Barber, Matthew D. MD, MHS; Visco, Anthony G. MD; Wyman, Jean F. RN, PhD; Fantl, J. Andrew MD; Bump, Richard C. MD; for the Continence Program for Women Research Group
Difficulties with sexual function have been associated with urinary incontinence and pelvic organ prolapse.1,2 Sutherst and Brown found that 43% of study participants with urinary incontinence felt that their urinary disorder had adversely affected sexual relations.3 Walters et al found that women with genuine stress incontinence and detrusor instability reported more sexual dysfunction than did continent controls.4 Others, however, have reported that sexual function in women with prolapse and urinary incontinence does not differ from that in continent women without prolapse.5
Data are limited on the effect of therapy for incontinence or prolapse on sexual function. Haase and Skibsted reported that after reconstructive surgery, sexual function improved in 24% of women with urinary incontinence or genital descensus and deteriorated in 5%.2 Information on the effect of more conservative therapies, such as behavioral therapy or estrogen therapy, is lacking.
The Continence Program for Women was a prospective multicenter trial that included women with advanced pelvic organ prolapse and women with urinary incontinence. The trial had several therapeutic arms, including surgical therapy, behavioral therapy, and hormonal therapy. Sexual function and satisfaction were measured by using a condition‐specific sexual function questionnaire at baseline and at several intervals after therapy. Using data from this large sample of community‐dwelling women, we sought to describe and compare sexual function and satisfaction in women with urinary incontinence and pelvic organ prolapse. We also aimed to describe and compare the effect of surgical, behavioral, and estrogen therapy on sexual function and satisfaction in women with urinary incontinence and pelvic organ prolapse.
MATERIALS AND METHODS
We performed a secondary analysis of three clinical trials conducted by the Continence Program for Women and funded by the National Institute on Aging. The study population of the Continence Program for Women consisted of 348 women recruited from three academic medical centers (Medical College of Virginia [n = 222], Bowman Gray Medical Center [n = 113], and Duke University Medical Center [n = 13]) as part of a prospective multi‐armed clinical trial between January 1989 and June 1994. All women were 45 years or older, community dwelling, ambulatory, and mentally competent (Mini‐Mental Status Examination score > 23). Three hundred sixteen patients had urinary incontinence, defined as involuntary urine loss at least once per week and no pelvic organ prolapse beyond the hymen with straining, and 32 patients had stage III or IV pelvic organ prolapse. Women with urinary incontinence were recruited through newspaper advertisements and articles and from the referral practices of the senior study investigators. Women with prolapse were recruited only from the referral practices of the senior study investigators. Exclusion criteria were reversible causes of urinary incontinence, uncontrolled metabolic conditions, functional disability limiting use of the toilet, indwelling catheterization, genitourinary fistula, and chronic urinary tract infections. Before enrollment, all participants gave informed consent as approved by the institutional human subjects research committee; institutional review board approval was obtained at each of the centers involved in the study. Five women (2 with urinary incontinence and 3 with pelvic organ prolapse) did not complete the sexual function questionnaire at baseline and were excluded from this analysis, leaving 343 women (314 with urinary incontinence and 29 with prolapse) in our study.
