Abstract
Background:
Menopause and the decline in systemic estrogen are associated with development of pelvic floor disorders such as prolapse, urinary incontinence, overactive bladder, and vulvovaginal atrophy symptoms. Past evidence suggests that postmenopausal women with symptomatic prolapse gain benefit from preoperative application of intravaginal estrogen, but it is unknown whether they would experience improvement in other pelvic floor symptoms when treated with intravaginal estrogen.
Objectives:
To determine effects of intravaginal estrogen (compared to placebo) on stress and urgency urinary incontinence, urinary frequency, sexual function/ dyspareunia, and vaginal atrophy symptoms and signs in postmenopausal women with symptomatic prolapse.
Study Design:
Planned ancillary analysis of a randomized, double-blind trial, “Investigation to Minimize Prolapse Recurrence Of the Vagina using Estrogen” (IMPROVE), which included participants with ≥stage 2 apical and/or anterior prolapse scheduled for transvaginal native tissue apical repair at 3 US sites. The intervention was 1g conjugated estrogen intravaginal cream (0.625mg/g) or identical placebo (1:1), inserted nightly for 2 weeks, then twice weekly for ≥5 weeks total before surgery and continued twice weekly for 1 year postoperatively. For this analysis, question responses were compared from participants’ baseline and preoperative visits: lower urinary tract symptoms (UDI-6 questionnaire); sexual health questions including dyspareunia (PISQ-IR questionnaire); and atrophy-related symptoms (dryness, soreness, dyspareunia, discharge, itching; each scored 1-4, 4 being quite-a-bit bothersome). Masked examiners assessed vaginal color, dryness, and petechiae (each scored 1-3, total range 3-9, with 9 being most estrogenized-appearing). Data were analyzed by intent-to-treat and “per-protocol” (i.e., those adherent with ≥50% of expected intravaginal cream use, per objective tube before/after weights).
Results:
Of 199 participants randomized (mean age 65 years) and contributing baseline data, 191 also had preoperative data. Characteristics were similar between groups. Total UDI-6 scores showed minimal change during this median time of 7 weeks between baseline and preoperative visits, but for those with at least moderately bothersome baseline stress urinary incontinence (n=32 estrogen group, n=21 placebo group), 16 (50%) and 9 (43%) showed improvement, respectively (P=0.78). Likewise, 43% and 31% showed improvement in urgency urinary incontinence (P=0.41), and 41% and 26% showed improvement in urinary frequency (P=0.18). There was minimal change in PISQ-IR among sexually active women; dyspareunia rates did not differ between intravaginal estrogen and placebo at the preoperative assessment: 42% and 48%, respectively, P=0.49. The maximum score for most bothersome atrophy symptom (among those with baseline symptoms and adherent to study cream) improved slightly more with intravaginal estrogen [adjusted mean difference of −0.33 points (95% CI: −0.98, 0.31)], but this was not statistically significant, P=0.19. However, on examination, among adherent participants, objective signs of atrophy were more improved with intravaginal estrogen treatment [+1.54 vs. +0.69, mean difference 0.85 (95% CI: 0.05, 1.65), P=0.01].
Conclusions:
Despite objective changes in vaginal epithelium consistent with increased estrogenization among drug-adherent participants, results were inconclusive regarding whether 7 weeks of preoperative intravaginal estrogen cream in postmenopausal women with symptomatic pelvic organ prolapse was associated with improved urinary, sexual function and dyspareunia symptoms, or other symptoms commonly attributed to atrophy. Additional study is needed.
Keywords: dyspareunia, genitourinary syndrome of menopause, oestrogen, sexual dysfunction, sexual function, stress urinary incontinence, urgency urinary incontinence, urinary frequency, vaginal atrophy
Tweetable statement:
In postmenopausal patients with pelvic organ prolapse, 7wk of preoperative intravaginal estrogen cream (vs placebo) improves tissue quality but impact upon urinary incontinence, sexual function, and atrophy symptoms remains uncertain.
INTRODUCTION
The decline in available estrogen with menopause contributes to vulvovaginal atrophy symptoms but likely also to other pelvic floor disorders including pelvic organ prolapse, stress urinary incontinence (SUI), urgency urinary incontinence (UUI), urinary frequency, and sexual dysfunction/dyspareunia.1 Besides being found in the vestibule and vagina, estrogen receptors have been identified in the distal urethra, bladder trigone, pelvic muscles, and surrounding supportive ligaments. These structures are responsive to estrogen status, suggesting that decreased levels of estrogen can adversely impact collagen and muscles of the pelvis.2
Intravaginal estrogen—in its many formulations as creams, tablets/inserts, or rings—has been shown to effectively treat genitourinary syndrome of menopause (GSM), in which patients’ most bothersome symptoms include vaginal dryness and associated dyspareunia.3 The evidence is less certain whether intravaginal estrogen is a useful treatment—or adjunctive therapy—in postmenopausal women with SUI, UUI, or other overactive bladder symptoms.4 A Cochrane Review of estrogen therapy for urinary incontinence found that effects were disparate for systemic vs. vaginal estrogen: the combined result of six trials of systemic administration of estrogen resulted in worse incontinence compared to placebo [risk ratio (RR) 1.32, 95% CI 1.17, 1.48] while vaginal estrogens were found to improve incontinence (RR 0.74, 95% CI 0.64, 0.86) and urinary urgency (RR 0.38, 95% CI 0.15, 0.99).5 Effects on SUI specifically are not well characterized, but one small trial found that local estrogen decreased the pulsatility index in the urethral vascular network (indicating greater blood flow) and improved subjective SUI bother in 60% of patients.6 Of note, the Cochrane Review’s summaries of improvements with local estrogen were based on just 2–4 trials, and further study was recommended.5
It is unknown whether postmenopausal women with symptomatic pelvic organ prolapse will experience improvement in other pelvic floor disorders when exposed to intravaginal estrogen. The Investigation to Minimize Prolapse Recurrence Of the Vagina using Estrogen (IMPROVE)7 trial provided a well-characterized cohort of postmenopausal patients randomized to intravaginal estrogen vs. placebo cream for five or more weeks preceding standardized native tissue transvaginal surgical repair. The aims of this ancillary analysis were to determine the effects of intravaginal estrogen—compared to placebo—on SUI, UUI, sexual function and dyspareunia, and vaginal atrophy symptoms and signs.
