Oestrogens

StudyStudy type and ELNo. of patientsPatient characteristicsInterventionComparisonLength of follow-upOutcome measuresEffect sizeAdditional comments
Dessole 2004431RCT
EL = 1+
88Postmenopausal women, (mean ~7 years since menopause), mean age ~57 years, stress UI, vaginal atrophy, recurrent UTI
Exclusions: uterovaginal prolapse, cystocele, rectocele, (grade II or III), severe systemic disease, thromboembolic disease, biliary lithiasis, previous breast or uterine cancer, abnormal uterine bleeding, BMI ≥ 25 kg/m2
Estriol ‘ovule’ intravaginally (1 mg/day for 2 weeks, then 2 mg once /week), n = 44Placebo intravaginally (n = 44)6 months txSubjective improvement of stress UI68% vs 16%, P < 0.01Funding: none declared.
Not a blinded study.
UPP (mean changes)MUP (cmH2O): +22% vs −6%, P < 0.05
MUCP (cmH2O): +26% vs −3%, P < 0.05
Pressure transmission ratio (%): +23% vs 0, P < 0.05
UrodynamicsNo significant difference between groups in volume at 1st sensation, bladder capacity or compliance, or max. bladder pressure.
Sig. greater increase in UPP in estriol grp vs placebo
Adverse events2 from each grp withdrew owing to localised adverse reactions; vaginal irritation, burning, itching.
No systemic AE reported
Fantl 1996432DB RCT
EL = 1+
83F, hypoestrogenic, aged ≥ 45 years (mean 67), UI at least once/week. 45% stress UI, 26% DO, 29% stress UI + DO
Exclusions: institutionalisation, permanent catheterisation, cognitive impairment, functional disability, neuropathic or uncontrolled metabolic conditions, chronic UTI, reversible causes of UI
Conjugated equine oestrogen 0.625 mg /30 days + medroxyprogerterone for 10 days each cycle (n = 44)Placebo (n = 39)3 months txLeakage episodes/week (mean change)−3 vs −3%, P = NSFunding: National Institute on Aging, National Institutes of Health, Wyeth-Ayerst, Upjohn.
Sig. differences between grps at baseline in vaginal parity (3.1 vs 2.4), and use of diuretics (32% vs 26%), P < 0.05.
Fluid loss (standardised pad test; mean change, g)−20 vs −13% P = NS
Frequency/week (mean change)Diurnal: −4 vs −6%, P = NS
Nocturnal: −11% vs 0%, P = NS
QOLIIQ: −25 vs −23 (−20 vs −19%), P = NS
UDI: −12 vs −20 (−10.5 vs −16.5%), P = NS
Perception of improvement (5 pt ordinal scale)Much or somewhat better 54% vs 45%, P = NS
Jackson 1999433DB RCT
EL = 1+
67Postmenopausal F (at least 1 year) mean age 63 years, genuine stress UI (mean 1 leakage episode/day and 1/night), not taken HRT in last 12 months
Exclusions: breast, endometrial or liver cancer; endometrium > 4 mm thick
Estradiol valerate 2 mg/day (n = 33)Placebo (n = 34)6 months txBFLUTSNo sig. difference between grps in % reporting symptomsFunding: MRC. Unit at which study based funded by educational grant from Pfizer.
Progestogen not given in addition to estradiol, in order to maintain blinding of treatment. Norethisterone 5 mg for 12 days/month given if breakthrough bleeding occurred with estradiol.
*baseline values 10 vs 3 g.
SF-36No sig. difference between grps in changes in scores
1 h pad test (mean change, g)+3 vs +6 g (30% vs 200%)* P = NS
Frequency–volume chart parameters (change in median values/24 h)Frequency: +1 vs −7%
Nocturia: −10 vs +10%
Mean voided vol.: +12 vs −1%
Leaks/day: +0.4 (40%) vs 0
(P = NS for all comparisons)
Urodynamics (change in median values)1st desire to void: +2% vs +9%
Functional capacity: +11% vs +7%
MUCP increased in both grp, P = NS between grps
Urodynamic cure14% both grps
Wilson 1987434DB RCT
EL = 1+
36 randomised, 34 analysedPostmenopausal F mean age 57 years (47–72), genuine stress UI, and stable detrusor function, no HRT in past 3 monthsPiperazine estrone sulphate, 3 mg at night (n = 16)Placebo (n = 18)3 monthsSubjective assessmentMuch improved 7 vs 5, improved 5 vs 5, no better 4 vs 8; P = NSFunding: none declared.
