Oestrogen therapy for treating pelvic organ prolapse in postmenopausal women

Abstract

Background

Pelvic organ prolapse (POP) is the descent of a woman's uterus, bladder, or rectum into the vagina. It affects 50% of women over 50 years old who have given birth to at least one child, and recognised risk factors are older age, higher number of births, and higher body mass index. This review assesses the effects of oestrogen therapy, alone or in combination with other treatments, on POP in postmenopausal women.

Objectives

To assess the benefits and harms of local and systemic oestrogen therapy in the management of pelvic organ prolapse symptoms in postmenopausal women, and to summarise the principal findings of relevant economic evaluations.

Search methods

We searched the Cochrane Incontinence Specialised Register (up to 20 June 2022), which includes CENTRAL, MEDLINE, two trials registers, and handsearching of journals and conference proceedings. We also checked the reference lists of relevant articles for additional studies.

Selection criteria

We included randomised controlled trials (RCTs), quasi‐RCTs, multi‐arm RCTs, and cross‐over RCTs that evaluated the effects of oestrogen therapy (alone or in combination with other treatments) versus placebo, no treatment, or other interventions in postmenopausal women with any grade of POP.

Data collection and analysis

Two review authors independently extracted data from the included trials using prespecified outcome measures and a piloted extraction form. The same review authors independently assessed the risk of bias of eligible trials using Cochrane's risk of bias tool. Had data allowed, we would have created summary of findings tables for our main outcome measures and assessed the certainty of the evidence using GRADE.

Main results

We identified 14 studies including a total of 1002 women. In general, studies were at high risk of bias in terms of blinding of participants and personnel, and there were also some concerns about selective reporting. Owing to insufficient data for the outcomes of interest, we were unable to perform our planned subgroup analyses (systemic versus topical oestrogen, parous versus nulliparous women, women with versus without a uterus).

No studies assessed the effects of oestrogen therapy alone versus no treatment, placebo, pelvic floor muscle training, devices such as vaginal pessaries, or surgery. However, we did identify three studies that assessed oestrogen therapy in conjunction with vaginal pessaries versus vaginal pessaries alone and 11 studies that assessed oestrogen therapy in conjunction with surgery versus surgery alone.

Authors' conclusions

There was insufficient evidence from RCTs to draw any solid conclusions on the benefits or harms of oestrogen therapy for managing POP symptoms in postmenopausal women. Topical oestrogen in conjunction with pessaries was associated with fewer adverse vaginal events compared with pessaries alone, and topical oestrogen in conjunction with surgery was associated with reduced postoperative urinary tract infections compared with surgery alone; however, these findings should be interpreted with caution, as the studies that contributed data varied substantially in their design.

There is a need for larger studies on the effectiveness and cost‐effectiveness of oestrogen therapy, used alone or in conjunction with pelvic floor muscle training, vaginal pessaries, or surgery, for the management of POP. These studies should measure outcomes in the medium and long term.

Plain language summary

Does oestrogen therapy help to treat pelvic organ prolapse in postmenopausal women?

Key messages

• Although 14 studies met the inclusion criteria for this review, no studies addressed our comparisons of greatest interest, so we are uncertain about the effects of oestrogen therapy on pelvic organ prolapse in postmenopausal women.

• Further research is needed to assess the effects of oestrogen therapy, alone or in combination with other treatments, on pelvic organ prolapse in postmenopausal women.

What is pelvic organ prolapse?

Pelvic organ prolapse is when a woman's uterus (womb), bladder, or rectum moves down from its normal position into the vagina. It is a common condition, affecting 50% of women over 50 years old who have had at least one child. Between 6% and 12% of women who have had a hysterectomy (surgical removal of the uterus) will experience pelvic organ prolapse. Women who are older, who have had more children, and who are overweight are more likely to have pelvic organ prolapse. Women with pelvic organ prolapse may have the feeling of "something coming down" into the vagina, or other symptoms that can negatively affect their quality of life and body image, such as discomfort during sex or urinary problems.

What did we want to find out?

Many clinicians prescribe oestrogen therapy (a type of hormonal therapy) to treat pelvic organ prolapse, sometimes in combination with other treatments such as a pessary (a device inserted into the vagina to provide support) or surgery. However, the benefits of this approach are unclear. We wanted to find out whether using oestrogen therapy, either alone or together with other treatments, could improve the symptoms of pelvic organ prolapse in women after their menopause.

What did we do?

We searched for studies that investigated the effects of oestrogen therapy, either alone or in combination with other treatments such as pessaries or surgery, on postmenopausal women. We compared and summarised the results of these studies and rated our confidence in the results based on aspects such as methods and number of participants.

What did we find?

We found 14 studies involving a total of 1002 women. Ten studies recruited women with different severities of pelvic organ prolapse. There were differences across studies in terms of the location of pelvic organ prolapse, the number of children the women had had, whether the women had had a hysterectomy, and the type of oestrogen therapy being investigated.

Main results

Although we identified 14 eligible studies, no studies addressed our main comparisons of interest (oestrogen therapy alone compared with no treatment, compared with pelvic floor muscle exercises, compared with surgery, or compared with devices such as vaginal pessaries). Four studies evaluated oestrogen therapy alongside vaginal pessaries compared with vaginal pessaries alone, and 10 studies evaluated oestrogen therapy alongside surgery compared with surgery alone.

What are the limitations of the evidence?

The evidence is very uncertain because of concerns about the studies' methods. The women were often aware of the treatments they were receiving, which may have affected the results. Additionally, many studies included few women.

How up to date is this evidence?

The evidence is up to date to 20 June 2022.

Summary of findings

Summary of findings 1. Oestrogen therapy alone versus no treatment or placebo.

Oestrogen therapy alone versus no treatment or placebo
Patient or population: women with POP Setting: primary or secondary care Intervention: oestrogen therapy Comparison: no treatment or placebo
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with oestrogens	Risk with no treatment or placebo
Subjective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Improvement or cure of associated sexual symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Objective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Generic quality of life measures Follow‐up: 12 months	—	—	—	—	—	Not reported
Participant adherence to intervention Follow‐up: various time points	—	—	—	—	—	Not reported
Minor adverse effects: vaginal events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
Major adverse events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; POP: pelvic organ prolapse.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Summary of findings 2. Oestrogen therapy alone versus pelvic floor muscle training.

Oestrogen therapy alone versus pelvic floor muscle training
Patient or population: women with POP Setting: primary or secondary care Intervention: oestrogen therapy Comparison: pelvic floor muscle training
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with oestrogens	Risk with PFMT
Subjective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Improvement or cure of associated sexual symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Objective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Generic quality of life measures Follow‐up: 12 months	—	—	—	—	—	Not reported
Participant adherence to intervention Follow‐up: various time points	—	—	—	—	—	Not reported
Minor adverse effects: vaginal events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
Major adverse events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; PFMT: pelvic floor muscle training; POP: pelvic organ prolapse.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Summary of findings 3. Oestrogen therapy alone versus devices such as vaginal pessaries.

Oestrogen therapy alone versus devices such as vaginal pessaries
Patient or population: women with POP Setting: primary or secondary care Intervention: oestrogen therapy Comparison: devices such as vaginal pessaries
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with oestrogens	Risk with no treatment or placebo
Subjective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Improvement or cure of associated sexual symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Objective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Generic quality of life measures Follow‐up: 12 months	—	—	—	—	—	Not reported
Participant adherence to intervention Follow‐up: various time points	—	—	—	—	—	Not reported
Minor adverse effects: vaginal events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
Major adverse events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; POP: pelvic organ prolapse.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Summary of findings 4. Oestrogen therapy alone versus surgery.

Oestrogen therapy alone versus surgery
Patient or population: women with POP Setting: primary or secondary care Intervention: oestrogen therapy Comparison: surgery
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with oestrogens	Risk with no treatment or placebo
Subjective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Improvement or cure of associated sexual symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Objective improvement or cure of POP symptoms Follow‐up: 12 months	—	—	—	—	—	Not reported
Generic quality of life measures Follow‐up: 12 months	—	—	—	—	—	Not reported
Participant adherence to intervention Follow‐up: various time points	—	—	—	—	—	Not reported
Minor adverse effects: vaginal events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
Major adverse events Follow‐up: longest time point reported	—	—	—	—	—	Not reported
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; POP: pelvic organ prolapse.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Background

For a glossary of terms, see Appendix 1.

Description of the condition

Pelvic organ prolapse (POP) is common, affecting 50% of women aged over 50 years who have given birth to at least one child (DeLancey 1993). POP is defined by the International Continence Society (ICS) and the International Urogynecological Association (IUGA) as the "descent of one or more of the anterior vaginal wall, posterior vaginal wall, the uterus (cervix) or the apex of the vagina [...correlated with] relevant POP symptoms" (Haylen 2016). Symptoms can include a bothersome vaginal lump or bulge, and can impact on urinary, bowel, and sexual health. Research has shown that the condition adversely impacts women's quality of life and body image (Jelovsek 2006). There are a number of recognised risk factors for POP, including older age, increasing number of births (parity), higher body mass index (BMI), and vaginal delivery (Vergeldt 2015). Age is a particularly important risk factor for POP; both the incidence and prevalence of the condition increase with age (Salvatore 2017). Additionally, 6% to 12% of women who have undergone a hysterectomy will develop prolapse (Maher 2017). Nonetheless, the aetiology is often multifactorial (Dietz 2008).

As well as affecting women's quality of life, POP has considerable resource implications that impact the health care system. The estimated annual cost of managing POP in 2005 was over EUR 144.2 million in Germany, more than EUR 83 million in France, and just over EUR 81 million in England (Subramanian 2009).

Description of the intervention

There are various conservative and surgical options for treating POP (Hagen 2011; Mowat 2018). When deciding on the most appropriate option, factors to consider include the woman's preference, age, comorbidities, lifestyle factors, type of prolapse, and previous pelvic floor surgery (NICE 2019). Conservative management includes lifestyle modifications (such as weight loss, minimising heavy lifting, and preventing or treating constipation; NICE 2019), pelvic floor muscle training (PFMT) for mild or moderate POP (Hagen 2011, NICE 2019), and vaginal pessaries (small mechanical devices placed in the vagina to provide support; Bugge 2020).

Clinicians commonly prescribe oestrogen‐only hormone replacement therapy (or oestrogen therapy) to treat POP. This therapy consists of either natural or synthetic oestrogens, and can be administered either systemically, through oral tablets, transdermal skin patches, or subcutaneous implants; or locally, through vaginal creams, tablets, or rings or pessaries impregnated with oestrogen (Stevenson 2009). Locally applied oestrogens result in less systemic absorption and hence a lower incidence of systemic side effects. When administered for the treatment of vaginal atrophy, there is no evidence of a difference between the various topical preparations with regard to efficacy or side‐effect profile (Lethaby 2016).

Clinicians can prescribe oestrogen therapy as a standalone option or in conjunction with other treatments. Oestrogen therapy alongside PFMT can strengthen the vaginal wall, improve sensation (North American Menopause Society 2007), and possibly improve pelvic floor muscle function (Sipilä 2001). When used with vaginal pessaries, oestrogen therapy can help prevent or treat vaginal atrophy (a thinning, drying, and inflammation of the vaginal walls that can occur if a woman's body has less oestrogen) before pessary insertion (NICE 2019). If conservative treatments for POP fail, or if a woman declines these treatments, she may be offered surgery (NICE 2019). Surgical methods include anterior and posterior vaginal wall repair, hysterectomy or hysteropexy for uterine prolapse, and sacrospinous fixation or sacrocolpopexy for apical prolapse (Maher 2019). Surgery has a 30% POP recurrence rate (Olsen 1997), and an estimated 11.5% of women who have POP surgery will have repeat POP surgery within 15 to 20 years (Løwenstein 2018). Repeat surgery most frequently involves the anterior compartment (cystocele). Some of these surgeries use native tissue and others involve the insertion of mesh. However, health authorities in several countries have suspended procedures involving vaginal mesh owing to concerns over long‐term complications. Like any operation, POP surgical procedures involve a risk of anaesthetic, thrombotic, cardiovascular, and surgical complications, as well as failure and recurrence (NICE 2019). One key objective of prescribing preoperative local oestrogen is to improve vaginal tissues for surgery to avoid tearing or reduce infection. One feasibility study published in 2020 showed that a randomised controlled trial (RCT) designed to evaluate the efficacy of this approach would be feasible, with high fidelity (Verghese 2020).

How the intervention might work

The female genital tract is sensitive to oestrogen. After the menopause, there is a reduction in circulating oestrogen, which leads to atrophy, loss of vaginal folds (rugae), vaginal dryness, weakening and thinning of vaginal epithelium and pelvic‐supporting fascia and ligaments, and changes in vaginal pH (DeLancey 2002; Naumova 2018; Whiteside 2005). Oestrogen therapy can improve many of these issues (Naumova 2018). The exact mechanism of oestrogen therapy in the treatment of POP is unclear, but experts believe that collagen degradation in the pelvic floor connective tissue decreases in the presence of oestrogen (Bodner‐Adler 2020).

When used with other therapies, oestrogen could improve their efficacy to reduce POP symptoms and also minimise their associated side effects by optimising the strength of pelvic organ supporting ligaments, muscles, and the vaginal epithelium. For example, thinner vaginal tissues are more susceptible to trauma from the friction of a pessary, and locally applied oestrogen could improve the elasticity, strength, and moistness of the vaginal wall and reduce pessary‐related complications (Lindahl 2014).

Why it is important to do this review

The incidence of POP will likely increase as the population continues to age (Vergeldt 2015). Therefore, evidence is needed to guide clinicians in optimising the management of postmenopausal women with POP. Many clinicians prescribe oestrogens in daily clinical practice to treat POP, sometimes in conjunction with other therapies such as pessaries or surgery, without clear supporting evidence. With this review, we aimed to synthesise information from RCTs to confirm the appropriateness of current practice, ensure that the management of POP with oestrogens is evidence‐based, and highlight the potential need for future research.

Objectives

To assess the benefits and harms of local and systemic oestrogen therapy in the management of pelvic organ prolapse symptoms in postmenopausal women, and to summarise the principal findings of relevant economic evaluations.

Methods

Criteria for considering studies for this review

Types of studies

We included RCTs and quasi‐RCTs (studies where allocation to groups is not strictly random, e.g. by day of the week), including multi‐arm and cross‐over trials.

Types of participants

We included studies that enrolled postmenopausal women (at least one year since last menses) with any grade or type of POP. We considered trials with a mix of postmenopausal and premenopausal women if there were separate data for postmenopausal women.

Types of interventions

We included studies that used any compound containing natural or synthetic oestrogen in any form or dose, for any length of time. We had planned to analyse systemic and local routes of administration via subgroup analyses.

We had planned to prepare summary of findings tables for the following comparisons.

• Oestrogen therapy alone versus no treatment or placebo

• Oestrogen therapy alone versus PFMT

• Oestrogen therapy alone versus devices such as vaginal pessaries

• Oestrogen therapy alone versus surgery

We were also interested in analysing the following comparisons.

• Oestrogen therapy in conjunction with PFMT versus PFMT alone

• Oestrogen therapy in conjunction with vaginal pessaries versus vaginal pessaries alone

• Oestrogen therapy in conjunction with surgery versus surgery alone

Types of outcome measures

Appendix 2 presents the definitions of abbreviations for outcome measurement tools.

Primary outcomes

• Participant‐reported subjective improvement or cure of POP symptoms, measured with tools such as a visual analogue scale (VAS; Ulrich 2014) or the Patient Global Impression of Improvement (PGI‐I; Srikrishna 2010)

Secondary outcomes

• Improvement or cure of associated urinary symptoms, measured by validated questionnaires such as the King's Health Questionnaire (KHQ; Kelleher 1997), the International Consultation on Incontinence Questionnaire‐Urinary Incontinence Short Form (ICIQ‐UI‐SF; Avery 2004), or the International Consultation on Incontinence Questionnaire Female Lower Urinary Tract Symptoms Modules (ICIQ‐FLUTS; Brookes 2004)

• Improvement or cure of associated bowel symptoms, measured by validated questionnaires such as KHQ (Kelleher 1997)

• Improvement or cure of associated sexual symptoms, measured by validated questionnaires such as the Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire, International Urogynecological Association‐Revised (PISQ‐IR; Constantine 2017)

• Objective improvement or cure of POP symptoms, measured by validated instruments such as the Pelvic Organ Prolapse Symptom Score (POP‐SS; Hagen 2009), the International Consultation on Incontinence Questionnaire Vaginal Symptoms Module (ICIQ‐VS; Price 2006), the Pelvic Organ Prolapse Quantification System (POP‐Q; Bump 1996), or the Australian Pelvic Floor Questionnaire (APFQ; Baessler 2009)

• Condition‐specific quality of life, measured by prolapse‐specific quality of life questionnaires such as Pelvic Floor Distress Inventory (PFDI; Barber 2001), the Pelvic Floor Impact Questionnaire (PFIQ; Barber 2001), the Prolapse Quality of Life Questionnaire (P‐QOL; Digesu 2005), or the electronic Personal Assessment Questionnaire Pelvic Floor (ePAQ‐PF; Radley 2006)

• Generic quality of life or health status, measured by generic quality of life questionnaires such as the Quality of Life Inventory (QOLI; Frisch 2009), the Global Quality of Life Scale (GQOL; Hyland 1996), or the Assessment of Quality of Life (AQoL; Hawthorne 1999)

• Psychological or emotional well‐being outcome measures and ability to cope with daily activities, measured by generic quality of life questionnaires such as the Quality of Life Questionnaire (QLQ; Evans 1985)

• Participant adherence to the intervention, including discontinuation of pessary use and declining surgical procedures

• Reduction in minor adverse events

  • Urinary tract infection (UTI)

  • Vaginal events: vaginal itching, abrasion, ulcer, vaginitis, rash, bleeding, pain, vaginosis

  • Urinary storage events: urinary retention, de novo urinary incontinence

  • Sexual adverse events: dyspareunia or partner discomfort during intercourse

• Major adverse events

  • Endometrial hyperplasia

  • Thromboembolism

  • Grade II Clavien‐Dindo complications (e.g. surgical site infection, haemorrhage, blood loss)

• Surgical outcomes

  • Operative time

  • Mesh exposure

Timing of outcome assessment

We intended to assess outcomes at six months, 12 months, five years, and 10 years, except minor adverse events, major adverse events, and surgical outcomes, which we measured at the latest time point reported by study authors.

Minimum important difference for outcomes

We determined the reported minimum important difference (MID) based on specific symptom and quality of life questionnaires. Some reported MIDs relate to surgical and non‐surgical management of prolapse. We looked for reported MIDs in the effect of oestrogen(s), either alone or alongside another treatment, or other treatment for POP. For example, a relevant MID would be the desired change resulting from the administration of local oestrogen cream either alongside pessary use for POP, or before and after surgery for POP. We found the following MIDs for various outcome measurement tools.

• ICIQ‐UI‐SF: a reduction of four points following non‐surgical intervention (Lim 2009)

• KHQ: a change from baseline of five points following an intervention (Kelleher 2004)

• POP‐SS: a reduction of 1.5 points or more (Hagen 2010)

• PFDI: an improvement of eight points in the stress subscale for non‐surgical treatments (Barber 2009)

• PFIQ: an improvement of 16 points in the urinary impact questionnaire for non‐surgical treatment (Barber 2009)

• AQOL: 0.06 utility points over time (Hawthorne 2005)

• APFQ: one point after POP surgery (Baessler 2019)

• PISQ‐IR: 0.31 following surgery for POP (Pruijssers 2021)

• PISQ: an improvement of six points (Mamik 2014)

We found no MIDs for PGI‐I, ICIQ‐FLUTS, PISQ‐IR, ICIQ‐VS, P‐QOL, QOLI/GQOL, or QLQ. Available MID data for ePAQ‐PF are for surgical treatment of urinary incontinence only.

Main outcomes for summary of findings tables

• Subjective improvement or cure of POP symptoms (at 12 months)

• Improvement or cure of associated sexual symptoms (at 12 months)

• Objective improvement or cure of POP symptoms (at 12 months)

• Generic quality of life measures (at 12 months)

• Participant adherence to intervention (at 12 months)

• Minor adverse effects: vaginal events (at latest time point reported)

• Major adverse events (at latest time point reported)

Search methods for identification of studies

When searching for eligible studies, we imposed no restrictions on language, publication date, or publication status.

Electronic searches

Search for clinical effectiveness studies

We identified relevant trials from the Cochrane Incontinence Specialised Register. For more details of the search methods used to build the Specialised Register, see the Group's webpages, which provide details of the Register's development (from inception) and the most recent searches performed to populate the Register. To summarise, the Register contains trials identified from the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, MEDLINE In‐Process, MEDLINE Epub Ahead of Print, ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP), and handsearching of journals and conference proceedings.

The date of the most recent search was 20 June 2022.

Appendix 3 presents the terms we used to search the Cochrane Incontinence Specialised Register.

Search for economic evaluations

We performed additional searches for the brief economic commentary (BEC). We searched the UK National Health Service Economic Evaluation Database (NHS EED) on the Centre for Reviews and Dissemination (CRD) website (covering from the earliest record in NHS EED, dating from 1968, up to and including 31 December 2014, when their coverage ended).

As NHS EED is no longer actively updated, we performed additional searches of the following databases to identify eligible studies added to these databases from 1 January 2015.

