The Epidemiology, Natural History and Prevention of Pelvic Floor Disorders

Pelvic floor disorders (PFD) constitute a huge global health problem affecting hundreds of millions of women throughout the world. The prevalence of at least one form of PFD (urinary incontinence (UI), pelvic organ prolapse (POP) or fecal incontinence (FI)) has been reported to be as high as 46%.¹ In addition many women may have a combination of PFDs. PFDs have a profound influence on well-being and quality of life as well as being of immense economic importance for the health service.^{1, 2, 3} Numerous risk factors for the development of PFDs have been identified¹ and many of these risk factors are shared by two or more forms of PFD. Vaginal childbirth is probably the most important factor in the etiology of PFD and definitely contributes to the occurrence of some or all types of PFD.¹

The structure of the female pelvis may be seen as a compromise between the preconditions for a regulated and controlled evacuation of urine and feces, and also allowing the passage of a fetus through the pelvic hiatus at vaginal delivery.

Viewed from above the pelvic floor is a hexagonally shaped entity consisting of striated muscles, connective tissue, and the organs that fill the lower pelvic cavity. The muscles of the pelvic diaphragm are collectively referred to as the levator ani muscles. These muscles form three regions. Region 1 is composed of the ileococcygeus muscle that builds a flat, horizontally orientated plate behind the anal canal from the arcus tendineus and the pelvic sidewall on one side to the opposite side passing in front of the sacrum in the midline. Region 2 encompasses the urogenital hiatus – through which the urethra and the vagina pass – and includes the pubovisceral muscles (the pubovaginal and puboanal muscle) that arise from either side of the symphysis and attach to and encircle the urethra, the vagina, and the perineal body. Contraction of these muscles closes the urogenital hiatus against the opening force of the intra-abdominal pressure. Region 3 consists of the puborectal muscle, which is a muscle sling that arises lateral to the pubovisceral muscles at the symphysis and fuses behind the rectoanal junction cephalic to the external anal sphincter.^{4, 5}

In contrast to other striated muscles of the human body, the muscles of the pelvic floor exhibit continuous activity both at rest and during sleep. The tonic, continuous activity is considered to originate from a sacral spinal center.⁶ Postural changes such as movements of the body, abdominal straining, speaking, coughing, etc. transiently increase the tone of the pelvic floor muscles.⁷ Phasic contractions of the levator ani muscles are coordinated and exert a force in a ventro-cephalic direction, which keeps the urogenital hiatus closed against increases in intra-abdominal pressure. A voluntary contraction exerts an even stronger force to close the urogenital hiatus. The hydrostatic pressure carried by the pelvic floor in the upright position is about 40 cm H₂O and 140 cm H₂O during maximum cough.⁸

The endopelvic fascia attaches the bladder, uterus, and rectum to the pelvic walls. The uterosacral and cardinal ligaments are important fascial condensations involved in supporting the uterus and upper vagina. In the distal vagina its wall is directly attached to the surrounding structures and fuses with the urethra anteriorly and to the perineal body posteriorly.

The anal sphincter complex consists of three muscular components: the internal anal sphincter (IAS), the external anal sphincter (EAS) and the puborectal muscle sling. The EAS surrounds the IAS and extends more caudally. All three sphincter components exhibit constant tonic activity to close the anal canal.^{9, 10} The EAS and the levator ani complex are also responsible for voluntary contraction and phasic reflex contractions. During defecation, by distending the rectum, the IAS is relaxed, which is mediated by the rectoanal inhibitory reflex along the intramural myenteric plexus of the rectum.¹¹ The IAS is the strongest smooth muscle sphincter of the human body and is responsible for maintaining anal continence at rest. Pudendal nerve branches from the S2-S4 segments travel through Alcocks canal into the ischiorectal fossa on both sides and innervate the levator ani muscles on their inferior surface. Direct branches from the S2-S4 segments innervate the cranial surface of the levator ani muscles.¹²