Frequency/24 h (mean change)−16 vs −3%, P = NS
Urilos nappy test (change in mean, ml/2 h) (n = 22)+1 ml (20%) vs −2 ml (40%)
MUCP (change in mean, cmH2O)−4 (7%) vs −3 (6%), P = NS
Adverse events2 in oestrogen grp withdrew (1 palpitations and trembling, 1 posterior subendocardial infarct). ‘other minor symptoms e.g. leg pain, breast discomfort, chest pain, nausea, were uncommon’
Cardozo 1993435DB RCT
EL = 1−
64 randomised, 56 completed and analysedPostmenopausal F, ambulant, mean age ~59 years, sensory or motor urge UI* (21 had stress UI, 48 urge UI)
Exclusions: symptoms present for 3 months before the menopause, voiding difficulty, pelvic anatomical defect requiring surgery, neurological disease, recent oestrogen use, contraindications to oestrogen use
Estriol 3 mg/day (n = 34 randomised, 31 analysed)Placebo (n = 30 randomised, 25 analysed)3 months txLeakage episodes/day (change in mean)Diurnal: −21 vs −59%
Night: −81 vs −70%
Funding: none declared.
*sensory urge UI: 1st desire to void at < 150 ml and cystometric capacity < 400 ml in the absence of detrusor activity); motor urge UI: uninhibited detrusor contractions > 15 cmH2O.
No between-group statistical comparisons reported.
[EL = 1−] only completers analysed.
Cure of UIStress UI: 6/11 vs 6/10
Urge UI: 11/25 vs 7/23
Frequency (change in mean/day)Diurnal: −16 vs −32% (P < 0.05 vs baseline for both grps)
Nocturia: −25 vs −46% (P < 0.05 vs baseline for plac grp)
UrgencyChange on severity scale of 0–3: −1.1 vs −1.1
Cure: 7/29 vs 9/25 (P < 0.05 vs baseline for both grps)
1st desire to void (change in mean)Bladder volume at:
+28 vs +24%
Detrusor pressure at::
0 vs +29%
Cystometric capacity (change in mean)Bladder volume:
+2 vs +13%
Detrusor pressure:
−19 vs +52%
Adverse effectsNone
Lose 2000436RCT
EL = 1+
251Postmenopausal F ( > 2 years), mean age 66 years, at least 1 bothersome lower urinary tract symptom. 23% had cystocele. 70% had urgency, 59% frequency, 52% urge UI, 50% stress UI, 50% nocturia, 19% dysuria
Exclusions: oestrogen-dependent neoplasia, or breast, ovarian or endometrial cancer, undiagnosed vaginal bleeding, liver disease, porphyria, uterovaginal prolapse grade II or III, sex hormone tx in last 6 months
Estradiol-releasing ring, 7.5 mg/24 h (n = 134)Estriol pessaries 0.5 mg (1/day for 3 weeks, then every other day), n = 1176 months txUrgencyResponder rate 51% vs 56%
Cure 27% vs 33%
Funding: none declared. Second author medical director of Pharmacia and Upjohn.
No sig. difference identified between grps in any outcome.
Responder rate not defined; assumed to be ≥ 50% reduction in symptoms.
Not a blinded study.
