Faecal Incontinence In Adults - Review

Review

Faecal incontinence in adults Robert D Madoff, Susan C Parker, Madhulika G Varma, Ann C Lowry

Lancet 2004; 364: 621–32

Faecal incontinence can affect individuals of all ages and in many cases greatly impairs quality of life, but incontinent patients should not accept their debility as either inevitable or untreatable. Education of the general public and of health-care providers alike is important, because most cases are readily treatable. Many cases of mild incontinence respond to simple medical therapy, whereas patients with more advanced incontinence are best cared for after complete physiological assessment. Recent advances in therapy have led to promising results, even for patients with refractory incontinence. Health-care providers must make every effort to communicate fully with incontinent patients and to help restore their self-esteem, eliminate their self-imposed isolation, and allow them to resume an active and productive lifestyle. Faecal incontinence is a surprisingly common disorder that receives little attention in general and medical publications. It can contribute to medical morbidity (such as urinary-tract infections and decubitus ulcers) and can burden patients with substantial, continuous financial expenses,1 but its main effect is on quality of life. Patients with faecal incontinence suffer embarrassment, shame, and sometimes depression; some must plan their life around maintaining easy and rapid access to a toilet. Not uncommonly, patients curtail or even entirely avoid activities that other members of society take for granted: shopping, going to the cinema, dining out, or having sexual intercourse. Sadly, many of these lifestyle limitations are unnecessary, because most cases of faecal incontinence are treatable. Barriers to treatment include the unwillingness of patients to broach the subject with health-care providers because of embarrassment or lack of knowledge about the availability of treatment. Even worse, many health-care providers are similarly reticent or ignorant.

Epidemiology Faecal incontinence is the involuntary loss of rectal contents through the anal canal. The true prevalence is unknown, owing to the lack of standard definitions based on severity and frequency, differences in data collection, under-reporting of symptoms by patients, and variations in the populations sampled. International population-based studies have provided widely varying estimates of prevalence, ranging from 0·004% to 18%.2–6 A US telephone survey found a prevalence of 2·2%; of the people who reported symptoms, 30% were over 65 years old and 63% were female.5 Most reported clinical series have a substantial predominance of female patients, but epidemiological studies tend to show an equal sex distribution. The reason for this discrepancy is not known, but it might relate partly to the age and sex of individuals who actively seek treatment. The prevalence of faecal incontinence strongly depends on the population studied. In Switzerland,7 the prevalence was 4·4% in the community, 5·6% for general outpatients, 6·7% for antenatal patients, and 15·9% for urogynaecology patients. A US study of www.thelancet.com Vol 364 August 14, 2004

outpatients2 found an overall prevalence of 18·4%. The prevalence was twice as high for patients visiting a gastroenterologist (26·0%) as for those seeing a primary-care physician (13·4%). The US data, stratified by frequency of episodes, showed that incontinence occurred daily in 2·7% of patients, weekly in 4·5%, and monthly or less in 7·1%. Symptomatic faecal incontinence occurs in 21% of women presenting with urinary incontinence, pelvic-organ prolapse, or both.7 Faecal incontinence disproportionately affects individuals with severe physical and mental disabilities. Patients living in institutions have an extremely high rate of faecal incontinence. Poor functional status, impaired cognitive ability, and limited mobility all contribute to incontinence in nursing-home residents,8 and the rates of incontinence rise with the length of time spent in nursing homes.9 A Canadian study of long-term hospital patients found a prevalence of 46%.10 Similarly, in a US survey of patients in nursing homes, 47% had faecal incontinence.3 One explanation for these strikingly high proportions is the advanced age, severe debility, and numerous associated medical problems of this population. However, another explanation could be that faecal incontinence affecting a child or parent simply cannot be managed by most families and that its development generally mandates nursing-home placement. Two studies of older patients (one focused on patients in nursing homes, the other on those living at home) found an association between severe faecal incontinence and increasing mortality.9,11 Under-reporting of symptoms by patients is a major reason for undertreatment. Only a third of symptomatic patients in the USA discuss their faecal incontinence with their physicians.2 In the United Arab Emirates, 60% of multiparous women with faecal incontinence do not seek medical advice because of embarrassment, the

Division of Colon and Rectal Surgery, Department of Surgery, University of Minnesota, Minneapolis, MN, USA (R D Madoff MD, S C Parker MD, A C Lowry MD) and Department of Surgery, University of California, San Francisco, CA (M G Varma MD) Correspondence to: Dr Robert D Madoff, 393 Dunlap Street North, Suite 500, St Paul, MN 55104, USA madof001@ umn.edu

Search strategy and selection criteria MEDLINE was used to search for articles related to faecal incontinence, emphasising those published from January, 1998. The Cochrane Database of Systemic Reviews was also queried for reviews related to faecal incontinence.

For personal use. Only reproduce with permission from Elsevier Ltd

621

Review

Puborectalis Rectum X-section Anus Pelvis

Sacrum Rectum

Internal sphincter External sphincter Puborectalis Internal sphincter External sphincter

Figure 1: The anal sphincter complex

hope that the problem will resolve spontaneously, the assumption that faecal incontinence is normal, or low expectations of medical care.12

Aetiology The anal sphincter complex consists of the internal and external anal sphincter and the puborectalis muscle (figure 1).13 The smooth muscle of the internal anal sphincter has autonomic innervation and contributes about 55% of the resting tone of the anal canal.14 The external anal sphincter is a striated muscle that has a predominance of slow-twitch, fatigue-resistant muscle fibres.15 It is continuously active and provides 20–30% of the anal resting tone;14 the remainder is provided by expansion of the anal vascular cushions.14,16 The external anal sphincter and puborectalis muscle function as one unit, bring about voluntary sphincter contraction, and normally double the sphincter pressure of the anal canal during voluntary contraction.17 The external anal sphincter is innervated by the inferior branch of the pudendal nerve, which emerges from Alcock’s canal on the medial aspect of the ischium and traverses the ischiorectal fossa. The puborectalis muscle is also directly innervated by the S3 and S4 sacral nerves.18,19 Sensation is provided by receptors in the pelvic floor that detect rectal distension20 and by various receptors in the anal transition zone.20,21 Congenital malformations such as imperforate anus, rectal agenesis, and cloacal defect all can cause faecal incontinence. The severity depends on the bulk and development of pelvic-floor muscles and the degree of impairment of sensory mechanisms. A greater proportion of cases of faecal incontinence are acquired. Sphincter disruption resulting from vaginal delivery is the most common sphincter injury. Other causes of anatomical defects, including anorectal surgery and trauma from impalement or pelvic fracture, 622