• MEDLINE on OvidSP (1 January 1946 to May Week 1 2022)

• Embase on OvidSP (1 January 1974 to 2022 Week 19)

The date of the most recent search was 16 May 2022.

Appendix 4 provides details of the searches.

Searching other resources

We searched the reference lists of all relevant reviews and included trial reports to identify further relevant studies.

Data collection and analysis

We conducted data collection and analysis in accordance with methods specified in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022a).

Selection of studies

Two review authors (AT and EJ) independently screened titles and abstracts of all the results returned by the search, then assessed the eligibility of the retrieved full‐text articles. We resolved any disagreements by discussion or by consulting a third review author.

Data extraction and management

Two review authors (AT and EJ) independently extracted data from the included studies using a piloted data extraction form. We compared the completed data extraction forms for disagreements, which we resolved by consensus. We entered all outcome data into Review Manager 5 (Review Manager 2020).

Assessment of risk of bias in included studies

Two review authors (AT and EJ) independently assessed the risk of bias for each included study using Cochrane's risk of bias tool (RoB 1), which covers the following domains (Higgins 2011).

• Random sequence generation

• Allocation concealment

• Blinding of participants and personnel

• Blinding of outcome assessment

• Incomplete outcome data

• Selective outcome reporting

• Other bias

We assessed each source of bias as either high, low, or unclear and provide a justification for our judgement in the risk of bias tables. We resolved any disagreements by discussion.

Measures of treatment effect

We based our analyses on available data from all included trials relevant to the comparisons and outcomes of interest. We summarised effect estimates for dichotomous outcomes using risk ratios (RRs) and corresponding 95% confidence intervals (CIs). For continuous outcomes, we used the mean difference (MD) or standardised mean difference (SMD) and corresponding 95% CI. We used final scores for continuous outcomes where change from baseline data were unavailable.

Unit of analysis issues

We had planned to incorporate cross‐over trial data using the methods described in Chapter 23 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022b). We had planned to analyse multi‐arm studies that compared two interventions arms with one control group using the methods described in Chapter 23 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022b). We would have prevented double‐counting of individuals by analysing each treatment arm separately against the common control group and dividing the sample size of the common control group proportionately across each intervention comparison.

Dealing with missing data

We contacted study authors to request missing data, where possible, and documented all attempts to contact authors for missing data or information in the Characteristics of included studies table.

Assessment of heterogeneity

We performed a qualitative assessment of clinical heterogeneity among studies (i.e. differences in study populations, interventions, and outcomes). Where studies were sufficiently similar to be included in a meta‐analysis, we assessed statistical heterogeneity by visually inspecting the forest plots, performing the Chi² test (P ≤ 0.10), and calculating the I² statistic (Deeks 2022). We planned to interpret I² values as follows.

• 0% to 40%: represents low heterogeneity

• 30% to 60%: may represent moderate heterogeneity

• 50% to 90%: may represent substantial heterogeneity

• 75% to 100%: represents considerable heterogeneity

We investigated possible causes of statistical heterogeneity, such as differing populations or interventions. We had intended to explore heterogeneity further through subgroup and sensitivity analyses.

Assessment of reporting biases

We searched registries of prospectively registered trials to identify completed but unpublished trials. Had we included 10 or more trials in any meta‐analysis, we would have investigated potential reporting bias by generating funnel plots (Egger 1997; Sterne 2011).

Data synthesis

We combined outcome data from the individual trials in meta‐analyses using Review Manager 5 (Review Manager 2020).

Where appropriate, we pooled effect estimates across similar studies using fixed‐effect or random‐effects meta‐analysis techniques. We intended to use the random‐effects model where studies showed heterogeneity, defined as an I² statistic over 50%. Otherwise, we intended to use a fixed‐effect model to pool the data, using the Mantel‐Haenszel method for dichotomous data and the inverse variance method for continuous data (Mantel 1959). Had data allowed, we would have used change from baseline data for continuous outcomes; instead, we used raw outcome data. Where a study collected data for a single outcome using multiple tools, we used data from the tool with the highest validity according to Cronbach's alpha. Additionally, where data were available for multiple types of adverse event, we used the outcome with the most events to avoid double‐counting.

Where outcome data from individual trials were not sufficiently similar to be combined quantitatively, we provided a narrative description of results from clinically comparable trials using the synthesis without meta‐analysis (SWiM) approach (Campbell 2020).

Subgroup analysis and investigation of heterogeneity

Had data allowed, we would have performed the following subgroup analyses.

• Systemic versus topical oestrogen

• Parous versus nulliparous women

• Women with versus without a uterus

Sensitivity analysis

Had data allowed, we would have excluded studies with a high risk of bias from our meta‐analyses for the primary outcomes to assess the robustness of our conclusions.

Incorporating economic evidence

Following the search outlined in the Search methods for identification of studies, we developed a BEC to summarise the availability and principal findings of the full economic evaluations that assess oestrogen therapy for treating POP in postmenopausal women (Aluko 2022). This BEC encompassed full economic evaluations (i.e. cost‐effectiveness analyses, cost‐utility analyses, and cost‐benefit analyses) conducted as part of a single empirical study, such as an RCT, a model based on such as study, or a model based on several such studies.

Summary of findings and assessment of the certainty of the evidence

We prepared summary of findings tables for the four main comparisons listed in Types of interventions. We planned to use the GRADE approach to assess the certainty of evidence related to the primary and secondary outcomes for the comparisons presented in the summary of findings tables (Schünemann 2021). The GRADE approach considers five domains (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of the body of evidence for each outcome. Had we included any data in our summary of findings tables, we would have justified all decisions to downgrade the certainty of evidence in footnotes.

Results

Description of studies

Results of the search

Search for clinical effectiveness studies

The literature search produced 239 records. After the title and abstract screen, we retrieved the full‐text articles of 74 potentially eligible records, and after reading these articles, we included 14 studies (47 reports) in the review. We excluded nine studies (14 reports), providing reasons for exclusion in the Characteristics of excluded studies table. Additionally, we identified seven reports of four ongoing studies (see Characteristics of ongoing studies table) and six reports of three studies still awaiting classification (see Studies awaiting classification). Figure 1 shows the flow of literature through the study selection process.

1.

PRISMA diagram – clinical effectiveness searches.

Search for economic evaluations

The literature search returned 212 records. After screening, we retrieved the full‐text article of only one record. This was an ongoing study and was not included in the BEC for this version of the review. Figure 2 shows the flow of literature through the study selection process.

2.

PRISMA diagram – cost effectiveness searches.

Included studies

We included 14 studies (Baracat 2001; Caruso 2017; Chiengthong 2022; Coelho 2020; Felding 1992; Karp 2012; Marschalek 2021; Nunes 2011; Rahn 2014; Stipic 2012; Sun 2016; Tontivuthikul 2016; Vaccaro 2013; Verghese 2020). For further details, see the Characteristics of included studies table. We contacted the authors of two included studies for further details (Caruso 2017; Verghese 2020), but only received a response regarding Verghese 2020.

Design

Thirteen studies were parallel‐arm RCTs (Baracat 2001; Caruso 2017; Chiengthong 2022; Coelho 2020; Felding 1992; Karp 2012; Marschalek 2021; Rahn 2014; Stipic 2012; Sun 2016; Tontivuthikul 2016; Vaccaro 2013; Verghese 2020). Nunes 2011 randomised two of its arms, but the third arm (no intervention) was not randomised, and the participants did not have POP.

Karp 2012 had three study arms, and Baracat 2001 had four study arms. Caruso 2017 initially randomised participants into two groups and then re‐randomised participants of the intervention group following surgery to either continue or cease the intervention. Vaccaro 2013 was initially a three‐arm study, but the first group were excluded early in the trial, leaving only two arms.

Sample size

The smallest study had eight participants (Baracat 2001), while the largest had 186 (Sun 2016).

Setting

The included studies were set in Austria (Marschalek 2021), Brazil (Baracat 2001; Coelho 2020; Nunes 2011), China (Sun 2016), Croatia (Stipic 2012), Denmark (Felding 1992), Italy (Caruso 2017), Thailand (Chiengthong 2022; Tontivuthikul 2016), the UK (Verghese 2020), and the USA (Karp 2012; Rahn 2014; Vaccaro 2013).

Nine studies were conducted in a single centre (Baracat 2001; Caruso 2017; Chiengthong 2022; Coelho 2020; Karp 2012; Rahn 2014; Stipic 2012; Sun 2016; Tontivuthikul 2016), while four were multicentre studies (Felding 1992; Marschalek 2021; Vaccaro 2013; Verghese 2020). Nunes 2011 did not report the number of participating sites.

Participants

In terms of POP type and severity, 10 studies recruited women with varying degrees of POP severity (Caruso 2017; Chiengthong 2022; Coelho 2020; Karp 2012; Marschalek 2021; Rahn 2014; Sun 2016; Tontivuthikul 2016; Vaccaro 2013; Verghese 2020). Three recruited women with central compartment (uterine) POP (Baracat 2001; Nunes 2011; Stipic 2012), while one recruited women with anterior compartment or central compartment (uterine) POP (Caruso 2017). In two studies, participants had anterior, central (apical), or posterior prolapse, or more than one of those types (Coelho 2020; Marschalek 2021). In Rahn 2014, participants had uterine or anterior compartment POP, or both. The participants in Sun 2016 had POP affecting either the anterior, apical, or posterior compartments.

Four studies did not report POP severity (Baracat 2001; Felding 1992; Nunes 2011; Stipic 2012), while six did not report the affected compartments (Chiengthong 2022; Felding 1992; Karp 2012; Tontivuthikul 2016; Vaccaro 2013; Verghese 2020).

In Chiengthong 2022, 100% of women in the intervention group and 84.6% in the non‐intervention group had one or more children, while in nine studies, most participants had two or more children (Caruso 2017; Coelho 2020; Karp 2012; Marschalek 2021; Rahn 2014; Stipic 2012; Sun 2016; Tontivuthikul 2016; Vaccaro 2013). In Verghese 2020, participants were roughly balanced in terms of whether they had two or fewer children (53%) or more than two children (47%). In Nunes 2011, average parity across the groups was at least four births. Two studies did not report parity (Baracat 2001; Felding 1992).

In three studies, none of the women had undergone a hysterectomy (Caruso 2017; Nunes 2011; Stipic 2012). In Chiengthong 2022, under 20% of women had had a hysterectomy, while in Karp 2012, the proportion of women with previous hysterectomy ranged from 27% to 33% between groups. In Marschalek 2021, prior rates of hysterectomy ranged from 10% in the control group to 28% in the intervention group. In Tontivuthikul 2016, 25% of women in the intervention group had undergone hysterectomy, compared to 20% in the control group. Hysterectomy rates across groups in Verghese 2020 ranged from 16% to 24%. All participants in Vaccaro 2013 had previously undergone hysterectomy. Five studies did not report previous hysterectomy (Baracat 2001; Coelho 2020; Felding 1992; Rahn 2014; Sun 2016).

The average age of participants was under 65 years in seven studies (Baracat 2001; Caruso 2017; Felding 1992; Marschalek 2021; Nunes 2011; Rahn 2014; Stipic 2012), and 65 years or over in five studies (Chiengthong 2022; Coelho 2020; Karp 2012; Tontivuthikul 2016; Verghese 2020). In Vaccaro 2013, the average age of participants was under 65 years in the intervention group and over 65 years in the control group. Sun 2016 reported the median age of participants at diagnosis of POP, which was 66 years in the intervention group and 65 years in the control group.

Average BMI of the women was greater than 25 kg/m² in nine studies (Caruso 2017; Coelho 2020; Karp 2012; Marschalek 2021; Nunes 2011; Rahn 2014; Stipic 2012; Vaccaro 2013; Verghese 2020), and under 25 kg/m² in three studies (Chiengthong 2022; Sun 2016; Tontivuthikul 2016). Two studies did not report BMI (Baracat 2001; Felding 1992).

Interventions

Oestrogen therapy plus surgery versus surgery alone

Eleven studies evaluated oestrogen therapy plus surgery versus surgery alone (Baracat 2001; Caruso 2017; Marschalek 2021; Nunes 2011; Rahn 2014; Stipic 2012; Sun 2016; Vaccaro 2013; Verghese 2020; Felding 1992; Karp 2012).

The types and doses of included topical oestrogens were as follows.

• Oestriol 50 μg vaginal gel (Caruso 2017)

• Oestradiol hemihydrate 1 g (Marschalek 2021; Stipic 2012)

• Promestriene 0.5 g cream (Sun 2016)

• Conjugated equine oestrogen vaginal cream (Premarin) 1.0 g (Vaccaro 2013)

• Oestradiol hemihydrate (Vagifem) 10 μg (Verghese 2020, Felding 1992)

• Oestradiol vaginal ring (Estring; Karp 2012)

The types and doses of included systemic oestrogens were as follows.

• Conjugated equine oestrogen (Premarin) 0.625 mg/g (Baracat 2001; Rahn 2014)

• Conjugated equine oestrogen (Premarin) 1.26 mg (Baracat 2001)

• 17‐beta‐oestradiol 50 g transdermal (Baracat 2001)

• "oral oestrogen" (no other details; Nunes 2011)

The types of surgery were as follows.

• Hysterectomy (Baracat 2001; Nunes 2011; Rahn 2014)

• Hysterectomy and anterior colporrhaphy (Caruso 2017)

• Hysterectomy with or without salpingo‐oophorectomy, anterior repair, posterior repair, or sacrospinous fixation (Verghese 2020)

• Hysterectomy, anterior and posterior colporrhaphy, Manchester‐Fothergill operation, or a combination of these (Felding 1992). These participants were also offered PFME.

• Reconstructive surgery, including hysterectomy and "kit" procedures (Karp 2012)

• Pelvic reconstructive surgery using mesh (Sun 2016)

Vaccaro 2013 described the surgery as either vaginal, robotic, or abdominal with no further details. Two studies did not describe the surgical procedures (Marschalek 2021; Stipic 2012).

Topical oestrogen therapy plus pessary versus pessary alone

Three studies evaluated topical oestrogen therapy plus pessary versus pessary alone (Chiengthong 2022; Coelho 2020; Tontivuthikul 2016). All three studies recruited women who were already using a pessary. Tontivuthikul 2016 also included pretreatment with oestrogen cream for all participants.

Outcomes

Two studies reported usable data for our primary outcome of participant‐reported subjective cure of POP (Sun 2016; Verghese 2020).

There was usable data for the following secondary outcomes.

• Improvement or cure of associated urinary symptoms (Caruso 2017; Sun 2016; Verghese 2020)

• Improvement or cure of associated bowel symptoms (Verghese 2020)

• Improvement or cure of associated sexual symptoms (Caruso 2017; Sun 2016; Verghese 2020)

• Objective improvement or cure of POP symptoms (Rahn 2014; Sun 2016; Verghese 2020)

• Condition‐specific quality of life (Sun 2016)

• Generic quality of life or health status (Caruso 2017)

• Participant adherence to the intervention (Rahn 2014; Vaccaro 2013; Verghese 2020)

• Adverse events (Caruso 2017; Chiengthong 2022; Coelho 2020; Felding 1992; Karp 2012; Marschalek 2021; Rahn 2014; Sun 2016; Tontivuthikul 2016; Vaccaro 2013; Verghese 2020)

• Surgical outcomes (Marschalek 2021; Karp 2012; Rahn 2014; Sun 2016; Verghese 2020)

No study reported psychological or emotional well‐being. Three studies provided no usable data for our outcomes of interest (Baracat 2001; Nunes 2011; Stipic 2012).

The Characteristics of included studies table presents details of the tools used to measure outcomes.

Excluded studies

We excluded eight studies: five had an ineligible population (Chinthakanan 2019; Chou 2013; Moalli 2012; Lind 2014; Paraiso 2020), and three evaluated an ineligible intervention (Barber 2002; Jiang 2017; Valente 2000). For further information, see the Characteristics of excluded studies table.

Studies awaiting classification

For three studies, we have requested information from the study authors to clarify eligibility but have yet to receive a response (Pinhat 2013; Vardy 2003; Zhu 2014). The Studies awaiting classification table provides further information.

Ongoing studies

We identified four ongoing studies (Rahn 2021; Roovers 2018; Vincent 2013; Zhu 2020). The Characteristics of ongoing studies table provides further information.

Risk of bias in included studies

The Characteristics of included studies table provides further information on risk of bias assessments. For a graphical summary of risk of bias judgements, see Figure 3 and Figure 4.

3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

4.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Random sequence generation

Eight studies adequately reported the method of randomisation, leading to a low risk of bias judgement for this domain (Chiengthong 2022; Coelho 2020; Karp 2012; Marschalek 2021; Rahn 2014; Sun 2016; Tontivuthikul 2016; Verghese 2020).

Four studies provided insufficient information for us to make a judgement, so we considered them at unclear risk of bias (Baracat 2001; Caruso 2017; Felding 1992; Stipic 2012). We also judged Vaccaro 2013 at unclear risk for this domain: although its initial randomisation procedure seemed adequate, one arm was subsequently excluded and the handling of these participants may have been open to bias.

We judged Nunes 2011 at high risk of bias for this domain because it provided no details on the randomisation procedure for two arms, while the third arm included non‐randomised women without POP.

Allocation concealment

Allocation concealment was adequate and clearly described in three studies, which we judged at low risk of bias (Chiengthong 2022; Coelho 2020; Karp 2012). Ten studies provided insufficient information for us to make a judgement, so we judged them at unclear risk (Baracat 2001; Caruso 2017; Felding 1992; Marschalek 2021; Nunes 2011; Rahn 2014; Stipic 2012; Sun 2016; Tontivuthikul 2016; Verghese 2020). Vaccaro 2013 noted that blinding was not possible during enrolment, so we judged it at high risk of bias.

Blinding

Performance bias

Two studies adequately blinded participants and personnel to the allocated treatment, leading to a low risk judgement for performance bias (Marschalek 2021; Rahn 2014). In eight studies, participants would have been aware of their allocated group owing to the nature of the interventions, or it was stated that blinding did not occur; we rated these studies at high risk of performance bias (Caruso 2017; Chiengthong 2022; Coelho 2020; Stipic 2012; Sun 2016; Tontivuthikul 2016; Vaccaro 2013; Verghese 2020). Four studies provided insufficient information to permit judgement so were at unclear risk of performance bias (Baracat 2001; Felding 1992; Karp 2012; Nunes 2011).

Detection bias

We judged six studies at low risk of detection bias as outcome assessment was adequately blinded (Coelho 2020; Felding 1992; Marschalek 2021; Rahn 2014; Stipic 2012; Sun 2016). Seven studies provided insufficient information to permit judgement so were at unclear risk of detection bias (Baracat 2001; Caruso 2017; Chiengthong 2022; Karp 2012; Nunes 2011; Tontivuthikul 2016; Verghese 2020). In Vaccaro 2013, some outcomes appeared to be blinded but not all, so we judged the study at unclear risk.

Incomplete outcome data

Attrition was low in 10 studies, leading to a low risk of bias judgement (Baracat 2001; Caruso 2017; Chiengthong 2022; Felding 1992; Karp 2012; Marschalek 2021; Nunes 2011; Stipic 2012; Sun 2016; Tontivuthikul 2016). We judged Coelho 2020 at high risk of attrition bias because of the high rate of loss to follow‐up in both groups. Rahn 2014 and Vaccaro 2013 were also at high risk of attrition bias because of large differential dropouts between groups. Verghese 2020 was at unclear risk because it was unclear whether the study authors performed an intention‐to‐treat (ITT) analysis, and there was a small differential loss to follow‐up.

Selective reporting

We judged five studies at low risk of selective reporting because they appeared to report all prespecified outcomes (Baracat 2001; Chiengthong 2022; Felding 1992; Marschalek 2021; Verghese 2020).

Four studies were at high risk of reporting bias. In Caruso 2017, all participants underwent transvaginal scans for endometrial thickness, but the results were not reported. The trial registration for Coelho 2020 did not list the urinary outcomes assessed in the study, and the report provided no explanation for the addition of these outcomes. Furthermore, adherence and medication use were monitored and recorded but not reported. Sun 2016 listed postoperative vaginal swab culture result as an outcome in its trial registration but did not report this in the results. Vaccaro 2013 only reported cytological outcomes in the full‐text article, while the abstract reported quality of life; we found no explanation for this inconsistency.

Four studies published after 2010 did not have an associated trial registration, so we were unable to judge which outcomes had been prespecified. We judged these studies at unclear risk of reporting bias (Karp 2012; Nunes 2011; Stipic 2012; Tontivuthikul 2016). Although Rahn 2014 reported all measures listed in its trial registration, it performed a post‐hoc analysis without explanation; for this reason, we also judged it at unclear risk of reporting bias.

Other potential sources of bias

There was no indication of any other sources of bias in 11 studies (Baracat 2001; Caruso 2017; Chiengthong 2022; Coelho 2020; Felding 1992; Karp 2012; Marschalek 2021; Rahn 2014; Sun 2016; Tontivuthikul 2016; Verghese 2020).

There were major baseline differences between groups in Nunes 2011, suggesting issues with the randomisation process and leading to a high risk judgement for other bias. We also judged Vaccaro 2013 at high risk of other bias because one arm of the study was excluded and participants were evaluated at between two and 12 weeks of treatment rather than at a standardised follow‐up time.

We considered Stipic 2012 at unclear risk of other bias because it did not report any baseline respiratory variables, although the study's primary endpoint was lung function.