FrequencyResponder rate 61% vs 58%
Cure 34% vs 44%
Urge UIResponder rate 58% vs 58%
Cure 33% vs 34%
Stress UIResponder rate 53% vs 59%
Cure 34% vs 41%
NocturiaResponder rate 51% vs 54%
Cure 31% vs 35%
Improvement of symptoms (VAS)+21.1 mm vs +23.2 mm
Adverse effects‘most’, including vaginal discomfort/discharge and breast pain were mild and transient
3.7 vs 2.6% withdrew owing to an adverse event
Ouslander 2001437RCT
EL = 1−
32 randomised, 21 completed txF nursing home residents aged ≥ 65 years (mean 88), UI at least dailyOestrogen 0.625 mg + medroxyprogesterone acetate 2.5 mg (n = 15 randomised, 9 completed)Placebo (n = 17 randomised, 12 completed)6 months txWet rate (% checks wet during 3 days prompted voiding); change in mean−10 vs −11%, P = NSFunding: National Institute on Aging. Materials supplied by Wyeth-Ayerest.
F also received prompted voiding when wet checks (primary outcome) were carried out (3 days for 8 h).
[EL = 1−] owing to high drop out rate.
Not a blinded study
Appropriate toileting rate (continent voids/total), change in mean+20 vs +23%, P = NS
Incontinent volume (change in mean)−1 vs −20%, P = NS
Bladder capacity (change in mean, ml)−5 vs +4%, P = NS
Adverse effectsActive tx grp: 2 vaginal spotting; ~10% breast tenderness
Rufford 2003438DB RCT
EL = 1+
40F with ‘urge syndrome’ (29 urge UI, 17 stress UI), age not stated, postmenopausal ( > 1 year) or estradiol < 150 pmol/l if had hysterectomy. 2 had colposuspension
Exclusions: any medication for urge syndrome, diabetes mellitus or insipidus, diuretics, HRT within 3 months or hormone implant /IM hormone injection within 1 year, endometrial thickness > 4 mm in women with intact uterus, UTI, haematuria, pelvic masses or urogenital prolapse
17-beta estradiol subcutaneous implant (n = 20)Placebo implant (n = 20)6 months txSelf-reported cure35% vs 30% urge UI
20% vs 15% stress UI
15% vs 10% urgency
Funding: Organon.
At final visit all women with intact uterus given norethisterone 5 mg/day for 2 weeks, repeated until no further vaginal bleeding.
Number of women recruited did not reach the target set in the power calculation.
No significant difference identified between groups for any outcome.
*first sensation, maximum capacity, pressure rise on filling, volume pressure > 15 cmH2O.
Leakage episodes (median, IQR)0 (0, 1.8) vs 0 (0, 0.5)
Frequency/24 h (median, IQR)8.6 (6.5, 11.4) vs 8.0 (7.0, 9.8)
Volume voided (median ml, IQR)177 (143, 209) vs 161 (107, 200)
Urodynamics* (at 3 months)No between-group analysis
Adverse effects (n)Estradiol; 9/12 women with intact uterus had irregular bleeding (5 requiring hysterectomy); 1 angina
Other; UTI (8 vs 11), breast tenderness (4 vs 1)
Simunic 2003439DB RCT
EL = 1+
1612Postmenopausal F ( > 1 year) with urogenital symptoms (28% had UI, 43% frequency/nocturia)
Exclusions: oestrogen tx within 6 months, systemic disease or infection, suspected or proven malignant disease, unexplained uterine bleeding, hysterectomy or surgery for stress UI, acute urogynaecological infection
17-beta estradiol intravaginal tablet (n = 828; 371 with frequency/nocturia, 245 UI)Placebo (n = 784; 315 with frequency/nocturia, 206 UI)1 year txFrequency/nocturia prevalence (mean change)−38% (P = 0.001 vs baseline) vs −10.1%Funding: none declared.
Treatment given daily for 2 weeks then twice a month for 12 months. Tx interrupted for 1 month every 4 months.