Women underwent the same pretherapy and post‐therapy evaluations according to an established, standardized, written protocol. Before therapy was given, each participant completed a standardized urogynecologic interview, a complete physical examination, a directed neurologic examination, a prospective bladder diary,6 quantitative pad test,7 and an urodynamic evaluation. Details of the techniques used for urodynamic evaluation are published elsewhere.8 Urodynamic techniques and measurements, terms, and diagnostic criteria conform to the recommendations of the International Continence Society.9,10
Figure 1 shows the design of the Continence Program for Women. Participants with urinary incontinence were stratified on the basis of urodynamic diagnosis, estrogen status, and treatment preference to receive estrogen therapy, behavioral therapy, or surgery. Women who had urinary incontinence and were hypoestrogenic (plasma estradiol level ≤ 30 pg/mL) were enrolled in a trial of estrogen therapy and were randomized to receive placebo or conjugated equine estrogens (0.625 mg) continuously with medroxyprogesterone (10 mg) for 10 days each month. Participants with urinary incontinence who were estrogenized were enrolled in the behavioral therapy trial or the surgical trial. Those with a urodynamic diagnosis of detrusor instability or mixed urinary incontinence were enrolled in the behavioral trial. Participants with a urodynamic diagnosis of genuine stress incontinence were given the choice of enrolling in the behavioral arm or surgical arm of the study. Those in the behavioral trial were randomized to receive bladder retraining, pelvic floor muscle exercises, or a combination of these treatments. Patients with genuine stress incontinence who enrolled in the surgical arm of the study received a Burch retropubic colposuspension. Participants with prolapse were enrolled in a randomized surgical trial. None of the patients with prolapse had preoperative genuine stress incontinence, and none of the patients with urinary incontinence had prolapse beyond the hymen. Twenty‐one percent (6 of 29) of women with prolapse also had a urodynamic diagnosis of detrusor instability. All women with prolapse underwent vaginal reconstructive surgery and were randomized to undergo plication of the urethrovesical junction endopelvic fascia or abdominal‐vaginal needle colposuspension.
In the clinical trials, randomization was performed in a blocked fashion (randomly selected block sizes of 4 to 6) by using sealed, opaque envelopes. Details of the methods, concurrent surgical procedures, and results of these trials are published elsewhere.11–15
Before therapy, all participants completed two validated condition‐specific quality‐of‐life questionnaires, the Urogenital Distress Inventory and the Incontinence Impact Questionnaire‐Revised,16 and a condition‐specific sexual function questionnaire. All participants except those enrolled in the estrogen trial were asked to repeat the sexual function questionnaire and the quality‐of‐life scales at their 3‐ and 6‐month follow‐up visits. Participants who received estrogen therapy were followed for only 3 months and repeated the three questionnaires at the end of therapy.
The condition‐specific sexual function questionnaire consisted of nine questions. The first two questions asked participants if they were in an intimate relationship and if they were sexually active. If these two questions were answered “yes,” participants were asked to complete the seven remaining questions. If participants were not currently sexually active, they were asked to skip the final seven questions but were asked to indicate the reason for their sexual inactivity. One item in the Incontinence Impact Questionnaire was also included in this report. It asked respondents whether urine leakage or prolapse had affected their ability to have sexual relations. Participants answered this question regardless of whether they were currently sexually active. “Sexual activity” was not specifically defined but was left to the interpretation of the participants.
Categorical data were analyzed for statistical significance by using the χ2 or Fisher exact test, as appropriate. Continuous variables were analyzed by performing analysis of variance. The results of the baseline sexual function questionnaire of each diagnostic group (genuine stress incontinence, detrusor instability, mixed urinary incontinence, and pelvic organ prolapse) were compared by using the Kruskal–Wallis test. The results of the sexual function questionnaire at the 3‐ and 6‐month follow‐up visits were compared with baseline findings by using Friedman two‐way analysis of variance. Pairwise comparisons were performed using Bonferroni correction and the Wilcoxon signed‐rank test, Wilcoxon rank‐sum test, McNemar test, χ2 test, or Student t test, as appropriate. Logistic regression analysis was used to adjust for covariates and to identify variables that were independently associated with overall sexual satisfaction and the degree to which urinary incontinence or pelvic organ prolapse affected a participant's ability to have sexual relations. Covariates included in the analysis were age, race, body mass index, education, income, previous hysterectomy, and previous urogynecologic surgery. The 309 participants who had data recorded for all of these covariates were included in the multivariable analysis. Statistical analysis was performed by using SAS software, 6.12 (SAS Institute Inc., Cary, NC).
Table 1 shows baseline characteristics of the participants, stratified by urodynamic diagnosis. Participants with genuine stress incontinence were similar in age to those with pelvic organ prolapse but were significantly younger than those with detrusor instability or mixed incontinence. Participants with prolapse or genuine stress incontinence were more likely to be estrogenized at baseline than were participants with mixed incontinence or detrusor instability. Among women with urinary incontinence, those with genuine stress incontinence had lower scores (less impact) on the Incontinence Impact Questionnaire and less urine loss on a quantitative pad test at baseline than did those with detrusor instability or mixed incontinence. However, the number of incontinence episodes per week and pads used per week recorded in a 1‐week bladder diary and the results of the baseline Urinary Distress Inventory were similar.
