Pelvic floor relaxation is a common condition among postmenopausal women and a major indication for gynecologic surgery. The lifetime risk of undergoing a single operation for pelvic organ prolapse or urinary incontinence by 80 years of age was approximately 11% in one North American study.1 For women who were 50–59 years and 60–69 years of age, the annual procedure rate for surgically managed prolapse and incontinence was 3.30 and 5.24 per 1000 woman-years, respectively.1 Although surgically managed pelvic floor relaxation represents only a subset of symptomatic patients, the costs associated with the procedure, as well as trauma to the patient, suggests that this is a substantial medical concern for older women.
The pathogenesis and etiology of pelvic floor relaxation is not well understood. However, a number of factors have been associated with this condition, including age, parity, smoking, obesity, and chronic respiratory disease.2 Damage to the connective tissues and neuromuscular support of the pelvic floor is thought to result from trauma at childbirth and increased intra-abdominal pressure resulting from obesity and chronic cough.3,4 Connective tissues may also be weakened during the aging process as a result of decreases in collagen content.5 Lumbar spine lordosis was recently reported as a potential risk factor for pelvic floor relaxation, but this requires confirmation.6
The role of hormonal status in the development of pelvic floor relaxation is also not clear. Low estrogen levels generally associated with the postmenopausal state have been considered to be a causal factor for atrophy of the pelvic floor and the subsequent increased incidence of pelvic floor relaxation in older women.7 Estrogen replacement therapy has been traditionally considered to have positive effects on pelvic floor relaxation, although no randomized controlled trials have validated this perception.8 Understanding the exact role of hormonal status in the pathophysiology of pelvic floor relaxation is further complicated by recent findings of increased pelvic organ prolapse with two selective estrogen receptor modulators (SERMs), levormeloxifene and idoxifene, which were previously in clinical development for the treatment and prevention of postmenopausal osteoporosis.9
Raloxifene is a benzothiophene SERM approved in the United States and Europe for the treatment and prevention of postmenopausal osteoporosis. Large-scale randomized placebo-controlled trials have demonstrated that raloxifene preserves bone mineral density, reduces the risk for fractures and for invasive breast cancer, and improves cardiovascular risk factors in postmenopausal women with osteoporosis.10–14 Several clinical trials have demonstrated a neutral or antiestrogenic effect of raloxifene on the endometrium and uterus.12,15–17 Overall, the effects of raloxifene on the vagina have been neutral; relative to placebo, raloxifene does not increase the incidence of events related to vaginal atrophy.18
Because of reports of pelvic floor relaxation and uterine prolapse associated with two other SERMs, we investigated whether raloxifene had a similar effect on the genitourinary system of postmenopausal women, specifically pelvic floor relaxation. We analyzed pooled data from three randomized, placebo-controlled osteoporosis prevention and treatment trials published previously10,11,19 and report here the effects of up to 3 years of raloxifene therapy on the frequency of surgery for pelvic floor relaxation.
MATERIALS AND METHODS
The current analysis comprises integrated data from three randomized, double-masked, placebo-controlled clinical trials of raloxifene (Table 1). Studies 1 and 2 were identical osteoporosis prevention trials of younger postmenopausal women enrolled at 20 sites in Western Europe and in North America.10,19 Study 3 was an osteoporosis treatment trial of older postmenopausal women enrolled at 180 sites in 25 countries.11 Women were excluded from the studies if during the previous 6 months they had taken systemic estrogen, progestins, androgens, or systemic corticosteroids. Women with unexplained vaginal bleeding at initial evaluation were also excluded. Prior hysterectomy was allowed in all three studies.
Women were assigned equally by randomized block design to receive placebo, raloxifene 30, 60, or 150 mg per day in Studies 1 and 2 or to placebo, raloxifene 60 or 120 mg per day in Study 3. Study medication was packaged in numbered kits using a random number table; kits were assigned in sequence at each study site, beginning with the lowest number available. Neither the investigator nor the subject was aware of treatment assignment. Study medication and placebo were formulated in tablets identical in appearance, taste, and odor. Further design details and results have been published previously.10,11,19
Because endometrial safety was a secondary endpoint in these studies, routine transvaginal ultrasonography was scheduled for all women enrolled in osteoporosis prevention trials and for 2155 women enrolled at designated centers in the osteoporosis treatment trial. In addition, some centers in the osteoporosis treatment trial elected to perform transvaginal ultrasonography in a subset of their participants. Investigators were required to report adverse events, including all surgical procedures, to the study sponsor. Surgical events were retrieved from the clinical trial safety database, and indications for any reported pelvic surgeries were reviewed by sponsor personnel blinded to treatment group. Procedures performed for pelvic organ prolapse or urinary incontinence, including hysterectomy, Burch retropubic bladder neck suspension, Marshall-Marchetti, and colpocystorrhaphy, were included in this analysis. Patients with a prior history of hysterectomy were excluded to avoid the potential confounding effect that prior surgery may have on pelvic floor support. The safety data included in this report were obtained from a total of 6926 postmenopausal women with uteri at baseline, 969 from the prevention studies, and 5957 from the treatment study, followed for up to 3 years.
