Urinary incontinence is a common problem in adult women, affecting as much as 28–55% of the population.1 Female sexual dysfunction is shown to have a direct relation to the severity of stress urinary incontinence (SUI).2 Patients often cite numerous reasons for urinary incontinence contributing to sexual dissatisfaction, including embarrassment, psychological distress, fear of coital incontinence, loss of self-esteem, and dyspareunia.3
There is a growing body of literature concerning female sexual function after treatment for urinary incontinence. Pelvic floor muscle therapy has been shown to improve sexual function as well as urinary incontinence symptoms.4,5 Surgical treatment, on the other hand, has had unclear effects on sexual function. Several studies have shown an improvement in sexual function after surgical correction.6–8 Though even in studies that show an overall improvement in sexual function, de novo dyspareunia may occur.8 A systematic review and meta-analysis of surgery for SUI showed that coital incontinence improves after intervention, but the authors did not observe an improvement or deterioration in overall female sexual function.9 The relationship between surgical treatment and improvement in sexual function has yet to be fully elucidated.
This study evaluates four of the standard surgical interventions for SUI (Burch colposuspension, autologous fascial sling, retropubic and transobturator polypropylene midurethral slings) and their relationship to postoperative sexual function. Each of these methods is shown to be highly effective for SUI treatment, with the autologous fascial pubovaginal sling and the retropubic midurethral sling being slightly more effective in each of the individual studies from this secondary analysis.10,11 Although the four surgical approaches are all effective for resolving symptoms of SUI, there may be important differences between them regarding their effect on sexual function. Our primary aim was to determine whether there was a difference in 24-month postoperative sexual function between treatment groups. Secondarily, we sought to explore the effects of subjective and objective surgical cure rates and concomitant surgical procedures on 24-month sexual function. We hypothesized that women with SUI who were undergoing surgical treatment would experience an overall improvement in sexual function from baseline to 24 months and that there would be no significant difference between surgical treatment options.
This is a secondary analysis of the SISTEr (Stress Incontinence Surgical Treatment Efficacy Trial) and TOMUS (Trial of Mid-Urethral Slings). Both were multicenter randomized trials that compared Burch colposuspension to autologous fascial sling and retropubic to transobturator midurethral slings. The SISTEr study was conducted from February 2002 through June 2006. The TOMUS study was conducted from April 2006 through June 2010. The data were obtained by accessing the Central Repository for the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). This study was classified as exempt by the Virginia Commonwealth University Institutional Review Board (IRB HM20012510) because these are both publicly available data sets.
The specifics of the designs and inclusion and exclusion criteria for both trials have been published previously.12,13 In brief, the inclusion criteria for the original studies included adult women older than age 21 years who had stress-predominant urinary incontinence for at least 3 months and were planning on surgical intervention for SUI. We included women who reported they were sexually active in the preceding 6 months at any of the study visits (baseline study visit, 12 months or 24 months postoperative study visit [Fig. 1]). To be included in the analysis, patients needed to report sexual activity at any of the study visits and complete a short form of the PISQ-12 (Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire) questionnaire at one of the included study visits.
The primary aim of this study was to assess the effect of surgical intervention for SUI on 2-year postoperative condition specific sexual function and to determine whether type of anti-incontinence surgical procedure affects postoperative sexual function. In both trials, female sexual function information was collected at study visits using the short version of the PISQ-12 PISQ-12.14 Each of the 12 items in PISQ-12 has a score range of 0–4, with higher scores indicating better sexual function. The PISQ-12 scores of each patient was obtained at baseline, 12-month postoperative visit, and 24-month postoperative visit. Only participants who had been sexually active in the 6 months before the visit were eligible to answer the PISQ-12. The PISQ-12 scores were compared at each timepoint between treatment groups. There is no minimum important difference established for the PISQ-12; thus, we used one half the baseline standard deviation to determine clinical significance.15 Secondary aims of the study were to evaluate whether subjective failure rate, objective failure rate, or other concomitant procedures at the time of anti-incontinence surgery were related to postoperative sexual function.