Effects of interventions

See: Table 1; Table 2; Table 3; Table 4

Oestrogen therapy alone versus no treatment or placebo

No studies evaluated oestrogen therapy alone versus no treatment or placebo.

Oestrogen therapy alone versus pelvic floor muscle training

No studies evaluated oestrogen therapy alone versus PFMT.

Oestrogen therapy alone versus devices such as vaginal pessaries

No studies evaluated oestrogen therapy alone versus devices such as vaginal pessaries.

Oestrogen therapy alone versus surgery

No studies evaluated oestrogen therapy alone versus surgery.

Oestrogen therapy in conjunction with pelvic floor muscle training versus pelvic floor muscle training alone

No studies evaluated oestrogen therapy in conjunction with PFMT versus PFMT alone.

Oestrogen therapy in conjunction with vaginal pessaries versus vaginal pessaries alone

Three studies evaluated topical oestrogen therapy in conjunction with vaginal pessaries versus vaginal pessaries alone (Chiengthong 2022; Coelho 2020; Tontivuthikul 2016).

Primary outcome

Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms

No studies reported participant‐reported subjective improvement or cure of POP symptoms.

Secondary outcomes

Improvement or cure of associated urinary symptoms

No studies reported improvement or cure of associated urinary symptoms.

Improvement or cure of associated bowel symptoms

No studies reported improvement or cure of associated bowel symptoms.

Improvement or cure of associated sexual symptoms

No studies reported improvement or cure of associated sexual symptoms.

Objective improvement or cure of pelvic organ prolapse symptoms

No studies reported objective improvement or cure of POP symptoms.

Condition‐specific quality of life

No studies reported condition‐specific quality of life.

Generic quality of life or health status

No studies reported generic quality of life or health status.

Psychological or emotional well‐being outcome measures and ability to cope with daily activities

No studies reported psychological or emotional well‐being outcome measures or ability to cope with daily activities.

Participant adherence to the intervention

Only Coelho 2020 reported participant adherence to topical oestrogen through discontinuation of pessary use. Slightly more women in the control group discontinued pessary use (Analysis 1.1).

1.1. Analysis.

Comparison 1: Oestrogen therapy (OT) in conjunction with vaginal pessaries versus vaginal pessaries alone, Outcome 1: Participant adherence to the intervention (including discontinuation of pessary use and declining surgical procedures)

Participant adherence to the intervention (including discontinuation of pessary use and declining surgical procedures)
Study	OT + pessary	Pessary alone	Time‐point	Nature of adherence
Coelho 2020	8/46	12/52	Not reported	Discontinued pessary

Reduction in minor adverse events: urinary tract infection

No studies reported reduction in UTIs.

Reduction in minor adverse events: vaginal events

All three studies reported reduction in vaginal events. Topical oestrogen therapy plus pessary may lead to a reduction in minor adverse vaginal events compared with pessary alone (RR 0.48, 95% CI 0.25 to 0.91; 3 studies, 185 participants; Analysis 1.2).

1.2. Analysis.

Comparison 1: Oestrogen therapy (OT) in conjunction with vaginal pessaries versus vaginal pessaries alone, Outcome 2: Reduction in minor adverse events: vaginal events

Reduction in minor adverse events: urinary storage events

No studies reported reduction in urinary storage events.

Reduction in minor adverse events: sexual adverse events

No studies reported reduction in sexual adverse events.

Major adverse events

Two studies recorded the number of major adverse events experienced by women (Coelho 2020; Tontivuthikul 2016). No major adverse events occurred in either group in either study (Analysis 1.3).

1.3. Analysis.

Comparison 1: Oestrogen therapy (OT) in conjunction with vaginal pessaries versus vaginal pessaries alone, Outcome 3: Major adverse events

Major adverse events
Study	OT + pessary	Pessary alone	Time‐point	Description of adverse event
Coelho 2020	0/46	0/52	During treatment	Major adverse events
Tontivuthikul 2016	0/20	0/20	Not reported	Major adverse events

Surgical outcomes: operative time

No studies reported operative time.

Surgical outcomes: mesh exposure

No studies reported mesh exposure.

Oestrogen therapy in conjunction with surgery versus surgery alone

Eleven studies evaluated topical or systemic oestrogen therapy in conjunction with surgery versus surgery alone (Baracat 2001; Caruso 2017; Felding 1992; Karp 2012; Marschalek 2021; Nunes 2011; Rahn 2014; Stipic 2012; Sun 2016; Vaccaro 2013; Verghese 2020).

Primary outcome

Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms

Two studies reported subjective improvement or cure of POP symptoms, but we were unable to pool the data in a meta‐analysis (Sun 2016; Verghese 2020). Both studies found no evidence of a difference in subjective improvement of POP symptoms between topical oestrogen therapy plus surgery and surgery alone at six months (265 participants; Analysis 2.1) and at 12 months (263 participants; Analysis 2.2).

2.1. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 1: Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms: 6 months

Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms: 6 months
Study	OT + surgery	Surgery alone	Tool used	Tool direction
Sun 2016	93.5% n = 93	90.3% n = 93	Rate of satisfaction on Patient Global Impression of Change (PGI‐C)	Good response on PGI‐C: at month 12, defined as "much better" or "a little better" when participants were asked "compared with how you were doing before your recent pelvic floor operation, would you say that now you are much better/a little better/about the same/a little worse/much worse?"
Verghese 2020	45.4 ± 43.8 n = 38	45.0 ± 37.1 n = 41	Pelvic Floor Impact Questionnaire (PFIQ‐7)	Higher score = worse outcome

2.2. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 2: Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms: 12 months

Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms: 12 months
Study	OT + surgery	Surgery alone	Tool used	Tool direction
Sun 2016	97.8% n = 93	95.7% n = 93	Rate of satisfaction on Patient Global Impression of Change (PGI‐C)	Good response on PGI‐C: at month 12, defined as "much better" or "a little better" when participants were asked "compared with how you were doing before your recent pelvic floor operation, would you say that now you are much better/a little better/about the same/a little worse/much worse?"
Verghese 2020	49.8 ± 44.1 n = 36	51.1 ± 47.7 n = 41	Pelvic Floor Impact Questionnaire (PFIQ‐7) summary score	Score 0–300; higher score = worse outcome

Secondary outcomes

Improvement or cure of associated urinary symptoms

Only Verghese 2020 reported improvement or cure of associated urinary symptoms, finding that topical oestrogen therapy plus surgery compared with surgery alone may result in worse urinary symptoms, measured using the Urinary Impact Questionnaire (UIQ‐7) subscale of the PFIQ‐7, at six months (79 participants; Analysis 2.3). However, the same study reported no difference in urinary symptoms between the two groups at 12 months (79 participants; Analysis 2.4). There was no clinically important difference at either time point (six or 12 months).

2.3. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 3: Improvement or cure of associated urinary symptoms: 6 months

Improvement or cure of associated urinary symptoms: 6 months
Study	OT + surgery	Surgery alone	Tool used	Tool direction
Verghese 2020	9.6 ± 21.7 n = 38	5.9 ± 11.2 n = 41	Urinary Impact Questionnaire (UIQ‐7); score 0–100	Higher score = worse outcome

2.4. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 4: Improvement or cure of associated urinary symptoms: 12 months

Improvement or cure of associated urinary symptoms: 12 months
Study	OT + surgery	Surgery alone	Tool used	Tool direction
Verghese 2020	8.9 ± 16.8 n = 37	8.0 ± 13.9 n = 42	Urinary Impact Questionnaire (UIQ‐7); score 0–100	Higher score = worse outcome

Improvement or cure of associated bowel symptoms

Only Verghese 2020 reported improvement of associated bowl symptoms, finding that topical oestrogen therapy plus surgery compared with surgery alone may result in worse bowel symptoms, measured using the Colorectal‐Anal Impact Questionnaire (CRAIQ‐7) subscale of the PFIQ‐7, at six months (79 participants; Analysis 2.5) and at 12 months (79 participants; Analysis 2.6). However, there was no clinically important difference at either time point (six or 12 months).

2.5. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 5: Improvement or cure of associated bowel symptoms: 6 months

Improvement or cure of associated bowel symptoms: 6 months
Study	OT + surgery	Surgery alone	Tool used	Tool direction
Verghese 2020	4.9 ± 13.1 n = 38	2.4 ± 6.1 n = 41	Colorectal‐Anal Impact Questionnaire (CRAIQ‐7); score 0–100	Higher score = worse outcome

2.6. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 6: Improvement or cure of associated bowel symptoms: 12 months

Improvement or cure of associated bowel symptoms: 12 months
Study	OT + surgery	Surgery alone	Tool used	Tool direction
Verghese 2020	4.5 ± 16.3 n = 36	2.9 ± 6.1 n = 42	Colorectal‐Anal Impact Questionnaire (CRAIQ‐7); score 0–100	Higher score = worse outcome

Improvement or cure of associated sexual symptoms

Two studies reported improvement or cure of associated sexual symptoms (Sun 2016; Verghese 2020). At six months, topical oestrogen therapy plus surgery may result in a small improvement in sexual symptoms compared with surgery alone, although the difference is not considered to be clinically important (MD −1.02, 95% CI −3.52 to 1.48; 2 studies 211 participants; Analysis 2.7). It is uncertain whether topical oestrogen therapy with surgery has an effect on associated sexual symptoms compared with surgery alone after 12 months; the difference is not considered to be clinically important (MD −0.57, 95% CI −1.98 to 0.85; 2 studies, 207 participants; Analysis 2.8).

2.7. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 7: Improvement or cure of associated sexual symptoms: 6 months

2.8. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 8: Improvement or cure of associated sexual symptoms: 12 months

Objective improvement or cure of pelvic organ prolapse symptoms

Three studies reported objective improvement or cure of POP symptoms (Rahn 2014; Sun 2016; Verghese 2020), though we were only able to pool data from Sun 2016 and Verghese 2020. The meta‐analysis suggested there may be no difference between topical oestrogen therapy plus surgery and surgery alone in terms of objective improvement of POP symptoms at six months (RR 1.01, 95% CI 0.94 to 1.08; 241 participants; Analysis 2.9). Rahn 2014 also found no evidence of a difference in objective improvement at six months (20 participants; Analysis 2.10).

2.9. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 9: Objective improvement or cure of pelvic organ prolapse: 6 months (meta‐analysis)

2.10. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 10: Objective improvement or cure of pelvic organ prolapse: 6 months (other data)

Objective improvement or cure of pelvic organ prolapse: 6 months (other data)
Study	OT + surgery	Surgery alone	Tool used
Rahn 2014	Postoperative anterior compartment stage Median: 0.5 IQR: 0.5, 0.8 n = 8	Postoperative anterior compartment stage Median: 1.0 IQR: 0.0, 1.3 n = 12	Assumed Pelvic Organ Prolapse Quantification System (POP‐Q); not clear from paper

Sun 2016 measured objective improvement or cure of POP symptoms at 12 months, finding no difference between topical oestrogen therapy plus surgery and surgery alone (186 participants; Analysis 2.11).

2.11. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 11: Objective improvement or cure of pelvic organ prolapse: 12 months

Objective improvement or cure of pelvic organ prolapse: 12 months
Study	OT + surgery	Surgery alone	Tool used
Sun 2016	100% n = 93	98.9% n = 93	Anatomic success rate % measured by Pelvic Organ Prolapse Quantification System (POP‐Q) stage < 2

Condition‐specific quality of life

Only Sun 2016 reported condition‐specific quality of life, finding no difference between topical oestrogen therapy plus surgery and surgery alone at six months (186 participants; Analysis 2.12) or 12 months (186 participants; Analysis 2.13).

2.12. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 12: Condition‐specific quality of life: 6 months

Condition‐specific quality of life: 6 months
Study	OT + surgery	Surgery alone	Tool used	Direction of tool
Sun 2016	12.4 ± 5.7 n = 93	13.8 ± 5.6 n = 93	Pelvic Floor Impact Questionnaire (PFIQ‐7)	Higher score = better result

2.13. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 13: Condition‐specific quality of life: 12 months

Condition‐specific quality of life: 12 months
Study	OT + surgery	Surgery alone	Tool used	Direction of tool
Sun 2016	7.2 ± 3.4 n = 93	7.8 ± 4.3 n = 93	Pelvic Floor Impact Questionnaire (PFIQ‐7)	Higher score = better result

Generic quality of life or health status

No studies reported generic quality of life or health status.

Psychological or emotional well‐being outcome measures and ability to cope with daily activities

No studies reported psychological or emotional well‐being outcome measures or ability to cope with daily activities.

Participant adherence to the intervention

Three studies reported participant adherence to the intervention (Felding 1992; Marschalek 2021; Verghese 2020), but we were only able to pool data from Felding 1992 and Marschalek 2021). The meta‐analysis showed no difference in adherence to the intervention between women who received oestrogen therapy plus surgery compared with those who received surgery alone (RR 1.07, 95% CI 0.49 to 2.32; 2 studies, 114 participants; Analysis 2.14).

2.14. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 14: Participant adherence to intervention (meta‐analysis)

Verghese 2020 reported that 20 of 28 women in the topical oestrogen group used the oestrogen pessaries at least 75% of the expected time (Analysis 2.15).

2.15. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 15: Participant adherence to intervention (other data)

Participant adherence to intervention (other data)
Study	OT + surgery	Surgery alone	Definition of compliance	Time‐point
Verghese 2020	20/28	Not applicable (no oestrogen therapy)	Pessaries used ≥ 75% of the expected time	6 months

Reduction in minor adverse events: urinary tract infection

Five studies reported reduction in UTIs (Felding 1992; Karp 2012; Marschalek 2021; Sun 2016; Verghese 2020). Topical oestrogen therapy plus surgery may lead to fewer cases of UTI compared with surgery alone (RR 0.49, 95% CI 0.26 to 0.90; 483 participants; Analysis 2.16).

2.16. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 16: Reduction in minor adverse events: urinary tract infection

Reduction in minor adverse events: vaginal events

Four studies reported reduction in vaginal events (Karp 2012; Marschalek 2021; Sun 2016; Vaccaro 2013). From the results of the meta‐analysis, it is uncertain whether topical oestrogen therapy plus surgery affects the occurrence of adverse vaginal events compared with surgery alone, as the 95% CI indicates both benefit and harm (RR 2.23, 95% CI 0.74 to 6.79; 371 participants; Analysis 2.17).

2.17. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 17: Reduction in minor adverse events: vaginal events

Reduction in minor adverse events: urinary storage events

Two studies reported reduction in urinary storage events (Marschalek 2021; Sun 2016). Meta‐analysis of their results showed no difference in risk of urinary storage issues between women who received topical oestrogen plus surgery and those who received surgery alone (RR 0.94, 95% CI 0.46 to 1.92; 289 participants; Analysis 2.18).

2.18. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 18: Reduction in minor adverse events: urinary storage events

Reduction in minor adverse events: sexual adverse events

Two studies reported sexual adverse events, but we were unable to pool their results (Karp 2012; Sun 2016). Karp 2012 found little evidence of a difference between the groups in the number of partners experiencing discomfort during intercourse at six weeks (1/22 in the oestrogen therapy plus surgery group versus 2/43 in the surgery alone group; Analysis 2.19). Sun 2016 found that the number of women with dyspareunia was slightly lower in the oestrogen therapy plus surgery group compared with the surgery alone group at one year (17/93 versus 21/93; Analysis 2.19).

2.19. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 19: Reduction in minor adverse events: sexual adverse events

Reduction in minor adverse events: sexual adverse events
Study	OT + surgery	Surgery alone	Symptom	Time‐point
Karp 2012	1/22	2/43	Partner discomfort with intercourse	6 weeks
Sun 2016	17/93	21/93	Dyspareunia	Any time within 1 year

Major adverse events

Five studies reported major adverse events (Felding 1992; Marschalek 2021; Rahn 2014; Sun 2016; Verghese 2020). It is uncertain whether topical or systemic oestrogen therapy plus surgery affects the occurrence of major adverse events compared with surgery alone, because the 95% CI indicates both benefit and harm (RR 1.02, 95% CI 0.39 to 2.66; 453 participants; Analysis 2.20).

2.20. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 20: Major adverse events: total number of events

Surgical outcomes: operative time

Only Marschalek 2021 reported operative time. Operative time in minutes may be slightly longer for women receiving topical oestrogen plus surgery compared with those receiving surgery alone (103 participants; Analysis 2.21).

2.21. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 21: Surgical outcomes: operative time

Surgical outcomes: operative time
Study	OT + surgery	Surgery alone	Measurement
Marschalek 2021	88.73 ± 30.3 n = 51	78.46 ± 32.3 n = 52	Minutes

Surgical outcomes: mesh exposure

Two studies reported mesh exposure, but we were unable to pool their results (Karp 2012; Sun 2016). Both studies found little or no difference in the number of women with mesh exposures between the topical oestrogen plus surgery and surgery alone groups (Analysis 2.22).

2.22. Analysis.

Comparison 2: Oestrogen therapy (OT) in conjunction with surgery versus surgery alone, Outcome 22: Surgical outcomes: mesh exposure

Surgical outcomes: mesh exposure
Study	OT + surgery	Surgery alone	Measurement
Karp 2012	0/19	0/40	Number of mesh exposures on examination at 12 weeks after surgery
Sun 2016	15/93	11/93	Mesh exposures ≤ 1 cm within 1 year

Discussion

A previous Cochrane Review evaluated oestrogens for treatment or prevention of POP in postmenopausal women (Ismail 2010). To aid clarity, it was decided to divide the topic into two separate reviews covering prevention and treatment. This review covers the treatment of POP with oestrogen therapy in postmenopausal women.

Summary of main results

We identified 14 studies including a total of 1002 women in this version of the review. No studies assessed the effects of oestrogen therapy alone versus no treatment or placebo, versus PFMT, versus devices such as vaginal pessaries, or versus surgery. However, three studies assessed topical oestrogen therapy in conjunction with vaginal pessaries versus vaginal pessaries alone, and 11 studies evaluated oestrogen therapy in conjunction with surgery versus surgery alone.

Overall completeness and applicability of evidence

We conducted a comprehensive search for studies, conducted citation chaining, and attempted to obtain unpublished data where necessary (e.g. by contacting study authors). However, the 14 studies included in this review only covered two of our prespecified comparisons, and none contributed to any of our four main comparisons.

Many studies were small, and the number of participants ranged from six to 186. Although the studies were set in a mix of high‐income countries and low‐ and middle‐income countries, three were conducted in Brazil, and three in the USA. As we identified only 14 studies, the applicability of our findings to different populations may be limited. In addition, many studies did not report the ethnicities of participants.

The studies were clinically heterogenous in terms of the outcomes assessed, the tools used, and the time points of measurement. Additionally, many studies focused on Vaginal Maturation Index (VMI) and cytological outcomes that are associated more with genitourinary symptoms of menopause (GSM) than with POP, or used as surrogate markers for outcomes.

Furthermore, we were unable to perform our prespecified subgroup analyses owing to insufficient data. Because we could not analyse the results of topical and systemic oestrogen therapy separately, some of our findings may not take into account the route of administration. Although there is evidence of a difference in the effect of systemic and local oestrogen on urinary symptoms (Cody 2012), similar evidence is not available for POP. Benefits from systemic oestrogen such as improvement of vaginal dryness and vaginal epithelial quality may also be relevant for POP symptoms (Rahn 2015).

Quality of the evidence

In general, the evidence on oestrogen therapy for treating POP is uncertain. We were unable to assess the evidence using GRADE, as no studies evaluated our main comparisons of interest.

The meta‐analyses we performed often included too few participants to give a precise result; this imprecision was reflected in wide CIs. In terms of risk of bias, many studies were at high risk of performance bias because they did not blind participants or personnel, and we also had some concerns about selective reporting.

Potential biases in the review process

We made every effort to minimise bias within the review process. This included conducting a comprehensive literature search, including citation chaining, and having two review authors independently screen studies, extract data, and assess risk of bias. Had data allowed, two independent review authors would have performed the GRADE assessment.

However, we changed the PICO from that prespecified in the protocol, particularly surrounding outcome measures (see Differences between protocol and review). We made most of these changes after starting data extraction. Our intention was to make the review more clinically useful and accessible for stakeholders (including patients, the public, and clinicians) by focusing on outcomes that would be more consistent across our prespecified comparisons. We made every attempt not to be data‐driven when making these decisions.

Agreements and disagreements with other studies or reviews

Two previous reviews investigated the role of oestrogen in the treatment of POP (Rahn 2015; Weber 2015). Both concluded that the available evidence was scarce and of low quality, and the included studies assessed vaginal signs and symptoms of atrophy or hypo‐oestrogenisation rather than specifically signs and symptoms of POP.

Brief economic commentary

To supplement the main systematic review of oestrogen therapy for treating POP in postmenopausal women, we sought to identify relevant economic evaluations. We found no economic studies that analysed the use of oestrogen therapy in this population. The apparent shortage of relevant economic evaluations indicates a paucity of economic evidence on the efficiency of oestrogen therapy as a management strategy for treating POP in postmenopausal women. It should be noted that one potentially relevant ongoing trial includes an economic evaluation (Vodegel 2021). The ongoing trial is a multicentre, double‐blind, randomised, placebo‐controlled trial based in the Netherlands, comparing the administration of vaginal oestrogen cream or placebo cream from four to six weeks after surgery until 12 months after surgery. The economic evaluation will include a cost‐effectiveness analysis (with the outcomes measured by the PFDI‐20) and a cost‐utility analysis (with the outcomes measured by quality adjusted life years (QALYs)).