UI prevalence (mean change)−17.8% (P = 0.002 vs baseline) vs −10.2%
CystometryMax. cystometric capacity, bladder volume at 1st or strong sensation to void sig. increased in estradiol grp (P ≤ 0.05 from baseline); no sig. change in plac grp
Adverse effectsIncreased vaginal discharge 2.7 vs 0.4%
vaginal bleeding 0.6% vs 0%
erythema 0.8 vs 0.1%
itching 0.5 vs 0.1%
UTI 0.4 vs 0.1%
labial oedema 0.3 vs 0.1%
‘other’ 2.4% vs 0%
Walter 1978440DB RCT
EL = 1−
29Postmenopausal F mean age 56 years (46–69), UI (21 frequency, urgency and urge UI, 29 stress UI), no detrusor hyperreflexiaEstradiol 2 mg + estriol 1 mg orally for 20 days, then 8 day break/cycle (n = 15)Placebo (n = 14)4 months txChange in frequency, urgency, urge UICure 7/11 vs 1/10Funding: none declared.
[EL = 1−] No baseline data reported
Aim of study was to evaluate effects of oestrogen on symptoms related to vaginal atrophy and urge UI /MUCP
*baseline values not given
MUCP (29 pts with stress UI)Mean change (cmH2O): +4.6 vs +0.17, P = NS*
Adverse effects‘no subjective adverse effects’
Samsioe 1985441DB RCT, cross-over
EL = 1−
34Postmenopausal F aged 77–78 years, (32% stress UI, 41% urge UI, 26% mixed)Estriol 3 mg orally (n = 34)Placebo (n = 34)Unclear whether 3 or 6 months txIncontinenceUrge UI: two thirds reported symptoms alleviated
Mixed UI: 6/8 reported symptoms ‘ameliorated’
Funding: AB Leo supplied tablets.
[EL = 1−] Duration of tx unclear (3 months or 2×3 months).
Main aim of study was to evaluate effects of oestrogen on symptoms related to vaginal atrophy.
Molander 1990442DB RCT
EL = 1+
35F with urogenital oestrogen deficiency syndrome including urinary frequency and dysuriaEstriol 3 mg/day for 4 weeks then 2 mg/day for 6 weeks (n = 18)Placebo (n = 17)10 weeks txSeverity of urinary frequency on 4 point scaleNo change from 2.5 in estriol grp; change from 2.4 to 2.3 in placebo grpFunding: none declared.
Main aim of study was to assess effects of tx on vaginal bacterial flora, cytology, and urogenital symptoms.
Eriksen 1992443DB RCT
EL = 1+
164 randomised, 154 completed and analysedF mean age 58 years (45–70) with atrophic vaginitis; 53% in estradiol grp and 41% placebo grp had urological symptoms
Exclusions: history of cancer or thromboembolism, unexplained vaginal bleeding
17-beta estradiol intravaginal tablet, 25 μg (n = 81 randomised, 75 analysed)Placebo (n = 83 randomised, 79 analysed)3 months txImprovement in urological symptoms (frequency, dysuria, urge or stress UI)63% vs 32% P < 0.001Funding: none declared.
Daily dose given for 2 weeks then 2×/week.
10 withdrew: 6 estradiol, 4 placebo.
Main aim of study was to evaluate effects of oestrogen on symptoms related to vaginal atrophy.
Adverse effectsEstradiol (7 reports); local effects (alopecia, slight pain, rash, smell); aggravation of psoriasis, dizziness, GI pain
Placebo (11 reports): local effects (itching, burning, eczema, discharge); palpitations, sweats, leg itching, weight gain (2), leg cramp, pain
Grady 2001444 (Analysis of data from HERS study#)445DB RCT
EL = 1++
1,525 (55% of the HERS study population)Postmenopausal F < 80 years with established coronary heart disease, not had a hysterectomy; included in the HERS study who had at least 1 episode incontinence weekly (23% stress UI, 51% mixed, 26% urge). 78% had ≥ 2 UI episodes/week, 23% had ≥ 7 episodes/week; mean 5.6 (SD 9.1)Conjugated oestrogen 0.625 mg with medroxyprogesterone acetate 2.5 mg (n = 768)Placebo (n = 757)Mean duration of tx 4.1 years% markedly improved or improved*20.9% vs 26%, P = 0.001Funding: Wyeth-Ayerst.
#aim of HERS study was to determine whether oestrogen + progesterone alters the risk of coronary events in postmenopausal women with established coronary heart disease.
