(From Chapple CR, Artibani W, Cardozo LD, et al. The role of urinary urgency and its measurement in the overactive bladder symptom syndrome: current concepts and future prospects. BJU Int 2005;95:335–40.)

Figure 63–2 Overlap between conditions. OAB, overactive bladder; SUI, stress urinary incontinence; UUI, urgency urinary incontinence.

(From Wein AJ, Rackley RR. Overactive bladder: a better understanding of pathophysiology, diagnosis and management. J Urol 2006;175:S5–10.)

Coughing may induce a detrusor contraction—hence the sign of stress incontinence is only a reliable indication of urodynamic stress incontinence when leakage occurs synchronously with the first proper cough and stops at the end of that cough.

• Uncategorized incontinence is the observation of involuntary leakage that cannot be classified into one of the above categories on the basis of signs and symptoms.

Definition and Classification of Pelvic Organ Prolapse

More than 40% of women with urethral sphincter incompetence will have a significant cystocele (Cardozo and Stanton, 1980). Conversely, a history of stress incontinence associated with a mild or moderate cystocele is not specific for diagnosing genuine stress incontinence (Summitt et al, 1992). “Occult or latent” incontinence is urethral sphincteric incompetence masked by the presence of pelvic prolapse (Rosenzweig et al, 1992a, 1992b). Not infrequently, incontinent women may note the decrease or disappearance of stress incontinence episodes as the degree of prolapse worsens. The sign of occult stress incontinence is facilitated by the use of a speculum or pessary to reduce the prolapse while a stress maneuver is performed. The method to reduce vaginal prolapse to evaluate latent incontinence is not universally agreed on or standardized at this time.

A large rectocele may cause incomplete bowel evacuation and tenesmus. Sexual function consists of a complex interaction of biologic, psychologic, and social factors that affect both patient and partner. A complex relationship exists between anatomic descent of pelvic organs and physiologic function, and often these components do not correlate well with each other.

Vaginal examination allows the description of observed and palpable anatomic abnormalities and the assessment of pelvic floor muscle function. Segments of the lower reproductive tract should be considered as a substitute for terms as “cystocoele, rectocoele, enterocoele, or urethrovesical junction” because these terms convey an unrealistic certainty relating to the structures associated with the vaginal bulge, especially following previous prolapse surgery. This section draws heavily on the ICS standardization report, which commented on the evaluation of pelvic organ prolapse (Bump et al, 1996) as modified by the ICS standardization report 2002.

• Pelvic organ prolapse is defined as the descent of one or more of the following: anterior vaginal wall, posterior vaginal wall, and apex of the vagina (cervix/uterus) or vault (cuff) after hysterectomy. Absence of prolapse is defined as stage 0 support; prolapse can be staged from stage I to stage IV.

Pelvic organ prolapse can occur in association with UI and other lower urinary tract dysfunction and may on occasion mask incontinence.

• Anterior vaginal wall prolapse is defined as descent of the anterior vagina so that the urethrovesical junction (a point 3 cm proximal to the external urinary meatus) or any anterior point proximal to this is less than 3 cm above the plane of the hymen.

• Prolapse of the apical segment of the vagina is defined as any descent of the vaginal cuff scar (after hysterectomy) or cervix, below a point that is 2 cm less than the total vaginal length above the plane of the hymen.

• Posterior vaginal wall prolapse is defined as any descent of the posterior vaginal wall so that a midline point on the posterior vaginal wall 3 cm above the level of the hymen or any posterior point proximal to this is less than 3 cm above the plane of the hymen.

Pelvic floor muscle function can be qualitatively defined by the tone at rest and the strength of a voluntary or reflex contraction as strong, weak, or absent or by a validated grading system (e.g., Oxford 1 to 5). A pelvic muscle contraction may be assessed by visual inspection, palpation, electromyography, or perineometry. Factors that must be assessed are strength, duration, displacement, and repeatability. Rectal examination allows the description of observed and palpable anatomic abnormalities and is the easiest method of assessing pelvic floor muscle function in children and men. In addition, rectal examination is essential in children with UI to rule out fecal impaction.

Relationship Between Incontinence and Prolapse

Patients with severe prolapse may develop voiding symptoms as a result of urethral kinking, leading to obstruction that is worsened during straining effort (Richardson et al, 1983). For instance, a moderate or severe cystocele may promote urethral compression and kinking, pressure dissipation, and an increase in maximum urethral closure pressures (Bergman et al, 1988; Versi et al, 1998). Clearly, a number of storage urinary symptoms may occur in combination with prolapse. Risk factors contributing to the symptoms of urgency and frequency include age and urogenital atrophy. One study found that women with mild cystoceles had a 20% incidence of detrusor overactivity, and the incidence increased to 52% in those with moderate to severe cystoceles (Enhorning, 1961).

More than 40% of women with urethral sphincter incompetence will have a significant cystocele (Cardozo and Stanton, 1980). A complaint of stress incontinence associated with the appearance of a mild or moderate cystocele, however, is not specific for diagnosing genuine stress incontinence (Walters and Shields, 1988; Summitt et al, 1992).

Occult or latent incontinence is urethral sphincteric incompetence masked by the presence of pelvic prolapse (Rosenzweig et al, 1992). Not infrequently, incontinent women may note the decrease or disappearance of stress incontinence episodes as the degree of prolapse worsens. The office demonstration of the sign of occult stress incontinence is facilitated by the use of a speculum or pessary to reduce the prolapse while a stress maneuver is performed. Its presence may influence management options given to the patient. However, the method to reduce vaginal prolapse to evaluate latent incontinence is not universally agreed on or standardized at this time.

A large rectocele may cause incomplete bowel evacuation and tenesmus, leading some to splint or manually reduce the posterior vaginal segment or perineum to assist in defecation. Patients may describe stool becoming trapped in the rectocele pocket itself. Constipation and straining may worsen the symptoms and lead to left lower quadrant abdominal pain if impaction occurs.

The prevalence of fecal incontinence increases to 17% in populations with pelvic organ prolapse and UI (Jackson et al, 1997) compared with 2% to 3% in the general population (Leigh and Turnberg, 1982; Thomas et al, 1984). The most common mechanisms are an incompetent sphincteric mechanism (secondary to a structural defect or pudendal nerve damage) and overflow incontinence.

Sexual function consists of a complex interaction of biologic, psychologic, and social factors that affect both patient and partner. Various studies have described the presence of sexual problems in patients who present with pelvic organ prolapse and UI (Haase and Skibsted, 1988; Field and Hilton, 1993).

Pathophysiology of Incontinence And Prolapse

Continence

To understand the pathophysiology of urinary incontinence, it is important to be familiar with the micturition cycle and the physiology of normal storage and emptying of urine. The details of the physiology of micturition are provided in Chapter 60. The function of the lower urinary tract is to temporarily store a continuously increasing amount of urine at low pressure and expel it under appropriate circumstances. This is dependent on the coordinated activity of smooth and striated muscles in the bladder, urethra, and pelvic floor. The bladder and urethra constitute a functional unit, which is controlled by a complex interplay between the central and peripheral nervous systems and local regulatory factors (Andersson and Wein, 2004).

Bladder emptying and, conversely, urine storage rely on a complex series of finely tuned and integrated neuromuscular events that involve anatomic and neurologic mechanisms. It is a complex pattern of efferent and afferent signaling in parasympathetic, sympathetic, somatic, and sensory nerves. Malfunction at various levels may result in bladder control disorders, which roughly can be classified as disturbances of filling/storage or disturbances of voiding/emptying. Failure to store urine may lead to various forms of incontinence (mainly urgency and stress incontinence), and failure to empty can lead to urinary retention, which may result in overflow incontinence. A disturbed filling/storage function can, at least theoretically, be improved by agents decreasing detrusor activity, increasing bladder capacity, and/or increasing outlet resistance.

During bladder filling at physiologic rates, detrusor pressure remains nearly constant because of this property of accommodation (Klevmark, 1974), otherwise known as “tonus” or receptive relaxation. Accommodation accounts for the nearly flat cystometric curve that is seen during normal bladder filling. The viscoelastic properties of the bladder, based on its composition of smooth muscle, collagen, and elastin, normally produce a highly compliant structure. When accommodation is impaired, low bladder compliance ensues. This is manifest as a steep rise in detrusor pressure during bladder filling. In addition to the viscoelastic properties of the bladder, the neural control of the lower urinary tract and the anatomy and support of the sphincteric unit are important factors in the maintenance of urinary continence. The micturition reflex is normally under voluntary control and is organized in the rostral brainstem (the pontine micturition center [PMC]). Micturition results from a release from the negative inhibitor effect of the higher centers on the PMC. It requires integration and modulation by the parasympathetic and somatic components of the sacral spinal cord (the sacral micturition center) and the thoracolumbar sympathetic components (Fowler et al, 2008).

The reservoir function of the bladder requires a permanently low intravesical pressure over a wide volume range and a bladder outlet that remains firmly closed, even under conditions of abdominal pressure rises, during the filling phase of the bladder, which represents the majority of its activity cycle. Conversely, during voiding the bladder should be able to develop a sustained contraction of sufficient strength with opening of the bladder outlet offering a low resistance to urinary flow to assure adequate emptying of the bladder. During micturition, the first recorded event is sudden and complete relaxation of the striated sphincteric muscles, characterized by complete electrical silence of the sphincter electromyogram (Tanagho and Miller, 1970). This is followed almost immediately by a rise in detrusor pressure and concomitant fall in urethral pressure as the bladder and proximal urethra become isobaric (Yalla et al, 1980). The bladder neck and urethra open, and voiding ensues. Voluntary interruption of the stream is accomplished by a sudden contraction of the striated periurethral musculature, which, through a reflex mechanism, shuts off the detrusor contraction, aborting micturition. The sphincteric system is also designed to resist physiologic increases in abdominal pressure and thereby prevent the development of incontinence.

Normal storage of urine is therefore dependent on the following:

1. An intact normally functioning innervated lower urinary tract (bladder, urethra, sphincters, and pelvic floor)

2. Spinal reflex mechanisms that allow centripetal afferent nerve traffic to activate sympathetic and somatic pathways to the bladder outlet

3. Tonic inhibitory systems in the brain that suppress parasympathetic excitatory outflow to the bladder (Morrison et al, 2005)

It is important to identify the underlying bladder abnormality in a reproducible manner. Bladder abnormalities that cause UI include detrusor overactivity and low bladder compliance.

Disordered lower urinary tract function can result from the following:

• Disruption of the normal peripheral or central nervous system (CNS) control mechanisms

• Disordered bladder muscle function, either primary (of unknown etiology) or secondary to an identifiable pathology such as prostatic-mediated bladder outflow obstruction

Patients who have disordered lower urinary tract function in routine clinical practice represent a heterogeneous collection for most of whom there is no identifiable neurologic abnormality. Some of these patients will have a primary neural or muscular disorder (e.g., primary idiopathic detrusor overactivity) in contrast to postobstructive secondary detrusor overactivity, in which the major etiologic factor is likely to be peripheral disruption of local neuromuscular function.

Detrusor overactivity is a urodynamic observation characterized by involuntary detrusor contractions during the filling phase, which may be spontaneous or provoked. Detrusor overactivity may be phasic, terminal, or both (Abrams et al, 2002). There are certain patterns of detrusor overactivity:

• Phasic detrusor overactivity is defined by a characteristic wave form and may or may not lead to urinary incontinence. Phasic detrusor contractions are not always accompanied by any sensation or may be interpreted as a first sensation of bladder filling or as a normal desire to void.

• Terminal detrusor overactivity is defined as a single, involuntary detrusor contraction, occurring at cystometric capacity, which cannot be suppressed and results in incontinence usually resulting in bladder emptying (voiding). Terminal detrusor overactivity is a new ICS term: It is typically associated with reduced bladder sensation (e.g., in the elderly stroke patient when urgency may be felt as the voiding contraction occurs). However, in complete spinal cord injury patients there may be no sensation whatsoever.

• Detrusor overactivity incontinence is incontinence due to an involuntary detrusor contraction. In a patient with normal sensation, urgency is likely to be experienced just before the leakage episode.

• Other patterns of detrusor overactivity are seen (e.g., the combination of phasic and terminal detrusor overactivity, the sustained high pressure detrusor contractions seen in spinal cord injury patients when attempted voiding occurs against a dyssynergic sphincter).

Provocative maneuvers are used during urodynamics in an effort to provoke detrusor overactivity (e.g., rapid filling, use of cooled or acid medium, postural changes, hand washing).

Detrusor overactivity may also be qualified, when possible, according to cause. Examples follow:

• Neurogenic detrusor overactivity, when there is a relevant neurologic condition. This term replaces the term detrusor hyperreflexia.

• Idiopathic detrusor overactivity when there is no defined cause. This term replaces detrusor instability.

In clinical and research practice, the extent of neurologic examination/investigation varies. It is likely that the proportion of neurogenic-to-idiopathic detrusor overactivity will increase if a more complete neurologic assessment is carried out (Table 63–1).

Table 63–1 Causes of Detrusor Overactivity

Interpretation of Vesicourethral Dysfunction

Because so many different aspects of vesicourethral function can be impaired in neurologic conditions, interpretation may be difficult. The most important points to clarify in clinical practice are the presence or absence of neurogenic detrusor overactivity and the behavior of the distal urethral sphincter mechanism during voiding.

