Abstract
Background
Endometriosis is defined as the presence of endometrial tissue outside the uterine cavity. This chronic and recurring condition occurs in women of reproductive age. It is a common cause of pain or infertility and can cause non‐specific symptoms such as lower back pain, dyspareunia (pain during or after intercourse), and dysmenorrhoea (menstrual pain). Endometriosis is an oestrogen‐dependent disease. Medical treatment aims to relieve symptoms and shrink lesions by suppressing the normal menstrual cycle. In this review, we consider medication specifically aiming to modulate oestrogen receptors as an alternative method of treatment.
Objectives
To evaluate the effectiveness and safety of selective oestrogen receptor modulators (SERMs) in the management of endometriosis.
Search methods
We searched for trials in the following databases (from their inception to 28 May 2020): Cochrane Gynaecology and Fertility Group Specialised Register, Cochrane Central Register of Studies (CRS Online), MEDLINE, Embase, CINAHL, PsycINFO, and registers of ongoing trials. In addition, we searched all reference lists of included trials, and we contacted experts in the field, in an attempt to locate trials.
Selection criteria
We included randomised controlled trials (RCTs) comparing selective oestrogen receptor modulators (SERMs) with placebo, no treatment, other medical treatment, or surgery for endometriosis.
Data collection and analysis
We used standard methodological procedures recommended by Cochrane. Two review authors independently selected trials for inclusion, assessed risk of bias, and extracted data using data extraction forms. We used risk ratios (RRs) with 95% confidence intervals (CIs) for reporting dichotomous data. Primary review outcomes were relief of pelvic pain and adverse events. Secondary outcomes included quality of life, recurrence rate, and economic and fertility outcomes.
Main results
We included only one RCT, which included 93 women, comparing the SERM raloxifene with placebo in biopsy‐proven endometriosis. All women first underwent complete surgical excision of all lesions. Evidence was of very low quality: the main limitation was imprecision ‐ with very sparse data from only one small study, which included only women after surgical treatment.
Relief of pelvic pain
The included study did not specifically measure the primary outcome of pain relief. Study authors reported that time to return of pelvic pain (defined as two months of pain equal to or more severe than pain at study entry) was more rapid in the raloxifene group (P = 0.03).
Adverse events
The included study reported adverse events such as pelvic pain, ovarian cyst, headache, migraine, and depression. We are uncertain whether raloxifene improves the incidence of pelvic pain (RR 1.25, 95% CI 0.63 to 2.45), ovarian cysts (RR 1.57, 95% CI 0.55 to 4.43), headache (RR 1.09, 95% CI 0.49 to 2.43), migraine (RR 0.73, 95% CI 0.28 to 1.95), depression (RR 1.96, 95% CI 0.63 to 6.06), or other adverse events (RR 0.08, 95% CI 0.00 to 1.30) (all: 1 study, n = 93; very low‐quality evidence).
Quality of life
The study described a statistically significant difference in mental health quality of life (QoL) by 12 months, in favour of placebo treatment (mean difference 11.1, 95% CI 0.01 to 21.19). Other QoL data did not differ between groups but were not reported in detail.
Recurrence rate, fertility, and economic outcomes
We are uncertain whether raloxifene improves the recurrence rate of endometriosis, proven by biopsy, when compared to placebo (RR 1.20, 95% CI 0.66 to 2.21; 1 study, n = 93; very low‐quality evidence). This suggests that if 28% of women taking placebo have biopsy‐proven recurrence of endometriosis, between 19% and 62% of those taking raloxifene will do so. These outcomes are prone to bias, as not all women had an actual second laparoscopy. Recurrence based on symptoms (non‐menstrual pain, dysmenorrhoea, or dyspareunia) was described; in these cases, symptoms improved after use of raloxifene as well as after use of placebo.
The included study did not report data on economic outcomes. No comparative data were available on pregnancy, as the study included only women who agreed to postpone pregnancy until after the study endpoint; the few pregnancies that did occur were uneventful but were regarded as an adverse event.