Statistical analyses were performed in an intention-to-treat analysis using Statistical Analysis Software Version 6.08 (SAS Institute Inc., Cary, NC). Statistical significance was defined as P ≤ .05 (two-tailed test). Continuous baseline characteristics were analyzed using two-way analysis of variance with therapy and protocol as fixed effect in the model and discrete parameters using the Cochran-Mantel-Haenszel technique, stratified by protocol. The incidence of surgery for pelvic floor relaxation in the raloxifene group was compared with the incidence in the placebo group. We also compared the incidence of pelvic floor surgery in the treated and untreated groups stratified for age. The interaction of age (at least 60 versus under 60 years) and therapy was tested by using a two-way analysis of variance with age, therapy, and age-by-therapy interaction as fixed effects. Adjusted odds ratio (OR) and 95% confidence interval (CI) were calculated using logistic regression. Cumulative incidence of pelvic floor surgery was plotted by treatment assignment, and the statistical significance of the difference was assessed by the log rank test.
Of 8850 women enrolled in Studies 1, 2, and 3, 6926 had an intact uterus at study entry and were included in this analysis. A total of 2246 were assigned to placebo (247 from the prevention studies and 1999 from the treatment study) and 4680 to raloxifene (722 from the prevention studies and 3958 from the treatment study). The demographics of the trial populations are presented in Table 2. There were no significant differences in baseline characteristics between the placebo group and the raloxifene group after adjusting for protocol. The incidence of urinary incontinence or pelvic organ prolapse at baseline was similar between the groups.
A total of 103 surgical procedures potentially indicated for pelvic floor relaxation were identified in the database. Thirty-four procedures were hysterectomies that were performed for reasons other than pelvic floor relaxation (eg, bleeding, leiomyoma, severe dysplasia, endometrial cancer, ovarian cancer, ovarian cyst, and cervical cancer) and were excluded from the analysis. Surgical procedures indicated for pelvic floor relaxation were reported for 69 women, 34 (1.51%) placebo subjects and 35 (0.75%) women assigned to raloxifene (Figure 1). After adjusting for protocol, the OR for pelvic floor surgery in the raloxifene group, as compared with placebo, was 0.50 (95% CI 0.31, 0.81). Raloxifene therapy was associated with 50% reduction in the risk of pelvic floor surgery (P < .005). A trend towards a reduced incidence of surgery associated with raloxifene treatment was observed within 9 months of treatment and was sustained throughout the 3-year treatment period (P = .003) (Figure 2). The percentage of women who reported urinary incontinence or pelvic organ prolapse at the time of enrollment was similar between the placebo (50%) and raloxifene (65.7%) groups (P = .176).
To determine whether age influences the effect of raloxifene on the risk for pelvic floor surgery, the incidence of surgery was determined in women younger than 60 years (n = 2042, mean age 55.2 years) and in those 60 years and older (n = 4884, mean age 68.8 years). In women younger than age 60, 13 (0.64%) operations were performed, five (0.85%) in the placebo group (n = 588) and eight (0.55%) in the raloxifene group (n = 1454) (P = .494). Raloxifene therapy was associated with a nonsignificant reduction in the incidence of surgery in younger women (adjusted OR 0.68, 95% CI 0.22, 2.08). In women 60 years of age or older, 56 (1.15%) operations were performed, 29 (1.75%) in placebo subjects (n = 1658) and 27 (0.84%) in those assigned to raloxifene (n = 3226) (P < .005). The adjusted OR for pelvic floor surgery in these older women taking raloxifene was 0.47 (95% CI 0.28, 0.80). There was no statistically significant difference in treatment effect on pelvic surgery incidence in the younger population compared with the older population (treatment-by-age interaction, P = .594).
The incidence rate of pelvic floor surgery in our studies was 4.2 per 1000 person-years overall, 3.1 per 1000 for women less than 60 years of age, and 4.6 per 1000 for women up to 60 years of age. The total raloxifene exposure was 11,080 person-years, as compared with 5421 person-years for placebo. The overall incidence rate for women on raloxifene was 3.2 per 1000 person-years, as compared with 6.3 per 1000 for placebo.
Pelvic floor relaxation is common among postmenopausal women. Its clinical manifestations range from urinary incontinence to total vault prolapse. Traditional treatment of these disorders has often included estrogen replacement therapy, although no clinical trials have confirmed the validity of this approach.8 Severe pelvic floor relaxation often involves surgical intervention, which is associated with a number of risks, including operative injury, recurrent prolapse, persistent or unmasked stress incontinence, voiding difficulties, impairment of sexual function, and worsening of preexisting perineal neuropathy.20–22 Thus, postmenopausal therapies for otherwise healthy women should not cause adverse effects in the genitourinary system and would ideally improve pelvic floor relaxation.