Variables collected include baseline characteristics and sociodemographic factors (age, body mass index, race–ethnicity, marital or partner status, pregnancy history, smoking status, menopausal status, prior hormone therapy, prior incontinence treatment including prior surgical treatment, prolapse severity, and concomitant surgery at time of procedure), surgical factors (type of procedure performed; subjective failure, defined by negative results from Medical, Epidemiological and Social Aspects of Aging Questionnaire; 3-day void diary; and no retreatment for stress incontinence [behavioral, pharmacologic, or surgical]), and objective failure, defined by negative results from both provocative stress test and 24-hour pad test as well as having received no retreatment for stress incontinence.12,13 Sexual function data included total PISQ-12 score at baseline,12 and 24 month postoperative study visits, and sexual domain specific responses (behavioral and emotive, physical, partner).14
Differences in baseline characteristics of the women across treatments were assessed with Pearson χ2 or one-way analysis of variance. To account for excluding women without PISQ-12 data, we evaluated the rates of sexual activity for patients within each treatment group and the change therein across treatment groups using cross classification with treatments. A linear mixed effect model was used to predict the PISQ-12 scores using time (baseline, 12 month follow-up, 24 month follow-up), treatment group, and other baseline or treatment variables related to the outcome. This model both allows for the assessment of predictors of continuous outcome and is able to account for longitudinal measurements. A parsimonious set of baseline and treatment variables were obtained from a backward selection procedure that always included time, treatment, and the interaction of treatment and time. A first-order auto-regressive error structure was assumed. Because the final model indicated no interaction, mean changes between timepoints were calculated averaging over the treatment groups.
Generalized linear regression models were then created to evaluate the 24-month PISQ-12 score as a factor of subjective and objective failure rates of the surgical procedure as well as a factor of a concomitant surgery. These models adjusted for baseline PISQ-12 score, as well as any baseline differences between treatment groups identified. A one-way analysis of variance was then used to evaluate differences in individual responses to the PISQ-12 questions as well as the sexual domains (behavioral or emotive, physical, partner) between treatment groups.
All statistical analyses were performed in SAS 9.4 or JMP Pro 14.0.0, and graphical results were generated using R. An a priori alpha of 0.05 was used to determine statistical significance.
Nine hundred twenty-four women met inclusion criteria (Fig. 1). Before proceeding with analysis of the included patients, an additional analysis was performed of women who were sexually active and inactive; there was no difference or significant change in the number of sexually active women across the four treatment groups (baseline visit P=.85, 12-month postoperative P=.56, 24-month postoperative P=.49) (Table 1). Of the 1,252 patients from the original studies, 924 were sexually active and completed PISQ-12 questionnaires at either baseline or at 12 month or 24 month follow up visits, and were included in the final analyses (Fig. 1).
Of the 924 total women included in final analyses: 249 (26.9%) had an autologous fascial sling, 239 (25.9%) underwent Burch colposuspension, 216 (23.3%) had a retropubic midurethral sling placed and 220 (23.8%) had transobturator midurethral sling placed. Baseline characteristics among all four treatment groups were similar overall (Table 2). Exceptions to this were differences observed in race–ethnic group (P<.01), the number of vaginal deliveries (P=.04), prolapse stage (P<.01), and concomitant prolapse repair surgery (P<.01).
There were no significant differences in mean PISQ-12 scores between the four treatment groups at the time of baseline (P=.07), the 12 month visit (P=.42), and the 24-month visit (P=.50). Patients in the two studies exhibited an overall improvement in sexual function over the 24-month study period (Fig. 2 and Table 3). Specifically, sexual function scores were higher at 12 month and 24 month follow-up time periods compared with baseline after adjusting for baseline differences (baseline PISQ-12: 32.6, 33.1, 31.9, 31.4; 12-month PISQ-12: 37.7, 37.8, 36.9, 37.1; 24-month PISQ-12: 37.7, 37.8, 36.9, 37.1). The majority of the improvement in sexual function was observed at the first 12 months after surgery visit and was maintained over the second postoperative year. There was no significant difference in mean PISQ-12 scores between the 12-month follow-up and the 24-month follow-up (P=.97). There is no published minimum important difference for the PISQ-12, so we conservatively estimated the minimum important difference as half the standard deviation of the baseline score.15 The SD of PISQ-12 at baseline is 7.08 for all surgical treatment groups. The mean scores for each treatment improved beyond this estimate for minimum important difference regardless of treatment type (Table 3).