Authors' conclusions
Based on a single, small RCT and incomplete data, we are uncertain of the effects of SERMs on pain relief in surgically treated patients with endometriosis. The included study was stopped prematurely because of higher pain scores among women who took SERMs when compared to scores among those receiving placebo. Further research is needed to fully evaluate the role of SERMs in endometriosis.
Plain language summary
Selective oestrogen receptor modulators (SERMs) for women with biopsy‐proven endometriosis
Review question
Cochrane Review authors reviewed the evidence on efficacy and safety of drugs called selective oestrogen receptor modulators (SERMs) for women with biopsy‐proven endometriosis.
Background
Endometriosis is a disease that affects 1 in 10 women of reproductive age. Endometriosis can cause abdominal and menstrual pain and infertility. Spreading and activity of the disease are dependent on the hormone oestrogen. Medical treatment consists of pain medication to relieve symptoms, oral contraceptive pills to suppress the menstrual cycle, or other hormonal treatments such as progestogen medication or hormonal intrauterine devices. SERMs aim to block the effects of oestrogen and could offer promise as medical treatment for endometriosis. We compared the effects and safety of SERMs versus placebo or other treatment for endometriosis.
Study characteristics
Only one randomised controlled trial was included in this review. This study compared the SERM 'raloxifene' versus placebo in 93 women who had undergone complete surgical excision of all endometriosis lesions. Evidence is current to 28 May 2020.
Key results
The included study did not report the primary outcome of pain relief. Adverse events such as pelvic pain, ovarian cyst, headache, migraine, and depression did not differ between study groups. We are uncertain whether raloxifene improves adverse outcomes when compared with placebo (1 study, 93 women; very low‐quality evidence). We are uncertain whether raloxifene improves the recurrence rate of endometriosis, as proven by biopsy, when compared to placebo (1 study, 93 women; very low‐quality evidence). This suggests that if 28% of women taking placebo have biopsy‐proven recurrence of endometriosis pelvic pain, between 19% and 62% of those taking raloxifene will do so. These outcomes are limited, as not all women had an actual second laparoscopy to take biopsy samples. However, study authors reported that women in the raloxifene group were likely to experience return of pain sooner than women in the placebo group. The included study reported that quality of life measurements did not differ between raloxifene and placebo groups, except in the domain of mental health. By 12 months, mental health quality of life scores favoured placebo treatment (mean difference 11.1, 95% confidence interval (CI) 0.01 to 21.19). There were no usable data on fertility nor on economic outcomes.
Based on the single study included in this review, reviewers found no evidence of a beneficial effect of SERMs as treatment for pain relief in surgically treated endometriosis. On the contrary, the study was prematurely stopped because women who used SERMs experienced a return of pain sooner than women who used placebo.
Quality of the evidence
Evidence was of very low quality. The main limitations were inclusion of only one study with a small number of participants and lack of both primary and secondary outcome data.
Summary of findings
Summary of findings 1. Raloxifene versus placebo for endometriosis.
