Volume 59, Number 10

OBSTETRICAL AND GYNECOLOGICAL SURVEY

Copyright © 2004
by Lippincott Williams & Wilkins CME REVIEWARTICLE 28
CHIEF EDITOR’S NOTE: This article is part of a series of continuing education activities in this Journal through which a total
of 36 AMA/PRA category 1 credit hours can be earned in 0. Instructions for how CME credits can be earned appear on the
last page of the Table of Contents.

Exercise and Urinary Incontinence

in Women
Kanli Jiang, PhD,* Joseph M. Novi, DO,† Suzanne Darnell,‡ and
Lily A. Arya, MD§
*Medical Student, University of Pennsylvania, Philadelphia, Pennsylvania; †Fellow, Division of
Urogynecology and Reconstructive Pelvic Surgery, Department of Obstetrics and Gynecology, University of
Pennsylvania, Philadelphia, Pennsylvania; ‡Medical Student, University of Minnesota Minneapolis,
Minnesota; and §Assistant Professor, Department of Obstetrics and Gynecology, Division of Urogynecology
and Reconstructive Pelvic Surgery, University of Pennsylvania, Philadelphia, Pennsylvania

Urinary incontinence is a common problem in women and may significantly impair their quality of
life. Although women often report stress urinary incontinence during exercise, current data indi-
cates that most types of exercise are not a risk factor for the development of urinary incontinence.
However, certain extreme high-impact sports such as parachute jumping may cause pelvic organ
support defects that result in stress urinary incontinence. Eating disorders also increase the risk of
urinary incontinence in athletes. Overall, women should be encouraged to pursue physical activity
that will benefit their general health without the risk of development of urinary incontinence later in
life. Women athletes should be counseled about the increased risk of urinary incontinence with
ultra high-impact sports and eating disorders.
Target Audience: Obstetricians & Gynecologists, Family Practitioners
Learning Objectives: After completion of this article, the reader should be able to list the most common
types of urinary incontinence, to outline the risk factors for the development of urinary incontinence, and
to describe the pathogenesis of exercise-associated urinary incontinence.

Urinary incontinence is a common problem affect- urinary incontinence (14%) (2). According to the
ing women of all ages. It is defined by the Interna- International Continence Society, stress urinary in-
tional Continence Society as “the complaint of any continence is “the complaint of involuntary leakage
involuntary leakage of urine that is objectively de- of urine on effort or exertion, or on sneezing or
monstrable and is a social and hygienic problem” (1). coughing”; and urge urinary incontinence is “the
The prevalence of urinary incontinence ranges from complaint of involuntary leakage of urine accompa-
5% to 54% depending on the definition used and the nied by or immediately preceded by urgency” (1).
population studied, with a mean prevalence of 28% Mixed incontinence describes the coexistence of both
(2). The most common type of female urinary incon- symptoms. Urinary incontinence impacts many as-
tinence is stress urinary incontinence, affecting 50% pects of a woman’s life, including health, psychoso-
of all incontinent women. The next most common cial, occupation, and leisure (3).
type is mixed incontinence (32%), followed by urge
Reprint requests to: Joseph M. Novi, DO, Fellow, Division of Urogy-
necology and Reconstructive Pelvic Surgery, Department of Obstetrics RISK FACTORS FOR URINARY
and Gynecology, University of Pennsylvania, 5 Penn Tower, 3400 INCONTINENCE
Spruce Street, Philadelphia, PA 19104. E-mail: [email protected].
The authors have disclosed no significant financial or other Multiple risk factors have been found to be asso-
relationship with any commercial entity. ciated with urinary incontinence. These include age,
717
718 Obstetrical and Gynecological Survey

