Berek and Novak's Gynecology 15th Ed.

28 Anorectal Dysfunction

Robert E. Gutman

Geoffrey W. Cundiff

• Defecatory dysfunction and fecal incontinence are common conditions for women that have tremendous psychosocial and economic implications.

• The differential diagnosis for anorectal dysfunction is broad and can be classified into systemic factors, anatomic and structural abnormalities, and functional disorders.

• A thorough history and physical examination are critical for the evaluation of fecal incontinence and defecatory dysfunction, as well as appropriate ancillary testing.

• Treatment of anorectal dysfunction should focus on treatment of the underlying condition, with nonsurgical management attempted before surgery.

• Overlapping sphincteroplasty is the procedure of choice for fecal incontinence caused by a disrupted anal sphincter.

Anorectal dysfunction encompasses a variety of conditions that disrupt normal anorectal function. Such conditions can be subdivided as those that cause defecatory dysfunction and fecal incontinence. Although anorectal dysfunction transcends any individual medical specialty, the pathophysiology, evaluation, and management of conditions relevant to obstetricians/gynecologists are presented in this chapter.

Normal Colorectal Function

Anal continence and defecation are complex physiologic processes that require intact and coordinated neurologic and anatomic function, including colonic absorption and motility, rectal compliance, anorectal sensation, and the multifaceted continence mechanism. An understanding of normal physiology and pathophysiology is essential to the treatment of women with anorectal dysfunction.

Stool Formation and Colonic Transit

The colon plays an important role in absorption and regulation of water and electrolytes. As much as 5 L of water and associated electrolytes can be absorbed in one day. Parasympathetic-mediated peristaltic contraction of colonic smooth muscle transfers fecal material to the rectum. A delay in stool transit at the rectosigmoid region of the colon allows for maximal absorption of water and sodium.

Storage

As stool accumulates in the rectosigmoid, rectal distention triggers a transient decrease in the internal anal sphincter (IAS) tone and an increase in the external anal sphincter (EAS) tone, known as the rectoanal inhibitory reflex. Exposure of the anal canal to fecal matter facilitates sampling, whereby the anal canal and its abundant sensory nerves determine stool consistency (i.e., solid, liquid, or gas). Accommodation occurs as the normally compliant rectal vault relaxes in response to increased volume. This cycle, combined with increased rectal distention, stimulates an urge to defecate.This urge can be voluntarily suppressed through cortical control, resulting in further accommodation and activation of the continence mechanism.

Continence Mechanism

Muscles

The key muscles of the continence mechanism are the puborectalis, IAS, and EAS. The puborectalis muscle originates from the pubic rami at the level of the arcus tendineus levator ani and passes laterally to the vagina and rectum in a U-shaped configuration, creating a sling around the genital hiatus. Contraction of the puborectalis muscle narrows the genital hiatus, developing the near 90-degree anorectal angle. The resting tone of the puborectalis muscle serves as the primary continence mechanism for solid stool. The IAS and EAS are essential for continence of flatus and liquid stool. The internal sphincter maintains most of the resting tone for the sphincter complex through autonomic reflex arcs and is essential for passive continence. Although the external sphincter also maintains constant resting tone, it is ultimately responsible for preventing fecal urgency and stress incontinence associated with sudden increases in intra-abdominal pressure. This function is under both voluntary and involuntary control. The anal cushions act as the final anatomic barrier. They fill with blood, causing occlusion of the anal canal.

Nerves

Many pathologic states disrupt normal function through denervation. The IAS receives its sympathetic supply from L5, which passes through the pelvic plexus via the hypogastric plexus. The parasympathetic supply from S2–4 synapses at the pelvic plexus, where it joins the sympathetic nerves. In addition to the parasympathetic and sympathetic components, the autonomic nervous system of the gut has an enteric nervous system (ENS). The ENS provides local circuitry that can contract or relax the gut muscles, as well as impact absorption and secretion. The autonomic ganglia of the ENS, located in the gut, are interconnected to provide local integration and processing of information. The IAS acts through reflex arcs at the spinal cord without voluntary control. The puborectalis (levator ani) is innervated by branches of the S2–4 sacral roots and does not receive direct innervation from the pudendal nerve (1). The EAS is innervated bilaterally by the pudendal nerve (S2–4) via Alcock’s canal. The pudendal nerve fibers cross over at the level of the spinal cord, allowing preservation of EAS function in the event of unilateral damage. The rich sensory supply from the anal canal travels along the inferior rectal branch of the pudendal nerve.

Evacuation

Initiation of defecation is normally under cortical control. As previously discussed, delivery of stool to the rectum stimulates the rectoanal inhibitory reflex, permitting sampling followed by accommodation. Further rectal distention results in an urge to defecate. Evacuation occurs with voluntary relaxation of the pelvic floor muscles (puborectalis muscle and EAS) in conjunction with increased intra-abdominal and intrarectal pressure from Valsalva. This results in widening of the anorectal angle and shortening of the anal canal, which facilitates emptying. Coordinated peristaltic activity of the rectosigmoid assists evacuation. After this process is complete, the closing reflex is initiated, resulting in contraction of the pelvic floor muscles and activation of the continence mechanism.

Epidemiology

The epidemiology of anorectal dysfunction has been best defined in terms of the incidence and prevalence of fecal incontinence. Few studies have been done to assess the incidence and prevalence of defecatory dysfunction.

Defecatory Dysfunction

The term defecatory dysfunction often is used synonymously with the symptom of constipation. Constipation is an imprecise term used by patients to report a variety of symptoms, including infrequent stools, dyschezia, straining, variation in stool consistency and caliber, incomplete emptying, bloating, and abdominal pain. The most common symptoms associated with constipation are straining and hard stools (2,3). Defecatory dysfunction is defined by many physicians as infrequent stools, typically fewer than three bowel movements per week. This definition is based on stool frequency studies in which 95% of women have more than three bowel movements per week. Using this definition, the prevalence of constipation should be 5% (4). However, the prevalence of constipation has been estimated to range from 2% to 28%, depending on the definition applied (5–7).

There is an increased prevalence of constipation among women and elderly individuals, nonwhite individuals, and those with low income and low education levels (5–7). Based on an estimated 2.5 million visits to US physicians per year for constipation, with an average cost of $2,752 per patient, the annual cost for evaluation of constipation would be approximately $6.9 billion (8,9). An estimated 85% of physician visits results in a prescription; including drug costs would increase this amount substantially (8). More recently, evaluation of 76,854 California Medicaid patients without supplemental insurance found somewhat lower annual total direct costs to care for constipation at almost $19 million ($246 per patient) for this subset of the US population (10). Constipation has a detrimental effect on health-related quality of life (3,10). Constipation contributed to decreased mental and physical scores for quality of life on the SF-36 Health Survey in a Canadian-based population (11).

Fecal Incontinence

The reported prevalence of fecal incontinence varies between 2% and 3% for community-dwelling individuals, 3% to 17% for those of increased age, and 46% to 54% for nursing home residents(12). A prevalence of 28% has been reported among patients seeking benign gynecologic care and 36% of primary care patients surveyed (13,14). The prevalence of fecal incontinence in the United States is expected to increase 59% from 10.6 million in 2010 to 16.8 million in 2050 as the population ages (15). Epidemiologic studies of fecal incontinence are compromised by social stigmata and the lack of a uniform definition. Definitions of fecal incontinence vary with respect to the type of material passed (solid, liquid, or gas), the frequency and duration of events (once in a lifetime to twice a week), and the impact on quality of life. Most authors agree that the true prevalence of this condition is underestimated in the current scientific literature. A large health survey in the United States found age, female gender, physical limitations, and poor general health to be independent risk factors associated with fecal incontinence (16).

Fecal incontinence has tremendous psychosocial and economic implications for individuals and society as a whole. The loss of such a basic function can be emotionally devastating, leading to poor self-esteem, depression, social isolation, and decreased quality of life (13,14,17). Fecal incontinence is the second leading reason for nursing home placement in the United States, even though less than one-third of individuals with this condition seek medical attention (13,17). The overall annual cost to treat fecal incontinence is difficult to pinpoint, but accounts for more than $400 million per year in the cost of adult incontinence products alone (17).

Symptom-Based Approach to Colorectal Disorders

Several medical conditions cause defecatory dysfunction, fecal incontinence, or combined symptoms. Following is the differential diagnosis—a proposed classification system based on systemic factors, anatomic and structural abnormalities, and functional disorders.

Differential Diagnosis

Disordered Defecation

Causes of defecatory dysfunction have traditionally been divided into systemic disorders and idiopathic constipation (all nonsystemic causes). Idiopathic constipation can be subdivided into anatomic and structural abnormalities and functional disorders (Table 28.1).

Table 28.1 Causes of Defecatory Dysfunction and Fecal Incontinence

Fecal Incontinence		Defecatory Dysfunction
	Systemic Factors
	Metabolic/Endocrine
●	Diabetes mellitus	●
●	Thyroid disease	●
	Hypercalcemia	●
	Hypokalemia	●
	Neurological
●	Central Nervous System	●
	Multiple sclerosis, Parkinson disease, stroke, tumor, dementia
●	Peripheral Nervous System	●
	Hirschsprung disease, spina bifida, autonomic neuropathy, pudendal neuropathy
	Infectious
●	Bacterial, viral, parasitic diarrhea
	Collagen Vascular/Muscle Disorder
	Systemic sclerosis, amyloidosis, myotonic dystrophy, dermatomyositis	●
	Idiopathic/Autoimmune
●	Inflammatory bowel disease
●	Food allergy
	Medications
●	Prescription, over the counter	●
	Anatomical/Structural Abnormalities
	Pelvic Outlet Obstruction
●	Pelvic organ prolapse	●
●	Descending perineum syndrome	●
	Anismus/rectosphincteric dyssynergia	●
●	Intussusception, rectal prolapse	●
	Volvulus	●
●	Neoplasia	●
●	Benign strictures	●
●	Hemorrhoids	●
	Anal Sphincter Disruption/Fistula
●	Obstetrical trauma
●	Surgical trauma
●	Anal intercourse
●	Injury (trauma, radiation proctitis)
	Functional
	Motility Disorders
	Global motility disorder	●
	Colonic inertia/slow-transit constipation	●
●	Irritable bowel syndrome	●
	Functional constipation	●
●	Functional diarrhea
	Functional Limitations
●	Decreased mobility	●
●	Decreased cognition	●

Diabetes, hypothyroidism, and pregnancy are the most common endocrinologic systemic factors that cause constipation, and all have a component of decreased gastrointestinal motility and intestinal transit. In one study, gastrointestinal symptoms were present in 76% of patients with diabetes, including constipation, which occurred in 60% (18). In patients with diabetes, constipation is believed to be secondary to intestinal autonomic neuropathy, resulting in delayed or absent gastrocolic reflex and decreased bowel motility. This enteric neuropathy may also cause gastroparesis and diarrhea. Although diabetes has been classified with the endocrinologic causes, it should also be grouped with the enteric neuropathies. Pregnancy is not considered a disease state; however, there is an 11% to 38% prevalence of constipation that is believed to result from the effect of progesterone on smooth muscle (19,20). Iron supplements and prior constipation treatment are also associated with constipation during pregnancy (20).

The neurologic systemic factors can be divided into central and peripheral processes. Spinal cord lesions, multiple sclerosis, and Parkinson disease affect the autonomic nervous system. Trauma to the sacral nerves often leads to severe constipation from decreased left-sided colonic motility, decreased rectal tone and sensation, and increased distention. These findings are also seen in patients with meningomyelocele, damage to the lumbosacral spine, and pelvic floor trauma (21,22). Higher spinal cord lesions result in delayed sigmoid transit and decreased rectal compliance. In these upper motor neuron lesions, colonic reflexes are intact, and defecation can be initiated by digital stimulation of the anal canal (23,24). Individuals with multiple sclerosis can have no gastrocolic reflex, decreased colonic motility, decreased rectal compliance, and even rectosphincteric dyssynergia (25,26). Constipation worsens with the duration of illness and may be compounded by the side effects of medical therapy. Similar findings of rectosphincteric dyssynergia and medication side effects are present with Parkinson disease.

Among the peripheral neurogenic disorders, dysfunction occurs at the level of the ENS. The ultimate example of this is congenital aganglionosis (Hirschsprung disease). The absence of intramural ganglion cells in the submucosal and myenteric plexuses of the rectum causes loss of the rectosphincteric inhibitory reflex. Patients with this illness usually present with functional obstruction and proximal colonic dilation. In most patients, the condition is diagnosed within 6 months of age, although milder cases can be seen later in life.

Other systemic factors to consider are collagen vascular and muscle disorders. Importantly, some of the most commonly used prescription and over-the-counter medications, including aluminum antacids, beta-blockers, calcium channel blockers, anticholinergics, antidepressants, and opiates, cause defecatory dysfunction (Table 28.2). Lifestyle issues, such as inadequate fiber intake and insufficient fluid intake, can exert similar effects independently or in conjunction with other disorders.

Table 28.2 Drugs Associated with Constipation

Over-the-Counter Medications
Antidiarrheals (loperamide, Kaopectate)
Antacids (with aluminum or calcium)
Iron supplements
Prescription Medications
Anticholinergics	Others
Antidepressants	Iron
Antipsychotics	Barium sulfate
Antispasmodics	Metallic intoxication (arsenic, lead, mercury)
Antiparkinsonian drugs	Opiates
Antihypertensives	Nonsteroidal anti-inflammatory agents
Calcium channel blockers	Anticonvulsants
Beta-blockers	Vinca alkaloids
Diuretics	5HT3 antagonists (ondansetron, granisetron)
Ganglionic blockers

Structural abnormalities refer to the obstructive disorders, such as pelvic organ prolapse, perineal descent, intussusception, rectal prolapse, and tumors. Functional disorders are those that do not have an identifiable anatomic or systemic etiology. Most functional disorders are motility disorders, such as slow-transit constipation or colonic inertia, irritable bowel syndrome (constipation predominant), and functional constipation. The Rome III criteria created strict definitions for these idiopathic conditions that are believed to result from the complex interaction of psychosocial factors and altered gut physiology via the gut-brain-gut axis (27). Patients also may have functional limitations, such as decreased mobility and cognition. It is important to understand that this classification system is somewhat arbitrary, and several of these conditions are interrelated.

Fecal Incontinence

Anal continence depends on a complex interaction of cognitive, anatomic, neurologic, and physiologic mechanisms. The continence mechanism can often compensate for a deficiency in one of these processes, but it can be overwhelmed with increased severity or decreased function over time. Systemic etiologies of fecal incontinence often are due to disease states that cause diarrhea. The rapid transport of large volumes of liquid stool to the rectum can produce urgency and incontinence even in healthy individuals (28). Fecal incontinence frequently results from infectious diarrhea caused by bacteria (e.g., Clostridium, Escherichia coli, Salmonella, Shigella, Yersinia, Campylobacter), viruses (e.g., Rotavirus, Norwalk, HIV), and parasites (e.g., Entamoeba, Giardia, Cryptosporidium, Ascaris). Numerous medications and dietary items cause diarrhea and fecal incontinence (Table 28.3). Endocrine factors that can lead to fecal incontinence include diabetes mellitus and hyperthyroidism. With diabetes, diarrhea can develop from autonomic dysfunction, bacterial overgrowth, osmotic diarrhea with sugar substitutes, and pancreatic insufficiency. Inflammatory bowel disease is considered an idiopathic or autoimmune systemic factor. Ulcerative colitis and Crohn disease cause fecal incontinence during exacerbations with bouts of bloody diarrhea. Inflammatory bowel disease can also result in structural abnormalities, such as anal fissures, fistulas, abscesses, and operative complications that lead to fecal incontinence.

Table 28.3 Drugs and Dietary Items Associated with Diarrhea

Over-the-Counter Medications
Laxatives
Antacids (with magnesium)
Prescription Medications
Laxatives	Chemotherapy
Diuretics	Colchicine
Thyroid preparations	Cholestyramine
Cholinergics	Neomycin
Prostaglandins	Para-aminosalicylic acid
Dietary Items
Dietetic foods, candy or chewing gum, and elixirs with sorbitol, mannitol, or xylitol
Olestra
Caffeine
Ethanol
Monosodium glutamate

As with defecatory dysfunction, neurologic causes of fecal incontinence can be divided into central and peripheral disorders. Among the central nervous system disorders, upper motor neuron lesions above the level of the defecation center (located in the sacral cord) cause spastic bowel dysfunction. Cortical communication is disrupted, resulting in impaired cognitive control and sensory deficit. The anal sphincter is under spastic contraction, but digital stimulation can be performed to initiate reflex evacuation. Head trauma, neoplasms, and cerebral vascular accidents that damage portions of the frontal lobe result in loss of control of both micturition and defecation. Greater loss of inhibition is present when the lesion is located more anteriorly in the frontal lobe. Spinal cord trauma and lower motor neuron lesions above the defecation center tend to cause permanent loss of cortical control. For 2 to 4 weeks following spinal cord injury, “spinal shock” occurs, resulting in a temporary loss of reflexes below the level of the lesion, flaccid bowel function, constipation, and fecal impaction. After the initial shock, spastic paralysis ensues with hyperactive bowel function. The gastrocolic reflex, along with digital stimulation, initiates reflex evacuation in the absence of cortical inhibition. Fortunately, IAS tone is maintained despite the loss of EAS control for stress and urge situations. Both constipation and fecal incontinence can occur in these patients.

The demyelination that is seen in multiple sclerosis is randomly distributed and can occur at any level in the central nervous system. In addition to the somatic disruption that is similar to spinal cord injury, autonomic dysfunction frequently is present. People with dementia and other degenerative disorders that cause cognitive impairment frequently have fecal incontinence caused by overflow incontinence. Although sensory nerves are functioning properly, these individuals lack the cognitive awareness necessary to inhibit defecation until a socially acceptable time, and they develop overflow incontinence.

Lower motor neuron lesions occurring at or below the level of the defecation center in the sacral cord cause flaccid bowel dysfunction. Cortical communication is disrupted, resulting in impaired cognitive control and sensory deficit. The bowel reflexes, including the bulbocavernosus and anal reflexes, are interrupted. The anal sphincter is flaccid, and fecal retention with overflow incontinence usually occurs. Digital disimpaction and Valsalva often are required for evacuation. Digital stimulation has no effect, and medications tend to work poorly. Examples of motor neuron lesions include tumor or trauma to the cauda equina, tabes dorsalis, spina bifida, and peripheral neuropathy.

The classic example of peripheral neuropathy is congenital aganglionosis (Hirschsprung disease), which was discussed earlier. The most common peripheral neuropathy occurs with diabetes. Approximately 20% of individuals with diabetes have fecal incontinence (29). The cause tends to be multifactorial with the exact mechanism uncertain. Fecal incontinence can occur with diabetic diarrhea or years later from progressive disease. Individuals with diabetes frequently experience intestinal autonomic neuropathy, an abnormal gastrocolic reflex, and chronic constipation. The subsequent pelvic floor denervation causes fecal incontinence by sensory neuropathy, failure of the rectoanal inhibitory reflex, and sphincter dysfunction (30). Consequently, fecal incontinence from peripheral neuropathy can be the result of defective sampling, a disrupted rectoanal inhibitory reflex, or pudendal neuropathy with sphincter dysfunction. Patients may experience stress or urge incontinence as well as overflow incontinence.

Anatomic and structural causes of fecal incontinence are usually due to obstetric or surgical trauma. Damage or dysfunction of the IAS, EAS, and puborectalis can result in varying degrees of fecal incontinence. Those with impaired resting tone from a defective IAS will have passive incontinence (incontinence at rest), which is worse during sleep because of decreased EAS activity (31). An inability to respond to sudden distention and to suppress defecation is often seen with external sphincter dysfunction. External and internal sphincter dysfunction often causes incontinence of liquid stool. Incontinence of solid stool is usually seen with widening of the anorectal angle from damage to the puborectalis muscles. Damage to the anal cushions usually causes minor soiling. Other anatomic and structural abnormalities associated with fecal incontinence include obstructive disorders such as pelvic organ prolapse, descending perineum syndrome, anismus, and intussusception; fistulas from diverticulitis, inflammatory bowel disease, cancer, or surgical trauma; and decreased rectal compliance from inflammatory bowel disease, cancer, and radiation. Decreased compliance results in higher intraluminal pressures with smaller volumes of stool, poor storage capacity, urgency, and incontinence (32).

Functional disorders associated with fecal incontinence include irritable bowel syndrome (diarrhea variant), functional diarrhea, decreased mobility, and decreased cognition.

Combined Disorders of Defecation and Fecal Incontinence

Several conditions have the potential to cause both defecatory dysfunction and fecal incontinence (Table 28.1). Most of these disorders cause combined symptoms through the development of fecal impaction followed by overflow incontinence. This situation can be seen with many of the neurologic conditions, pelvic outlet obstructive disorders, functional disorders of irritable bowel syndrome, decreased mobility, and decreased cognition. The cause of these symptoms is often multifactorial.

Structural versus Functional Disorders

Disordered Defecation

Disordered defecation can result from outlet obstruction or functional motility disorders.