Ninety‐eight percent of patients completed the sexual function questionnaire at baseline. Eighty‐nine percent and 76% of participants completed the sexual function questionnaire at 3 and 6 months of follow‐up, respectively. Nonresponders were significantly younger (59 versus 62 years, P < .01) and more likely to report an annual income less than $20,000 (P < .05) than were responders. Other demographic characteristics were similar, however. Neither baseline diagnosis nor therapeutic group significantly affected the response rate.
At baseline, 67% of participants (231 of 343) reported that they were married or in an intimate relationship. Patients with detrusor instability were less likely than patients in other groups to be married or in an intimate relationship (Table 2). Of patients who were involved in an intimate relationship, 70% (161 of 231) reported that they were currently sexually active. The rate of sexual activity was similar among diagnostic groups. Of participants who reported that they were married or in an intimate relationship but not sexually active, the most commonly cited reason for sexual inactivity was a physical ailment of the participant's partner that made sexual relations embarrassing or uncomfortable (44 of 70 participants [63%]). Twenty percent (14 of 70) indicated that the reason for their sexual inactivity was embarrassment or discomfort, resulting from urinary incontinence or pelvic organ prolapse, on the part of themselves or their partner. Patients with prolapse or detrusor instability were significantly more likely to be sexually inactive because of their pelvic floor symptoms than were patients with genuine stress incontinence or mixed incontinence. Other reasons cited for sexual inactivity included “I am too tired” (4 of 70 [5%]), “My partner is too tired” (8 of 70 [11%]), “I am not interested” (17 of 70 [24%]), “My partner is not interested” (16 of 70 [22%]), “I have a physical illness other than urine leakage or prolapse that makes sexual relations difficult or uncomfortable” (8 of 70 [11%]), and “other” (6 of 70 [8%]). Forty‐eight percent (34 of 70) of sexually inactive participants indicated more than one reason for their sexual inactivity.
Table 3 shows results of the baseline sexual function questionnaire. Overall, 81% (131 of 161) of sexually active participants reported that sexual relations were “somewhat” or “very” satisfactory. Baseline diagnosis did not significantly affect the response to this question. Twenty‐two percent (25 of 108) of sexually active participants with genuine stress incontinence were “moderately” to “extremely” worried that sexual activity would cause urine loss, significantly more than any other group. One third of participants (7 of 21) with pelvic organ prolapse felt that prolapse or urine leakage affected their ability to have sexual relations, compared with 10% of participants (28 of 267) with urinary incontinence (P = .02). However, the responses to all other sexual function questions, including questions about sexual frequency, vaginal dryness, and dyspareunia, did not significantly differ among diagnostic groups at baseline.
After using multivariable analysis to control for the demographic characteristics in Table 1, both age (odds ratio [OR] 1.13; 95% confidence interval [CI] 1.1, 1.2) and a baseline diagnosis of pelvic organ prolapse (OR 2.3; 95% CI 1.4, 3.9) were found to be independently associated with the degree to which urinary incontinence or prolapse affected a participant's ability to have sexual relations. No patient characteristics or diagnoses were found to be significantly associated with the response to the question about overall sexual satisfaction.
Fewer patients with prolapse reported that their prolapse affected their ability to have sexual relations “moderately” or “greatly” 6 months after vaginal reconstructive surgery (0 of 18) than at baseline (7 of 21 [33%], P < .01). Among participants with genuine stress incontinence who were enrolled in the surgical trial and were sexually active (n = 21), the proportion who were “moderately” to “extremely” worried about urine loss during intercourse decreased significantly from 29% (6 of 21) at baseline to 9% (1 of 11) 6 months after retropubic urethropexy (P = .02). Responses to all other questions on the sexual function questionnaire did not significantly differ from baseline to 3 or 6 months after surgery in the group with prolapse or the group with genuine stress incontinence who were managed surgically. In addition, the number of patients who were sexually active did not change after surgery in either surgical group.
