and Ridzuan Farouk2
Department of Surgery, Royal Berkshire Hospital, London Road, Reading, Berkshire, RG1 5AN, UK
Department of Surgery, Royal Berkshire NHS Trust, Reading, Berkshire, UK
Faecal incontinence is thought to have a prevalence of between 1 and 10 %. This chapter considers the reasons why normal continence mechanisms break down. Several theories of continence exist including the flutter valve theory, the flap valve theory and sphincteric theories. Any disruption to these processes, such as damage to the internal or external sphincters or their nerve supply, can reduce continence. In addition rapid, uncontrolled delivery of stool into the rectum, or a loss of rectal compliance, may overwhelm the normal processes of continence. Abnormalities in anorectal sensation in conditions such as rectal prolapse may contribute to a loss of continence, as well as abnormal sampling, reliant on the anorectal inhibitory complex. Understanding the mechanism of continence helps to understand the aetiology of incontinence including obstetric injuries, trauma and previous surgery. This chapter ends by considering how to take a history from a patient with incontinence, examine and investigate them as well as how treatments may address the underlying problems described. Surgical methods such as sphincter repair will be discussed but also newer methods such as secca and sacral nerve stimulation.
The rectum and anus, as the terminal segment of the gastrointestinal tract, have the primary function of storage and expulsion of faeces and flatus. With education this function is modified so that expulsion is reserved until the desire to expel rectal contents is sufficiently strong and social circumstances are appropriate.
Quality of life in patients with faecal incontinence as measured by the Faecal Incontinence Severity Index and Faecal Incontinence Quality of Life Scale is generally poorer than that of a ‘normal’ population when assessing social functioning and mental health. Faecal incontinence, when defined as the uncontrolled loss of solid or liquid stool, is difficult to quantify as not all patients will seek medical attention but is thought to have a prevalence of between 1 and 10 %. This chapter will consider the reasons why continence mechanisms break down. Firstly, consideration will be given to the major theories of continence, followed by a discussion of all the factors that usually contribute to continence and where they can fail, leading to incontinence.
Theories of Continence
A number of different theories have been proposed to explain anorectal continence.
The Flutter Valve Theory
Following similar principles to mechanisms of gastro-oesophageal continence to prevent acid reflux, this theory tries to fit a mechanical explanation to continence . In this theory it is supposed that raised intra-abdominal pressure occludes the high-pressure lower rectum, preventing movement of rectal contents into the anal canal. In this theory it was supposed that the sphincters were limited to fine control of continence and against sudden increases in rectal pressure. However, many studies have not found supporting evidence for this theory; for instance, Duthie  found that the mid-portion of the anal canal was the only zone that had a consistently high pressure.
The Flap-Valve Theory
Popularised by Sir Alan Parks , this theory developed the mechanical concept of continence but suggested that it was due to intra-abdominal forces acting against an acute resting anorectal angle. The puborectalis muscle acts like a sling and helps maintain the angle, and as the intra-abdominal pressure rises, so the anterior rectal wall ‘plugs’ the proximal anus. Some studies using videoproctography have not confirmed this, showing no contact between the anterior rectal wall and anal canal during Valsalva . Indeed recent studies have shown little difference in the anorectal angle between incontinent patients and a normal group.
In Parks’ theory a too obtuse angle, associated with perineal descent, would result in incontinence. To correct this, the postanal repair was developed. However, improvement in continence following this procedure has not been associated with an increased acuity in the anorectal angle .
Sphincteric theories of continence explain the effects of damage to the anal sphincters on continence, which can now be better demonstrated by endoanal ultrasound and MRI. The most important mechanism at rest is thought to be the internal anal sphincter (IAS) that forms a high-pressure zone at approximately 2 cm from the anal verge thus preventing entry of rectal contents. The internal anal sphincter is under a combination of controls, most coming from myogenic tone but with some augmentation from sympathetic nerves. Parasympathetic nerves are inhibitory. The IAS alone is unable to maintain complete occlusion, and the canal sits as an anteroposterior slit. Anal endocushions may generate up to 15 % of resting anal pressures and contribute to this occlusion .
The external sphincters are more important for generating additional pressure at times of threatened continence. Bartolo  found that continence was maintained in subjects in whom their rectum was filled with liquid by recruiting the external anal sphincter and puborectalis muscle. When a subject is asked to voluntarily contract their anal sphincters, the canal pressures are seen to rise by 175–270 %. The distribution of the contraction is symmetrical apart from the anterior where the pressures are significantly lower .
Breakdown in Normal Continence
Normal continence is dependent on a complex interaction between local and reflex mechanisms and voluntary intervention. Colonic transit and the consistency of stool being passed are important additional integral factors to maintenance of continence. The rectum is able to act as a reservoir, increasing in volume until a threshold is reached when the subject feels a desire to defecate. The normal subject can exhibit control over when it is appropriate to defecate, as well as discriminate between gas, fluid and solid faeces. Any breakdown in this system can lead to varying degrees of incontinence. It is important therefore to examine in detail the normal processes of rectal filling, sensation and defecation to understand the causes of incontinence.
Rate and Consistency of Delivery of Stool
Rapid rectal filling of very liquid stool can result in urgency and sometimes overcome the normal processes of continence. Any underlying sphincteric weakness may make individuals prone to leakage when stools are too liquid. Any cause of diarrhoea needs to be established before looking too closely for an underlying anatomical cause in the anorectum.