MATERIALS AND METHODS
This was a planned ancillary analysis of a 3-center, randomized superiority trial (NCT02431897), the design of which has been published previously.7 The IMPROVE protocol and statistical analysis plan were approved by institutional review boards at each site and by an independent Data and Safety Monitoring Board (DSMB) assembled by the National Institute on Aging. All participants provided written informed consent.
The eligible study population included postmenopausal patients 48 years of age or greater with symptomatic prolapse of the vaginal apex and/or anterior wall at least to the hymen planning an elective native-tissue transvaginal surgical repair. Participants need not necessarily have had baseline complaints of SUI, UUI, sexual dysfunction or pain, or vulvovaginal atrophy symptoms. Those with contraindications to intravaginal estrogen therapy, body mass index greater than 35 kg/m2, current tobacco use, or with prior vaginal apical prolapse repair or use of mesh for prolapse repair were excluded.
Participants were randomized 1:1 to conjugated estrogen or identical placebo cream via stratified randomization lists with a consistent block factor of size 4. Stratifications were made for clinical site, duration since menopause (less than 10 and ≥10 years), and hysterectomy status. Patients and outcomes assessors were masked to assignment. The study intervention was conjugated estrogen cream 0.625mg/g (Premarin) 1g or identical placebo, inserted vaginally nightly for 2 weeks, then twice weekly for a minimum of 5 weeks total (goal: 6-8 weeks) before prolapse surgery, consistent with a trial by Manonai et al and the pilot study preceding IMPROVE.8,9 The intravaginal cream could be applied by either the provided plastic applicator or fingertip.
At baseline (i.e., time of randomization and start of study cream) and at their preoperative appointment, patients’ urinary symptom bother was queried via the short version of the Urinary Distress Inventory, i.e., the UDI-6.10 This questionnaire ascertains presence of and bother with urinary frequency, leakage associated with urgency, leakage associated with stress maneuvers, small amounts of leakage (drops), difficulty with bladder emptying, and pain or discomfort in the lower abdomen, pelvic, or genital area. Sexual function was assessed with the Pelvic Organ Prolapse/Urinary Incontinence Sexual Function Questionnaire (PISQ-IR).11 Vaginal atrophy symptoms were assessed with a questionnaire designed for IMPROVE which asked about symptoms of vaginal dryness, soreness/pain, pain with intercourse (if sexually active), discharge, and itching. Each of these symptoms were either not present (0) or present and scored from not at all bothersome (1) to quite a bit bothersome (4). Finally, masked examiners completed an objective vaginal atrophy assessment implemented for IMPROVE, rating dryness/moisture, color/pallor, and presence/absence of petechiae, each scored from 1-3, total range 3-9, with 9 being most estrogenized appearing.
Outcomes were analyzed according to the assigned treatment group (i.e., intention-to-treat) in all participants who were randomized, received study cream, and contributed data at baseline and a preoperative visit. A per-protocol analysis was also planned for those who were at least 50% adherent to their study cream as assessed by objective before-and-after tube weights. Fifty-percent adherence was consistent with a pivotal dosing paper of conjugated estrogens cream by Bachmann et al.12 At baseline, continuous measures were compared between groups with Student’s t-test (or Mann-Whitney U test) and chi-square (or Fisher’s exact test) for categorical measures. Scores from repeated surveys with continuous outcomes were analyzed using a mixed effects model with repeated measures and subject as a random effect; this approach facilitates the use of all participant data, including those participants who may not have data from both baseline and the preoperative visit. Tests for interactions were performed between time effect and treatment effect. For measures that may not be assumed normal, the data were transformed prior to analysis. Comparisons of atrophy symptoms used analysis of covariance, with treatment and study center as factors and baseline values as the covariate. Adjusted means, mean difference, 95% CIs, and P values comparing continuous outcomes of the intervention groups are obtained from mixed effects models adjusted for site and treatment group.
RESULTS
A total of 199 participants were randomized in IMPROVE, and 186 ultimately underwent surgery with 93 in the estrogen and 93 in the placebo cream arms. However, 191 contributed data at both the baseline and preoperative assessments. Baseline demographics and clinical characteristics of participants are noted in Table 1. There was no difference between arms, with mean age 67 (standard deviation [SD] 6.7) years. Ninety-one percent were white, and 8 percent were black. Twenty-four percent were of Hispanic ethnicity. Approximately 45% had no atrophy symptoms at baseline, but of those who did, dryness was the most bothersome symptom, occurring in 45 of 199 (23%). Approximately two-thirds of participants had stage 3 pelvic organ prolapse; the remainder were stage 2 or 4. There were no differences in atrophy findings on examination at baseline with an overall median score of 6 (interquartile range [IQR], 5-7); the full range (3-9) was seen in both groups. There were no differences between groups in urinary distress or sexual dysfunction/dyspareunia symptoms (Table 1).
Table 1.