Among sexually active participants with genuine stress incontinence (n = 77) who were enrolled in the behavioral therapy trial, the proportion of those who were “moderately” to “extremely” worried about urine loss during sexual intercourse decreased significantly from baseline to 3 and 6 months after initiation of therapy (17 of 77 [22%] versus 7 of 66 [11%] and 7 of 66 [11%], respectively; P = .001). Sexually active participants with detrusor instability and mixed incontinence who were enrolled in the behavioral trial (n = 10 and 16, respectively), were significantly less likely than participants with stress incontinence to worry about urine loss during intercourse at baseline (0% and 10%, respectively); responses to this question did not change significantly after therapy. The responses to all other questions about sexual function were unchanged after behavioral therapy in all three groups with urinary incontinence. Similarly, estrogen therapy had no effect on any of the items on the sexual function questionnaire, including the questions about vaginal dryness and dyspareunia.
The perception of sexual satisfaction depends on many complex interactions. Emotional well‐being, intimacy with one's partner, quality of life, and physical health are some contributing factors. In this prospective study of older women with urinary incontinence and pelvic organ prolapse, we found that satisfaction with sexual relations was high at baseline and relatively unaffected by diagnosis or therapy. These findings are similar to those reported by Weber et al,5 who found that 84% of their patients with urinary incontinence or pelvic organ prolapse reported satisfaction with sexual relations. In that study, women with prolapse and urinary incontinence did not differ from continent women without prolapse in measures of sexual function. They concluded that age was the most important predictor of sexual satisfaction. Age was not an independent predictor of sexual satisfaction in our study, but this may be explained by the older average age of our patients. Of note, the rate of sexual activity and sexual satisfaction in our study was similar to those in other studies of women of similar age in the general population.17
Although urinary incontinence and prolapse did not seem to significantly influence sexual satisfaction in our study, pelvic floor symptoms, particularly those of prolapse, affected sexual relations. Of the patients with stage III or IV pelvic organ prolapse who were sexually active, more than one third felt that their prolapse affected their ability to have sexual relations. Similar results were reported by Weber et al, who found that increasing grade of prolapse predicted interference with sexual activity but did not affect frequency of intercourse or sexual satisfaction.5 Approximately one fifth of participants in our study cited urinary incontinence or prolapse as a reason for sexual inactivity. Participants with prolapse were more likely than those with urinary incontinence to cite their condition as a reason for sexual inactivity. Surgical correction of prolapse, particularly posterior colporrhaphy, has been associated with worsening sexual function and increased dyspareunia.18,19 A recent study found that the combination of Burch urethropexy and posterior colporrhaphy was especially likely to result in dyspareunia.19 In contrast, Haase and Skibsted reported that sexual relations improved after reconstructive surgery in 24% of women with urinary incontinence or genital descensus and deteriorated in only 5%.2 In our study, vaginal reconstructive surgery eliminated the perception that sexual function was affected by pelvic organ prolapse. Sexual satisfaction remained high, and symptoms of dyspareunia were unchanged from baseline to 6 months after surgery.
Many investigators have reported an association between sexual dysfunction and urinary incontinence; rates of dysfunction are as high as 26% to 43% in this condition.3,4,20,21 Data from studies comparing the sexual function of women with genuine stress incontinence, detrusor instability, and mixed incontinence are conflicting, however. Sutherst and Brown found that women with detrusor instability had a significantly higher prevalence of sexual dysfunction than did women with genuine stress incontinence,3 whereas Walters et al found that women with genuine stress incontinence felt that their urinary tract symptoms interfered with sex more than did women with detrusor instability.4 One study compared the prevalence of urinary leakage during intercourse in women with genuine stress incontinence, detrusor instability, and mixed incontinence and found no difference between groups, whereas others have shown a clear predominance of one diagnosis over the others.21,22 Our data suggest that urinary leakage during intercourse is more common in women with genuine stress incontinence and that effective therapies for this condition, such as surgical or behavioral therapy, reduce leakage during intercourse. Because estrogen therapy has been shown to be an ineffective treatment for genuine stress incontinence,11 it is not surprising that it would also be ineffective in reducing urine leakage during intercourse. The proportion of participants with detrusor instability and mixed incontinence who worried about urinary leakage during intercourse was small in our study. As a result, there were too few patients to demonstrate a potential treatment effect with behavioral or estrogen therapy in these groups.