*markedly improved: reduction of ≥ 5 UI episodes/week;
improved: reduction of 2–4 UI episodes/week;
unchanged: no greater than change of 1 UI episode/week;
worsened: increase of 2–4 UI episodes/week;
markedly worsened: increase of ≥ 5 UI episodes/week.
Compliance: ≥ 80% of study medication taken by 82% vs 88% at 1 year, and 69% vs 74% at 4 years.
6% vs 4% took tx for UI during the study, and 1.3 vs 0.9% had continence surgery.
% worsened or markedly worsened38.8% vs 27%, P = 0.001 (OR for worsening by > 1 category 1.51 [95% CI 1.26 to 1.82])
Leakage episodes /week (change in mean)+0.7 vs −0.1, P < 0.001
Frequency‘essentially unchanged’; no numerical data
Total HERS study population445,945(n = 1380)(n = 1383)Mean duration of tx 4.1 yearsAdverse effectsConfirmed venous thromboembolism: 6.3 vs 2.2 per 1000 woman/years (RH 2.89, 95% CI 1.50 to 5.58), P = 0.002
Gallbladder disease, 6 vs 4.4% (RH 1.38, 95% CI 1.00 to 1.92), P = 0.05
No significant differences between grps in the risk of breast or endometrial or other cancer, or fracture
Steinauer 2005450As Grady4441208F from the HERS study who did not report UI at baseline (any episodes of stress or urge UI in the last week)Conjugated oestrogen 0.625 mg with medroxyprogesterone acetate (n = 597)Placebo (n = 611)As Grady444% pts reporting weekly UI episode at least once (Odds Ratio, 95% CI)Any type of UI:
64% vs 49%, P ≤ 0.01 (OR 1.6, 95% CI 1.3 to 1.9)
Urge UI:
48% vs 36%, P < 0.001 (OR 1.5 to 95% CI 1.2 to 1.8)
Stress UI:
54% vs 38%, P < 0.001 (OR 1.7, 95% CI 1.5 to 2.1)
Funding: Wyeth-Ayerst.
Women in placebo grp were on average 1 year older than HRT grp.
Hendrix 2005446 (Analysis of data from the Women’s Health Initiative [WHI] study#)447,448DB RCT
EL = 1++
23,296 who had baseline and 1 year UI data (85% of WHI study population)Postmenopausal F aged 50–79 years. Across tx grps, 34.1–36.9% had no UI; 24–26.7% stress UI, 21.9–24.2% urge UI, 15–17.9% mixed UI, 0.2–0.5% UI only at night. Of those with UI, leakage episodes < 1/month 30.9%, ≥ 1/month but < 1/week 26.4%, ≥ 1/week but < 1 day 27.9%, daily 14.7%
Exclusions: breast cancer, other invasive cancer in last 10 years, venous thromboembolism, hypertriglyceridaemia, severe menopausal symptoms at end of 3 month HRT washout period prior to start of study
Conjugated equine oestrogen (CEE) 0.625 mg + medroxyprogesterone acetate (MPA) 2.5 mg (n = 7247)Placebo (n = 7056)1 year tx and follow-upIncident UIAny: 11.5 vs 7.9%, RR 1.39 (95% CI 1.27 to 1.52)
Stress UI: 5.9 vs 3.1%, RR 1.87 (95% CI 1.61 to 2.18)
Mixed UI: 1.3 vs 0.1%, RR 1.49 (95% CI 1.10 to 2.01)
Urge UI: 4.2 vs 3.9%, RR 1.15 (95% CI 0.99 to 1.34)
Funding: National Heart Lung and Blood Institute. Wyeth-Ayerst Research provided study medication.
#aim of WHI study was to assess effects of HRT on coronary heart disease, and other risks and benefits. Those with an intact uterus randomised to CEE + MPA or placebo; those who had hysterectomy randomised to CEE or placebo. Both studies were stopped early; CEE+MPA vs placebo at a mean of 5.6 years owing to more harm than benefit, and CEE vs placebo study at mean 7.1 years owing to increased risk of stroke and no benefit for CHD.