parity, vaginal delivery, obesity, race, and lifestyle a risk factor for stress urinary incontinence (OR,
factors. 1.62; 95% CI, 1.18–2.22) (8). Interestingly, in the
same study, daily bread intake was found to be as-
sociated with a reduced risk of stress urinary incon-
Age
tinence (OR, 0.76; 95% CI, 0.61–0.96). High caf-
Several studies have identified increasing age as a feine intake has also been associated with a higher
major risk factor in the development of urinary in- risk of urge urinary incontinence (OR, 2.4; 95% CI,
continence. A recent review of the literature showed 1.1–6.5) (3,9). No statistically significant correlation
that there are 2 distinct peaks in the prevalence of between urinary incontinence and alcohol intake has
urinary incontinence, one in the fifth decade of life, been reported (2). One study has shown that eating
and the second in the eighth decade (2). The preva- disorders such as anorexia nervosa were associated
lence of significant urinary incontinence increases with increased risk of all urinary symptoms, includ-
from the second to the eighth decades (2). ing urgency, frequency, nocturia, stress and urge
urinary incontinence (10). This observation has also
been reported by Bo et al., who noted an increased
Pregnancy-Related Factors
prevalence of stress urinary incontinence in athletes
Almost 32% of women report stress urinary incon- with eating disorders (P ⫽ 0.03) (11).
tinence during pregnancy and most tends to be new
onset (4). It has been shown that women who report
stress urinary incontinence during their first preg- Smoking
nancy and those who develop stress incontinence Smoking has been shown to be a risk factor for
during the immediate postpartum period are at high both stress and urge urinary incontinence. In 1 study,
risk for persistent stress urinary incontinence 5 years women with detrusor instability were more likely to
later (3). Other independent obstetric risk factors of be current smokers than control subjects (9). Women
urinary incontinence are multiparity and vaginal de- smokers have also been reported to have a 2.5-fold
livery (2,3). Forceps-assisted vaginal delivery may increased risk for developing stress urinary inconti-
be an additional risk (5). Data regarding the impact nence than control subjects (12).
of the size of the baby, fetal head, and the length of
second stage of labor on the development of urinary
incontinence are controversial (3). Exercise
Women commonly report stress urinary inconti-
Obesity nence during physical activity; however, it is not
clear whether exercise is a causative factor in the
Increased body mass index (BMI) has been estab- development of stress urinary incontinence.
lished as an independent risk factor for the develop- The purpose of this article is to review the litera-
ment of urinary incontinence. It has been shown that ture regarding the role of physical exercise in the
with each unit increase in BMI, there is a 5% increase etiology of urinary incontinence.
in the odds of having urinary leakage (6).

EXERCISE AND INCONTINENCE

Race
Prevalence of Incontinence in Exercising
After adjusting for known risk factors such as age,
Women
parity, and BMI, several studies have demonstrated
that white race is a significant risk factor for the Nygaard et al. reported in 1990 that as many as 30%
development of stress urinary incontinence (2,3,7). of women complain of urinary leakage during physical
In 1 study, black race was a significant predictor of activity. The highest rates were reported in high-impact
urge incontinence (odds ratio [OR], 2.6; 95% confi- activities such as running (38%) and high-impact aer-
dence interval [CI], 1.45–4.80) (7). obics (36%). The mean age of women in that study was
38.5 years (13). A subsequent study by the same au-
thors showed that exercise-associated incontinence is
Diet
also a common problem in much younger, nulliparous
In a large longitudinal study, daily or more fre- women (mean age, 19.9 years) with as many as 28%
quent intake of carbonated beverages was found to be reporting stress urinary incontinence during exercise
 Exercise and Urinary Incontinence Y CME Review Article 719

(14). A more recent study in 2002 by Eliasson et al. 50.0%). This study suggests that women actively
described similar findings in nulliparous elite trampo- competing in high-impact sports report higher rates
linists with 80% reporting involuntary leakage (15). of stress urinary incontinence than those active in
Thus, there appears to be a high prevalence of stress low-impact sports. Importantly, this disparity does
urinary incontinence in women who participate in high- not appear to persist later in life (16).
impact exercise. In a large prospective study involving 660 elite
athletes and 765 nonathletes, Bo et al. reported no
significant difference in the prevalence of stress or
Exercise as a Risk Factor for Incontinence urge urinary incontinence between cases and control
Several studies have been published in the last few subjects. In this study, urinary incontinence was di-
years comparing the rate of urinary incontinence in agnosed by a standardized questionnaire and a clin-
women who exercise and those who do not. These ical interview. Athletes were divided into different
results are summarized in Table 1. sport groups, including technical, endurance, es-
A 1990 study by Nygaard et al. reported that there thetic, weight-dependent, ball games, power, and
was no difference in the prevalence of urinary incon- gravity. In addition, data on eating disorders were
tinence in women who exercised and nonexercisers. collected using Diagnostic and Statistical Manual of
However, this study was limited by a small sample Mental Disorders, 4th edition, criteria. There was no
size. statistically significant difference in the prevalence
In 1997, the same group performed a retrospective of stress or urge urinary incontinence between ath-
cohort study comparing the rate of urinary inconti- letes and control subjects (41% vs. 39% with stress
nence in women who participated in low-impact urinary incontinence and 16% vs. 19% with urge
(swimming) and high-impact (gymnastics and track urinary incontinence, respectively). However, the
and field) sports. A group of 104 female former prevalence of stress urinary incontinence (49.5%)
Olympians who had competed in these sport events and urge incontinence (20%) in athletes with eating
were identified. They were asked to recall urinary disorders was significantly higher than in healthy
incontinence when they were actively competing in athletes, 38.8% (P ⫽ 0.003) and 15% (P ⫽ 0.048),
their sports and also to report current symptoms of respectively.
urinary incontinence. Women who participated in A small study was conducted by Davis et al. in
high-impact sports reported a higher rate of urinary 1996 investigating the role of ultra high-impact
incontinence while they had been actively competing sports such as parachute jumping in the development
than those in low-impact sports. However, rates of of urinary incontinence. In this study of 512 female
current urinary incontinence were not statistically soldiers in an airborne infantry unit, 9 previously
different in these 2 groups of women (41.1% vs. continent women reported severe urinary inconti-