High-pressure bladders with detrusor–sphincter dyssynergia are prone to develop vesicoureteral reflux, which may ultimately lead to renal impairment. This can be clearly demonstrated by videocystometrography (VCMG). Preservation of renal function is of utmost importance in the management of patients who have chronic neurologic conditions. As a rule of thumb, those patients with a competent bladder outflow and an end filling pressure of 40 cm H₂O or higher before leakage occurs—(the detrusor leak point pressure) are at particular risk of developing upper urinary tract problems due to back pressure (McGuire et al, 1996).

Neurogenic detrusor overactivity can be defined as abnormal function of the bladder and urethra due to lesions affecting their innervation, either within the central nervous system or in the peripheral nerves of the lower urinary tract.

Urodynamic investigations are essential to characterize the nature of the detrusor and sphincteric abnormality to do the following:

• Identify patients who are at risk of deteriorating renal function as a result of the abnormally high bladder pressures

• Formulate the best treatment strategies to achieve efficient bladder emptying and continence and reduce the incidence of urinary tract infections and autonomic dysreflexia

The interpretation of urodynamics in such complex cases is often difficult and is best performed in specialist centers using VCMG.

Neurologic conditions can alter vesicourethral function by impairing the following:

• Detrusor activity

• Striated sphincter activity

• Bladder and urethral sensation

• Urethral smooth muscle activity

These alterations may occur in isolation or in combination. Identical abnormalities may also occur without clinical evidence of a neurologic deficit.

Classification of Neuropathic Bladder Dysfunction

Neuropathic bladder dysfunction may be classified into supraspinal, suprasacral, and infrasacral, according to the level of the lesion in relation to the pontine and sacral micturition centers:

• Supraspinal lesions include all cerebral diseases such as cerebral hemorrhage and thrombosis, dementia, tumors, arteriosclerosis, and Parkinson disease.

• Suprasacral lesions are all spinal cord lesions above the conus.

• Infrasacral lesions involve the conus or sacral roots and may be further subdivided into cauda equina lesions (within the spinal canal) and peripheral lesions (outside the spinal canal).

Neurologic lesions generally affect bladder and urethral function in a relatively consistent fashion, depending on the area affected and the completeness of the lesion. A basic understanding of the bladder dysfunction associated with each level of injury is necessary to interpret the results of urodynamics properly.

Idiopathic Detrusor Overactivity

Idiopathic detrusor overactivity can be associated with a variety of non-neurogenic clinical conditions such as bladder outlet obstruction, dysfunctional voiding, and inflammatory reactions (Hebjorn et al, 1976; Awad and McGinnis, 1983; Blaivas, 1988; Resnick et al, 1989; Fantl et al, 1990; Brading and Turner, 1994; Carlson et al, 2001). Elbadawi and colleagues (1993a, 1993b, 1993c) have proposed a possible explanation for age-related detrusor overactivity based on detailed ultrastructural studies of the bladder. They showed a characteristic structural pattern in bladder biopsy specimens obtained from elderly patients with detrusor overactivity. Electron microscopic changes (abundant, distinctive protrusion junctions and abutments) in patients with detrusor overactivity are thought to result in a diminished electrical resistance between detrusor muscle cells and, hence, a hyperexcitable state that results in involuntary detrusor contractions.

Bladder outlet obstruction causes increased voiding pressures and a prolonged duration of voiding. It has been hypothesized that this results in ischemic damage to the intramural ganglia and consequently to a partial denervation of the detrusor muscle (Brading and Turner, 1994). Experiments have shown that denervation of smooth muscle cells may result in supersensitivity to agonists, increased excitability, and increased electrical coupling. These changes may make the detrusor muscle susceptible to the development of involuntary contractions. The detrusor muscle hypertrophy on obstruction does not necessarily play a causative role (Brading and Turner, 1994). A somewhat contrasting and more evidence-based explanation comes from the observation of increased levels of nerve growth factor in the bladders of obstructed rats and humans and increased dimensions of afferent and efferent rat neurons (Steers et al, 1991).

In many cases the etiology is unclear—the cause of idiopathic detrusor overactivity is likely to be neurogenic rather than myogenic, particularly in view of the association between detrusor overactivity and aging. An increased number of putative afferent nerves in the bladder wall of female patients has been reported (Smet et al, 1997). Although no effect of intravesical capsaicin on idiopathic detrusor overactivity was found (Fowler, 2002), efficacy was found in 12 patients treated with the more potent analog resiniferatoxin (Cruz, 2002). This suggests the involvement of the afferent C fibers. The role of non-neurogenic factors cannot be excluded because abnormal cell-to-cell communications have been reported in the bladders of patients with idiopathic overactivity (Elbadawi et al, 1993b).

Bladder Compliance during Filling Cystometry

Bladder compliance describes the relationship between change in bladder volume and change in detrusor pressure. The observation of reduced bladder compliance during conventional filling cystometry is often related to relatively fast bladder filling: the incidence of reduced compliance is markedly lower if the bladder is filled at physiologic rates, as in ambulatory urodynamics.

• Compliance is calculated by dividing the volume change (ΔV) by the change in detrusor pressure (ΔPdet) during that change in bladder volume (C = V/ΔPdet). It is expressed in mL/cm H₂O.

• A variety of means of calculating bladder compliance has been described. The ICS recommends that two standard points be used for compliance calculations, but the investigator may define additional points. The standards points are as follows:

1. the detrusor pressure at the start of bladder filling and the corresponding bladder volume (usually zero), and

2. the detrusor pressure (and corresponding bladder volume) at cystometric capacity or immediately before the start of any detrusor contraction that causes significant leakage (and therefore causes the bladder volume to decrease, affecting compliance calculation). Both points are measured excluding any detrusor contraction.

The viscoelastic properties and neurologic innervation of the bladder under normal conditions allow it to store increasing volumes of urine at low pressure. Thus with normal compliance there is little or no pressure change as the bladder fills. Compliance (Δvolume/Δpressure) is expressed as mL/cm H₂O.

Although it is well established that high storage pressures can lead to incontinence and upper tract deterioration (McGuire et al, 1981; Comiter et al, 1997), at the present time there are no standardized or international agreed values to define normal, high, and low compliance. The calculated value of compliance is probably less important than the actual bladder pressure during filling. This is because the compliance value can change depending on the volume over which it is calculated. For this reason, numeric values of compliance are rarely reported. For example, sustained detrusor pressures of 35 to 40 cm H₂O or greater during storage, regardless of the bladder volume, can lead to upper tract damage, but similar damage can also occur at lower pressures (McGuire et al, 1981; Churchill et al, 1987). When the detrusor pressure exceeds the pressure capacity of the sphincter to resist that pressure, incontinence results.

Low bladder compliance denotes an abnormal volume-pressure relationship in which there is a high incremental rise in detrusor pressure during bladder filling. Low bladder compliance may be caused by changes in the elastic and viscoelastic properties of the bladder, changes in detrusor muscle tone, or combinations of the two. Clinically, low bladder compliance is most commonly seen in a variety of neurologic conditions, especially lower motor neuron lesions such as spina bifida and cauda equina syndrome. Obstructive uropathy, because of its effect on bladder ultrastructure, is known to cause low bladder compliance and has been identified as a urodynamic risk factor. Leng and McGuire (2003) showed that relief of obstruction by transurethral incision or resection of the prostate can improve compliance.

The clinical causes of low bladder compliance are listed in Table 63–2.

Table 63–2 Causes of Low Bladder Compliance

Neurogenic

Myelodysplasia

Shy-Drager syndrome

Suprasacral spinal cord injury/lesion

Radical hysterectomy

Abdominoperineal resection

Non-Neurogenic (Increased Collagen)

Tuberculous cystitis

Radiation cystitis

Chronic indwelling catheter

Bladder outlet obstruction

Sphincteric Mechanisms

The urethra has two main functions:

• Provide an effective continence mechanism, for the majority of the time (>99%) (storage phase)

• Allow adequate emptying from the bladder with minimal resistance during micturition (voiding phase)

The innermost mucosal layer in both sexes is organized in longitudinal folds, and during the storage phase, when the urethra is “closed,” this appears in a stellate configuration on cross section. Such a configuration allows significant distensibility, which is necessary during urethral “opening.”

The submucosal layer contains a vascular plexus, which may be involved in improving the seal of a “closed urethra” by transmitting the tension of the urethral muscle to the mucosal folds.

Apart from the obvious anatomic differences (the longer urethra and presence of a prostate gland in men), there are important differences in the configuration of the sphincter mechanisms between males and females. There are two powerful sphincter mechanisms in the male compared with the single weaker intrinsic sphincter mechanism with a weaker bladder neck and also a shorter urethra in the female.

Traditionally, the urethral sphincter is considered to be composed of two components: the internal sphincter, which represents a direct continuation of the detrusor smooth muscle, and the striated external sphincter (Tanagho, 1992). From a clinical standpoint, the bladder neck–proximal urethra normally functions as a sphincter in both sexes. Anatomically there is a unique mixture of smooth and striated muscle, intracellular matrix, and mucosal components contributing to the functional sphincter.

Male Sphincteric Mechanisms

In the male there are two important sphincteric mechanisms: (Hadley et al, 1986).

1. A proximal “bladder neck mechanism” consisting of the bladder neck, prostate, and prostatic urethra to the level of the verumontanum. This receives a dual innervation from the autonomic nervous system via parasympathetic and sympathetic fibers. The main motor control is likely to be provided by the sympathetic component, although there is some evidence for a contribution from the detrusor muscle where the predominant innervation is parasympathetic. This portion of the continence mechanism is removed during prostatectomy, leaving only the distal urethral sphincter mechanism to prevent urinary leakage.

2. A distal urethral mechanism at the apex of the prostate. This extends from the verumontanum to the proximal bulb and is composed of a number of structures that help to maintain continence. The male distal sphincter complex is composed of the prostatomembranous urethra, cylindrical rhabdosphincter (external sphincter muscle) surrounding the prostatomembranous urethra, and extrinsic paraurethral musculature and connective tissue structures of the pelvis. The rhabdosphincter is a concentric muscular structure consisting of longitudinal smooth muscle and slow-twitch (type I) skeletal muscle fibers that can maintain resting tone and preserve continence (Gosling et al, 1981; Turner-Warwick, 1983). Skeletal muscle fibers of the rhabdosphincter have been shown to intermingle with smooth muscle fibers of the proximal urethra, suggesting a dynamic and coordinated interaction (Burnett and Mostwin, 1998). The rhabdosphincter is invested in a fascial framework and supported below by a musculofascial plate that fuses with the midline raphe, which is also a point of origin for the rectourethralis muscle (Burnett and Mostwin, 1998). Superiorly, the fascial investments of the rhabdosphincter fuse with the puboprostatic ligaments (Steiner, 2000). This dorsal and ventral support probably contributes to the competence of the sphincter. The striated fibers of the extrinsic paraurethral muscle (levator ani complex), on the other hand, are of the fast-twitch (type II) variety (Gosling et al, 1981). During sudden increases in abdominal pressure, these fibers can contract rapidly and forcefully to provide continence. Continence has been shown to be maintained after inducing paralysis of the striated sphincter (Lapides et al, 1957; Krahn and Morales, 1965), indicating that this structure is not solely responsible for continence in men with an intact functioning bladder neck.

The proximal sphincter in the male bladder neck provides a powerful mechanism in both maintaining urinary continence and also prevents retrograde ejaculation of semen during sexual activity. In patients with a damaged distal urethral sphincter (e.g., a pelvic fracture-associated urethral disruption), continence can be maintained solely by the proximal bladder neck mechanism. Ultrastructurally, it consists of a powerful inner layer of muscle bundles arranged in a circular orientation.

The distal sphincteric mechanism is also extremely important, as evidenced by its ability to also maintain continence even when the proximal bladder neck mechanism has been rendered totally incompetent by surgical bladder neck incision or a prostatectomy. It is confined to the 3- to 5-mm thickness of the wall of the membranous urethra from the level of the verumontanum down to the distal aspect of the membranous urethra. It is composed mainly of extrinsic striated muscle, which is capable of the sustained contraction necessary for continence and to a lesser degree by intrinsic smooth muscle.

Unlike that in the female, where urethral support can be compromised as a result of childbirth and aging, in the male, compromise to the rhabdosphincter usually occurs after trauma or surgery (e.g., prostatectomy).

Female Sphincteric Mechanisms

Females are much more likely to suffer from UI due to sphincteric deficiency than males because of the much less powerful sphincteric mechanisms. The bladder neck is a far weaker structure than the male bladder neck and is often incompetent, even in nulliparous young women (Chapple et al, 1989). The bladder neck is poorly defined with the muscle fibers having a mainly longitudinal orientation.

Urinary continence usually relies on the integrity of the urethral sphincteric mechanism in females (analogous to the distal sphincter in men). Like the male distal mechanism, it is composed of a longitudinal intrinsic urethral smooth muscle and a larger extrinsic striated muscle component. This sphincter extends throughout the proximal two thirds of the urethra, being most developed in the middle one third of the urethra. Therefore in women the majority of the urethra should be considered to be sphincter active. Damage to the innervation of the urethral sphincter (in particular the pudendal nerve) by obstetric trauma reduces the effectiveness of this mechanism and predisposes to stress UI (Table 63–3).

Table 63–3 Comparison of Male and Female Sphincteric Mechanisms *

	MALE	FEMALE
Proximal bladder neck mechanism	Powerful	Weak
Distal urethral mechanism	Powerful	Extending along the majority of the length of the female urethra and prone to the effect of exogenous influences such as pelvic floor weakness and damage or denervation consequent on childbirth
Prostate	Further increases bladder outlet resistance	Not present
Urethra	Long	Short (≈3.5 cm)

* Indicates why females are much more likely to develop an incompetent urethral mechanism and are prone to urinary incontinence from intrinsic sphincter deficiency.