Compliance at 1 year: ≥ 80% of study medication taken by 74% vs 81% in CEE+MPA vs placebo arm, and by 77.4 vs 81.4% in CEE vs placebo arm. % who stopped taking medication: 9.7 vs 6.6%, and 8.4% vs 8%.
3 year data published for 8.6% of the 23,296; not reproduced here.
CHD = coronary heart disease.
Worsening of prevalent UI
RR (95% CI)
Quantity of leakage: 1.20 (1.06, 1.36)
Leakage episodes 1.38 (1.28, 1.49)
Limitations in daily activities related to UI 1.18 (1.06, 1.32)
Bother factor 1.22 (1.13, 1.32)
Conjugated equine oestrogen (CEE) 0.625 mg (n = 4476)Placebo (n = 4517)1 year tx and follow-upIncident UIAny: 12.4 vs 8.1%, RR 1.53 (95% CI 1.37 to 1.71)
Stress UI: 5.9 vs 2.9%, RR 2.15 (95% CI 1.77 to 2.62)
Mixed UI: 1.7 vs 1.1%, RR 1.79 (95% CI 1.26 to 2.53)
Urge UI: 4.7 vs 4.0%, RR 1.32 (95% CI 1.10 to 1.58)
Worsening of prevalent UI
RR (95% CI)
Quantity of leakage: 1.59 (1.39, 1.82)
Leakage episodes 1.47 (1.35, 1.61)
Limitations in daily activities related to UI 1.29 (1.15, 1.45)
Bother factor 1.50 (1.37, 1.65)
Total population447,448CEE + MPA (n = 8506)Placebo (n = 8102)Mean 5.2 years follow-upAdverse effects (annualised %, i.e. events per 10,000 person years [hazard ratio, 95% CI])CHD 0.37 vs 0.30 (HR 1.29, 95% CI 1.02 to 1.63)
Stroke 0.29 vs 0.21 (HR 1.41, 95% CI 1.07 to 1.85)
Venous thromboembolism 0.34 vs 0.16 (HR 2.11, 95% CI 1.58 to 2.82)
Invasive breast cancer 0.38 vs 0.30 (HR 1.26, 95% CI 1.0 to 1.59)
CEE (n = 5310)Placebo (n = 5429)Mean 6.8 years follow-upAdverse effectsCHD 0.49 vs 0.54 (HR 0.91, 95% CI 0.75 to 1.12)
Stroke 0.44 vs 0.32 (HR 1.39, 95% CI 1.10 to 1.77)
Venous thromboembolism 0.28 vs 0.21 (HR 1.33, 95% CI 0.99 to 1.79)
Invasive breast cancer 0.26 vs 0.33 (HR 0.77, 95% CI 0.59 to 1.01)
Goldstein 2005449RCT
EL = 1+
619F enrolled in a placebo-controlled study evaluating raloxifene and oestrogen for osteoporosis prevention in postmenopausal women
F had prior hysterectomy, mean age 53 years. Prevalence of UI at baseline: 4% CEE grp, 3% raloxifene 60 mg, 3% raloxifene 150 mg, 6% placebo
Conjugated equine oestrogen 0.625 mg o.d. (n = 158)Raloxifene 60 mg o.d. (n = 152)
Raloxifene 150 mg o.d. (n = 157)
Placebo (n = 152)
3 yearsNew or worsening UI7 vs 0.7 vs 0.6 vs 1.3%, P < 0.02 for CEE grp vs othersFunding: Eli Lilly and Co.
60% of pts still taking study medication at 3 years. No difference between grps in reasons for discontinuing tx.
% patients with improvement in existing UI (% of 4 vs 3 vs 3% vs 6%)50 vs 100 vs 100% vs 89%

From: Evidence tables for included studies

Cover of Urinary Incontinence
Urinary Incontinence: The Management of Urinary Incontinence in Women.
NICE Clinical Guidelines, No. 40.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2006 Oct.
Copyright © 2006, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.