| Raloxifene versus placebo for endometriosis | ||||||
|
Population: women with endometriosis, following complete surgical excision of lesions
Settings: outpatient clinic
Intervention: raloxifene Comparison: placebo | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect (95% CI) | No. of participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | Raloxifene | |||||
| Relief of pelvic pain | Not reported | |||||
| Adverse events ‐ pelvic pain | 239 per 1000 |
299 per 1000 (151 to 586) |
RR 1.25 (0.63 to 2.45) |
93 (1 study) | ⊕⊝⊝⊝ very lowa,b,c |
|
| Adverse events ‐ ovarian cyst | 109 per 1000 |
171 per 1000 (60 to 482) |
RR 1.57 (0.55 to 4.43) |
93 (1 study) | ⊕⊝⊝⊝ very lowa,b,c |
|
| Adverse events ‐ headache | 196 per 1000 |
213 per 1000 (96 to 475) |
RR 1.06 (0.48 to 2.37) |
93 (1 study) | ⊕⊝⊝⊝ very lowa,b,c |
|
| Adverse events ‐ migraine | 174 per 1000 |
127 per 1000 (49 to 339) |
RR 0.73 (0.28 to 1.95) |
93 (1 study) | ⊕⊝⊝⊝ very lowa,b,c |
|
| Adverse events ‐ depression | 87 per 1000 |
170 per 1000 (55 to 527) |
RR 1.91 (0.62 to 5.92) |
93 (1 study) | ⊕⊝⊝⊝ very lowa,b,c |
|
| Biopsy‐proven recurrence of endometriosis Follow‐up: mean 18 months | 283 per 1000 | 339 per 1000 (187 to 625) | RR 1.20 (0.66 to 6.21) | 93 (1 study) | ⊕⊝⊝⊝ very lowb,d,e,f | |
| *The basis for the assumed risk is the mean control group risk. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio. | ||||||
| GRADE Working Group grades of evidence. High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate. | ||||||
aRisk of bias: downgraded by one level for serious risk in the domains incomplete outcome, selective reporting, and 'other' bias; a large number of women reduced dose or stopped taking drugs because of possible adverse events.
bImprecision: downgraded by two levels due to small sample size and few events (< 300); wide confidence intervals.
cNo issues with indirectness, inconsistency (one study included), or publication bias.
dHigh risk of bias in the domains incomplete outcome, selective reporting, and 'other bias; not all women underwent second laparoscopy and had a chance of biopsy; incomplete data because of loss to follow‐up.
eIndirectness: downgraded by one level, as included women are a select population of women who had undergone therapeutic surgery with complete excision of visible endometriosis and no pain at study entry.
fNo issues with Inconsistency (only one study included) or publication bias.
Background
Description of the condition
Endometriosis is a benign gynaecological disease that affects 2% to 10% of women at reproductive age and up to 50% of infertile women (Eskenazi and Warner 1997; Meuleman 2009). In this condition, endometrial tissue is dispersed outside the uterus, for example, on the peritoneum, ovaries, rectovaginal septum, bladder, or intestines. Severe disease leads to altered pelvic anatomy and extensive adhesions, which can cause pain and lead to infertility. Pelvic pain, dyspareunia (pain during and after intercourse), and dysmenorrhoea (menstrual pain) are the most common symptoms of endometriosis, but one‐third of women are asymptomatic or present with atypical symptoms.
Because of the wide variety of symptoms, women with endometriosis encounter quite a delay in diagnosis, leaving their symptoms largely untreated until they experience progressive disease. As it mainly affects pre‐menopausal women, endometriosis imposes a huge social and economical burden (Bianconi 2007; Jones 2002; Marques 2004).
A lot of young women with suspected endometriosis start empirical treatment with analgesics and hormonal therapy. The definitive diagnosis requires surgically retrieved histology. However, surgery is not always necessary, as empirical therapy might be effective without the need for a confirmed diagnosis.
Although evidence on the use of hormonal therapy is limited, the European Society of Human Reproduction and Embryology (ESHRE) compiled recommendations regarding medical treatment (ESHRE Guideline 2014). Endometriosis‐associated pelvic pain may be reduced by the use of a combination of analgesics and oestrogen‐blocking therapy such as hormonal contraceptives, progestogens, and anti‐progestogens (Brown 2012), along with gonadotropin‐releasing hormone (GnRH) agonists (Brown 2010; Vercellini 1993). Besides medical treatment, surgical coagulation or excision of endometriosis is an option. However, surgery carries risk of complications (Kondo 2011), which makes medical treatment the first step in controlling endometriosis‐associated symptoms.
A lot of research conducted over the last couple of years has examined the development of new medical treatments. The fact that endometriosis is an oestrogen‐dependent disease suggests that there could be a role for drugs called selective oestrogen receptor modulators (SERMs).
Description of the intervention
Endometriosis is a chronic and multi‐factorial condition for which aetiology and pathophysiology remain largely unknown. Hormonally active endometrial lesions cause an excessive amount of oestrogen and an inflammatory reaction of the body, leading to pain. Suppression of oestrogen could lead to a reduction in pelvic pain.