Baseline Participant Characteristics
| Treatment Group | ||
|---|---|---|
| Characteristic | CE (n=102)a | Placebo (n=97)a |
| Age, mean (SD), y | 64.9 (6.1) | 65.2 (7.3) |
| Race, No. (%) | ||
| White | 92 (90) | 89 (92) |
| Black | 10 (10) | 5 (5) |
| Asian | 0 (0) | 1 (1) |
| American Indian/Alaskan Native | 0 (0) | 0 (0) |
| Other | 0 (0) | 2 (2) |
| Hispanic or Latina, No. (%) | 24 (24) | 24 (25) |
| Health Insurance, No. (%) | ||
| Medicare/Medicaid | 34 (33) | 33 (34) |
| Private/HMO | 47 (46) | 44 (45) |
| Self-pay | 1 (1) | 1 (1) |
| Other | 20 (20) | 19 (20) |
| Medical History | ||
| Gravidity, median (IQR) | 3 (2, 3) | 2 (2, 3) |
| Vaginal deliveries, median (IQR) | 2 (2, 3) | 2 (2, 3) |
| Caesarean deliveries, median (IQR)b | 0 (0, 0) | 0 (0, 0) |
| Smoking status | ||
| Never-smoker, No. (%) | 71 (70) | 70 (72) |
| Former smoker, quit <6mo prior, No. (%) | 0 (0) | 2 (2) |
| Former smoker, quit ≥6mo prior, No. (%) | 31 (30) | 25 (26) |
| Diabetes mellitus, No. (%) | 20 (20) | 19 (20) |
| ≥3 urinary tract infections in preceding year, No. (%) | 5 (5) | 5 (5) |
| Use of overactive bladder medication | 14 (14) | 19 (20) |
| Surgical History | ||
| Prior stress urinary incontinence surgery, No. (%) | 6 (6) | 4 (4) |
| Prior pelvic organ prolapse surgery, No. (%) | 6 (6) | 4 (4) |
| Exam | ||
| BMI, mean (SD) | 27.3 (3.7) | 27.8 (4.0) |
| POP-Q Measurement (cm), median (IQR)c | ||
| Ba | 2 (1, 3) | 2 (1, 3) |
| Bp | −1 (−2, 0) | −1.5 (−2, 0) |
| C | −0.5 (−3, 3) | −2 (−3, 2) |
| TVL | 9 (8.5, 10) | 9 (8, 10) |
| GH | 4 (3, 5) | 4 (3.4, 5) |
| POP-Q Stage, No. (%)d | ||
| 2 | 29 (29) | 28 (29) |
| 3 | 63 (62) | 64 (66) |
| 4 | 9 (9) | 5 (5) |
| Oxford Pelvic Muscle Assessment, median (IQR)e | 2 (2, 3) | 2 (2, 3) |
| Vaginal epithelial assessment (erosion), No. (%)f | 2 (2) | 2 (2) |
| Vaginal Atrophy Assessment total, median (IQR)g | 6 (5, 7) | 6 (5, 7) |
|
| ||
| Patient-reported outcome scores, mean (SD) | ||
| Urogenital Distress Inventory-6h, median (IQR) | 37.5 (20.8, 58.3) | 37.5 (12.5, 54.2) |
| PISQ-IR score of sexually active womeni | (n=39) | (n=39) |
| Mean (SD) | 3.2 (0.7) | 3.3 (0.5) |
| Dyspareunia, No. (%), from PISQ-IR | (n=98) | (n=93) |
| 41 (42) | 47 (51) | |
| Vaginal Atrophy Symptoms Questionnaire, mean scorei | 0.78 (0.2, 1.6) | 0.75 (0, 1.5) |
| Vaginal Atrophy Symptoms Questionnaire, max score (i.e. of most bothersome symptom)j | 2 (1, 3) | 2 (0, 3) |
| Vaginal Atrophy Symptoms Questionnaire, max score (among those with at least one bothersome symptom)j | 3 (3, 4) | 3 (2, 4) |
| Vaginal Atrophy Symptoms Questionnaire, Most Bothersome Symptom, No. (%)ij | ||
| Not applicable, none cause this much bother | 47 (46) | 43 (44) |
| Dryness | 27 (26) | 18 (19) |
| Soreness/ pain | 9 (9) | 12 (12) |
| Pain with intercourse | 6 (6) | 11 (11) |
| Discharge | 8 (8) | 5 (5) |
| Itching | 5 (5) | 8 (8) |
Abbreviations: CE, conjugated estrogen cream; SD, standard deviation; IQR, interquartile range; BMI, body mass index; POP-Q, Pelvic Organ Prolapse Quantification; TVL, total vaginal length; GH, genital hiatus; PISQ-IR, Pelvic Organ Prolapse/Incontinence Sexual Function Questionnaire, IUGA-Revised
Scores from repeated surveys with continuous outcomes were analyzed using a mixed effects model with repeated measures and subject as a random effect; this approach facilitates the use of all participant data, including those participants who may not have data from both baseline and the preoperative visit. These N’s of 102 and 97 represent all randomized participants.
Majority of patients had no cesarean deliveries. For interest, the maximum values for CE group was 4; the maximum value for placebo group was 3.
POP-Q point C represents the most distal edge of the cervix or vaginal cuff (if post-hysterectomy); point Ba, the most distal point of any part of the upper anterior vaginal wall, ranging from −3cm to TVL; point Bp, the most distal point of any part of the upper posterior vaginal wall, ranging from −3cm to TVL; TVL represents the distance from the posterior fornix to the hymen when point C is reduced to its full normal position. Negative values indicate absence of prolapse, i.e., measurements cephalad to/above the hymen, whereas positive values indicate the number of centimeters of prolapse outside the hymen.