account for much of the faecal incontinence seen in men. Lateral internal sphincterotomy,22,23 fistulotomy,22 haemorrhoidectomy, and anal dilation can all result in incontinence.24,25 Sphincter-sparing colorectal resections can also produce incontinence, because of loss of the rectal reservoir and stretching of the sphincter during surgery.26,27 Isolated degeneration of the internal anal sphincter has been reported as a cause of soiling in both men and women.28 Physical examination significantly underestimates the frequency of postpartum sphincter disruption.29 Occult sphincter defects have been detected by ultrasonography in up to 35% of primiparous women after normal vaginal delivery; the associated incontinence rate has ranged from 13% to 23%.30,31 More recent reports indicate a lower rate of sphincter defects, 8–16%, with little or no evidence of incontinence in the immediate postpartum period.32,33 A meta-analysis of 717 vaginal deliveries found a frequency of sphincter defects of 26·9% in primiparous women and a frequency of new sphincter defects of 8·5% in multiparous women.34 Although only 29·7% of sphincter defects were symptomatic in the postparum period, the probability of faecal incontinence resulting from these defects was calculated to be 76·8–82·8%. Disruption of the sphincter complex is associated with diminished sphincter pressures,33 and defects occur more frequently in women who have undergone forceps delivery.30,32,35 Other risk factors for incontinence in women include undergoing the first delivery, giving birth to a baby weighing over 4 kg, and delivering a baby who is in the occiput posterior position.36 Postal surveys of postpartum women have reported incontinence rates of 9·6% to 17·0%,37,38 with up to 50% of women having faecal urgency or incontinence after third-degree obstetric lacerations.36,39 Traction injury to the pudendal nerve commonly accompanies obstetric sphincter laceration and contributes to faecal incontinence.40 The injury is caused by descent of the perineal floor, which stretches the nerve as it emerges from its fixed site of exit from Alcock’s canal. The result is muscle denervation and subsequent reinnervation by nerve sprouting.41 An identical neuropathy causes incontinence in patients with rectal prolapse,42 chronic straining at stool,43,44 and pelvic-floor descent.44 Faecal impaction is an important cause of incontinence, particularly in older people and those living in institutions. Impaction-associated overflow incontinence has generally been attributed to chronic reflex inhibition of the internal anal sphincter, but physiological studies have shown that the cause is a combination of decreased anorectal sensation and reduced sphincter pressures. Factors contributing to impaction include inadequate fibre and water intake, metabolic disorders (hypercalcaemia, hypokalaemia, hypothyroidism), immobility, dementia, depression, and www.thelancet.com Vol 364 August 14, 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Review

drugs (eg, narcotics, antipsychotics, antidepressants, diuretics, and calcium-channel blockers).45 Despite the importance of adequate pelvic-floor function, maintenance of continence depends on an intact chain of anatomical structures and physiological mechanisms, a sequence extending from the anus to the brain. Patients must have the awareness and desire to maintain continence. Many patients with dementia are incontinent because of a lack of interest in, or awareness of, bowel function. Neurological disease or injury that affects the brain, spinal cord, or peripheral nerves can cause incontinence. Congenital neurological causes include spina bifida, myelomeningocele, and meningocele. Acquired neurological causes include stroke, tumours, spinal-cord injury, multiple sclerosis, and diabetic autonomic neuropathy. Abnormal gastrointestinal function can also cause faecal incontinence. Excessive stool volume and rapid gut transit can overwhelm an entirely normal pelvic floor. Contributory abnormalities include intestinal malabsorption, inflammatory bowel disease, and infectious diarrhoea. Finally, we should emphasise that in many cases faecal incontinence arises from a combination of factors. Pudendal neuropathy commonly accompanies obstetric sphincter injury, as already noted. Similarly, women with pre-existing irritable bowel syndrome have more postpartum defecatory urgency and incontinence to flatus than those without, despite a similar frequency of sphincter injury.46 The fact that many women with sphincter injuries do not develop incontinence until later in life also suggests a cumulative, multifactorial process. The possible causes of incontinence are summarised in the panel.

Assessment Many patients find the subject of faecal incontinence difficult to discuss, so they may provide limited or misleading information. Common complaints include urgency, pruritus (which is caused by soiling), and “diarrhoea,” a term commonly used by patients to denote incontinence. Clinicians eliciting such complaints should investigate further, particularly if the patient’s history (eg, recent vaginal delivery) or physical findings suggest faecal incontinence. True incontinence must be differentiated from perineal soiling due to inadequate hygiene or prolapsing haemorrhoids.

History When taking the history, clinicians should attend to these essential elements: the onset and type of incontinence (flatus, liquid, or solid stool); the frequency of episodes; and any related changes in bowel function or stool consistency. Solid stool is easier to control than liquid, so the loss of solid stool generally indicates a greater degree of physiological impairment. However, loss of liquid stool is more troublesome to patients than www.thelancet.com Vol 364 August 14, 2004

Panel: Causes of incontinence Congenital Imperforate anus Rectal agenesis Cloacal defects Myelomeningocele Meningocele Anatomical Obstetric injury, vaginal delivery Anorectal surgery Sphincter-sparing bowel resection Pelvic fracture Anal impalement Neurological Diabetes mellitus Multiple sclerosis Stroke Dementia Central nervous system tumour, infection, trauma Spina bifida Pudendal neuropathy Functional Psychiatric disorder Malabsorption Inflammatory bowel disease Radiation proctitis Hypersecretory tumours Rectal intussusception, prolapse Faecal impaction Physical disabilities

infrequent loss of solid stool.47 The degree of incontinence can also be inferred from the patient’s use of pads or other protection and from restrictive changes in lifestyle, but these behavioural factors may be more strongly influenced by the patient’s anxiety than the actual severity of the disorder. Incontinence can be classified into two categories on basis of the history: passive (unconscious loss of stool) and urge (inability to control a perceived impending bowel movement).48 Many patients with physical and psychological impairments have faecal incontinence, so assessment of the patient’s functional status is important. Physical disabilities can impede access to a toilet, preclude transfer to or sitting on a commode, or prevent cleaning up after bowel movements. Cognitive impairment can affect the ability to sense the need to defecate or the desire to defecate in a controlled way. Environmental assessment in some cases uncovers contributory factors to incontinence, such as a lack of accessible toilets.

Scoring systems Planning of treatment and assessment of the results requires a judgment of the severity of incontinence. In

For personal use. Only reproduce with permission from Elsevier Ltd

623

Review

systems compared in one study correlated with the physician’s clinical impression.50,52 Several issues must be resolved before an ideal scoring system is developed: the definition of incontinence must be standardised; the optimum method of data collection must be decided on (ie, diary versus patients’ recall); the need for data beyond type and frequency must be assessed; and the assignment of numerical values to the combinations of type and frequency must be validated. The developers of the faecal incontinence severity index used patients’ input to assign numerical values, but further work is necessary.47,50 Interpretation of publications on incontinence is difficult because of the lack of standardisation, and even more important, because various methods are used for data collection. For example, maintenance of a daily diary of bowel movements is a far more stringent method than patients’ recall, so results of treatment may appear to be worse. Even the diary approach is limited by the confounding factor of the patient’s activity: the most severely affected patient can appear continent by refusing to venture from a nearby toilet. Quality of life should be assessed independently of severity, because the two measures do not necessarily correlate. Overall health-status instruments and gastrointestinal quality-of-life instruments have been used to measure quality of life related to faecal incontinence, with mixed results.53,54 A recently validated incontinence-specific quality-of-life scale measures four variables (lifestyle, coping/behaviour, depression/selfperception, and embarrassment); it is more sensitive than global scales.55

Normal appearance

External sphincter defect

Internal sphincter defect

Physical examination

Figure 2: Endoanal ultrasonography

contrast to many other disorders, no physiological index of faecal incontinence accurately reflects clinical severity.49 That fact led to the development of scoring systems based on patients’ reports of symptoms.50 The most widely used system includes as variables the frequency and type of incontinence, the extent of lifestyle changes, and the need to wear a pad.51 Although none of the currently used scoring systems has been psychometrically validated, the scores with all four 624

Physical examination reveals the cause of the incontinence in many cases. Pertinent findings include a thinned or deformed perineal body and scars from previous surgery or trauma. Breakdown of the perianal skin is a consequence, not a cause of incontinence, but skin condition should be noted and addressed. Gaping of the anus suggests rectal prolapse, which can usually be demonstrated with Valsalva’s manoeuvre. Digital rectal examination can be used to diagnose faecal impaction associated with overflow incontinence; such an examination is essential to exclude tumours in the anal canal or low rectum. Furthermore, the examiner can assess both resting anal-sphincter tone and the patient’s ability to augment it with voluntary squeeze effort. Diminished perianal sensation and the absence of an anal wink suggest a neurogenic cause of incontinence. Endoscopy to exclude a mass or inflammatory condition is a key adjunct to the physical examination. Flexible sigmoidoscopy is adequate in most cases, but complete colonoscopy should be done if the patient has unexplained diarrhoea, bleeding, or changed bowel habits. The patient’s ability to retain a 100 mL water www.thelancet.com Vol 364 August 14, 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Review

enema is a useful bedside measure of sphincter function.