SERMs are used widely for oestrogen‐related conditions that affect a woman's health including treating and reducing the risk of breast cancer and treating osteoporosis and dyspareunia in postmenopausal women. These drugs differ from pure oestrogen agonists and antagonists by their different mechanisms in various tissues, which grants them the opportunity to selectively stimulate or inhibit oestrogen response (Maximov 2013).
Raloxifene is a SERM that has been used to treat osteoporosis since 1999. It has a beneficial oestrogen agonist effect on bone density and can reduce the incidence of atherosclerosis, but raloxifene stimulates neither the endometrium nor the breasts in postmenopausal women (Eng ‐Wong 2004; Francucci 2005; Scott 1990; Silfen 1999; Thiebaud 2001). This suggests that raloxifene might be an effective treatment for endometriosis pain, without the risk of bone density loss that accompanies a prolonged hypo‐oestrogenic status. However raloxifene has been associated with adverse effects including increased risk of thromboembolism, vaginal dryness, generalised rash, and abdominal cramps.
How the intervention might work
The growth and survival of endometrial tissue depend on oestrogen. In the human body, oestrogen is derived from multiple sources. Endometrial tissue consists of intrinsic aromatase activity leading to conversion of cholesterol to oestrogen. Prostaglandin‐E2 receptors also express steroidogenic genes by activating cyclic adenosine monophosphate (cAMP) activity, leading to increased production of oestrogen. Peripheral fat produces oestrogen, contributing to its increased presence (Rafique 2017). SERMs have oestrogen‐antagonistic effects on endometrial tissue. They interact with oestrogen receptors and block the hormonal signalling pathway, which leads to reduction in oestrogen activity and possible reduction in endometriosis pain (Delmas 1997; Simsa 2007).
Why it is important to do this review
The chronic aspect of the disease shows the necessity for therapies that provide long‐term benefit, including prevention of progression and recurrence of pain (Becker 2017). With the debilitating effects of pelvic pain on quality of life and the large social and economic burden that endometriosis inflicts on women, the search for a new, safe therapy without long‐term effects on, for example, bone density, is ongoing (Vitale 2016).
Objectives
To evaluate the effectiveness and safety of selective oestrogen receptor modulators (SERMs) in the management of endometriosis.
Methods
Criteria for considering studies for this review
Types of studies
We included all randomised controlled trials (RCTs) evaluating the use of selective oestrogen receptor modulators (SERMs) for treatment of pain associated with endometriosis. We considered cross‐over trials eligible, but we would include only pre‐cross‐over data in the meta‐analysis, as the cross‐over is not a valid design in this context. We excluded quasi‐randomised and non‐randomised studies (case control studies, cohort studies).
Types of participants
We included women of reproductive age with diagnosed pain and suspected endometriosis. This could be clinically diagnosed, visually diagnosed, or biopsy‐proven endometriosis (Dunselman 2014; RCOG 2006).
Types of interventions
We considered trials comparing SERMs with any route of administration versus any other active medical or surgical intervention, placebo, or no treatment as eligible for inclusion. We included any dosage or duration of treatment. Available SERMs included raloxifene, arzoxifene, levormeloxifene, ospemifene, lasofoxifene, bazedoxifene, centchroman, arzoxifene, tamoxifen, toremifene, droloxifene, and idoxifene.