A POP-Q of Stage 2 indicates the vagina is prolapsed between 1cm above and 1cm below the hymen; stage 3, the vagina is prolapse more than 1cm beyond the hymen but is not everted within 2cm of TVL; stage 4, the vagina is everted to within 2cm of TVL
The Oxford Pelvic Muscle Assessment measures levator strength of contraction as 0, no contraction; 1, “flicker”; 3, moderate strength with lift of pelvic floor; to 5, strong contraction, with elevation of the examiner’s fingers against resistance.
Erosion at baseline refers to visible defects in the externalized vaginal mucosa related to prolapse
Three missing observations. The Vaginal Atrophy Assessment tool assesses dryness/moisture, color/pallor, and presence/absence of petechiae each from 1, most atrophic, to 3, most estrogenized-appearing, for a total possible score range of 3 to 9.
The Urogenital Distress Inventory-6 is scaled from 0 (least distress) to 100 (most distress).
The PISQ-IR mean score ranges from 1 (worse sexual experience) to 5 (better sexual experience) with midrange scores seen commonly with women with pelvic floor disorders
The Vaginal Atrophy Symptoms Questionnaire asks respondents to rate—over the preceding month—their experiences of vaginal dryness, soreness or pain, pain with intercourse (if sexually active), discharge, and itching, each scored as 0, not experienced; 1, present but not bothersome; 2, somewhat bothersome; 3, moderately bothersome; or 4, quite a bit bothersome. Of those symptoms causing at least moderate bother, the respondent is asked to identify the most bothersome symptom.
Time from baseline to preoperative assessment was approximately 7 weeks: median (IQR) of 47 (33-66) days for the estrogen group and 43 (33-63) days for the placebo group, P = 0.43. Adherence, defined by use of at least 50% of anticipated study cream by weight, was similar by group: 76 of 94 (80.1%) for the estrogen group and 70 of 93 (75.3%) for the placebo group, P = 0.36.
Urinary symptom changes are detailed in Table 2. There were minimal changes in the total UDI-6 scores for either group. When isolating individual symptom components of the questionnaire and considering changes for those with at least moderate bother at baseline, SUI symptoms were resolved or improved in 16 of 32 (50%) in the estrogen arm compared to 9 of 21 (43%) of the placebo arm, P = 0.78; UUI symptoms resolved or improved in 13 of 30 (43%) in the estrogen arm and 11 of 36 (31%) in the placebo arm, P = 0.41; and urinary frequency resolved or improved in 18 of 44 (41%) in the estrogen arm and 12 of 47 (26%) in the placebo arm, P = 0.18. Of note, frequencies of overactive bladder medication use were similar at baseline: 14 of 102 (14%) in the estrogen arm and 19 of 97 (20%) in the placebo arm, P = 0.27.
Table 2.
Urinary Symptoms
| Estrogen | Placebo | Difference (95% CI) | P value | |||
|---|---|---|---|---|---|---|
| Urogenital Distress Inventory-6 a | (n=102) | (n=97) | ||||
| Intent-to-Treat | Baseline | Preop | Baseline | Preop | ||
| Adjusted mean (SE) | 40.2 (2.7) | 38.2 (2.7) | 38.1 (2.8) | 37.7 (2.8) | ||
| Adjusted mean difference (95% CI) | −2.0 (−5.5, 1.4) | −0.4 (−3.9, 3.1) | −1.6 (−8, 4.8) | 0.52 | ||
| Per Protocol (i.e., adherent to >50% of expected cream, by weight) | (n=76) | (n=70) | ||||
| Baseline | Preop | Baseline | Preop | |||
| Adjusted mean (SE) | 39.8 (3.1) | 39.2 (3.1) | 37.7 (3.2) | 36.9 (3.2) | ||
| Adjusted mean difference (95% CI) | −0.6 (−4.6, 3.5) | −0.8 (−5.0, 3.4) | 0.2 (−7.4, 7.8) | 0.94 | ||
|
| ||||||
| Resolved or improving stress urinary incontinence, No./total (%)b | (n = 49) | (n = 36) | ||||
| 20 (41) | 17 (47) | −6.4 (−27.7, 14.9) | 0.66 | |||
| Resolved or improving urgency urinary incontinence, No./total (%)b | (n = 47) | (n = 52) | ||||
| 17 (36) | 16 (31) | 5.4 (−13.2, 24.0) | 0.72 | |||
| Resolved or improving urinary frequency, No./total (%)b | (n = 62) | (n=64) | ||||
| 20 (32) | 15 (23) | 8.8 (−6.8, 24.4) | 0.37 | |||
| Resolved or improving stress urinary incontinence, No./total (%)c | (n = 32) | (n = 21) | ||||
| 16 (50) | 9 (43) | 7.1 (−20.2, 34.5) | 0.78 | |||
| Resolved or improving urgency urinary incontinence, No./total (%)c | (n = 30) | (n = 36) | ||||
| 13 (43) | 11 (31) | 12.7 (−10.5, 36.0) | 0.41 | |||
| Resolved or improving urinary frequency, No./total (%)c | (n = 44) | (n = 47) | ||||
| 18 (41) | 12 (26) | 15.4 (−3.8, 34.5) | 0.18 | |||
SE, Standard error; CI, confidence interval; adjusted means, mean difference, 95% CIs, and P values comparing continuous outcomes of the intervention groups are obtained from mixed effects models adjusted for site and treatment group
The Urinary Distress Inventory-6 (UDI-6) is scaled from 0 (least distress) to 100 (most distress)
Denominator is participants answering “Yes” to the symptom question at baseline and with some degree of bother; numerator represents those answering “No” (or “Yes” but with less bother) at the preoperative assessment
subanalysis for those participants who were at least moderately bothered at baseline
Sexual function information is in Table 3. Between baseline and time of preoperative assessment, there was no change in the PISQ-IR scores of sexually active women. In this cohort of women with prolapse, dyspareunia (i.e., stating pain during sexual intercourse occurring sometimes, usually, or always, or not sexually active due to pain) was still endorsed by almost half of participants at the time of the preoperative assessment, with no statistically significant differences in frequency between the estrogen and placebo groups.