Urinary incontinence

The reported prevalence rates of UI in women vary considerably due to differences in target population, definitions, study designs and methods chosen for statistical analysis¹ which complicates the comparison of results. The prevalence of UI in women varies during their lifetime also depending on the effect of age, weight, parity, lifestyle factors, and general health status. In population-based studies the prevalence of UI (any) ranged between 5% and 69%, although most studies reported 25–45%. In most studies the prevalence of isolated stress UI was 10–39%, the prevalence of mixed UI 7.5–25%, and isolated urgency UI in 1–7%.¹

UI is, however, not static but dynamic and many factors may contribute to incidence, progression or remission. There are only a few studies describing progression as well as remission, in the short term, of UI in the general population as well as in selected groups of the population. The mean annual incidence of UI seems to range from 1 to 9% while estimates of remission are more varying, 4–30%.^{13, 14, 15}

Wennberg et al.¹⁵ studied the prevalence of UI in the same women (aged > 20 years) over time in order to assess possible progression or regression. A self-administered postal questionnaire with questions regarding UI, OAB and other lower urinary tract symptoms (LUTS) was sent to a random sample of the total population of women in 1991. The same women who responded to the questionnaire in 1991 and who were still alive and available in the population register 16 years later were re-assessed using a similar self-administered postal questionnaire. The overall prevalence of UI, increased from 15 to 28% from 1991 to 2007 and the incidence rate of UI was 21% while the corresponding remission rate was 34%.

Numerous risk factors for UI have been identified and are summarized in Table 1.¹ Many of these risk factors for UI are also shared for by FI and POP. Vaginal childbirth is probably the most important factor in the etiology of PFD and results in the combination of some or all of the following conditions: UI, FI and POP.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. Body mass is an important, modifiable risk factor for UI. 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. 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. Diabetes is a risk factor for UI in most studies. While 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. Menopause, as generally defined, does not appear to be an independent risk factor for stress UI. Hysterectomy remains a possible risk factor for later UI, but the evidence is inconsistent. Moderate to severe dementia in older women is a moderate to strong independent risk factor for UI. Whether interventions to maintain or improve cognitive functioning also reduce UI has not been evaluated. 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. Data from twin studies suggests that there is a substantial genetic component to UI. Other potential risk factors, including smoking, diet, depression, constipation, UTIs, and exercise, while associated with UI, have not been established as etiological risk factors and are in fact difficult to study with observational data because of the potential for unmeasured confounding factors and questions of direction of the association.

Table 1. Risk factors for urinary and fecal incontinence

The prevalence of UI in nulliparous women of childbearing age has been reported to be 10–15%.^{16, 17, 18, 19, 20} UI preceding pregnancy in nulliparous women has been shown to be a strong indicator for increased prevalence of UI 4–12 years postpartum.^{21, 22} Pregnancy in itself, independent of labor and delivery practices, seems to be a risk factor for postpartum UI,^{23, 24} especially if the incontinence started during the first trimester.²⁵ During pregnancy the prevalence of UI increases with gestational age,²⁶ so that more than half of all women report UI during the third trimester.^{27, 28, 29} Stress UI and mixed UI increased the most during pregnancy compared to before pregnancy, whereas urgency UI did not change during the same period.³⁰ During the first 3 months postpartum UI prevalence was 30% and most women had stress UI.³¹ In uncomplicated courses of pregnancy and labor UI usually declines rapidly during the first 3 months following childbirth, indicating that most symptoms are part of a normal pregnancy.²³ Several studies have also demonstrated that postpartum UI is a risk factor for UI after longer (7 months to 6 years) terms of follow-up.^{24, 32, 33, 34}

The first delivery is considered to increase the prevalence of UI the most, and recent studies have demonstrated a further increase for each delivery.^{35, 36, 37, 38, 39} Many cross-sectional and several longitudinal studies show a protective effect of cesarean section for UI.^{40, 41, 42} BMI is considered to be an established risk factor for UI,³⁸ whereas the association between UI and age is complicated by confounders.⁴³