Table 1
Relationship between exercise and urinary incontinence
Mean age
Author Year Design Subjects N (years) Prevalence of UI Comments
Control Exercise P
Nygaard 1990 Cross-sectional Female volunteers 326 38.5 N/A 30% NS Small number of
controls
Davis 1996 Case series Airborne infantry 9 21–34 N/A 9 with severe UI N/A UI confirmed with
trainees urodynamics
Nygaard 1997 Retrospective Former female 104 46.2 N ⫽ 48 N ⫽ 56 NS Control women
Cohort Olympians were low-impact
sports Olympians
v. 42.4 50.0% 41.1%
Bo 2001 Case-control Elite athletes v. 1425 21.5 N ⫽ 765 N ⫽ 660 No objective
non-athletes measurements
for UI
v. 24.2 SUI (39%) SUI (41%) NS
UUI (19%) UUI (16%) NS
UI ⫽ urinary incontinence; SUI ⫽ stress urinary incontinence; UUI ⫽ urge urinary incontinence; N/A ⫽ non-applicable; NS ⫽ non-
significant.
720 Obstetrical and Gynecological Survey

nence on completion of training. Objective testing fascia and lead to an increased risk for urinary in-
revealed that 3 of the 9 women had urge urinary continence. Furthermore, women with eating disor-
incontinence and 6 had stress urinary incontinence. ders that involve self-induced vomiting may be sub-
On physical examination, the 6 women with stress jecting the pelvic floor and endopelvic fascia to
urinary incontinence showed evidence of bladder significant repetitive pressures might have a cumu-
neck hypermobility, anterior vaginal wall prolapse, lative negative impact on these structures.
and loss of urine with Valsalva maneuvers. Interest-
ingly, 5 of them could recall a specific traumatic
event associated with a parachute jump after which CONCLUSION
her urinary incontinence began. All 5 recalled a pain-
ful feeling in the pelvic region at the time of the Current data on the role of exercise in the devel-
incident. All 6 opted for surgical intervention at opment of urinary incontinence is limited. Existing
which time the pubocervical fascia was noted to be studies are compromised by retrospective design and
separated from the pelvic side wall (17). Despite the failure to control for confounding factors such as
small size of this study, it appears that activities that diet, hormonal factors, and BMI. Additional prospec-
predispose women to pelvic organ support damage tive studies are required to fully evaluate the role of
could be a risk factor for urinary incontinence. exercise in the development of urinary incontinence.
Nevertheless, based on existing literature, it is pos-
Pathogenesis of Exercise-Associated sible to draw some conclusions. Health problems
Incontinence reported in women athletes include an increased in-
Stress urinary incontinence commonly reported cidence of sports-related injuries, arthritis, and men-
during physical activity is likely the effect on the strual disturbances (19). As a result of its potential
normal continence mechanism of excessive increases for serious social and psychologic impacts, the high
in intraabdominal pressure that occur during exer- prevalence of urinary incontinence associated with
cise. Normally, when intraabdominal pressure rises physical activities deserves special attention. Based
during coughing or exercise, continence is main- on our review of the available data, it appears that no
tained because this pressure increase results in direct causal relationship has been identified between most
mechanical compression of urethral sphincter against forms of exercise and the development of urinary
the supporting endopelvic fascia of the vagina (18). incontinence. However, extreme high-impact sports
This effectively closes the urethral sphincter. It is may increase the risk of developing urinary inconti-
possible that during high-impact sports such as vig- nence by causing direct damage to the pelvic organ
orous running, gymnastics, or intense aerobics, intra- support structures. Additionally, women athletes who
abdominal pressure rises high enough to allow intra- have eating disorders are at high risk for urinary
vesical pressure to exceed intraurethral pressure. incontinence even if they participate in less intense
This excessive rise may overwhelm even a normally exercise.
functioning continence mechanism, leading to stress Regular and moderate exercise is known to promote
urinary incontinence. It seems unlikely that even cardiovascular health, weight control, muscle strength,
high-impact sports result in direct physical damage to bone density, as well as psychologic health. In light of
the urethral support system (16). However, extreme our present knowledge, we recommend that women be
high-impact sports such as parachute jumping may encouraged to pursue physical activity that will benefit
cause direct physical damage to the pelvic floor such their general health. They should be counseled about
that separation of the pubocervical fascia from the the increased risk of urinary incontinence with ultra
pelvic side wall occurs (17). This would result in high-impact sports. All women should be encouraged to
poor support of the bladder neck with consequent maintain normal eating habits and body weight to aid in
bladder neck hypermobility and stress incontinence. preserving continence.
The high prevalence of urinary incontinence in
women athletes who also have eating disorders is
harder to explain. In 1992, Bump et al. postulated REFERENCES
that urinary incontinence in smokers may be the 1. Abrams P, Cardozo L, Fall M, et al. The standardisation of
result of the damaging effect of nicotine on the terminology in lower urinary tract function: Report from the
endopelvic fascia (12). It is possible that the lack of Standardisation Sub-committee of the International Conti-
nence Society. Urology 2003;61:37–49.
appropriate nutrients, as may occur in women with 2. Minassian VA, Drutz HP, Al-Badr A. Urinary incontinency as a
eating disorders, might also weaken the endopelvic worldwide problem. Int J Gynecol Obstet 2003;82:327–338.
 Exercise and Urinary Incontinence Y CME Review Article 721