Outlet Obstruction

Anismus/Rectosphincteric Dyssynergia

Anismus is otherwise known as rectosphincteric dyssynergia, pelvic floor dyssynergia, spastic floor syndrome, and paradoxical puborectalis syndrome. The anorectal angle narrows as a result of paradoxical contraction of the puborectalis and external anal sphincter during defecation. Frequent symptoms include dyschezia, straining, hard stools, incomplete emptying, and tenesmus. A recent prospective study of 120 patients with dyssynergic defecation found a higher prevalence in women (77%) (33,34). The need for digital assistance (digital disimpaction or splinting) to evacuate the rectum occurs in up to 58% of patients. Psychosocial factors, such as a history of sexual abuse, depression, eating disorder, obsessive-compulsive disorder, and stress, may play an important role in this disease. In this study, 22% reported a history of sexual abuse, and 31% reported a history of physical abuse. One-third believed the problem began during childhood, and 24% reported a precipitating illness or surgery was related to a particular event. Five percent of women claimed that pregnancy or childbirth was a precipitating factor. This condition also is seen in young children with constipation and dyschezia. The response to biofeedback and pelvic floor physical therapy, as well as the aforementioned patient characteristics, indicate a learned response mechanism is involved (33,34). Although this is often categorized as an outlet obstruction, the Rome III criteria for functional gastrointestinal disorders places this in the category of functional defecation disorders. The specific Rome III diagnostic criteria for dyssynergic defecation includes “inappropriate contraction of the pelvic floor or less than 20% relaxation of basal resting sphincter pressure with adequate propulsive forces during attempted defecation” (35).

Pelvic Organ Prolapse

Pelvic organ prolapse bears special mention because it is often seen by gynecologists but inconsistently associated with defecatory dysfunction. Prolapse is very common, although many women with this condition are asymptomatic. Those with symptoms may report incomplete evacuation and the need to apply digital pressure to the posterior vaginal wall or perineum to aid in evacuation of stool (digitation or splinting). It is important to rule out other causes of constipation, because these symptoms are nonspecific, and rectocele can result from chronic straining and increased intra-abdominal pressure due to other etiologies of defecatory dysfunction. Defecatory dysfunction related to pelvic organ prolapse can result from rectocele, enterocele, or perineal descent, either individually or in combination.

Rectocele is a herniation of the rectal mucosa through a defect in the rectovaginal septum. These site-specific defects can be transverse or longitudinal through the inferior, middle, or superior regions of the rectovaginal septum (36). Enterocele is a herniation of a peritoneal sac and bowel through the pelvic floor, typically between the uterus or vaginal cuff and rectum. It is more common following hysterectomy and retropubic urethropexy.There are two theories surrounding the formation of an enterocele. The first theory implicates a defect in the fibromuscular endopelvic fascia of the vagina, allowing peritoneum and bowel to herniate. The second theory attributes its formation to a support defect with full thickness protrusion, including endopelvic fascia (37). Ultimately, the mechanism might be attributed to a combination of the two theories because some support defects are secondary to superior breaks in the rectovaginal and pubocervical fascia. Patients with rectocele and enterocele may have similar symptoms, including pelvic pressure, vaginal protrusion, obstipation, fecal incontinence, and sexual dysfunction. Although associations have been made between defecatory dysfunction and advanced stages of pelvic organ prolapse, a causal relationship remains to be established. Controversy remains as to whether anatomic herniation is the cause of these symptoms or the effect of underlying colonic dysfunction, chronic constipation, and straining.

Descending perineum syndrome is defined as descent of the perineum (at the level of the anal verge) beyond the ischial tuberosities during Valsalva. Excessive perineal descent was first described in the colorectal literature by Parks et al. in 1966 (38,39). It occurs as a result of inferior detachment of the rectovaginal septum from the perineal body. As the condition progresses, the patient can develop pudendal neuropathy from stretch injury. Perineal descent has been associated with a variety of defecatory disorders, including constipation, fecal incontinence, rectal pain, solitary rectal ulcer syndrome, rectocele, and enterocele (40).

Rectal Intussusception

Rectal intussusception or intrarectal prolapse is the circumferential prolapse of the upper rectal wall into the rectal ampulla but not through the anal verge. It occurs most often in women in their fourth and fifth decades. The most common symptoms are obstructive, including incomplete emptying, manual disimpaction, splinting, pain with defecation, and bleeding. Other symptoms include fecal incontinence, decreased urge to defecate, inability to distinguish between gas and feces, and mucus discharge with pruritus ani. Bleeding often originates from a solitary rectal ulcer or localized proctitis of the involved bowel segment (41). Intussusception is seen in as many as one-third of women with defecatory dysfunction and other symptoms, such as constipation, rectal pain, and fecal incontinence(42). It has also been seen in 29% of asymptomatic patients (43). The intussusception rarely develops into total rectal prolapse (44).

Functional Motility Disorders

Functional Bowel Disorders

Functional bowel disorders, as defined by the Rome III criteria, consist of irritable bowel syndrome, functional bloating, functional constipation, functional diarrhea, and unspecified functional bowel disorders. In this section we will focus primarily on irritable bowel syndrome (45).

Irritable bowel syndrome (IBS) has been estimated to have a prevalence of 10% to 20% and is more common in women and younger individuals. It accounts for 25% to 50% of all referrals to gastrointestinal clinics.Irritable bowel syndrome has distinct diagnostic criteria, including the exclusion of structural or metabolic abnormalities. These patients often have other gastrointestinal, genitourinary, and psychological illness, including gastroesophageal reflux disease, fibromyalgia, headache, backache, chronic pelvic pain, sexual dysfunction, lower urinary tract dysfunction, depression, and anxiety. Stressful life events seem to correlate with the onset and exacerbation of symptoms. A detailed history frequently reveals past physical or sexual abuse (46). Currently, specific criteria allow for classification of IBS into diarrhea-, constipation-, and pain-predominant categories (Table 28.4). The constipation variant is most commonly associated with defecatory dysfunction, whereas the diarrhea variant causes fecal incontinence. The pain or spastic variant causes predominantly abdominal discomfort but can also be associated with both defecatory dysfunction and fecal incontinence. After excluding organic disease, the criteria listed in Table 28.4 have a sensitivity of 65%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 76% (47).

Table 28.4 Irritable Bowel Syndrome

Diagnostic Criteriona

Recurrent abdominal pain or discomfortb at least 3 days per month in the last 3 months associated with two or more of the following:

1. Improved with defecation 2. Onset associated with a change in frequency of stool 3. Onset associated with a change in form (appearance) of stool

aCriterion fulfilled for the last 3 months with symptom onset at least 6 months prior to diagnosis. b“Discomfort” means an uncomfortable sensation not described as pain.

In pathophysiology research and clinical trials, a pain/discomfort frequency of at least 2 days a week during the screening evaluation is recommended for subject eligibility.

From Drossman DA, Corazziari E, Talley NJ, et al., eds. Rome III: the functional gastrointestinal disorders. 3nd ed. McLean, VA: Degnon Associates, 2006:885–897, Appendix A, with permission.

Functional constipation is a term created by the Rome II criteria as a unifying definition of constipation (Table 28.5). The rationale for the criteria listed in Table 28.5 stems from the variability in patient definitions of constipation (46).

Table 28.5 Functional Constipation

Diagnostic Criteriaa*

1. Must include two or more of the following:   a. Straining during at least 25% of defecations   b. Lumpy or hard stools in at least 25% of defecations   c. Sensation of incomplete evacuation for at least 25% of defecations   d. Sensation of anorectal obstruction/blockage for at least 25% of defecations   e. Manual maneuvers to facilitate at least 25% of defecations (e.g., digital evacuation, support of the pelvic floor)   f. Fewer than three defecations per week 2. Loose stools are rarely present without the use of laxatives, and there are insufficient criteria for IBS. 3. Insufficient criteria for irritable bowel syndrome

aCriteria fulfilled for the last 3 months with symptom onset at least 6 months prior to diagnosis

From Drossman DA, Corazziari E, Talley NJ, et al., eds. Rome III: the functional gastrointestinal disorders. 3nd ed. McLean, VA: Degnon Associates, 2006:885–897, Appendix A, with permission.

Functional Defecation Disorders

Functional defecation disorders are divided into dyssynergic defecation and inadequate defecatory propulsion (colonic inertia). (For the purposes of this chapter, dyssynergic defecation has been included in the structural category of outlet obstruction; however, it is important to recognize that Rome III considers it a functional disorder.) Both of the functional defecation disorders require the presence of functional constipation. Table 28.6lists the criteria for diagnosing these conditions.

Table 28.6 Functional Defecation Disorders

Diagnostic Criteriaa

1. The patient must satisfy diagnostic criteria for functional constipation (Table 28.5) 2. During repeated attempts to defecate must have at least two of the following:   a. Evidence of impaired evacuation, based on balloon expulsion test or imaging   b. Inappropriate contraction of the pelvic floor muscles (i.e., anal sphincter or puborectalis) or less than 20% relaxation of basal resting sphincter pressure by manometry, imaging, or EMG   c. Inadequate propulsive forces assessed by manometry or imaging

aCriteria fulfilled for the last 3 months with symptom onset at least 6 months prior to diagnosis

From Drossman DA, Corazziari E, Talley NJ, et al., eds. Rome III: the functional gastrointestinal disorders. 3nd ed. McLean, VA: Degnon Associates, 2006:885–897, Appendix A, with permission.

Colonic Inertia/Slow-Transit Constipation

Severe constipation, defined as fewer than three stools per week and refractory to therapy, is relatively rare; however, these patients frequently suffer from motility disorders such as global motility disorder and colonic inertia. Women are more likely to be affected than men. Colonic inertia or slow-transit constipation is defined as the delayed passage of radiopaque markers through the proximal colon without retropulsion of markers from the left colon and in the absence of systemic or obstructive disorders. The cause remains unclear. Patients with this disorder have impaired phasic colonic motor activity and diminished gastrocolic reflexes (48,49). Studies on the role of laxatives, absorption, hormones, psychological abnormalities, and endogenous opioids have been inconclusive. Current literature suggests a possible neurologic or smooth muscle disorder (49,50).

Fecal Incontinence

Sphincter Disruption

In young women, obstetric injury is the most common cause of fecal incontinence. The mechanism of injury can be from anatomic disruption of the anal sphincter complex, pelvic floor denervation, or a combination of the two conditions. The risk factors for anal sphincter laceration are primiparity, high birth weight, forceps delivery, and episiotomy (51–53). Recent work suggests that women with anal sphincter injuries have slower labor, without the normal deceleration phase, and with late descent of the fetal head (54). Although there are limited long-term prospective studies demonstrating the natural history of anal sphincter injury, pelvic floor neuropathy, and the progression of these conditions to fecal incontinence, current literature supports the relationship of early-onset symptoms to sphincter damage and delayed-onset symptoms to neuropathy (55). This relationship would account for the large discrepancy in the prevalence of fecal incontinence between younger men and women that decreases as the population ages (56).

Obstetric Trauma

Third- and fourth-degree lacerations at delivery are associated with an increased risk of fecal incontinence (odds ratio [OR] 3.09)[MB2] (55). Whereas the incidence of clinically documented third- and fourth-degree anal sphincter tears is between 0.5% and 5.9% (51,53,57), occult third- and fourth-degree defects are present in 28% to 35% of primiparous women and 44% of multiparous women, and approximately one-third of these patients have symptoms of anal incontinence. Patients with occult anal sphincter tears are 8.8 times more likely to have fecal incontinence (53,58). Forceps-assisted vaginal delivery significantly increases this risk, but the data on vacuum-assisted delivery are less conclusive (52,59,60). Elective cesarean delivery, in contrast with emergency cesarean delivery, was believed to prevent anal incontinence, but recent studies argue against any protective effect with cesarean delivery, irrespective of timing (46,51,53,59,61,62). A recent Cochrane review concludes that there is insufficient evidence to support primary elective cesarean delivery for the purpose of preserving fecal continence (63). Midline episiotomy is strongly linked to sphincter damage and fecal incontinence(52,64). One study of a large population found conflicting results, with an overall protective effect seen with episiotomy (OR 0.89). The likelihood of fourth-degree laceration was increased (OR 1.12) and of third-degree laceration was decreased (OR 0.81) (51). A Cochrane review supports the restrictive use of both midline and mediolateral episiotomy due to less posterior perineal trauma, less suturing, and fewer healing complications. There were no differences in severe trauma, pain, dyspareunia, or urinary incontinence, but there was an increase in anterior perineal trauma with restrictive use (65). An important finding in another study was that one-half of patients who underwent immediate repair of a third-degree laceration had symptoms of anal incontinence, and 85% had persistent sphincter defects on endoanal ultrasonography (66).

Surgical Trauma

Iatrogenic injury follows obstetric trauma as the second most common cause of direct sphincter damage. Surgical procedures that have been associated with fecal incontinence include anal fistula repair, anal sphincterotomy, hemorrhoidectomy, and anal dilation. Fistulotomy is the most common procedure that results in fecal incontinence. Rectovaginal or anovaginal fistulas can develop after obstetric injury, operative complications during pelvic surgery, and inflammatory bowel disease exacerbations. Fistulas cause fecal incontinence, and the degree of postoperative dysfunction depends on the location of the fistula and the amount of sphincter that is disrupted during the surgical repair. It also depends on the preoperative level of sphincter function and pudendal nerve function. Anal sphincterotomy to treat painful anal fissures can lead to incontinence by disruption of rectal sensory innervation and anal cushions and transection of the anal sphincter (67,68). Hemorrhoidectomy often results in minor soiling as a result of resection of the anal cushions, which act as the final mucosal barrier. Similar to sphincterotomy, rectal sensory innervation can be disrupted, and injury to the internal sphincter can occur during sharp dissection (68,69).

Sphincter Denervation

Idiopathic (primary neurogenic) fecal incontinence results from denervation of both the anal sphincter and pelvic floor muscles. Denervation injury related to obstetric trauma accounts for approximately three of four cases of idiopathic fecal incontinence and is the most common overall cause of fecal incontinence (70,71).

Obstetric Trauma

The two proposed mechanisms of pudendal neuropathy are stretch injury during the second stage of labor and compression of the nerve as it exits Alcock’s canal (70). Established risk factors for pelvic floor neuropathy include multiparity, high birth weight, forceps delivery, prolonged active second stage, and third-degree laceration (72,73). Several studies have shown increased pudendal nerve terminal motor latencies following vaginal delivery, especially after sphincter laceration (53,71,74). Most women will recover function within a few months postpartum. Others will have evidence of injury several years later, which may represent the cumulative effects of subsequent deliveries (71,75). However, fecal incontinence will develop in only a fraction of patients with neuropathy (73).

Descending Perineum Syndrome

As noted previously, prolonged straining for any reason could cause descending perineum syndrome. This syndrome is defined as descent of the perineum beyond the ischial tuberosities during Valsalva (38,39). Pudendal neuropathy results from stretching and entrapment of the pudendal nerve. This diagnosis is supported by findings of elongation of the pudendal nerve, prolonged pudendal nerve motor terminal latency, and decreased anal sensation in women with perineal descent (76–78). As pudendal neuropathy progresses, it ultimately leads to fecal incontinence (40,79).

Functional Bowel Disorders

Functional Fecal Incontinence

The Rome III criteria established well-defined guidelines for functional causes of fecal incontinence (Table 28.7). The criteria essentially exclude systemic and anatomic abnormalities; however, minor abnormalities of sphincter innervation or structure are permitted.

Table 28.7 Functional Fecal Incontinence

Diagnostic Criteriaa

1. Recurrent uncontrolled passage of fecal material in an individual with a developmental age of at least 4 years and one or more of the following:   a. Abnormal functioning of normally innervated and structurally intact muscles   b. Minor abnormalities of sphincter structure and/or innervation   c. Normal or disordered bowel habits, (i.e., fecal retention or diarrhea)   d. Psychological causes  AND 2. Exclusion of all of the following   a. Abnormal innervation caused by lesion(s) within the brain (e.g., dementia), spinal cord, or sacral nerve roots, or mixed lesions (e.g., multiple sclerosis), or as part of a generalized peripheral or autonomic neuropathy (e.g., due to diabetes)   b. Anal sphincter abnormalities associated with a multisystem disease (e.g., scleroderma)   c. Structural or neurogenic abnormalities believed to be the major or primary cause of fecal incontinence.

aCriteria fulfilled for the last 3 months

From Drossman DA, Corazziari E, Talley NJ, et al., eds. Rome III: the functional gastrointestinal disorders. 3nd ed. McLean, VA: Degnon Associates, 2006:885–897, Appendix A, with permission.

Irritable Bowel Syndrome

The diarrhea variant of irritable bowel syndrome is often associated with fecal incontinence as well as disordered defecation. The criteria for diagnosis are presented in Table 28.4.

Functional Diarrhea

The Rome III criteria create a unifying definition of diarrhea called functional diarrhea (Table 28.8). The rationale for the criteria listed in Table 28.8 stems from the variability in patients' descriptions of diarrhea (46).

Table 28.8 Functional Diarrhea

Diagnostic Criteriona

Loose (mushy) or watery stools without pain occurring in at least 75% of stools

aCriteria fulfilled for the last 3 months with symptom onset at least 6 months prior to diagnosis

From Drossman DA, Corazziari E, Talley NJ, et al., eds. Rome III: the functional gastrointestinal disorders. 3nd ed. McLean, VA: Degnon Associates, 2006:885–897, Appendix A, with permission.

Pitfalls for the Pelvic Floor Surgeon

It sometimes is easy to overlook or misinterpret signs and symptoms of constipation and defecatory dysfunction. Any acute change in bowel habits must be evaluated thoroughly, and malignancy must be considered in the differential diagnosis. Even in the presence of chronic disease, malignancy must still be excluded. Persistent symptoms after an empiric trial of medical therapy should prompt further evaluation, including colonoscopy or flexible sigmoidoscopy. It is also possible to mistakenly attribute symptoms of defecatory dysfunction and constipation to pelvic organ prolapse when prolapse is actually the result of an underlying bowel disorder. In this case, surgical treatment of prolapse will have little lasting benefit if the underlying bowel disorder remains untreated.

History and Physical Examination

History

A thorough history and physical examination are critical to the evaluation of fecal incontinence and defecatory dysfunction. The history of present illness should focus on the bowel habits, including frequency and consistency of bowel movements (hard vs. soft, formed vs. unformed, diarrhea vs. constipation). Determining the duration and severity of symptoms, as well as exacerbating factors, is important for understanding the impact on quality of life. Patients should be questioned about straining with bowel movements, symptoms of incomplete emptying, and splinting of the perianal region, perineal body, or posterior vaginal wall to assist with evacuation. Patients should also be asked about the need to perform digital disimpaction because they are unlikely to volunteer this information. With respect to fecal incontinence, information should be obtained about leakage with solids, liquid, and flatus and the ability to discriminate between these different types of stool (sampling). Similar to urinary incontinence, fecal incontinence can be stress related, urge related, or unconscious. Questions about alternating diarrhea and constipation, mucus or blood in the stools, constitutional symptoms, and changes in stool caliber can help the investigator uncover systemic and functional etiologies. Finally, it is important to ask about adaptive behaviors, incontinence product usage, and past and present treatments, including surgery, physical therapy, and medications.

A large amount of information can be obtained efficiently through questionnaires. Validated questionnaires quantify symptoms, which are subjective in nature, to objectively measure response to treatment. A valuable survey to assess defecatory dysfunction is the Colorectal-Anal Distress Inventory (CRADI), which has been incorporated into the Pelvic Floor Distress Inventory (PFDI) (80). The latter is a useful tool for evaluating symptoms of prolapse, urinary incontinence, fecal incontinence, voiding dysfunction, and defecatory dysfunction. Other useful symptom scales and bother scores for fecal incontinence include the Wexner Score, Fecal Incontinence Severity Index, and Fecal Incontinence Quality of Life Scale (81–83).

The medical history, surgical history, family history, and review of systems should focus on uncovering potential systemic and obstructive disorders shown in Table 28.1. A complete obstetric history should include the number of vaginal deliveries, operative vaginal deliveries, or presence of a third- or fourth-degree laceration, which is critical for patients with fecal incontinence. Length of the second stage of labor, birth weight, and the use of episiotomy should be ascertained because they may pose risk factors for sphincter damage and denervation. The sexual history should include questions about rape, anal intercourse, and dyspareunia. Use of over-the-counter, prescription, and illegal drugs should be recorded as well as food allergies.

Physical Examination

The evaluation of anorectal dysfunction requires a basic general examination as well as a focused abdominal and pelvic examination. The general physical survey should include a global assessment of mobility and cognitive function. Routine examination of the abdomen involves inspection, palpation, and auscultation to rule out the presence of masses, organomegaly, and areas of peritoneal irritation. This examination should be followed by a detailed evaluation of the vagina, perineum, and anorectum. The goals of the pelvic examination are to define objectively the degree of prolapse and determine the integrity of the connective tissue, neurologic function, and muscular support of the pelvic organs.

