Etiology of Pelvic Organ Prolapse and Urinary Incontinence
DANIEL M. MORGAN
Pelvic organ prolapse (POP) and urinary incontinence (UI) are distinct conditions but often treated and studied concurrently, as they are in this chapter. The role of pelvic relaxation in treating both of these conditions may account for how they have become linked in our thinking. Nonetheless, it is worthwhile to remember that the association between prolapse and UI is weak at best. Many women with prolapse are not affected by UI and vice versa.1
The lifetime risk of surgical intervention for POP and UI increases with age2 and ranges from 11% to 20%3,4,5,6 by age 80 to 85 years. Women are increasingly more likely to seek care for prolapse and UI as they age.7 With an aging population in the United States, it is estimated that by 2050, the number of women with POP will increase from 3.3 to 4.9 million and the number with UI will increase from 18.3 million to 28.4 million. Surgical management is projected to increase annually for POP from 166,000 to 245,970 and for stress urinary incontinence (SUI) from 210,700 to 310,050.8 An analysis of a large managed care organization has raised concerns that there may be a shortage of health care personnel to meet this need.9
Consensus is developing around definitions for POP. To consider it a clinical problem, there should be a combination of symptoms and findings consistent with the condition.
The American College of Obstetricians and Gynecologists (ACOG) and the American Urogynecologic Society (AUGS) define POP as descent of the vagina or uterus that allows the bladder or bowel to herniate into the vaginal space. They advise that prolapse be considered a problem when it causes symptoms of bulging or pressure or when it alters bladder or bowel function. Mild relaxation is common and should not be considered pathologic.10 The International Continence Society (ICS) and the International Urogynecological Association (IUGA) define prolapse as the “descent of one or more of the anterior vaginal wall, posterior vaginal wall, uterus (cervix), or apex of the vagina (vaginal vault or cuff scar after hysterectomy)” and advise that “the presence of any such sign should be correlated with relevant POP symptoms.” These symptoms of POP are described by ICS/IUGA as the “departure from normal sensation, structure or function, experienced by the woman in reference to the position of her pelvic organs.” ICS/IUGA further recognize how symptoms may be “generally worse … when gravity makes prolapse worse (e.g. long periods of standing or exercise) … better when … lying supine,” and “more prominent at times of abdominal straining, e.g. defecation.”
Both societies also recognize that there is a threshold in prolapse at which symptoms are more common. ACOG/AUGS have noted that protrusion of the vagina 0.5 cm beyond the hymen is associated with symptoms of prolapse,11 whereas ICS/IUGA have commented that the “correlation” of signs and symptoms would most often “occur at the level of the hymen or beyond.”12
PROLAPSE STAGING AND PREVALENCE
The Pelvic Organ Prolapse Quantification (POP-Q) staging system is used most often to describe pelvic organ support. Prolapse is assessed with respect to the hymen, with negative numbers indicating support above it, positive numbers beyond it, and zero at the hymen. In Figure 6.1, support of the vagina and uterus are shown with respect to common reference points of vaginal and vulvar anatomy.13 The proportion of women who have support in these different stages—also analyzed by centimeters in relationship to the hymen—is illustrated in Figure 6.2.14 Due to the definition of stage 2 prolapse being within 1 cm of the hymen, it is important to recognize that two-thirds of women have vaginal support that is categorized as having some prolapse.
Symptom questionnaires and exam findings are useful in establishing the prevalence of vaginal prolapse considered a clinical issue. Vaginal prolapse that is bothersome enough for treatment is reported by 6% of women15 and prolapse beyond the hymen is also present in 5% to 5.7%.14,16 These findings make it appear that there is a nearly perfect correlation between subjective and objective findings of POP. Although symptoms are important in establishing a diagnosis of POP, there are limitations—especially when support is at or just above the hymen. When the criterion for prolapse on pelvic exam is vaginal support 0.5 cm proximal to the hymen, only 72% were correctly identified with the question “Can you feel with your hand or see something bulging out of your vagina?”