Anorectal physiology testing Testing of anorectal physiology is useful both as a diagnostic tool and as a way to quantify the magnitude of the physiological defect. Such testing is particularly important when the results of the physical examination are normal, when rectal prolapse is suspected but not demonstrable, or when surgery is contemplated. Useful laboratory studies include endoanal ultrasonography, anal manometry, electromyography, and defecography.

Imaging

Endoanal ultrasonography is a simple and rapid technique that permits accurate delineation of analsphincter anatomy (figure 2). When done by an experienced ultrasonographer, the method has sensitivity and specificity of almost 100% in identifying defects of the internal and external sphincter.56 It can also reveal unsuspected sphincter injuries in a patient thought to have neurogenic incontinence or rule out significant anatomical pathology in a patient with suspected sphincter disruption; either result might lead to a change in the planned therapy.57 Sphincter abnormalities are shown on endoanal ultrasonography in up to 90% of women whose sole risk factor for faecal incontinence is obstetric trauma; thus, this imaging technique is essential for complete assessment of incontinent parous women. MRI is another approach to pelvic-floor imaging (figure 3).58 The need for an endoanal coil to optimise imaging remains controversial. Advantages of MRI include lower dependency on the operator, a wider field of view, and the ability to undertake dynamic studies of pelvic-floor function. Endoanal ultrasonography and MRI have similar accuracy in diagnosing defects of the external anal sphincter, but endoanal ultrasonography is more accurate in diagnosing injuries of the internal anal sphincter.59

Manometry

Anal manometry assesses function of the internal and external anal sphincter, the rectoanal inhibitory reflex, and rectal sensation. It uses a microballoon, a waterperfused catheter, or a solid-state transducer. In a relaxed patient, resting pressures mainly reflect function of the internal anal sphincter; squeeze pressures represent voluntary contraction of the external anal sphincter. Normal values of both resting and squeeze pressures vary among patients: they are lower in women than men and in older patients of both sexes.17 Despite a general relation between sphincter pressure and continence, sphincter pressure varies substantially in both continent and incontinent populations.17,49 That fact helps emphasise the multifactorial nature of incontinence. Furthermore, successful treatment of www.thelancet.com Vol 364 August 14, 2004

Figure 3: MRI of the normal pelvis Sagittal phased-array images of the anal canal and perianal structures. AC=anococcygeal body; BS=bulbospongiosus; EAS=external anal sphincter; PB=perineal body; PU=pubic bone. Reprinted with permission.58

incontinent patients will not necessarily correct manometric abnormalities, and in fact it commonly does not. The rectoanal inhibitory reflex causes relaxation of the internal anal sphincter in response to rectal distension. It is demonstrated by a drop in resting anal pressure in response to inflation of a rectal balloon. This reflex could

For personal use. Only reproduce with permission from Elsevier Ltd

625

Review

Neurophysiological tests

Faecal incontinence

Diarrhoea?

History and physical examination

Yes

• Assess/treat aetiology of diarrhoea: colitis, hypersecretory tumour, radiation, overflow • Medical treatment: fibre, dietary, barrier cream, antidiarrhoeal agents, bowel regimen

Does not resolve

Improves

No

Anorectal physiology testing • Anorectal manometry • Pudendal nerve testing • Endoanal ultrasonography • Defecography (optional)

Sphincter defect? Yes

No

Major defect?

Yes

No Biofeedback

Overlapping sphincteroplasty

Improves

Fails

Endoanal ultrasonography: persistent sphincter defect? Yes

No

Improves

Fails

Consider indications, age, comorbidities, technical issues

Repeat sphincteroplasty with or without biofeedback

Improves

Fails

Dynamic graciloplasty Artifical sphincter Sacral stimulation

Defecography Stoma

Figure 4: Algorithm for incontinence

permit “sampling” of the rectal contents by the transitional-zone receptors, a process that assists in the decision to initiate or defer defecation. The exact mechanism of the decision process remains uncertain. This reflex is absent in patients with Hirschsprung’s and Chagas’ diseases; its loss after low rectal anastomosis is associated with poor functional outcome.60 Rectal sensory testing includes volumetric measurements of the first detectable sensation, the sensation of fullness, and the maximum tolerated volume by balloon distension. Hypersensitivity can be seen with inflammatory disorders, after irradiation, and with irritable bowel syndrome;61 it can lead to urge incontinence. Blunted sensation can also contribute to incontinence;62,63 it is associated with megarectum and with neurogenic disorders such as diabetes and multiple sclerosis.64,65 626

Neurophysiological assessment of incontinent patients is done by single-fibre electromyography or pudendalnerve terminal motor latency (PNTML) testing. Singlefibre electromyography shows multiphasic action potentials in the external anal sphincter, a finding diagnostic of muscle denervation and subsequent reinnervation.41 Although it provides more direct evidence of denervation, single-fibre electromyography has largely been supplanted in many centres by PNTML testing, which uses a glove-mounted intra-anal electrode rather than a needle electrode.66 PNTML testing measures conduction time from stimulation of the nerve at the ischial spine to contraction of the external anal sphincter. Because PNTML testing measures conduction time in the fastest remaining nerve fibres, significant nerve damage is sometimes overlooked. Results are also influenced by the patient’s body type and the technician’s expertise.67 Long latencies are associated with traction injury to the nerve as well as with primary neuropathies. Pudendal neuropathy is seen in up to 70% of patients with faecal incontinence, and in more than 50% of patients with sphincter injury.68 Some investigators have related the presence of pudendal neuropathy to poor results after sphincteroplasty,69 but others have not been able to demonstrate an association.70 Enthusiasm for PNTML testing has been tempered by the lack of consensus about its accuracy and predictive value for outcome after surgery.

Defecography examines rectal emptying of a soft barium paste under fluoroscopy. Although useful for assessment of patients with obstructed defecation, it has limited value in most incontinent patients. Its main role in patients with faecal incontinence is to help diagnose occult rectal prolapse or other suspected pelvic-floor abnormalities (such as a poorly emptying rectocele).

Treatment Medical therapy An algorithm for the treatment of incontinence is given in figure 4. Initial treatment should be conservative. Dietary changes (eg, avoidance of foods that cause diarrhoea or urgency), the addition of supplementary fibre, and bowel habit training are useful for most patients. They may even be the only treatment necessary for those with mild incontinence. Barrier creams, cotton wicks at the anus, and rectal washouts can prevent or ameliorate anal excoriation secondary to leakage. In patients with diarrhoea, assessment of the cause and specific treatment directed at the underlying cause are needed. Patients with idiopathic diarrhoea are treated with antidiarrhoeal medications, such as loperamide, diphenoxylate, and bile-acid binders. Loperamide decreases intestinal motility and secretion and increases www.thelancet.com Vol 364 August 14, 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Review

sphincter pressure.71 Bulking agents, such as psyllium compounds, are also used, because control of solid stool is easier than control of liquid stool. Conversely, patients with constipation and faecal impaction may need routine enemas or laxatives to empty the rectum, so that overflow incontinence does not occur. Incontinent patients with associated irritable bowel syndrome present a therapeutic challenge, given their alternating symptoms of diarrhoea and constipation. Novel approaches to the medical treatment of incontinence include oestrogen replacement therapy in postmenopausal women,72 amitriptyline,73 and valproate sodium.74 Topical 10% phenylephrine increased resting anal pressure in healthy volunteers75 and improved seepage in patients with ileoanal reservoirs,76 but it was ineffective in a randomised controlled trial of incontinent patients.77