Neurologic Examination

Important elements of the neurologic examination are assessment of cranial nerve function, sensation and strength of the lower extremities, and reflexes for the lower extremities, bulbocavernosus, and anal wink. These examinations evaluate the function of the lower lumbar and sacral nerve roots, recognizing the importance of the second through fourth sacral nerve roots in pelvic floor dysfunction. The perineal reflexes can be elicited by stroking the labia majora and perianal skin or tapping the clitoris with a cotton-tipped swab. The anal wink, bulbocavernosus, and cough reflexes all test the integrity of motor innervation to the external anal sphincter (S2–4). Sensation over the inner thigh, vulva, and perirectal areas should be tested for symmetry to light touch and pinprick.

Muscle Strength

The integrity of the pelvic floor muscles should be assessed at rest and with voluntary contraction to determine strength, duration, and anterior lift. The ability to relax these muscles and tenderness on palpation should also be evaluated. Several standardized systems have been described to objectively measure muscle strength, but none has been accepted as a standard. The puborectalis muscle should be readily palpable posteriorly as it creates a 90-degree angle between the anal and rectal canals. Voluntary contraction of this muscle “lifts” the examining finger anteriorly toward the pubic rami. An intact external anal sphincter muscle that has decreased tone and contractility often indicates pudendal neuropathy. Similarly, neuropathy affecting the puborectalis can be recognized by an obtuse anorectal angle and weak voluntary contraction. Similar to the urethral axis, the anorectal angle can also be tested using a cotton-tipped swab, although this test is rarely performed. Deflection is measured in the supine position at rest, with strain, and with squeeze.

Vaginal Support

The salient points of pelvic organ prolapse (see Chapter 27 for patients with defecatory dysfunction are the support of the vaginal apex, posterior wall, and perineal body, although some experts believe anterior wall defects can also affect defecatory dysfunction. The posterior wall is assessed while supporting the vaginal apex and anterior wall with a Sims speculum. This permits the examiner to focus on identifying specific locations of rectovaginal fascial defects. A rectovaginal examination aids in identification of defects in the rectovaginal fascia or perineal body. Loss of vaginal rugation has also been reported overlying the site of a rectovaginal fascial tear (84). This technique is especially useful for enteroceles, which have a smooth, thin epithelium over the enterocele sac or peritoneum.

Normally, the perineum should be located at the level of the ischial tuberosities, or within 2 cm of this landmark. A perineum below this level, either at rest or with straining, represents perineal descent. Subjective findings of perineal descent include widening of the genital hiatus and perineal body, as well as a flattening of or a convex appearance of the intergluteal sulcus. Women with perineal descent also tend to have less severe stages of pelvic organ prolapse based on the Pelvic Organ Prolapse Quantification (POP-Q) staging system because it measures descent from the hymenal ring (85). An increase in the length of the perineal body and genital hiatus consistent with straining suggests perineal descent. The degree of perineal descent can also be measured objectively with a St. Mark’s perineometer, although a thin ruler placed in the posterior introitus at the level of the ischial tuberosities also can be used. Descent is measured as the distance the perineal body moves when the patient strains. Although pelvic floor fluoroscopy is the standard technique for measuring perineal descent, this technique is most useful in patients with symptoms of severe defecatory dysfunction and evidence of perineal descent on pelvic examination.

Anorectal Examination

Visual and digital inspection of the vagina and anus will help to identify structural abnormalities such as prolapse, fistulas, fissures, hemorrhoids, or prior trauma. As previously mentioned, a rectovaginal examination provides useful information regarding the integrity of the rectovaginal septum and can demonstrate laxity in the support of the perineal body. The rectovaginal examination is helpful in the diagnosis of enteroceles, which can be felt as protrusion of bowel between the vaginal and rectal fingers with straining. Digital rectal examination should be performed at rest, with squeeze, and while straining. The presence of fecal material in the anal canal may suggest fecal impaction or neuromuscular weakness of the anal continence mechanism. Circumferential protrusion of the upper rectum around the examining finger during straining suggests intussusception, which often occurs in combination with laxity of the posterior rectal support along the sacrum.

The integrity of the external anal sphincter and puborectalis muscle can be evaluated by observation and palpation of these structures during voluntary contraction. Evidence of dovetailing of the perianal skin folds and the presence of a perineal scar with an asymmetric contraction often indicates a sphincter defect. When a patient is asked to contract her pelvic floor muscles, two motions should be present: The external anal sphincter should contract concentrically, and the anal verge should be pulled inward. These actions should also be apparent on digital rectal examination. As mentioned previously, the 90-degree angle created by the puborectalis should be readily palpable posteriorly and, with voluntary contraction, the examining finger should be lifted anteriorly toward the pubic rami. Both the puborectalis and external anal sphincter should relax during Valsalva effort. Patients with anismus may experience a paradoxical contraction of these muscles during straining. Finally, defects in the anterior aspects of the external anal sphincter may be detected by digital examination.

Testing

Sophisticated diagnostic testing is currently being used in clinical research and in anorectal physiology laboratories to quantify the function of the colon and anorectum. Following is a description of these techniques as they relate to the management of fecal incontinence and disordered defecation.

Fecal Incontinence

Endoanal Ultrasonography

Endoanal ultrasonography permits accurate imaging of both the internal and external anal sphincters. It can assess the continuity and thickness of the muscle and currently is considered the single best method for detecting anal sphincter defects. Endoanal ultrasonography is performed using a Bruel-Kjaer (Copenhagen, Denmark) ultrasound scanner with a 360-degree rectal endoprobe (type 1850) with a 7.0 MHz transducer (focal length, 2–5 cm) housed within a plastic cone (Fig. 28.1). The normal IAS is a continuous hypoechoic band of smooth muscle surrounded by the thick echogenic layer of the striated EAS. A sphincter defect occurs when there is disruption in these muscle bands. Location and severity of the defect can be described by circumferential distance in degrees, percentage of thickness, and distance from the anal verge (Fig. 28.2). Measurements are usually taken in the proximal, middle, and distal anal canal. It is important to recognize the physiologic split in the proximal EAS as it merges with the puborectalis muscle of the levator ani. Misinterpretation of this finding as a sphincter defect can result in an increased prevalence of reported defects. The puborectalis muscle appears as a U-shaped or V-shaped thick echogenic layer outside the IAS in the proximal anal canal. Magnetic resonance imaging (MRI) may be equally as effective or better at diagnosing sphincter defects, especially with the use of a vaginal or rectal coil. For this purpose, MRI is more expensive, and currently its use is largely investigational. It may be beneficial in cases in which endoanal ultrasonography results are inconclusive or the quality of the study is poor.

Figure 28.1 Bruel-Kjaer (Copenhagen, Denmark) ultrasound probe (type 1850) with a 7.0 MHz transducer (focal length, 2 to 5 cm) housed with a plastic cone.

Figure 28.2 A: Endoanal ultrasound image from the distal anal canal demonstrating defects in the internal sphincter from 10 to 3 o'clock and the external sphincter from 10 to 2 o'clock. B: Endoanal ultrasound image from the middle anal canal demonstrating defects in the internal sphincter from 12 to 2 o'clock and the external sphincter from 10 to 1 o'clock. C: Endoanal ultrasound image from the proximal anal canal demonstrating an intact IAS and a normal physiologic split in the external sphincter.

Electromyography

Electromyography (EMG) is used to evaluate neuromuscular integrity of the EAS following a traumatic injury such as childbirth, as well as to document the presence of pelvic floor neuropathy(86). This technique measures the electrical activity arising in muscle fibers during contraction and at rest. Different types of electrodes may be employed, including surface electrodes, concentric needle electrodes, and single-fiber electrodes. Surface electrodes are less invasive because they are applied near or within the anal canal, but they are capable only of recording basic anal sphincter activity. This technique often is used in conjunction with biofeedback therapy. Concentric needle electrodes are most commonly used in anorectal physiology laboratories to selectively survey an individual muscle’s activity. Insertion of the thin needlelike cannulas containing steel wire electrodes can be painful. Even smaller single-fiber EMG electrodes are used to record the activity of single muscle fibers, which can be quantified to calculate fiber density. Following denervation injury, increased muscle fiber density occurs during reinnervation. Thus, single-fiber EMG can provide indirect evidence of neurologic injury by mapping the EAS and identifying injured areas. This technique is used rarely in clinical practice. Endoanal ultrasonography offers increased patient comfort and more reliable results than EMG and has replaced this technique for the detection of EAS disruption because of increased patient comfort and more reliable results.

Motor nerve conduction studies provide another means of measuring pelvic floor neuropathy. The axon of a nerve is stimulated, and the time it takes the action potential to reach the muscle supplied by the nerve is recorded. The delay between stimulation and the muscle response is called the nerve latency. Pudendal nerve terminal motor latency (PNTML) can be determined by transrectal stimulation of the pudendal nerve using a St. Mark’s electrode (87). A nerve stimulator is mounted on an examination glove at the fingertip (Fig. 28.3) and positioned transrectally over each ischial spine. A stimulus of up to 50 mV for 0.1 milliseconds is applied, and the latency of the EAS muscle contraction is measured. A value of 2.2 milliseconds or less is considered normal. A recent study evaluating normative values for pudendal and perineal nerve latencies observed increased latencies with increased age (88). Prolongation of the PNTML is indicative of damage to that nerve or the presence of a demyelinating condition. Pudendal nerve function has prognostic value in the surgical repair of traumatic sphincter injuries and is useful in preoperative counseling (89).

Figure 28.3 St. Mark’s electrode used for measuring pudendal nerve motor terminal latency. The stimulating electrode is on the fingertip, and the receiving electrode is on the proximal finger near the knuckle.

Anal Manometry

Anal manometry is used to quantify function of the anal sphincter mechanism. Water-perfused manometry catheters or water-filled balloons are most often used to measure anal canal pressures. Resting anal canal pressures reflect IAS function, and pressures in the lower anal canal during maximal voluntary contraction reflect EAS function. Vector analysis can be used to detect asymmetry within the anal sphincter. Anal manometry provides indirect evidence of sphincter injury; low resting tone indicates IAS injury, and decreased maximum squeeze pressure indicates EAS injury. Anal pressures are influenced by a variety of factors, including tissue compliance and muscular tone. Consequently, anal manometry results are difficult to interpret and correlate poorly with the specific anatomic defect. Interpretation is further complicated by the wide variation of normal pressure values that change with age and parity. Significant overlaps occur between manometric values for incontinent patients and those without incontinence. Thus, anal manometry may be of limited value in the evaluation and treatment of anal sphincter defects and fecal incontinence.

Proctoscopy and Flat Tire Test

Proctoscopy has an important role in the evaluation of fecal incontinence. It can be performed independently or during colonoscopy, flexible sigmoidoscopy, and the flat tire test. Proctoscopy can detect anorectal pathology, such as prolapsing hemorrhoids, intussusception, ulcerative or radiation proctitis, or a solitary rectal ulcer. The flat tire test is added when a rectovaginal or colovaginal fistula is suspected but cannot be visualized on routine office evaluation. This test usually is performed under anesthesia but can also be done in the office setting. Saline or water is placed in the vagina with the patient in Trendelenburg position. Using a proctoscope or rigid sigmoidoscope, air is instilled into the rectum. Vaginal retractors provide visualization of the posterior vaginal epithelium and vaginal apex. Observation of bubbling into the vaginal fluid confirms the diagnosis and location of a rectovaginal or colovaginal fistula. The rectal site of the fistula usually is identifiable, depending on the size and location of the fistula as well as the quality of the bowel preparation.

Disordered Defecation

Sitzmark Study

Colonic transit studies are performed using ingested radiopaque markers followed by serial abdominal radiography. Patients are asked to follow a high-fiber diet over the test period and avoid the use of laxatives, suppositories, or enemas. A capsule containing 20 to 24 markers is ingested initially, and abdominal radiography is performed either daily or on the fourth day, the seventh day, and every 3 days thereafter until all the markers are gone. Segmental transit times are then calculated using a mathematical formula. Colonic transit study results are used to classify patients with constipation into delayed transit, normal transit, and outlet obstruction. After day 6, there should be fewer than five markers remaining in the colon. With slow transit, more than five markers are scattered throughout the colon. With outlet obstruction, more than five markers are in the rectosigmoid region, and transit is normal throughout the rest of the colon.

Pelvic Floor Fluoroscopy and Magnetic Resonance Imaging

Pelvic fluoroscopy permits radiological evaluation of pelvic floor and anorectal anatomy and physiology. It is particularly useful in obstructive defecation disorders, such as intussusception, rectocele, enterocele, anismus, and perineal descent. The patient is placed on a radiolucent commode, and contrast material is instilled into the rectum. The addition of vaginal, bladder, and oral contrast material is helpful diagnostically when multicompartmental prolapse is suspected. A series of lateral still images or continuous imaging using videography are made with fluoroscopy while the patient is at rest, during defecation, and with contraction of the anal sphincter. Similar films can be obtained for evacuation of the bladder. Pelvic fluoroscopy has many names, including defecography, defecating proctography, defecating cystoproctography, and colpocystoproctography, depending on the technique used. The measurements obtained include size of the rectal ampulla, length of the anal canal, anorectal angle, puborectalis motion, and pelvic floor descent. Severity of prolapse and pelvic floor descent is quantified in relation to the pubococcygeal line. Pelvic fluoroscopy is superior to physical examination for diagnosing enterocele, and this technique has the advantage of being able to distinguish enteroceles from sigmoidoceles (90). Rectosphincteric dyssynergia may be present when the patient experiences incomplete relaxation of the puborectalis muscle during rectal evacuation, the anorectal angle is preserved, and there is incomplete emptying. Pelvic fluoroscopy is considered the definitive test for diagnosing intussusception, and it is the preferred technique for quantifying perineal descent (91).

Dynamic MRI with luminal contrast is an imaging modality similar to pelvic fluoroscopy. Its ability to detect prolapse is similar to that of fluoroscopy, but MRI can visualize pelvic floor musculature and soft tissue, thus giving it the advantage of detecting ballooning of the levator muscles and levator ani hernias. The supine position of the testing is a drawback; however, there are isolated reports of upright dynamic MRI using open scanners that show results comparable to fluoroscopy for detection of anorectal pathology (92). Fluoroscopy and dynamic MRI can be used in situations involving severe multicompartmental prolapse or in which the severity of the symptoms is disproportionate to examination findings.

Anal Manometry

Anal manometry is used to determine maximum resting pressure, maximum squeeze pressure, rectal sensation and compliance, as well as the presence of an intact rectoanal inhibitory reflex. With disordered defecation, it can be used to diagnose Hirschsprung disease and anismus. The addition of surface EMG to document relaxation helps exclude anismus as a cause of obstructed defecation. Failure of the anal sphincter to relax with defecation and increased electrical activity of the EAS and puborectalis are seen in patients with anismus. In contrast, there should be no increase in the electrical activity measured by surface electrodes for patients with Hirschsprung disease. A rectal balloon expulsion test can also be of assistance in the evaluation of rectal emptying and may be valuable during physiotherapy for diagnosing dyssynergic defecation.

Colonoscopy and Proctoscopy

Standard gastrointestinal evaluation for patients with symptoms of disordered defecation should include a barium enema or colonoscopy to eliminate the possibility of colorectal malignancy. Proctoscopy should be included as part of the routine examination because it may reveal anorectal pathology.

Therapeutic Approach to Fecal Incontinence

Treatment of fecal incontinence should first focus on nonsurgical options, including dietary modification, medical therapy, and biofeedback. Any underlying systemic conditions or gastrointestinal disorders should be treated before initiating an extensive evaluation for other causes of fecal incontinence. If symptoms persist, further investigation should be undertaken. If the evaluation discloses an underlying EAS defect and conservative therapy has been unsuccessful, it is reasonable to proceed with surgical treatment.

Following is an overview of treatment options and the efficacy of each approach. The lack of consistent outcome measures makes it difficult to compare efficacy among treatments. Some studies base success on strict conformity with criteria for continence, but the results vary for continence of flatus, liquid, or solid stool. Other studies base success on more subjective criteria, such as improvement following treatment. Daily diaries can be maintained, but the results may be unreliable. Even if a validated symptom survey and quality-of-life scale are employed, few studies use the same outcome measure.

Nonsurgical Treatment

Nonsurgical management focuses on maximizing the continence mechanism through alteration of stool characteristics or behavioral modification. Stool consistency and volume can be manipulated by dietary and pharmacologic means to achieve passage of one to two well-formed stools per day. The rationale for this approach is that formed stool is easier to control than liquid stool. Additionally, behavior modification can be employed using bowel regimens that focus on the predictable elimination of feces. Physical therapy and biofeedback can also be useful for strengthening the continence mechanism.

Pharmacologic Approaches

Dietary Modification and Fiber

Dietary modification for treatment of fecal incontinence frequently involves avoidance of foods that precipitate loose stools and diarrhea. Common dietary irritants include spicy foods, coffee and other caffeinated beverages, beer and alcohol, and citrus fruits. Avoidance of dairy products or the addition of lactase dietary supplements is essential for those with lactose intolerance. The addition of fiber may improve fecal incontinence by functioning as a stool bulking agent to increase volume and density. The average individual in the United States consumes less than half the recommended daily fiber intake (25–35 g). Various fiber sources are listed in Table 28.9, with the highest content found in high-fiber cereals. It is difficult to consume the recommended daily amount from diet alone, and fiber supplements often are required. Although the increased stool volume and density helps many individuals maintain continence, excessive fiber with inadequate fluid intake may predispose elderly patients to fecal impaction.

Table 28.9 Fiber Sources

Constipating Agents

Constipating agents have the most value in patients with chronic loose stools or diarrhea. They can also help improve symptoms in patients with fecal frequency and urgency. Loperamide (Imodium) and diphenoxylate hydrochloride with atropine (Lomotil) are the most commonly used agents. Loperamide has been shown to prolong transit time and stimulate anal sphincter function. With either of these agents, careful titration is recommended to prevent the primary side effect of constipation. It is generally preferable to begin using 2 to 4 mg of loperamide daily and then titrate up to 4 mg three to four times per day. A 4-mg dose before meals has been shown to increase anal tone and improve continence (93). Lomotil is started at a dose of one to two tablets every day or every other day and titrated up to one to two tablets three to four times a day as needed. Caution should be exercised for patients taking other anticholinergic medications. Anticholinergic side effects include dry mouth, drowsiness, lightheadedness, and tachycardia. Codeine can also be used as a constipating agent. It should be used judiciously in those with chronic disorders and in elderly patients because of side effects common to narcotics, including addiction with prolonged usage and central nervous system and respiratory depression. A study of 82 geriatric patients documented the efficacy of pharmacologic treatment for fecal incontinence (94). Patients were treated based on the underlying cause. Those with fecal impaction received lactulose and enemas, whereas those with neurogenic fecal incontinence received codeine phosphate as a constipating agent and enemas. The rate of cure for fecal incontinence was 60% in the treatment group versus 32% for controls (P <.001).

Medications for Irritable Bowel Syndrome

Dietary treatment of IBS consists of avoiding foods that are associated with symptoms, including alcohol, caffeine, sorbitol, and foods that increase gas production. Although increased dietary fiber or fiber supplementation has been shown to improve the constipation-predominant form of this illness, fiber supplementation has little effect on the diarrhea variant associated with fecal incontinence. Pharmacologic therapy is directed toward the predominant symptom. Loperamide and Lomotil tend to be useful first-line agents for treating diarrhea. Tricyclic antidepressants improve abdominal discomfort and are also valuable in diarrhea-predominant patients because of their constipating effect. The serotonin type 3 (5HT3) antagonist alosetron (Lotronex) has been approved by the U.S. Food and Drug Administration (FDA) for the treatment of severe diarrhea-predominant IBS refractory to treatment. It has shown improvement in global assessment measures, but its use is limited because of multiple isolated case reports of ischemic colitis. The recommended dose is 1 mg once or twice daily. It does not appear to be effective for the spastic-pain variant of IBS. Anticholinergics (dicyclomine, hyoscyamine) and antispasmodics (mebeverine, pinaverine[MB3]) are targeted at the pain and bloating symptoms but may also be useful for the diarrhea variant because of their constipating side effects. Studies comparing anticholinergic medications to placebo show inconclusive results with only modest benefits. Antispasmodic agents may also be of value and are available in many countries but are not approved for use in the United States. Currently, additional 5HT3 antagonists and 5HT4 antagonists are under development and are approved for use in Europe but not in the United States. Most studies are poorly designed and difficult to interpret because of a high placebo response rate that often exceeds 30% (95,96).

Behavioral Approaches

Biofeedback

Biofeedback can be an effective therapeutic modality provided patients are motivated and comprehend instructions. The two proposed mechanisms through which biofeedback improves fecal continence are afferent and efferent training. Afferent training focuses on improving sensation in the anorectal canal through recruitment of adjacent neurons to decrease the sensory threshold of volume stimulation. The goal of this training is to enhance and restore anal sensation and the rectoanal inhibitory reflex. Efferent training enhances and restores voluntary contraction of the EAS, which permits additional recruitment of motor units and stimulates muscle hypertrophy. These two methods of training can be performed independently but are often combined for additional therapeutic benefit. The most common training method uses an intrarectal balloon. The balloon acts to stimulate rectal distention and provide pressure feedback from coordinated or synchronized contraction of the pelvic floor muscles. Other techniques focus on strength training of the EAS alone using anal pressure feedback or EMG or afferent training alone using an intrarectal balloon without pelvic floor muscle contraction in response to the stimulus.