The principles underlying normal sphincteric function are integrated interactions among a number of factors:

1. Watertight apposition of the urethral lumen. Four urethral wall factors promote continence: (1) wall tension or external compression, (2) inner wall softness, (3) a filler material beneath the mucosa that helps to deform the mucosal folds into apposition, and (4) a lining of mucus provides the stickiness that enables cooptation of these mucosal folds (Zinner et al, 1980). Histologic cross sections of the urethra show that the urethra is not simply a closed tube; rather, there are numerous mucosal folds, between which are potential spaces for urine leakage. In women, estrogen and a cushion effect of the submucosal vasculature have been suggested as ancillary factors (Raz et al, 1972; Tulloch, 1974).

2. Compression of the wall around the lumen. External compression of the urethral lumen is achieved by (1) smooth and striated muscle tone, (2) phasic contractions of the smooth and striated musculature, (3) elastic and viscoelastic properties of the extracellular matrix, (4) mechanical factors related to transmission of abdominal pressure, and (5) structural (anatomic) support of the posterior urethral wall.

3. Structural support to keep the proximal urethra from moving during increases in pressure

4. A means of compensating for abdominal pressure changes (pressure transmission)

5. Neural control

Conditions Causing Sphincter Abnormalities

The causes of sphincteric dysfunction are different in men and women.

In men, sphincter abnormalities are most commonly caused by anatomic disruption after prostate surgery or, less commonly, by trauma or neurologic abnormalities.

In women with stress incontinence it is likely that there is a degree of sphincter weakness in all cases. The sphincter abnormality will lie on a spectrum from a mild degree where the predominant pathophysiologic abnormality is urethral hypermobility (urethral support defect), to those where irrespective of the urethral support defect there is severe intrinsic sphincteric insufficiency (ISD). Urethral hypermobility and support defects in women are most commonly associated with pregnancy and vaginal delivery, pelvic surgery, and chronic abdominal straining (e.g., chronic constipation). In some cases, loss of support many be secondary to neurologic injury. In such cases, nerve injury (e.g., pudendal nerve) can lead to muscle atrophy and subsequent failure of the support mechanism.

Sphincter Abnormalities in Women

Labor and delivery have long been thought to be risk factors for stress incontinence, urogenital prolapse, and anal incontinence. Vaginal delivery may be associated with direct injury to the pelvic soft tissues, as well as partial denervation of the pelvic floor and thus may play a role in the etiology of stress urinary incontinence (SUI).

There are four major mechanisms by which vaginal delivery might lead to sphincteric damage and incontinence:

1. Injury to connective tissue support by the mechanical process of vaginal delivery

2. Vascular damage to the pelvic structures as a result of compression by the fetus

3. Damage to the pelvic nerves or muscles or both as a result of trauma during parturition

4. Direct injury to the urinary tract during labor and delivery

Vaginal delivery can cause partial denervation with consequent reinnervation in most first deliveries (Allen et al, 1990). A growing body of evidence indicates that multiparity; forceps delivery; increased duration of the second stage of labor, partially related to epidural anesthesia; third-degree perineal tear; and high birth weight (>4000 g) are important factors leading to pudendal nerve injury (Snooks et al, 1986; Handa et al, 1996; Brown and Lumley, 1998; Rortveit et al, 2003). Snooks and colleagues (1984) studied the postpartum innervation of the external anal sphincter and found that the external anal sphincter and its innervation might be damaged by vaginal delivery but not by cesarean section. The abnormalities found were most marked in multiparas and correlated most strongly with a prolonged second stage of labor and forceps delivery.

Several conditions are associated with increased risk for ISD in women:

1. Previous urethral or periurethral surgery (e.g., anti-incontinence surgery, urethral diverticulectomy) may result in postoperative ISD from periurethral fibrosis, scarring, or denervation (Haab et al, 1996). Prevalence of ISD after two or more failed anti-incontinence operations was found to be as high as 75% (McGuire et al, 1980).

2. Neurologic insult may cause ISD. Sacral neurologic lesions have a variable effect on micturition, depending on the extent to which the neurologic injury affects the parasympathetic, sympathetic, and somatic systems (Gerstenberg et al, 1980; Blaivas and Barbalias, 1983; Wheeler et al, 1986; McGuire et al, 1988). In complete parasympathetic lesions, the bladder is areflexic and the patient is in urinary retention. When, in addition to a parasympathetic lesion, there is a sympathetic lesion, the proximal urethra loses its sphincteric function. Clinically, this results in incomplete bladder emptying, caused by the acontractile detrusor, and sphincteric incontinence, caused by the nonfunctioning proximal urethra (Gerstenberg et al, 1980; Blaivas and Barbalias, 1983). Somatic neurologic lesions affect pudendal afferent and efferent nerves (Blaivas, 1982). In addition to loss of perineal and perianal sensation, these lesions abolish the bulbocavernosus reflex and impair the ability to contract the urethral and anal sphincters voluntarily. Sacral neurologic lesions are caused by herniated disks, diabetic neuropathy, multiple sclerosis, and spinal cord tumors. They are also commonly encountered after extensive pelvic surgery such as abdominoperineal resection of the rectum and radical hysterectomy (Gerstenberg et al, 1980; Blaivas and Barbalias, 1983; Chang and Fan, 1983; McGuire et al, 1988).

3. Pelvic radiation therapy has been associated with damage to the mucosal seal coaptation of the urethra and local neurologic damage (Haab et al, 1996).

Sphincter Abnormalities in Men

Sphincter abnormalities in men are caused by trauma or neurologic injury. With prostatectomy, whether for benign disease or with radical prostatectomy, the distal urethral sphincter (particularly the rhabdosphincter) can be damaged by direct injury or injury to the nerve supply or supporting structures (Nitti, 2002). In some cases, there may be preexisting damage to the sphincter that cannot be accurately diagnosed preoperatively (Nitti, 2002). As in women (as noted earlier), radiation and neurologic lesions can cause sphincteric dysfunction. Finally, pelvic trauma or instrumentation that results in trauma to the distal urethral sphincter can result in incontinence, particularly when the PUS is absent or deficient.

Postprostatectomy incontinence, like any urinary incontinence, may be caused by bladder dysfunction, sphincter dysfunction, or a combination of both and may be associated with other lower urinary tract symptoms. Urodynamic investigations are helpful to rule out bladder outlet obstruction or significant bladder dysfunction. Urodynamic studies have demonstrated that the sphincter incompetence occurs as the sole cause in more than two thirds of patients. Isolated bladder dysfunction (detrusor overactivity, poor compliance, detrusor underactivity during voiding) is uncommon, occurring in less than 10% (Ficazzola and Nitti, 1998; Groutz et al, 2000). However, sphincter and bladder dysfunction can coexist in at least one third of incontinent patients.

Bladder dysfunction may occur de novo after prostatectomy, perhaps induced by bladder denervation; may be caused by outlet obstruction; or may be related to preexisting factors such as age. Impaired detrusor contractility and poor compliance resolved in the majority of patients within 8 months (Giannantoni et al, 2004). Decreased sphincter resistance may be due to tissue scarring in some cases and is reflected by a low urethral compliance; however, this parameter is difficult to measure (Groutz et al, 2000). Scarring may lead to an anastomotic stricture and is clinically suspected when both incontinence and decreased force of stream coexist. The preoperative length of the membranous urethra determined on magnetic resonance imaging (MRI) has been shown to be significantly related to time to postoperative continence. When urethral length was greater than 12 mm, 89% of the patients were continent at 1 year, versus 77% with 12 mm or less than this length (Coakley et al, 2002). Urodynamic studies revealed that a reduced functional urethral length was a predictive parameter of incontinence (Hammerer and Huland, 1997; Van Kampen et al, 1998; Wei et al, 2000). Different components of the urethra may also be involved. The urethral intrinsic component responsible for passive continence, as well as the extrinsic component responsible for active continence, may be involved as has been demonstrated in a study using urodynamic evaluation and alpha blockade (Pfister et al, 2002). This may explain passive incontinence despite a high voluntary urethral pressure or that measured during an active squeeze by the patient. Postoperative disruption of the innervation of the posterior urethra may also be involved and can affect both motor and sensory function (John et al, 2000; Bader et al, 2001).

Risk factors for incontinence after transurethral resection of the prostate (TURP) have not been clearly defined, probably because the incidence is so low. However, previous brachytherapy does predispose to post-TURP incontinence, as does significant bladder overactivity and preoperative urinary incontinence. Preoperative lower urinary tract symptoms including urgency incontinence and “overflow incontinence” may improve with de-obstruction secondary to extirpative surgery. Whenever a functional abnormality of the bladder or sphincter abnormalities is suspected, preoperative urodynamic assessment is advised before TURP.

UI occurring after radical prostatectomy (RP) is a significant problem. Although its rate has lessened in recent years primarily due to a better understanding of pathophysiology and improvements in surgical technique, its prevalence has increased due to the dramatic increase of laparoscopic and robotic-assisted laparoscopic radical prostatectomy as standard treatments for men with prostate cancer in developed countries (Hu et al, 2003, 2004) within the past decade. Based on the body of available data on postoperative continence, it would appear that incontinence rates are similar between open and laparoscopic/robotic approaches. Several studies have compared the techniques either prospectively in nonrandomized fashion (Anastasiadis et al, 2003; Jacobsen et al, 2007) retrospectively (Ahlering et al, 2004) or via limited meta-analysis (Rassweiler et al, 2003; Salomon et al, 2004), and similar continence rates were found. Further prospective comparative studies with open surgery are necessary. Perineal prostatectomy is done by only a limited number of urologists but is still advocated for obese patients. The continence rate was reported as similar to the retropubic route (Weldon et al, 1997; Gray et al, 1999; Harris, 2003).

Data from large multicenter studies and prostate cancer databases suggest that following RP, 1% to 40% of patients complain of persistent urinary incontinence. The incidence of postprostatectomy incontinence (PPI) depends on the definition of UI and the length of follow-up (Olsson et al, 2001; Krupski et al, 2003; Rodriguez et al, 2006). Recent reports of large cohorts use definitions that include “total control/perfect continence,” “occasional leakage but no pad,” and “less than one pad.” Because one third to one half of men who do not wear pads will have occasional leakage of urine, it is important to distinguish among those men who leak enough to require pad use and those who do not. It has been demonstrated that health-related quality of life is strongly correlated with the level of incontinence, and wearing a pad more significantly affects the quality of life than wearing no pad (Litwin et al, 2000). In addition, not all men who leak will elect to have further treatment. Most large cohort studies indicate that between 6% and 9% of patients undergo subsequent surgical treatment for PPI following prostate cancer surgery (Stanford et al, 2000; Begg et al, 2002; Steineck et al, 2002; Penson et al, 2005).

Reported risk factors for incontinence following radical prostatectomy include the following:

• Patient age at surgery

• Stage of disease

• Surgical technique including nerve sparing

• Preoperative bladder function and urinary continence status

• Prior radiation therapy

• Preoperative length of the membranous urethra

• Prior TURP

• Vascular comorbidities

Preoperative sphincteric insufficiency (demonstrated by either the preexisting clinical sign of SUI or the urodynamic finding of lower maximal urethral closure pressure) predicts postoperative SUI (Wei et al, 2000; Majoros et al, 2006). Preoperative bladder dysfunction can also contribute to postoperative incontinence. Preexisting abnormalities of detrusor function may predispose to leakage following surgery, especially neurogenic detrusor overactivity due to Parkinson disease, dementia, or spinal cord injury (Khan et al, 1991).

Advancing age as a risk factor is supported by several studies (Leandri et al 1992; Zincke at al, 1994; Eastham et al, 1996; Catalona et al, 1999; Horie et al, 1999; Moore et al, 2007). Others have found advancing age and the number of comorbidities to have a negative impact on the recovering time for continence during the first year after radical prostatectomy (Young et al, 2003). Rogers and colleagues demonstrated that age affected postoperative continence status following laparoscopic RRP. In those younger than 50 years old, 100% achieved 0 to 1 pad per day continence at 1 year, which decreased to 91% and 81% for those 50 to 59 years old and older than 60 years, respectively (P < 0.01) (Rogers et al, 2006). Strasser and colleagues (1998) hypothesized that age-related sphincteric changes may be responsible for the age-related increase in postoperative SUI and successfully demonstrated a progressive reduction in sphincter striated muscle cells with age.

Most large series have found no correlation between the stage of disease and incontinence rates (Eastham et al, 1996; Jonler et al, 1996; Lowe, 1996; Catalona et al, 1999; Pierorazio et al, 2007). However, in certain cases, the stage of disease may affect the surgical technique (i.e., nerve sparing) and incontinence rates may be higher, but this appears to be a reflection on surgical technique and not disease stage (Zincke et al, 1994).

Regarding surgical technique, the many parameters involved in continence have resulted in difficulties in understanding the benefit of certain technical points.

Bladder neck preservation has been reported to improve continence at 3 months (Lowe, 1996), but no difference was evident at 6 and 12 months (Poon et al, 2000; Srougi et al, 2001).