Biofeedback Biofeedback is commonly the first treatment recommended if medical therapy fails. It is appealing because it is simple, cheap, and without adverse physical effects. The goal is improved contraction of the external anal sphincter in response to rectal distention. Various protocols and feedback equipment have been reported.78–86 In general, three different protocols are used: coordination training, which teaches patients to contract the sphincter muscle in response to rectal distension; sensory training, which teaches patients to recognise progressively smaller volumes of rectal distension; and strength training, which teaches patients to isolate and exercise the sphincter muscle without using rectal distension. The length and the number of sessions recommended vary widely. In most centres, manometry equipment or an electromyographic rectal probe is used to provide information to patients. The technique chosen does not seem to affect the likelihood of success.87 Success rates after biofeedback range from 38% to 100%.64,78,88–90 The presence of a sphincter defect limits but does not preclude the possibility of a good response.90 Pudendal neuropathy does not adversely affect clinical outcome,91,92 though poor results have been reported in patients with severe neurogenic faecal incontinence who lack rectal sensation.93 The reported success rates for coordination training and for strength training do not differ significantly. Interpretation of reports on biofeedback is difficult. Most studies have been retrospective. Parallel design and randomisation have rarely been used. The definition of success varies widely, and follow-up is short. In most studies, the sample size is small and criteria for selection of patients are not reported. Few studies include controls.78,94 Studies comparing biofeedback with attentive medical care alone are rare and have given conflicting results.88,95 However, a randomised controlled trial showed no advantage of biofeedback over standard www.thelancet.com Vol 364 August 14, 2004

medical and nursing care (advice) or advice plus sphincter exercises.96 Many issues about biofeedback remain unresolved. It is not uniformly available and not covered by all health insurance schemes. No clear criteria for selection of patients have been identified. The optimum protocol, equipment, and duration of treatment are unknown. The mechanism of improvement is poorly understood. No consistent change in sphincter pressures, rectal sensation, or duration of contraction has been reported. Long-term follow-up data are limited, albeit promising, in a few small studies.97–99

Surgery To restore the anal aperture by repairing sphincter defects is the cornerstone of surgery for incontinence. At the time of a recognised obstetric injury, immediate direct repair is advocated, yet persistent defects are common.29,35,100 If immediate repair is not attempted, patients should wait at least 3 months before surgery, so that the magnitude of the functional deficit can be defined, physiological assessment can be done, and local tissue inflammation and oedema can resolve.

Sphincteroplasty

Established injuries in symptomatic patients are treated by overlapping sphincteroplasty. In this technique, a curvilinear incision is made over the perineal body, and the scarred sphincter remnant is dissected back to healthy muscle on either side. The scar is transected, but not excised, and is used as part of the overlapping repair to restore an intact ring of muscle. Many surgeons plicate the levator muscles anteriorly, in an effort to add length to the restored anal canal, but no objective data support this approach. Other surgeons avoid levatormuscle plication out of fear of causing dyspareunia. Individual repair of the internal and external anal sphincter has been advocated by some researchers, but it is not widely practised; moreover, proof of its superiority to single-layer repair is lacking.101 Routine sphincteroplasty is generally done without creation of a diverting stoma; randomised trials have shown no benefit from diversion and morbidity was increased.102,103 As with all therapies for incontinence, the reported success of sphincteroplasty depends on the definitions of incontinence and on the data-collection method. Still, most series report that 60–88% of patients achieve an excellent or good outcome, defined as perfect continence or as incontinence to flatus with minor staining.69,70,104–106 About 15–20% experience no change or a worse outcome. However, several studies have shown that the results of sphincteroplasty deteriorate substantially with time.105,107–109 Some69,106,109,110 but not all series69,103,108 suggest that the presence of pudendal neuropathy adversely affects the outcome of overlapping sphincter repair. Patients for whom sphincteroplasty fails should undergo follow-up

For personal use. Only reproduce with permission from Elsevier Ltd

627

Review

ultrasonography to ensure that the muscle wrap is intact; patients with persisting defects can undergo repeat repair after 6–12 months.111,112 Biofeedback can be effective salvage therapy for patients with suboptimum results after sphincteroplasty.91

Diversion

If all other therapies have failed or if comorbidities preclude more aggressive therapy, faecal diversion remains an excellent alternative. Many patients express reluctance to live with a stoma, but counselling by the physician and an enterostomal therapist can be persuasive. Colostomy admittedly does not confer continence, but it does restore control of bowel evacuation and permits resumption of a normal personal and social life. Emphasis should be placed on creating a well-constructed stoma at an appropriate site, to limit difficulty with its management.

Innovations Restoration of continence to patients when traditional treatment fails, or when traumatic or neurogenic injuries are extensive, remains a challenge. Even after an initial successful result with biofeedback or anterior sphincteroplasty, only about 50% of patients retain the post-treatment degree of continence for 3 years or longer.97,106,108 Options available to such patients include dynamic graciloplasty, an artificial anal sphincter, and sacral-nerve stimulation. Dynamic graciloplasty and artificial anal sphincters are advanced variations of anal encirclement. The earliest and simplest version of anal encirclement was the use of silver wire, described by Thiersch in 1891. Later, Pickrell and colleagues described anal encirclement by use of the gracilis muscle, and other researchers have favoured use of gluteal-muscle flaps.113 Functional results with passive muscle wraps are limited by the inability of patients to maintain, at all times, continuous voluntary contraction of the muscle. Dynamic graciloplasty combines transposition of the gracilis muscle with electrical stimulation via an implantable pulse generator. Application of graded electrical stimulation allows conversion of the fasttwitch, fatiguable gracilis muscle to a slow-twitch, fatigue-resistant muscle that more closely resembles the anal sphincter.114 Baeten and co-workers first reported the use of electrical stimulation with a gracilis-muscle wrap for faecal incontinence in 1988115 and subsequently reported a 72% continence rate. However, multicentre studies have been unable to replicate this degree of success without substantial morbidity and high reoperation rates.116,117 This discrepancy appears to arise at least partly from a relative lack of experience with the technique. Surgeons new to the operation have higher morbidity rates and lower success rates than those with substantial experience.116 Dynamic graciloplasty 628

remains an option for refractory incontinence in a limited number of centres, but it is not approved for use in the USA. The artificial anal sphincter (Acticon Neosphincter, American Medical Systems, Minneapolis, MN, USA) maintains continence via a fluid-filled cuff that surrounds and compresses the anal canal. The patient controls the device via a pump placed in the scrotum or labia majora. Squeezing the pump nine to 12 times forces the fluid from the cuff into a reservoir balloon, which is implanted behind the pubic bone in preperitoneal tissues. Once the cuff is deflated, the anal canal is open, allowing the passage of stool. The cuff then gradually reinflates to occlude the anal canal until defecation is again desired. Christiansen and Lorentzen first reported implantation of an artificial anal sphincter for faecal incontinence in 1987.118 Since then, variable results with the technique have been reported.119–124 Lehur and colleagues reported successful results in 75% of patients, with 29% requiring at least temporary removal of the device.123 By contrast, Malouf and co-workers reported successful results in only 38% of patients; infection, particularly with meticillin-resistant Staphylococcus aureus, was the major cause of failure.124 In a multicentre clinical trial, 67% of patients had a functional device in place at 1 year postoperatively.125 The mean incontinence score (range 0–120) dropped from 105 (incontinent to liquids and solids daily) to 48 (seepage). The infection rate necessitating surgical revision was 25%. In all, 46% of patients (51 of 112) required surgical revision; 37% (41 of 112) required device removal, seven of whom (17%) underwent successful reimplantation. The artificial anal sphincter is suitable for many of the same patients who are candidates for dynamic graciloplasty. However, the perineal soft tissue must be sufficient to allow placement of a Silastic cuff around the anal canal. The device is now available in Europe, Canada, and the USA. Sacral-nerve stimulation, like the artificial anal sphincter, was initially devised for urinary incontinence. Matzel and colleagues introduced the use of such stimulation in 1995 to treat patients with functional, but not anatomical, deficits of the anal sphincter muscle.126 The procedure entails placing an electrode in a sacral foramen (generally S3) to stimulate the sacral nerves. The desired effect is maximum contraction of pelvic muscles, with the minimum possible stimulation of the fibres to the leg. Once the optimum site has been selected, the lead is connected to a temporary external pulse generator for 2 weeks of test stimulation. If function improves adequately, a permanent pulse generator is implanted. Both the initial operation for lead placement and the subsequent one for placement of the pulse generator are done under local anaesthesia with conscious sedation. www.thelancet.com Vol 364 August 14, 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Review