More than 35 studies have been done to evaluate the efficacy of biofeedback for treatment of fecal incontinence, and several excellent review articles and meta-analyses have determined the effects of individual treatments and predictors of patient response to treatment (97–99). The results of all of these studies uniformly agree that biofeedback and pelvic floor exercises improve fecal incontinence and have a role in clinical practice. They also agree that the existing literature is fraught with methodologic problems and lacks validated outcomes and controls. Thus it is difficult to compare directly the study results.

Biofeedback is an ideal first-line therapy because it offers an effective, minimally invasive treatment without any reported adverse events. Biofeedback also appears to provide a higher probability of successful outcome than standard medical care for treating functional fecal incontinence (67% vs. 36%, respectively, P <.001) (97).

A Cochrane review of biofeedback and exercises for treatment of fecal incontinence found only five randomized or quasi-randomized control trials that qualified for inclusion (100). The authors concluded that there is insufficient evidence to evaluate the efficacy of exercises and biofeedback for treatment of fecal incontinence. Specifically, they were not able to determine which patients were suitable for treatment nor which method of treatment was optimal. A meta-analysis of biofeedback techniques included a review of 13 studies using strength training alone, 4 studies with sensory training alone, and 18 with coordinated sensory and strength training (99). The authors found no advantages between coordinated training (67% improved) and strength training (70% improved). However, strength training using EMG appeared to be better than strength training with anal canal pressure biofeedback (74% vs. 64% improved, respectively, P <.04). The limitations of this study and the literature were acknowledged.

A large randomized control trial of biofeedback for fecal incontinence in 171 patients, divided into four treatment groups, showed no significant benefit when comparing standard care to similar care with the addition of biofeedback (54% vs. 53% improvement, respectively) (101). In all groups, there was a high median rating of change of symptoms and median satisfaction with benefits relatively maintained at 1-year follow-up. All groups displayed improvement in the validated symptom surveys and quality-of-life measures as well as in anal sphincter function. The authors concluded that interactions with the therapist, patient education, and development of better coping strategies seem to be the most important factors for improvement rather than pelvic muscle exercises or biofeedback. Additional benefit may be derived with augmented biofeedback using electrical stimulation (102).

There are no clear indicators to predict which patients will benefit from biofeedback. Potential factors include age, duration and severity of incontinence, prior treatments or surgery, and severity of neurologic or physical damage. Controversy exists as to whether response to biofeedback is dependent on the presence of a structurally intact anal sphincter or normal pudendal nerve function (103–106). Similarly, there is insufficient evidence that electrical stimulation improves fecal incontinence (107). There is an obvious need for well-designed control trials using validated symptom surveys and quality-of-life instruments. More objective measures are desirable, and studies should carefully document duration of treatment and length of follow-up.

Bowel Regimens

The goal of bowel regimens is to achieve predictable elimination of feces. This can be accomplished by using the gastrocolic reflex as well as by dietary and pharmacologic means. Defecation immediately following meals involves the physiologic response of the gastrocolic reflex to facilitate predictable emptying. The strength of the gastrocolic reflex varies among individuals and may be hypoactive or hyperactive with certain systemic disorders, such as diabetes and multiple sclerosis. This technique can be especially useful in the morning to give the individual freedom from fecal incontinence throughout the day. The use of suppositories or enemas in the morning or at night in conjunction with the gastrocolic reflex may provide further relief of daytime symptoms. The goal is to leave the rectum empty between evacuations. Enema use, typically once or twice daily, should be titrated to the patient’s baseline colonic activity. Regular toileting in elderly patients in nursing homes can improve fecal incontinence caused by overflow incontinence from fecal impaction. The use of cone-tip colostomy-irrigation catheters is reserved for patients in whom other therapeutic modalities have failed. These catheters avoid the risk of rectal perforation and provide a dam to prevent efflux of the irrigating solutions (108).

Surgical Treatment

In general, surgical treatment should be employed after conservative measures have failed. Although there may be exceptions to this principle, most surgeons follow this recommendation because of the poor long-term outcomes and high complication rates with surgery for fecal incontinence.

Overlapping Sphincteroplasty

Overlapping sphincteroplasty is the procedure of choice for fecal incontinence caused by a disrupted anal sphincter. Most authorities believe that an overlapping technique is superior to an end-to-end repair, although there are few direct comparisons in the literature. The rationale for the overlapping technique is that a more secure repair can be accomplished by placing sutures through the scarred connective tissue rather than the sphincter muscle. Sutures should be less likely to tear through or pull out of connective tissue than muscle. Therefore, the key component of the overlapping technique is preservation of the scarred ends of the ruptured EAS for suture placement.

Technique

The initial step involves wide mobilization of the ruptured EAS without excision of the scarred ends of the sphincter. This is accomplished through an inverted semilunar perineal incision or a transverse incision near the posterior vaginal fourchette with inferiolateral extension. The latter incision facilitates repair in patients with damage to the rectovaginal septum attachment to the perineal body. Patients with EAS defects have either a band of intervening fibrous scar tissue between the viable muscular ends of sphincter or a complete separation with scar tissue present only on the ruptured ends of the sphincter. In the presence of complete separation of scar tissue, perineal body reconstruction usually is indicated at the time of repair to restore normal anatomy. A Peña muscle stimulator aids in identification of the distal ends of the EAS and differentiates viable muscle tissue from scar tissue. The stimulator can be used to outline the sphincter before incision as well as during the dissection. It is important to apprise the anesthesiologist of the stimulator usage so that paralytic agents are avoided.

Excessive lateral dissection of the EAS past the 3 and 9 o'clock positions should be avoided as this is where the inferior rectal branches of the pudendal nerve innervate the EAS. Moderate bleeding often is encountered during this dissection, and the use of needlepoint electrocautery can maximize hemostasis. Controversy exists regarding the need to separate the EAS and IAS before repair. Identification of the intersphincteric groove facilitates dissection of the EAS. Dissection in this plane is relatively simple and avoids damage to either sphincter. Defects in the IAS can be more difficult to visualize because this muscle is intimately associated with the rectal mucosa. Examination with a finger in the anal canal is often helpful.

The reconstruction begins with repair of an existing IAS defect using a 3-0 delayed absorbable monofilament suture. Next, the EAS defect is repaired with the primary goal of overlapping at least 2 to 3 cm to ensure adequate bulk of sphincter muscle encircling the anal canal. The EAS is overlapped using three to four mattress sutures of 2-0 delayed absorbable monofilament suture through the distal scar tissue. Once the sutures are tied, there should be resistance palpable with placement of a finger in the anal canal. Copious irrigation is performed throughout the procedure. Following sphincter repair, a perineal body reconstruction and rectocele repair should be undertaken, if indicated, to maximize the normal continence mechanism. Finally, the perineal skin is closed with interrupted absorbable monofilament sutures. Closure frequently requires modification of the initial incision because of changes in the perineal architecture that result from the repair. The most common approach is an inverted Y-shaped closure of the incision (Fig. 28.4).

Figure 28.4 Overlapping sphincteroplasty procedure. A: Inverted semilunar perineal incision with the distal ends of the external sphincter outlined using the Peña muscle stimulator. B: The external sphincter has been dissected, the scar divided in the midline, and the internal sphincter repaired. C: The external sphincter is overlapped using three mattress sutures of 2–0 delayed absorbable monofilament suture through the distal scar tissue. D: The sutures are tied. E: The skin is closed.

Some surgeons recommend the overlapping repair regardless of whether it is performed immediately postpartum, delayed postpartum, or several years after obstetric injury. Performance of the overlapping technique is difficult immediately postpartum and requires adequate anesthesia, exposure, and equipment. Many surgeons believe that this can only be accomplished in the operating room. This repair lacks the theoretical advantage of using scar tissue to improve suture holding; however, it maximizes surface area for scarification of the sphincter ends. For a delayed postpartum repair, it is recommended to wait 3 to 6 months to permit complete resolution of inflammation and reinnervation.

Four randomized control trials compared end-to-end approximation to overlapping sphincteroplasty after acute obstetrical injury (109–112). Fitzpatrick et al. randomized 112 primiparous women undergoing immediate repair of a third- or fourth-degree sphincter tear (109). The authors did not detect any significant differences in objective or subjective outcomes between either of the two repairs at 3 months follow-up. Approximately one-half of the women had minor alteration in fecal continence, whereas 7 (6%) had daily soiling. Despite good symptomatic results, 74 (66%) had full-thickness EAS defects on endoanal ultrasonography. Williams et al. randomized 112 women to overlapping or end-to-end repair with either polyglactin (Vicryl) or polydioxanone (PDS) (110). They, too, found no difference in outcomes between the different methods of repair or type of suture used. Fernando et al. randomized 64 women and discovered a lower rate of fecal incontinence symptoms and fecal urgency after overlapping repair compared to end-to-end repair (111). Farrell et al. randomized 149 women to one of the two techniques and was the only study to discover higher rates of fecal incontinence of flatus in the overlapping group compared to the end-to-end group (61% vs. 39%; OR 2.44) (112). There was a trend toward overall higher rates of fecal incontinence; however, these were not statistically different (15% vs. 8%) (112). A Cochrane review that included the first three randomized controlled trials concluded that the overlapping repair technique resulted in less fecal incontinence and fecal urgency, but the data were insufficient to recommend one repair over the other (113). Another randomized controlled trial involved a delayed repair in 23 patients more than 1 year following delivery (114). The scar was preserved for each repair, and a puborectalis plication was performed. At a median follow-up of 18 months, there were no detectable differences in continence scores; however, the study was clearly underpowered. Consequently, the authors believe there is no definitive evidence to support an overlapping technique at the time of acute obstetrical injury and repair but prefer an overlapping technique as a delayed repair.

Efficacy

Despite the many large series reporting the outcomes of overlapping sphincteroplasty, almost all are retrospective in nature and lack validated measures of symptom severity and quality-of-life considerations. Several overlapping sphincteroplasty series with a total of 891 patients were evaluated from 1984 to 2001. Although the length of follow-up was variable, the results showed excellent and good outcomes in approximately two-thirds of patients (median 67%, range 52%–83%) (115). None of these studies had long-term outcomes.

More recent studies suggest poor long-term outcomes for the overlapping sphincteroplasty. In a series of 55 women who underwent overlapping sphincteroplasty for fecal incontinence secondary to obstetric trauma, researchers contacted 47 (86%) patients by postal questionnaire and telephone interview with a median time since surgery of 77 months (range 60–96 months) (116). The investigators observed less symptomatic improvement when compared with the results at 15 months postoperative evaluation. After excluding one patient because of Crohn disease, eight (17%) failed because they required additional surgery, such as colostomy, postanal repair, and artificial bowel sphincter. Among the remaining 38 patients, 27 (71%) reported improved bowel control, 5 (13%) were unimproved, and 6 (16%) were worse. No patient was fully continent to solid and liquid stool and flatus. Only 23 (50%) patients had “good” outcomes defined as not requiring further continence surgery and fecal incontinence less than once per month.

In another study, investigators contacted 49 (69%) of 71 patients by telephone interview (117). All underwent overlapping sphincteroplasty with a median follow-up of 62.5 months (range 47–141 months). Only 6 (12%) patients were totally continent, and another 18 (37%) were continent to liquid and solid stool. More than half of the patients had incontinence to liquid or solid stool. The largest series with long-term follow-up involved contact of 130 (71%) of 191 patients using a postal or telephone questionnaire (118). The median time from surgery for respondents was 10 years (range 7–16 years). Of those who responded, 6% had no incontinence, 16% were incontinent of flatus only, 19% had soiling only, and 57% were incontinent of solid stool. These outcomes were significantly worse than the previously reported 3-year assessment (119). Despite the fact that 61% had a poor outcome defined as having fecal incontinence or requiring additional surgery for incontinence, 62% still considered their bowel control to be better than before surgery, and 74% were satisfied with the results of their surgery. Although control may be improved when compared with preoperative status, continence outcomes do not seem to be maintained at long-term follow-up.

The cause of this deterioration in long-term outcomes is unknown. Possible explanations include weakening of the muscles with normal aging, repair breakdown, and underlying nerve damage from either obstetric injury or the repair itself. A problem with most studies is the lack of a follow-up ultrasonography to determine whether the repair was still intact. The effect of pudendal nerve function on overlapping sphincteroplasty is somewhat controversial. Significantly lower success rates have been shown in a comparison of those with normal pudendal nerve terminal motor latencies to those with abnormal latencies (63% vs. 17%; P <.01) (120). Other studies have confirmed this finding (79,89,106,107,121,122), but the more recent studies fail to show a difference based on preoperative neurophysiologic testing (116,118). Other controversial factors that may affect outcome include age, duration of fecal incontinence, size of the defect, and anal manometry results.

Although there are many controversial aspects to overlapping sphincteroplasty, the literature is in agreement that diverting colostomy is not necessary; bowel confinement does not improve outcomes; clinical improvement correlates with postoperative endoanal ultrasonography results; and prior sphincteroplasty does not affect outcomes (114,116,120,123–128).

Subsequent Deliveries

Multiple studies confirm the impact of anal sphincter laceration during the first delivery on the risk of a sphincter laceration in a second delivery (129–132). These studies have calculated odds ratios ranging between 2.5 and 5.3 for a second sphincter disruption. Two recent population-based studies revealed adjusted odds ratios of 4.2 (95% confidence interval [CI], 3.9–4.6) and 4.3 (95% CI, 3.8–4.8) (130,131). These odds ratios probably represent underestimates because they do not take into account higher cesarean delivery rates in subsequent births for women with a history of sphincter laceration. Both of these studies observed significantly increased risk of recurrent sphincter laceration associated with increased birth weight. The studies estimated that approximately 25 cesarean deliveries have to be performed to prevent one recurrent sphincter laceration. In fact, only 10% of women with anal sphincter lacerations at second delivery had a history of prior sphincter laceration. Although a history of prior sphincter laceration increases the risk of recurrent sphincter laceration, the risk remains relatively small. It is important to accurately counsel expectant mothers about their risk of sphincter laceration. Using this information, they can decide whether the risk of recurrent laceration outweighs the risk of elective cesarean birth. The risk of subsequent vaginal delivery on symptoms of fecal incontinence is unknown for women with a repaired anal sphincter. The presence or absence of preexisting fecal incontinence, as well as the estimated fetal weight, should be considered in counseling for a subsequent pregnancy.

Graciloplasty

Surgical reconstruction with a muscle flap should be considered in cases in which there is insufficient muscle to repair the EAS and all conservative measures have failed. Insufficient muscle can be caused by trauma or severe atrophy that results from denervation injury and congenital disease. Most patients considering this procedure have already undergone an overlapping sphincteroplasty that failed. Graciloplasty, first described by Pickrell et al. in 1952, is a skeletal muscle transposition procedure that uses the gracilis to create a new anal sphincter (133). There are three suitable muscles for this type of procedure: the gracilis, sartorius, and gluteus maximus. The muscle should be easily mobilized and transposed but not essential for locomotion or posture. The sartorius and gluteus maximus are suboptimal because the sartorius receives segmental vascularization, which restricts rotation, and the gluteus maximus is important in daily activities such as running, climbing stairs, and rising from a sitting position. The gracilis is a better choice because it can easily be mobilized without damage. As the most superficial adductor, it receives neurovascular supply proximally and has no important independent function.

Technique

Either one long incision or three small incisions are made in the medial thigh. The gracilis muscle is identified and mobilized toward its insertion onto the medial aspect of the tibia where the tendon is divided. Anterior and posterior perianal incisions are made approximately 1.5 cm from the anal verge. Tunnels are developed in the extrasphincteric space and from the proximal thigh to the anterior perianal incision. The gracilis muscle is then gently delivered to the anterior perianal incision, guided around the anus to the posterior perianal incision, and returned to the anterior incision encircling the anal canal. The distal tendon of the gracilis is passed behind the muscle and anchored to the contralateral periosteum of the ischium. In cases when there is inadequate length, it can be sutured to the ipsilateral ischium. This procedure can also be performed bilaterally. In patients with a large rectovaginal fistula or cloaca, a myocutaneous flap can be mobilized and used to help close the defect. Improvement of fecal incontinence is caused by passive increase of the resistance of the anal canal by the bulk of the encircling muscle (Fig. 28.5).

Figure 28.5 Graciloplasty. A: The gracilis muscle is identified and mobilized toward its insertion onto the medial aspect of the tibia where the tendon is divided. B: Anterior and posterior perianal incision are made approximately 1.5 cm from the anal verge. The muscle is then tunneled around the extrasphincteric space circumferentially. The distal tendon is passed behind the muscle and anchored to the contralateral periosteum of the ischium.

Experimental efforts to improve the efficacy of this procedure have focused on developing resting tone in the transposed muscle through the use of an implanted neurostimulator. The intent of the stimulated graciloplasty is to convert the fast-twitch muscle fibers into slow-twitch muscle fibers, which are more fatigue resistant. Initially, implantation of the pacemaker was performed at 6 weeks after the graciloplasty, but now most are performed concomitantly. Stimulation can be applied directly to the obturator nerve or intramuscularly to the nerve branches inside the muscle. The muscle is stimulated at a cyclic frequency, with gradual increases every 2 weeks. After 2 months, continuous stimulation is performed. Stimulation is adjusted to maintain tonic contraction around the anus, and it is interrupted or turned off to defecate.

Efficacy

An exhaustive review of the published literature identified 37 articles on patients undergoing dynamic graciloplasty (134). Most of these articles were case series, and there were no randomized trials or cohort studies evaluating safety and efficacy. Mortality rate was 1% (range 0%–13%; 95% CI, 1%–3%) after excluding cancer deaths. There was a high rate of morbidity (1.12 events per patient). Most patients will have at least one adverse event, and several will have multiple complications. There is also a very high reoperation rate. The most common complications were infections (28%), stimulator and lead faults (15%), and leg pain (13%). Satisfactory continence was achieved 42% to 85% of the time, although satisfaction was not defined consistently across studies. The authors concluded that dynamic graciloplasty appeared to have equal or better efficacy than colostomy but carried a higher morbidity rate. Another review of the three largest case series found success rates ranging from 55% to 78% (135–137). Major infections were found in 13% to 29%, pain in 27% to 28%, and device or lead problems in 12% to 18%. More recent series containing large numbers of patients found similar results for efficacy, morbidity, and reoperation rate (138,139). High rates of disturbed evacuation also were reported.

Artificial Sphincter

The artificial anal sphincter is an alternative to a graciloplasty. This is a modification of the device originally designed to treat urinary incontinence. The current device is the Acticon^® Neosphincter (American Medical Systems, Minnetonka, Minnesota) (Fig. 28.6). The indications for its use are similar to those for the graciloplasty.

Figure 28.6 Acticon^® Neosphincter. This device includes an inflatable cuff placed around the anal canal, a balloon reservoir stored behind the pubic bone, and a pump located in the labia. (Courtesy of American Medical Systems, Inc. Minnetonka, Minnesota, www.AmericanMedicalSystems.com.)

Technique

Implantation of the artificial anal sphincter is performed, similar to the graciloplasty, through perianal tunnels. A silastic inflatable cuff is placed around the native sphincter to occlude the anal canal. A pressure-regulating balloon containing radio-opaque solution is situated in the retropubic space, and a control pump is positioned in the labia majora. Activation of the control pump deflates the cuff, permitting defecation (Fig. 28.6).

Efficacy

An extensive review of the literature summarized 13 case series and one case report from 1996 to 2003. There were no randomized trials or cohort studies (140). The largest series consisted of 112 patients (141). There was one series with 53 patients, and all others had fewer than 28 patients each (142). Explantation was required in 17% to 41% of patients. Reasons for explantation included infection, erosion, device malfunction, pain, incontinence, and dissatisfaction, with infection being the most common. Surgical revision was necessary in 13% to 50% of the reports. Almost everyone had at least one adverse event, and more than one-third of these events required surgical intervention. Reasons for surgical revision were similar to those for explantation. Rates of fecal impaction ranged from 6% to 83%. All studies recorded statistically and clinically significant improvement in continence scores for patients with a functional artificial sphincter; however, most did not report the continence status for those in whom the device was explanted. The proportion of patients with a functional device ranged from 49% to 85%. The authors concluded that there is insufficient evidence on the safety and effectiveness of the artificial sphincter for fecal incontinence.

One randomized control trial of 14 patients compared an artificial sphincter with a program of supportive care (143). Supportive care included all aspects of conservative management, such as physiotherapy, dietary advice, pharmacotherapy, and advice regarding skin care, odor management, anxiety reduction, and use of incontinence aids or appliances. Significant improvements in continence scores and quality-of-life measures were seen in the artificial sphincter group but not in the control group at 6 months follow-up. Explantation rate was 14% (one of seven patients). Two other patients had complications, including severe fecal impaction and perineal wound erosion requiring reoperation. The authors conclude that the artificial sphincter is safe and effective compared with supportive care alone. They anticipate perioperative and late complications, which may require explantation in up to one-third of patients. It is also remarkable that only one patient (14%) whose condition was managed conservatively had significant improvement based on continence scores, whereas the status of all others was relatively unchanged.