An interesting finding in our study was that detrusor instability was significantly more likely than stress incontinence or mixed incontinence to be cited as a cause of sexual inactivity. The reason for this is unclear, especially considering that urine leakage during intercourse was less common and the rate of dyspareunia was similar to that in the other two groups. It is possible that symptoms other than urinary leakage, such as urinary urgency and frequency, contribute to sexual inactivity in these women.
Our study had several limitations. First, there was no control group. As a result, we cannot make causal inferences about sexual function and urinary incontinence or prolapse. Second, participants with urinary incontinence were stratified into treatment groups rather than randomized. Randomization occurred only after assignment to hormonal, behavioral, or surgical therapy. In the absence of randomization, the treatment groups were not identical. Although known confounders were adjusted for by using multivariable analysis, unknown (or unmeasured) confounders may still exist. Third, sexual function was not a primary outcome of the Continence Program for Women. Consequently, the sample sizes for each diagnostic and therapeutic group were calculated to detect important differences in the primary end points rather than differences in the responses to the sexual function questionnaire. In addition, many of the questions on the sexual function questionnaire were intended to be answered only by participants who were sexually active, reducing the sample size of some questions by up to 50%. As a result, our study may not have enough power to identify some important differences in sexual function. This is particularly true for the prolapse trial, which had relatively few patients in it. However, this lack of power does not diminish the validity of findings for which statistically significant differences were demonstrated. Finally, the reliability, validity, and internal consistency of our sexual function questionnaire are unknown. However, a modified version of this questionnaire studied by our group demonstrated good psychometric properties (unpublished data). Despite these limitations, our study is useful because it provides a prospective evaluation of sexual function in a well‐defined group of women with urinary incontinence and pelvic organ prolapse in which standardized outcomes before and after three different therapies were used.
Sexual dysfunction has many causes, and disorders of the female pelvic floor, such as urinary incontinence and pelvic organ prolapse, can influence sexual function and satisfaction. However, as we and Weber et al5 found, general assessments of sexual satisfaction do not demonstrate a significant effect of pelvic floor dysfunction on sexual function in most women. However, specific questions about the effect of prolapse on sexual function and about urinary leakage during intercourse demonstrated differences between groups with various diagnoses of pelvic floor dysfunction. A general assessment of sexual function may not fully capture the effect of pelvic floor function on sexual satisfaction. Questions specifically designed to elucidate the complex interaction between disorders of the pelvic floor and sexual function are probably necessary. Condition‐specific sexual function questionnaires should be developed and widely adopted for use in epidemiologic surveys and clinical trials of pelvic floor dysfunction so that we may understand fully the effect of such conditions as urinary incontinence and pelvic organ prolapse, and their treatment, on sexual function.
MEMBERS OF THE CONTINENCE PROGRAM FOR WOMEN RESEARCH GROUP
J. Andrew Fantl, Jean F. Wyman, Richard C. Bump, Donna McClish, and Denise M. Elser, Virginia Commonwealth University, Richmond, Virginia; Curt D. Furberg, Samuel F. Lentz, Timothy M. Morgan, Deirdre Robinson, Sally A. Shumaker, and John S. Uebersax, Bowman Gray School of Medicine, Wake Forest University, Winston‐Salem, North Carolina; and Richard C. Bump and James P. Theofrastous, Duke University, Durham, North Carolina.
1. Field SM, Hilton P. The prevalence of sexual problems in women attending for urodynamic investigation. Int Urogynecol J 1993;4:212–5.
2. Haase P, Skibsted L. Influence of operations for stress incontinence and/or genital descensus on sexual life. Acta Obstet Gynecol Scand 1988;67:659–61.