Disease models are helpful in conceptualizing the development of prolapse and UI. In 1998, Bump and Norton17 described how risk factors can be sorted into those that predispose, incite, promote, or decompensate structure or function and how medication or surgery can intervene and return structure or function to normal. In 2008, DeLancey et al.18 described a lifespan model (Fig. 6.3) to emphasize the temporal relationship between these factors and development of symptoms. Phase 1 consists of the interaction of genetics, environment, and social training that lead to variation in health status. Phase 2 is characterized by inciting factors such as childbirth. In Phase 3, there are intervening factors (considered promoting and decompensating in the Bump and Norton model) that can hasten the onset of symptoms. Aging and conditions such as constipation that leads to excessive straining, obesity that increases abdominal pressure, or any number of medical conditions that require chronic corticosteroids can hasten deterioration in tissue health and function and cause a patient to cross the symptom threshold.
There is a complex interplay of genetics and environment with evidence that heritability is a factor in the development of prolapse. A twin study identified a higher concordance rate for surgically treated POP and SUI among monozygotic than dizygotic twins,19 and the sisters of women presenting for treatment with stage 3 or 4 prolapse were fivefold more likely to develop prolapse than the general population.20 Many investigators have focused on genes that are associated with connective tissue and extracellular matrix. A meta-analysis of the genetic epidemiology of prolapse identified 4.79 increased odds (95% confidence interval [CI], 1.91 to 11.98) for developing prolapse when a gene that affects type 3 collagen fibers (COL3A1 rs1900255 genotype AA) is present. This type of collagen is found predominantly in the loose areolar tissue surrounding the vagina and pelvic organs.21
Patients with spina bifida, characterized by an incomplete closure of the spinal column, are also at higher risk of symptomatic prolapse than women in the general population. The underdeveloped sacral nerves, which supply innervation to the levator ani and other pelvic floor muscles, lead to a flaccid pelvic floor and high rates of POP. In two cohorts of patients with spina bifida, the rates of prolapse approached 50%. Among the 24 nulliparous women, 50% had at least stage 2 POP.22 Bladder management is associated with their risk for prolapse. The rate of de novo vaginal prolapse or rectal intussusception for patients with acontractile neurogenic bladder who empty the bladder by Valsalva was 32.1% compared to 3.7% in those using clean intermittent catheterization with 5-year follow-up.23
Social determinants of health are important to consider when comparing the prevalence of prolapse between populations. Differences observed among women analyzed by race are likely related to socioeconomic factors. For instance, women who have physically demanding occupations and fewer financial resources may be at higher risk of developing prolapse.24 These findings are beginning to demonstrate how race is “more accurately described as a social construct and not a biological one.”25
Vaginal childbirth influences the risk of developing prolapse. The cumulative incidence rates for developing prolapse at 5 and 15 years are 1.6% and 30% for spontaneous vaginal birth and 4% and 45% for operative vaginal delivery, compared to 0.2% and 9.4% for cesarean delivery.26
Many events of labor are potentially associated with the development of POP. These include length of the second stage of labor, use of episiotomy, mode of delivery (spontaneous vs. instrumented vs. cesarean), anal sphincter injury, and occult injuries (levator ani avulsion and connective tissue injuries). Memon et al. (Fig. 6.4) illustrated the clustering of childbirth events in 418 primiparous women to show how individual women may have one or multiple exposures.27
Levator avulsion injury is associated with a significant increase in risk for POP. The prevalence of levator ani injury with first birth ranges from 15% to 36%.28,29,30,31,32 Several studies have confirmed the importance of this injury in the development of prolapse.33,34 In age- and parity-matched groups of women with prolapse and normal support, DeLancey et al.34 found that 55% of women with prolapse had magnetic resonance imaging consistent with a levator ani injury compared to 16% of controls—a sevenfold increased odds of prolapse. With a median of 11 years follow-up after first vaginal birth, Handa et al.32 found that 55% of women with levator avulsion injury diagnosed by ultrasound had developed prolapse compared to 21% of those without levator avulsion—a fourfold increased risk.
Levator ani injury is associated with functional and physiologic change in the pelvic floor, but it is not clear how this injury mediates prolapse. It has been argued that levator ani injury leads to an increase in levator hiatus size and decreased contraction strength.35 However, there is also evidence that a levator ani defect accounts for only a proportion of the changes in the size of the levator hiatus and prolapse.36 Among patients recruited 5 to 10 years after delivery, Handa et al.37 found that a larger genital hiatus led to a significantly increased risk of prolapse and that the effect of parity was attenuated with adjustment for genital hiatus.