Another study compared the effectiveness of artificial sphincter with dynamic graciloplasty (144). Two surgeons each performed four consecutive operations with each technique to minimize the learning curve of a new operation. Each started with a different procedure to avoid discrepancies in the time of follow-up. This prospective cohort study involved eight patients in each group who had similar demographic variables. Length of follow-up was 44 months in the artificial sphincter group and 39 months in the dynamic graciloplasty group. Early postoperative complications were similar in each group at 50%, as were late complications, with both groups reporting a high reoperation rate of 63%. There were six (75%) late complications in the artificial sphincter group, of which three (38%) were nonreversible and required explantation. Postoperative continence scores were significantly lower with the artificial sphincter than with graciloplasty. The authors conclude that artificial sphincter has better efficacy and similar morbidity compared with dynamic graciloplasty. The rate of late complications for the artificial sphincter exceeded that reported in the literature, which may indicate poor long-term durability. Postoperative continence scores reflect those reported for artificial sphincter but are far worse than those for dynamic graciloplasty. The authors feel that the learning curve with the artificial sphincter is less important than that with graciloplasty.

Sacral Nerve Root Stimulator

Sacral neuromodulation (InterStim^®, Medtronic, Minneapolis, Minnesota) was approved by the FDA for treatment of urinary urge incontinence in 1997 and for nonobstructive urinary retention and urgency in 1999. It has been employed experimentally for the treatment of fecal incontinence and is currently undergoing FDA approval for this intervention. In Europe, it was approved for treatment of both urinary and fecal incontinence in 1994. The exact mechanism of action has not been fully elucidated. The goal of sacral nerve stimulation is to recruit residual function of the continence mechanism through electrical stimulation of its peripheral nerve supply. Initially, indications were confined to patients with deficient EAS and levator ani function without gross morphologic defects and intact neuromuscular connections. More recently, the acceptable indications have expanded to include deficiency of the IAS, limited structural defects, and functional deficits of the internal and external anal sphincter.

Technique

The device is instilled exactly the same way as for treatment of urinary incontinence. Current application is performed as a two-stage outpatient surgical procedure. The first stage involves instillation of the electrodes. The electrode is placed through the S2–4 foramen using minimally invasive surgical technique. During the test phase, multiple electrodes can be employed either bilaterally or at different levels to determine the site with the best response. Proper location is confirmed intraoperatively using fluoroscopy as well as visualization of an appropriate pelvic floor muscle response (bellows) with minimal plantar flexion of the first and second toes, which usually corresponds to S3 stimulation. An interval testing phase utilizes an external pulse generator that typically lasts 1 to 2 weeks. Those with a good response (decrease in fecal incontinence episodes of at least 50% documented by bowel-habit diary) will proceed to the second stage, implantation of the permanent pulse generator (IPG). Typically only one electrode is left in place at the end of the second stage. Once the permanent pulse generator is implanted, all adjustments are made using telemetry. The patient has a basic remote control that enables her to turn the device on or off and adjust the amplitude of the stimulation.

Efficacy

By the end of 2003, sacral nerve stimulation had been used to treat more than 1,300 patients with fecal incontinence (145). Despite this large number, the analysis of the results was limited to several small case series. In all studies, significant improvements in continence scores lasting up to 99 months occurred. Most patients experienced at least a 75% improvement in continence scores, and improvement also occurred in the frequency of incontinence episodes, the ability to postpone defecation, and bowel emptying. Intent-to-treat analysis revealed 80% to 100% therapeutic success. There also were significant improvements in quality-of-life measures using validated measurement scales. Complications occurred in 0% to 50% of patients, with the most common complications consisting of pain at the electrode or IPG site, electrode migration, infection, or worsening of bowel symptoms. No permanent sequelae occurred, however. Effects of anorectal physiology varied among the published studies, highlighting the fact that the precise mechanism of action remains unclear.

A Cochrane review of three crossover trials concluded that the limited data suggest improvement in both fecal incontinence and constipation with sacral nerve stimulation (146). Large multicenter trials have been conducted to evaluate the efficacy of sacral nerve stimulation for fecal incontinence. Wexner et al. prospectively followed 120 patients (110 female, 10 male) for a mean of 28 (2–70) months (147,148). Participants completed validated quality-of-life measures and bowel diaries. Success, defined as a 50% or greater improvement and reduction in fecal incontinence episodes per week, was achieved in 83% at 1 year and 85% at 2 years. Forty-one percent achieved total continence at 1 year, and mean incontinent episodes decreased from 9.4 to 1.9 per week. The same group reported an overall 10.8% infection rate (13/120), with nine occurring within the first month and four occurring after 1 year. Among the early infections, five responded to antibiotics, one resolved spontaneously, and three were treated surgically. For the late infections, all four had to be removed. Another large multicenter study of 200 patients revealed decreased severity scores and improved quality of life with sacral nerve stimulation (149). Loose stool consistency and low stimulation intensity were predictive of successful outcomes on multivariate analysis. Ultrasound findings, manometry, age and gender did not impact outcomes. Tjandra et al. performed a randomized controlled trial of 120 participants with 1-year follow-up comparing sacral nerve stimulation to best supportive therapy consisting of pelvic floor exercises, bulking agent, and dietary manipulation (150). Fecal incontinence outcomes improved dramatically in the stimulation group, with 47% achieving complete continence. In contrast, there were no significant improvements observed in the supportive therapy group. Thus, sacral nerve stimulation appears to be a promising new treatment for fecal incontinence with relatively limited complications. FDA approval for fecal incontinence seems likely in the near future, and the minimally invasive nature of this procedure makes it a desirable first-line surgical option.

Therapeutic Approach to Constipation

As with fecal incontinence, it is imperative to attempt conservative management of constipation and defecatory dysfunction before performing surgery. Initial evaluation should focus on identifying any underlying systemic conditions (Table 28.1) associated with disordered defecation and optimizing treatment for these conditions. In the absence of systemic etiologies, it is reasonable to proceed with empiric, nonsurgical management, such as diet, fiber supplementation, and toileting behavior changes. Biofeedback and laxatives can be used in more severe cases. Initially, disimpaction with regular enemas or laxatives is essential if the patient has fecal impaction. Symptoms that persist despite a trial of conservative management indicate the need for further evaluation of colonic and anorectal function. A diagnostic algorithm for idiopathic (nonsystemic) constipation is given in Figure 28.7. Treatment should then be tailored to the underlying cause. Some conditions associated with disordered defecation are best treated using nonsurgical techniques, whereas others may benefit from surgery once conservative management has failed. As with fecal incontinence, the lack of consistent outcome measures in the published literature makes it difficult to compare efficacy among treatments.

Figure 28.7 Diagnostic Algorithm for Idiopathic Constipation. (From Ellerkmann MR, Kaufman H. Defecatory dysfunction. In: Bent AE, Ostergard DR, Cundiff GW, et al., eds. Ostergard’s urogynecology and pelvic floor dysfunction. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2002:358, with permission.)

Nonsurgical Treatment

Nonsurgical management focuses on maximizing anorectal function through alteration of stool characteristics or behavioral modification. Stool consistency and volume can be manipulated by dietary and pharmacologic means to achieve passage of one stool every day or every other day. Additionally, behavior modification can be employed using regular toileting to prevent fecal impaction. Physical therapy and biofeedback can also be useful for coordinating pelvic floor and anal sphincter relaxation with defecation.

Pharmacologic Approaches

Dietary Modification and Fiber

The role of increased fluid and fiber intake for the treatment of constipation is controversial. It has been a commonly accepted belief that constipation is caused by low fluid intake and can be improved by increasing consumption. Several studies showed no association between fluid intake and constipation (151–153). However, one large study of 21,012 nursing home residents found a weak association between decreased fluid intake and constipation with an odds ratio of 1.49 (154). In one interventional study, increased fluid intake failed to improve stool frequency, consistency, or defecatory dysfunction in children (155). Another interventional study using fiber and mineral water displayed an increase in stool frequency and a decrease in laxative use in adults with constipation (156). This study lacked baseline data collection resulting in recall bias, and the use of mineral water containing magnesium may confound the results because of its mild laxative effect. Overall, the existing data do not support increased fluid intake to treat constipation unless there is evidence of dehydration (151).

The addition of fiber may decrease constipation through several mechanisms. Fiber acts as a stool bulking agent and improves stool consistency through water absorption. It can also act as a substrate for bacterial proliferation and gas production. These mechanisms of action are believed to result in increased colonic motility, decreased transit time, and increased stool frequency.

Fiber therapy appears to have a beneficial effect in the treatment of diverticular disease, constipation of pregnancy, and possibly IBS (19,157,158). Its efficacy for idiopathic (nonsystemic) constipation remains uncertain. Dietary fiber intake for patients with constipation was similar to that of controls in several studies (152,153). A meta-analysis of 36 randomized trials using laxatives or fiber therapy for the treatment of constipation showed that the use of fiber or laxatives resulted in increased stool frequency and improved symptoms without the presence of severe side effects (159). Conversely, another meta-analysis showed an inability to restore transit time and stool weight in constipated patients using dietary fiber (158). Approximately half of the patients in another study responded to fiber treatment, but a much better response occurred in patients without an identifiable structural or motility disorder (160). Consequently, a low-fiber diet may be a contributing factor in chronic constipation, and an empiric trial of fiber therapy can be expected to help some patients. Side effects of increased gas production may limit compliance with treatment, so doses should be slowly titrated. Fiber therapy should be avoided in patients with impaction, megacolon or megarectum, or obstructive gastrointestinal lesions. Fiber therapy should also be used with caution in patients with cognitive dysfunction (dementia), difficulty with ambulation, and underlying neurogenic disease for fear of worsening the condition. There is no evidence to substantiate the recommendation for extra water intake with fiber supplements (161).

Laxatives

Laxatives are commonly used to treat constipation and disordered defecation. Many classes of laxatives are available over the counter.

Bulk-Forming Laxatives

These come in natural forms (psyllium) as well as synthetic form (Metamucil, Konsyl, Citrucel) and are felt to be the safest laxatives. They have mechanisms of action and side effects similar to that of fiber (162).

Hyperosmolar Laxatives

These consist of poorly absorbed substances that increase intraluminal osmolarity and water absorption. This action results in greater stool volume with decreased consistency. Examples include nonabsorbable sugars (lactulose and sorbitol), glycerin, and polyethylene glycol (GoLytely, MiraLAX). Polyethylene glycol is a common preoperative bowel preparation. Side effects are diarrhea, increased flatus, and abdominal cramping (162).

Emollient Laxatives

These agents are divided into two subsets: docusate salts and mineral oil. The docusate salts have hydrophilic and hydrophobic properties similar to detergents. They soften stool and decrease surface tension by increasing stool water and lipid content. Examples include docusate calcium (Surfak), docusate potassium (Dialose, Kasof), and docusate sodium (Colace, Comfolax). They also improve the absorption of other laxatives and are combined in preparations with stimulant laxatives such as Correctol, Peri-Colace, and Feen-a-Mint. The limited absorption of mineral oil allows it to penetrate and soften the stool. It can be used orally or rectally. Prolonged daily use can lead to decreased absorption of the fat-soluble vitamins A, D, E, and K. Use of mineral oil should be avoided in elderly and debilitated patients, as well as in those with esophageal motility disorders because of the potential for aspiration pneumonia. Side effects include diarrhea, anal leakage, and pruritus ani (162–164).

Saline Laxatives

These usually contain magnesium cations and phosphate anions that are relatively nonabsorbable and produce an osmotic gradient with increased water absorption. They also stimulate intestinal motility by increasing cholecystokinin release. Fast-acting effects can be seen with both oral (2–6 hours) and rectal (15 minutes) preparations. Examples include magnesium citrate, magnesium hydroxide (Milk of Magnesia), magnesium sulfate, sodium phosphate, and biphosphate (Phospho-soda, Fleet enema). Although generally well tolerated, electrolyte abnormalities can occur. These side effects should be avoided in patients with renal insufficiency because of the potential for magnesium toxicity (162–164).

Stimulant Laxatives

These are found in three basic types: castor oil, anthraquinones, and diphenylmethanes. A metabolite of castor oil, ricinoleic acid, increases intestinal motility and secretion. Anthraquinones (cascara sagrada, senna [Senokot], casanthranol [aloe], and danthron) are absorbed by the small intestine and stimulate motility by increasing intraluminal fluid and electrolyte content. Diphenylmethanes (phenolphthaleins[Feen-a-Mint, Correctol] and bisacodyl[Dulcolax]) have a mechanism of action similar to anthraquinones. These agents are potent and are intended for short-term use in cases refractory to bulk or osmotic laxatives. It has been a long-standing belief that prolonged use can lead to a dilated atonic colon known as cathartic colon syndrome, melanosis coli, or neuronal degeneration. A recent article refutes the theory that stimulant laxatives damage the autonomic nervous system when used at recommended doses (151). Other side effects include cramping, nausea, and abdominal pain (162–164).

Prokinetic Agents

Medications that stimulate gastrointestinal motility primarily through neuromodulation of acetycholine levels include metoclopramide, cisapride, cholinergic agonists (bethanechol), cholinesterase inhibitors (neostigmine), and serotonin agonists. Their efficacy in the treatment of chronic idiopathic constipation is uncertain. Metoclopramide is better for upper gastrointestinal motility disorders, whereas cisaprideappears to exert its effect at the level of the colon (162–164).

Behavioral Approaches

Behavioral techniques such as biofeedback and bowel regimens may have a role in certain conditions associated with constipation and defecatory dysfunction. Overall, these approaches have far less application to disordered defecation than to fecal incontinence. Biofeedback is important in the treatment of dyssynergic defecation. Relaxation techniques and behavioral modification may be helpful for IBS. Bowel regimens in conjunction with laxatives, suppositories, and enemas can facilitate emptying by optimizing the gastrocolic reflex and increased peristaltic activity.

Efficacy of Nonsurgical Treatment

Irritable Bowel Syndrome

The most commonly used first-line treatment for the constipation variant of IBS is fiber supplementation and osmotic laxatives. The efficacy of bulking agents for this condition is controversial, and many studies, including meta-analyses, exhibit an effect similar to placebo. There may be benefit to the use of fiber because of the high placebo effect with IBS treatment and lack of serious adverse events associated with its use. However, patients may experience exacerbation of bloating and abdominal discomfort with fiber therapy. Randomized trials and meta-analysis revealed a beneficial effect of polyethylene glycol (MiraLAX) over placebo for the treatment of chronic constipation. In addition, polyethylene glycol was better than tegaserod and lactulose for chronic constipation (165–167). They can be useful as adjunctive treatment options, but can also exacerbate abdominal pain and discomfort. A newer class of drugs, serotonin 5HT4 agonist tegaserod (Zelnorm), stimulates peristalsis, increases colonic motility, decreases intestinal transit times, and reduces visceral hypersensitivity. The recommended dose is 6 mg twice daily. Randomized trials have consistently shown an approximately 10% greater improvement in global IBS symptoms when compared with placebo. Improvement in the bloating and pain symptoms also occurred. No episodes of ischemic colitis or cardiac toxicity have been reported with the use of this medication, and the most common side effects are diarrhea and headache. Cisapride, a 5HT4 agonist with partial 5HT3 antagonist actions, has been withdrawn from use secondary to rare cardiac toxicity. Additional 5HT4 agonists, 5HT3 agonists, and cholecystokinin antagonists are in development (95,96).

Colonic Inertia and Slow-Transit Constipation

Patients with slow-transit constipation tend to respond poorly to fiber supplementation, although most have already tried an empiric trial of fiber before testing to confirm the diagnosis (160). Some patients may benefit from regular toileting, either in the morning or after meals when there is increased colonic motor activity. Biofeedback may have modest short-term benefits, but the long-term effect is questionable (168). Enemas and suppositories can be used in conjunction with bowel regimens. It is also reasonable to attempt a trial of any of the laxatives listed in Table 28.10. Stimulant laxatives are commonly used, but questions remain about the development of neuronal degeneration with prolonged usage. It is imperative that patients adhere to and not exceed the recommended dosages. Data regarding laxative use for this condition have failed to show a significantly better response than placebo. Prokinetic agents are intuitively the ideal choice to stimulate colonic motility. Currently there is only one available prokinetic agent, tegaserod, approved for the treatment of constipation that improves colonic transit. Its data are almost entirely based on treatment of IBS, and there is a lack of information for its use in slow-transit constipation. Other prokinetic agents in various stages of testing include bethanechol, neostigmine, cholecystokinin antagonists, misoprostol, colchicine, neurotrophin-3, and other 5HT4 agonists such as prucalopride and mosapride (49).

Table 28.10 Laxatives for the Treatment of Disordered Defecation

Dyssynergic Defecation

As with slow-transit constipation, initial management using bowel regimens, laxatives, enemas, suppositories, and fiber supplementation is appropriate for patients with dyssynergic defecation, yet many will have already tried conservative management before undergoing testing to confirm the diagnosis. These treatments are relatively well tolerated with few serious side effects. They have not been shown to have greater efficacy when compared with placebo, and their role in the treatment of dyssynergic defecation remains uncertain. Specific treatment for this condition tends to focus on biofeedback because of studies indicating that this is an acquired behavioral disorder of defecation. Modalities such as diaphragmatic muscle training, simulated defecation, and manometric or electromyography-guided anal sphincter and pelvic muscle relaxation have been employed independently or combined with other techniques. These techniques have yielded symptomatic improvement in approximately 60% to 80% of patients. Many patients with dyssynergic defecation also have abnormal rectal sensation, so rectal sensory conditioning may provide additional benefit (33,49,169,170). Others have tried botulinum toxin injections to paralyze the puborectalis and anal sphincter muscle. Small case series have shown modest early improvement, but the results do not appear to be long lasting (171,172). A recent randomized trial of 48 participants comparing botulinum toxin to biofeedback retraining found better initial improvement with the botulinum toxin (70% vs. 50%) but no difference at 1 year (33% vs. 25%) (173).

Pessary for Treatment of Pelvic Organ Prolapse

Pessaries of various shapes and sizes have been used for centuries to treat pelvic organ prolapse (174). They are a safe alternative to surgery, with the most common complications being increased vaginal discharge and erosion or ulceration of the vaginal wall. Although pessaries represent a common therapeutic modality, there are limited data regarding fitting and management (175). Even less is known about which type of pessary is better for enteroceles and rectoceles, although the site of prolapse does not appear to affect the ability to retain a pessary (176). Pessaries can be divided into subtypes of supportive and space occupying (177). Some of the space-occupying pessaries, such as the Gellhorn and cube, use a suction mechanism to maintain vaginal retention, whereas others, like the donut, do not. In theory, space-occupying pessaries and those that exert forces against the posterior wall and vaginal apex (donut, inverted Gehrung) should aid in treatment of rectoceles and enteroceles. However, there are few data regarding the efficacy of pessaries for relieving symptoms of disordered defecation. One prospective study found that stage III or IV posterior vaginal wall prolapse was an independent predictor for discontinuation of pessary use in favor of surgical repair (178). The only randomized crossover trial comparing different types of pessaries (ring and Gelhorn) found improvement in quality-of-life measures with each pessary that did not differ. Protrusion and voiding dysfunction symptoms were most improved (179). More research is needed to determine the role of pessaries for treatment of rectoceles and enteroceles as well as symptoms that are likely to be improved using a pessary.

Surgical Treatment

Following is a review of the efficacy of various surgical treatments for specific conditions associated with constipation and disordered defecation.

Slow Transit/Colonic Inertia

Subtotal colectomy with ileosigmoid or ileorectal anastomosis is considered by many to be the surgical treatment of choice for slow-transit constipation refractory to medical management. Most surgeons restrict the use of this surgical procedure to the most extreme cases and typically operate on fewer than 10% of patients. Strict criteria for surgery include the following: chronic, severe, disabling symptoms unresponsive to medical therapy; slow transit in the proximal colon; no evidence of pseudo-obstruction; and normal anorectal function (162). Success rates are variable and depend on several factors. An extensive review of colectomy for slow-transit constipation analyzed 32 studies from 1981 to 1988 and found satisfaction rates ranged from 39% to 100% (180). Higher success rates occurred in studies in the United States (n = 11, median 94%, range 75%–100%) and prospective studies (n = 16, median 90%, range 50%–100%). Superior outcomes occurred in those who had a complete physiologic evaluation and proven slow-transit constipation. Patients with anismus had higher rates of recurrent symptoms and lower satisfaction levels (181). Poorer outcomes occurred with ileosigmoid and cecorectal anastomosis than with ileorectal anastomosis. Those with segmental resection (hemicolectomy) had the worst outcome. None of the studies had a comparison group, and outcomes were variable and lacking validated measures. Morbidity associated with the operation included small bowel obstruction (median 18%, range 2%–71%), need for reoperation (median 14%, range 0%–50%), diarrhea (median 14%, range 0%–46%), fecal incontinence (median 14%, range 0%–52%), recurrent constipation (median 9%, range 0%–33%), persistent abdominal pain (median 41%, range 0%–90%), and permanent ileostomy (median 5%, range 0%–28%). Mortality ranged from 0%–6% (182). A quality-of-life study revealed that the score correlated poorly with frequency of bowel movements. However, a lower score was seen in those patients who had persistent abdominal pain, diarrhea, fecal incontinence, and permanent ileostomies. Overall satisfaction with the procedure was very high and correlated with the quality-of-life score (183).