Authors' conclusions

Implications for practice.

The available evidence suggests that topical oestrogen may reduce the risk of minor adverse vaginal events when used in conjunction with pessaries for pelvic organ prolapse (POP). In addition, when used in conjunction with surgery for POP, topical oestrogen may reduce the risk of early urinary tract infections. There is no evidence of significant harm or adverse events attributable to vaginal oestrogen. However, these findings should be interpreted with caution, as the studies that contributed data varied substantially in their design. Therefore, we are unable to provide evidence‐based recommendations regarding the use of oestrogen for treating POP, either alone or in conjunction with surgery or support pessaries, as the evidence is uncertain.

Implications for research.

Topical oestrogen may play an adjunctive role in the treatment of POP when used alongside surgery or support pessaries. The data appear to imply that minor adverse vaginal events related to support pessaries are less common among women who are also using topical oestrogen. There is a need for sufficiently powered studies to assess appropriate outcomes over a longer time frame. No studies investigated whether oestrogen can improve the long‐term integrity of surgical repair or prolonged use of pessaries. Short‐term complications following surgical repair are insignificant compared to the burden of longer‐term failure of the procedure (prolapse recurrence).

The data suggest that preoperative use of topical oestrogen may reduce the risk of early postoperative urinary tract infection. It seems logical that preoperative use of topical oestrogen would improve the vaginal tissue quality to provide a better substrate for suture placement for surgical repair and improve postoperative tissue healing. However, many of the outcome measures in the included studies, such as vaginal epithelial thickness, are surrogate outcomes for tissue quality. If preoperative oestrogen does improve tissue quality as hypothesized, then postoperative use until full healing may also provide further benefits. Meaningful studies demonstrating the mode of effect of preoperative oestrogen on long‐term surgical outcomes should be adequately powered and randomised, comparing tissue quality between oestrogenised and non‐oestrogenised groups and subsequent long‐term surgical outcomes.

In addition, there is a need to develop appropriate measures for assessing adherence to oestrogen therapy and reasons for discontinuation.

History

Protocol first published: Issue 5, 2021

Notes

A previous Cochrane Review evaluated oestrogens for treatment or prevention of POP in postmenopausal women (Ismail 2010). To aid clarity, it was decided to divide the topic into two separate reviews covering prevention and treatment. This review covers the treatment of POP with oestrogen therapy in postmenopausal women.

Appendices

Appendix 1. Glossary of terms

Term	Definition
Anterior vaginal wall	The front wall of the vagina (anterior compartment).
Cuff scar	A scar at the apex of the vagina following hysterectomy.
Hysterectomy	Surgical removal of the uterus.
Hysteropexy(sacrohysteropexy)	A surgical procedure for correcting prolapse of the uterus in women who wish to preserve their uterus.
Nulliparous	Having never given birth.
Parous	Having given birth at least once.
Pelvic floor muscle training (PFMT)	Intensive and regular pelvic floor exercises for strengthening the muscles of the pelvic floor.
Pelvic organ prolapse	Bulging or herniation of one or more pelvic organs (uterus, vagina, bowel, or bladder) into or out of the vagina.
Posterior vaginal wall	The back wall of the vagina (posterior compartment).
Sacrocolpopexy	A surgical procedure for correcting prolapse of the vaginal vault in women who have had a hysterectomy. The vaginal vault is supported by using mesh attached to the sacrum (lower part of the spine).
Sacrospinous fixation	A surgical procedure to restore support to the uterus or vaginal vault, where sutures are placed to attached the vaginal vault or cervix to the sacrospinous ligament in the pelvis.
Topical oestrogens	Oestrogen applied directly to the vagina, either in the form of a vaginal tablet, cream, gel, or ring pessary.
Vaginal epithelium	The inner lining of the vagina.
Vaginal pessaries	Vaginal devices that come in various shapes and sizes to provide mechanical support to the prolapsed organs.
Vaginal vault	The apex or top of the vagina.

Appendix 2. Abbreviations: outcome measurement tools

Abbreviation	Definition
APFQ	Australian Pelvic Floor Questionnaire
AQoL	Assessment of Quality of Life
ePAQ‐PF	Electronic Personal Assessment Questionnaire Pelvic Floor
GQOL	Global Quality of Life Scale
ICIQ‐FLUTS	International Consultation on Incontinence Questionnaire Female Lower Urinary Tract Symptoms Modules
ICIQ‐UI‐SF	International Consultation on Incontinence Questionnaire‐Urinary Incontinence Short Form
ICIQ‐VS	International Consultation on Incontinence Questionnaire Vaginal Symptoms Module
KHQ	King's Health Questionnaire
P‐QOL	Prolapse Quality of Life Questionnaire
PFDI	Pelvic Floor Distress Inventory
PFIQ	Pelvic Floor Impact Questionnaire
PGI‐I	Patient Global Impression of Improvement
PISQ‐IR	Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire, International Urogynecological Association‐Revised
POP‐Q	Pelvic Organ Prolapse Quantification System
POP‐SS	Pelvic Organ Prolapse Symptom Score
QLQ	Quality of Life Questionnaire
QOLI	Quality of Life Inventory
VAS	Visual analogue scale

Appendix 3. Search of the Cochrane Incontinence Specialised Register – for clinical effectiveness studies

The search terms used to search the Cochrane Incontinence Specialised Register are given below:

((design.cct*) OR (design.rct*))
AND
topic.prolapse*
AND
intvent.chem.horm*

All searches were of the 'keywords' field of EndNote 2018.

Appendix 4. Search methods for the brief economic commentary (BEC)

We performed electronic searches designed to identify published reports of relevant economic evaluations to inform the BEC (see 'Incorporating economic evidence' in the Methods section).

We searched the Cochrane Incontinence Specialised Register of Economic Evaluations (EEs) which, for pelvic organ prolapse, consists of searches of the NHS Economic Evaluation Database (NHS EED) on the UK Centre for Reviews and Dissemination (CRD) website (covering from the earliest record in NHS EED, dating from 1968, up to and including 31 December 2014 when their coverage ended).

As NHS EED is no longer actively updated, we performed additional searches of the following databases to identify eligible studies added to these databases from 1 January 2015 onwards.

• MEDLINE on OvidSP (1946 to May Week 1 2022)

• Embase on OvidSP (1974 to 2022 Week 19)

We used the following terms to search the Cochrane Incontinence Specialised Register of Economic Evaluations (EEs):

• topic.prolapse*AND intvent.chem.horm*

All searches were of the 'keywords' field of EndNote 2018. The date of the most recent search was 16 May 2022.

The economic evaluation search filters that we applied to our MEDLINE and Embase search strategies were those formerly used by the CRD to identify published reports of full economic evaluations for indexing on NHS EED. These economic evaluation search filters remain freely available on the CRD Database website (CRD 2015). The other search lines in the MEDLINE and Embase search strategies were adapted from the electronic search strategies run for our Cochrane Incontinence Specialised Register of clinical effectiveness studies, along with additional terms for this population developed specifically for this review. Similarly, our NHS EED search strategy was adapted from search strategies run for our Specialised Register and based on text word and MeSH terms (capturing relevant P‐I‐C concepts) used to identify eligible studies of intervention effects. We followed the current economic methods guidance (Aluko 2022).

The search for the Cochrane Incontinence Specialised Register of EEs was only current up to 7 April 2021, so one of the review authors (AK) screened an updated search of MEDLINE and Embase (conducted on 16 May 2022) to bring the Specialised Register search for EEs up to date. The search strategies used in MEDLINE and Embase to populate the Cochrane Incontinence Specialised Register of EEs are presented below.

The original search of NHS EED to begin populating the Specialised Register of EEs is provided at the end of this Appendix.

MEDLINE on OvidSP (covering 1 January 1946 to May Week 1 2022) was searched using the strategy given below.

1. Economics/

2. exp "costs and cost analysis"/

3. Economics, Dental/

4. exp economics, hospital/

5. Economics, Medical/

6. Economics, Nursing/

7. Economics, Pharmaceutical/

8. (economic$ or cost or costs or costly or costing or price or prices or pricing or pharmacoeconomic$).ti,ab.

9. (expenditure$ not energy).ti,ab.

10. value for money.ti,ab.

11. budget$.ti,ab.

12. or/1‐11

13. ((energy or oxygen) adj cost).ti,ab.

14. (metabolic adj cost).ti,ab.

15. ((energy or oxygen) adj expenditure).ti,ab.

16. or/13‐15

17. 12 not 16

18. letter.pt.

19. editorial.pt.

20. historical article.pt.

21. or/18‐20

22. 17 not 21

23. exp animals/ not humans/

24. 22 not 23

25. prolapse/

26. uterine prolapse/

27. Rectocele/

28. (prolaps$ adj5 (pelvi$ or vagin$ or genit$ or uter$ or vault$ or apical or urethr$ or segment$ or wall$)).tw.

29. cystoc?ele$.tw.

30. rectoc?ele$.tw.

31. urethroc?ele$.tw.

32. enteroc?ele$.tw.

33. proctoc?ele$.tw.

34. sigmoidoc?ele$.tw.

35. (pelvi$ adj3 dysfunct$).tw.

36. (pelvi$ adj3 (disorder$ or relax$)).tw.

37. (vagin$ adj3 defect$).tw.

38. (urogen[SW1] ital adj5 prolaps$).tw.

39. (cervi$ adj5 prolaps$).tw.

40. exp Pelvic Organ Prolapse/

41. Pelvic Floor/

42. Pelvic Floor Disorders/

43. (pelvi$ adj2 floor[SW2] ).tw.

44. perineomet$.tw.

45. (prolaps$ adj5 rect$).tw.

46. or/25‐45

47. 24 and 46

Embase (on OvidSP) (covering 1 January 1974 to 2022 Week 19) was searched using the strategy given below.

1. Health Economics/

2. exp Economic Evaluation/

3. exp Health Care Cost/

4. pharmacoeconomics/

5. (econom$ or cost or costs or costly or costing or price or prices or pricing or pharmacoeconomic$).ti,ab.

6. (expenditure$ not energy).ti,ab.

7. (value adj2 money).ti,ab.

8. budget$.ti,ab.

9. or/1‐8

10. letter.pt.

11. editorial.pt.

12. note.pt.

13. or/10‐12

14. 9 not 13

15. (metabolic adj cost).ti,ab.

16. ((energy or oxygen) adj cost).ti,ab.

17. ((energy or oxygen) adj expenditure).ti,ab.

18. 15 or 16 or 17

19. 14 not 18

20. animal/

21. exp animal experiment/

22. nonhuman/

23. (rat or rats or mouse or mice or hamster or hamsters or animal or animals or dog or dogs or cat or cats or bovine or sheep).ti,ab,sh.

24. 20 or 21 or 22 or 23

25. exp human/

26. human experiment/

27. 25 or 26

28. 24 not (24 and 27)

29. 19 not 28

30. conference abstract.pt.

31. 29 not 30

32. Uterus Prolapse/

33. Rectocele/

34. Vagina Prolapse/ or Cystocele/ or enterocele/

35. prolapse/ or pelvic floor prolapse/ or exp pelvic organ prolapse/ or perineal descent/ or exp visceral prolapse/

36. (prolaps$ adj5 (pelvi$ or vagin$ or genit$ or uter$ or vault$ or apical or urethr$ or segment$ or wall$)).tw.

37. (prolaps$ adj5 rect$).tw.

38. cystoc?ele$.tw.

39. rectoc?ele$.tw.

40. urethroc?ele$.tw.

41. enteroc?ele$.tw.

42. proctoc?ele$.tw.

43. sigmoidoc?ele$.tw.

44. (pelvi$ adj3 dysfunct$).tw.

45. (pelvi$ adj3 (disorder$ or relax$)).tw.

46. (vagin$ adj3 defect$).tw.

47. (urogenital adj5 prolaps$).tw.

48. (cervi$ adj5 prolaps$).tw.

49. (pelvi$ adj2 floor).tw.

50. perineomet$.tw.

51. pelvis floor/

52. or/32‐51

53. 31 and 52

NHS Economic Evaluation Database (NHS EED) on the UK Centre for Reviews and Dissemination (CRD) website was searched on 13 February 2020 (in all fields) using the strategy given below.

Search	Hits
1	(MeSH DESCRIPTOR Prolapse EXPLODE ALL TREES) IN NHSEED
2	(MeSH DESCRIPTOR Cystocele EXPLODE ALL TREES) IN NHSEED
3	(MeSH DESCRIPTOR Pelvic Organ Prolapse EXPLODE ALL TREES) IN NHSEED
4	(MeSH DESCRIPTOR Rectal Prolapse EXPLODE ALL TREES) IN NHSEED
5	(MeSH DESCRIPTOR Uterine Prolapse EXPLODE ALL TREES) IN NHSEED
6	(MeSH DESCRIPTOR Visceral Prolapse EXPLODE ALL TREES) IN NHSEED
7	(MeSH DESCRIPTOR Rectocele EXPLODE ALL TREES) IN NHSEED
8	(MeSH DESCRIPTOR Pelvic Floor EXPLODE ALL TREES) IN NHSEED
9	(MeSH DESCRIPTOR Pelvic Floor Disorders EXPLODE ALL TREES) IN NHSEED
10	((pelvi* ADJ5 prolaps*)) OR ((prolaps* ADJ5 pelvi*)) IN NHSEED
11	((vagin* ADJ5 prolaps*)) OR ((prolaps* ADJ5 vagin*)) IN NHSEED
12	((genit* ADJ5 prolaps*)) OR ((prolaps* ADJ5 genit*)) IN NHSEED
13	((uter* ADJ5 prolaps*)) OR ((prolaps* ADJ5 uter*)) IN NHSEED
14	((vault* ADJ5 prolaps*)) OR ((prolaps* ADJ5 vault*)) IN NHSEED
15	((apical ADJ5 prolaps*)) OR ((prolaps* ADJ5 apical)) IN NHSEED
16	((urethr* ADJ5 prolaps*)) OR ((prolaps* ADJ5 urethr*)) IN NHSEED
17	((segment* ADJ5 prolaps*)) OR ((prolaps* ADJ5 segment*)) IN NHSEED
18	((wall* ADJ5 prolaps*)) OR ((prolaps* ADJ5 wall*)) IN NHSEED
19	((rect* ADJ5 prolaps*)) OR ((prolaps* ADJ5 rect*)) IN NHSEED
20	((urogen* ADJ5 prolaps*)) OR ((prolaps* ADJ5 urogen*)) IN NHSEED
21	((cervi* ADJ5 prolaps*)) OR ((prolaps* ADJ5 cervi*)) IN NHSEED
22	((pelvi* ADJ5 dysfunct*)) OR ((dysfunct* ADJ5 pelvi*)) IN NHSEED
23	((pelvi* ADJ5 disorder*)) OR ((disorder* ADJ5 pelvi*)) IN NHSEED
24	((pelvi* ADJ5 relax*)) OR ((relax* ADJ5 pelvi*)) IN NHSEED
25	((pelvi* ADJ5 floor*)) OR ((floor* ADJ5 pelvi*)) IN NHSEED
26	((vagin* ADJ5 defect*)) OR ((defect* ADJ5 vagin*)) IN NHSEED
27	(perineomet*) IN NHSEED
28	(cystocele*) OR (cystocoele*) IN NHSEED
29	(rectocele*) OR (rectocoele*) IN NHSEED
30	(urethrocele*) OR (urethrocoele*) IN NHSEED
31	(enterocele*) OR (enterocoele*) IN NHSEED
32	(proctocele*) OR (proctocoele*) IN NHSEED
33	(sigmoidocele*) OR (sigmoidocoele*) IN NHSEED
34	#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33

Key: $ or * = truncation symbol; tw. = titles and abstracts; / = MeSH or Emtree term; exp = exploded MeSH or Emtree term; adjn = within n words, in any order on Ovid, in fixed order on CRD; ? character may or may not be present.

Data and analyses

Comparison 1. Oestrogen therapy (OT) in conjunction with vaginal pessaries versus vaginal pessaries alone.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Participant adherence to the intervention (including discontinuation of pessary use and declining surgical procedures)	0		Other data	No numeric data
1.2 Reduction in minor adverse events: vaginal events	3	185	Risk Ratio (M‐H, Fixed, 95% CI)	0.48 [0.25, 0.91]
1.3 Major adverse events	0		Other data	No numeric data

Comparison 2. Oestrogen therapy (OT) in conjunction with surgery versus surgery alone.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms: 6 months	0		Other data	No numeric data
2.2 Participant‐reported subjective improvement or cure of pelvic organ prolapse symptoms: 12 months	0		Other data	No numeric data
2.3 Improvement or cure of associated urinary symptoms: 6 months	0		Other data	No numeric data
2.4 Improvement or cure of associated urinary symptoms: 12 months	0		Other data	No numeric data
2.5 Improvement or cure of associated bowel symptoms: 6 months	0		Other data	No numeric data
2.6 Improvement or cure of associated bowel symptoms: 12 months	0		Other data	No numeric data
2.7 Improvement or cure of associated sexual symptoms: 6 months	2	211	Mean Difference (IV, Random, 95% CI)	‐1.02 [‐3.52, 1.48]
2.8 Improvement or cure of associated sexual symptoms: 12 months	2	207	Mean Difference (IV, Fixed, 95% CI)	‐0.57 [‐1.98, 0.85]
2.9 Objective improvement or cure of pelvic organ prolapse: 6 months (meta‐analysis)	2	241	Risk Ratio (M‐H, Fixed, 95% CI)	1.01 [0.94, 1.08]
2.10 Objective improvement or cure of pelvic organ prolapse: 6 months (other data)	0		Other data	No numeric data
2.11 Objective improvement or cure of pelvic organ prolapse: 12 months	0		Other data	No numeric data
2.12 Condition‐specific quality of life: 6 months	0		Other data	No numeric data
2.13 Condition‐specific quality of life: 12 months	0		Other data	No numeric data
2.14 Participant adherence to intervention (meta‐analysis)	2	114	Risk Ratio (M‐H, Random, 95% CI)	1.07 [0.49, 2.32]
2.15 Participant adherence to intervention (other data)	0		Other data	No numeric data
2.16 Reduction in minor adverse events: urinary tract infection	5	483	Risk Ratio (M‐H, Random, 95% CI)	0.49 [0.26, 0.90]
2.17 Reduction in minor adverse events: vaginal events	4	371	Risk Ratio (M‐H, Fixed, 95% CI)	2.23 [0.74, 6.79]
2.18 Reduction in minor adverse events: urinary storage events	2	289	Risk Ratio (M‐H, Fixed, 95% CI)	0.94 [0.46, 1.92]
2.19 Reduction in minor adverse events: sexual adverse events	0		Other data	No numeric data
2.20 Major adverse events: total number of events	5	453	Risk Ratio (M‐H, Fixed, 95% CI)	1.02 [0.39, 2.66]
2.21 Surgical outcomes: operative time	0		Other data	No numeric data
2.22 Surgical outcomes: mesh exposure	0		Other data	No numeric data

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Baracat 2001.

Study characteristics
Methods	Design: RCT Dates study conducted: NR
Participants	Number of participants: 8 Country: Brazil Setting: single centre Severity of POP: NR Affected compartment: uterine Parity: NR Hysterectomy: NR Age: 45 to 55 BMI: NR Inclusion criteria Age 44–55 years Prolapsed uterus ≥ 5 years after menopause Exclusion criteria Use of any other hormonal treatment Endocrinological disease Any other medication
Interventions	Group I (n = 2): oral CEO 0.625 mg/day, administered for 28 days before vaginal hysterectomy to correct the prolapsed uterus Group II (n = 2): oral CEO 1.25 mg/day, administered for 28 days before vaginal hysterectomy to correct the prolapsed uterus Group III (n = 2): transdermal 17 beta‐oestradiol 50 μg/day, administered for 28 days before vaginal hysterectomy to correct the prolapsed uterus Group IV (n = 2): no hormonal treatment (control group) Co‐interventions: surgery for all participants (vaginal hysterectomy)
Outcomes	Ultrastructural analysis
Notes	ITT analysis: NR Power calculation: NR Length of follow‐up: NR Conflicts of interest: NR Study funding sources: NR
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "randomly divided into four groups". Comment: not explained how women were randomised.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not reported.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All samples were assessed.
Selective reporting (reporting bias)	Low risk	Seems to report all outcomes stated.
Other bias	Low risk	Nothing to suggest any other sources of bias.

Caruso 2017.