Table 3.
Sexual Function and Dyspareunia
| Estrogen | Placebo | Difference (95% CI) | P value | |||
|---|---|---|---|---|---|---|
| PISQ-IR score of sexually active women * | (n=39) | (n=39) | ||||
| Intent-to-Treat | Baseline | Preop | Baseline | Preop | ||
| Adjusted mean (SE) | 3.16 (0.10) | 3.13 (0.10) | 3.23 (0.10) | 3.12 (0.10) | ||
| Adjusted mean difference (95% CI) | −0.03 (−0.18, 0.11) | −0.11 (−0.25, 0.03) | 0.07 (−0.18, 0.33) | 0.47 | ||
| Per Protocol (i.e., adherent to >50% of expected cream, by weight) | (n=31) | (n=34) | ||||
| Baseline | Preop | Baseline | Preop | |||
| Adjusted mean (SE) | 3.13 (0.10) | 3.12 (0.10) | 3.20 (0.10) | 3.09 (0.10) | ||
| Adjusted mean difference (95% CI) | −0.02 (−0.19, 0.15) | −0.10 (−0.27, 0.06) | 0.09 (−0.21, 0.38) | 0.47 | ||
| Dyspareunia at time of preoperative assessment | ||||||
| All treated participants, No./total (%) | (n = 88) | (n = 91) | ||||
| 37 (42) | 44 (48) | −6.3 (−20.9, 8.2) | 0.49 | |||
SE, Standard error; CI, confidence interval; adjusted means, mean difference, 95% CIs, and P values comparing continuous outcomes of the intervention groups are obtained from mixed effects models adjusted for site and treatment group
PISQ-IR, Pelvic Organ Prolapse/Urinary Incontinence Sexual Function Questionnaire: mean score ranges from 1 (worse sexual experience) to 5 (better sexual experience) with midrange scores seen commonly with women with pelvic floor disorders
Vaginal atrophy symptoms and examination-related findings are presented in Table 4. Those with atrophy-related symptom bother at baseline had similar numerical improvements in the most bothersome symptom score for both groups, but these still corresponded with moderate bother at baseline and the time of preoperative assessment. Conversely, on objective assessment, among those participants who were adherent to their study cream, the intravaginal estrogen group did appear more estrogenized by blinded evaluation, improving from an adjusted mean (SE) score of 5.8 (0.2) to 7.3 (0.2), while the placebo group improved from 5.8 (0.2) to 6.5 (0.2), P = 0.01.
Table 4.
Vaginal Atrophy Symptoms and Signs
| Estrogen | Placebo | Difference (95% CI) | P value | |||
|---|---|---|---|---|---|---|
| Vaginal Atrophy Symptom Questionnaire, max score (among those with baseline bother)a | (n=55) | (n=54) | ||||
| Intent-to-Treat | Baseline | Preop | Baseline | Preop | ||
| Adjusted mean (SE) | 3.21 (0.14) | 2.74 (0.15) | 2.85 (0.14) | 2.67 (0.15) | ||
| Adjusted mean difference (95% CI) | −0.47 (−0.79, 0.15) | −0.18 (−0.50, 0.14) | −0.29 (−0.88, 0.30) | 0.21 | ||
| Per Protocol (i.e., adherent to >50% of expected cream, by weight) | (n=44) | (n=39) | ||||
| Baseline | Preop | Baseline | Preop | |||
| Adjusted mean (SE) | 3.19 (0.15) | 2.58 (0.16) | 2.93 (0.16) | 2.65 (0.16) | ||
| Adjusted mean difference (95% CI) | −0.61 (−0.95, 0.27) | −0.28 (−0.64, 0.08) | −0.33 (−0.98, 0.31) | 0.19 | ||
|
| ||||||
| Most bothersome symptom at time of preoperative assessment, No. (%) | (n=94) | (n=93) | 0.37 | |||
| None | 44 (47) | 45 (48) | ||||
| Dryness | 18 (19) | 18 (19) | ||||
| Soreness/pain | 9 (10) | 10 (11) | ||||
| Pain with intercourse | 2 (2) | 6 (6) | ||||
| Discharge | 8 (9) | 8 (9) | ||||
| Itching | 13 (14) | 6 (6) | ||||
|
| ||||||
| Vaginal Atrophy Assessment, total b | (n=93) | (n=93) | ||||
| Intent-to-Treat | Baseline | Preop | Baseline | Preop | ||
| Adjusted mean (SE) | 5.78 (0.15) | 7.09 (0.15) | 5.73 (0.16) | 6.51 (0.15) | ||
| Adjusted mean difference (95% CI) | 1.30 (0.92, 1.69) | 0.78 (0.39, 1.16) | 0.52 (−0.19, 1.24) | 0.06 | ||
| Per Protocol (i.e., adherent to >50% of expected cream, by weight) | (n=76) | (n=70) | ||||
| Baseline | Preop | Baseline | Preop | |||
| Adjusted mean (SE) | 5.79 (0.16) | 7.33 (0.16) | 5.82 (0.17) | 6.50 (0.17) | ||
| Adjusted mean difference (95% CI) | 1.54 (1.11, 1.97) | 0.69 (0.24, 1.13) | 0.85 (0.05, 1.65) | 0.01 | ||
SE, Standard error; CI, confidence interval; adjusted means, mean difference, 95% CIs, and P values comparing continuous outcomes of the intervention groups are obtained from mixed effects models adjusted for site and treatment group
The Vaginal Atrophy Symptoms Questionnaire asks respondents to rate—over the preceding month—their experiences of vaginal dryness, soreness or pain, pain with intercourse (if sexually active), discharge, and itching, each scored as 0, not experienced; 1, present but not bothersome; 2, somewhat bothersome; 3, moderately bothersome; or 4, quite a bit bothersome. Of those symptoms causing at least moderate bother, the respondent is asked to identify the most bothersome symptom.