Surgical alternatives to subtotal colectomy include ileostomy, cecostomy with antegrade continence enemas, and sacral nerve stimulation. Subtotal colectomy has never been directly compared with ileostomy, but those who had a permanent diversion after subtotal colectomy had lower quality-of-life scores. Patients undergoing cecostomy with antegrade continence enemas can expect to have satisfactory function approximately half of the time, with most requiring additional revision procedures secondary to stomal complications (184). Although sacral nerve stimulation has primarily been used for fecal incontinence, the results of a few small studies evoke optimism for its use in chronic constipation and slow-transit constipation (146,185–187).

Pelvic Organ Prolapse

The variety of surgical treatment techniques for the repair of rectocele include posterior colporrhaphy, defect-directed repair, posterior fascial replacement, transanal repair, and abdominal repair with sacral colpopexy. When an enterocele is present, a culdoplasty is usually performed. In cases of perineal descent, abdominal sacral colpoperineopexy is the procedure of choice. Suture rectopexy can be performed in conjunction with sacral colpoperineopexy if rectal prolapse is present. Despite the routine use of these procedures, data are limited regarding symptomatic improvement of disordered defecation. Greater detail regarding the specific techniques for many of these procedures is provided in Chapter 27 This section will focus on surgical outcomes, including anatomic cure of prolapse, improvement of defecatory dysfunction symptoms, and morbidity associated with the procedure.

Posterior Colporrhaphy

Posterior colporrhaphy has been the surgical procedure for rectocele repair preferred by gynecologic surgeons for more than 100 years. Traditional posterior colporrhaphy narrows the vaginal caliber through plication of the rectovaginal septum and usually includes a perineorrhaphy, which narrows the introitus. Despite it broad use, there are few data regarding long-term anatomic success, symptomatic improvement, and sexual function following the procedure. The outcomes of several studies are summarized in Table 28.11 (189–207). Anatomic cure and relief of vaginal bulge occurred in 76% to 96% of patients. In these studies, the procedure was ineffective at treating constipation, vaginal digitations (splinting), and fecal incontinence. Dyspareunia developed in 8% to 26% of patients with and without levator plication (188–192,207). As early as 1961, high rates of dyspareunia have been reported with this procedure in as many as 50% of patients (208).

Table 28.11 Rectocele Repairs

Many feel that the successful anatomic support obtained with this procedure is offset by the modest relief of functional symptoms and high rate of de novo dyspareunia. However, a recent prospective case series of 38 women undergoing posterior colporrhaphy along with concomitant procedures for rectocele and obstructed defecation revealed markedly different results (209). Fascial plication was performed without levator plication, and perineal body reconstruction rather than routine perineorrhaphy was employed when indicated. Anatomic cure rate was 87% at 12 months and 79% at 24 months. Subjective cure rate was 97% at 12 months and 89% at 24 months. There was significant improvement in preoperative and postoperative symptoms for constipation (76% vs. 24%), digitations (100% vs. 16%), awareness of prolapse (100% vs. 5%), obstructed defecation (100% vs. 13%), and dyspareunia (37% vs. 5%). There was no difference in fecal incontinence and only one case of de novo dyspareunia. The authors attribute their improved anatomic and functional outcomes and combined improvement in dyspareunia to exclusion of levator plication, perineorrhaphy, and excision of vaginal epithelium. An additional benefit may be derived during mobilization of the vaginal epithelium, when scar tissue from prior episiotomy or surgery is divided. They also found that preoperative defecating proctography was of limited value and have stopped its routine use as part of the preoperative evaluation for women with symptomatic rectoceles and obstructive defecation.

Defect-Directed Repair

The goal of a defect-directed repair or site-specific repair is to restore normal anatomy (36). This procedure can be combined with a perineal body reconstruction, if necessary, but usually does not routinely involve perineorrhaphy. Table 28.11 lists the anatomic and functional outcomes for this type of repair. Anatomic cure rates range from 82% to 100%, which are similar to those for posterior colporrhaphy. This procedure also resulted in modest improvement for symptoms of difficult evacuation, vaginal bulge, and vaginal digitations, which appear to be slightly better than for posterior colporrhaphy (193–197,207). Constipation symptoms significantly decreased in only one study (193). All studies reported low rates of de novo dyspareunia with good functional and anatomic outcomes, but the long-term durability of the procedure is unknown. All but one of these studies included concomitant prolapse and urinary incontinence procedures.

A randomized clinical trial of 106 women with stage II or greater posterior vaginal wall prolapse compared posterior colporrhaphy, defect-directed rectocele repair, and defect-directed repair augmented with a porcine small intestinal submucosal (210). Participants completed validated pelvic floor instruments at baseline and 6 months, 1 year, and 2 years after surgery. Anatomic failure was defined as POP-Q system point Bp greater than or equal to −2 at 1 year. There was a significant improvement in prolapse and colorectal scales in all groups with no differences between groups. The proportion of subjects with functional failures was 15% overall and not significantly different between groups. Posterior colporrhaphy and defect-directed repairs resulted in similar anatomic and functional outcomes, although the addition of a porcine-derived graft did not improve anatomic outcomes. On average, all bowel symptoms evaluated were significantly improved 1 year after surgery, with no differences between treatment groups. The development of new “bothersome” bowel symptoms after surgery was uncommon (11%). After controlling for age, treatment group, comorbidities, and preoperative bowel symptoms, corrected postoperative vaginal support (stage 0 or I) was associated with a reduced risk of postoperative straining (adjusted OR 0.17; 95% CI, 0.03–0.9) and feeling of incomplete emptying (adjusted OR 0.1; 95% CI, 0.01–0.52). This led the authors to conclude that resolution or improvement in bowel symptoms can be expected in the majority of women after rectocele repair (211).

Transanal Repair

Transanal repair involves repair of the rectocele through a transanal incision with excision of redundant rectal mucosa and plication of the rectovaginal septum and rectal wall. The procedure was developed and is primarily used by colorectal surgeons to treat constipation or obstructed defecation associated with “low” or distal rectoceles. The advantages of this approach include excision of redundant rectal mucosa and the ability to treat other anorectal pathology, such as hemorrhoids or anterior rectal wall prolapse (212). Disadvantages include the inability to repair higher rectoceles, enteroceles, cystoceles, uterine prolapse, and defects in the perineal body or anal sphincter (213). Major complications of infection (6%) and rectovaginal fistula (3%) are relatively rare (203). Most studies did not require vaginal bulging or protrusion symptoms as a prerequisite for surgery. The results of several studies are summarized in Table 28.11. The anatomic cure rate was 70% to 98%, and symptoms of constipation, difficult evacuation, and vaginal digitations appear to improve (202–207).

Recent reviews have compared transanal with transvaginal rectocele repair using the results of two small, randomized control trials (214–217). Women with compromised sphincter function and other symptomatic prolapse were excluded. The results for transvaginal repair were superior to those for transanal repair with respect to subjective failure rate (relative risk [RR] 0.36; 95% CI, 0.13–1.0) and objective failure rate (RR 0.24; 95% CI, 0.09–0.64) (214). In one study, a significant decrease occurred in the depth of rectocele on postoperative defecography for the transvaginal group compared with the transanal group (2.73 vs. 4.13 cm, respectively) (217). The transvaginal group had fewer problems with bowel evacuation, but this finding was not statistically significant. In one study, researchers discovered that 38% of patients developed fecal incontinence following transanal repair (188). In the two randomized trials, no significant differences were seen in the rate of fecal incontinence or dyspareunia, but the studies were underpowered to detect a difference (216,217). Although a vaginal approach has been considered superior to a transanal approach for rectocele repair, studies are retrospective and impossible to compare because the indications for transanal repairs are generally different from those for transvaginal repairs. A prospective, randomized trial with adequate power to evaluate symptomatic outcomes of bowel and sexual function along with anatomic cure is warranted.

Posterior Fascial Reinforcement

Rectocele repair with graft augmentation is becoming more common, despite a paucity of supporting evidence indicating its benefits over standard procedures. The reason for its emergence is the theory that vaginal hernia repairs behave similar to abdominal hernia repairs, which have a documented decrease in recurrence when augmented with grafts. A variety of graft materials have been employed with posterior colporrhaphy and defect-directed repairs including autograft, allograft, xenograft, and synthetic mesh. There are no comparison data to aid in selecting the optimal graft. The purpose of the graft is debatable. It can either be intended to replace existing fascia as a permanent barrier or to provide an absorbable scaffold for collagen deposition, scar formation, and remodeling. The ideal material should have a low erosion rate, be relatively inexpensive, and decrease recurrence rates without causing bowel or sexual dysfunction. The outcomes for rectocele repair using graft materials placed either vaginally or abdominally appear in Table 28.11. High anatomic cure rates of 89% to 100% occurred, and symptoms of constipation, difficult evacuation, and vaginal bulge also appeared to improve.

One randomized trial used absorbable vaginal mesh for rectocele repair (192). Patients were randomly assigned to fascial replacement with polyglactin 910 mesh at the time of anterior and posterior colporrhaphy. There were no differences in recurrence rates when comparing 70 women with a traditional colpoperineorrhaphy with 73 women having a traditional repair plus mesh: 10% versus 8%, respectively. This study did not describe changes in bowel or sexual function, and there were no mesh-related adverse events. As previously noted, the randomized controlled trial comparing posterior colporrhaphy, defect-directed repair, and defect-directed repair with porcine small intestine submucosal graft augmentation revealed higher anatomical failure rates in the graft-augmented group compared to the site-specific alone or the posterior colporrhaphy group (46% vs. 22% and 14%, respectively; P = .02), with no difference in symptom outcomes (210). Importantly, the risks of vaginal mesh erosion and severe complications may be relatively low but carry significant morbidity, including rectovaginal fistula, persistent vaginal bleeding and discharge, dyspareunia, and the need for additional surgery (199,207). Nonsynthetic grafts appear to be safer, with fewer erosions compared with synthetic grafts; however, there is no evidence to suggest that the addition of a graft to the posterior compartment improves outcomes (210,218,219).

Abdominal Rectocele Repair

The abdominal approach to rectocele repair may be of value when a superior defect in the rectovaginal fascia occurs in a patient with accompanying enterocele, uterine prolapse, or vault prolapse.If a patient is undergoing an abdominal or laparoscopic procedure such as a sacral colpopexy, the graft can be extended along the posterior vaginal wall to correct proximal defects in the rectovaginal septum (220). There are limited data regarding the efficacy of abdominal rectocele repair. The indication for this procedure, as well as the need for additional vaginal repair of distal defects, is often determined intraoperatively. An ancillary study from the Pelvic Floor Disorders Network evaluating bowel symptoms 1 year after sacrocolpopexy found that the majority of bothersome bowel symptoms resolve after this procedure. There was no difference in postoperative bowel symptoms among those who underwent a concomitant rectocele repair and those who did not. It is important to note that the study was not developed to evaluate the impact of concomitant rectocele repair on bowel symptom resolution, and those who underwent a rectocele repair had more severe baseline bowel symptoms including worse obstructive symptom (221).

Sacral Colpoperineopexy for Perineal Descent

Sacral colpoperineopexy is a modification of sacral colpopexy aimed at correction of apical prolapse combined with rectocele and perineal descent (39). A continuous graft is placed from the anterior longitudinal ligament of the sacrum down to the perineal body. This procedure can be accomplished either through a total abdominal approach or a combined abdominal and vaginal procedure. If performing a total abdominal approach, the rectovaginal space is opened, and the rectum is dissected off the posterior vaginal wall and rectovaginal septum toward the perineal body. The graft is then sutured to the perineal body or as close to it as possible. A rectovaginal examination with the surgeon’s nondominant hand facilitates this attachment by supporting the perineal body. The graft is secured to additional points along the posterior vaginal wall and apex, and sacral colpopexy is completed in the usual fashion.

If performing a combined abdominal and vaginal approach, the graft is secured to the perineal body vaginally. The posterior vaginal wall is opened, and a defect-directed rectocele repair is performed. Sacral colpopexy is accomplished in the usual fashion except that the vaginal dissection is opened superiorly, creating a window to the abdominal dissection. The graft can then be passed down from the abdominal field to the vaginal field and anchored inferiorly to the perineal body and laterally to the arcus tendineus fascia rectovaginalis (Fig. 28.8).

Figure 28.8 Abdominal sacral colpoperineopexy with sigmoid resection and suture rectopexy. This sagittal view shows the posterior graft sutured to the rectovaginal fascia and perineal body after defect-directed rectocele repair. The anterior graft is sutured to the pubocervical fascia. Both sheets will be secured to the sacral periosteum to the right of the rectum. Rectopexy sutures (left) have not yet been tied and secured. (Courtesy of Geoffrey W. Cundiff, M.D.)

Short-term outcomes for 19 patients who underwent sacral colpoperineopexy indicated good anatomic results for apical and posterior support as well as for perineal descent (39). Complete cessation of defecatory dysfunction symptoms was accomplished in 66% of patients. In a report of outcomes for a slightly different variation of the sacral colpoperineopexy, the authors' technique involved attachment of Marlex mesh to the perineal body using a needle carrier (222). The failure rate was 25% and mesh erosion rate was 5% for 205 patients with up to 10-year follow-up. A study of Mersilene mesh erosion rates related to sacral colpopexy and sacral colpoperineopexy noted similar erosion rates between sacral colpopexy and colpoperineopexy when the vagina was not opened (3.2% vs. 4.5%, respectively) (223). However, the erosion rate was 16% with vaginal suture placement and 40% when the mesh was placed vaginally. The use of nonsynthetic grafts such as dermal allograft and xenograft may help prevent high erosion rates. In a case series of 11 patients, researchers performed sigmoid resection (if indicated) and suture rectopexy in conjunction with sacral colpoperineopexy using AlloDerm for women with coexistent rectal prolapse, perineal descent, and defecatory dysfunction. Early follow-up (12.5 ± 7.7 months) revealed excellent improvement of defecatory dysfunction symptoms and quality-of-life considerations, with an 82% cure of perineal descent (224). A recently published retrospective cohort of 38 women revealed high satisfaction following abdominal sacral colpoperineopexy, despite the persistence of obstructed defecation symptoms 5 years after surgery (225). Sacral colpoperineopexy may have value for a select group of patients, but larger prospective series with long-term anatomic and symptomatic outcomes are necessary to evaluate the durability of this procedure.

Rectal Prolapse

Numerous surgical procedures have been described for the treatment of rectal prolapse and are broadly categorized into perineal or abdominal approaches. Most surgeons prefer an abdominal procedure because of lower recurrence rates, reserving perineal procedures for more debilitated patients.

Abdominal Procedures

Abdominal procedures vary with respect to the extent of rectal mobilization, method of rectal fixation, and inclusion or exclusion of bowel resection. During abdominal rectopexy, the mesorectal plane is developed and the rectum mobilized down to the pelvic floor posteriorly, with care taken to identify and preserve the hypogastric nerves. Division of the lateral ligaments may or may not be performed. The concern is that division of the lateral ligaments will lead to rectal denervation and increased postoperative constipation. If performing a suture rectopexy, the fascia propria of the rectum is secured to the sacral periosteum from S-1 to S-3 (226). If performing a sigmoid resection with the rectopexy (Frykman-Goldberg resection rectopexy), the bowel resection is performed after mobilization and before suturing (227). The theoretical advantages of a rectosigmoid resection are creation of a dense area of fibrosis between the anastomotic suture line and the sacrum; removal of abundant rectosigmoid, avoiding torsion or volvulus; additional fixation through straightening of the left colon and decreased mobility from the phrenocolic ligament; and relief of constipation in select patients. It is typically reserved for patients with a long redundant sigmoid colon, although specific criteria have not been proposed. Mesh rectopexies are usually avoided because of concern for increased complications and infections associated with placement of a foreign body at the time of bowel resection. There are two basic types of mesh rectopexy: posterior mesh rectopexy and anterior sling rectopexy (Ripstein procedure) (228,229). A variety of materials have been used for this procedure, including absorbable and permanent mesh. The assumption is that placement of this material will provide increased support through increased fibrous tissue formation. During the Ripstein procedure, an anterior sling of fascia lata or synthetic mesh is placed in front of the rectum and sutured to the sacrum. Most surgeons avoid this procedure because of fear of obstructed defecation. Modifications using a posterolateral wrap have been developed to resolve this problem.

In a series of more than 10 patients, there were five open series and five laparoscopic reports for suture rectopexy (230). The recurrence rates ranged between 0% and 9%. Most reports showed an improvement in fecal incontinence symptoms, but the results for constipation were variable. There were no mortalities noted and no difference between laparoscopic and open results. For posterior mesh rectopexy, there were 14 open series and five laparoscopic reports. The recurrence rates ranged between 0% and 6%. As with suture rectopexy, there was general improvement in fecal incontinence, mixed results for constipation, and no differences between laparoscopic and open outcomes. The mortality rate was between 0% and 3%, with increased rates of infection if resection rectopexy was performed. For anterior sling rectopexy (Ripstein procedure), there were eight studies with a recurrence rate between 0% and 12%. Again, there was a trend toward improvement of fecal incontinence and mixed response for constipation. Mortality ranged from 0% to 3%. For resection rectopexy (Frykman-Goldberg procedure), there were nine open series and three laparoscopic reports. Recurrence ranged between 0% and 5%. There was general improvement in continence as well as an overall reduction in constipation observed in most studies. Mortality rate was 0% for all studies but one, in which it was 6.7% (231). This study was a small, randomized trial comparing 15 patients undergoing resection rectopexy to 15 patients undergoing absorbable mesh rectopexy. The patient who died was in the resection group and had a myocardial infarction. The authors concluded that sigmoid resection did not seem to increase operative morbidity but tended to diminish postoperative constipation, possibly by causing less outlet obstruction. The study was underpowered to detect a difference in morbidity or mortality.

The laparoscopic series demonstrated similar safety and efficacy to the open techniques, and the effect on continence and constipation tended to mirror the type of rectopexy performed. In a small, randomized trial, there were significant short-term benefits with laparoscopic rectopexy compared with open rectopexy, including earlier ambulation, more rapid return to normal diet, shorter hospital stay, and lower morbidity (232). Most surgeons believe that there are no differences in recurrence rates between suture and mesh rectopexy. Consequently, the role of mesh in these procedures is suspect. The role of division of the lateral ligaments is somewhat controversial. A Cochrane review performed in 2008 concluded that division of the lateral ligaments was associated with less recurrent prolapse but more postoperative constipation (233). The authors acknowledged the limitations of their review, which consisted of very few trials with small sample sizes and methodological weakness. A review of seven open and four laparoscopic series involving division of the lateral ligaments revealed a general improvement in fecal incontinence and either no change or worsening of constipation (230). Conversely, there were 15 open and 4 laparoscopic series with preservation of the ligaments that displayed improved continence and a trend toward reduced constipation. This study suggests that preservation of the lateral ligaments is associated with an improvement in fecal incontinence and constipation symptoms.

Perineal Procedures

Perineal procedures are more easily tolerated because they avoid laparotomy. Thus, they are ideal for patients at high risk for perioperative and postoperative morbidity and mortality. There are basically two perineal procedures: the Delorme procedure and perineal rectosigmoidectomy (Altemeier operation). Perianal encirclement procedures such as the Thiersch procedure are not recommended because of poor success rates, high recurrence rates, and fecal impaction.

The Delorme procedure was first described in 1900 and involves separation of the rectal mucosa from the sphincter and muscularis propria, followed by resection of the rectal mucosa and plication of the distal rectal wall (muscularis propria) (234) (Fig. 28.9). A review of 10 series found a recurrence rate ranging between 4% and 38% and mortality rates of 0% to 4% (230).The low mortality rates are impressive considering the higher-risk population; however, the recurrence rates make it a less desirable procedure among healthy patients. There was a general improvement in fecal incontinence and constipation. Fecal incontinence (presumably indicating anal sphincter disruption or denervation), chronic diarrhea, and severe perineal descent are associated with failure of this procedure (235). The Delorme operation may be preferred in cases when the prolapsing segment is shorter than 3 to 4 cm or there is no circumferential full-thickness prolapse, making perineal rectosigmoidectomy difficult to perform (230,236).

Figure 28.9 Delorme’s procedure. After mucosal stripping to the full extent of the prolapse, the circular smooth muscle or the rectum is plicated. A mucosa-to-mucosa anastomosis is then performed.