The prevalence of subtypes of UI is age dependent. Pooled data from several studies showed that the prevalence of stress UI (SUI) peaks in the 4th decade and then declines gradually with increasing age, whereas mixed UI (MUI) starts to increase in the 4th and urgency UI (UUI) in the 6th decade. The prevalence of all three subtypes subsequently converges to approximately 30% in the 8th decade.⁴⁴

Pelvic organ prolapse

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. Prolapse is thus a continuous condition when measured by visual inspection of the vaginal wall during valsalva. For clinical purposes, the degree of POP is commonly described as above the introitus, at the introitus, or beyond the introitus with or without valsalva. Determining POP based on self-reported symptoms is difficult because of the lack of specificity and sensitivity of most symptoms attributed to POP and the fact that prolapse above the level of the hymeneal ring is usually asymptomatic.

However, assessments of the prevalence of symptomatic pelvic organ prolapse (sPOP) in women are remarkably consistent with a prevalence between 4% and 11%.^{1, 45, 46, 47} POP is a rare condition in nulliparous women and in women after one or several cesarean sections indicating that mode of delivery is more important than pregnancy alone.^{1, 48, 49} Vaginal delivery is associated with an increased risk of POP and increasing number of births is positively associated with increased risk of POP.^{1, 46, 48, 49, 50, 51} There is increasing evidence to suggest that cesarean section is protective for POP.^{1, 50, 51}

Several risk factors for POP have been identified.¹ Childbirth is associated with an increased risk of pelvic organ prolapse later in life and an increasing number of childbirths is positively associated with the risk. Current evidence also suggests that some surgical procedures, e.g. hysterectomy, increase the risk  of subsequent 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. Life style factors and socioeconomic indices may be associated with the risk of POP in both industrialized and non-industrialized countries.¹

The main risk factor for POP is vaginal delivery.^{1, 47, 48, 52} Several other obstetric interventions and events have been associated with POP.¹ In one study, maternal age and use of epidural analgesia was associated with an increased need for POP surgery.⁵³ A case–control study found no significant association with maternal age, instrumental delivery (forceps or vacuum), or length of delivery when comparing women with prolapse to randomly selected controls.⁴⁹ However, other studies have reported an increased risk of POP following a forceps delivery.¹ The inconsistency and wide variety in the magnitude of the risk estimates reported in these studies suggest that most studies so far lack sufficient statistical power for valid conclusions. BMI has been reported to be a risk factor for POP in several studies.^{1, 48, 50, 51, 54}

Fecal and anal incontinence

In a recent cross-sectional study on fecal incontinence ≥ 1/month in adults (aged >18 years) the overall community prevalence of FI was 12.4%.⁵⁵ Many earlier studies have reported much lower prevalence rates,⁵⁶ but those figures are almost certainly too low, as it has been shown that more than every second person with FI does not report the symptoms to their doctor because of embarrassment or pessimism about its curability.⁵⁷

Reported prevalence of anal incontinence (AI) and FI after childbirth vary considerably due to lack of consistency in definition, type of questionnaire, selected population, and mainly to length of follow-up.⁵⁸ During pregnancy AI was reported by 10% of nulliparous women, of which 90% had incontinence of flatus only.⁵⁹ In another report on FI during the pregnancy of nulliparous women no increase of incontinence occurred during the second and third trimester.⁶⁰

Fecal/anal incontinence is common during the first months postpartum,^{61, 62, 63} but the majority of women with these early problems will recover.⁶⁴ In the study of MacArthur et al. women were followed for 12 years. The prevalence of FI among those who had a spontaneous vaginal delivery was 11.5% compared to 11.7% women delivered exclusively by cesarean section.⁶⁵

Age is widely considered to be a major risk factor for FI.⁵⁸ However, in parous women the association between FI and age is probably not linear, since FI usually has a late onset, in most women after the age of 40.^{66, 67} A number of studies have shown an association between BMI and fecal/anal incontinence.^{68, 69} Weight reduction has also been shown to result in improvement of FI.⁷⁰