The current analysis, based on data from 6926 postmenopausal women, demonstrates that raloxifene was associated with a 50% reduction in the risk of surgery for pelvic floor relaxation in women treated for up to 3 years. Similar to results from epidemiologic studies, our data demonstrate a surgery incidence rate of 4.2 per 1000 person-years,1 suggesting that the observations made in this study may be applicable to a broader population of women. Our data also demonstrate an overall increased incidence of pelvic floor surgery with age (0.64% in women under 60 years and 1.15% in women at least 60 years), similar to findings in previous studies.1 A trend towards a reduced rate of surgical repair associated with raloxifene therapy was observed in younger postmenopausal women, but this effect was not statistically significant. In the older postmenopausal population, raloxifene therapy was associated with a 50% reduction in the risk of pelvic floor surgery. A trend towards a decreased incidence of pelvic floor repair associated with raloxifene use was observed within approximately 9 months of therapy initiation; no differences between placebo and raloxifene were observed in the first 9 months of therapy. The results of this analysis indicate that raloxifene does not contribute to pelvic floor relaxation. The data raise the interesting hypothesis that raloxifene may actually contribute to a reduction in the incidence of pelvic floor relaxation, which could be a reflection of estrogen agonist activities on tissues of the pelvic floor.
Several methodologic shortcomings in this study must be recognized. The clinical trials described herein were not designed to assess the effect of raloxifene in the pelvic floor; pelvic floor examinations were not mandated at any time during the study, and there was no systematic evaluation for pelvic organ relaxation.23 An additional weakness of this analysis is the lack of baseline data around the obstetric history of patients. However, the large number of women who were included in this analysis, the highly significant reduction in pelvic floor surgery associated with raloxifene therapy, and the lack of therapy difference in the first 9 months of treatment suggest that the current findings are not spurious.
The findings of this study on the effect of raloxifene on the genitourinary system contrast with the effects recognized for two other SERMs, idoxifene, and levormeloxifene. Idoxifene, a triphenylethylene SERM related to tamoxifen, was associated with an increased incidence of pelvic organ prolapse within 1 year of therapy initiation in postmenopausal women participating in clinical trials of osteoporosis.9 An increased incidence of pelvic organ prolapse and urinary incontinence was also observed for levormeloxifene, a centchroman SERM, within 1 year of therapy initiation.9 Because genitourinary adverse affects would not be tolerable to a relatively healthy postmenopausal population, clinical development of both idoxifene and levormeloxifene for osteoporosis indications was discontinued. In contrast, neither tamoxifen nor toremifene, two clinically available SERMs, have been associated with pelvic floor relaxation.24,25
The mechanism by which raloxifene might affect pelvic floor relaxation is not clear. The presence of estrogen receptors in connective tissues and muscles of the pelvic floor indicates that this region is a target for estrogen and may be responsive to selective estrogen receptor modulators.26 Turnover and remodeling of the pelvic connective tissue is an ongoing process that probably changes with aging and menopause. For example, changes in one marker of tissue remodeling, namely collagen metabolism, have been observed in bone27 and skin5 in association with aging and menopause. In addition, abnormal collagen metabolism has been observed in the vaginal tissues of women with genitourinary prolapse.28 Thus, agents that positively affect tissue remodeling may restore pelvic tone and reduce the incidence of pelvic floor relaxation. Although no data yet supports a remodeling effect of raloxifene in pelvic floor tissues, it is well known that raloxifene decreases collagen turnover in the skeleton,10,29 an affect that may contribute in part to the 30–50% reduction in risk for vertebral fracture in postmenopausal osteoporotic women on raloxifene therapy.11
1. Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 1997;89:501–6.
2. Bump RC, Norton PA. Epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin North Am 1998;25:723–46.
3. Gilpin SA, Gosling JA, Smith AR, Warrell DW. The pathogenesis of genitourinary prolapse and stress incontinence of urine. A histological and histochemical study. Br J Obstet Gynaecol 1989;96:15–23.
4. DeLancey JO. Anatomy and biomechanics of genital prolapse. Clin Obstet Gynecol 1993;36:897–909.
5. Affinito P, Palomba S, Sorrentino C, Di Carlo C, Bifulco G, Arienzo MP, et al. Effects of postmenopausal hypoestrogenism on skin collagen. Maturitas 1999;33:239–47.
6. Nguyen JK, Lind LR, Choe JY, McKindsey F, Sinow R, Bhatia NN. Lumbosacral spine and pelvic inlet changes associated with pelvic organ prolapse. Obstet Gynecol 2000;95:332–6.