Perineal rectosigmoidectomy (Altemeier operation) has become the perineal procedure of choice (237). Among 12 studies, performance of full thickness excision of the rectosigmoid was associated with recurrence rates from 0% to 16% and mortality rates of 0% to 5%. Patients generally have minimal pain and a relatively uneventful postoperative course. Recurrent prolapse probably reflects inadequate resection. Incontinence results are modest at best but seem to improve substantially with the addition of levatorplasty. The addition of levatorplasty also appears to decrease the short-term recurrence rate, but there is no significant change in constipation with this procedure (238). Most agree that perineal rectosigmoidectomy with levatorplasty is the best procedure for very elderly patients and those with profound comorbidity. This is the preferred approach for patients with incarcerated, strangulated, or even gangrenous prolapsed rectal segment who are not candidates for abdominal rectopexy. Although there is a general consensus that abdominal rectopexy is better than perineal rectosigmoidectomy, there is only one small, prospective, randomized controlled trial comparing these procedures. This study did not have the power to detect a difference in recurrence rates but found that patients undergoing abdominal resection rectopexy had less fecal incontinence and better physiological results than patients who had perineal rectosigmoidectomy (233,239).

References

1. Barber MD, Bremer RE, Thor KB, et al. Innervation of the female levator ani muscles. Am J Obstet Gynecol 2002;187:64–71.

2. Harari D, Gurwitz JH, Avorn J, et al. How do older persons define constipation? Implications for therapeutic management. J Gen Intern Med 1997;12:63–66.

3. Glia A, Lindberg G. Quality of life in patients with different types of functional constipation. Scand J Gastroenterol 1997;32:1083–1089.

4. Cundiff GW, Nygaard I, Bland DR, et al. Proceedings of the American Urogynecologic Society Multidisciplinary Symposium on Defecatory Disorders. Am J Obstet Gynecol 2000;182:S1–S10.

5. Higgins PD, Johanson JF. Epidemiology of constipation in North America: a systematic review. Am J Gastroenterol 2004;99:750–759.

6. Sonnenberg A, Koch TR. Epidemiology of constipation in the U.S. Dis Colon Rectum 1989;32:1–8.

7. Drossman DA, Li Z, Andruzzi E, et al. U.S. household survey of functional gastrointestinal disorders: prevalence, sociodemography and health impact. Dig Dis Sci 1993;38:1569–1580.

8. Sonnenberg A, Koch TR. Physician visits in the United States for constipation: 1958 to 1986. Dig Dis Sci 1989;34:606–611.

9. Rantis PC Jr, Vernava AM III, Daniel GL, et al. Chronic constipation—is the work-up worth the cost? Dis Colon Rectum 1997;40:280–286.

10. Singh G, Lingala V, Wang H, et al. Use of health care resources and cost of care for adults with constipation. Clin Gastroenterol Hepatol 2007;5:1053–1058.

11. Irvine EJ, Ferrazzi S, Pare P, et al. Health-related quality of life in functional GI disorders: focus on constipation and resource utilization. Am J Gastroenterol 2002;97:1986–1993.

12. Nelson RL. Epidemiology of fecal incontinence. Gastroenterology 2004;126:S3–S7.

13. Boreham MK, Richter HE, Kenton KS, et al. Anal incontinence in women presenting for gynecologic care: prevalence, risk factors, and impact upon quality of life. Am J Obstet Gynecol 2005;192:1637–1642.

14. Dunivan GC, Heymen S, Palsson OS, et al. Fecal incontinence in primary care: prevalence, diagnosis, and health care utilization. Am J Obstet Gynecol 2010;202:493.e1–e6.

15. Wu J, Hundley AF, Fulton RG, et al. Forecasting the prevalence of pelvic floor disorders in U.S. Women: 2010 to 2050. Obstet Gynecol 2009;114:1278–1283

16. Nelson R, Norton N, Cautley E, et al. Community-based prevalence of anal incontinence. JAMA 1995;274:559–561.

17. Johanson JF, Lafferty J. Epidemiology of fecal incontinence: the silent affliction. Am J Gastroenterol 1996;91:33–36.

18. Feldman M, Schiller LR. Disorders of gastrointestinal motility associated with diabetes mellitus. Ann Intern Med 1983;98:378–384.

19. Jewell DJ, Younge G. Interventions for treating constipation in pregnancy. Cochrane Database Syst Rev 2001;2:CD001142.

20. Bradley CS, Kennedy CM, Turcea AM, et al. Constipation in pregnancy: prevalence, symptoms and risk factors. Obstet Gynecol 2007;110:1351–1357.

21. Devroede G, Lamarche J. Functional importance of extrinsic parasympathetic innervation to the distal colon and rectum in man. Gastroenterology 1974;66:273–280.

22. Devroede G, Arhan P, Duguay C, et al. Traumatic constipation. Gastroenterology 1979;77:1258–1267.

23. Read NW, Timms JM. Defecation and the pathophysiology of constipation. Clin Gastroenterol 1986;15:937–965.

24. Glick ME, Meshkinpour H, Haldeman S, et al. Colonic dysfunction in patients with thoracic spinal cord injury. Gastroenterology 1984;86:287–294.

25. Weber J, Grise P, Roquebert M, et al. Radiopaque markers transit and anorectal manometry in 16 patients with multiple sclerosis and urinary bladder dysfunction. Dis Colon Rectum 1987;30:95–100.

26. Glick ME, Meshkinpour H, Haldeman S, et al. Colonic dysfunction in multiple sclerosis. Gastroenterology 1982;83:1002–1007.

27. Longstreth GF, Thompson WG, Chey WD, et al. Functional bowel disorders. Gastroenterology 2006;130:1480–1491.

28. Barnett JL. Anorectal diseases. In: Yamada T, ed. Textbook of gastroenterology. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 1999.

29. Wald A, Tunuguntla AK. Anorectal sensorimotor dysfunction in fecal incontinence and diabetes mellitus: modification with biofeedback therapy. N Engl J Med 1984;310:1282–1287.

30. Schiller LR, Santa Ana CA, Schumulen AC, et al. Pathogenesis of fecal incontinence in diabetes mellitus: evidence for internal-anal-sphincter dysfunction. N Engl J Med 1982;307:1666–1671.

31. Harris RL, Cundiff GW. Anal incontinence. Postgrad Obstet Gynecol 1997;17:1–6.

32. Ihre T. Studies on anal function in continent and incontinent patients. Scand J Gastroenterol 1974;9:1–80.

33. Rao SS. Dyssynergic defecation. Gastroenterol Clin North Am 2001;30:97–114.

34. Rao SS, Tuteja AK, Vellema T, et al. Dyssynergic defecation: demographics, symptoms, stool patterns, and quality of life. J Clin Gastroenterol 2004;38:680–685.

35. Drossman DA. The functional gastrointestinal disorders and the Rome III process. Gastroenterology 2006;130:1377–1390.

36. Richardson AC. The rectovaginal septum revisited: its relationship to rectocele and its importance in rectocele repair. Clin Obstet Gynecol 1993;36:976–982.

37. Tulikangas PK, Walters MD, Brainard JA, et al. Enterocele: is there a histologic defect? Obstet Gynecol 2001;98:634–637.

38. Parks AG, Porter NH, Hardcastle J. The syndrome of the descending perineum. Proc R Soc Med 1966;59:477–482.

39. Henry MM, Parks AG, Swash M. The pelvic floor musculature in the descending perineum syndrome. Br J Surg 1982;69:470–472.

40. Cundiff GW, Harris RL, Coates K, et al. Abdominal sacral colpoperineopexy: a new approach for correction of posterior compartment defects and perineal descent associated with vaginal vault prolapse. Am J Obstet Gynecol 1997;177:1345–1355.

41. Ihre T. Intussusception of the rectum and the solitary ulcer syndrome. Ann Med 1990;22:419–423.

42. Thompson JR, Chen AH, Pettit PD, et al. Incidence of occult rectal prolapse in patients with clinical rectoceles and defecatory dysfunction. Am J Obstet Gynecol 2002;187:1494–1500.

43. Freimanis MG, Wald A, Caruana B, et al. Evacuation proctography in normal volunteers. Invest Radiol 1991:26;581–585.

44. Mellgren A, Schultz I, Johansson C, et al. Internal rectal intussusception seldom develops into total rectal prolapse. Dis Colon Rectum 1997;40:817–820.

45. Bharucha AE, Wald, A, Enck P, et al. Functional anorectal disorders. Gastroenterology 2006;130:1510–1518.

46. Thompson WG, Longstreth GF, Drossman DA, et al. Functional bowel disorders and functional abdominal pain. In: Drossman DA, Corazziari E, Talley NJ, et al., eds. Rome II: the functional gastrointestinal disorders. 2nd ed. McLean, VA: Degnon Associates, 2000:351–432.

47. Vanner SJ, Depew WT, Paterson WG, et al. Predictive value of the Rome criteria for diagnosing the irritable bowel syndrome. Am J Gastroenterol 1999;94:2912–2917.

48. Bassotti G, Imbimbo B, Betti C, et al. Impaired colonic motor response to eating in patients with slow-transit constipation. Am J Gastroenterol 1992;87:504–508.

49. Rao SS. Constipation: evaluation and treatment. Gastroenterol Clin North Am 2003;32:659–683.

50. Knowles CH, Martin JE. Slow transit constipation: a model of human gut dysmotility. Review of possible aetologies. Neurogastroenterol Motil 2000;12:181–196.

51. Handa VL, Danielsen BH, Gilbert WM. Obstetric anal sphincter lacerations. Obstet Gynecol 2001;98:225–230.

52. Fenner DE, Genberg B, Brahma P, et al. Fecal and urinary incontinence after vaginal delivery with anal sphincter disruption in an obstetrics unit in the United States. Am J Obstet Gynecol 2003;189:1543–1550.

53. Sultan AH, Kamm MA, Hudson CN, et al. Anal-sphincter disruption during vaginal delivery. N Engl J Med 1993;329:1905–1911.

54. Nguyen T, Handa VL, Hueppchen N, et al. Labour curve findings associated with fourth degree sphincter disruption: the impact of labour progression on perineal trauma. J Obstet Gynaecol Can 2010;32:21–27.

55. De Leeuw JW, Vierhout ME, Struijk PC, et al. Anal sphincter damage after vaginal delivery: functional outcome and risk factors for fecal incontinence. Acta Obstet Gynecol Scand 2001;80:830–834.

56. Nygaard IE, Rao SS, Dawson JD. Anal incontinence after anal sphincter disruption: a 30-year retrospective cohort study. Obstet Gynecol 1997;89:896–901.

57. Kamm MA. Faecal incontinence. BMJ 1998;316:528–532.

58. Faltin DL, Boulvain M, Irion O, et al. Diagnosis of anal sphincter tears by postpartum endosonography to predict fecal incontinence. Obstet Gynecol 2000;95:643–647.

59. MacArthur C, Glazener CM, Wilson PD, et al. Obstetric practice and faecal incontinence three months after delivery. Br J Obstet Gynaecol 2001;108;678–683.

60. Sultan AH, Johanson RB, Carter JE. Occult anal sphincter trauma following randomized forceps and vacuum delivery. Int J Gynaecol Obstet 1998;61:113–119.

61. Lal M, Mann CH, Callender R, et al. Does cesarean delivery prevent anal incontinence? Obstet Gynecol 2003;101:305–312.

62. Borello-France D, Burgio KL, Richter HE, et al. Pelvic Floor Disorders Network. Fecal and urinary incontinence in primiparous women. Obstet Gynecol 2006;108:863–872.

63. Nelson RL, Furner SE, Westercamp M, et al. Cesarean delivery for the prevention of anal incontinence. Cochrane Database Syst Rev 2010;2:CD006756.

64. Signorello LB, Harlow BL, Chekos AK, et al. Midline episiotomy and incontinence. BMJ 2000;320:86–90.

65. Carroli G, Mignini L. Episiotomy for vaginal birth. Cochrane Database Syst Rev 2009;1:CD000081.

66. Sultan AH, Kamm MA, Hudson CN, et al. Third degree obstetric anal sphincter tears: risk factors and outcome of primary repair. BMJ 1994;308:887–891.

67. Walker WA, Rothenberger DA, Goldberg SM. Morbidity of internal sphincterotomy for anal fissure and stenosis. Dis Colon Rectum 1985;28:832–835.

68. Zbar AP, Beer-Gabel M, Chiappa AC, et al. Fecal incontinence after minor anorectal surgery. Dis Colon Rectum 2001;44:1610–1623.

69. Read MG, Read NW, Haynes WG, et al. A prospective study of the effect of haemorrhoidectomy on sphincter function and faecal incontinence. Br J Surg 1982;69:396–398.

70. Snooks SJ, Henry MM, Swash M. Faecal incontinence due to external anal sphincter division in childbirth is associated with damage to the innervation of the pelvic floor musculature: a double pathology. Br J Obstet Gynaecol 1985;92:824–828.

71. Snooks SJ, Setchell M, Swash M, et al. Injury to the innervation of the pelvic floor sphincter musculature in childbirth. Lancet 1984;1:546–550.

72. Ryhammer AM, Bek KM, Laurberg S. Multiple vaginal deliveries increase the risk of permanent incontinence of flatus and urine in normal premenopausal women. Dis Colon Rectum 1995;38:1206–1209.

73. Handa VL, Harris TA, Ostergard DR. Protecting the pelvic floor: obstetric management to prevent incontinence and pelvic organ prolapse. Obstet Gynecol 1996;88:470–478.

74. Allen RE, Hosker GL, Smith AT, et al. Pelvic floor damage and childbirth: a neurophysiological study. Br J Obstet Gynaecol 1990;97:770–779.

75. Smith ARB, Hosker GL, Warrell DW. The role of partial denervation of the pelvic floor in the aetiology of genitourinary prolapse and stress incontinence of urine: a neurophysiologic study. Br J Obstet Gynaecol 1989;96:24–28.

76. Henry MM, Parks AG, Swash M. The anal reflex in idiopathic fecal incontinence: an electrophysiological study. Br J Surg 1980;67:781–783.

77. Ho YH, Goh HS. The neurophysiological significance of perineal descent. Int J Colorectal Dis 1995;10:107–111.

78. Gee AS, Mills A, Durdey P. What is the relationship between perineal descent and anal mucosal electrosensitivity? Dis Colon Rectum 1995;38:419–423.

79. Berkelmans I, Heresbach D, Leroi AM, et al. Perineal descent at defecography in women with straining at stool: a lack of specificity or predictive value for future anal incontinence. Eur J Gastroenterol Hepatol 1995;7:75–79.

80. Barber MD, Kuchibhatla MN, Pieper CF, et al. Psychometric evaluation of 2 comprehensive condition-specific quality of life instruments for women with pelvic floor disorders. Am J Obstet Gynecol 2001;185:1388–1395.

81. Jorge JM, Wexner SD. Etiology and management of fecal incontinence. Dis Colon Rectum 1993;36:77–97.

82. 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;42:1525–1532.

83. 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;43:9–17.

84. Richardson AC. Female pelvic floor support defects [editorial]. Int Urogynecol J Pelvic Floor Dysfunct 1996;7:241.

85. Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–17.

86. Swash M. Electromyography in pelvic floor disorders. In: Henry MM, ed. Coloproctology and the pelvic floor. 2nd ed. London: Butterworth-Heinemann, 1992:184–195.

87. Swash M, Snooks SJ. Motor nerve conduction studies of the pelvic floor innervation. In: Henry MM, ed. Coloproctology and the pelvic floor. 2nd ed. London: Butterworth-Heinemann, 1992:196–206.

88. Olsen AL, Ross M, Stansfield RB, et al. Pelvic floor nerve conduction studies: establishing clinically relevant normative data. Am J Obstet Gynecol 2003;189:1114–1119.

89. Laurberg S, Swash M, Henry MM. Delayed external sphincter repair for obstetric tear. Br J Surg 1988;75:786–788.

90. Hock D, Lombard R, Jehaes C, et al. Colpocystodefecography. Dis Colon Rectum 1993;36:1015–1021.

91. Kelvin FM, Maglinte DD, Hornback JA, et al. Pelvic prolapse: assessment with evacuation proctography (defecography) Radiology 1992;184:547–551.

92. Lamb GM, de Jode MG, Gould SW, et al. Upright dynamic MR defaecating proctography in an open configuration MR system. Br J Radiol 2000;73:152–155.

93. Read M. Effects of loperamide on anal sphincter function in patients complaining of chronic diarrhea with fecal incontinence and urgency. Dig Dis Sci 1982;27:807–814.

94. Tobin GW, Brocklehurst JC. Fecal incontinence in residential homes for the elderly: prevalence, aetiology and management. Age Ageing 1986;15:41–46.

95. Talley NJ. Pharmacologic therapy for the irritable bowel syndrome. Am J Gastroenterol 2003;98:750–758.

96. Schoenfeld P. Efficacy of current drug therapies in irritable bowel syndrome: what works and does not work. Gastroenterol Clin North Am 2005;34:319–335.

97. Palsson OS, Heymen S, Whitehead WE. Biofeedback treatment for functional anorectal disorders: a comprehensive efficacy review. Appl Psychophysiol Biofeedback 2004;29:153–174.

98. Norton C. Behavioral management of fecal incontinence in adults. Gastroenterology 2004;126:S64–S70.

99. Heymen S, Jones KR, Ringel Y, et al. Biofeedback treatment of fecal incontinence: a critical review. Dis Colon Rectum 2001;44:728–736.

100. Norton C, Hosker G, Brazzelli M. Biofeedback and/or sphincter exercises for the treatment of faecal incontinence in adults. Cochrane Database Syst Rev 2000;2:CD002111.

101. Norton C, Chelvanayagam S, Wilson-Barnett J, et al. Randomized controlled trial of biofeedback for fecal incontinence. Gastroenterology 2003;125:1320–1329.

102. 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;42:753–758.

103. Norton C, Kamm MA. Outcome of biofeedback for fecal incontinence. Br J Surg 1999;86:1159–1163.

104. Leroi AM, Dorival MP, Lecouturier MF, et al. Pudendal neuropathy and severity of incontinence but not presence of an anal sphincter defect may determine the response to biofeedback therapy in fecal incontinence. Dis Colon Rectum 1999;42:762–769.

105. Rieger NA, Wattchow DA, Sarre RG, et al. Prospective trial of pelvic floor retraining in patients with fecal incontinence. Dis Colon Rectum 1997;40:821–826.

106. van Tets WF, Juijpers JH, Bleijenberg G. Biofeedback treatment is ineffective in neurogenic fecal incontinence. Dis Colon Rectum 1996;39:992–994.

107. Hosker G, Cody JD, Norton CC. Electrical stimulation for faecal incontinence in adults. Cochrane Database Syst Rev 2007;3:CD001310

108. Madoff RD, Williams JG, Caushaj PF. Fecal incontinence. N Engl J Med 1992;326:1002–1007.

109. Fitzpatrick M, Behan M, O'Connell PR, et al. A randomized clinical trial comparing primary overlap with approximation repair of third-degree obstetric tears. Am J Obstet Gynecol 2000;183:1220–1224.

110. Williams A, Adams EJ, Tincello DG, et al. How to repair an anal sphincter injury after vaginal delivery: results of a randomized controlled trial. BJOG 2006;113:201–207.

111. Fernando RJ, Sultan AH, Kettle C, et al. Repair techniques for obstetrical anal sphincter injuries: a randomized controlled trial. Obstet Gynecol 2006;107:1261–1268.

112. Farrell SA, Gilmour D, Turnbull GK, et al. Overlapping compared with end-to-end repair of third- and fourth-degree obstetric anal sphincter tears: a randomized controlled trial. Obstet Gynecol 2010;116:16–24.

113. Fernando R, Sultan AH, Kettle C, et al. Methods of repair of obstetric anal sphincter injury. Cochrane Database Syst Rev 2006;3:CD002866

114. Tjandra JJ, Han WR, Goh J, et al. Direct repair vs. overlapping sphincter repair: a randomized, controlled trial. Dis Colon Rectum 2003;46:937–942.

115. Madoff RD. Surgical treatment options for fecal incontinence. Gastroenterology 2004;126:S48–S54.

116. Malouf AJ, Norton CS, Engel AF, et al. Long-term results of overlapping anterior anal-sphincter repair for obstetric trauma. Lancet 2000;355:260–265.

117. Halverson AL, Hull TL. Long-term outcome of overlapping anal sphincter repair. Dis Colon Rectum 2002;45:345–348.

118. Bravo Gutierrez A, Madoff RD, Lowry AC, et al. Long-term results of anterior sphincteroplasty. Dis Colon Rectum 2004;47:727–731.

119. Buie WD, Lowry AC, Rothenberger DA, et al. Clinical rather than laboratory assessment predicts continence after anterior sphincteroplasty. Dis Colon Rectum 2001;44:1255–1260.

120. Gilliland R, Altomare DF, Moreira H Jr, et al. Pudendal neuropathy is predictive of failure following anterior overlapping sphincteroplasty. Dis Colon Rectum 1998;41:1516–1522.

121. Londono-Schimmer EE, Garcia-Duperly R, Nicholls RJ, et al. Overlapping anal sphincter repair for faecal incontinence due to sphincter trauma: five year follow-up results. Int J Colorectal Dis 1994;9:110–113.