The Vaginal Atrophy Assessment tool assesses dryness/moisture, color/pallor, and presence/ absence of petechiae each from 1, most atrophic, to 3, most estrogenized-appearing, for a total possible score range of 3 to 9.
A sub-analysis was performed comparing those with at least one moderately bothersome atrophy symptom at baseline (n=81) to those without (n=118). Those with atrophy-related bother had worse urinary dysfunction with median (IQR) UDI-6 scores of 41.7 (29.2-70.8) vs. 29.2 (12.5-50.0), P<0.01. Likewise, the symptomatic atrophy subgroup had significantly greater proportions who were also at least moderately bothered by SUI [n=30 (37%) vs. n=26 (22%), P = 0.021], UUI [39 (48%) vs. 29 (25%), P = 0.001], urinary frequency [50 (62%) vs. 46 (39%), P = 0.003], and dyspareunia [46 of 78 sexually active participants (59%) vs. 42 of 113 sexually active participants (37%), p=0.003]. However, when just the n=81 with bothersome atrophy symptoms were compared according to treatment group (n=43 with estrogen vs. n=38 with placebo), there were not significant differences observed in changes in UDI-6, PISQ-IR, maximum vaginal atrophy symptom scores, or objective atrophy assessment scores (data not shown).
COMMENT
Principal findings:
For this planned ancillary analysis of a randomized trial in postmenopausal women with symptomatic pelvic organ prolapse, it remains uncertain whether preoperative application of intravaginal estrogen cream for 7 weeks results in improved symptoms of SUI, UUI, urinary frequency, overall sexual function or dyspareunia, or bothersome symptoms associated with vulvovaginal atrophy. These inconclusive findings were despite objective changes assessed by clinic examiners masked to treatment assignment in the vaginal epithelium consistent with increased estrogenization among drug-adherent participants.
Results in the context of what is known:
Systematic reviews of use of intavaginal estrogen for pelvic floor disorders—apart from management of vulvovaginal atrophy symptoms, which have robust evidence—generally have found poor-to-moderate evidence quality, but there is the suggestion of benefit.4,13 With respect to local estrogen’s impact on urinary incontinence, Weber et al concluded that subjective, semi-objective, and urodynamic outcomes changed in favor of the vaginal estrogen groups compared to placebo. Further, they found largely similar treatment effects compared to non-hormonal agents such as tolterodine, an anticholinergic agent for overactive bladder treatment.13 Nonetheless, authors of the latest Cochrane Database review warn that current evidence needs to be interpreted with caution as conclusions about treatment effects were based on relatively few studies and numbers of patients, and there are a variety of types of delivery, dosages, and durations of treatment reported.5 To this point, the present study followed participants for a median of just 7 weeks of estrogen treatment, but most trials examining effects of estrogen on overactive bladder required 12 weeks of intervention.14–18
A prior randomized placebo-controlled trial of preoperative intravaginal estrogen cream in postmenopausal women with vaginal prolapse suggested that 6 or more weeks of topical estrogen therapy improved quality of vaginal tissue by increasing biogenesis of the extracellular matrix and reducing its degradation. Specifically, pre-operative intravaginal estrogen application increased synthesis of mature collagen, decreased degradative enzyme activity, and increased thickness of the vaginal wall.9 However, this past study did not attempt to correlate laboratory and histological findings with patient-reported outcomes of various pelvic floor symptoms, as this ancillary analysis of IMPROVE achieves.
Clinical implications:
In a cohort of postmenopausal women with symptomatic pelvic organ prolapse planning a surgical repair, it is possible that the degree of prolapse-related bother could overshadow or mask minor improvements in SUI, UUI, or urinary frequency symptoms. Of those participants with at least moderate bother in urinary symptoms at baseline, this study did identify modestly greater proportions of participants with resolution or improvement in these symptoms after using intravaginal estrogen, although this was not statistically significant compared to placebo; this points to the need for future studies that are adequately powered. Similarly, these patients with prolapse commonly had dyspareunia at baseline, which may have been related more to their prolapse—or numerous other possible causes of dyspareunia—than to vulvovaginal atrophy. Thus, preoperative application of intravaginal estrogen did not impact sexual function or dyspareunia. Further, vaginal atrophy symptoms improved modestly in both groups without a significantly greater improvement between groups. Taken together, this study cannot argue for or against standard use of intravaginal estrogen for 7 weeks to address urinary symptoms of SUI, UUI, or urinary frequency, or to improve sexual function or dyspareunia, in patients with bothersome prolapse. Further study is needed to determine whether increased time of use may provide benefit or if cream application would be more impactful in a cohort with greater symptom bother at baseline.
Strengths and Limitations:
The main strengths of this study include its randomized, placebo-controlled design and well-characterized patient population of postmenopausal participants with prolapse from three geographically diverse sites. Validated questionnaires were utilized to explore important patient-reported outcomes related to urinary incontinence, urinary frequency, and sexual function (specific to women with prolapse and/or incontinence) and dyspareunia.