7. Rekers H, Drogendijk AC, Valkenburg HA, Riphagen F. The menopause, urinary incontinence and other symptoms of the genitourinary tract. Maturitas 1992;15:101–11.
8. Casper RF, McPherson R, Ste-Marie LG. The Canadian Consensus Conference on Menopause and Osteoporosis. J Soc Obstet Gynaecol Can 1998;20:1243–72.
9. Silfen SL, Ciaccia AV, Bryant HU. Selective estrogen receptor modulators: Tissue specificity and differential uterine effects. Climacteric 1999;2:268–83.
10. Delmas PD, Bjarnason NH, Mitlak BH, Ravoux A-C, Shah AS, Huster WJ, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N Engl J Med 1997;337:1641–7.
11. Ettinger B, Black DM, Mitlak BH, Knickerbocker RK, Nickelsen T, Genant HK, et al. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: Results from a 3-year randomized clinical trial. JAMA 1999;282:637–45.
12. Cummings S, Eckert S, Krueger K, Grady D, Powles T, Cauley J, et al. The effect of raloxifene on risk of breast cancer in postmenopausal women. JAMA 1999;281:2189–97.
13. Walsh BW, Kuller LH, Wild RA, Paul S, Farmer M, Lawrence JB, et al. Effects of raloxifene on serum lipids and coagulation factors in healthy postmenopausal women. JAMA 1998;279:1445–51.
14. Walsh BW, Paul S, Wild RA, Dean RA, Tracy RP, Cox DA, et al. The effects of hormone replacement therapy and raloxifene on C-reactive protein and homocysteine in healthy postmenopausal women: A randomized, controlled trial. J Clin Endocrinol Metab 2000;85:214–8.
15. Cohen FJ, Watts S, Shah A, Akers R, Plouffe L. Uterine effects of three-year raloxifene therapy in postmenopausal women under age 60. Obstet Gynecol 2000;95:104–10.
16. Goldstein SR, Scheele WH, Rajagopalan SK, Wilkie JL, Walsh BW, Parsons AK. A 12-month comparative study of raloxifene, estrogen, and placebo on the postmenopausal endometrium. Obstet Gynecol 2000;95:95–103.
17. Davies GC, Huster WJ, Shen W, Mitlak BH, Plouffe L, Shah A, et al. The endometrial response to raloxifene compared with placebo, cyclical hormone replacement therapy, and unopposed estrogen in postmenopausal women. Menopause 1999;6:188–95.
18. Davies GC, Huster WJ, Lu Y, Plouffe L, Lakshmanan M. Adverse events reported by postmenopausal women in controlled trials with raloxifene. Obstet Gynecol 1999;93:558–65.
19. Johnston CC Jr, Bjarnason NH, Cohen FJ, Shah AS, Lindsay R, Mitlak BH, et al. Long-term effects of raloxifene on bone mineral density, bone turnover, and serum lipids in early postmenopausal women. 3-year data from two double-blind, randomized, placebo-controlled trials. Arch Intern Med 2000;160:3444–50.
20. Zivkovic F, Tamussino K. Effects of vaginal surgery on the lower urinary tract. Curr Opin Obstet Gynecol 1997;9:329–31.
21. Burnett LS, Buckley SL. Surgical failures in the management of pelvic floor relaxation. Curr Opin Obstet Gynecol 1993;5:465–70.
22. Zivkovic F, Tamussino K, Ralph G, Schied G, Auer-Grumbach M. Long-term effects of vaginal dissection on the innervation of the striated urethral sphincter. Obstet Gynecol 1996;87:257–60.
23. Bump RC, Mattiasson A, Bo K, Brubaker LP, DeLancey 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.
24. Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, et al. Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371–88.
25. Maenpaa JU, Ala-Fossi SL. Toremifene in postmenopausal breast cancer. Efficacy, safety and cost. Drugs Aging 1997;11:261–70.
26. Smith P, Heimer G, Norgren A, Ulmsten U. Localization of steroid hormone receptors in the pelvic muscles. Eur J Obstet Gynecol Reprod Biol 1993;50:83–5.
27. Bailey AJ, Sims TJ, Ebbesen EN, Mansell JP, Thomsen JS, Mosekilde L. Age-related changes in the biochemical properties of human cancellous bone collagen: Relationship to bone strength. Calcif Tissue Int 1999;65:203–10.
28. Jackson SR, Avery NC, Tarlton JF, Eckford SD, Abrams P, Bailey AJ. Changes in metabolism of collagen in genitourinary prolapse. Lancet 1996;347:1658–61.
29. Prestwood KM, Gunness M, Muchmore DB, Lu Y, Wong M, Raisz LG. A comparison of the effects of raloxifene and estrogen on bone in postmenopausal women. J Clin Endocrinol Metab 2000;85:2197–202.