currently in progress. Surgical treatment of faecal incontinence improves symptoms but does not cure and the outcome deteriorates with time.

Ancillary