Study characteristics
Methods	Design: RCT Dates study conducted: January 2014–February 2015
Participants	Number of participants: 75 Country: Italy Setting: single centre (Obstetrical Gynecological Unit, Department of General Surgery and Medical Surgical Specialities, School of Medicine, University of Catania) Severity of POP Total: stage II: n = 19 (25.3%); stage III: n = 32 (42.6%); stage IV: n = 24 (32%) Group I: stage II: n = 10 (26.3%); stage III: n = 16 (42.1%); stage IV: n = 12 (31.6%) Group II: stage II: n = 9 (24.3%); stage III: n = 18 (48.4%); stage IV: n = 10 (27.1%) Affected compartment: cystocele and uterine Parity 1 child: group I: n = 3 (7.9%); group II: n = 2 (5.4%) ≥ 2 children: group I: n = 35 (92.1%); group II: n = 35 (94.6%) Hysterectomy: none in either group Age: group I: mean 60.3 years (SD 3.7); group II: mean 59.5 years (SD 4.3); Total: mean 58.6 years (SD 3.5) BMI: group I: mean 25.8 kg/m2 (SD 4.8); group II: mean 26.7 kg/m2 (SD 5.5) Inclusion criteria Natural menopause Relevant vaginal bulging because of POP requiring surgical repair Symptoms or signs of GSM Genital dryness, irritation/burning/itching of the vulva or vagina, decreased lubrication during sexual activity, discomfort or pain during sexual activity, decreased arousal or orgasm desire, or dysuria Exclusion criteria Previous POP surgery Urinary incontinence BMI ≥ 35 kg/m2 Endometrial thickness ≥ 4 mm measured by transvaginal ultrasound before study initiation Abdominal uterine bleeding Hormone‐dependent malignancies History of thromboembolic disease Liver disease Use of systemic or local hormone therapy for < 3 months Use of phytoestrogens or non‐hormonal moisturisers or lubricants within 1 month before the start of the study Partner affected by a sexual disorder Unhappy relationship with partner
Interventions	Group I (n = 38): estriol vaginal gel 1 g (containing 50 ug of estriol (0.005%)). Once daily for 3 weeks then twice weekly for a further 9 weeks for a complete cycle of treatment of 12 weeks. Participants were invited to administer the gel at night by inserting the applicator deep inside the vagina. These participants were re‐randomised 1 week after surgery to either continue receiving oestrogen (group 1A) or discontinue oestrogen therapy (group 1B). Group II (n = 37): no vaginal gel Group IA (n = 19): estriol vaginal gel 1g (containing 50 ug of estriol (0.005%)) after surgery. Once daily for 3 weeks then twice weekly for a further 9 weeks for a complete cycle of treatment of 12 weeks. Participants were invited to administer the gel at night by inserting the applicator deep inside the vagina. Group IB (n = 19): no vaginal gel after surgery Co‐interventions: surgery for all participants. Each participant underwent vaginal hysterectomy with surgical wall epithelium and anterior colporrhaphy. Colpoperineoplasty was performed if indicated. The same team of 3 surgeons performed all of the operations.
Outcomes	Improvement in GSM (including vaginal dryness and dysuria): severity of each symptom was self‐assessed and received a grade on a 4‐point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe). Change in severity was used to evaluate symptomatic improvement. Effect of estriol gel on vaginal epithelium: measured using VMI. Vaginal smears were taken and sent to the Cytological Laboratory of the Institute of Pathological Anatomy for staining, followed by analysis. Vaginal pH: measured by inserting a pH test strip into the upper wall of the vagina. pH levels > 5.0 associated with decreased serum E2. Vaginal moisture, vaginal fluid secretion, type and consistency, overall elasticity, appearance of the vaginal mucosa: each assessed using VHI (range 5–25) on scales ranging from 1 (none) to 5 (excellent) and the mean sum calculated. Quality of life: assessed using SF‐36. Women were instructed to place a mark on a 0–100 scale and mean values were calculated. Higher scores = better functioning. Sexual function: obtained via interview and assessed using the PISQ‐12 (composed of 12 items examining 3 factors). Only totals used. Sexual events: recorded using a participant diary. Adverse events: recorded using a participant diary.
Notes	ITT analysis: NR Power calculation: NR Length of follow‐up: part A: 12 weeks (before surgery); part B: 12 weeks (from 1 week after surgery) Conflicts of interest: "None reported" Study funding sources: "None reported" Correspondence: we contacted the study authors on 25 October 2021 and again on 9 November 2021 for further information regarding the PISQ‐12 and SF‐36 assessments, but we received no response.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "underwent randomization 1:1". Comment: unclear method of randomisation.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants in the control groups would have been aware that they were not receiving any additional treatment; unclear whether personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	At the end of the first phase of the study, all participants were followed up. At the end of the second phase, 17/19 in Group 1 and 16/19 in Group 2 were followed up. No reasons for loss to follow‐up given, but the attrition is not substantial or imbalanced between groups.
Selective reporting (reporting bias)	High risk	Participants all had transvaginal scans for endometrial thickness but the results were not reported.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Chiengthong 2022.

Study characteristics
Methods	Design: RCT Dates study conducted: April 2018–August 2020
Participants	Number of participants: 78 Country: Thailand Setting: single centre (Female Pelvic Medicine and Reconstructive Clinic at Chulalongkorn Memorial Hospital, Bangkok) Severity of POP Group I: stage II: n = 8 (20.5%); stage III: n = 16 (41.0%); stage IV: n = 15 (38.5%) Group II: stage II: n = 2 (5.1%); stage III: n = 19 (48.7%); stage IV: n = 18 (46.2%) Affected compartment: NR Parity Nulliparity: group I: n = 0; group II: n = 6 (15.4%) ≥ 1 child: group I: n = 35 (reported as 100%); group II: n = 33 (84.6%) Hysterectomy: group I: 6 (15.4%); group II: 6 (15.4%) Age: group I: mean 71.2 years (SD 6.8); group II: mean 67.1 years (SD 9.3) BMI: group I: mean 24.7 kg/m2 (SD 3.5); group II: mean 24.8 kg/m2 (SD 3.5) Inclusion criteria Postmenopausal Thai women Symptomatic POP Ability to complete the questionnaire Successful pessary fitting Exclusion criteria History of vaginal bleeding of unknown origin Oestrogen‐dependent cancers Immunosuppression Clinically manifested sexually transmitted diseases Bacterial vaginosis Trichomoniasis Vulvovaginal candidiasis Known hypersensitivity to the anti‐infective agent of test preparation
Interventions	Group I (n = 39): intravaginal oestriol 0.03 mg plus Lactobacillus acidophilus 100 million viable cell tablets (Gynoflor): daily for 12 days followed by 1 tablet on Monday and Thursday for 12 weeks Group II (n = 39): no additional intervention Co‐interventions: pessary for all participants (ring pessary without support, ring pessary with support, Gellhorn pessary, or and cube pessary). All participants were instructed to remove and clean their pessary every day.
Outcomes	Primary outcome Incidence of bacterial vaginosis: measured by the Amsel criteria during vaginal examination by a single investigator using a cotton swab on the vaginal wall. Criteria include the presence of a homogenous vaginal discharge, vaginal pH > 4.5, ≥ 20% clue cells per high‐power field, and a positive whiff test. Bacterial vaginosis was established if 3 or 4 criteria were met. Secondary outcomes Normal vaginal flora index: determined by numbers of lactobacilli, pathogenic microorganisms, and leucocytes examined after gram staining and scored from 0–3 for each parameter. Vaginal pH measured from 0–3. Higher total score = more vaginal flora. Objective assessment of vaginal symptoms, associated sexual matters and impact on quality of life: measured by the Thai version of ICIQ‐VS questionnaire Severity of vaginal symptoms: measured using a VAS for pain Incidence of vaginal abrasion and vaginal bleeding: method not reported
Notes	ITT analysis: NR Power calculation: yes: "Sample size was calculated by using a 32% incidence of bacterial vaginosis in pessary users and the incidence of bacterial vaginosis in women using a pessary (with intravaginal estrogen administration) from the pilot study (5%). The level of significance (alpha = 5%) and a power of 80% were used. A 20% dropout rate was considered. Total participants required for this study were 39 women per group." Length of follow‐up: 2 weeks and 14 weeks Conflicts of interest: "None" Study funding sources: "This study was funded by the Ratchadapisek Sompoch research grant (RA61/027)."
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization was done in a 1:1 ratio by computer generated blocks (block size of four)". Comment: adequate method.
Allocation concealment (selection bias)	Low risk	Quote: "The allocation was concealed with an opaque sealed envelope". Comment: adequate method of concealment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants in the control groups would have been aware that they were not receiving any additional treatment; unclear whether personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "The blind analysis was conducted by a single laboratory technician from the Microbiology Department of King Chulalongkorn Memorial Hospital" (for normal flora). Comment: other outcomes unclear.
Incomplete outcome data (attrition bias) All outcomes	Low risk	At the 14‐week follow‐up, 35/39 in the intervention group and 32/39 in the control group were lost to follow‐up; reasons are explained in the flow diagram. Although no ITT was performed, attrition was not largely differential between the 2 groups.
Selective reporting (reporting bias)	Low risk	All outcomes from the trial registration seem to be reported.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Coelho 2020.

Study characteristics
Methods	Design: RCT Dates study conducted: August 2018–October 2019
Participants	Number of participants: 98 Country: Brazil Setting: single centre ("Urogynecology Outpatient Clinic of the University of Campinas (UNICAMP) in the Department of Obstetrics and Gynecology") Severity of POP: POP‐Q staging Anterior 0: oestrogen: n = 1 (2.17%); no oestrogen: n = 1 (1.92%) Anterior 1: oestrogen: n = 0; no oestrogen: n = 2 (5.77%) Anterior 2: oestrogen: n = 4 (8.70%); no oestrogen: n = 4 (7.69%) Anterior 3/4: oestrogen: n = 41 (89.13%); no oestrogen: n = 44 (84.62%) Posterior 0: oestrogen: n = 26 (56.5%); no oestrogen: n = 24 (46.1%) Posterior 1: oestrogen: n = 5 (10.8%); no oestrogen: n = 5 (9.6%) Posterior 2: oestrogen: n = 0; no oestrogen: n = 4 (7.6%) Posterior 3/4: oestrogen: n = 15 (32.5%); no oestrogen: n = 19 (36.4%) Apical 0: oestrogen: n = 2 (4.3%); no oestrogen: n = 1 (1.9%) Apical 1: oestrogen: n = 3 (6.52%); no oestrogen: n = 4 (7.69%) Apical 2: oestrogen: n = 15 (32.6%); no oestrogen: n = 7 (13.4%) Apical 3/4: oestrogen: n = 26 (56.5%); no oestrogen: n = 40 (76.8%) Affected compartment: anterior, posterior, and apical in different severities across groups (see above) Parity: oestrogen: mean 4.28 births (SD 2.4); no oestrogen: mean 4.85 births (SD 3.2) Hysterectomy: NR Age: oestrogen: mean 69.8 years (SD 8.0); no oestrogen: mean 68 years (SD 8.9) BMI: oestrogen: mean 26.5 kg/m2 (SD 3.8); No oestrogen: mean 25.8 kg/m2 (SD 3.6) Inclusion criteria Menopause POP symptoms and POP‐Q score stage 3 or 4 Informed consent Successfully fitted support pessary Exclusion criteria Contraindications for local oestrogen (e.g. known or suspected breast/endometrium cancer or thromboembolic disorders) Systemic hormonal therapy Inability to follow protocol Uncontrolled or untreated psychiatric disorders Refusal to participate
Interventions	Group I (n = 46): topical oestrogen cream. Promestirene 10 mg/g cream, 0.5 g 3 times a week. Participants were advised to stop using the cream 3 days before vaginal sampling collection and not to have sexual intercourse 24 hours before the vaginal examination. Group II (n = 52): no oestrogen cream or lubricant. This group followed the same protocol as the intervention group regarding avoidance of sexual intercourse prior to vaginal examination. Co‐interventions: pessary. All participants used a standardised ring format "so that selection bias could be reduced." All pessaries were made of silicone with a varying diameter in 10 different sizes from 57 mm to 82 mm. The same physician chose the ring pessary's size. The size of the pessary was considered to be correct when the physician could place a single finger between the pessary and the vaginal wall, prolapse was reduced to above the hymen, the participant felt comfortable following insertion, and the participant could retain the pessary after a Valsalva manoeuvre in both standing and supine positions. Participants were instructed on how to prevent expulsion and the correct position during defecation. The pessary was removed and cleaned during a medical appointment every 3 months. Quote: "In Brazil, many patients may be unwilling or unable to manage pessary care themselves. Therefore, in this study, we advised the women not to periodically remove or replace the pessary themselves."
Outcomes	Primary outcomes Vaginal discharge, vaginal odour, itching, and vaginal ulceration: assessed 3 and 6 months after pessary insertion. Participants were asked to return to the clinical if vaginal odour, discharge, or bleeding/ulceration occurred. They underwent a physical examination and vaginal discharge analysis; they were also aware that these symptoms would require investigation for infection or erosion. Secondary outcomes Stress urinary incontinence: defined by IUGA/ICS consensus Urinary urgency: defined by IUGA/ICS consensus Urgency incontinence: defined by IUGA/ICS consensus Urgency incontinence: defined by IUGA/ICS consensus Increased urinary frequency: defined by IUGA/ICS consensus Nocturia: defined by IUGA/ICS consensus Nocturnal enuresis: defined by IUGA/ICS consensus Adherence to intervention: monitored by a researcher once a month (note: measured but not listed with other outcomes)
Notes	ITT analysis: yes Power calculation: yes: "The sampling size was calculated using Intercooled Stata version 13.0. The process used a two‐sided 5% significance level, 80% study power, and a maximum difference of 25% in the proportion between the estrogen and control groups that was found for vaginal discharge in previous studies; the sample size obtained was 98 women." Length of follow‐up: 3 and 6 months Conflicts of interest: "None" Study funding sources: NR Other: While the inclusion criteria say "menopausal", the mean ages of the women involved are above what would be considered postmenopausal. Quote: "In the case of serious adverse events (fibrosis, fistulas, bilateral hydroephrosis with urosepsis, bowel obstruction, ureteric obstruction, hydronephrosis, or cancer), either from the pessary or related to oestrogen/control groups, women were removed from the study and received appropriate treatment". 14 women (30.4%) in the oestrogen group and 13 women (25%) in the no oestrogen group had previously had surgery, but it is unclear what procedure this was.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "After a successful initial fitting, a computer‐generated randomization table (a block of 10) was used to allocate participants to one of the groups (estrogen or control). The randomization sequence was kept locked, and one researcher performed this step". Comment: adequate method of randomisation.
Allocation concealment (selection bias)	Low risk	Quote: "The allocation sequence was concealed from the researcher responsible for enrolling and assessing participants. Numbers were inserted sequentially in opaque, sealed, and stapled envelopes. The corresponding envelopes were opened only after the enrolled participants completed all of the baseline assessments and it was time to allocate an intervention". Comment: adequate method of allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	The study notes that it is "single blind," likely the outcome assessor, but blinding of participants and personnel not reported. However, would not have been possible to blind participants to intervention or control.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "During their follow‐up consultations (after 3 and 6 months), the blinded intervention group health professional assessed the women by using a questionnaire regarding the presence of vaginal symptoms, a physical examination, and vaginal sampling for microbiological analysis". Comment: outcome assessor likely blinded.
Incomplete outcome data (attrition bias) All outcomes	High risk	38/46 in intervention group and 40/52 in control group completed and were analysed. No major differential losses between groups, but overall attrition rate across the study was more than 20%.
Selective reporting (reporting bias)	High risk	Trial registration available. No mention of the urinary outcomes in the trial registration; some of the outcomes from the physical examination and microbiological analysis in the manuscript are also not reported in the trial registration. Unclear why these outcomes were added. Adherence and medication use was monitored in a patient diary and by researcher contact but results of this not reported.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Felding 1992.

Study characteristics
Methods	Design: RCT Dates study conducted: NR
Participants	Number of participants: 48 Country: Denmark Setting: multicentre (2 gynaecological departments at University Hospitals (Gentofte and Glostrup County Hospitals)) Severity of POP: NR Affected compartment: NR Parity: NR Hysterectomy: NR Age (measure not clear): group I: 63.5 years (range 44–82); group II: 64.0 years (range 50–77) BMI: NR Inclusion criteria postmenopause (≥ 1 year) no hormone treatment for ≥ 3 months prior to admission Exclusion criteria Previous breast or uterine malignancy or other hormone‐dependant neoplasms Genital bleeding of unknown origin Acute thrombophlebitis or earlier thrombo‐embolic episodes in relation to oestrogen therapy Cardiac insufficiency treated with digitalis Use of steroid medication
Interventions	Group I (n = 22): oestrogen pessary. Participants used a 6‐mm diameter pessary (designed to adhere to the vaginal mucosa) for 3 weeks preoperatively. Pessaries contained 25 ug oestradiol in a hydrophilic cellulose‐derived matrix (Vagifem). The pessaries were inserted deep into the vagina with a specially designed applicator. All participants also received PFME training from a study author. Group II (n = 23): placebo pessary. Participants used a 6‐mm diameter pessary (designed to adhere to the vaginal mucosa) for 3 weeks preoperatively (Novo‐Nordisk A/S, Copenhagen, Denmark). The pessaries were inserted deep into the vagina with a specially designed applicator. All participants also received PFME training from a study author. Co‐interventions: surgery for all participants. Either a vaginal hysterectomy, anterior and posterior colporrhaphy, a Manchester‐Fothergill operation, or a combination of these. All participants treated with an indwelling catheter for 3 to 5 days postoperatively. Everyone also offered PFME.
Outcomes	Hormonal status of the vagina: evaluated cytologically and histologically Vaginal epithelium status: evaluated histologically at a punch biopsy taken from the lateral vaginal wall before and after treatment Possible effect on the endometrium: evaluated by a specimen obtained by current age at surgery Changes in systemic hormonal markers: evaluated by taking venous blood samples before treatment, after 3 weeks of therapy, and 4 weeks after surgery. Serum levels of oestradiol and follicle‐stimulating hormone measured by radio‐immunoassay.
Notes	ITT analysis: NR Power calculation: NR Length of follow‐up: 3 weeks, 7 weeks, 3 years Conflicts of interest: NR Study funding sources: NR
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "randomised, double‐blind design". Comment: unclear method of randomisation.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Quote: "double‐blind". Comment: participants received pessary and either oestrogen or placebo, but unclear if they were blinded because there are insufficient details of administration alongside the pessary. Unclear if personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "All smears and biopsies were evaluated by the same pathologist without knowledge of the treatment". Comment: although the details of the blood tests are not reported, assumed low as the major outcome measure assessments were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Two patients (8%) in the treatment group and one (4%) patient in the placebo group declined surgery […] leaving 22 patients in the treatment group and 23 patients in the placebo group". Comment: attrition is low and not differential, and reasons for dropouts are explained. Follow‐up rate for 3‐year questionnaire was 93%, also suggesting low overall attrition.
Selective reporting (reporting bias)	Low risk	All outcomes prespecified in the methods seem to be reported.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Karp 2012.