The role of nerve sparing on subsequent continence is debated. Nerve sparing has no significant impact according to Steiner and colleagues (1991), and Lepor and colleagues (2004) recently confirmed this. Others did find benefit (Wei et al, 2000). In particular, Nandipati and colleagues reported that in a cohort of 152 patients followed prospectively, bilateral nerve-sparing surgery was associated with a shorter time to regain continence, as well as improved long-term continence rates compared with non–nerve-sparing surgery. They additionally found that increased age was a risk factor for postprostatectomy incontinence (Nandipati et al, 2007). Burkhard and colleagues similarly demonstrated a positive effect of nerve-sparing surgery on postoperative continence. In a prospective cohort study of 536 patients, PPI developed in 1 of 75 (1.3%), 11 of 322 (3.4%), and 19 of 139 (13.7%) with attempted bilateral, attempted unilateral, and without attempted nerve sparing, respectively. Attempted nerve sparing was, in fact, the only statistically significant factor influencing urinary continence after RRP in this cohort (Burkhard et al, 2006).

Pelvic Floor Support Mechanisms

Two major factors govern the predisposition of the human female to develop either stress UI or prolapse, or both.

1. After the evolutionary changes in pelvic floor anatomy resulting from the change from a four-limb to an upright two-limb posture; the pelvic floor has had to adapt from its mainly muscular utility (of moving the tail) in four-legged animals to a supportive function in the human. This has generated a decrease in reliance on muscle mass and an increase in the importance of connective tissue support mechanisms.

2. The human reproductive process consists of a relatively large fetus with a large bony cranium that must pass through the pelvic floor during labor and delivery.

Certainly, repeated episodes of vaginal traction can stretch, tear, or weaken urethral muscles and contribute to a chronically weakened urethra characterized by low VLPP or low urethral closure pressures. This is suggested to be typical of intrinsic sphincter deficiency (ISD). Clearly, however, there is no adequate categorization of ISD because most of the measures have limited sensitivity by virtue of their nature and inability to separate out the different components of urethral function. Electrophysiologic studies, MRI, and sonography can be used to quantify the morphologic and functional decline in striated sphincter tone, but analysis of the etiologic factors is difficult because the pathophysiology develops over many years. Successful operations can restore urethral position but probably do not reverse damage to urethral sphincter function, and there are still many question marks about whether this is ever possible.

The complex etiology of stress UI and pelvic organ prolapse and the long time period over which these complications and problems develop strongly supports the need for prospective studies taking account of both nature and nurture. This may allow the investigation of early timely interventions to prevent the further deterioration of their support tissues and urethral closure mechanisms.

From Dichotomy to Continuum between Hypermobility and Intrinsic Sphincter Deficiency

An integrated approach involving anatomy and physiology is essential to an improved understanding of pathophysiology. Clearly a continent bladder outlet in the females depends on multiple factors (resting tone, active contraction, external compression, pressure transmission, integrity of configuration), and the focal point appears to be the midpoint of the urethra. In patients with stress incontinence there is clearly a spectrum between the two components of ISD and urethral hypermobility. Clearly, some patients have primary sphincteric problems, whereas others have an adequately functioning sphincter but significant hypermobility. The majority lie somewhere between these two extremes.

In recent years there has been a gradual change from this dichotomic classification of stress incontinence: hypermobility or ISD. ISD alone is rare, and urethral hypermobility may occur commonly without significant ISD, but usually there is a combination of both in individual patients.

Figure 63–3 demonstrates the potential balance between the relative proportions of ISD and pelvic floor muscle weakness hypermobility seen in any individual patient.

Figure 63–3 The potential balance between the relative proportions of intrinsic sphincter deficiency (ISD) and pelvic floor muscle weakness hypermobility seen in any individual patient.

This evolution in our understanding in part followed the development of the concept of Valsalva leak point pressure (VLPP) introduced by McGuire in 1995 and also from analysis of long-term results of incontinence surgery. Despite lacking standardization of recording methods, as well as lacking consensus on how to deal with any associated prolapse, low VLPP (inferior to 60 cm H₂O has been suggested) has been widely considered as an indicator of ISD. Reports in the literature vary considerably regarding the extent of correlation between VLPP and outcome. Similarly, urethral pressure profilometry has been extensively studied in this context but there is a similar lack of correlation both in terms of predictive value and outcome. For instance, a low-pressure urethra may not leak, whereas the high-pressure urethra may. Also, a number of studies have failed to show any correlation between postoperative UPP and outcome. Cystography, although an invasive technique requiring radiologic input, provides a reproducible and easily interpreted technique, which allows the relative contribution of both sphincteric weakness and hypermobility to be assessed (Blaivas and Olsson, 1988).

Consequently, some degree of ISD may exist in many patients where hypermobility is believed to be the only cause of stress incontinence (Kayigil et al, 1999). This understanding supports the increasing use of suburethral slings to treat stress incontinence, whereas formerly this procedure was exclusively suggested for patients with recurrent stress incontinence or severe ISD (Chaikin et al, 1998). Clearly, vaginal support is an important factor in maintaining urinary continence, but in addition one needs to consider the importance of ISD, a discussion hampered by the lack of adequate techniques to evaluate ISD at present.

Urethral Support Mechanisms

The urethral support mechanism comprises all the structures extrinsic to the urethra that offer a supportive backplate on which the proximal urethra and midurethra lie.

The anatomic support is derived from the following:

1. Fascial structures

In recent years the important potential role of connective tissue in the overall physiologic function of the lower urinary tract and pelvic floor has been progressively recognized. The age-related weakening of connective tissues can influence both tissue and organ function, and it is likely that the increasing attention on connective tissue changes will provide a better understanding of pathophysiology, which hopefully will be useful in leading to the more effective management of stress UI and vaginal prolapse.

2. Musculature

Urethral hypermobility (increased mobility) is thought to be due to loss of the normal extrinsic support of the urethra by a weakness of both endopelvic fascia and pelvic floor muscles, usually initiated by childbirth but worsened by aging and alterations of hormone level. Stretching, tearing, and avulsion of the levator muscles results in the urogenital hiatus becoming longer and wider. This results in chronic anterior displacement of pelvic organs with loss of any organ support at rest and especially during straining. Stretching or tearing of the cardinal and utero-sacral ligaments may result in anterior displacement of the uterus at rest or during straining. This results in stretching of the vaginal wall and loss of the normal superior vaginal sulcus and vaginal folds.

The consequence of this is a rotational descent of the proximal urethra away from its retropubic position and the eventual development of SUI.

Certainly in lateral cystourethrograms, the main anatomic change is loss of the posterior urethra-vaginal angle with the urethra and trigone falling into the same plane (Jeffcoate and Roberts, 1952; Hodgkinson, 1953; Hodgkinson, 1970).

Consequently, radiographic studies cannot distinguish between lateral or central defects in vaginal wall support because they appear in the same sagittal plane. It is necessary to examine the patient to determine which defect is present and to what extent. Because the proximal urethra rotates out of the focal plane of ultrasonographic probes or MRI, coronal images of vaginal relaxation cannot provide adequate anatomic information during leakage.

Despite the extensive data about anatomic defects, it is difficult to correlate the influence of these defects, the vaginal position, and the urethral closure mechanism. Not all women with stress incontinence have vaginal prolapse; clearly, prolapse repairs do not always cure the stress incontinence and, conversely, women who redevelop stress incontinence after an apparently successful operation do not always have a recurrence of prolapse (Wall et al, 1994).

In conclusion, support and suspension of the pelvic organs is dependent on a healthy pelvic floor striated muscle, intact robust connective tissue, and their attachment to the bony frame of the pelvis. A useful analogy to visualize these factors is to consider a ship lying moored in dock. One can consider the ligaments as the mooring ropes and the pelvic floor musculature as the water on which the ship sits. If the dock is emptied, the influence of the support of the water disappears, leaving all of the tension applied to the mooring robes, which will inevitably weaken and fracture (Fig. 63–4).

Figure 63–4 Analogy demonstrating the support provided by the A, pelvic floor (water) and ligaments (ropes) to the pelvic organs (ship). B, Consequences of a pelvic floor muscle weakness with increasing strain placed on ligamentous structures. C, Ligamentous damage as a consequence of loss of pelvic floor muscle weakness.

Berglas and Rubin (1953) showed that in the nulliparous patient, the lower one third of the vagina is oriented more vertically, while the upper two thirds deviate horizontally, thereby maintaining the vaginal axis in an almost horizontal position (Fig. 63–5A and B). This configuration is maintained by the posterior attachments of the cervix with the cardinal and utero-sacral ligaments and by the anterior position of the urogenital hiatus. During stressful maneuvers such as coughing or straining, the levator hiatus is shortened anteriorly by contraction of the pubococcygeus muscles (Fig. 63–5C). In the case of genital prolapse when the levator ani support is lost, the vaginal axis becomes more vertical, the urogenital hiatus broadens, and fascial supports are strained (Fig. 63–5D).

Figure 63–5 A, In the nulliparous patient, the lower one third of the vagina is oriented more vertically, whereas the upper two thirds deviate horizontally, thereby maintaining the vaginal axis in an almost horizontal position. B, In the nulliparous patient, the lower one third of the vagina is oriented more vertically, whereas the upper two thirds deviate horizontally, thereby maintaining the vaginal axis in an almost horizontal position. C, During stressful maneuvers such as coughing or straining, the levator hiatus is shortened anteriorly by contraction of the pubococcygeus muscles. D, In the case of genital prolapse when the levator ani support is lost, the vaginal axis becomes more vertical, the urogenital hiatus broadens, and fascial supports are strained.

Connective Tissue

The pubourethral ligament attaches the midurethra to the inferior side of the pubic symphysis and prevents its downward rotational descent (Zacharin, 1963). It works in conjunction with the pubourethralis muscle, a subdivision of the levator ani muscle that forms a sling around the proximal urethra. Together they form the midurethral complex. In particular it has been postulated that an elongation of the posterior pubourethral ligaments may be a significant contributory factor to the loss of urethral support seen in stress incontinence. An important fascial support for the urethra is provided by the urethro-pelvic ligament, which attaches the urethra to the tendineus arc (Rovner et al, 1997).

In addition, the endo-pelvic fascia/pubocervical fascia, extending between the bladder and the vagina, suspends and attaches the vagina and cervix to the pelvic sidewall and to each arcus tendineus fascia pelvis, thereby offering posterior support to the bladder and bladder neck. It has two surfaces: the perivesical fascia on the vaginal side and the endo-pelvic fascia on the abdominal side. Its upper zone supports the bladder above the cervix, the middle zone supports the trigone, and the lower zone supports the bladder neck. Laxity of the fascia in each of the zones will result in uterine prolapse, cystocele, and urethrocele, respectively (DeLancey, 2001).

The arcus tendineus fasciae pelvis are tensile structures located bilaterally on either side of the urethra and vagina that act like the ropes of a suspension bridge and provide the necessary support needed to hang the urethra on the anterior vaginal wall. They originate as fibrous bands from the pubic bone and broaden out as aponeurotic structures moving dorsally to insert into the ischial spine (Fig. 63–6).

Figure 63–6 The arcus tendineus fasciae pelvis and its relationship to bony structures. The x’s refer to placement of sutures during a paravaginal or vagino–obturator repair.

The cardinal ligaments and the more medially placed utero-sacral ligaments support the uterus and cervix; their relaxation results in uterine prolapse (Fig. 63–7).

Figure 63–7 Relationship of ligaments to pelvic organs.

Pelvic Floor Musculature

The pelvic floor musculature, represented by the levator ani muscles, carries the weight of the pelvic contents and prevents the abdominal pressure from stretching the ligamentous support structures. The levator ani includes the puborectalis muscle, which surrounds the rectum connecting the pubic bones anteriorly in a U-shaped configuration, the pubococcygeus muscle, which crosses from the pubis to the coccyx, and the iliococcygeus muscle. The last one arises laterally from the arcus tendineus levator ani and forms a horizontal sheet that spans the posterior opening of the pelvis, providing a shelf on which the pelvic organs lie. The urethra and vagina pass through an aperture in the levator musculature, the urogenital hiatus. The levator ani contains not only type I fibers providing resting tone but also type II (fast-twitch) fibers that, under stress, maintain the urethral closure and prevent stretching of the pelvic ligaments.

The constant muscle tone (Critchley et al, 1980) maintained by predominantly type I (slow-twitch) striated muscle fibers compresses the vagina and urethra anteriorly toward the pubic bone and keeps the hiatus closed (DeLancey and Hurd, 1998).

The pudendal nerve provides somatic innervation to the striated muscle of levator ani, as well as to the striated muscle within the external anal and urethral sphincters. The intrapelvic somatic fibers travel along the anterior vaginal wall from the sacral segments S2 to S4 to provide asomatic nerve supply to the pelvic floor (Borirakchanyavat et al, 1997). Neuromuscular injuries have been proposed as an important factor in predisposing to pelvic floor dysfunction. Childbirth and age are the two major factors predisposing to pelvic floor denervation (Smith et al, 1989). Other factors include pelvic surgery (rectal and vaginal surgery with extensive pelvic dissection), radiotherapy, and congenital neurologic conditions such as spina bifida and muscular dystrophy (Fig. 63–8).

Figure 63–8 Disposition of the levator ani.

Current Theories Relating to the Maintenance of Continence

The traditional view has been that urethral support is critical to the maintenance of continence during increases in abdominal pressure. It was thought that loss of structural urethral support resulted in varying degrees of descent of the bladder neck and urethra, and it was thought that the resultant urethral hypermobility is the principal cause of SUI (Hodgkinson, 1953; Green, 1968; McGuire et al, 1976; Hodgkinson, 1978; Blaivas, 1988; Walters and Jackson, 1990). It was also thought that the normally supported bladder neck and proximal urethra sit in a high retropubic position and that increased intra-abdominal pressure is transmitted equally to both bladder and urethra. In cases of stress-induced urethral descent, the urethra attains a low, dependent position. Increased intra-abdominal pressure is then transmitted unequally to the urethra and bladder, and when bladder pressure exceeds urethral pressure, SUI ensues (McGuire and Herlihy, 1977; Constantinou and Govan, 1982; Westby et al, 1982; Constantinou, 1985; Bump et al, 1988; Rosenzweig et al, 1991). However, the fact that many women are totally continent despite clinical evidence of urethral hypermobility somewhat refutes that theory. This implies that the normal urethra remains closed even during stress-induced descent (Walters and Diaz, 1987).