Recent series of patients treated by sacral-nerve stimulation have shown promising results with little morbidity.127,128 Success rates after short-term stimulation approach 90%.129 The great majority of patients for whom test stimulation is successful remain continent after permanent implantation.127,130–132 However, infection and lead displacement remain challenging in up to 25% of patients.133 Because of continuing pain, device repositioning or explantation has been needed for several patients.127,130 Most studies of sacral-nerve stimulation have shown improvement in both resting and squeeze anal pressures as well as increased rectal sensation.133 Ambulatory manometry has shown reduced rectal contractility and suppression of spontaneous anal relaxation.129 The mechanism by which these effects are mediated remains uncertain, though many researchers believe that sacral-nerve stimulation works by modulating local reflex arcs. The mechanism could well be multifactorial. Sacral-nerve stimulation is available in Europe; in the USA, qualifying patients with faecal incontinence can enrol in a trial approved by the Food and Drug Administration. Recent work has investigated novel minimally invasive approaches to faecal incontinence. One option, the addition of a bulking agent to the anal canal to augment resting tone, stems from the routine use of bulking agents for treatment of urinary incontinence caused by intrinsic urinary-sphincter deficiency. The successful use of implantable microballoons,134 carbon-coated beads,135 autologous fat,136 silicone,137 and collagen138 have each been reported in small series with low morbidities. However, both the magnitude and the durability of improvement have varied with such techniques.139 A second investigational approach has used radiofrequency energy to apply a series of small submucosal burns to the anal canal. A pilot series showed promising results,140 but a larger trial showed a lesser degree of improvement.141

Conclusion Faecal incontinence is an embarrassing and sometimes debilitating disorder. Although incontinence is generally treatable, many patients remain untreated because they do not report their symptoms or because their healthcare provider is not familiar with available treatment options. Conservative therapy is successful for many cases of mild incontinence, but more severe cases should be formally assessed before treatment is undertaken. Recent advances have provided new therapeutic options for patients with refractory incontinence. Conflict of interest statement Robert D Madoff consults for Medtronic, manufacturer of the hardware used for dynamic graciloplasty and sacral-nerve stimulation, and for Solvay Pharmaceuticals, manufacturer of topical phenylephrine hydrochloride. Susan C Parker consults for Medtronic and for American

www.thelancet.com Vol 364 August 14, 2004

Medical Systems, manufacturer of the Acticon Neosphincter artificial bowel sphincter. None of the authors received payment (except from The Lancet) for writing this review. Acknowledgments We thank Alexandra A Broek for assistance in preparation of the review, and Mary E Knatterud for editorial help. References 1 Mellgren A, Jensen LL, Zetterstrom JP, Wong WD, Hofmeister JH, Lowry AC. Long-term cost of fecal incontinence secondary to obstetric injuries. Dis Colon Rectum 1999; 42: 857–65. 2 Johanson JF, Lafferty J. Epidemiology of fecal incontinence: the silent affliction. Am J Gastroenterol 1996; 91: 33–36. 3 Nelson R, Furner S, Jesudason V. Fecal incontinence in Wisconsin nursing homes: prevalence and associations. Dis Colon Rectum 1998; 41: 1226–29. 4 Thomas TM, Egan M, Walgrove A, Meade TW. The prevalence of faecal and double incontinence. Community Med 1984; 6: 216–20. 5 Nelson R, Norton N, Cautley E, Furner S. Community-based prevalence of anal incontinence. JAMA 1995; 274: 559–61. 6 Drossman DA, Li Z, Andruzzi E, et al. US householder survey of functional gastrointestinal disorders: prevalence, sociodemography, and health impact. Dig Dis Sci 1993; 38: 1569–80. 7 Faltin DL, Sangalli MR, Curtin F, Morabia A, Weil A. Prevalence of anal incontinence and other anorectal symptoms in women. Int Urogynecol J Pelvic Floor Dysfunct 2001; 12: 117–20. 8 Porell F, Caro FG, Silva A, Monane M. A longitudinal analysis of nursing home outcomes. Health Serv Res 1998; 33: 835–65. 9 Chassagne P, Landrin I, Neveu C, et al. Fecal incontinence in the institutionalized elderly: incidence, risk factors, and prognosis. Am J Med 1999; 106: 185–90. 10 Borrie MJ, Davidson HA. Incontinence in institutions: costs and contributing factors. CMAJ 1992; 147: 322–28. 11 Nakanishi N, Tatara K, Shinsho F, et al. Mortality in relation to urinary and faecal incontinence in elderly people living at home. Age Ageing 1999; 28: 301–06. 12 Rizk DE, Hassan MY, Shaheen H, Cherian JV, Micallef R, Dunn E. The prevalence and determinants of health care-seeking behavior for fecal incontinence in multiparous United Arab Emirates females. Dis Colon Rectum 2001; 44: 1850–56. 13 Jorge JM. Anorectal anatomy and physiology. In: Wexner SD, ed. Fundamentals of anorectal surgery, 2nd edn. Philadelphia: W B Saunders, 1998: 1–24. 14 Lestar B, Penninckx F, Kerremans R. The composition of anal basal pressure: an in vivo and in vitro study in man. Int J Colorectal Dis 1989; 4: 118–22. 15 Johnson MA, Polgar J, Weightman D, Appleton D. Data on the distribution of fibre types in thirty-six human muscles: an autopsy study. J Neurol Sci 1973; 18: 111–29. 16 Gibbons CP, Trowbridge EA, Bannister JJ, Read NW. Role of anal cushions in maintaining continence. Lancet 1986; 1: 886–88. 17 McHugh SM, Diamant NE. Effect of age, gender, and parity on anal canal pressures: contribution of impaired anal sphincter function to fecal incontinence. Dig Dis Sci 1987; 32: 726–36. 18 Matzel KE, Schmidt RA, Tanagho EA. Neuroanatomy of the striated muscular anal continence mechanism: implications for the use of neurostimulation. Dis Colon Rectum 1990; 33: 666–73. 19 Snooks SJ, Swash M. The innervation of the muscles of continence. Ann R Coll Surg Engl 1986; 68: 45–49. 20 Parks AG. Anorectal incontinence. Proc R Soc Med 1975; 68: 681–90. 21 Duthie HL, Gairns FN. Sensory nerve endings and sensation in the anal region of man. Br J Surg 1960; 47: 585–95. 22 Garcia-Aguilar J, Belmonte C, Wong WD, Lowry AC, Madoff RD. Open vs closed sphincterotomy for chronic anal fissure: long-term results. Dis Colon Rectum 1996; 39: 440–43. 23 Nyam DC, Pemberton JH. Long-term results of lateral internal sphincterotomy for chronic anal fissure with particular reference to incidence of fecal incontinence. Dis Colon Rectum 1999; 42: 1306–10. 24 MacIntyre IM, Balfour TW. Results of the Lord non-operative treatment for haemorrhoids. Lancet 1972; 1: 1094–95.