122. Sangwan YP, Coller JA, Barrett RC, et al. Unilateral pudendal neuropathy: impact on outcome of anal sphincter repair. Dis Colon Rectum 1996;39:686–689.

123. Rosenberg J, Kehlet H. Early discharge after external anal sphincter repair. Dis Colon Rectum 1999;42:457–459.

124. Hasegawa H, Yoshioka K, Keighley MR. Randomized trial of fecal diversion for sphincter repair. Dis Colon Rectum 2000;43:961–964.

125. Nessim A, Wexner SD, Agachan F, et al. Is bowel confinement necessary after anorectal reconstructive surgery? A prospective, randomized, surgeon-blinded trial. Dis Colon Rectum 1999;42:16–23.

126. Pinedo G, Vaizey CJ, Nicholls RJ, et al. Results of repeat anal sphincter repair. Br J Surg 1999;86:66–69.

127. Savoye-Collett C, Savoye G, Koning E, et al. Anal endosonography after sphincter repair: specific patterns related to clinical outcome. Abdom Imaging 1999;24:569–573.

128. Giordano P, Renzi A, Efron J, et al. Previous sphincter repair does not affect the outcome of repeat repair. Dis Colon Rectum 2002;45:635–640.

129. Peleg D, Kennedy CM, Merrill D, et al. Risk of repetition of a severe perineal laceration. Obstet Gynecol 1999;93:1021–1024.

130. Elfaghi I, Johansson-Erneste B, Rydhstroem H. Rupture of the sphincter ani: the recurrence rate in second delivery. Br J Obstet Gynaecol 2004;111:1361–1364.

131. Spydslaug A, Trogstad LI, Skrondal A, et al. Recurrent risk of anal sphincter laceration among women with vaginal deliveries. Obstet Gynecol 2005;105:307–313.

132. Lowder JL, Burrows LJ, Krohn MA, et al. Risk factors for primary and subsequent anal sphincter lacerations: a comparison of cohorts by parity and prior mode of delivery. Am J Obstet Gynecol 2007;196:344.e1–e5.

133. Pickrell KL, Broadbent TR, Masters FW, et al. Construction of a rectal sphincter and restoration of anal continence by transplanting the gracilis muscle; a report of four cases in children. Ann Surg 1952;135:853–862.

134. Chapman AE, Geerdes B, Hewett P, et al. Systematic review of dynamic graciloplasty in the treatment of faecal incontinence. Br J Surg 2002;89:138–153.

135. 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;116:549–556.

136. Baeten CG, Bailey HR, Bakka A, et al. Safety and efficacy of dynamic graciloplasty for fecal incontinence: report of a prospective multicenter trial. Dynamic Graciloplasty Therapy Study Group. Dis Colon Rectum 2000;43:743–751.

137. Geerdes BP, Heineman E, Konsten J, et al. Dynamic graciloplasty: complications and management. Dis Colon Rectum 1996;39:912–917.

138. Rongen MJ, Uluday O, El Naggar K, et al. Long-term follow-up of dynamic graciloplasty for fecal incontinence. Dis Colon Rectum 2003;46:716–721.

139. Penninckx F, Belgian Section of Colorectal Surgery. Belgian experience with dynamic graciloplasty for faecal incontinence. Br J Surg 2004;91:872–878.

140. Mundy L, Merlin TL, Maddem GJ, et al. Systemic review of safety and effectiveness of an artificial bowel sphincter for faecal incontinence. Br J Surg 2004;91:665–672.

141. 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;45:1139–1153.

142. [MB4]Devesa JM, Rey A, Hervas PL, et al. Artificial anal sphincter: complications and functional results of a large personal series. Dis Colon Rectum 2002;45:1154–1163.

143. O'Brien PE, Dixon JB, Skinner S, et al. A prospective, randomized, controlled clinical trial of placement of the artificial bowel sphincter (Acticon Neosphincter) for the control of fecal incontinence. Dis Colon Rectum 2004;47:1852–1860.

144. Ortiz H, Armendariz P, DeMiguel M, et al. Prospective study of artificial anal sphincter and dynamic graciloplasty for severe anal incontinence. Int J Colorectal Dis 2003;18:349–354.

145. Matzel KE, Stadelmaier U, Hohenberger W. Innovations in fecal incontinence: sacral nerve stimulation. Dis Colon Rectum 2004;47:1720–1728.

146. Mowatt G, Glazener C, Jarrett M. Sacral nerve stimulation for faecal incontinence and constipation in adults. Cochrane Database Syst Rev 2007;3:CD004464.

147. Wexner SD, Coller JA, Devroede G, et al. Sacral nerve stimulation for fecal incontinence: results of a 120-patient prospective multicenter study. Ann Surg 2010;251:441–449.

148. Wexner SD, Hull T, Edden Y, et al. Infection rates in a large investigational trial of sacral nerve stimulation for fecal incontinence. J Gastrointest Surg 2010;14:1081–1089.

149. Gallas S, Michot F, Faucheron JL, et al. Predictive factors for successful sacral nerve stimulation in the treatment of faecal incontinence: Results of trial stimulation in 200 patients. Colorectal Dis 2011;13:689–696..

150. Tjandra JJ, Chan MK, Yeh CH, et al. Sacral nerve stimulation is more effective than optimal medical therapy for severe fecal incontinence: a randomized, controlled study. Dis Colon Rectum 2008;51:494–502.

151. Muller-Lissner SA, Kamm MA, Scarpignato C, et al. Myths and misconceptions about chronic constipation. Am J Gastroenterol 2005;100:232–242.

152. Preston DM, Lennard-Jones JE. Severe chronic constipation of young women: “idiopathic slow transit constipation”. Gut 1986;27:41–48.

153. Towers AL, Burgio KL, Locher JL, et al. Constipation in the elderly: influence of dietary, psychological, and physiological factors. J Am Geriatr Soc 1994;42:701–706.

154. Robson KM, Kiely DK, Lembo T. Development of constipation in nursing home residents. Dis Colon Rectum 2000;43:940–943.

155. Young RJ, Beerman LE, Vanderhoof JA. Increasing oral fluids in chronic constipation in children. Gastroenterol Nurs 1998;21:156–161.

156. Anti M, Pignataro G, Armuzzi A, et al. Water supplementation enhances the effect of high-fiber diet on stool frequency and laxative consumption in adult patients with functional constipation. Hepatogastroenterology 1998;45:727–732.

157. Brodribb AJ. Treatment of symptomatic diverticular disease with a high-fiber diet. Lancet 1977;1:664–666.

158. Muller-Lissner SA. Effect of wheat bran on weight of stool and gastrointestinal transit time: a meta analysis. BMJ 1988;296:615–617.

159. Tramonte SM, Brand MB, Mulrow CD, et al. The treatment of chronic constipation in adults: a systematic review. J Gen Intern Med 1997;12:15–24.

160. Voderholzer WA, Schatke W, Muhldorfer BE, et al. Clinical response to dietary fiber treatment of chronic constipation. Am J Gastroenterol 1997;92:95–98.

161. Ziegenhagen DJ, Tewinkel G, Kruis W, et al. Adding more fluid to wheat bran has no significant effects on intestinal functions of healthy subjects. J Clin Gastroenterol 1991;13:525–530.

162. Wald A. Constipation. Med Clin North Am 2000;84:1231–1246.

163. Ellerkmann MR, Kaufman H. Defecatory dysfunction. In: Bent AE, ed. Ostergard’s Urogynecology. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2002:355–389.

164. Wald A. Approach to the patient with constipation. In: Yamada T, ed. Textbook of gastroenterology. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 1999.

165. Belsey JD, Geraint M, Dixon TA. Systematic review and meta analysis: polyethylene glycol in adults with non-organic constipation. Int J Clin Pract 2010;64:944–955.

166. Dipalma JA, Cleveland MV, McGowan J, et al. A randomized, multicenter, placebo-controlled trial of polyethylene glycol laxative for chronic treatment of chronic constipation. Am J Gastroenterol 2007;102:1436–1441.

167. Dipalma JA, Cleveland MV, McGowan J, et al. A randomized, multicenter comparison of polyethylene glycol laxative and tegaserod in treatment of patients with chronic constipation. Am J Gastroenterol 2007;102:1964–1971.

168. Battaglia E, Serra AM, Buonafede G, et al. Long-term study on the effects of visual biofeedback and muscle training as a therapeutic modality in pelvic floor dyssynergia and slow-transit constipation. Dis Colon Rectum 2004;47:90–95.

169. Ho YH, Tan M, Goh HS. Clinical and physiologic effects of biofeedback in outlet obstruction constipation. Dis Colon Rectum 1996;39:520–524.

170. Enck P. Biofeedback training in disordered defecation: a critical review. Dig Dis Sci 1993;38:1953–1960.

171. Ron Y, Avni Y, Lukovetski A, et al. Botulinum toxin type-A in therapy of patients with anismus. Dis Colon Rectum 2001;44:1821–1826.

172. Friedenberg F, Gollamudi S, Parkman HP. The use of botulinum toxin for the treatment of gastrointestinal motility disorders. Dig Dis Sci 2004;49:165–175.

173. Farid M, El Monem HA, Omar W, et al. Comparative study between biofeedback retraining and botulinum neurotoxin in the treatment of anismus patients. Int J Colorectal Dis 2009;24:115–120.

174. Shah SM, Sultan AH, Thakar R. The history and evolution of pessaries for pelvic organ prolapse. Int Urogynecol J Pelvic Floor Dysfunct 2006;17:170–175.

175. Adams E, Thomson A, Maher C, et al. Mechanical devices for pelvic organ prolapse in women. Cochrane Database Syst Rev 2004;2:CD004010.

176. Clemons JL, Aguilar VC, Tillinghast TA, et al. Risk factors associated with an unsuccessful pessary fitting trial in women with pelvic organ prolapse. Am J Obstet Gynecol 2004;190:235–250.

177. Cundiff GW, Weidner AC, Visco AG, et al. A survey of pessary use by members of the American Urogynecological Society. Obstet Gynecol 2000;95:931–935.

178. Clemons JL, Aguilar VC, Sokol ER, et al. Patient characteristics that are associated with continued pessary use versus surgery after 1 year. Am J Obstet Gynecol 2004;191:159–164.

179. Cundiff GW, Amundsen CL, Bent AE, et al. The PESSRI study: symptom relief outcomes of a randomized crossover trial of the ring and Gellhorn pessaries. Am J Obstet Gynecol 2007;196:405.e1–e8.

180. Knowles CH, Scott M, Lunniss PJ. Outcomes of colectomy for slow transit constipation. Ann Surg 1999;230:627–638.

181. Bernini A, Madoff RD, Lowry AC, et al. Should patients with combined colonic inertia and nonrelaxing pelvic floor undergo subtotal colectomy? Dis Colon Rectum 1998;41:1363–1366.

182. Walsh PV, Peebles-Brown DA, Watkinson G. Colectomy for slow transit constipation. Ann R Coll Surg Engl 1987;69:71–75.

183. FitzHarris GP, Garcia-Aguilar J, Parker SC, et al. Quality of life after subtotal colectomy for slow-transit constipation: both quality and quantity count. Dis Colon Rectum 2003;46:433–440.

184. Lees NP, Hodson P, Hill J, et al. Long-term results of the antegrade continent enema procedure for constipation in adults. Colorectal Dis 2004;6:362–368.

185. Malouf AJ, Wiesel PH, Nicholls T, et al. Short-term effects of sacral nerve stimulation for idiopathic slow transit constipation. World J Surg 2002;26:166–170.

186. Kenefick NJ, Vaizey CJ, Cohen CR, et al. Double-blind placebo-controlled crossover study of sacral nerve stimulation for idiopathic constipation. Br J Surg 2002;89:1570–1571.

187. Holzer B, Rosen HR, Novi G, et al. Sacral nerve stimulation in patients with severe constipation. Dis Colon Rectum 2008;51:524–529.

188. Arnold MW, Stewart WR, Aguilar PS. Rectocele repair: four years' experience. Dis Colon Rectum 1990;33:684–687.

189. Mellgren A, Anzen B, Nillson BY, et al. Results of rectocele repair: a prospective study. Dis Colon Rectum 1995;38:7–13.

190. Kahn MA, Stanton SL. Posterior colporrhaphy: its effects on bowel and sexual function. Br J Obstet Gynaecol 1997;104:82–86.

191. Weber AM, Walters MD, Piedmonte MR. Sexual function and vaginal anatomy in women before and after surgery for pelvic organ prolapse and urinary incontinence. Am J Obstet Gynecol 2000;182:1610–1615.

192. Sand PK, Koduri A, Lobel RW, et al. Prospective randomized trial of polyglactin 910 mesh to prevent recurrence of cystoceles and rectoceles. Am J Obstet Gynecol 2001;184:1357–1364.

193. Cundiff GW, Weidner AC, Visco AG, et al. Anatomic and functional assessment of the discrete defect rectocele repair. Am J Obstet Gynecol 1998;179:1451–1457.

194. Porter WE, Steele A, Walsh P, et al. The anatomic and functional outcomes of defect-specific rectocele repairs. Am J Obstet Gynecol 1999;181:1353–1359.

195. Kenton K, Shott S, Brubaker L. Outcome after rectovaginal fascia reattachment for rectocele repair. Am J Obstet Gynecol 2000;79:1360–1364.

196. Glavind K, Madsen H. A prospective study of the discrete fascial defect rectocele repair. Acta Obstet Gynecol Scand 2000;79:145–147.

197. Singh K, Cortes E, Reid WM. Evaluation of the fascial technique for surgical repair of isolated posterior vaginal wall prolapse. Obstet Gynecol 2003;101:320–324.

198. Oster S, Astrup A. A new vaginal operation for recurrent and large rectocele using dermis transplant. Acta Obstet Gynecol Scand 1981;60:493–495.

199. Goh JT, Dwyer PL. Effectiveness and safety of polypropylene mesh in vaginal prolapse surgery. Int Urogynecol J 2001;12:S90.

200. Kohli N, Miklos JR. Dermal graft-augmented rectocele repair. Int Urogynecol J Pelvic Floor Dysfunct 2003;14:146–149.

201. Mercer-Jones MA, Sprowson A, Varma JS. Outcome after transperineal mesh repair of rectocele: a case series. Dis Colon Rectum 2004;47:864–868.

202. Sullivan ES, Leaverton GH, Hardwick CE. Transrectal perineal repair: an adjunct to improved function after anorectal surgery. Dis Colon Rectum 1968;11:106–114.

203. Sehapayak S. Transrectal repair of rectocele: an extended armamentarium of colorectal surgeons. A report of 355 cases. Dis Colon Rectum 1985;28:422–433.

204. Janssen LW, van Dijke CF. Selection criteria for anterior rectal wall repair in symptomatic rectocele and anterior rectal wall prolapse. Dis Colon Rectum 1994;37:1100–1107.

205. van Dam JH, Huisman WM, Hop WC, et al. Fecal continence after rectocele repair: a prospective study. Int J Colorectal Dis 2000;15:54–57.

206. Ayabaca SM, Zbar AP, Pescatori M. Anal continence after rectocele repair. Dis Colon Rectum 2002;45:63–69.

207. Cundiff GW, Fenner D. Evaluation and treatment of women with rectocele: focus on associated defecatory and sexual dysfunction. Obstet Gynecol 2004;104:1403–1421.

208. Francis WJ, Jeffcoate TN. Dyspareunia following vaginal operations. J Opt Soc Am 1961;68:1–10.

209. Maher CF, Qatawneh AM, Baessler K, et al. Midline rectovaginal fascial plication for repair of rectocele and obstructed defecation. Obstet Gynecol 2004;104:685–689.

210. Paraiso MF, Barber MD, Muir TW, Walters MD. Rectocele repair: a randomized trial of three surgicval techniques including graft augmentation. Am J Obstet Gynecol 2006;195:1762–1771.

211. Gustilo-Ashby AM, Paraiso MF, Jelovsek JE, et al. Bowel symptoms 1 year after surgery for prolapse: further analysis of a randomized trial of rectocele repair. Am J Obstet Gynecol 2007;197:76.e1–e5.

212. Zbar AP, Leinemann A, Fritsch H, et al. Rectocele: pathogenesis and surgical management. Int J Colorectal Dis 2003;18:369–384.

213. Goh JT, Tjandra JJ, Carey MP. How could management of rectoceles be optimized? ANZ J Surg 2002;72:896–901.

214. Maher C, Baessler K, Glazener CM, et al. Surgical management of pelvic organ prolapse in women. Cochrane Database Syst Rev 2010:4:CD004014.

215. Maher C, Baessler K. Surgical management of posterior vaginal wall prolapse: an evidence-based literature review. Int Urogynecol J Pelvic Floor Dysfunct 2006;17:84–88.

216. Kahn MA, Stanton SL, Kumar D, et al. Posterior colporrhaphy is superior to the transanal repair for treatment of posterior vaginal wall prolapse. Neurourol Urodyn 1999;18:70–71.

217. Nieminen K, Hiltunen K, Laitinen J, et al. Transanal or vaginal approach to rectocele repair: a prospective randomized pilot study. Dis Colon Rectum 2004;47:1636–1642.

218. Watson SJ, Loder PB, Halligan S, et al. Transperineal repair of symptomatic rectocele with Marlex mesh: a clinical, physiological and radiologic assessment of treatment. J Am Coll Surg 1996;183:257–261.

219. Murphy M, Society of Gynecologic Surgeons Systematic Review Group. Clinical practice guidelines on vaginal graft use from the society of gynecologic surgeons. Obstet Gynecol 2008;112:1123–1130.

220. Addison WA, Cundiff GW, Bump RC, et al. Sacral colpopexy is the preferred treatment for vaginal vault prolapse. J Gynecol Tech 1996;2:69–74.

221. Gutman RE, Bradley CS, Ye W, et al. Pelvic Floor Disorders Network. Effects of colpocleisis on bowel symptoms among women with severe pelvic organ prolapse. Int Urogynecol J Pelvic Floor Dysfunct 2010;21:461–466.

222. Sullivan ES, Longaker CJ, Lee PY. Total pelvic mesh repair: a ten-year experience. Dis Colon Rectum 2001;44:857–863.

223. Visco AG, Weidner AC, Barber MD, et al. Vaginal mesh erosion after abdominal sacral colpopexy. Am J Obstet Gynecol 2001;184:297–302.

224. Kaufman HS, Cundiff G, Thompson J, et al. Suture rectopexy and sacral colpoperineopexy with Alloderm^® for perineal descent. Dis Colon Rectum 2000;43:A16.

225. Grimes CL, Quiroz LH, Gutman RE, et al. Long-term impact of abdominal sacral colpoperineopexy on symptoms of obstructed defecation. Female Pelvic Med Reconstr Surg 2010;16:234–237.

226. Cutait D. Sacro-promontory fixation of the rectum for complete rectal prolapse. Proc R Soc Med 1959;52[Suppl]:105.

227. Frykman HM, Goldberg SM. The surgical treatment of rectal procidentia. Surg Gynecol Obstet 1969;129:1225–1230.

228. Ripstein CB. Treatment of massive rectal prolapse. Am J Surg 1952;83:68–71.

229. Ripstein CB. Surgical care of muscle rectal prolapse. Dis Colon Rectum 1965;8:34–38.

230. Madiba TE, Baig MK, Wexner SD. Surgical management of rectal prolapse. Arch Surg 2005;140:63–73.

231. Luukkonen P, Mikkonen U, Jarvinen H. Abdominal rectopexy with sigmoidectomy vs rectopexy alone for rectal prolapse: a prospective, randomized study. Int J Colorectal Dis 1992;7:219–222.

232. Solomon MJ, Young CJ, Eyers AA, et al. Randomized clinical trial of laparoscopic versus open abdominal rectopexy for rectal prolapse. Br J Surg 2002;89:35–39.

233. Tou S, Brown SR, Malik AI, et al. Surgery for complete rectal prolapse in adults. Cochrane Database Syst Rev 2008;4:CD001758.

234. Delorme R. Sur le traitment des prolapses du rectum totaux pour l'excision de la muscueuse rectale ou rectocolique. Bull Mem Soc Chir Paris 1900;26:499–518.

235. Sielezneff I, Malouf A, Cesari J, et al. Selection criteria for internal rectal prolapse repair by Delorme’s transrectal excision. Dis Colon Rectum 1999;42:367–373.

236. Takesue Y, Yokoyama T, Murakami Y, et al. The effectiveness of perineal rectosigmoidectomy for the treatment of rectal prolapse. Surg Today 1999;29:290–293.

237. Altemeier WA, Culbertson WR, Schwengerdt C, et al. Nineteen years' experience with the one-stage perineal repair of rectal prolapse. Ann Surg 1971;173:993–1007.

238. Agachan F, Reissman P, Pfeifer J, et al. Comparison of three perineal procedures for the treatment of rectal prolapse. South Med J 1997;90:925–932.

239. Deen KI, Grant E, Billingham C, et al. Abdominal resection rectopexy with pelvic floor repair versus perineal rectosigmoidectomy and pelvic floor repair for full-thickness rectal prolapse. Br J Surg 1994;81:302–304.