The most commonly held view of the pathophysiology of stress incontinence secondary to urethral hypermobility is based on DeLancey’s theory of urethral support, the so-called hammock theory (DeLancey, 1994). In anatomic specimens in which increases in abdominal pressure were simulated, he found that the urethra lies in a position where it can be compressed against a hammock-like musculofascial layer on which the bladder and urethra rest (DeLancey, 1994). In this model, it is the stability of this supporting layer rather than the position of the urethra that determines stress continence. Urethral support is supplied by connective tissue and muscle arranged to resist the downward pressure created by increases in abdominal pressure. The urethra is intimately connected to the anterior vaginal wall, and the connections of these two structures to the levator ani muscle complex and arcus tendineus fascia pelvis (ATFP) determine the stability of the urethra. The ATFP is a fibrous fascial band that stretches from the pubic bone anteriorly to the ischial spine. The fascial covering of the levator ani consists of two leaves: the endopelvic fascia (abdominal side) and the pubocervical fascia (vaginal side). The two leaves fuse laterally to attach on to the ATFP, creating a hammock of support under the urethra and bladder neck. Normally, with rises in intra-abdominal pressure, the urethra is compressed against the supporting structures, which act like a backboard and prevent loss of urine (DeLancey, 1997). When the supporting structures fail, there can be rotational descent of the bladder neck and proximal urethra during increases in abdominal pressure. If the urethra opens concomitantly, SUI ensues. If this supporting layer establishes its stability, although at a lower level, continence may still be preserved (DeLancey, 1996). These findings have important clinical implications that support the contention that the primary goal of surgery for sphincteric incontinence in women is to provide a backboard against which the bladder neck and proximal urethra can be compressed during increases in abdominal pressure. In fact, some of the most popular procedures for the treatment of stress incontinence do not change the position of the urethra or even correct mobility (Lo et al, 2001; Sarlos et al, 2003; Minaglia et al, 2005). An additional observation by DeLancey (1990) is that the medial portion of the levator ani muscle has a direct connection to the endopelvic fascia and the vaginal wall, and contraction of the levators contributes to stabilization of the urethra during sudden increases in abdominal pressure (e.g., a cough).

In a series of elegant MRI studies of the pelvic floor, stress incontinence was found to occur when there was unequal movement of the anterior and posterior walls of the bladder neck and proximal urethra during stress. With good anterior wall support and a lax posterior wall, the urethral lumen is literally pulled open as the posterior wall moves away from the anterior wall, resulting in incontinence (Yang et al, 1991; Yang et al, 1993; Mostwin et al, 1995). Conversely, one would expect less incontinence if there is similar laxity of the anterior wall.

Furthermore, ultrasound imaging of the bladder neck and proximal urethra, in both normal and stress-incontinent patients, demonstrated that opening (funneling) of the bladder neck was the common denominator underlying stress incontinence, not urethral hypermobility. Anterior support of the urethra is provided by the pubourethral ligaments, attaching the midurethra to the inferior aspect of the pubic bone. These work in conjunction with the pubourethralis muscle of the levator ani complex, which forms a sling around the proximal urethra and prevents rotational descent (Plzak and Staskin, 2002). Together, the two structures form the midurethral complex, which is now thought to play a significant role in the maintenance of stress continence along with the suburethral hammock described earlier. According to Mostwin and colleagues (1995): “It is the relative, not the absolute, position of the urethra with respect to the pubis that may contribute to urethral opening in stress incontinence. The length and strength of the suburethral complex may vary in patients. In those with shorter, stronger anterior fascial support, only a small loss of suburethral support may be sufficient to permit distraction (of the anterior and posterior urethral walls) during rotational descent.” In contrast to the explanation that proposes a pure transmitted pressure effect, the mechanism of shear force suggests that transmitted energy mechanically opens the proximal and middle urethra (Plzak and Staskin, 2002).

The exact point of continence in the urethra, if such a point exists, has been debated. For years it was thought to be the bladder neck; however, the work of Constantinou and Govan (1981) and Petros and Ulmstem (1990, 1998) suggested that the midurethra supported anteriorly by the pubourethral ligaments is the most important zone of continence. Constantinou and Govan (1981) showed that pressure change ratios in the urethra/bladder are highest at the midurethra in continent control patients and in postsurgical patients after endoscopic bladder neck suspension, whereas the high midurethral pressure is lost in women with stress incontinence. In 1990 Petros and Ulmstem published the integral theory of stress and urge incontinence. According to the theory, female stress incontinence and urgency incontinence have a common etiology. They proposed that support of the anterior vaginal wall is provided by three separate but synergistic mechanisms. The anterior pubococcygeus muscle lifts the anterior vaginal wall to compress the urethra; the bladder neck is closed by traction of the underlying vaginal wall in a backward and downward fashion; and the pelvic floor musculature, under voluntary control, draws the hammock upward, closing the bladder neck. Overall laxity of the anterior vaginal wall causes dissipation of all of these forces, resulting in stress incontinence. They further suggested that laxity of the anterior vaginal wall causes activation of stretch receptors in the bladder neck and proximal urethra that can trigger an inappropriate micturition reflex, resulting in detrusor overactivity. These theories are not mutually exclusive, and there is probably a least some truth in each. The fact that both bladder neck and midurethral slings have been shown to be successful in treating stress incontinence supports all these theories (Ulmsten et al, 1998; Olsson and Kroon, 1999; Nilsson and Kuuva, 2001; Rezapour and Ulmsten, 2001; Nilsson et al, 2004). Further reports of successful treatment of mixed incontinence with slings lend credence to the integral theory.

In 1980 McGuire and colleagues first introduced the concept of intrinsic sphincteric deficiency (ISD) when, in a retrospective analysis of the results of anti-incontinence surgery, they noted that some women who had undergone multiple retropubic operations had a deficient sphincteric mechanism characterized by an open bladder neck and proximal urethra at rest with minimal or no urethral descent during stress. They stated that ISD denotes an intrinsic malfunction of the urethral sphincter, regardless of its anatomic position. Contemporary theories suggest that all patients with sphincteric incontinence have some degree of ISD. Considerations that support this theory include the fact that the normal urethra is intended to remain closed no matter what the degree of stress or rotational descent. Furthermore, many women with urethral hypermobility remain continent (Versi et al, 1986). Finally, urethral hypermobility and ISD may (and often do) coexist in the same patient. Because urethral hypermobility and ISD often coexist, we believe that they do not necessarily define discrete classes of patients (Nitti and Combs, 1996; Fleischmann et al, 2003). Thus these parameters may be used to characterize incontinence but not necessarily classify patients. Classification systems have been devised to characterize stress incontinence by the presence or absence of urethral mobility, the degree of mobility, and the presence of rotational descent (Green, 1962; McGuire et al, 1976; Blaivas and Olsson, 1988). However, it is recommended that rather than “classifying” stress incontinence, it makes more sense to simply characterize it by two parameters—the degree of urethral mobility and sphincter strength.

Epidemiology of Urinary Incontinence and Prolapse

The authors would like to acknowledge the major contribution made by the International Consultation on Incontinence Committee 1 on Epidemiology of Urinary and Faecal Incontinence and Pelvic Organ Prolapse (Milsom et al, 2009), on which the following section is based.

Background Terminology

Epidemiology is the scientific study of the distribution and determinants of disease in people. The epidemiology and natural history of UI vary greatly between the genders. Causes and risk factors are different. Therefore it is most practical to discuss the epidemiology of incontinence in men and women separately.

Prevalence is defined as the probability of experiencing a symptom or having a condition or a disease within a defined population and at a defined time point. The concept is important for establishing the distribution of the condition in the population and for projecting the need for health and medical services.

Incidence is defined as the probability of developing the condition under study during a defined time period. Incidence is usually reported for a 1-, 2-, or 5-year time interval.

When discussing the epidemiology and impact of incontinence, it is important to distinguish between its prevalence and incidence.

The relative risk (RR) estimates the magnitude of an association between exposure and a condition and indicates the likelihood of having the condition in the exposed group relative to those who are not exposed (e.g., do not have the risk factor). An RR of 1.0 indicates that the rates in the exposed and nonexposed groups are identical and thus that there is no association between the exposure and the condition in that specific dataset. A value greater than 1.0 indicates a positive association or an increased risk. An RR of 2.5 for UI indicates that there is a 2.5 times increased risk or that the persons in question are 250% more likely to have incontinence than those without the risk factor.

The odds ratio (OR) is the odds for having a risk factor in persons with a condition divided by the odds among those without the condition. An OR of 2.5 for UI may be interpreted as meaning that in this sample the odds in favor of having incontinence are 2.5 times higher among those with the risk factor than among those without.

For a condition with high prevalence, like UI or pelvic organ prolapse, OR and RR will not be identical, but in practice the results can be interpreted similarly. Results should always be given with a 95% confidence interval (CI).

Incontinence

UI is a common symptom that may affect women at all ages, and there is a wide range of severity and nature of symptoms. UI is not a life-threatening disease, but the symptoms may seriously influence the physical, psychologic, and social well-being of the affected individuals.

1. Pregnancy, labor, and vaginal delivery versus caesarean section are significant risk factors for later UI, but the strength of this association diminishes substantially with age (level 1 evidence).

2. Although several specific parturition factors such as instrumental delivery and birth weight are risk factors for UI in the postpartum period, their association with UI in later life is weak or nonexistent, suggesting that changes in birthing practices in developed countries are unlikely to affect UI in older age (level 2 evidence).

3. Additional evidence has now established body mass as important, modifiable risk factors for UI (level 1 evidence).

4. Physical function also appears to be an independent risk factor for UI in older women. Whether improvement in physical function leads to a reduction in UI remains to be established (level 2 evidence).

5. Evidence from two blinded, randomized controlled trials indicate that oral estrogen, with or without progestogen, is a significant risk factor for UI in women age 55 and older (level 1 evidence).

6. Diabetes is a risk factor for UI in most studies. Although diabetic neuropathy and/or vasculopathy are possible mechanisms by which diabetes could lead to UI, no mechanism has been established, nor is it clear whether prevention or treatment of diabetes, separate from weight reduction, will reduce the risk of UI (level 2 evidence).

7. Menopause, as generally defined, does not appear to be an independent risk factor for stress UI (level 2 evidence).

8. Hysterectomy remains a possible risk factor for later UI, but the evidence is inconsistent (level 2 evidence).

9. Moderate to severe dementia in older women is a moderate to strong independent risk factor for UI (level 2 evidence). Whether interventions to maintain or improve cognitive functioning also reduce UI has not been evaluated.

10. Mild loss of cognitive function in community-dwelling women, separated from physical function and other factors, increases the risk of UI slightly, if at all, but may increase the impact of UI (level 2 evidence).

11. Data from twin studies suggests that there is a substantial genetic component to UI (level 1 evidence).

12. Other potential risk factors including smoking, diet, depression, constipation, UTIs, and exercise, although associated with UI, have not been established as etiologic risk factors and are, in fact, difficult to study with observational data because of the potential for unmeasured confounding and questions of direction of the association (level 3 evidence).

General Comments

Virtually all epidemiologic studies rely on participant self-reporting of UI. Multiple studies have examined the association between self-reported incontinence and clinically demonstrable incontinence and by type of incontinence based on self-reported compared with type based on clinical diagnosis. Although self-reporting and clinical diagnosis are highly correlated, they are inherently different. Sandvik and colleagues (1995) validated diagnostic questions used in a survey against a final diagnosis made by a gynecologist after urodynamic evaluation. The percentage of stress-only incontinence increased from 51% to 77%, and mixed incontinence was reduced from 39% to 11% while the proportion with urgency-only UI remained virtually the same (10% vs. 12%). For public health purposes, self-report has the advantage of reflecting the woman’s experience and allows further characterization of the incontinence by frequency and quantity of urine loss. Despite the uncertainty in determining the type of incontinence, the differences in the epidemiology of urgency and stress incontinence are potentially important given the presumed differences in underlying pathophysiology. Differences in the association of stress and urgency UI (based on self-report) with respect to age, race/ethnicity, and risk factors suggest that questionnaire-based determination of UI type is useful and should be included in all epidemiologic studies of UI (Thom, 1998).

Usually incontinence is defined by the episode frequency (e.g., one or more episodes in the past year or past month or typically leaking at least once a week or daily). UI can also be defined by severity (a combination of frequency and quantity). Probably the most widely used measure of severity is the Sandvik Severity index (Sandvik et al, 1993, 2000), which is calculated by multiplying the reported frequency (four levels) by the amount of leakage (three levels). Incontinence can also be characterized by impact, which is usually assessed either in terms of bothersomeness or by the degree to which it restricts a woman’s activity (e.g., the Incontinence Impact Questionnaire) (Shumaker et al, 1994).

The vast majority of epidemiologic studies of female UI have been conducted in the United States, Canada, Europe, Australia, or Japan. A few additional studies have been reported from other Asian countries. Thus we know little about the prevalence, incidence, and risk factors for UI in Africa or other parts of the developing world.