Study characteristics
Methods	Design: RCT Dates study conducted: October 2008–January 2010
Participants	Number of participants: 65 Country: USA Setting: single centre (Urogynecology and Reconstructive Pelvic Surgery service within the Department of Gynecology at Cleveland Clinic Florida) Severity of POP: group I: median prolapse stage IV (range II–IV); group II: median prolapse stage IV (range II–IV); group III: median prolapse stage IV (range III–IV) Affected compartment: NR Parity: group I: median 3 vaginal births (range 1–7); group II: median 2.5 vaginal births (range 1–4); group III: median 2 vaginal births (range 0–5) Hysterectomy: group I: n = 6 (27%); group II n = 6 (33%); group III n = 7 (32%) Age: group I: mean 65 years (SD 7.4); group II: mean 66 years (SD 7.9); group III: mean 65 years (SD 7.8) BMI: group I: mean 26.7 kg/m2 (SD 3.2); group II: 26.8 kg/m2 (SD 3.4); group III: 28.7 kg/m2 (SD 4.7) Inclusion criteria ≥ 2 years since spontaneous or surgical menopause Symptomatic urogenital atrophy and POP Decision to undergo reconstructive vaginal surgery ≥ 1 symptom (vaginal dryness, vulvar pruritis, dyspareunia, dysuria, urinary urgency) or sign (vaginal pallor, petechiae, friability, vaginal dryness) of atrophic vaginitis Exclusion criteria Contraindication to oestrogen use (vaginal bleeding, oestrogen‐dependent cancers, hepatic or thrombotic disease) Allergy to silicone or oestradiol Absence of vulvovaginal atrophy on examination Vaginal pH ≤ 4.0 Use of vaginal or systemic oestrogen in the previous 6 months
Interventions	Group I (n = 22): oestradiol‐releasing vaginal ring. Estring placed vaginally immediately after reconstructive surgery and vaginal packing left in place for 24 hours postoperatively. Group II (n = 21): placebo vaginal ring. Vaginal ring placed vaginally immediately after reconstructive surgery and vaginal packing left in place for 24 hours postoperatively. The placebo ring was of identical size and shape to the Estring, and was manufactured and provided free of charge by the pharmaceutical company (Pfizer, New York City, NY). Group III (n = 22): no vaginal ring (control) Co‐interventions: surgery for all participants. All participants underwent reconstructive vaginal surgery, including hysterectomy and "kit" reconstructive procedures (including Apogee, Perigree, and Elecative with Intexen), concomitant anti‐incontinence surgery with synthetic suburethral sling (including Monarc transobturator sling, tension‐free vaginal tape, and I‐Stop). Native tissue‐sutured reconstruction included native tissue anterior and posterior colporrhaphy, McCall colposuspension and high uterosacral suspension, and vaginal enterocoele repair.
Outcomes	Primary outcome Tissue quality after surgery: measured by vaginal maturation index (primary surrogate marker for tissue quality). Vaginal smears were used to determine the vaginal maturation index. Maturation value of 0–49 indicated low oestrogen effect, 50–64 indicated moderate oestrogen effect, and 65–100 indicated high oestrogen effect Secondary outcomes Presence and severity of vaginal dryness, pruritus, dyspareunia, dysuria, urinary urgency: measured by participant‐reported symptom and severity questionnaires. Severity ratings defined as none (symptom not present), mild (aware of symptom but easily tolerated), moderate (enough to interfere with activities of daily living), severe (incapacity to work or carry out activities of daily living). Presence and severity of vaginal pallor, petechiae, friability, dryness: assessed by a physician. Symptom questionnaires and vaginal health assessments modified from those used in other urogenital atrophy studies. Graded on a 4‐point scale from none to severe based on similar models. Vaginal pH: obtained in standardised manner by swabbing fresh indicator strips with a pH range of 4.0–7.5 in intervals of 0.5 against the midvagina and interpreted by the examiner according to colour change from yellow to blue as compared with a standardised colour chart Vaginal health assessments and POP quantification: examiners asked to complete a preprinted vaginal health assessment Presence of granulation tissue, microscopic evidence of inflammation, subjective bother of atrophy; visible healing abnormalities (mesh erosion, suture line disruption, vaginal haematoma): assessed by examiner using a preprinted vaginal health assessment Cure or improvement: self‐reported by participants using a global self‐assessment ("cured", "greatly improved", "somewhat improved", "not improved", "worsened") Patient acceptance of the ring: including whether the ring had been removed or fallen out during treatment period, whether there was discomfort for the participant or their partner during coitus, whether the participant experienced discomfort related to wearing a vaginal ring
Notes	ITT analysis: NR Power calculation: no – "Because this was considered a pilot trial, no power analysis was performed a priori." Length of follow‐up: 6 and 12 weeks after surgery Conflicts of interest: "Dr Jean‐Michel is a consultant for Warner‐Chilcott and Boston Scientific. Dr Davila is a consultant for Astellas, American Medical Systems, CL Medical. The other authors declare that they have nothing to disclose". Study funding sources: "The authors thank Pfizer (New York, NY) for providing 22 placebo vaginal rings for this study".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "eligible subjects were randomly assigned via computer generation in blocks of 20 to one of 3 groups". Comment: seems adequate.
Allocation concealment (selection bias)	Low risk	Quote: "Allocation of randomization group was made by the primary author via sealed envelopes on the day of surgery once the patient was under anesthesia in the operating room". Comment: seems adequate.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Quote: "The patients remained blinded to the study group throughout the study unless unblinding became necessary owing to surgical complications"; "Thus although blinding of examiners to ring versus no ring was impossible, examiners remained blinded to active versus placebo ring". Comment: given that 1 of the groups did not have a pessary, they probably knew which group they were in. Also unclear how much the personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "Slides for maturation value and microscopic inflammatory cells were processed by the cytology laboratory by standard laboratory protocols and then analyzed by a cytopathologist (Y.J.) blinded to patient identity, treatment group, and preoperative/postoperative time period" Comment: cytopathologist blinded but unclear if other outcomes were blind.
Incomplete outcome data (attrition bias) All outcomes	Low risk	19/22 (86%) in the Estring group, 18/21 (86%) in the placebo ring group and 22/22 (100%) in the control group completed the study. Reasons for dropout are reported in the results section and were mainly due to participants wishing the vaginal ring to be removed. Attrition was not differential between intervention groups.
Selective reporting (reporting bias)	Unclear risk	No trial registration. Unclear which outcomes were specified before the trial began.
Other bias	Low risk	Nothing to suggest any other sources of bias.

Marschalek 2021.

Study characteristics
Methods	Design: RCT Dates study conducted: 2017–2020
Participants	Number of participants: 120 Country: Austria Setting: multicentre (2 centres in Vienna and Tulin) Severity of POP: POP‐Q Stage II: group I: n = 9 (17%); group II: n = 13 (25%) Stage III: group I: n = 41 (80%); group II: n = 38 (73%) Stage IV: group I: n = 1 (0%); group II: n = 1 (0%) Overall mean: group I: 2.8 (SD 0.4); group II: 2.7 (SD 0.5) Affected compartment Anterior (exclusive): group I: n = 3 (6%); group II: n = 4 (8%) Posterior (exclusive): group I: n = 1 (2%); group II: n = 1 (2%) All 3 (anterior, apical, posterior): group I: n = 47 (92%); group II: n = 47 (90%) Parity: group I: mean 2.4 births (SD 2.1); group II: mean 2.3 births (SD 1.1) Hysterectomy: group I: n = 14 (28%); group II: n = 5 (10%) Age: group I: mean 64.3 years (SD 9.7); group II: mean 61. 2 years (SD 10.1); total: mean 62.8 years (SD 10.0) BMI: group I: 26.9 kg/m2 (SD 4.0); group II: 27.3 kg/m2 (SD 4.8) Inclusion criteria Postmenopause Symptomatic POP Planned surgical prolapse repair Ability to apply vaginal cream Ability to comprehend the informed consent Ability to comprehend the delivered questionnaire Exclusion criteria Suspicion or history of malignancies Postmenopausal bleeding History of deep vein thrombosis Inherited or acquired blood clotting disorders History of transient ischaemic attack Myocardial infarction or ischaemic heart disease Hypersensitivity Contraindications to oestrogen Language barrier
Interventions	Group I (n = 60): oestradiol cream containing Linoladiol (Montavit company, Absam, Austria). Active ingredient of Linoladiol Estradiol‐Emulsion is oestradiol 0.10 mg in 1 g cream; chemically and biologically identical to endogenous human oestradiol. Women asked to use the cream intravaginally once daily for 1 week, every 48 hours the following week, then twice weekly for the remaining 4 weeks. They documented the self‐administered application in a patient diary. Trial co‐ordinators assessed adherence to the regimen at 6 weeks. Group II (n = 60): placebo cream contained cetyl alcohol, propylene glycol, triglycerides, hostacerin T3, polysorbate, almond oil, benzyl alcohol, purified water. Women asked to use the cream intravaginally once daily for 1 week, every 48 hours the following week, then twice weekly for the remaining 4 weeks. They documented the self‐administered application in a patient diary. Trial co‐ordinators assessed adherence to the regimen at 6 weeks. Co‐interventions: surgery for all participants. Undertaken by 3 senior consultants trained and specialising in pelvic floor surgery.
Outcomes	Primary outcome Surgeon's individual assessment of tissue quality during pelvic surgery: assessed by an 8‐item questionnaire; surgeons were also asked whether they thought the participants had received pretreatment with local oestrogen cream or not. Secondary outcomes Length of stay (days) Frequent use of analgesics: administration of intravenous analgesics > 2 times per day Postoperative use of antibiotics Defined postoperative complications: complications associated with the pelvic floor surgery, including postoperative UTIs, postoperative urinary retention, surgical site infection, secondary bleeding, and readmission within 30 days of discharge
Notes	ITT analysis: yes Power calculation: yes – "To discard this null hypothesis, the sample size was calculated using preliminary, unpublished data from preliminary tests of the surgical questionnaire as follows: The G‐Power 3.1.9.2 software (Universität Düsseldorf, Düsseldorf, Germany) was used for calculation of sample size using Student's t test for unpaired groups with a level of significance at 5%, a test power of 80% and an effect size of 0.45. Thus, 60 patients per group were calculated". Length of follow‐up: 6 weeks Conflicts of interest: "The authors declare no conflict of interest". Study funding sources: "This study was funded by Montavit Ges.m.b.H., Salzbergstraße 96, A‐6067 Absam/Tirol, Austria. Role of Funder: The sponsor had no role in the design and conduct of the study, collection, analysis and interpretation of result as well as submission process".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The randomization was conducted by the Pharmaceutical Company Montavit. The allocation sequence was computer‐generated by Rancode Professional 3.6 software (IDV, Gauting, Germany) A randomization in blocks of four was performed and carried out from numbers 1 to 400. On the basis of the randomization list, all labels were produced, and information of the principal investigator as well as the randomization number was included". Comment: seems adequate.
Allocation concealment (selection bias)	Unclear risk	Quote: "The allocation sequence was computer‐generated by Rancode Professional 3.6 software (IDV, Gauting, Germany)" but also says "The research team was unaware of each participant's allocated treatment group. The only unblinded person was the study coordinator of the company Montavit Ges.m.b.H.". Comment: unclear of the impact of the above.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "The research team was unaware of each participant's allocated treatment group". Comment: seems that personnel were adequately blinded. Assumed that participants are also blinded because the cream and placebo cream were administered in the same way, and they were given the same instructions.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The main variables of interest included conditions during surgery determined by the surgeon blinded to treatment allocation and assessed by an 8‐item questionnaire". Comment: the surgeon undertaking the questionnaire was blind to treatment allocation. No other self‐reported outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	51/60 in treatment group and 52/60 from control group were analysed. No differential attrition. Reasons for loss to follow‐up adequately explained in flow diagram. ITT not performed, but overall attrition (14%) not enough to cause concern.
Selective reporting (reporting bias)	Low risk	All outcomes stated in the trial registration seem to be reported across the study.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Nunes 2011.

Study characteristics
Methods	Design: RCT + 1 non‐randomised arm Dates study conducted: NR
Participants	Number of participants: 60 in total; 40 randomised Country: Brazil Setting: NR Severity of POP: NR Affected compartment: uterine Parity: group I: mean 4.1 births (SD 3.0); group II: mean 4.6 births (SD 3.0); group III: mean 4.3 births (SD 3.0) Hysterectomy: none at baseline (all participants underwent hysterectomy as part of this study) Age: group I: mean 45.6 years (SD 5.8); group II: mean 64.6 years (SD 8.3); group II: mean 63.3 years (SD 9.3) BMI: group I: mean 26.7 kg/m2 (SD 4.1); group II: mean 28.6 kg/m2 (SD 4.6); group III: mean 29.4 kg/m2 (SD 4.5) Inclusion criteria Postmenopause Stage II, III, or IV uterine prolapse according to POP‐Q Exclusion criteria Known connective tissue disease Diabetes Vulvovaginal infection Tobacco use Use of oestrogen, progestins, or androgens in previous 12 months
Interventions	Group I (n = 20): oral oestrogen 30 days before surgery Group II (n = 20): oral placebo 30 days before surgery Group III (n = 20): these control cases did not have uterine prolapse and were premenopausal but had received abdominal hysterectomies for benign gynaecological disorders Co‐interventions: surgery for all participants (vaginal hysterectomy)
Outcomes	Evaluation of glycosaminoglycans (GAGs) in the parametrium, paraurethral tissue, and vaginal apex: measured using laboratory testing of sample of the parametrium and vaginal apex obtained from all participants after hysterectomy
Notes	ITT analysis: NR Power calculation: NR Length of follow‐up: at surgery: "Samples of the parametrium and vaginal apex were obtained from all 60 patients after the uterus had been removed during hysterectomy" Conflicts of interest: "We confirm that all of the authors have no actual or potential conflicts of interest to disclose regarding any financial, personal or other relationships with other people or organizations within three years of beginning the submitted work that could inappropriately influence, or be perceived to influence, our work". Study funding sources: "Nil"
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	2 arms of the study described as "randomized" (with no further details on randomisation process) but the third study arm was a non‐randomised set of 20 women who did not have uterine prolapse.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Quote: "Double‐blind". Comment: participants received pessary and either oestrogen or placebo but unclear if they were blinded because there are insufficient details of administration. Unclear if personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Assumed low as no dropouts or losses to follow‐up are reported in any of the groups.
Selective reporting (reporting bias)	Unclear risk	No trial registration. Unclear which outcomes were specified before the trial began.
Other bias	High risk	There was a major difference in average age between group 1 and the other 2 groups, suggesting that there may be issues with the randomisation process.

Rahn 2014.

Study characteristics
Methods	Design: RCT Dates study conducted: enrolment completed in 2012
Participants	Number of participants: 30 Country: USA Setting: single centre (University of Texas Southwestern Medical Center and Parkland Health and Hospital System, Dallas, Texas) Severity of POP: group I: median anterior vaginal wall prolapse stage II (IQR II–III); group II: median anterior vaginal wall prolapse stage II (IQR II–III) Affected compartment: uterine, anterior vaginal wall, or both Parity: group I: median 3 births (IQR 2–4); group II: median 4 births (IQR 3–6) Hysterectomy: NR Age: group I: mean 55.1 years (SD 5.4); group II: mean 58.9 years (SD 5.1) BMI: group I: mean 29.2 kg/m2 (SD 5.4); group II: mean 31.7 kg/m2 (SD 4.1) Inclusion criteria 1–10 years beyond last menses Age 40–70 years Symptomatic uterine or anterior vaginal wall prolapse (≥ stage 2) Wish for surgical repair Willingness to undergo total hysterectomy No oestrogen replacement by any route in preceding month Physical capability to apply vaginal cream Exclusion criteria BMI > 35 kg/m2 Prior surgical repair including anterior colporrhaphy or prior hysterectomy Corticosteroid therapy in preceding month History of connective tissue disease (e.g. Ehlers‐Danlos or Marfan syndromes) Prior vaginal radiation Concurrent use of steroid cream (e.g. treatment of lichen sclerosis) Current tobacco use History of vaginal infection in preceding month Contraindication to oestrogen replacement therapy (i.e. history of spontaneous deep vein thrombosis, oestrogen responsive malignancy, unexplained vaginal bleeding)
Interventions	Group I (n = 15): vaginal conjugated oestrogen cream (Premarin, Pfizer). A pharmacy with expertise in drug compounding re‐tubed the Premarin (0.625 mg drug per 1 g cream) into unbranded white tubes that were identical to the placebo cream. Participants were instructed to apply 1 g cream intravaginally via an applicator every night for 2 weeks and then twice weekly until surgery. A total of 6 (minimum 4, maximum 8) weeks of treatment was required before surgery. Group II (n = 15): placebo cream. The placebo cream was contained in unbranded white tubes identical to the oestrogen cream. Participants were instructed to apply 1 g cream intravaginally via an applicator nightly for 2 weeks and then twice a week until surgery. A total of 6 (minimum 4, maximum 8) weeks of treatment was required before surgery. Co‐interventions: surgery for all participants. Total hysterectomy, which could be either vaginal or abdominal, as part of a planned prolapse repair.
Outcomes	Mucosa and muscularis thickness, connective tissue thickness, connective tissue synthesis and degradation: all assessed by analysing biopsies of the full‐thickness anterior apical vaginal wall at the time of hysterectomy Serum levels of oestrone and 17 beta‐oestradiol: analysed using highly sensitive chromatography‐tandem mass spectrometry mRNA levels in vaginal tissues: assessed using quantitative PCR Adherence to treatment: assessed by comparing tube weights at the time of surgery to the anticipated weight if used according to protocol. Participants were also given a diary to complete indicating the nights they had applied cream.
Notes	ITT analysis: yes (they also planned per‐protocol analyses) Power calculation: no – based on experience with premenopausal women and animal studies Length of follow‐up: at surgery: "At the time of surgery, a sample of vaginal wall was excised" Conflicts of interest: "The authors have nothing to disclose", but in an abstract they note "J. I. Schaffer: Speaker, Astellas, Speaker, Cadence, Consultant, Ferring". Study funding sources: "This work was supported by the American Urogynecologic Society Foundation’s Astellas Research Award and National Institutes of Health Grant AG028048".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "A computer‐generated randomization table (blocks of four) was used to allocate participants to either vaginal conjugated estrogen cream (Premarin; Pfizer) or placebo cream". Comment: seems adequate.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "A pharmacy with expertise in drug compounding (Drug Crafters) retubed commercially available Premarin cream (0.625 mg drug per 1 g cream) and its placebo match into identical unbranded white tubes". Comment: very likely that the participants were blinded. Also mentions that the department in charge of the randomisation sequence "was the only party unblinded to drug allocation".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Thickness of the vaginal epithelium, lamina propria, and muscularis was determined as previously described by examiners blinded to treatment allocation"; "Hysterectomy specimens were examined by a blinded pathologist for endometrial and myometrial thickness and pathology". Comment: seems adequate.
Incomplete outcome data (attrition bias) All outcomes	High risk	8/15 in the oestrogen group and 12/15 in the control group were analysed per protocol. Reasons for dropouts and non‐adherence are explained in the flow diagram, but there is differential attrition between the groups that may potentially affect results.
Selective reporting (reporting bias)	Unclear risk	Prospectively registered. Although all outcome measures specified in trial registration are reported, unclear why the investigators carried out a posthoc analysis on stage of prolapse at 6 months.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Stipic 2012.

Study characteristics
Methods	Design: RCT Dates study conducted: March 2009–November 2010
Participants	Number of participants: 60 Country: Croatia Setting: single centre (Department of Gynecology and Obstetrics, University Hospital in Split) Severity of POP: NR Affected compartment: uterine Parity: group I: mean 2.0 births (SD 1.0); group II: mean 2.0 births (SD 1.0) Hysterectomy: all participants Age: group I: mean 57.54 years (SD 4.27); group II: mean 57.61 years (SD 4.98) BMI: group I: mean 26.74 kg/m2 (SD 3.32); group II: mean 26.19 kg/m2 (SD 3.57) Inclusion criteria Postmenopausal women with POP admitted to the hospital for surgery "Previously, all women had been in postmenopause for more than 2 years and had undergone hysterectomy and bilateral salpingo‐oophorectomy" Exclusion criteria Previous pelvic surgery or pelvic radiation therapy Pelvic inflammatory disease Pelvic or other malignancies Age > 70 years Cardiopulmonary disease, especially COPD or asthma Corticosteroid use Hypertension Anaemia Diabetes mellitus Anticoagulant therapy Neurologic disease that may be associated with bladder disorders Anticholinergic or duloxetine medication Immobility Previous HRT Previous active tobacco smoking
Interventions	Group I (n = 30): oestrogen replacement therapy. Oestradiol hemihydrate 1 mg administered orally every day Group II (n = 30): no oestrogen replacement therapy Co‐interventions: surgery for all participants. Surgery for uterine prolapse (no other details)
Outcomes	Lung function tests (FVC, FEV1, peak expiratory flow, forced expiratory flow at 75% of FVC, forced expiratory flow at 50% of FVC, and forced expiratory flow at 25% of FVC): all measured via spirometry using a computed spirometer, with the participants asked to breathe spontaneously followed by inhaling their vital capacity and then complete, most forceful expiration through the mouthpiece
Notes	ITT analysis: NR Power calculation: NR Length of follow‐up: prior to surgery and 6 months Conflicts of interest: "We certify that all of the authors did not have any relationship with companies that may have a financial interest in the information contained in the manuscript". Study funding sources: "This study was supported by the Croatian Ministry of Science, Education and Sports project (No. 216‐0000000‐0535); head researcher Professor Tomislav Strinic, M.D., Ph.D.".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "randomly divided". Comment: unclear method of randomisation.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants in the control groups would have been aware that they were not receiving any additional treatment; unclear whether personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Evaluation was generally performed between 9 a.m. and 11 a.m., with the respiratory laboratory personnel being unaware of the subject's status". Comment: seems adequately blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Appears all study participants from both arms had complete data.
Selective reporting (reporting bias)	Unclear risk	No trial registration. Unclear which outcomes were specified before the trial began.
Other bias	Unclear risk	No demographic information on respiratory variables (e.g. passive smoking, smoking history).

Sun 2016.