3. Sutherst J, Brown M. Sexual dysfunction associated with urinary incontinence. Urol Int 1980;35:414–6.
4. Walters MD, Taylor S, Schoenfeld LS. Psychosexual study of women with detrusor instability. Obstet Gynecol 1990;75:22–6.
5. Weber AM, Walters MD, Schover LR, Mitchinson A. Sexual function in women with uterovaginal prolapse and urinary incontinence. Obstet Gynecol 1995;85:483–7.
6. Wyman JF, Choi SC, Harkins SW, Wilson MS, Fantl JA. The urinary diary in evaluation of incontinent women: A test-retest analysis. Obstet Gynecol 1988;71:812–7.
7. Fantl JA, Harkins SW, Wyman JF, Choi SC, Taylor JR. Fluid loss quantitation test in women with urinary incontinence: A test-retest analysis. Obstet Gynecol 1987;70:739–43.
8. Bump RC, Copeland WE Jr, Hurt WG, Fantl JA. Dynamic urethral pressure/profilometry pressure transmission ratio determinations in stress-incontinent and stress-continent subjects. Am J Obstet Gynecol 1988;159:749–55.
9. Abrams P, Blaivas JG, Stanton SL, Andersen JT. The standardisation of terminology of lower urinary tract function. The International Continence Society Committee on Standardisation of Terminology. Scand J Urol Nephrol Suppl 1988;114:5–19.
10. Bump RC, Mattiasson A, Bo K, Brubaker LP, De Lancey JO, Klarskov P, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7.
11. Fantl JA, Bump RC, Robinson D, McClish DK, Wyman JF. Efficacy of estrogen supplementation in the treatment of urinary incontinence. The Continence Program for Women Research Group. Obstet Gynecol 1996;88:745–9.
12. Bump RC, Hurt WG, Theofrastous JP, Addison WA, Fantl JA, Wyman JF, et al. Randomized prospective comparison of needle colposuspension versus endopelvic fascia plication for potential stress incontinence prophylaxis in women undergoing vaginal reconstruction for stage III or IV pelvic organ prolapse. The Continence Program for Women Research Group. Am J Obstet Gynecol 1996;175:326–33.
13. Wyman JF, Fantl JA, McClish DK, Bump RC. Comparative efficacy of behavioral interventions in the management of female urinary incontinence. The Continence Program for Women Research Group. Am J Obstet Gynecol 1998;179:999–1007.
14. Bump RC, Hurt WG, Addison WA, Fantl JA, McClish DK. Reliability and correlation of measurements during and after bladder neck surgery. The Continence Program for Women Research Group. Br J Urol 1998;82:628–33.
15. Bump RC, Hurt WG, Elser DM, Theofrastous JP, Addison WA, Fantl JA, et al. Understanding lower urinary tract function in women soon after bladder neck surgery. The Continence Program for Women Research Group. Neurourol Urodyn 1999;18:629–37.
16. Shumaker SA, Wyman JF, Uebersax JS, McClish D, Fantl JA. Health-related quality of life measures for women with urinary incontinence: The Incontinence Impact Questionnaire and the Urogenital Distress Inventory. Continence Program in Women (CPW) Research Group. Qual Life Res 1994;3:291–306.
17. Johannes CB, Avis NE. Gender differences in sexual activity among mid-aged adults in Massachusetts. Maturitas 1997;26:175–84.
18. Francis WJA, Jeffcoate TNA. Dyspareunia following vaginal operations. J Obstet Gynaecol British Commonwealth 1961;68:1–10.
19. Weber AM, Walters MD, Piedmonte MR. Sexual function and vaginal anatomy in women before and after surgery for pelvic organ prolapse and urinary incontinence. Am J Obstet Gynecol 2000;182:1610–5.
20. Iosif CS. Sexual function after colpo-urethrocystopexy in middle-aged women. Urol Int 1988;43:231–3.
21. Hilton P. Urinary incontinence during sexual intercourse: A common, but rarely volunteered, symptom. Br J Obstet Gynaecol 1988;95:377–81.
22. Clark A, Romm J. Effect of urinary incontinence on sexual activity in women. J Reprod Med 1993;38:679–83.