Prevalence

Estimates of UI prevalence differ for a variety of reasons other than variation in true prevalence:

1. Sampling frame. Representative samples drawn from a large, usually geographically defined population (e.g., via random digit dialing) are thought to provide prevalence estimates that are the most generalizable.

2. Response rates. Even population-based studies can provide biased prevalence estimates due to poor response rates (Fultz and Herzog, 1993). Although the degree of bias is often unknown, as a general rule response rates above 80% are considered good and those between 60% and 80% are marginal.

3. Threshold definition of UI. Usually incontinence is defined by a threshold frequency (e.g., one or more episodes in the past year or past month, or typically leaking at least once a week or daily). Severity of degree of bother may also be used. The threshold chosen to define UI obviously affects the prevalence and incidence.

4. Types of UI. UI in women is typically classified as stress, urgency, or mixed. Prevalence estimates for each type of UI are sensitive to the wording of the questions and to definitions used.

5. Survey methods. Although there are few data comparing differences in prevalence based on method of ascertainment (Herzog and Fultz, 1990; Fultz and Herzog, 2000), variation in ascertainment (e.g., in-person interviews, phone interviews, mailed questionnaires, diaries) affects prevalence estimates for other socially sensitive conditions.

6. Culture and language. Although this area has not been investigated, it is likely that there are cultural differences in the perception of and terms used to describe UI that may affect prevalence estimates. Similarly, differences in language may lead to differences in ascertainment.

Prevalence in General Population of Adult Women

A review of 36 general population studies from 17 countries in the 2004 Third ICI edition (Hunskaar, 2005) found that the prevalence estimates for the most inclusive definitions of UI (“ever,” “any,” or “at least once in the past 12 months”) ranged from 5% (Van Oyen and Van Oyen, 2002) to 69% (Swithinbank et al, 1999), with most studies reporting a prevalence of any UI in the range of 25% to 45%. For daily incontinence, prevalence estimates typically range between 5% and 15% for middle-aged and older women (Thomas et al, 1980; Yarnell et al, 1981; Holst and Wilson, 1988; Rekers et al, 1992; Lara and Nacey, 1994; Hannestad et al, 2000). Studies reported since 2004 have added important information on prevalence of incontinence in women younger than 30 and older than 80 years of age, particularly for prevalence of incontinence by type. These studies are consistent with previous studies reporting that older women are more likely to have mixed and urgency incontinence (Diokno et al, 1986; Molander et al, 1990; Roberts et al, 1998; Nuotio et al, 2003), whereas young and middle-aged women generally report stress incontinence (Hording et al, 1986; Holst and Wilson, 1988; Burgio et al, 1991; Samuelsson et al, 1997; Hannestad et al, 2000). Overall, approximately half of all incontinent women are classified as stress incontinent. A smaller proportion are classified as mixed incontinent, and the smallest fraction as urgency incontinent. A recent study that included the entire adult age range by Hannestad and colleagues (2000) demonstrated a fairly regular increase in prevalence of mixed incontinence across the age range and a decrease in prevalence of stress incontinence from the 40- to 49-year-old age group through the 60- to 69-year-old group.

Prevalence in Long-Term Care Facilities

Several studies have documented a higher prevalence of UI in women residing in long-term care facilities compared with community-dwelling women (Ouslander et al, 1982; Sgadari et al, 1997; Hunskaar et al, 1998; Aggazzotti et al, 2000; Saxer et al, 2008). Because UI is associated with dementia, limited mobility, and other comorbid conditions, prevalence of UI is higher in facilities with residents requiring a higher level of care. In addition to being more frequent, UI in long-term care residents tends to be more severe, costly, and more burdensome on caregivers compared with UI in the community (Ouslander and Schnelle, 1995).

Prevalence during Pregnancy and Postpartum

Studies of UI during pregnancy have reported period prevalences of 32% to 64% for all UI and 40% to 59% for stress (including mixed) UI. Period prevalence is higher in parous than in nulliparous women (Dimpfl et al, 1992; Marshall et al, 1998; Morkved and Bo, 1999; Hvidman et al, 2002), whereas new-onset UI (cumulative incidence) during pregnancy is higher in primigravida women (Marshall et al, 1998; Morkved and Bo, 1999; Hvidman et al, 2002). Point prevalence of UI is low in the first trimester, rises rapidly in the second trimester, and increases further, though less rapidly, in the third trimester (Marshall et al, 1998; Morkved and Bo, 1999). Severity of incontinence appears to increase during pregnancy as well (Burgio et al, 1996). Data recently reported from the Norwegian Mother and Child Cohort Study, a large (N = 43,279) population-based study, confirms these observations (Wesnes et al, 2007). In this study the prevalence of stress incontinence from before to during pregnancy rose from 9% to 31% in nulliparous women and from 24% to 42% in parous women. In contrast, mixed incontinence showed a similar rise in both groups (from 6% to 16% and from 8% to 20%, respectively). Urgency incontinence remained virtually unchanged in both groups at less than 5%.

Estimating the postpartum prevalence of UI is complex because in addition to differences in study design, method of ascertainment, and choices of definition of incontinence, prevalence may also depend on the number of previous births, type of delivery, and history of previous incontinence. Nearly all postpartum UI is stress related. Most studies did not provide information on the frequency of leakage or degree of bother. Overall, it appears that any incontinence is reported by 15% to 30% of women during the postpartum year, weekly or greater incontinence is reported by 5% to 10%, and daily UI is reported by less than 5% of women. Excluding women with UI before pregnancy reduced the prevalence of postpartum incontinence by 3% to 4% in the one study reporting separate prevalences for each group (Glazener et al, 2006). Studies that included women delivered by caesarean section and delivered vaginally consistently found a lower prevalence in those delivered by caesarean section. Studies reporting prevalence in women delivered by elective caesarean section and by caesarean section after initiation of labor showed a substantially lower prevalence of postpartum UI in the elective caesarean section group.

Prevalence in Different Racial and Ethnic Groups

Although most epidemiologic studies of UI have been conducted on Caucasian populations in the United States, Canada, and Europe, there are now a substantial number of studies in Asian countries (Herzog and Fultz, 1990; Dimpfl et al, 1992; Burgio et al, 1996; Marshall et al, 1998; Morkved and Bo, 1999; Hvidman et al, 2002; Jumadilova et al, 2005; Anger et al, 2006b; Glazener et al, 2006; Waetjen et al, 2007; Wesnes et al, 2007). Studies have also been conducted on racial/ethnic groups in the United States (Burgio et al, 1991; Fultz et al, 1999; Sampselle et al, 2002; Grodstein et al, 2003; Nygaard et al, 2003; Jackson et al, 2004; Anger et al, 2006a; Danforth et al, 2006; Thom et al, 2006; Waetjen et al, 2007; Dooley et al, 2008; Fenner et al, 2008; Markland et al, 2008, 2009). Because of the substantial variation in prevalence of incontinence between studies, it is best to examine differences by race and ethnicity in the same study. Most striking is the lower prevalence of stress UI in black and Asian compared with white women. In most studies, blacks have half to a quarter the prevalence of stress UI compared with white women but the same or slightly higher prevalence of urgency UI and a slightly lower prevalence of mixed UI. Thus the difference in prevalence of UI between white and black women appears to be almost entirely the result of a markedly lower prevalence of stress UI symptoms in black women. In contrast, the lower prevalence of UI seen in Asian women compared with white women reflects a lower prevalence of both stress and urgency UI. The comparison of prevalences between Hispanic and non-Hispanic white women is complex in that some studies have found a lower prevalence of UI in Hispanic women while others have reported the prevalence to be higher in this group. This heterogeneity may be explained, at least in part, by differences in prevalence among a subpopulation of Hispanic women. Studies such as the SWAN study (Sampselle et al, 2002) that include Hispanic women who are primarily of Caribbean origin have reported lower UI prevalence, whereas those enrolling primarily Mexican-American women have reported higher UI prevalences (Thom et al, 2006; Daneshgari et al, 2008; Dooley et al, 2008). This relationship is further supported by the NHANES data reported by Anger and colleagues, which found prevalences of any UI to be 36% in Mexican-American women and 30% in other Hispanic women (Anger et al, 2006a; Daneshgari et al, 2008).

Key Points: Prevalence of Urinary Incontinence

• The estimated prevalence of UI in middle-aged and older women in the general population appears to be in the range of 30% to 60% (increasing with age). The prevalence of daily UI ranges from 5% to 15%, rising to more than 15% in women older than age 70 who are institutionalized. Some studies have found prevalences outside theses ranges, demonstrating that there remains a large variation in the estimated prevalence of UI in women, even after taking into account differences in definitions, ascertainment, and demographic characteristics. At least part of this variation is likely to be due to the sensitivity of the subject and subtle differences in the conduct of studies (level 1 evidence).

• Multiple observational studies have confirmed that white, non-Hispanic women have a substantially higher prevalence of stress UI than black or Asian women that is not explained by differences in known risk factors for UI (level 1 evidence).

Potential Risk Factors

Many epidemiologic studies have investigated potential risk factors for UI. The design of most of these studies has been cross sectional using prevalent UI. In cross-sectional studies, the temporal relationship between some potential risk factors such as diet, physical activity, medications, or body weight is ambiguous because the development of UI can lead to changes in behavior. Since 2004, several prospective studies have examined risk factors for incident UI that have allowed for evaluation of the temporal relationship. Associations in all observational studies are subject to potential confounding, necessitating multivariate analysis to estimate the independent association between risk factors and UI. The risk factors of estrogen use and body mass (obesity) have been examined in randomized, controlled trials. With the exception of age, the following risk factors are considered to be at least potentially modifiable.

Key Points: Conditions Predisposing to Urinary Incontinence (Milsom et al, 2009)

• Age: Virtually all studies have found an increase of UI prevalence with age. As a result, while stress incontinence predominates in younger and middle-aged women, urgency and mixed incontinence are more common in older women.

• Pregnancy, parity, and parturition events: Stress UI is reported by most women at some time during pregnancy, usually in the third trimester, but is generally self-limited and resolves after delivery. The interpretation of this association is not clear. It may be that physiologic changes during pregnancy cause incontinence later in life, but it is also possible that the temporary physiologic changes during pregnancy may “uncover” women with a predisposition to incontinence who are destined to become incontinent later in life regardless of the effects of pregnancy. The vast majority of studies have found an association between parity and later UI.

• The effect of parturition on risk of incontinence is difficult to separate from the effect of pregnancy. Possible mechanisms by which vaginal birth could lead to incontinence include stretching damage to pudendal and other nerves and tearing of connective tissue, which reduces pelvic floor support for urethral competence.

• Obesity and body mass: Obesity is well established as a factor that can cause UI or contribute to the severity of the condition. It is believed that the added weight of obesity, like pregnancy, may bear down on pelvic tissues causing chronic strain, stretching, and weakening of the muscles, nerves, and other structures of the pelvic floor. Intervention studies for weight reduction have reported that loss of weight is associated with partial remission or complete resolution of UI. Thus there is strong evidence to support the causal role of excess weight in the development of UI.

• Hormones: Current evidence suggests that although oral estrogen as a risk factor for stress UI may be unexpected, it is likely to be important, at least in women age 55 and older.

• Diabetes: Many, though not all, cross-sectional studies have reported UI to be more common in women with type 2 diabetes than among women with normal glucose levels even after adjusting for common risk factors including age and obesity.

• Hysterectomy: Approximately 600,000 women have hysterectomies in the United States each year, usually for noncancerous conditions. There has been concern that hysterectomy may be associated with development of UI via damage to the pelvic nerves and pelvic supportive structures. On the basis of the available data, hysterectomy does not appear to increase the risk of UI in the short term.

• Urinary tract infections and lower urinary tract symptoms: Urinary tract infection (UTI) has long been considered to be a cause of transient UI. The role of UTIs with respect to the development of chronic UI is less clear. Without additional data, it is not possible to say if UTIs actually increase the risk of later UI, if UI increases the risk of being diagnosed with a UTI, or if both UI and UTIs are manifestations of a common underlying process.

• Impaired physical function: Functional impairments, particularly mobility limitations, a history of falls, arthritis, dizziness, the need to use a walking aid, and poor lower extremity strength, have been correlated with UI in many community-based and nursing home studies.

• Cognitive impairment: Research on UI in nursing home residents has consistently supported an association between UI and dementia. In studies using multivariate analyses, patients lacking mental orientation had a 3.6 times greater risk of being incontinent than those with normal mental status and the presence of dementia increased the odds of UI by 1.5 to 2.3.

• Depression: Although depression may predispose a woman to developing UI, it is also possible that UI leads to depression (e.g., by reducing a woman’s social network); that depression and UI share one or more common neurochemical or hormonal pathways; or that women with depression are more likely than nondepressed women to be bothered by, and hence report, having incontinence.

• Menopause: Historically, menopause has been considered a risk factor for UI. However, establishing a relationship between menopause and UI has been difficult, probably in part because changes in endogenous hormone levels vary over a period of at least several years and only approximately correspond to amenorrhea or vasomotor symptoms used to define clinical menopause.

• Physical activity: Evaluating physical activity as a potential risk factor or protective factor for incontinence has been difficult. Incontinence, particularly stress UI, can be a barrier to exercise, which could lead to the spurious appearance of a protective effect of physical activity on UI. Alternatively, engaging in physical activity could “unmask” an existing propensity toward stress incontinence by increasing intra-abdominal pressure, resulting in physical activity being identified as a risk factor in the short term even if it is protective in the long term.