For personal use. Only reproduce with permission from Elsevier Ltd

629

Review

25 Read MG, Read NW, Haynes WG, Donnelly TC, Johnson AG. A prospective study of the effect of haemorrhoidectomy on sphincter function and faecal continence. Br J Surg 1982; 69: 396–98. 26 Otto IC, Ito K, Ye C, et al. Causes of rectal incontinence after sphincter-preserving operations for rectal cancer. Dis Colon Rectum 1996; 39: 1423–27. 27 Ho YH, Tsang C, Tang CL, Nyam D, Eu KW, Seow-Choen F. Anal sphincter injuries from stapling instruments introduced transanally: randomized, controlled study with endoanal ultrasound and anorectal manometry. Dis Colon Rectum 2000; 43: 169–73. 28 Vaizey CJ, Kamm MA, Bartram CI. Primary degeneration of the internal anal sphincter as a cause of passive faecal incontinence. Lancet 1997; 349: 612–15. 29 Warshaw JS. Obstetric anal sphincter injury: incidence, risk factors, and repair. Sem Colon Rectal Surg 2001; 12: 90–97. 30 Sultan AH, Kamm MA, Hudson CN, Thomas JM, Bartram CI. Anal-sphincter disruption during vaginal delivery. N Engl J Med 1993; 329: 1905–11. 31 Faltin DL, Boulvain M, Irion O, Bretones S, Stan C, Weil A. Diagnosis of anal sphincter tears by postpartum endosonography to predict fecal incontinence. Obstet Gynecol 2000; 95: 643–47. 32 Varma A, Gunn J, Gardiner A, Lindow SW, Duthie GS. Obstetric anal sphincter injury: prospective evaluation of incidence. Dis Colon Rectum 1999; 42: 1537–43. 33 Zetterstrom J, Mellgren A, Jensen LL, et al. Effect of delivery on anal sphincter morphology and function. Dis Colon Rectum 1999; 42: 1253–60. 34 Oberwalder M, Connor J, Wexner SD. Meta-analysis to determine the incidence of obstetric anal sphincter damage. Br J Surg 2003; 90: 1333–37. 35 Varma A, Gunn J, Lindow SW, Duthie GS. Do routinely measured delivery variables predict anal sphincter outcome? Dis Colon Rectum 1999; 42: 1261–64. 36 Sultan AH, Kamm MA, Hudson CN, Bartram CI. Third degree obstetric anal sphincter tears: risk factors and outcome of primary repair. BMJ 1994; 308: 887–91. 37 Faltin DL, Sangalli MR, Roche B, Floris L, Boulvain M, Weil A. Does a second delivery increase the risk of anal incontinence? Br J Obstet Gynecol 2001; 108: 684–88. 38 MacArthur C, Glazener CM, Wilson PD, et al. Obstetric practice and faecal incontinence three months after delivery. Br J Obstet Gynecol 2001; 108: 678–83. 39 Sorensen M, Tetzschner T, Rasmussen OO, Bjarnesen J, Christiansen J. Sphincter rupture in child birth. Br J Surg 1993; 80: 392–94. 40 Snooks SJ, Swash M, Setchell M, Henry MM. Injury to innervation of pelvic floor sphincter musculature. Lancet 1984; 2: 546–50. 41 Neill ME, Swash M. Increased motor unit fibre density in the external anal sphincter muscle in ano-rectal incontinence: a single fibre EMG study. J Neurol Neurosurg Psychiatry 1980; 43: 343–47. 42 Parks AG, Swash M, Urich H. Sphincter denervation in anorectal incontinence and rectal prolapse. Gut 1977; 18: 656–65. 43 Snooks SJ, Barnes PR, Swash M, Henry MM. Damage to the innervation of the pelvic floor musculature in chronic constipation. Gastroenterology 1985; 89: 977–81. 44 Kiff ES, Barnes PR, Swash M. Evidence of pudendal neuropathy in patients with perineal descent and chronic straining at stool. Gut 1984; 25: 1279–82. 45 Wrenn K. Fecal impaction. N Engl J Med 1989; 321: 658–62. 46 Donnelly VS, O’Herlihy C, Campbell DM, O’Connell PR. Postpartum fecal incontinence is more common in women with irritable bowel syndrome. Dis Colon Rectum 1998; 41: 586–89. 47 Rockwood TH, Church JM, Fleshman JW, et al. Patient and surgeon ranking of the severity of symptoms associated with fecal incontinence: the fecal incontinence severity index. Dis Colon Rectum 1999; 42: 1525–32. 48 Engel AF, Kamm MA, Bartram CI, Nicholls RJ. Relationship of symptoms in faecal incontinence to specific sphincter abnormalities. Int J Colorectal Dis 1995; 10: 152–55. 49 Felt-Bersma RJ, Klinkenberg-Knol EC, Meuwissen SG. Anorectal function investigations in incontinent and continent patients: differences and discriminatory value. Dis Colon Rectum 1990; 33: 479–85.

630

50 Baxter NN, Rothenberger DA, Lowry AC. Measuring fecal incontinence. Dis Colon Rectum 2003; 46: 1591–605. 51 Jorge JM, Wexner SD. Etiology and management of fecal incontinence. Dis Colon Rectum 1993; 36: 77–97. 52 Vaizey CJ, Carapeti E, Cahill JA, Kamm MA. Prospective comparison of faecal incontinence grading systems. Gut 1999; 44: 77–80. 53 Eypasch E, Williams JI, Wood-Dauphinee S, et al. Gastrointestinal quality of life index: development, validation and application of a new instrument. Br J Surg 1995; 82: 216–22. 54 Sailer M, Bussen D, Debus ES, Fuchs KH, Thiede A. Quality of life in patients with benign anorectal disorders. Br J Surg 1998; 85: 1716–19. 55 Rockwood TH, Church JM, Fleshman JW, et al. Fecal incontinence quality of life scale: quality of life instrument for patients with fecal incontinence. Dis Colon Rectum 2000; 43: 9–16. 56 Sultan AH, Kamm MA, Talbot IC, Nicholls RJ, Bartram CI. Anal endosonography for identifying external sphincter defects confirmed histologically. Br J Surg 1994; 81: 463–65. 57 Liberman H, Faria J, Ternent CA, Blatchford GJ, Christensen MA, Thorson AG. A prospective evaluation of the value of anorectal physiology in the management of fecal incontinence. Dis Colon Rectum 2001; 44: 1567–74. 58 Morren GL, Beets-Tan RG, van Engelshoven JM. Anatomy of the anal canal and perianal structures as defined by phased-array magnetic resonance imaging. Br J Surg 2001; 88: 1506–12. 59 Malouf AJ, Williams AB, Halligan S, Bartram CI, Dhillon S, Kamm MA. Prospective assessment of accuracy of endoanal MR imaging and endosonography in patients with fecal incontinence. AJR Am J Roentgenol 2000; 175: 741–45. 60 Lewis WG, Martin IG, Williamson ME, et al. Why do some patients experience poor functional results after anterior resection of the rectum for carcinoma? Dis Colon Rectum 1995; 38: 259–63. 61 Prior A, Maxton DG, Whorwell PJ. Anorectal manometry in irritable bowel syndrome: differences between diarrhoea and constipation predominant subjects. Gut 1990; 31: 458–62. 62 Buser WD, Miner PB Jr. Delayed rectal sensation with fecal incontinence: successful treatment using anorectal manometry. Gastroenterology 1986; 91: 1186–91. 63 Gladman MA, Scott SM, Chan CL, Williams NS, Luniss PJ. Rectal hyposensitivity: prevalence and clinical impact in patients with intractable constipation and fecal incontinence. Dis Colon Rectum 2003; 46: 238–46. 64 Wald A, Tunuguntla AK. Anorectal sensorimotor dysfunction in fecal incontinence and diabetes mellitus: modification with biofeedback therapy. N Engl J Med 1984; 310: 1282–87. 65 Caruana BJ, Wald A, Hinds JP, Eidelman BH. Anorectal sensory and motor function in neurogenic fecal incontinence: comparison between multiple sclerosis and diabetes mellitus. Gastroenterology 1991; 100: 465–70. 66 Kiff ES, Swash M. Slowed conduction in the pudendal nerves in idiopathic (neurogenic) faecal incontinence. Br J Surg 1984; 71: 614–16. 67 Yip B, Barrett RC, Coller JA, et al. Pudendal nerve terminal motor latency testing: assessing the educational learning curve: can we teach our own? Dis Colon Rectum 2002; 45: 184–87. 68 Roig JV, Villoslada C, Lledo S, et al. Prevalence of pudendal neuropathy in fecal incontinence: results of a prospective study. Dis Colon Rectum 1995; 38: 952–58. 69 Gilliland R, Altomare DF, Moreira H Jr, Oliveira L, Gilliland JE, Wexner SD. Pudendal neuropathy is predictive of failure following anterior overlapping sphincteroplasty. Dis Colon Rectum 1998; 41: 1516–22. 70 Buie WD, Lowry AC, Rothenberger DA, Madoff RD. Clinical rather than laboratory assessment predicts continence after anterior sphincteroplasty. Dis Colon Rectum 2001; 44: 1255–60. 71 Hallgren T, Fasth S, Delbro DS, Nordgren S, Oresland T, Hulten L. Loperamide improves anal sphincter function and continence after restorative proctocolectomy. Dig Dis Sci 1994; 39: 2612–18. 72 Donnelly V, O’Connell PR, O’Herlihy C. The influence of oestrogen replacement on faecal incontinence in postmenopausal women. Br J Obstet Gynaecol 1997; 104: 311–15.