There are several risk factors for levator ani injuries. Forceps-assisted vaginal birth is associated with injury.38 Compared to noninstrumented vaginal delivery, operative delivery by forceps or vacuum significantly increases the odds for all pelvic floor disorders, with the highest increase for POP (odds ratio 7.5, 95% CI 2.7 to 20.9).
PHASE 3 FACTORS
Aging is frequently cited as a factor in prolapse.16,39,40 It seems likely that there is an interaction between aging and parity and other risk factors. In matched cohorts recruited from the Swedish Medical Birth Register, the rate of prolapse for those with one vaginal delivery increased from 3.8% at age 40 years to 13.8% by age 64 years. In contrast, the rate of symptomatic prolapse did not exceed 5% among nulliparous women and those who had one cesarean delivery.41 There was no difference in pelvic organ support assessed by POP-Q among nulliparous women stratified by deciles between ages 20 and 70 years.42 In a large epidemiologic study, the odds of prolapse did not increase for women up to age 84 years compared to those 25 to 39 years when adjustments were made for birth experience, menopausal status, hysterectomy, and obesity.15
Although posthysterectomy prolapse rates are less than 1%, epidemiologic analyses indicate that hysterectomy is associated with an increased risk of vaginal prolapse. This complication was most common among those who had a vaginal hysterectomy for prolapse, followed by those who had a vaginal hysterectomy for nonprolapse indications, and then by those who underwent a total abdominal hysterectomy for nonprolapse indications. The lowest rates of prolapse were identified among women who did not undergo hysterectomy.43 It is not possible to disentangle the relative impact of the underlying reasons for an increased rate of prolapse—it could be the hysterectomy itself, the pelvic anatomy allowing for vaginal hysterectomy, or the indication of uterovaginal prolapse.
Surgical technique is believed to reduce the risk of posthysterectomy prolapse. Shortening and plication of the uterosacral and cardinal ligaments have long been used to restore apical vaginal support. In 1927, Miller described reattachment of these ligaments to the vaginal cuff using chromic catgut as “lifting sutures passed through the peritoneum and underlying fascial and muscular structures at the base of the sacro-uterine ligament.”44 The technique of posterior culdoplasty to treat enterocele at the time of hysterectomy was published by McCall45 in 1957. The principles of this approach underlie the treatment of posthysterectomy vaginal vault prolapse.46
Obesity is a risk factor for POP. Studies with larger sample sizes are more likely to report an increased risk of prolapse.47 In a meta-analysis of 22 studies providing estimates for overweight (body mass index [BMI] 25 to <30) and obese (BMI ≥30), there was a 36% (risk ratio [RR] 1.36, 95% CI 1.20 to 1.53) and 47% (RR 1.47, 95% CI 1.35 to 1.59) higher risk of prolapse, respectively, compared to women of normal (<25) BMI.48
Chronic straining also plays a role in prolapse. Obstructed defecation syndrome, dyssynergic defecation, and constipation are often associated with prolapse—especially posterior vaginal prolapse. The anatomic defects leading to this type of prolapse were first described according to the relative location in the posterior vagina—low, middle, and upper vagina. These forms of prolapse correlate with injuries to the perineal membrane, failure of the levator ani muscles to close the genital hiatus, and failure of the uterosacral ligaments to provide upward suspension, respectively.
Physical stressors and socioeconomic status (SES) are risk factors about which little is known. In a cross-sectional analysis, women found to have POP at least 1 cm beyond the hymen were sevenfold more likely to work as laborers/factory workers and to have income <$10,000. It is not clear to what extent nutrition, access to care, and health status might confound this relationship.24 In the only other study evaluating SES, women living in Massachusetts were surveyed about whether or not a health care provider had advised them if they had uterine prolapse or vaginal prolapse involving the bladder or rectum. Those with income >$70,000 were more likely to be aware of such a diagnosis. These patients did not have a pelvic exam to describe the severity of prolapse and were not asked about symptoms.49
Prevalence and Impact
In population-based studies, UI symptoms affect 25% to 60% of women,39,50,51,52,53,54,55,56 with several clustering around a 30% prevalence for any incontinence. Managing UI symptoms is a burden for patients, caregivers, and the community. Women who are incontinent are more likely to experience anxiety and depression,57 in addition to being more likely to have unmet care needs and to be dependent on others.55 The average annual cost per woman to manage UI was $873 in community-dwelling settings and $5,325 for those older than age 65 years in an institutionalized setting. The total costs of UI for women in the United States was $11.2 billion in 2000.58