Types of outcome measures
Primary outcomes
-
Relief of pelvic pain, assessed on a visual analogue scale (VAS) (0 (no pain) to 10 (worst pain imaginable)), or subjective improvement (cured, better, same, or worse)
Pelvic pain as reported in studies
Painful symptoms defined as dysmenorrhoea, dyspareunia, or pelvic or lower abdominal pain
Adverse events resulting from treatment (hot flushes, endometrial cancer, vaginal bleeding, thrombosis, or other adverse events) during or after treatment
Secondary outcomes
Quality of life (by quality of life scores (as reported in included RCTs))
Recurrence rate of endometriosis (symptomatic or biopsy‐proven)
-
Fertility outcomes (only in women wishing to conceive)
Live birth: defined as delivery of a live foetus
Clinical pregnancy: defined as evidence of a gestational sac on ultrasound
Fertility‐related adverse events (miscarriage, multiple birth)
Economic outcomes (costs, time taken for treatment)
Search methods for identification of studies
We sought all published and unpublished RCTs of SERMs versus other interventions via the following search strategy, without language or date restriction, and in consultation with the Cochrane Gynaecology and Fertility Group Information Specialist. We identified relevant trials by searching electronic databases and other resources.
Electronic searches
We searched the following electronic databases, trial registers, and websites from their inception to 28 May 2020:
Cochrane Gynaecology and Fertility Group (CGFG) Specialised Register of controlled trials, ProCite platform, searched 28 May 2020 (Appendix 1);
CENTRAL via the Cochrane register of Studies Online (CRSO), Web platform, searched 28 May 2020 (Appendix 2);
MEDLINE, Ovid platform, searched from 1946 to 28 May 2020 (Appendix 3);
Embase, Ovid platform, searched from 1980 to 28 May 2020 (Appendix 4);
PsycINFO, Ovid platform, searched from 1806 to 28 May 2020 ( Appendix 5);
CINAHL (Cumulative Index to Nursing and Allied Health Literature), Ebsco platform, searched from 1982 to 28 May 2020 (Appendix 6);
Chinese language electronic databases: Chinese Biomedical Literature Database (CBM), Web platform, (http://www.sinomed.ac.cn/cross/advancedSearch.html) (1978 to 28 May 2020) and Chinese Medical Current Contents (CMCC) (http://librarian.wanfangdata.com.cn/) (1983 to 28 May 2020); using the corresponding Chinese terms for: “endometriosis” AND “SERMs”, “ERs”,“Raloxifene”;
ClinicalTrials.gov (http://clinicaltrials.gov/ct2/home) Web platform, searched 28 May 2020;
World Health Organization International Trials Registry Platform search portal (www.who.int/trialsearch/Default.aspx); Web platform, searched 28 May 2020;
Citation indexes (http://scientific.thomson.com/products/sci/), Web platform, searched 28 May 2020;
Conference abstracts in the Web of Knowledge (http://wokinfo.com/), Web platform, searched 28 May 2020;
Latin American Caribbean Health Sciences Literature database via the Virtual Health Library Regional Portal, (http://regional.bvsalud.org/php/index), Web platform, searched 28 May 2020;
PubMed (http://www.ncbi.nlm.nih.gov/pubmed/), Web platform, searched 28 May 2020;
OpenGrey database (http://www.opengrey.eu/), Web platform, searched 28 May 2020;
Google for grey literature, Web platform, searched 28 May 2020.
The MEDLINE search was combined with the Cochrane highly sensitive search strategy for identifying RCTs, which appears in the searching chapter of the Cochrane Handbook for Systematic Reviews of Interventions. The Embase search was combined with trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN) (https://www.sign.ac.uk/what-we-do/methodology/search-filters/).
Searching other resources
We also did the following.
We searched the references lists of all included studies and relevant reviews to identify further relevant articles.
We contacted study authors and experts in the field to look for potential studies.
We handsearched the following Chinese journals.
Chinese Journal of Obstetrics and Gynaecology (from 1953 to 28 May 2020).
Chinese Journal of Practical Gynaecology and Obstetrics (from 1985 to 28 May 2020).
Journal of Practical Obstetrics and Gynaecology (from 1985 to 28 May 2020).
Maternal and Child Health Care of China (from 1986 to 28 May 2020).
Data collection and analysis
Selection of studies
After an initial screening of titles and abstracts retrieved by the search, conducted by CY and MH or SM, we retrieved the full texts of all potentially eligible studies. Two review authors (WQ and ZA or MH and SM) independently examined these full‐text articles for compliance with the inclusion criteria and selected eligible studies. We retrieved the full text if information given in titles, abstracts, and keywords suggested that the study used SERMs as an intervention in endometriosis. We corresponded with researchers as required, to clarify eligibility. We resolved disagreements by discussion with a third review author. If any reports required translation, we described the process used for data collection. We documented the selection process in a PRISMA flow chart.