www.thelancet.com Vol 364 August 14, 2004

For personal use. Only reproduce with permission from Elsevier Ltd

Review

73 Santoro GA, Eitan BZ, Pryde A, Bartolo DC. Open study of low-dose amitriptyline in the treatment of patients with idiopathic fecal incontinence. Dis Colon Rectum 2000; 43: 1676–81. 74 Kusunoki M, Shoji Y, Ikeuchi H, Yamagata K, Yamamura T, Utsunomiya J. Usefulness of valproate sodium for treatment of incontinence after ileoanal anastomosis. Surgery 1990; 107: 311–15. 75 Carapeti EA, Kamm MA, Evans BK, Phillips RK. Topical phenylephrine increases anal sphincter resting pressure. Br J Surg 1999; 86: 267–70. 76 Carapeti EA, Kamm MA, Nicholls RJ, Phillips RK. Randomized, controlled trial of topical phenylephrine for fecal incontinence in patients after ileoanal pouch construction. Dis Colon Rectum 2000; 43: 1059–63. 77 Carapeti EA, Kamm MA, Phillips RK. Randomized controlled trial of topical phenylephrine in the treatment of faecal incontinence. Br J Surg 2000; 87: 38–42. 78 Heymen S, Jones KR, Ringel Y, Scarlett Y, Whitehead WE. Biofeedback treatment of fecal incontinence: a critical review. Dis Colon Rectum 2001; 44: 728–36. 79 Wald A. Biofeedback therapy for fecal incontinence. Ann Intern Med 1981; 95: 146–49. 80 Engel BT, Nikoomanesh P, Schuster MM. Operant conditioning of rectosphincteric responses in the treatment of fecal incontinence. N Engl J Med 1974; 290: 646–49. 81 Norton C, Kamm MA. Anal sphincter biofeedback and pelvic floor exercises for faecal incontinence in adults: a systematic review. Aliment Pharmacol Ther 2001; 15: 1147–54. 82 Fynes MM, Marshall K, Cassidy M, et al. A prospective, randomized study comparing the effect of augmented biofeedback with sensory biofeedback alone on fecal incontinence after obstetric trauma. Dis Colon Rectum 1999; 42: 753–58. 83 Goldenberg DA, Hodges K, Hershe T, Jinich H. Biofeedback therapy for fecal incontinence. Am J Gastroenterol 1980; 74: 342–45. 84 Latimer PR, Campbell D, Kasperski J. A components analysis of biofeedback in the treatment of fecal incontinence. Biofeedback Self Regul 1984; 9: 311–24. 85 Miner PB, Donnelly TC, Read NW. Investigation of mode of action of biofeedback in treatment of fecal incontinence. Dig Dis Sci 1990; 35: 1291–98. 86 Constantinides CG, Cywes S. Fecal incontinence: a simple pneumatic device for home biofeedback training. J Pediatr Surg 1983; 18: 276–77. 87 Solomon MJ, Pager CK, Rex J, Roberts R, Manning J. Randomized, controlled trial of biofeedback with anal manometry, transanal ultrasound, or pelvic floor retraining with digital guidance alone in the treatment of mild to moderate fecal incontinence. Dis Colon Rectum 2003; 46: 703–10. 88 Whitehead WE, Burgio KL, Engel BT. Biofeedback treatment of fecal incontinence in geriatric patients. J Am Geriatr Soc 1985; 33: 320–24. 89 McHugh S, Walma K, Diamant NE. Fecal incontinence: a controlled trial of biofeedback. Gastroenterology 1986; 90: 154–55. 90 Norton C, Kamm MA. Outcome of biofeedback for faecal incontinence. Br J Surg 1999; 86: 1159–63. 91 Jensen LL, Lowry AC. Biofeedback improves functional outcome after sphincteroplasty. Dis Colon Rectum 1997; 40: 197–200. 92 Rieger NA, Wattchow DA, Sarre RG, et al. Prospective trial of pelvic floor retraining in patients with fecal incontinence. Dis Colon Rectum 1997; 40: 821–26. 93 van Tets WF, Kuijpers JH, Bleijenberg G. Biofeedback treatment is ineffective in neurogenic fecal incontinence. Dis Colon Rectum 1996; 39: 992–94. 94 Norton C, Hosker G, Brazzelli M. Biofeedback and/or sphincter exercises for the treatment of faecal incontinence in adults. Cochrane Database Syst Rev 2000: CD002111. 95 Loening-Baucke V. Efficacy of biofeedback training in improving faecal incontinence and anorectal physiologic function. Gut 1990; 31: 1395–402. 96 Norton C, Chelvanayagam S, Wilson-Barnett J, Redfern S, Kamm MA. Randomized controlled trial of biofeedback for fecal incontinence. Gastroenterology 2003; 125: 1320–29.