The association between sphincter injury at delivery and subsequent late incontinence of stool is complex. To date, much attention has been focused on overt sphincter and levator ani injuries, but due to methodological deficiencies results are conflicting even between studies of women at a time remote from the date of birth.^{71, 72, 73} In primiparous women clinical sphincter injury has been demonstrated in about 7% and by endoanal ultrasonography in up to 35%.⁷⁴ A 2nd degree or more tear has been shown to be associated with an almost doubled prevalence of late FI compared with women without  2nd degree or more  tears – 22.8 versus 13.9%.⁷⁵

The role of episiotomy in the short term for subsequent bowel continence is uncertain. A systematic review in 2005 concluded that the effect of episiotomy on the development of pelvic floor disorders remains unknown.⁷⁶ In a study of late FI it was shown that the rate of FI after an episiotomy was similar to that after cesarean section (11.1 versus 10.6%).⁷⁵

Pelvic floor disorders, such as urinary incontinence, are associated with a range of risk factors, e.g. obesity, mode of delivery, etc., many of which are modifiable. However, lifestyle changes with behavioral modifications that can prevent PFD need to be adopted from an early phase of life. Obesity, pregnancy and in particular labor and delivery practice, are potentially modifiable risk factors for UI.⁷⁷ There is a correlation between pelvic-floor related surgery and UI and FI. The evidence would suggest that in some cases incontinence can be prevented. With primary prevention, actions taken early, even in the absence of incontinence, may avoid or delay its onset. There is also evidence that some risk factors for incontinence are modifiable and can be reduced through preventive measures. However, it is unlikely that urinary and fecal incontinence can be eradicated and therefore it is important to improve preventive measures and identify new and better management modalities in order to improve the health of individuals afflicted and reduce the costs for society.⁷⁷

Vaginal delivery has been strongly implicated in the etiology of pelvic floor disorders. Prevention of PFD should be discussed with pregnant women as part of their routine antenatal care. This should include pelvic floor muscle training, modifiable risk factors (avoid smoking and constipation and maintaining a normal BMI)⁷⁸ and possible elective cesarean section for women at greatly increased risk of PFD. Up until now it has been difficult to counsel women on the risk factors for subsequent PFD as there have been few good quality long-term data available. We now have moderately robust epidemiological data at 12^{50, 65,} and 20 years after delivery (primiparous women)^{41, 42, 51, 75} and objective pathophysiological data (pudendal nerve trauma and levator avulsion).⁷⁹ Major risk factors for subsequent pelvic dysfunction are antenatal UI,¹ ethnicity,¹ age at first birth,^{1, 41, 42, 50, 51, 65, 75,} BMI,^{1, 41, 42, 51, 75} family history (mother and sister’s) of PFD,¹ baby’s weight^{1, 41, 42, 51, 78} and maternal height (if under 160 cm and with the baby >4 kg).^{1, 41, 42, 51}

Recently, cesarean section has been shown to provide partial protection for POP and to a lesser degree urinary incontinence. Regarding POP, exclusive cesarean section is associated with a reduced risk of objectively measured signs of prolapse 12 years after delivery⁵⁰ and symptoms of prolapse 20 years after delivery.⁵¹ There is also a significant increase in prolapse surgery with vaginal delivery in comparison to cesarean section and a dramatically increased risk with forceps in comparison to a cesarean delivery.⁸⁰

It has been suggested⁸¹ that the physical features of the mother and the baby (using known evidence based risk factors, UR-CHOICE – UI before pregnancy; Race ethnicity; Child bearing at what age; Height of mother; Overweight; Inheritance (family history); Children (number of children desired); and Estimated fetal weight) can be scored antenatally and may be considered along with the woman’s informed choice regarding her preferred route of delivery and the number of children she wishes in order to avoid future pelvic floor disorders. This idea is, however, not new and has been previously proposed and evaluated by colleagues at the Cleveland Clinic.^82, The application of the UR-CHOICE scoring system could help provide women with evidence based pre-labor advice. Following appropriate counselling with their proposed birth attendant, serious decisions and choices can be made in order to help maximize the chances of the mother achieving a problem-free vaginal birth and avoid unnecessarily high incidences of urinary and fecal incontinence and prolapsed pelvic organs requiring surgery in the future.