Data extraction and management
We extracted data from eligible studies using a special data extraction form. Two review authors (CY and WQ or MH and SM) independently extracted data, and we resolved any disagreements between review authors by discussion with a third review author (ZA). Data extracted included study characteristics and outcome data. When studies had multiple publications, we used the main trial as a reference and obtained additional details from secondary reports. We corresponded with trial authors to request further data on methods and/or results, as required.
Assessment of risk of bias in included studies
We assessed included studies for risk of bias using the Cochrane 'Risk of bias' assessment tool (Higgins 2011), including selection bias (random sequence generation, allocation concealment); performance bias (blinding of participants and personnel); detection bias (blinding of outcome assessors); attrition bias (incomplete outcome data); reporting bias (selective reporting); and other biases. Two review authors (MH and SM) assessed risk of bias and resolved disagreements by consensus or by discussion with a third review author (ZA). We searched for published protocols and compared outcomes between protocols and the published study. We included the risk of bias table in the Characteristics of included studies section and resolved disagreements by discussion with a third review author. We fully described all judgements. We presented conclusions in the 'Risk of bias' table and incorporated them into our interpretation of review findings by performing sensitivity analyses.
Measures of treatment effect
For dichotomous data, we recorded the number of participants experiencing the event in each group of the trial. We used risk ratios (RRs) with 95% confidence intervals (CIs) to report dichotomous data. For continuous outcomes, we extracted the final value and the standard deviation of the outcome of interest, along with the number of women assessed at endpoint in each treatment arm and at end of follow‐up. We used mean differences (MDs) with 95% CIs to report continuous data. In the case of outcomes with continuous data on different scales, we used standardised mean differences (SMDs) with 95% CIs. When data to calculate RRs or MDs were not reported, we utilised the most detailed numerical data available that facilitated similar analyses of included studies (e.g. test statistics, P values), or we reported the results in narrative form only.
Unit of analysis issues
The primary analysis was per woman randomised. We measured fertility outcomes among women seeking pregnancy. For certain fertility outcomes (e.g. miscarriage, multiple birth), we used per pregnancy data.
Dealing with missing data
We tried to obtain missing data from the original investigators. We analysed data on an intention‐to‐treat basis as far as possible (analysis included all randomised participants in the groups to which they were originally randomly assigned). When missing data could not be obtained, we undertook imputation of individual values for primary outcomes. If studies reported sufficient detail for calculation of mean differences but provided no information on associated standard deviations (SDs), we assumed the outcome to have an SD equal to the highest SD from other studies within the same analysis. For other outcomes, we analysed only available data.
Assessment of heterogeneity
We considered whether the clinical and methodological characteristics of included studies were sufficiently similar for meta‐analysis to provide a clinically meaningful summary. We assessed statistical heterogeneity by using the I² measure. An I² measurement greater than 50% was taken to indicate substantial heterogeneity (Higgins 2011). We planned to use a fixed‐effect model.
Assessment of reporting biases
We aimed to minimise publication bias and other reporting biases by ensuring a comprehensive search for eligible studies and by being alert for duplication of data. We assessed potential publication bias by using a funnel plot or other corrective analytical methods if 10 or more studies were included in an analysis (Egger 1997). We detected study reporting bias by searching published protocols and comparing outcomes between the protocol and the final published study.
Data synthesis
If sufficient clinically similar studies were available, we would have pooled data in meta‐analyses using Review Manager software (Review Manager 2014). For continuous outcomes, we determined the MD between treatment arms at end of follow‐up if all trials measured the outcome on the same scale; otherwise we calculated the SMD. An increase in the odds of a particular outcome would be displayed to the right of the centre line. Although the precise scaling of diary data would vary when a VAS score was used, trials used mean VAS score or reported a reduction in VAS score (with 50% reduction considered clinically significant). Different comparisons would be analysed separately.