We found that postoperative objective failure and subjective failure are significantly associated with lower PISQ-12 scores in a generalized linear regression, controlling for baseline differences (P<.01, P<.01). Concomitant prolapse repair surgery is significantly associated with higher PISQ-12 scores when controlling for baseline characteristic differences (P<.01).
We examined responses to all questions in the PISQ-12 questionnaire. The mean score for individual questions by surgical treatment group are shown at all timepoints in Table 4. Interestingly, statistically significant differences were observed for pain (question 4) and negative emotional reactions (question 9) with sexual intercourse at baseline that were no longer seen at 12 or 24 months postoperatively. Similarly, fear of incontinence at baseline and 12 months (question 7) and avoidance of sexual intercourse because of bulging in to the vagina (question 8) at baseline and 24 months were different between groups (Table 4). In terms of domains, at baseline the physical and partner domains showed significant differences between treatment groups (Table 5), but these differences were no longer detected postoperatively and no differences were noted in the behavioral or emotive domain at any timepoint.
Although the differences in sexual function domains and individual condition-specific questions show statistically significant differences, they are all less than 0.4 points different in mean scores between surgical treatment groups, which is lower than the baseline standard deviation for all questions and domains, and therefore unlikely clinically meaningful (Tables 4 and 5). The greatest improvement from baseline to postoperatively in sexual function postoperatively was seen in the specific questions 6 and 7 “are you incontinent of urine with sexual activity” or “does fear of incontinence restrict your sexual activity?” and was noted in the physical domain, although improvements in all three domains were apparent in these data.
The primary aim of this study was to identify the effect of surgical intervention for SUI on 2-year postoperative condition-specific sexual function and to determine whether type of anti-incontinence surgical procedure affects postoperative sexual function. There is an overall clinically meaningful improvement in sexual function 2 years after anti-incontinence surgery, independent of the type of procedure performed. This was true for overall sexual function as well as condition-specific questions regarding recurrent incontinence, prolapse, pain or sexual function domains (physical, behavioral or emotive, or partner). This supported our hypothesis. Interestingly, we did not find a significant difference in sexual function between 12 and 24 months, which suggests that most of the improvements in female sexual function as measured by PISQ-12 had occurred within the first 12-month period after treatment for SUI. This is a particularly salient point for care of and counseling patients who are undergoing anti-incontinence procedures that the improvements in sexual function initially seen do not significantly decrease over time.
This study is strengthened by the data obtained from the original SISTEr and TOMUS trials' being validated, condition-specific, and in a large randomized population. The PISQ-12 questionnaire is a strength of this study because it addresses multiple domains of sexual function for patients; the differentiation of the domains helps demonstrate that the greatest improvement in sexual function postoperatively derives from the physical domain in sexual function as it relates to coital incontinence or fear of coital incontinence. The study is also strengthened by the number of patients with responses out to 24 months, which is reflective of more stable, long-term quality of life metrics. However, the generalizability of this study is limited by the low degree of diversity among women in these trials.
Although the PISQ-12 is an excellent tool for measuring sexual function in our study population, there are limitations to its use. It does not address sexual stimulation or nonpenetrative vaginal intercourse. Additionally, it limits partner-related problems to erectile dysfunction and premature ejaculation; some eligible participants may be excluded secondary to sexual preferences given the assumptions inherent to the questionnaire that the partner is male. This does limit our ability to evaluate important aspects of sexual function for our patients such as self-stimulation, same sex partners and sexual activity beyond vaginal penetration.