The rate of delivery of stool to the rectum may influence its ability to act as a reservoir through compliance (see below). The angle of the rectosigmoid junction, the resistance provided by the valves of Houston and a reduction in frequency of contractions between the sigmoid and rectum may all play a part in controlling the rate of delivery of stool to the rectum. Continence may be impaired when the rectosigmoid is diseased, as in inflammatory colitis, reducing the control of flow into and from the rectum. Excision of any part of the left colon may also speed up the transit of contents into the lower rectum, leading to urgency.
Abnormal rectal function has been found in patients with faecal incontinence. Patients with normal anal pressure who are incontinent have been demonstrated to have abnormal rectal contractility . Rectal compliance measures the relationship between changes in rectal volume with pressure. Compliance is the change in volume or cross-sectional area divided by the change in pressure. Pressure/volume measurement during distention with a compliant balloon is the most commonly used method for computation of rectal compliance. However, intraindividual and interindividual variations exist. Other methods exist such as rectal distention by a large, noncompliant bag and rectal impedance planimetry for assessment of pressure-cross-sectional-area relations. This has been proposed as an alternative owing to the reduction of errors from elongation of the balloon within the rectal lumen. The rectum should be able to accommodate a significant volume without a significant rise in pressure. Interpretation of rectal compliance in the literature should consider the technique used .
Compliance can change with age but is also altered by disease states. Compliance is reduced in ulcerative colitis, Crohn’s disease and radiation proctitis. In such states there can be an increased tendency to incontinence. In other forms of incontinence, it is possible the reduction in rectal compliance seen is secondary rather than a primary phenomenon.
The anal canal is sensitive to touch, pain, temperature and movement . This aids continence through discrimination between fluid, flatus and faeces. Temperature discrimination is thought to be particularly important to the discrimination of solid stool from liquid stool and flatus. Temperature and electrical sensations are reduced in those with faecal incontinence , as well as with haemorrhoids and diabetic neuropathy.
Patients with rectal prolapse often suffer incontinence that is not associated with a desire to defecate, and soiling may only be discovered on inspection of underclothes. This may be attributed to the relatively insensitive rectal mucosa prolapsing into the anus. Furthermore, an improvement in anal continence status is associated with an improvement in anal sensation following anterior sphincter repair .
Patients who have undergone complete rectal resection and coloanal anastomosis continue to have some sensation, which improves over time, indicating that sensory receptors must also lie in the pelvic floor fascia or musculature .
Recto-anal Inhibitory Complex and Sampling
The anorectal inhibitory reflex occurs when there is a short-lived relaxation of the internal anal sphincter in response to rectal distension. This causes a transient reduction in resting anal pressures. Faecal matter or flatus then comes into contact with specialised anal receptors to allow sampling and discrimination of the state of the rectal contents to allow controlled passage of flatus without soiling. Crucially, the external sphincter is recruited during sampling to ensure continence. If defecation is not appropriate, the pelvic floor muscles return the contents upwards. The frequency of sampling reflexes varies depending on how it is measured (see below), but ambulatory manometry has suggested normal individuals have a rate of about 4 reflexes per hour that is increased following meals and prior to defecation . These increases may be related to daytime rectal motor complex activity. In this study about one-third of events were not consciously perceived as a desire to defecate.
The rectal receptors responsible for this reflex are blocked by topical anaesthesia, which suggests they are located in the mucosa. Distension results in inhibition of the internal anal sphincter muscle fibres, mediated by nitric oxide. The role of the extrinsic autonomic system in this reflex is less clear. Anal relaxation may be absent in progressive rectal distension with low infusion rates, suggesting the internal anal sphincter can maintain continence when the rectum is slowly filling.
The sampling reflex has been demonstrated in the laboratory with rectal insufflation with a balloon which results in IAS relaxation. Ambulatory manometry  has shown frequent IAS relaxations with equalisation of rectal and upper anal canal pressures. Whether transient internal sphincter relaxations are initiated by the movement of rectal contents or are centrally mediated is still unclear.
Transient sphincter relaxation is abnormal in some faecally incontinent patients. The rise in rectal pressure and fall in mid-anal canal pressure is of greater magnitude than normal controls. The frequency of such relaxations is significantly higher in incontinent groups. This is perhaps a reflection of the abnormal internal sphincter in such patients as evidenced by low resting pressures  as well as EMG abnormalities [14, 15] and increased collagen content . The fall in mid-anal canal pressures may also reflect poor external sphincter recruitment in incontinent patients, particularly those with obstetric injuries who are aware of impending leakage but are unable to recruit sufficient external sphincter to prevent this happening.
Full-thickness rectal prolapse is associated with incontinence in 35–100 % of cases. Improvement in continence may be seen following rectopexy in combination with an improvement in anal sensation and resting anal pressure. It is plausible that the prolapse, which starts as an intussusception of the upper rectum, acts like a faecal bolus and induces an anorectal inhibitory reflex which results in internal sphincter relaxation and a fall in anal pressures. In fact studies show that eventually the reflex may not be elicited, perhaps because the internal anal sphincter is maximally inhibited . In addition, patients with prolapse exhibit a different type of high-pressure rectal wave that results in prolonged anorectal inhibition and fall in anal pressures. They are recorded in patients with complete rectal prolapse  where anal pressures fall to levels that would promote faecal leakage.
The anorectal inhibitory reflex is not seen to occur in Hirschsprung’s disease where the myenteric plexus is absent in the rectum. After rectal excisional surgery and coloanal anastomosis, the reflex is initially absent but it rapidly recovers.