There were several limitations. Foremost, while this was a planned ancillary analysis of the main IMPROVE trial19, it was not powered to specifically detect changes in these pelvic floor disorder symptoms. Inclusion requirements centered on symptomatic prolapse but did not specifically require a minimum baseline degree of bother from SUI, UUI, urinary frequency, sexual dysfunction, or vulvovaginal atrophy symptoms. Prior research recommends a UDI-6 score threshold of 33.33 to distinguish symptomatic (i.e., higher distress/bother) from asymptomatic urinary incontinence.20 The median baseline UDI-6 score in the present study was 38, pointing to a relatively modest bother from urinary incontinence at baseline. Thus, finding a significant difference in symptomatic leakage after approximately 7 weeks of topical estrogen treatment would be challenging in this cohort. Similarly, almost half of participants had no baseline atrophy-related complaints. The measurements of vaginal atrophy symptoms and objective atrophy were not validated tools, although very similar metrics have been reported in past studies examining vaginal estrogens for treatment of GSM, and a past systematic review highlighted this shortcoming in the literature stating, “there were no universal, validated tools for the objective assessment of vulvovaginal atrophy or subjective patient-reported atrophy-related symptoms.”3 More recently, Mension et al concurred that literature for assessing GSM objectively is scant; the most common assessment tools for quantifying atrophy symptom severity were simple visual analog scales, and there was disagreement about whether the Vaginal Health Index (VHI), a clinical medical evaluation instrument, was a subjective or objective measure.21 The present study may have been strengthened by the concomitant reporting of change in vaginal pH and/or change in vaginal maturation index, but vaginal pH is easily confounded by the possible presence of bacterial vaginosis.21 Nonetheless, in this study with masked examiners, the IMPROVE trial tool identified an objective, physiologic change consistent with increased estrogenization in the vagina of participants who were adherent to study cream. While this suggests that the median of 7 weeks of treatment was sufficient to effect a change in the visual assessment of the vaginal epithelium, this may not have been adequate to impact symptoms such as dryness and atrophy-related dyspareunia,22 and it remains unknown whether this was a sufficient duration to impact other pelvic floor disorder symptoms such as SUI, UUI, and urinary frequency.
Conclusions and research implications:
In conclusion, it remains undetermined whether the addition of 7 weeks of preoperative intravaginal estrogen in postmenopausal women with symptomatic pelvic organ prolapse will improve concomitant urinary complaints, sexual function and dyspareunia, or symptoms commonly attributed to atrophy. There continues to be a need for well-designed, adequately powered, randomized trials of intravaginal estrogen to provide quality evidence about its potential as adjunctive—or primary—therapy for pelvic floor disorders, particularly for patients with bothersome baseline SUI, UUI, and urinary frequency.
AJOG at a Glance:
Why was this study conducted?
To determine whether intravaginal estrogen application given preoperatively to postmenopausal women planning a prolapse repair would experience improvements in other pelvic floor disorder symptoms.
What are the key findings?
Compared to placebo application, use of intravaginal estrogen for a median of 7 weeks did not result in significant improvements in symptoms of stress or urgency urinary incontinence, urinary frequency, sexual function, dyspareunia, or vaginal atrophy symptoms.
In participants adherent to their study medication, intravaginal estrogen did result in objectively less atrophic-appearing vaginal epithelium, per examiners masked to treatment assignment.
What does this study add to what is already known?
Despite objective changes in the vaginal epithelium consistent with increased estrogen effect, it remains uncertain whether addition of intravaginal estrogen cream in postmenopausal women with symptomatic pelvic organ prolapse results in improvement in other pelvic floor symptoms.
Sources of support:
National Institute on Aging, R01 AG047290
American Board of Obstetrics & Gynecology and The American Association of Obstetricians and Gynecologists Foundation, Bridge Funding Award
Pfizer, Inc.
Neither of the two funding sources (1 & 2) nor Pfizer, Inc. (3) had any part in the study design; collection, analysis, or interpretation of data; writing of the report; or decision to submit the article for publication. Pfizer, Inc. provided Premarin vaginal cream through an investigator-initiated research award.
Footnotes
The authors report no relevant conflicts of interest.
Presented at the 49th Annual Scientific Meeting of the Society of Gynecologic Surgeons, Tucson, AZ, March 19-22, 2023
CRediT author statement:
David Rahn: conceptualization, methodology, investigation, writing (original draft), project administration, funding acquisition; Holly Richter: investigation, writing (review and editing); Vivian Sung: investigation, writing (review and editing); Linda Hynan: methodology, data curation, writing (review and editing), supervision; Jessica Pruszynski: methodology, formal analysis, data curation, writing (review and editing)
- First registered May 1, 2015;
- Initial participant enrollment: December 2016;
- Clinical trial identification number: NCT02431897;
- URL of registration site: www.ClinicalTrials.gov;
- Data Sharing Information plan: not applicable (participant enrollment began before January 1, 2019)
REFERENCES
- 1.Weber MA, Limpens J, Roovers JP. Assessment of vaginal atrophy: a review. Int Urogynecology J 2015;26(1):15–28. DOI: 10.1007/s00192-014-2464-0. [DOI] [PubMed] [Google Scholar]