-
SERMs versus placebo, subgrouped by mode of administration and by dose (we planned to pool data: low dose oral versus high dose oral).
All SERMs versus placebo or no treatment.
All SERMs versus alternative active therapy, stratified by all SERMs versus medical treatment and versus surgery.
Subgroup analysis and investigation of heterogeneity
When data were available, we intended to explore the following potential sources of heterogeneity using subgroup analyses.
Individual types of SERMs versus placebo, no treatment, or each alternative active therapy.
Duration of treatment (three months, three to six months, six to 12 months, at least 12 months).
When substantial heterogeneity is present, we would explore possible explanations including individual study risk of bias and dose of SERM.
Sensitivity analysis
We considered sensitivity analyses for primary outcomes to determine whether the conclusions were sensitive to arbitrary decisions made regarding eligibility and analysis. These analyses included consideration of whether conclusions had differed if we excluded studies with high risk of bias.
Summary of findings and assessment of the certainty of the evidence
We prepared a 'Summary of findings' table using GRADEpro software (GRADEpro GDT 2014), along with Cochrane methods (Higgins 2011). This table evaluated the overall quality of the body of evidence for primary review outcomes (relief of pelvic pain and adverse events following SERM treatment) and disease recurrence for the main review comparison (SERMs versus placebo or no treatment), using GRADE criteria (Atkins 2004) (risk of bias, consistency of effect, imprecision, indirectness, and publication bias). We justified, documented, and incorporated into reporting of results judgements about evidence quality for each outcome (high, moderate, low, or very low).
Results
Description of studies
We briefly summarised characteristics of identified studies in the tables (Characteristics of included studies;Characteristics of excluded studies).
Results of the search
We obtained a total of 251 citations via the electronic search strategy on 29 September 2016. Handsearching of reference lists and journals resulted in retrieval of 10 additional reports. An updated search on 28 May 2020 yielded 25 citations. Of these 286 reports, 280 were review articles, which did not meet inclusion criteria and were excluded at title and abstract screening. We assessed the remaining studies for eligibility at full‐text review. We excluded four studies with reasons. We included only one RCT (two references) of raloxifene versus placebo in women with endometriosis. See the PRISMA chart at Figure 1 (Moher 2009).
1.
Study flow diagram.
Included studies
Design
The only included study was a randomised, double‐blind, placebo‐controlled study of women with biopsy‐proven endometriosis lesions (Stratton 2008). A total of 93 participants were eligible for inclusion in this trial. This single‐centre study was conducted at the Clinical Centre of the US National Institutes of Health (NIH) and was not commercially funded.
Participants
This study enrolled women aged 18 to 45 years with a history of pelvic pain for about three months and with biopsy‐proven endometriosis at study laparoscopy who had undergone complete excision of lesions. Patients were required to have had significant postoperative pelvic pain reduction and to have used only the trial medication, without additional treatment for endometriosis, within six months following laparoscopy.
Interventions
Raloxifene tablets (180 mg once a day) that were ground and blended with lactose and placebo and contained lactose and inert materials were used for a six‐month period in this study. At NIH, a Pharmacy Development Service formulated study agents and placebos for placebo‐controlled trials. As part of this service, each batch was tested to confirm its potency and underwent other standard quality assurance testing. In this clinical trial, raloxifene was formulated this way to make it similar in appearance to placebo. There was no other potential adverse impact with reformulating the tablets. Quality of life and pelvic pain questionnaires were completed every three months until pelvic pain had returned, or for 18 months after completion of study medication treatment. After November 2002, a telephone interview replaced the 3‐, 9‐, 15‐, and 21‐month visits for asymptomatic participants. Women were offered a second laparoscopy at two years, or earlier if pain returned. By excising any new lesions, the goal was to diminish pain again. This approach enabled us to see whether endometriosis had reoccurred with the return of pain.