• Smoking, cough, and chronic lung disease: Conflicting data from cross-sectional studies and lack of association between smoking and incident UI in prospective studies suggest that smoking per se is probably not an independent risk factor for UI.

• Diet: Studying diet as a risk factor for UI is challenging. Although some dietary constituents such as coffee, alcohol, or carbonated beverages have been suspected as worsening UI, there is little reason to suspect other dietary constituents as causing, or protecting from, UI.

• Family history and genetics: Several family history studies have found a twofold to threefold greater prevalence of stress UI among first-degree relatives of women with stress UI compared with first-degree relatives of continent women.

• Ischemic heart disease: It is not clear if heart disease is an independent risk factor for UI.

• Other factors: Additional variables reported as associated with UI include constipation; fecal incontinence; genital prolapse; congestive heart failure; use of diuretics, benzodiazepines, and other drugs; and childhood enuresis.

Impact and Consequences

The impact of UI is typically assessed by degree of bother or by scales that measure the impact of incontinence on functional status and quality of life. A single question asking patients to report the degree of bother they experience from their incontinence provides an important dimension to characterizing UI in addition to frequency and amount of urine lost. Typically, incontinence that is more severe, of longer duration, and is urgent or mixed is also more bothersome (van der Vaart et al, 2002). Validated scales measure the impact of UI. One of the first scales, the Incontinence Impact Questionnaire (Shumaker et al, 1994), uses 30 items asking whether incontinence limits activities in four domains. Several other similar scales exist, some specific for urgency or stress incontinence or for overactive bladder (Shumaker et al, 1994; Uebersax et al, 1995; Jackson et al, 1996; Wagner et al, 1996; Kelleher et al, 1997; Brown et al, 1999; DuBeau et al, 1999). In addition to severity of UI, interference with sexual activities and incontinence at night is associated with greater reported impact on quality of life (Huang et al, 2006).

Incontinence has been identified as a potential risk factor for falls and nursing home admissions. In a prospective study of more than 6000 older women, women with weekly urgency, but not stress, incontinence were more likely to fall during the 3-year follow-up, adjusting for multiple other factors (adjusted OR = 1.3, 95% CI = 1.1 to 1.4) (Brown et al, 2000). Similarly, in a study of 472 long-term care residents in Germany, UI at baseline predicted falls over the next 12 months after adjusting for a variety of patient factors and comorbid conditions (Kron et al, 2003). UI was the second strongest predictor of falls (a history of falls being first). However, it is possible that both falls and UI are markers for frailty that cannot completely be adjusted for. Whether treatment of UI would reduce the number of falls has not been established.

UI has also been reported as a risk factor for nursing home admissions even after adjusting for multiple comorbid conditions (Thom et al, 1997a, 1997b). However, another study that also adjusted for additional measures of frailty including activities of daily living found that the association became weaker and nonsignificant (Holroyd-Leduc et al, 2004).

UI has substantial, well-documented economic consequences as well. Annual costs of UI are estimated to be more than $20 billion in direct costs (primarily costs of routine care including out-of-pocket costs) (Wilson et al, 2001), which is greater than the annual costs for breast, cervical, uterine, and ovarian cancer combined (Subak et al, 2006). There are also substantial indirect costs from work loss and missed volunteer work (Thom et al, 2005).

Treatment Seeking and Treatment

The proportion of women with any UI in the past 12 months who report having sought medical treatment ranges from 12% (Sampselle et al, 2002) to 53% (Mardon et al, 2006) with about half the studies showing 25% to 30% (Rekers et al, 1992; Hannestad et al, 2002; Hagglund et al, 2003; Hsieh et al, 2008). All studies that have examined treatment seeking by type of incontinence have found that women with urgency or mixed incontinence are more likely to seek treatment than women with stress incontinence (Hannestad et al, 2002; Danforth et al, 2006; Zhu et al, 2008). As expected, the proportion of women seeking treatment is strongly associated with frequency, severity, and bothersomeness, with several studies reporting proportions of 50% to 80% of women with daily, moderate to severe, or bothersome incontinence (Hannestad et al, 2002; Kinchen et al, 2003; Diokno et al, 2004; Gasquet et al, 2006).

Reasons given by people for not seeking help include not regarding incontinence as abnormal or serious, embarrassment, considering incontinence to be a normal part of aging, having low expectations of treatment, and thinking they should cope on their own. Other characteristics such as age, general health, or attitude toward health care may also play a role.

Key Points: Consequences of Urinary Incontinence

• Typically, incontinence that is more severe, of longer duration, and is urgency or mixed incontinence is more bothersome.

• In addition to severity of UI, interference with sexual activities and incontinence at night are associated with greater reported impact on quality of life.

• Incontinence has been identified as a potential risk factor for falls and nursing home admissions.

• UI has also been reported as a risk factor for nursing home admissions even after adjusting for multiple comorbid conditions.

• UI therefore has substantial, well-documented, economic consequences.

• The proportion of women with any UI in the past 12 months who report having sought medical treatment ranges from 12% to 53% with about half the studies showing 25% to 30%.

• Women with urgency or mixed incontinence are more likely to seek treatment than women with stress incontinence.

• The proportion of women seeking treatment is strongly associated with frequency, severity, and bothersomeness.

Urinary Incontinence in Men

General Comments

Key Points: Urinary Incontinence in Men

• The epidemiology of UI in men has not been investigated to the same extent as for females. But it appears that UI is at least twice as prevalent in women as compared with men. There seems to be a more steady increase in prevalence with increasing age than for women (level 1 evidence).

• Most studies find a predominance of urgency incontinence, followed by mixed forms of UI and stress incontinence the least. Most studies have a large fraction of other/unclassified types (level 2 evidence).

• Literature on incidence and remission of male UI is still scarce.

• Few clear risk factors have been scientifically documented, but several medical correlates have been reported. Established risk factors predisposing men to UI include increasing age, presence of lower urinary tract symptoms (LUTS), urinary tract infections, functional and cognitive impairment, neurologic disorders, and prostatectomy (level 2 evidence).

• UI after radical prostatectomy is frequent, ranging from 2% to 57%. Rates steadily decline from the time of surgery and plateau at 1 to 2 years (level 1 evidence).

• Factors affecting postprostatectomy UI include the age at surgery, type of prostatectomy, and the modifications in the technique involving nerve and bladder neck preservation (level 2 evidence).

Prevalence

A systematic review of 21 studies reported a prevalence of UI in older men ranging from 11% to 34% (median = 17, pooled mean = 22%). In the same review, the prevalence of daily UI in men ranged from 2% to 11% (median = 4%, pooled mean = 5%) (Thom, 1998). A wide definition of UI, older age, inclusion of institutionalized men, and the use of self-reporting methods tend to result in higher prevalence rates (Koyama et al, 1998; Thom, 1998). The wide span of results may be explained by the variation in the population studied, the definition of incontinence used, and the methods used in the surveys.

For any definition of UI, there is a steady increase in prevalence with increasing age.

Incontinence Subtypes

Due to differences in pathologic anatomy and pathophysiology of UI in men and women, there is a different distribution in incontinence subtypes. Recent studies confirmed our previous reports of the predominance of urgency incontinence (40% to 80%), followed by mixed forms of UI (10% to 30%) and stress incontinence (<10%) (Herzog and Fultz, 1990).

The higher percentages of the urgency and mixed types of incontinence are more significant in studies involving older people. In fact, the increasing prevalence of any UI by age in men is largely due to the contribution of urgency incontinence rather than stress incontinence. One study demonstrated an increasing rate of urgency UI from 0.7% between ages 50 and 59 to 2.7% between 60 and 69 to 3.4% for respondents 70 years and older. Stress UI was steady at 0.5%, 0.5%, and 0.1% for these groups, respectively (Ueda et al, 2000). On the other hand, Maral and colleagues (2001) reported increasing prevalence of SUI with age: 0.9% between ages 35 and 44, 1.2% between ages 45 and 54, 3.8% between ages 55 and 64, and 4.9% at age 65 and older.

Most studies report a significant fraction of other/unclassified type of urinary incontinence. One study reported that a majority of men with UI had overflow and functional types of incontinence (Bortolotti et al, 2000), whereas another found constant dribbling in 7% of their respondents (Damian et al, 1998). Terminal dribbling or postvoid dribbling is another type of leakage in men that is difficult to assign to the conventional subtypes of UI. In an Australian survey, 12% of respondents reported frequent terminal dribbling (Sladden et al, 2000).

Severity

When it comes to severity, the distribution in men follows that of the women. Estimates for severe UI in older women tend to be about twice as high as for older men (Herzog and Fultz, 1990).

Race/Ethnicity

Few studies have looked at the impact of race or ethnicity on the prevalence of UI among men. A four-country study presented lower prevalences of reported UI among men from Korea (4%) and France (7%) than in men from Britain (14%) and Denmark (16%) (Boyle et al, 2003). On the other hand, unpublished data from the MESA study did not indicate differences in prevalence among white male respondents compared with African-American respondents.

Incidence and Remission

Literature on the incidence of male UI is scarce. The MESA study (Herzog and Fultz, 1990) found a 1-year incidence rate for men older than 60 years of 9% to 10%. In a population-based survey in the UK among men at least 40 years of age, the 1-year incidence of UI was noted to be 3.8% (McGrother et al, 2004). A review of a health organization database of males at least 65 years old revealed a UI incidence of 23.8 per 1000 person years. Malmsten (1997) analyzed the age of onset of UI for each age cohort. Mean starting age for all men was 63 years. The mean duration was about 8 to 10 years in the cohorts. Substantial remission rates for UI in males were noted by the MESA study, higher among men (27% to 32%) than women (11% to 13%) (Herzog and Fultz, 1990). A similarly high 1-year remission rate of 39.6% was noted among British males (McGrother et al, 2004). One possible explanation for the difference in the published incidence and remission rates in men compared with women is the predominance of urgency type incontinence among men and its close relation to overactive bladder with and without incontinence. Another factor is the close association among urgency UI and prostate gland disease, infections, or bowel dysfunction, all of which are relatively amenable to treatment or may improve even without treatment.

Potential Risk Factors for Urinary Incontinence

There is relatively little research concerning conditions and factors that may be associated with UI in men, and clear risk factors are poorly documented. However, a few available studies have identified potential risk factors, which are described next.

Age

As in women, increasing age is correlated with increasing prevalence of UI. Multivariate analysis in several studies has shown that age is an independent risk factor for incontinence (Muscatello et al, 2001; Boyle et al, 2003; Landi et al, 2003; Nelson and Furner, 2005; Diokno et al, 2007). Compared with women, however, there seems to be a steadier increase in prevalence in men with increasing age.

Lower Urinary Tract Symptoms and Infections

LUTS, like urgency, nocturia, a feeling of incomplete voiding, and reduced flow, are typically associated with UI (Diokno et al, 1986; Hunskaar, 1992; Schmidbauer et al, 2001). In one study, UI was reported by 15% of men without voiding symptoms, frequency, or urgency and by 34% of those with such symptoms (Diokno et al, 1986).

Studies have also reported that urinary tract infections and cystitis are strongly associated with male UI (Damian et al, 1998; Ueda et al, 2000), with an OR of 3.7 for UI in men reporting cystitis (Ueda et al, 2000) and an OR of 12.5 among men with recurrent infections (Bortolotti et al, 2000). It should be noted that reports indicating a positive association between UTI and incontinence involved men older than 60 years.

Functional and Cognitive Impairment, Physical Activity

Mobility problems such as use of a wheelchair or aids to walking, as well as diagnosed arthritis or rheumatism or having a fall during the past year, were significantly greater among incontinent than continent men (Diokno et al, 1990; Damian et al, 1998).

Neurologic Disorders

Many specific neurologic diseases may lead to UI (Resnick and Yalla, 1987). Neurogenic detrusor overactivity is seen commonly in meningo-myelocele patients and in spinal injuries, Parkinson disease, and multiple sclerosis. Underactive bladder dysfunction due to a cauda equina lesion or diabetes might cause overflow or a paralyzed pelvic floor and hence stress incontinence. Men who have suffered a stroke were at increased risk for incontinence with an OR of 7.1 (Ueda et al, 2000). A case control study of age-matched, long-term stroke male survivors with controls showed a higher prevalence of UI among stroke survivors compared with controls (17% vs. 9%). In addition, among UI sufferers, the stroke survivors were found to have higher frequency and more leakage than controls. In a study of 235 stroke patients, the occurrence of UI correlated with motor weakness (OR 5.4), visual defects (OR 4.8), and dysphagia (OR 4.0).

Prostatectomy

A well-known iatrogenic cause of male incontinence is prostatectomy, but the attributable risk for this factor in the population of men with UI is unknown. In a Norwegian survey of elderly men with UI, almost a third had undergone prostatectomy (Hunskaar, 1992).

In a cross-sectional study among men in Vienna, Schmidbauer and colleagues identified previous prostatectomy to be associated with UI (Schmidbauer et al, 2001).

TURP seems to be followed by an incidence of stress incontinence of about 1%. A randomized controlled trial comparing TURP, laser prostatectomy, and evaporization of the prostate for benign disease showed comparable incontinence rates immediately and up to 12 months postoperatively (van Melick et al, 2003).

Radical prostatectomy seems to induce UI at a much higher rate than TURP. The overall prevalence of postradical prostatectomy incontinence ranges from 2% to nearly 60%. This wide range may be explained by many factors including differences in study characteristics, population characteristics, study site, definition used, and timing of assessment of continence in relation to the surgery. The era of development of the procedure has also been found to be associated with the prevalence rates (Hu et al, 2003), as well as the various procedural modifications of the surgery (see later). Postprostatectomy incontinence rates elicited from symptoms reported by patients are generally two to three times higher than those from physicians’ observations. Studies that have performed both assessments in the same population confirm this observation that doctors underestimate postprostatectomy incontinence by as much as 75% (Ojdeby et al, 1996; Donnellan et al, 1997; McKammon et al, 1997; Olsson et al, 2001).