Study characteristics
Methods	Study design: RCT Dates study conducted: August 2013–May 2014
Participants	Number of participants: 186 Country: China Setting: single centre (Peking Union Medical College Hospital) Severity of POP: POP‐Q Anterior stage 3 or 4: group I: n = 93 (100%); group II: n = 93 (100%) Apical stage 3 or 4: group I: n = 93 (100%); group II: n = 93 (100%) Posterior stage 1: group I: n = 25 (26.9%); group II: n = 16 (17.2%) Posterior stage 2: group I: n = 47 (50.5%); group II: n = 57 (61.2%) Posterior stage 3 or 4: group I: n = 21 (22.6%); group II: n = 20 (21.5%) Affected compartment: uterine and anterior Parity: group I: median 2 births (range 1–3); group II: median 2 births (range 1–3) Hysterectomy: NR Age: group I: median age at diagnosis 66 years (range 50–75); group II: median age at diagnosis 65 years (range 51–74) BMI: group I: median 24.2 kg/m2 (range 18.4–28.9); group II: median 24.5 kg/m2 (range 17.7–30.0) Inclusion criteria Age 50–75 years Postmenopause (no menses for ≥ 12 months before the study; in cases of uncertainty, follicle‐stimulating hormone levels were measured via a blood test) Stage 3 or 4 uterine prolapse combined with stage 3 or 4 anterior vaginal wall prolapse (according to POP‐Q) Referral for transvaginal pelvic reconstructive surgery with mesh Exclusion criteria Oral or transdermal oestrogen therapy within 1 year of participating Any contraindication to local oestrogen application Any type of cancer Inability to tolerate the laparoscopic procedures History of anti‐incontinence or prolapse surgical repair Need for concomitant surgery Current stress or urgency urinary incontinence (although those with occult stress incontinence screened by the 1‐h pad test with pessary were allowed to continue)
Interventions	Group I (n = 93): oestrogen. Promestriene 0.5 g cream transvaginally twice weekly for 6 weeks before surgery with mesh. The cream was all from the same batch produced by Merck Serono. Group II (n = 93): no treatment. No oestrogen preparation before surgery. Co‐interventions: surgery. All participants underwent pelvic reconstructive surgery using mesh. A senior urogynaecologist performed all the surgeries in accordance with the original description of the procedure using the Avaulta Biosynthetic Support System.
Outcomes	Primary outcomes Occurrence of mesh exposure: defined as displaying, revealing, or exhibiting the mesh or making the mesh accessible. For each exposure case, the study authors also assessed the exposure area, the time to exposure, and the need to remove the mesh. Secondary outcomes Proportion of participants achieving anatomic success: defined as a POP‐Q stage score < 2 at 12 months Good response on PGI‐C: at month 12, defined as "much better" or "a little better" when participants were asked "compared with how you were doing before your recent pelvic floor operation, would you say that now you are much better/a little better/about the same/ a little worse/much worse?" Change in Pelvic Floor Impact Questionnaire Short Form (PFIQ‐7): defined as a decrease of ≥ 70% PISQ‐12 scores: defined as an improvement of ≥ 5 points
Notes	ITT analysis: yes (per‐protocol analysis also planned) Power calculation: yes – "Assuming a 90% response rate, a sample size of 186 participants (93 women in each arm) was estimated to provide at least 85% power to detect a predefined noninferiority margin of 8% in the proportion of VE‐ and non–VE‐treated women achieving mesh exposure at 1 year, at a one‐sided 2.5% level of significance". Length of follow‐up: 2, 6, and 12 months Conflicts of interest: "None reported" Study funding sources: "This work was mainly supported by the 5‐year National Project of China (3053750)".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "A research assistant with no clinical involvement in the trial allocated participants using a randomization list generated by computer". Comment: seems adequate.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "the surgeon and participants were not blinded to the medication". Comment: participants and personnel were not blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "the clinical parameters and outcomes were assessed at baseline and months 2, 6, and 12 by different trained evaluators who were blinded to randomization assignment and who were not the operating surgeon". Comment: outcome assessments appear to be blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "92.5% and 93.5% of participants in the two groups completed the 1‐year evaluation". Comment: attrition not differential or large. ITT analyses performed. Reasons for dropouts explained in the participant flow chart.
Selective reporting (reporting bias)	High risk	Trial registered prospectively. The outcome "Vaginal swab culture result post‐operation" is listed in the trial registration but does not appear to be reported in the results.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Tontivuthikul 2016.

Study characteristics
Methods	Study design: RCT Dates study conducted: NR
Participants	Number of participants: 40 Country: Thailand Setting: assumed single centre (Chiang Mai University) Severity of POP Stage 2: group I: n = 3 (15%); group II: n = 5 (25%) Stage 3: group I: n = 9 (45%); group II: n = 10 (50%) Stage 4: group I: n = 8 (40%); group II: n = 5 (25%) Affected compartment: NR Parity: group I: mean 2.9 births (SD 1.1); group II: mean 4.0 births (SD 2.5) Hysterectomy: group I: n = 5 (25%); group II: n = 4 (20%) Age: group I: mean 66.13 (SD 6.78); group II: mean 66.67 (SD 8.05) BMI: group I: mean 24.63 (SD 3.97); group II: mean 22.48 (SD 3.55) Inclusion criteria Postmenopause POP Ability to give informed consent and answer treatment‐related questions Use of pessary for 6 weeks and intention to continue using the pessary Exclusion criteria Contraindications to local oestrogen (e.g. known or suspected breast cancer, hormone‐dependent tumour, or thromboembolic disorder) Inability to follow protocol
Interventions	Group I (n = 20): vaginal oestrogen cream. After the initial period, the participants continued to use the conjugated equine oestrogen 0.625 mg/g (Premarin) once a week for the entire 24 weeks Group II (n = 20): no treatment (control) Co‐interventions: pessary and pretreatment with oestrogen cream for all participants. Conjugated equine oestrogen 0.625 mg/g (Premarin). All participants told to apply 0.5 g of the cream daily for 2 weeks then twice weekly until their follow‐up visit. Also advised to remove their pessary every night and reinsert it every morning. Pessaries included Ring, Gellhorn and Donut. Support pessary (group I: 85% ring, 10% Gellhorn, 5% donut; group II: 75% ring, 20% Gellhorn, 5% donut)
Outcomes	Primary outcomes Assessment of vaginal health: assessed by a self‐reported vaginal symptom score (dryness, soreness, itching/irritation, discharge). Self‐assessed with a 4‐point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe) Vaginal pH: measured by a pH indicator strip inserted into the vagina Vaginal cytology: vaginal smears taken from the upper third of the lateral vaginal wall and analysed blindly by the cytologist to determine the VMI Secondary outcomes Difficulty to insert/remove the pessary: self‐assessed using a 4‐point scale (0 = very easy, 1 = easy, 2 = difficult, 3 = very difficult) Endometrial thickness Adverse events: such as vaginal abrasion/ulcer or vaginal bleeding
Notes	ITT analysis: NR Power calculation: yes – "The sample size would provide an 80% power in detecting a difference in change from baseline score of 0.5 between two groups for the primary endpoint (vaginal symptom score of vaginal dryness, soreness, itching/irritation and discharge). We calculated that 18 subjects each arm would be required to demonstrate a p‐value of 0.05. Forty subjects were recruited, expecting 10% to drop out". Length of follow‐up: 12 and 24 weeks Conflicts of interest: "None" Study funding sources: "The present research was supported by the Faculty of Medicine Endowment Fund for Medical Research, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "A computer‐generated randomization table (block of ten) was used to allocate participants". Comment: seems adequate.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants in the control groups would have been aware that they were not receiving any additional treatment; unclear whether personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Vaginal maturation index was assessed "blindly by the cytologist" but it is not clear whether any of the other outcomes were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No dropouts or losses to follow‐up reported in either arm of the study.
Selective reporting (reporting bias)	Unclear risk	No trial registration. Unclear which outcomes were specified before the trial began.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

Vaccaro 2013.

Study characteristics
Methods	Study design: RCT Dates study conducted: January 2009–March 2011
Participants	Number of participants: 54 (4 women randomised to 0.5 g were excluded from final analysis; only the women in 1.0 g and control groups are considered in the full paper) Country: USA Setting: multicentre (the Division of Urogynecology and Reconstructive Pelvic Surgery at TriHealth Inc, Cincinnati, Ohio, and its affiliated hospitals) Severity of POP Group II: stage 2: n = 13 (59.1%); stage 3: n = 8 (36.4%); stage 4: n = 1 (4.5%) Group III: stage 2: n = 8 (40%); stage 3: n = 9 (45%); stage 4: n = 3 (15%) Affected compartment: NR Parity: group II: mean 2.7 births (SD 1.3); group III: mean 3.9 births (SD 2.4) Hysterectomy: 100% in both groups Age: group II: mean 64.3 years (SD 10); group III: mean 66.3 years (SD 10.2) BMI: group II: mean 29.7 years (SD 5.3); group III: mean 29.4 years (SD 4.6) Inclusion criteria Postmenopause, either natural (> 12 months since last menses and age ≥ 55 years) or surgical (history of bilateral salpingo‐oophorectomy and age ≥ 45 years) Clinical atrophic vaginitis and posthysterectomy POP (≥ stage 2 according to POP‐Q) Wish for surgical management within 2–12 weeks of enrolment Exclusion criteria Uterus present Well‐oestrogenised vagina Known or suspected history of breast carcinoma or hormone‐dependent tumour Genital bleeding of unknown cause Acute thrombophlebitis or thromboembolic disorder associated with oestrogen use Vaginal infection requiring treatment Allergy to oestrogen or its constituents Any serious disease or chronic condition that would interfere with study compliance or preoperative surgical clearance Use of exogenous corticosteroid or sex hormones (e.g. oestrogen, progesterone, testosterone, including homeopathic preparation) in the 8 weeks prior to recruitment
Interventions	Group I (n = 4): 0.5 mg vaginal cream. Participants instructed to use conjugated oestrogens vaginal cream (Premarin) nightly from enrolment until the surgery date (2–12 weeks). An instruction label was affixed to the medication box indicating the dosage, which stated: "Measure ____ gram of cream using the applicator, and then squeeze that amount onto your finger. Using your finger(s), rub the cream into the bulge in your vagina for at least 30 seconds (or longer until completely absorbed). **DO NOT USE THE APPLICATOR TO APPLY THE CREAM**". The dosage was written on the black line of the label. Group II (n = 31): 1.0 g vaginal cream. Participants instructed to use conjugated oestrogens vaginal cream (Premarin) nightly from enrolment until the surgery date (2–12 weeks). An instruction label was affixed to the medication box indicating the dosage, which stated: "Measure ____ gram of cream using the applicator, and then squeeze that amount onto your finger. Using your finger(s), rub the cream into the bulge in your vagina for at least 30 seconds (or longer until completely absorbed). **DO NOT USE THE APPLICATOR TO APPLY THE CREAM**". The dosage was written on the black line of the label. Group III (n = 23): no intervention (control) Co‐interventions: surgery for all participants. Either vaginal, robotic, or abdominal; no other details.
Outcomes	Primary outcomes VMI: assessed using vaginal cytology samples obtained from the most dominant compartment of the prolapse and analysed by a pathologist and the primary investigator Vaginal health composite score (VCHS – assessment of vaginal atrophy): evaluated the degree of epithelial atrophy via vaginal pH, fluid secretions, epithelial appearance, moisture, rugosity, and colour. Summed to a range of 6–24 with a lower score indicating greater vaginal atrophy Secondary outcomes Vaginal epithelial thickness Inflammation and chronicity: measured by analysing histologic samples. Inflammation classed as "none, mild, moderate and severe" and chronicity measured as "acute or chronic". Adherence: assessed by weighing the remainder of the vaginal cream in grams. The expected weight of cream was calculated by multiplying the days of use of the cream by the cream dosage and then subtracting from the reference weight. Vaginal symptomatology (dryness, soreness, and irritation): obtained via a VAS and the PFDI, PFIQ, and PISQ‐12
Notes	ITT analysis: NR Power calculation: yes – "A sample size calculation was performed using nQuery Advisor v.7 (Statistical Solutions, Saugus, Mass). On the basis of prior research demonstrating a mean 15% increase in VMI over 2 to 12 weeks of vaginal estrogen use, 20 women were to be enrolled in each of the 3 arms (control, 0.5‐g, and 1.0‐g group) to yield an 80% power with a significance level of 0.05.9 Additionally, Bachmann et al9 demonstrated a 16.6% to 19.6% increase in VMI after only 2 weeks of daily vaginal estrogen use. Despite eliminating the 0.5‐g treatment arm, the sample size calculation remained unchanged with 20 women in each arm (control and 1.0‐g group) to yield an 80% power with a significance level of 0.05". Length of follow‐up: on the day of surgery (somewhere between 2 and 12 weeks) Conflicts of interest: "The authors have declared they have no conflicts of interest." but also in an abstract: "R Pauls: Consultant, consulting fee, Researcher, research support Stock Options, Scientific Advisory Board". Study funding sources: "Supported by the Hatton Institute for Research and Education, Good Samaritan Hospital, Cincinnati, OH".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Originally, the randomisation seemed adequate: "Participants were randomized to 3 arms using a 1:1:1 ratio to receive vaginal cream (0.5 or 1.0 g) or control (no intervention). A randomization code was generated and assigned in blocks of 3"; However, after the 0.5 g arm excluded: "Participants were not re‐randomized, but rather, any randomized packet allocating to the 0.5‐g group was discarded and the next lowest number opened". This could leave the randomisation process open to bias.
Allocation concealment (selection bias)	High risk	Although "Each randomized participant was assigned to treatment or control via opening her sequentially numbered opaque envelope", "The treatment group dosage, either 0.5 or 1.0 g, was written on the blank line of the previously mentioned label by the enrolling investigator, thus blinding was not possible during enrollment." Therefore, allocation was unblinded.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "The treatment group dosage, either 0.5 or 1.0 g, was written on the blank line of the previously mentioned label by the enrolling investigator, thus blinding was not possible during enrollment". Comment: blinding did not occur.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "The VHCS, as well as cytology and histology specimens were collected by the primary investigator or co‐investigators, who were not blinded to the group assignment at the time of collection, as they were responsible for dispensing and collecting the vaginal estrogen cream"; "During cytologic assessment, the pathologist (G.M.) and primary investigator (C.V.) were blinded to the study group". Comment: some blinding of assessors but not for all outcomes.
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "Four women who had been in the 0.5‐g group were excluded from the final analysis. Eight women were removed from the study, that is, 4 women were excluded because they cancelled or postponed their scheduled surgery more than 12 weeks from enrollment; 2 women were excluded due to the following adverse effects: vaginal burning and/or pain; and 2 women voluntarily withdrew from the study without giving a reason". Comment: this means 22/31 in the treatment groups and 20/23 in the control group were analysed overall. Differential loss to follow‐up and no ITT analyses performed.
Selective reporting (reporting bias)	High risk	Manuscript only reports cytological outcomes, same population used in abstract reporting quality of life outcomes with different primary outcomes reported.
Other bias	High risk	Quote: "A randomization code was generated and assigned in blocks of 3. However, enrollment in the 0.5‐g arm was halted early in the trial due to recruiting difficulties and a projected inability to complete the study. Participants were not re‐randomized, but rather, any randomized packet allocating to the 0.5‐g group was discarded and the next lowest number opened". Comment: a whole arm of the study was cancelled so that only the 1.0 g and control groups remained. Additionally, participants were evaluated after receiving treatment for between 2 and 12 weeks – no standardisation of how long participants received the oestrogen and this could have affected treatment response.

Verghese 2020.

Study characteristics
Methods	Study design: RCT Dates study conducted: July 2015–August 2016
Participants	Number of participants: 100 Country: UK Setting: multicentre (6 urogynaecology centres) Severity of POP Stage 1: group I: n = 7 (14%); group II: n = 7 (14%); total: n = 14 (14%) Stage 2: group I: n = 23 (46%); group II: n = 23 (46%); total: n = 46 (46%) Stage 3/4: group I: n = 20 (40%); group II: n = 20 (40%); total: n = 40 (40%) Affected compartment: NR Parity ≤ 2 births: group I: n = 27 (54%); group II: n = 26 (52%); total: 53 (53%) > 2 births: group I: n = 23 (46%); group II: n = 24 (48%); total: 47 (47%) Hysterectomy: group I: n = 12 (24%); group II: n = 8 (16%); total: 20 (20%) Age: group I: mean 65.7 years (SD 8.2); group II: mean 65.9 years (SD 8.4); total: 65.8 years (SD 8.3) BMI: group I (2 participants missing): mean 28.1 kg/m2 (SD 5.1); group II (7 participants missing): mean 28.5 kg/m2 (SD 5.9); total (9 participants missing): 28.2 kg/m2 (SD 5.5) Inclusion criteria Postmenopause Planned POP surgery without the use of vaginal mesh Exclusion criteria HRT in past 12 months Previous breast/uterine malignancy or other hormone‐dependent neoplasms Genital bleeding of unknown origin Previous thromboembolic episodes in relation to oestrogen Known allergy to oestrogens ≥ 2 episodes of culture positive UTI in past 6 months Voiding dysfunction Previous POP surgery involving mesh or previous POP surgery in the same compartment
Interventions	Group I (n = 50): oestrogen. Oestradiol hemihydrate 10 μg pessaries (Vagifem) 6 weeks prior to surgery (once daily for 2 weeks then twice weekly for 4 weeks) until the night before surgery. Treatment was restarted 6 weeks postoperatively (twice weekly for 20 weeks). Participants were encouraged to insert the pessaries into the vagina at the same time of day, but if they missed a dose they were asked to administer as soon as possible thereafter, unless the next dose was already due. Group II (n = 50): no treatment (control); participants received the usual care of their randomising centre. Co‐interventions: surgery for all participants. The surgical approach to the POP repair was decided on the discretion of the urogynaecological surgeon. Anterior repair: 35 (80%) in oestrogen group, 34 (76%) in control group Posterior repair: 16 (36%) in oestrogen group, 19 (42%) in control group Vaginal hysterectomy with or without salpingo‐oophorectomy: 26 (59%) in the oestrogen group, 25 (56%) in the control group Sacrospinous fixation: 6 (14%) in the oestrogen group, 6 (13%) in the control group 5 (10%) women in oestrogen group and 8 (16%) women in the control group already had a pessary in place.
Outcomes	Primary outcome "To obtain estimates for important aspects of the protocol to allow development of a definitive trial" Secondary outcomes Assessment of effectiveness of patient identification and screening processes Assessment of the effectiveness of the randomisation process of participants Evaluation of the robustness of the data collection processes Proportion of participants followed up at 6 months Derivation of the preliminary data from clinical outcome measures (e.g. Pelvic Floor Distress Inventory Short Form 20 (PFDI‐SF20)) to help inform the sample size calculation for the substantive study Objective assessment of prolapse: assessed by POP‐Q Complications: including UTIs and vaginal infections Repeat surgery information: collected by the randomising or operating clinician Patient‐reported questionnaires: PFDI‐SF20, PFIQ‐7, PISQ‐12, PGI‐I Serious adverse events: reported by clinical staff through the duration of the study
Notes	ITT analysis: yes Power calculation: yes – "We planned to randomise 100 women; this number would allow us to measure recruitment and compliance rates with 95% CI between 10% and 20%. It would also be enough to estimate the SD of POPDI‐6 with reasonable confidence for future planning of a larger trial (95%CI for SD would be 7 points, assuming the SD is around 20)". Length of follow‐up: 6 weeks and 6 months; questionnaires collected at 6 and 12 months. Conflicts of interest: "PML received sponsorship from Astellas, Contura and Pfizer to attend meetings". Study funding sources: "The study was funded by a grant from the National Institute for Health Research for Patient Benefit (grant number: PB‐PG‐0213–30126)". Correspondence: we contacted the study authors on 26 October 2021 to clarify how many women in the oestrogen therapy arm failed the objective assessment of prolapse with POP‐Q at 6 months. The study authors responded and provided these data on 2 November 2021.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomisation was performed using a webbased central randomisation system (via Birmingham Clinical Trials Unit) to allocate patients to either oestrogen or no treatment in a 1:1 ratio. Minimisation was used to achieve balance between age (<65 or ≥65years), parity (≤2or >2 vaginal births), maximum stage of prolapse (I, II or III/IV) and whether a concomitant continence surgery was planned. An oversight committee was formed to provide independent guidance to the Trial Management Committee and to review accruing safety information during the period of recruitment". Comment: seems adequate.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "It was not possible to blind clinicians or participants due to the nature of the interventions". Comment: could not be blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	At 12 months, 37/50 in intervention group and 42/50 in control group who had been initially randomised were followed up. Small differential loss to follow‐up but reasons explained. ITT was planned, but unclear from the data whether ITT was performed.
Selective reporting (reporting bias)	Low risk	Prospectively registered. Seems to have reported all outcomes of interest as stated in the protocol.
Other bias	Low risk	Nothing to suggest any other potential sources of bias.

BMI: body mass index; CEO: conjugated equine oestrogen; CI: confidence interval; COPD: chronic obstructive pulmonary disease; FEV1: forced expiratory volume in 1 second; FVC: forced vital capacity; GSM: genitourinary symptoms of menopause; HRT: hormone replacement therapy; ICIQ‐VS: International Consultation on Incontinence Questionnaire Vaginal Symptoms Module; IQR: interquartile range; IUGA/ICS: International Urogynecological Association/International Continence Society; ITT: intention‐to‐treat; mRNA: messenger ribonucleic acid; NR: not reported; PCR: polymerase chain reaction: PFDI: Pelvic Floor Distress Inventory; PFIQ: Pelvic Floor Impact Questionnaire; PFME: pelvic floor muscle exercises; PGI‐C: Patient Global Impression of Change; PGI‐I: Patient Global Impression of Improvement; PISQ‐12: Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire; POP: pelvic organ prolapse; POP‐Q: Pelvic Organ Prolapse Quantifications System; RCT: randomised controlled trial; SD: standard deviation; SF‐36: 36‐item Short‐Form Health Survey; UTI: urinary tract infection; VAS: visual analogue scale; VHI: vaginal health index; VMI: vaginal maturation index.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Barber 2002	Ineligible intervention: participants who had POP received surgery.
Chinthakanan 2019	Ineligible population: no mention of need for the participants to be postmenopausal.
Chou 2013	Ineligible population: no mention of need for the participants to be postmenopausal.
Jiang 2017	Ineligible intervention: not an oestrogen therapy.
Lind 2014	Ineligible population: no mention of need for the participants to be postmenopausal.
Moalli 2012	Ineligible population: no mention of need for the participants to be postmenopausal.
Paraiso 2020	Ineligible population: participants are currently menopausal with GSM, not postmenopausal.
Valente 2000	Ineligible intervention: vitamin D and calcium versus HRT.

GSM: genito‐urinary symptoms of menopause; HRT: hormone replacement therapy; POP: pelvic organ prolapse.

Characteristics of studies awaiting classification [ordered by study ID]

Pinhat 2013.