- 2.Stenberg A, Heimer G, Ulmsten U. The prevalence of urogenital symptoms in postmenopausal women. Maturitas 1995;22 Suppl:S17–S20. DOI: 10.1016/0378-5122(95)00958-2. [DOI] [PubMed] [Google Scholar]
- 3.Rahn DD, Carberry C, Sanses TV, et al. Vaginal estrogen for genitourinary syndrome of menopause: a systematic review. Obstet Gynecol 2014;124(6):1147–56. DOI: 10.1097/AOG.0000000000000526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rahn DD, Ward RM, Sanses TV, et al. Vaginal estrogen use in postmenopausal women with pelvic floor disorders: systematic review and practice guidelines. Int Urogynecology J 2015;26(1):3–13. DOI: 10.1007/s00192-014-2554-z. [DOI] [PubMed] [Google Scholar]
- 5.Cody JD, Jacobs ML, Richardson K, Moehrer B, Hextall A. Oestrogen therapy for urinary incontinence in post-menopausal women. Cochrane Database Syst Rev 2012;10:CD001405. DOI: 10.1002/14651858.CD001405.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Long CY, Liu CM, Hsu SC, Chen YH, Wu CH, Tsai EM. A randomized comparative study of the effects of oral and topical estrogen therapy on the lower urinary tract of hysterectomized postmenopausal women. Fertil Steril 2006;85(1):155–60. DOI: 10.1016/j.fertnstert.2005.06.042. [DOI] [PubMed] [Google Scholar]
- 7.Rahn DD, Richter HE, Sung VW, Larsen WI, Hynan LS. Design of a randomized clinical trial of perioperative vaginal estrogen versus placebo with transvaginal native tissue apical prolapse repair (Investigation to Minimize Prolapse Recurrence Of the Vagina using Estrogen: IMPROVE). Female Pelvic Med Reconstr Surg 2021;27(1):e227–233. DOI: 10.1097/SPV.0000000000000899. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Manonai J, Theppisai U, Suthutvoravut S, Udomsubpayakul U, Chittacharoen A. The effect of estradiol vaginal tablet and conjugated estrogen cream on urogenital symptoms in postmenopausal women: a comparative study. J Obstet Gynaecol Res 2001;27(5):255–60. [DOI] [PubMed] [Google Scholar]
- 9.Rahn DD, Good MM, Roshanravan SM, et al. Effects of preoperative local estrogen in postmenopausal women with prolapse: a randomized trial. J Clin Endocrinol Metab 2014:jc20141216. DOI: 10.1210/jc.2014-1216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Barber MD, Walters MD, Bump RC. Short forms of two condition-specific quality-of-life questionnaires for women with pelvic floor disorders (PFDI-20 and PFIQ-7). Am J Obstet Gynecol 2005;193(1):103–13. DOI: 10.1016/j.ajog.2004.12.025. [DOI] [PubMed] [Google Scholar]
- 11.Rogers RG, Rockwood TH, Constantine ML, et al. A new measure of sexual function in women with pelvic floor disorders (PFD): the Pelvic Organ Prolapse/Incontinence Sexual Questionnaire, IUGA-Revised (PISQ-IR). Int Urogynecology J. DOI: 10.1007/s00192-012-2020-8. [DOI] [PubMed] [Google Scholar]
- 12.Bachmann G, Bouchard C, Hoppe D, et al. Efficacy and safety of low-dose regimens of conjugated estrogens cream administered vaginally. Menopause 2009;16(4):719–27. DOI: 10.1097/gme.0b013e3181a48c4e. [DOI] [PubMed] [Google Scholar]
- 13.Weber MA, Kleijn MH, Langendam M, Limpens J, Heineman MJ, Roovers JP. Local oestrogen for pelvic floor disorders: a systematic review. PloS One 2015;10(9):e0136265. DOI: 10.1371/journal.pone.0136265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nelken RS, Ozel BZ, Leegant AR, Felix JC, Mishell DR Jr. Randomized trial of estradiol vaginal ring versus oral oxybutynin for the treatment of overactive bladder. Menopause 2011;18(9):962–6. DOI: 10.1097/gme.0b013e3182104977. [DOI] [PubMed] [Google Scholar]
- 15.Serati M, Salvatore S, Uccella S, Cardozo L, Bolis P. Is there a synergistic effect of topical oestrogens when administered with antimuscarinics in the treatment of symptomatic detrusor overactivity? Eur Urol 2009;55(3):713–9. DOI: 10.1016/j.eururo.2008.06.051. [DOI] [PubMed] [Google Scholar]
- 16.Tseng LH, Wang AC, Chang YL, Soong YK, Lloyd LK, Ko YJ. Randomized comparison of tolterodine with vaginal estrogen cream versus tolterodine alone for the treatment of postmenopausal women with overactive bladder syndrome. Neurourol Urodyn 2009;28(1):47–51. DOI: 10.1002/nau.20583. [DOI] [PubMed] [Google Scholar]
- 17.Kobata SA, Girao MJ, Baracat EC, et al. Estrogen therapy influence on periurethral vessels in postmenopausal incontinent women using Doppler velocimetry analysis. Maturitas 2008;61(3):243–7. DOI: 10.1016/j.maturitas.2008.07.015. [DOI] [PubMed] [Google Scholar]
- 18.Ellington DR, Szychowski JM, Malek JM, Gerten KA, Burgio KL, Richter HE. Combined tolterodine and vaginal estradiol cream for overactive bladder symptoms after randomized single-therapy treatment. Female pelvic medicine & reconstructive surgery 2016;22(4):254–60. DOI: 10.1097/SPV.0000000000000256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rahn DD, Richter HE, Sung VW, Pruszynski JE, Hynan LS. Perioperative vaginal estrogen as adjunct to native tissue vaginal apical prolapse repair: a randomized controlled trial. JAMA 2023;330(7):615–625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Skorupska K, Grzybowska ME, Kubik-Komar A, Rechberger T, Miotla P. Identification of the Urogenital Distress Inventory-6 and the Incontinence Impact Questionnaire-7 cutoff scores in urinary incontinent women. Health Qual Life Outcomes 2021;19(1):87. DOI: 10.1186/s12955-021-01721-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mension E, Alonso I, Tortajada M, et al. Genitourinary Syndrome of Menopause assessment tools. J Midlife Health 2021;12(2):99–102. DOI: 10.4103/jmh.jmh_93_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pinkerton JV, Bushmaklin AG, Komm BS, Abraham L. Relationship between changes in vulvar-vaginal atrophy and changes in sexual functioning. Maturitas 2017;100:57–63. [DOI] [PubMed] [Google Scholar]