This was a secondary analysis comparing four surgical intervention groups; thus, we do not have a nonsurgical treatment or control group for comparison. Because we excluded participants who did not complete PISQ-12 survey, we limited the opportunity to allow those patients to provide a free-text response for why they were not sexually active, which may pertain to their SUI symptoms or may be completely unrelated. Though in our study there were no differences in sexual activity compared with inactivity between treatment groups. Additionally, the study is limited by the decreasing response rates and loss to follow up for patients over the 24 months study period. This has a potential to bias our results and overestimate improvement in sexual function over the study period.
These data contribute to the growing body of literature of postoperative sexual function after four commonly performed anti-incontinence procedures. This study and others demonstrate that sexual function improves with surgical improvement of stress incontinence which may suggest a possible association of urinary incontinence and sexual dysfunction.3,16 As we continue to explore the complex and multi-faceted problem of sexual dysfunction, further evaluation of the effect of pelvic floor disorders—and their treatments—will be important and necessary research.
1. Melville JL, Katon W, Delaney K, Newton K. Urinary incontinence in US women: a population-based study. Arch Intern Med 2005;165:537–42.
2. Stadnicka G, Łepecka-Klusek C, Pilewska-Kozak A, Jakiel G. Psychosocial problems of women with stress urinary incontinence. Ann Agric Environ Med 2015;22:499–503.
3. Duralde ER, Rowen TS. Urinary incontinence and associated female sexual dysfunction. Sex Med Rev 2017;5:470–85.
4. Bø K, Talseth T, Vinsnes A. Randomized controlled trial on the effect of pelvic floor muscle training on quality of life and sexual problems in genuine stress incontinent women. Acta Obstet Gynecol Scand 2000;79:598–603.
5. Serati M, Braga A, Dedda MCD, Sorice P, Peano E, Biroli A, et al. Benefit of pelvic floor muscle therapy in improving sexual function in women with stress urinary incontinence: a pretest-posttest intervention study. J Sex Marital Ther 2015;41:254–61.
6. Lemack GE, Zimmern PE. Sexual function after vaginal surgery for stress incontinence: results of a mailed questionnaire. Urology 2000;56:223–7.
7. Glavind K, Larsen T, Lindquist AS. Sexual function in women before and after tension-free vaginal tape operation for stress urinary incontinence. Acta Obstet Gynecol Scand 2014;93:986–90.
8. Mengerink BB, Leijsen SAV, Vierhout ME, Inthout J, Mol BW, Milani AL, et al. The impact of midurethral sling surgery on sexual activity and function in women with stress urinary incontinence. J Sex Med 2016;13:1498–507.
9. Jha S, Ammenbal M, Metwally M. Impact of incontinence surgery on sexual function: a systematic review and meta-analysis. J Sex Med 2012;9:34–43.
10. Albo ME, Richter HE, Brubaker L, Norton P, Kraus SR, Zimmern PE, et al. Burch colposuspension versus fascial sling to reduce urinary stress incontinence. N Engl J Med 2007;356:2143–55.
11. Richter HE, Albo ME, Zyczynski HM, Kenton K, Norton PA, Sirls LT, et al. Retropubic versus transobturator midurethral slings for stress incontinence. N Engl J Med 2010;362:2066–76.
12. Tennstedt S; Urinary Incontinence Treatment Network. Design of the stress incontinence surgical treatment efficacy trial (SISTEr). Urology 2005;66:1213–7.
13. Urinary Incontinence Treatment Network (UITN). The trial of mid-urethral slings (TOMUS): design and methodology. J Appl Res 2008;8:AlboVol8No1.
14. Rogers RG, Coates KW, Kammerer-Doak D, Khalsa S, Qualls C. A short form of the Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire (PISQ-12). Int Urogynecol J Pelvic Floor Dysfunct 2003;14:164–8.
15. Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation. Med Care 2003;41:582–92.
16. Morgan DM, Dunn RL, Stoffel JT, Fenner DE, DeLancey JO, McGuire EJ, et al. Are persistent or recurrent symptoms of urinary incontinence after surgery associated with adverse effects on sexual activity or function? Int Urogynecol J Pelvic Floor Dysfunct 2008;19:509–15.