Methods	Study design: RCT
Participants	Number of participants: 69 (actual) Country: Brazil Setting: single centre Severity of POP: NR (trial registration) Affected vaginal compartment: NR (trial registration) Parity: NR (trial registration) Hysterectomy: NR (trial registration) Age: NR (trial registration) BMI: NR (trial registration) Inclusion criteria Postmenopause Normal oncotic cytology and mammogram Any grade of POP with surgery indication and ultrasound with endometrium measure < 5 mm Exclusion criteria Vaginal bleeding Hormonal replacement in past 3 months Renal or hepatic disease Porphyria Past venous thromboembolism Ulcerated prolapse
Interventions	Group I: conjugated oestrogen. 1 g/day Group II: promestriene alone. 1 g/day Group III: oestriol alone. 1 g/day Group IV: vaginal moisturiser cream. 1 g/day Co‐interventions: NR
Outcomes	Primary outcomes Evaluate the histopathological effects on endometrium after each treatment Compare the endometrial thickness measured by ultrasound after each treatment Secondary outcomes Evaluate the changes of maturation index (Meisels Index) after each treatment (48 days) Compare the changes on Meisels Index after each treatment and among each other Evaluate the changes on climacteric complaints after each treatment (48 days) Compare the changes on systemic and genital complaints after each treatment, and among each other
Notes	Length of follow‐up: 48 days Study funding sources: NR Correspondence: we contacted the lead author on 14 December 2021 and again on 4 January 2022 to ask if there was a full publication associated with the study. The study authors responded that there was currently no full publication available. Reason for awaiting classification: trial ended in 2016, according to trial registration. Located thesis but only abstract in English with few details. Attempted to identify Portuguese translator.

Vardy 2003.

Methods	Study design: RCT (secondary analysis of a trial whose primary endpoint was osteoporosis) Dates study conducted: NR
Participants	Number of participants: 47 Country: USA Setting: single centre Severity of POP: "All groups were similar for stage of prolapse at baseline"; no other details Affected vaginal compartment: NR Parity: group I: mean 4.3 births (SD 1.9); group II: mean 3.2 births (SD 2.3); group III: mean 2.8 births (SD 1.2); group IV: mean 3.00 births (SD 1.7) Hysterectomy: none (women who had hysterectomy were excluded) Age: group I: mean 57.7 years (SD 5.2); group II: mean 54.3 years (SD 3.7); group III: mean 54.4 years (SD 3.4); group IV: mean 55.00 years (SD 3.2) BMI: group I: mean 28.5 kg/m2 (SD 5.8); group II: mean 26.6 kg/m2 (SD 4.3); group III: mean 22.5 kg/m2 (SD6.5); group IV: mean 25.7 kg/m2 (SD 3.8) Inclusion criteria Healthy, white, postmenopausal women (≥ 1 year since last menstrual period) Age 45–70 Normal mammogram within 6 months of study No acute or chronic medical disorder Exclusion criteria Non‐white women ("because there may be differences in bone metabolism that is associated with race") Previous hysterectomy or reconstructive pelvic surgery Oestrogen, calcitonin, fluoride steroid, or diuretic therapy within 6 months of investigation Known contraindication to HRT or either SORM
Interventions	Group I (n = 15): CEO 0.625 mg, identical in appearance to all other interventions and provided as a single non‐gelatin‐containing capsule and administered orally Group II (n = 15): tamoxifen 20 mg, identical in appearance to all other interventions and provided as a single non‐gelatin‐containing capsule and administered orally Group III (n = 15): Raloxifene 60 mg, identical in appearance to all other interventions and provided as a single non‐gelatin‐containing capsule and administered orally Group IV (n = 13): placebo, identical in appearance to all other interventions and provided as a single non‐gelatin‐containing capsule and administered orally Co‐interventions: pelvic floor exercises. All participants were taught Kegel exercises, verified digitally during baseline. They were encouraged to perform 4 sets of 10 Kegels each day. In addition, "one patient used a pessary, which was removed for prolapse evaluations". Unclear which group they were in.
Outcomes	Cytologic assessment: epithelial cells obtained from right lateral vaginal fornix with a wooden spatula. Karyopyknotic index assessed in samples from the lateral vaginal fornix and mid‐urethra and then translated into the quantitative maturation value Modified POP‐Q: a modified version of the POP‐Q was used to increase descriptive power. Serial changes were defined as at least 2 points of change of at least 1cm in 4 POP‐Q points. Cotton swab test: Q‐tip test performed only in those participants who also presented with SUI (32.7% participants); significant change was defined as >20‐degree change in angle of deflection or 20‐degree increase in resting angle between before and after evaluations
Notes	Reason for Awaiting Classification: it is not clear whether all of the participants had POP at the beginning of the study as only change in POP‐Q is reported. Correspondence: We contacted the authors on 23 November 2021 for clarification but received an undeliverable response. On 24 November 2021, we attempted to contact the lead author via a contact form on the website of their institution (www.englewoodhealth.org/about‐englewood‐health/contact‐us) and via a different email address. We emailed the lead author again on 13 December 2021 but did not receive a response.

Zhu 2014.

Methods	Study design: RCT Dates study conducted: NR
Participants	Number of participants: 86 Country: China Setting: single centre Severity of POP: NR (eligibility criteria included women with Stage III or IV POP on POP‐Q but unclear how many) Affected vaginal compartment: NR Parity: NR Hysterectomy: NR Age: NR BMI: NR Inclusion criteria Severe POP (defined as POP‐Q Stage III to IV) or recurrent POP Referral for surgery Exclusion criteria NR
Interventions	Group I (n = 44): vaginal oestrogen. Transvaginal oestrogen cream (0.5 g) twice per week for 6 weeks Group II (n = 42): no oestrogen Co‐interventions: surgery: transvaginal pelvic floor reconstruction with mesh 6 weeks after diagnosis
Outcomes	Primary outcome Mesh exposure rate: measured at least 24 months after surgery Secondary outcomes Anatomic success rate: defined as a POP‐Q score < Stage II Subjective satisfaction rate: defined as absence of symptoms of vaginal bulging Surgical complications Bacterial infection rates related to the procedure: diagnosed by culture from a vaginal swab at 6 weeks postoperatively
Notes	Length of follow‐up: 2 years Conflict of interest: "Lan Zhu: No disclosures. Zhixing Sun: No disclosures." Study funding sources: NR Reason for awaiting classification: the population, interventions, and outcomes are very similar to those of Sun 2016. The relationship between the 2 studies is unclear. We contacted the study authors for clarification on 26 October 2021 and again on 9 November 2021 but have received no response.

BMI: body mass index; CEO: conjugated equine oestrogen; HRT: hormone replacement therapy; NR: not reported; POP: pelvic organ prolapse; POP‐Q: Pelvic Organ Prolapse Quantification System; RCT: randomised controlled trial; SORM: selective oestrogen receptor modulator; SUI: stress urinary incontinence.

Characteristics of ongoing studies [ordered by study ID]

Rahn 2021.

Study name	Investigation to Minimize Prolapse Recurrence of the Vagina Using Estrogen (IMPROVE)
Methods	Study design: RCT
Participants	Number of participants: 197 randomised at time of protocol publication Country: USA Setting: multicentre Inclusion criteria Age ≥ 48 years ≥ 1 year postmenopause (or with prior bilateral oophorectomy, still with uterus and no menses for > 1 year, posthysterectomy with or without oophorectomy, and > 55 years old) Vulvovaginal atrophy exposed to no oral, vaginal, or transdermal medication containing oestrogens (including SORMs), androgens or progestins within 4 weeks of screening Symptomatic POP (≥ stage 2 apical or anterior prolapse) Planned elective transvaginal native tissue apical repair Fitness for elective screening Physical ability to apply or insert the study drug Availability for follow‐up for ≤ 1 year Exclusion criteria Concurrent use of steroid creams for other indications (e.g. lichen sclerosis) BMI > 35 kg/m2 Recent history (within last month) of vaginal infection or vaginitis Contraindications to oestrogen therapy (e.g. spontaneous DVT, stroke, breast or endometrial/ hormone‐responsive cancer, genital bleeding of unknown cause) History of connective tissue disease Any oral or transdermal oestrogen, SORM, or other medication impacting vaginal milieu History of vaginal irradiation Allergy to Premarin or its constituents Prior apical repair or use of mesh for prolapse repair Current tobacco use
Interventions	Group I: conjugated oestrogen cream (Premarin). 0.625 mg/1 g cream, 1 g applied vaginally nightly for 2 weeks then 2x/week for > 5 weeks before surgery, then 2x/week for 1 year after surgery. Participants were able to apply cream with either a plastic applicator or fingertip. Group II: placebo cream. 1 g applied vaginally nightly for 2 weeks then 2x/week for > 5 weeks before surgery, then 2x/week for 1 year after surgery. Placebo cream was "identical" Co‐interventions: prolapse repair surgery. All participants underwent either uterosacral ligament suspension or unilateral sacrospinous ligament fixation at their surgeon's discretion.
Outcomes	Primary outcome Surgical "success" or "failure": "failure" at 1 year was defined by either objective prolapse of the anterior or posterior walls beyond the hymen or the apex descending > 1/3 of the total vaginal length, subjective sense of bulge, or retreatment. Secondary outcomes Condition‐specific (i.e. POP) quality of life: as measured by PFDI‐20 Condition‐specific (i.e. POP) quality of life: as measured by PFIQ‐7 Generic quality of life: as measured by SF‐12 Urinary symptoms: as measured by urinary subscale questions of the PFDI‐20 (i.e. UDI‐6) at 8 weeks and 12 months Urinary symptoms: as measured by urinary subscale questions of the PFIQ‐7 (i.e. the Urinary Impact Questionnaire) at 8 weeks and 12 months Sexual function: as measured by the PISQ‐12 at 8 weeks and 12 months Number of participants with adverse events
Starting date	December 2016
Contact information	David Rahn, MD, Associate Professor, Dept. of Obstetrics & Gynecology, University of Texas Southwestern Medical Center Email: david.rahn@utsouthwestern.edu
Notes	Follow‐up time points: 8 weeks (time of surgery) and 12 months Conflicts of interest: "The authors have declared they have no conflicts of interest". Study funding sources: "National Institute on Aging AG047290; Pfizer, Inc: WI195371—study drug provision"

Roovers 2018.

Study name	Research into effectiveness and costs concerning the use of oestrogen before and after vaginal prolapse surgery in women after menopause
Methods	Study design: RCT
Participants	Number of participants: 300 (target size) Country: Netherlands Setting: 22 participating centres (university, teaching and non‐teaching hospitals) Inclusion criteria Postmenopause (> 1 year amenorrhoea) Age ≥ 18 years old POP‐Q stage ≥ 2 or higher Planned primary POP surgery with native tissue repair, including at least anterior or posterior vaginal wall repair Exclusion criteria Previous POP surgery in concerning compartment Prolapse repair using mesh Current vaginal infection Use of oestrogens in past 12 months Contraindication for use of topical oestrogen Known, past, or suspected oestrogen‐dependent malignant tumours (e.g. breast cancer, endometrial cancer) Insufficient knowledge or understanding of the Dutch language
Interventions	Group I: 0.5 mg oestriol cream. 1 mg/g, topical administration from 4–6 weeks after surgery until 12 months after surgery. First 2 weeks 0.5 mg once daily, thereafter 0.5 mg twice weekly. In the first 2 weeks after surgery, no cream will be used. The total duration will be 58 weeks (6 weeks before surgery and 52 weeks after surgery). Group II: placebo cream. From 4–6 weeks after surgery until 12 months after surgery. First 2 weeks once daily, thereafter twice weekly. Co‐interventions: prolapse repair surgery
Outcomes	Primary outcome Subjective improvement in POP: assessed by PGI‐I Secondary outcomes Compound measure: no prolapse in compartment of surgery or past the hymen; no bothersome complaints of prolapse; no re‐intervention for prolapse in the compartment of surgery within the follow‐up period QALYs: EQ‐5D‐6L Disease specific quality of life: micturition, defecation, and sexual function Vaginal pH Signs of vaginal atrophy during gynaecological examination Complaints of vaginal atrophy Morbidity and adverse events Interventions for pelvic floor pathology Costs Adherence to treatment
Starting date	September 2018
Contact information	Jan‐Paul W.R. Roovers MD, PhD Email: j.p.roovers@amc.uva.nl Phone: +31 (0)20 5666429
Notes	Follow‐up time points: 3 and 12 months after surgery Conflicts of interest: "The authors declare that they have no competing interests". Study funding sources: "The study received an unrestricted grant from the 'Leading the Change' program from Stichting Zorgevaluatie Nederland (project number 80‐85009‐98‐1009), a Dutch foundation that consist of the Dutch Medical Specialists Federation, the Dutch Health Insurers and the Dutch Patient Federation". Correspondence: we contacted the study authors on 13 December 2021 to ask whether the EQ‐5D‐6L reported in the trial registration was the EQ‐5D‐3L, but we received no response.

Vincent 2013.

Study name	Hormonal status on blood flow and tissue in pelvic organ prolapse
Methods	Study design: RCT
Participants	Number of participants: 10 (actual) Country: USA Setting: single centre Inclusion criteria Age 21–70 years Stage ≥ 2 POP Wish to undergo surgery to treat POP Willingness and ability to comply with study procedures Willingness and ability to provide written informed consent Exclusion criteria Contraindication for oestrogen cream Any medical condition that in the opinion of the investigator would place the subject at increased risk if they participated History of connective tissue disease Previous hysterectomy or POP surgery Known allergic reaction to any agent required by the protocol Use of hormone therapy in postmenopausal women in last 90 days Pregnancy or lactation History of prior non‐compliance or history of psychiatric condition that would in the opinion of the investigator make it difficult for the subject to comply with study procedures or follow instructions
Interventions	Group I: postmenopausal, topical vaginal oestrogen cream Group II: postmenopausal, topical placebo cream Group III: premenopausal, no topical vaginal cream Co‐interventions: prolapse repair surgery
Outcomes	Primary outcome Vaginal epithelial thickness: measure of vaginal epithelial thickness obtained by optical coherence tomography Secondary outcome Histologic assessment of excised tissue: evaluation of histology from surgery for collagen and elastin
Starting date	June 2013
Contact information	Kathleen Vincent, Galveston, Texas, United States, 77550
Notes	Follow‐up time points: 1 month Conflicts of interest: NR (trial registration) Study funding sources: NR Other: started in 2013, but primary completion date is not until 2024.

Zhu 2020.

Study name	The effect of intra‐vaginal conjugated estrogen cream on ring pessary use for pelvic organ prolapse
Methods	Study design: RCT
Participants	Number of participants: 420 (estimated) Country: China Setting: single centre Inclusion criteria Symptomatic POP (stage II, stage III, or stage IV) Postmenopause (serum FSH > 40 IU/L or amenorrhoea for 12 months) Fitted ring with support pessary Ability to attend the clinical trial and follow‐up Willingness and ability to participate for up to 1 year and to provide informed consent Willingness and ability to apply vaginal cream Exclusion criteria Acute infection of internal genital tract HRT in last 3 months Suspected or untreated lower genital tract tumour Genital fistula Abnormally elevated intra‐abdominal pressure (e.g. ascites, tumour) Life expectancy < 1 year Vaginal oestrogen contraindications, including known or suspected oestrogen‐dependent malignancies (endometrial carcinoma, melanoma) Endometrial hyperplasia (thickness of endometrium ≥ 5 mm) Thrombophlebitis or thromboembolic disease Undiagnosed irregular vaginal bleeding Known or suspected breast cancer Allergy to any component of the oestrogen cream Severe liver disease Volume of postvoiding residual > 250 mL
Interventions	Group I: vaginal oestrogen cream. 1 g/day for the first 2 weeks then 1 g twice weekly Group II: vaginal placebo cream. 1 g/day for the first 2 weeks then 1 g twice weekly Co‐interventions: pessary. Participants with symptomatic POP will have an initial pessary fitting and will be followed up in 2 weeks. If the prolapse is supported, with no discomfort, and the participant decides to continue using the pessary, it is considered successful. Those who then consent will be randomised to either arm while continuing to wear the pessary for 3 months.
Outcomes	Primary outcome Proportion of participant using the pessary with very much or much improvement on the patient impression of improvement questionnaire Sustained wearing with satisfaction: continuous wearing is defined as the use of the pessary after successful fitting test. Continuous wearing time interval is from the start of wearing the pessary to cessation of pessary use. PGI‐I divided into 7 levels: very much better, much better, a little bit better, no change, a little bit worse, much worse, and very much worse. The responses "very much better" and" much better" indicate improvement. Secondary outcomes Comparison of symptom‐specific distress (PFDI‐20) at 6 months and 1 year in participants using the pessary in the vaginal oestrogen group and the placebo group at 1 year: scores 0–300, higher score means symptoms of PFD are more serious. Comparison of symptom specific impact (PFIQ‐7) at 6 months and 1 year in participants using the pessary in the vaginal oestrogen group and the placebo group at 1 year: PFIQ‐7 scores 0–300, higher score means a more severe influence on participant. Comparison of sexual function (PISQ‐12) at 6 months and 1 year in participants using the pessary in the vaginal oestrogen group and the placebo group at 1 year: PISQ‐12 scores 0–48, higher score means a more severe influence on participant. Comparison of the adverse events of subjects wearing a pessary in the vaginal oestrogen group and the placebo group at 1 year: participants will be asked about complications during each follow‐up visit.
Starting date	24 May 2020
Contact information	Lan Zhu: +86‐69156874; zhu_julie@vip.sina.com Ying Zhou: +86‐69156204; shellypumch@163.com
Notes	Follow‐up time points: every 3 months to 1 year Conflicts of interest: NR (trial registration) Study funding sources: NR

BMI: body mass index; DVT: deep vein thrombosis; FSH: follicle‐stimulating hormone; HRT: hormone replacement therapy; NR: not reported; PFD: pelvic floor distress; PFDI‐20: Pelvic Floor Distress Inventory‐20; PFIQ‐7: Pelvic Floor Impact Questionnaire‐7; PGI‐I: Patient Global Impression of Improvement; PISQ‐12: Pelvic Organ Prolapse/Urinary Incontinence Sexual Function Questionnaire; POP: pelvic organ prolapse; QALY: quality‐adjusted life year; RCT: randomised controlled trial; SF‐12: 12‐item Short‐Form Health Survey; SORM: selective oestrogen receptor modulator; UDI‐6: Urinary Distress Inventory.

Differences between protocol and review

See Taithongchai 2021 (review protocol).

• For greater transparency, we clarified the definition of surgery or pessaries as an intervention in Types of interventions.

• We separated compliance from minor adverse events to make it a stand‐alone outcome measure. We also consolidated the original pessary management outcomes into participant adherence to intervention.

• We further refined major and minor adverse events, splitting minor adverse events into sub‐categories and defining specific major adverse events (endometrial hyperplasia, thromboembolism, and Clavien‐Dindo Grade II events).

• We further refined surgical events as two separate outcomes: operative time and mesh exposure.

• We adjusted the timing of outcome measurements so that minor adverse events, major adverse events, and surgical outcomes were measured at the latest time point reported by study authors. This was to capture as much data surrounding these outcomes as possible.

• We adjusted the outcomes originally planned for the summary of findings tables due to the changes in the outcome measures. Following the refinement of the secondary outcome measures, we chose outcomes that we considered to be of clinical importance, as well as outcomes that would be applicable across all summary of findings tables.

  • We removed improvement or cure of associated urinary and bowel symptoms but kept improvement or cure of associated sexual symptoms.

  • We included generic quality of life or health status because it was a more inclusive outcome measure than condition‐specific quality of life.

  • We combined declining surgery (due to presumed satisfaction with oestrogen or pessary plus oestrogen) and discontinuation of support pessary (due to presumed dissatisfaction or lack of improvement) into one outcome (participant adherence to intervention) as an overall indicator of oestrogen effecting management.

  • We included grade II Clavien‐Dindo complications (e.g. surgical specific infection, haemorrhage, blood loss) as major adverse events related to oestrogen use.

• We clarified that where a study used multiple tools to measure a single outcome, we used the tool with the best validity according to Cronbach's alpha.

• We clarified that where a study reported multiple adverse event outcomes, we used data from the adverse event with the highest number of events to avoid double‐counting.

• We used the synthesis without meta‐analysis (SWiM) approach to narratively synthesise results, where possible and appropriate (Campbell 2020).

• Where numerical data such as standard deviations or correlation coefficients were missing, we had originally planned to use the methods described in the Cochrane Handbook for Systematic Reviews of Interventions to attempt to calculate them from other available statistics, such as P values (Higgins 2021). In the review, we only contacted study authors for missing data or information.

• We originally planned to produce summary of findings tables using the GRADEpro GDT software but instead created these manually using RevMan Web.

Contributions of authors

AT: study selection, data extraction, risk of bias assessments, writing the review.
EEJ: study selection, data extraction, risk of bias assessments, data analysis, writing the review.
SII: took part in planning the review and editing the review.
EBM: took part in planning the review.
AK: undertook the brief economic commentary.
RT: took part in planning the review and editing the review.

Sources of support

Internal sources

• No sources of support provided

External sources

• National Institute for Health Research (NIHR), UK

  This project was supported by the National Institute for Health Research, via Cochrane Infrastructure funding to Cochrane Incontinence. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Evidence Synthesis Programme, the NIHR, NHS or the Department of Health and Social Care.

Declarations of interest

In accordance with Cochrane's Commercial Sponsorship Policy, the following declarations are applicable for the three years prior to the publication date of this review.

AT: none known
EEJ: is Assistant Managing Editor for Cochrane Incontinence. She did not take any part in the editorial process for this review.
SII: none known
EBM: none known
AK: is an Editor with Cochrane. She did not take any part in the editorial process for this review.
RT: none known

New