www.thelancet.com Vol 364 August 14, 2004

97 Ryn AK, Morren GL, Hallbook O, Sjodahl R. Long-term results of electromyographic biofeedback training for fecal incontinence. Dis Colon Rectum 2000; 43: 1262–66. 98 Enck P, Daublin G, Lubke HJ, Strohmeyer G. Long-term efficacy of biofeedback training for fecal incontinence. Dis Colon Rectum 1994; 37: 997–1001. 99 Guillemot F, Bouche B, Gower-Rousseau C, et al. Biofeedback for the treatment of fecal incontinence. Dis Colon Rectum 1995; 38: 393–97. 100 Nielsen MB, Hauge C, Rasmussen OO, Pedersen JF, Christiansen J. Anal endosonographic findings in the follow-up of primarily sutured sphincteric ruptures. Br J Surg 1992; 79: 104–06. 101 Wexner SD, Marchetti F, Jagelman DG. The role of sphincteroplasty for fecal incontinence reevaluated: a prospective physiologic and functional review. Dis Colon Rectum 1991; 34: 22–30. 102 Hasegawa H, Yoshioka K, Keighley MR. Randomized trial of fecal diversion for sphincter repair. Dis Colon Rectum 2000; 43: 961–64. 103 Young CJ, Mathur MN, Eyers AA, Solomon MJ. Successful overlapping anal sphincter repair: relationship to patient age, neuropathy, and colostomy formation. Dis Colon Rectum 1998; 41: 344–49. 104 Engel AF, Kamm MA, Sultan AH, Bartram CI, Nicholls RJ. Anterior anal sphincter repair in patients with obstetric trauma. Br J Surg 1994; 81: 1231–34. 105 Karoui S, Leroi AM, Koning E, Menard JF, Michot F, Denis P. Results of sphincteroplasty in 86 patients with anal incontinence. Dis Colon Rectum 2000; 43: 813–20. 106 Londono-Schimmer EE, Garcia-Duperly R, Nicholls RJ, Ritchie JK, Hawley PR, Thomson JP. Overlapping anal sphincter repair for faecal incontinence due to sphincter trauma: five year follow-up functional results. Int J Colorectal Dis 1994; 9: 110–13. 107 Halverson AL, Hull TL. Long-term outcome of overlapping anal sphincter repair. Dis Colon Rectum 2002; 45: 345–48. 108 Malouf AJ, Norton CS, Engel AF, Nicholls RJ, Kamm MA. Longterm results of overlapping anterior anal-sphincter repair for obstetric trauma. Lancet 2000; 355: 260–65. 109 Rothbarth J, Bemelman WA, Meijerink WJ, Buyze-Westerweel ME, van Dijk JG, Delemarre JB. Long-term results of anterior anal sphincter repair for fecal incontinence due to obstetric injury. Dig Surg 2000; 17: 390–93. 110 Sangwan YP, Coller JA, Barrett RC, et al. Unilateral pudendal neuropathy: impact on outcome of anal sphincter repair. Dis Colon Rectum 1996; 39: 686–89. 111 Pinedo G, Vaizey CJ, Nicholls RJ, Roach R, Halligan S, Kamm MA. Results of repeat anal sphincter repair. Br J Surg 1999; 86: 66–69. 112 Giordano P, Renzi A, Efron J, et al. Previous sphincter repair does not affect the outcome of repeat repair. Dis Colon Rectum 2002; 45: 635–40. 113 Devesa JM, Vicente E, Enriquez JM, et al. Total fecal incontinence–a new method of gluteus maximus transposition: preliminary results and report of previous experience with similar procedures. Dis Colon Rectum 1992; 35: 339–49. 114 Konsten J, Baeten CGMI, Havenith MG, Soeters PB. Morphology of dynamic graciloplasty compared with the anal sphincter. Dis Colon Rectum 1993; 36: 559–63. 115 Baeten C, Spaans F, Fluks A. An implanted neuromuscular stimulator for fecal continence following previously implanted gracilis muscle: report of a case. Dis Colon Rectum 1988; 31: 134–37. 116 Madoff RD, Rosen HR, Baeten CG, et al. Safety and efficacy of dynamic muscle plasty for anal incontinence: lessons from a prospective, multicenter trial. Gastroenterology 1999; 116: 549–56. 117 Baeten CG, Bailey HR, Bakka A, et al. Safety and efficacy of dynamic graciloplasty for fecal incontinence: report of a prospective, multicenter trial. Dis Colon Rectum 2000; 43: 743–51. 118 Christiansen J, Lorentzen M. Implantation of artificial sphincter for anal incontinence. Lancet 1987; 2: 244–45. 119 Altomare DF, Dodi G, La Torre F, Romano G, Melega E, Rinaldi M. Multicentre retrospective analysis of the outcome of artificial anal sphincter implantation for severe faecal incontinence. Br J Surg 2001; 88: 1481–86.

For personal use. Only reproduce with permission from Elsevier Ltd

631

Review

120 O’Brien PE, Skinner S. Restoring control: the Acticon Neosphincter artificial bowel sphincter in the treatment of anal incontinence. Dis Colon Rectum 2000; 43: 1213–16. 121 Christiansen J, Rasmussen OO, Lindorff-Larsen K. Long-term results of artificial anal sphincter implantation for severe anal incontinence. Ann Surg 1999; 230: 45–48. 122 Wong WD, Jensen LL, Bartolo DC, Rothenberger DA. Artificial anal sphincter. Dis Colon Rectum 1996; 39: 1345–51. 123 Lehur PA, Roig JV, Duinslaeger M. Artificial anal sphincter: prospective clinical and manometric evaluation. Dis Colon Rectum 2000; 43: 1100–06. 124 Malouf AJ, Vaizey CJ, Kamm MA, Nicholls RJ. Reassessing artificial bowel sphincters. Lancet 2000; 355: 2219–20. 125 Wong WD, Congliosi SM, Spencer MP, et al. The safety and efficacy of the artificial bowel sphincter for fecal incontinence: results from a multicenter cohort study. Dis Colon Rectum 2002; 45: 1139–53. 126 Matzel KE, Stadelmaier U, Hohenfellner M, Gall FP. Electrical stimulation of sacral spinal nerves for treatment of faecal incontinence. Lancet 1995; 346: 1124–7. 127 Matzel KE, Stadelmaier U, Hohenfellner M, Hohenberger W. Chronic sacral spinal nerve stimulation for fecal incontinence: longterm results with foramen and cuff electrodes. Dis Colon Rectum 2001; 44: 59–66. 128 Matzel KE, Kamm MA, Stosser M, et al. Sacral spinal nerve stimulation for faecal incontinence: multicentre study. Lancet 2004; 363: 1270–76. 129 Vaizey CJ, Kamm MA, Turner IC, Nicholls RJ, Woloszko J. Effects of short term sacral nerve stimulation on anal and rectal function in patients with anal incontinence. Gut 1999; 44: 407–12. 130 Malouf AJ, Vaizey CJ, Nicholls RJ, Kamm MA. Permanent sacral nerve stimulation for fecal incontinence. Ann Surg 2000; 232: 143–48. 131 Ganio E, Ratto C, Masin A, et al. Neuromodulation for fecal incontinence: outcome in 16 patients with definitive implant: the initial Italian Sacral Neurostimulation Group (GINS) experience. Dis Colon Rectum 2001; 44: 965–70.

632

132 Matzel KE, Kamm ME, Stosser M, et al. Sacral spine nerve stimulation for faecal incontinence: multicentre study. Lancet 2004; 363: 1270–76. 133 Rosen HR, Urbarz C, Holzer B, Novi G, Schiessel R. Sacral nerve stimulation as a treatment for fecal incontinence. Gastroenterology 2001; 121: 536–41. 134 Feretis C, Benakis P, Dailianas A, et al. Implantation of microballoons in the management of fecal incontinence. Dis Colon Rectum 2001; 44: 1605–09. 135 Weiss E, Efron J, Nogueras J, Wexner S. Submucosal injection of carbon-coated beads is a successful and safe office-based treatment of fecal incontinence. Dis Colon Rectum 2002; 45: A46–47 (abstr). 136 Shafik A. Perianal injection of autologous fat for treatment of sphincteric incontinence. Dis Colon Rectum 1995; 38: 583–87. 137 Malouf AJ, Vaizey CJ, Norton CS, Kamm MA. Internal anal sphincter augmentation for fecal incontinence using injectable silicone biomaterial. Dis Colon Rectum 2001; 44: 595–600. 138 Kumar D, Benson MJ, Bland JE. Glutaraldehyde cross-linked collagen in the treatment of faecal incontinence. Br J Surg 1998; 85: 978–79. 139 Kenefick NJ, Vaizey CJ, Malouf AJ, Norton CS, Marshall M, Kamm MA. Injectable silicone biomaterial for faecal incontinence due to internal anal sphincter dysfunction. Gut 2002; 51: 225–28. 140 Takahashi T, Garcia-Osogobio S, Valdovinos MA, Belmonte C, Barreto C, Velasco L. Extended two-year results of radio-frequency energy delivery for the treatment of fecal incontinence (the Secca procedure). Dis Colon Rectum 2003; 46: 711–15. 141 Efron J, Corman ML, Fleshman J, et al. Safety and effectiveness of temperature-controlled radio-frequency energy delivered to the anal canal (Secca procedure) for the treatment of fecal incontinence. Dis Colon Rectum 2003; 46: 1606–16.

www.thelancet.com Vol 364 August 14, 2004

For personal use. Only reproduce with permission from Elsevier Ltd