Incontinence rates after prostatectomy seem to decline steadily with time and plateau 1 to 2 years after surgery (Lowe, 1996; Weldon et al, 1997; Goluboff et al, 1998; Horie et al, 1999; Walsh et al, 2000; Moinzadeh et al, 2003). This emphasizes the need for a long follow-up period to establish continence status postprostatectomy. An actuarial study among 647 postprostatectomy men estimated UI rate of 13% at 1 year and 7% at 2 years postsurgery (Saranchuk et al, 2005).

The technique of radical prostatectomy affects UI rates. Modifications associated with lower UI rates include the perineal approach (Bishoff et al, 1998; Gray et al, 1999) and preservation of neurovascular bundle (Eastham et al, 1996; Van Kampen et al, 1998; Wei et al, 2000). Bladder neck preservation affords earlier return to continence compared with bladder neck resection, but with similar UI rates after 1 year (Poon et al, 2000; Srougi et al, 2001). One study showed earlier recovery of UI after tennis racquet reconstruction and bladder neck preservation compared with bladder neck resection with puboprostatic ligament preservation, but with similar UI rates at 1 year (Lowe, 1996; Noh et al, 2003). However, continence status assessed more than 12 months after surgery showed even lower rates of UI after bladder neck preservation (11.6%) compared with resection (4.9%) in one study (Lowe, 1996). UI after laparoscopic procedures seems to be higher initially but approximates that of open techniques by 1 year (Jacobsen et al, 2007).

Older age at time of surgery has been found to be associated with a higher prevalence of postprostatectomy UI (Eastham et al, 1996; Goluboff et al, 1998; Van Kampen et al, 1998; Gray et al, 1999; Wei et al, 2000; Kundu et al, 2004). One study showed a doubled risk for every 10 years of age beginning at age 40 (Catalona et al, 1999). Another study suggested that rather than absolutely affecting final continence prevalence, elderly men need a longer time to achieve continence after surgery (Horie et al, 1999). Two studies found no relation between age at surgery and UI (Donnellan et al, 1997; Kao et al, 2000).

Other factors have been found to be associated with a higher prevalence of postprostatectomy UI, although not consistently. Such factors include prior TURP, preoperative lower urinary tract symptoms, obesity, clinical stage, prostate-specific antigen (PSA), prostate volume, and Gleason score (Eastham et al, 1996; Moul et al, 1998; Van Kampen et al, 1998; Kundu et al, 2004; Lee et al, 2006). A retrospective analysis of 156 patients who had undergone preoperative MRI of the prostate and were followed up postprostatectomy showed that time to return to continence was associated with the variation in the shape of the prostatic apex. The prostatic apex that does not overlap with the membranous urethra was found to be significantly associated with an early return of continence (Lee et al, 2006). The 5-year cohort study of the Prostate Cancer Outcomes Study found that among prostatectomy patients, race and ethnic differences were related to urinary incontinence, with African-Americans having better recovery compared with non-Hispanic whites and Hispanics (Johnson et al, 2004).

Adjuvant radiotherapy has not been found to affect postprostatectomy incontinence rates when assessed beyond 1 year (Formenti et al, 2000; Hofmann et al, 2003; Kundu et al, 2004).

Factors of Unclear Association with Urinary Incontinence in Men

A 9-year study of Janssen (2007) showed increasing rates of UI with increasing body mass index (BMI) among the older men and women. However, multivariate analysis failed to show increased BMI (overweight and obese levels) as an independent risk factor for the development of UI. In a study including a younger population in Australia, obesity was noted to be associated with UI with an OR of 3.2 (1.2 to 9.0) (Muscatello et al, 2001). In this study, however, being merely overweight was not associated with UI. Several studies in the older persons have shown an association between physical activity and UI among women that is not seen among men (Boyle et al, 2003; Kikuchi et al, 2007).

Key Points: Epidemiology of Male Incontinence

• Relatively little research concerning conditions and factors that may be associated with UI in men has been conducted, and clear risk factors are poorly documented.

• Increasing age is correlated with increasing prevalence of UI.

• LUTS, like urgency, nocturia, feeling of incomplete voiding, and reduced flow, are typically associated with UI.

• Many specific neurologic diseases may lead to UI.

• TURP seems to be followed by an incidence of stress incontinence of about 1%.

• The overall prevalence of postradical prostatectomy incontinence ranges from 2% to nearly 60%.

• Postprostatectomy incontinence rates elicited from symptoms reported by patients are generally two to three times higher than those from physicians’ observations.

• Doctors underestimate postprostatectomy incontinence by as much as 75%.

• The technique of radical prostatectomy affects UI rates.

• Older age at time of surgery has been found to be associated with a higher prevalence of postprostatectomy.

• Other factors associated with a higher prevalence of postprostatectomy UI include prior TURP, preoperative lower urinary tract symptoms, obesity, clinical stage, PSA, prostate volume, and Gleason score.

Epidemiology of Overactive Bladder

The studies reported in the literature have used different populations and ways of selecting samples, and the definition of OAB has changed slightly after its initial description, although this was standardized after the 2002 ICS standardization report.

The prevalence of OAB in adult males varies from 10.2% to 17.4% and in females from 7.7% to 31.3%. Clearly different definitions, the different age groups studied and maybe cultural differences affect the rates (level 2 evidence). OAB_wet was less prevalent than OAB_dry except for one implausible-looking result. The response rate was quite low in many of these studies, and sometimes it was impossible to know the response rate (e.g., when quota sampling was used).

A common finding was that the rates increased with age, but this was not universal (level 2 evidence). Some studies found that after adjusting for state of health, the relationship with age went away, and some found that it only applied to OAB_wet. Usually OAB was more prevalent in females, and this was much more pronounced for OAB_wet, which sometimes meant that OAB_dry was the same for males and females. OAB was related to general health problems and in particular diabetes. It is likely to be related to urinary problems in childhood.

Key Points: Epidemiology of Overactive Bladder

• The prevalence of OAB in adult males varies from 10.2% to 17.4% and in females from 7.7% to 31.3%.

• Rates of OAB increase with age.

• OAB is more prevalent in females, particularly OAB_wet to the extent that the prevalence of OAB_dry can be the same for males and females.

Epidemiology of Pelvic Organ Prolapse

General Comments and Definitions

Pelvic organ prolapse (POP) refers to loss of support for uterus, bladder, colon, or rectum leading to prolapse of one or more of these organs into the vagina. The ICS Pelvic Organ Prolapse Quantification (POPQ) examination defines prolapse by measuring the descent of specific segments of the reproductive tract during Valsalva strain relative to a fixed point, the hymen. As noted earlier, the POPQ system describes the anatomic findings of pelvic organ prolapse without consideration for symptoms and bother perceived by the woman. Validation of this system has shown it to be highly reliable (Hall et al, 1996). The stages of prolapse severity are arbitrarily defined, and there is no clear differentiation between normal anatomic variation and mild POP. For research purposes there is consensus for use of the POPQ system until further evidence might clarify the distinction between normal variation and mild prolapse (Weber et al, 2001).

Determining POP on the basis of self-reported symptoms is difficult because of the lack of specificity and sensitivity of most symptoms attributed to pelvic organ prolapse (Ellerkmann et al, 2001) and the fact that prolapse above the level of the hymenal ring is usually asymptomatic (Lukacz et al, 2005). The only exception appears to be a sensation of bulging into the vagina, which is most strongly associated with prolapse at or below the hymeneal ring (Swift, 2000; Hendrix et al, 2002). A recent study of 110 women found that a question asking about a feeling of something bulging in or dropping out of their vagina had a sensitivity of 84% and a specificity of 94% for POP at or beyond the hymeneal ring on examination (Lukacz et al, 2005). Seeing prolapse would presumably be even more specific but is too uncommon to be useful as a definition.

Prevalence of PELVIC ORGAN PROLAPSE

The prevalence of POP based on a sensation of a mass bulging into the vagina was remarkably consistent, ranging between 5% and 10%. The study by Eva and colleagues (2003) reported a substantially higher prevalence included in the definition of POP, pelvic heaviness, or digital pressure on the perineum or in the vagina to aid with defecation. Reports from the Women’s Health Initiative (WHI) Oestrogen Plus Progestin Trial and randomized controlled trial (Handa et al, 2004; Nygaard et al, 2004; Bradley et al, 2007), although not actually population based, recruited the women in the trial from the community rather than from women seeking gynecologic care and provided important information on the prevalence of POP based on pelvic examination. The prevalence of observed prolapse in women enrolled in the WHI trial is similar to the prevalence found in the population-based study that also used pelvic examination (Samuelsson et al, 1999), although the prevalence of each type of prolapse was higher in the WHI study (Hendrix et al, 2002). In both studies, prolapse occurs most frequently in the anterior compartment, next most frequently in the posterior compartment, and least in the apical compartment.

Two studies that examined prolapse by race found that black women had the lowest prevalence and Hispanic women the highest after controlling for multiple other factors in multivariate analysis (Hendrix et al, 2002; Rortveit et al, 2007). The study reported by Rortveit and colleagues (2007) based on symptoms found adjusted ORs of 0.4 (95% CI = 0.2 to 0.8) for black and 1.3 (95% CI = 0.8 to 2.2) for Hispanic women, with white women as the referent group. Hendrix and colleagues (2002) reported adjusted ORs of 0.6 (95% CI = 0.5 to 0.8) for black and 1.2 (95% CI = 1.0 to 1.5) for Hispanic women compared with white women for POP based on genital examination.

Incidence

Only two studies that reported the incidence of new POP could be located. Both studies were done on subgroups of women enrolled in the WHI Oestrogen Plus Progestin Trial. The first study of 412 women enrolled at the University of California, Davis site used a standardized pelvic examination repeated every 2 years over 8 years (Handa et al, 2004). The incidence of new cystocele, rectocele, and uterine prolapse was 9%, 6%, and 2%, respectively. Annual rates of remission from grade 1 (prolapse to above introitus) were relatively common for each type of POP (24%, 22%, and 48%, respectively) but less common from grade 2 or 3 (prolapse to or beyond introitus) (9%, 3%, and 0%, respectively). In a second study of 259 women, postmenopausal women with a uterus were examined using the POP-Q at baseline and annually for 3 years. POP was defined as prolapse to or beyond the hymeneal ring. The incidence of new POP was 26% at 1 year and 40% at 3 years, with remission rates of 21% at 1 year and 19% at 3 years (Bradley et al, 2007).

Several studies have reported the annual incidence of surgery for POP in the United States and at least one in the United Kingdom. A longitudinal study of more than 17,000 women in the United Kingdom, age 25 to 39 at baseline, reported an annual rate of prolapse surgery of 0.16% (Mant et al, 1997). This rate is consistent with the rate of approximately 0.2% per year reported in the United States (Brown et al, 2002; Boyles et al, 2003). One U.S. study reported an annual incidence rising with age from 0.05% in women age 30 to 39 to 0.5% in women age 70 to 79 with an estimated lifetime cumulative risk of surgery from prolapse of 7% to 11% (Olsen et al, 1997). A recent U.S. study reported similar surgical rates: 0.07% for women age 18 to 39, 0.24% for women age 40 to 59, and 0.31% for women age 60 to 79 (Shah et al, 2008). Surgical rates drop substantially after age 80 (Olsen et al, 1997; Shah et al, 2008). Estimating rates of prolapse surgery has the advantage of use of hospital discharge data on procedures, which is highly accurate for the procedure performed but less accurate for the indications for the procedures, particularly when a procedure may have more than one indication.

Key Points: Epidemiology of Pelvic Organ Prolapse

• Most studies have used a cross-sectional design and there are limited longitudinal data to suggest a causal relationship between symptoms of obstructed defecation and pelvic organ prolapse or vice versa.

• Posterior vaginal wall prolapse and perineal descent are the compartment-specific defects most clearly associated with symptoms of obstructive defecation.

• A number of studies suggest that hysterectomy and other pelvic surgery may increase the risk for pelvic organ prolapse (level 2 evidence).

• Hysterectomy due to pelvic organ prolapse and other pelvic floor surgery are the strongest predictors of secondary pelvic floor surgery (level 2 evidence).

• SUI and pelvic organ prolapse have familial transmission patterns mediated by either genetic or environmental factors (level 2 evidence).

• SUI is more common in Caucasian women and pelvic organ prolapse in Caucasian and Hispanic women when compared with African-American women (level 2 evidence).

• Twin studies suggest that heritability contributes to the liability of developing pelvic floor disorders, but environmental effects are substantial.

• Childbirth is associated with an increased risk for pelvic organ prolapse later in life. Current evidence also suggests that increasing number of childbirths increases the risk (level 2 evidence).

• It remains undecided if caesarean section prevents the development of pelvic organ prolapse but most studies indicate that caesarean is associated with a decreased risk for subsequent pelvic floor morbidity in comparison to giving vaginal birth (level 2 evidence).

• There is a dearth in the understanding of how specific obstetrical events and the process of labor and delivery affect the risk for pelvic organ prolapse.

• Lifestyle factors and socio-economic indices appear to be associated with the risk for pelvic organ prolapse.

• A number of somatic diseases and conditions have been linked to the occurrence of prolapse, but the cause-effect relationship is undetermined.