3. Holroyd-Ledue JM, Straus SE. Management of urinary incon- 11. Bo K, Borgen JS. Prevalence of stress and urge urinary in-
tinence in women: Scientific review. JAMA 2004;291:986–995. continence in elite athletes and controls. Med Sci Sports
4. Viktrup L, Lose G, Rolff M, et al. The symptom of stress Exerc 2001;33:1797–1802.
incontinence caused by pregnancy or delivery in primiparas. 12. Bump RC, McClish DK. Cigarette smoking and urinary incon-
Obstet Gynecol 1992;79:945–949. tinence in women. Am J Obstet Gynecol 1992;167:1213–
5. Arya LA, Jackson ND, Myers DL, et al. Risk of new-onset 1218.
urinary incontinence after forceps and vacuum delivery in 13. Nygaard I, DeLancey JOL, Arnsdorf L, et al. Exercise and
primiparous women. Am J Obstet Gynecol 2001;185:1318– incontinence. Obstet Gynecol 1990;75:848–851.
1323. 14. Nygaard IE, Thompson FL, Svengalis SL, et al. Urinary incon-
6. Sampselle CM, Harlow SD, Skurnick J, et al. Urinary inconti- tinence in elite nulliparous athletes. Obstet Gynecol 1994;84:
nence predictors and life impact in ethnically diverse perim- 183–187.
enopausal women. Obstet Gynecol 2002;100:1230–1238. 15. Eliasson K, Larsson T, Mattsson E. Prevalence of stress in-
7. Graham CA, Mallett VT. Race as a predictor of urinary incon- continence in nulliparous elite trampolinists. Scand J Med Sci
tinence and pelvic organ prolapse. Am J Obstet Gynecol Sports 2002;12:106–110.
2001;185:116–120. 16. Nygaard IE. Does prolonged high-impact activity contribute to
8. Dallosso MM, McGrother CW, Matthews RJ, et al. The asso- later urinary incontinence? A retrospective cohort study of
ciation of diet and other lifestyle factors with overactive blad- female Olympians. Obstet Gynecol 1997;90:718–722.
der and stress incontinence: A longitudinal study in women. 17. Davis G, Goodman M. Stress urinary incontinence in nullipa-
BJU Int 2003;92:69–77. rous female soldiers in airborne infantry training. Journal of
9. Arya LA, Myers DL, Jackson ND. Dietary caffeine intake and Pelvic Surgery 1996;2:68–71.
the risk for detrusor instability: A case-control study. Obstet 18. Delancey JOL. Structural support of the urethra as it relates to
Gynecol 2000;96:85–89. stress urinary incontinence: The hammock hypothesis. Am J
10. Hextall AS, Majid L, Cardozo K, et al. A prospective controlled Obstet Gynecol 1994;170:1713–1723.
study of urinary symptoms in women with severe anorexia 19. Warren MP, Sjantha S. The female athlete. Bailliere’s Clinical
nervosa [Abstract]. Neurourol Urodyn 1999;18:398–399. Endocrinology and Metabolism 2000;14:37–53.