Outcomes
All outcome measures were observed for up to 18 months post intervention or for up to 2 years after first laparoscopy. Besides laparoscopy, no alternative treatment was described. Quality of life (according to the Duke Health Profile in Parkerson 1990) and pelvic pain questionnaires (in Brosens 1993) were completed preoperatively and then every three months until pelvic pain returned. Outcomes reported were recurrence of pain, presence and severity of pelvic pain rated on a VAS, quality of life, and menstrual cycle length. Adverse events were described.
Excluded studies
We excluded four studies for various reasons (Barra 2019; Harada 2018; Li 1999; Parker 2006). See Characteristics of excluded studies.
Risk of bias in included studies
We have summarised risk of bias from the included study in Figure 2 and Figure 3.
2.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
3.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
We rated the included study at low risk of selection bias related to sequence generation, as the Pharmaceutical Development Service created the allocation sequence, using a table of random numbers and alternating blocks of 8 and 10. The allocation sequence was accessible only to the pharmacy.
Blinding
Treatment assignment was concealed from study staff and participants until the study ended. So we rated the study as having low blinding risk.
Incomplete outcome data
The primary review outcome was pelvic pain relief. In this study, pelvic pain relief was not reported and per patient data on clinical symptoms of recurrence were not available.
The intention‐to‐treat principle was used in this trial. Non‐completing participants were categorised based on their pain status at last study visit. A total of 93 women entered the study; 20 of these were lost to follow‐up. In total, 37 patients with adverse events had dose reduced or stopped study medication early. In total, eight patients without adverse events delayed the start of treatment, stopped medication, or took less study medication than was recommended. Nine patients refused second surgery; eight became pregnant; and three fell pregnant while on medication and stopped medication after pregnancy was diagnosed. Another four participants became pregnant after completing medication at six months, which could possibly influence post‐study outcomes. This means that 16 of 46 (35%) in the raloxifene group and 20 of 47 (43%) in the placebo group did not complete the protocol.
Quality of life measures were reported as similar between raloxifene and placebo groups; detailed data were not provided.
For the above reasons, we rated risk of attrition bias as high.
Selective reporting
The study protocol of Stratton 2008 was available at ClinicalTrials.gov (https://clinicaltrials.gov/ct2/show/study/NCT00001848) and had been registered prospectively in 1999. However, trial investigators did not report the outcome measures for this study in the protocol. Therefore we consider it at high risk of reporting bias.
Adverse events named in the protocol were not reported in the study. It is unknown whether these adverse events did not occur, or whether they were not reported. Thus, we assigned high risk of selective reporting.
Other potential sources of bias
The prospectively registered study protocol did not encompass study outcomes. This might have induced other sources of bias.
We rated the risk of other potential bias as high. During the study, women with possible adverse effects were sometimes advised to reduce or stop study medication. It is unknown whether the adverse effects women may have experienced would have been actual adverse events, and whether women would have reduced or stopped the study medication by themselves. We found no potential sources of within‐study bias in the study.
Effects of interventions
See: Table 1
Primary outcomes
Relief of pelvic pain on a VAS (0 (no pain) to 10 (worst pain imaginable)), or subjective improvement
VAS and subjective verbal ratings from pain questionnaires for dysmenorrhoea, dyspareunia, and non‐menstrual pelvic pain were used to measure experienced pain at several moments of follow‐up in this study. The RCT described return of pain as a primary outcome measure. The raloxifene group had significantly earlier return of pain than the placebo group (P = 0.03). However, no data on relief of pelvic pain were provided.
Adverse events resulting from treatment with SERMs either during or following treatment
The included study has a table documenting adverse events in both groups. Stratton 2008 reported 37 patients (15/47 in the raloxifene group, 22/46 in the placebo group) with adverse effects who either reduced dose or stopped study medication early at a similar rate between groups. Adverse events consisted of pelvic pain, ovarian cysts, headache, migraine, and depression. We are uncertain whether raloxifene improves adverse outcomes when compared with no treatment (Analysis 1.1). We are uncertain whether raloxifene improves the incidence of pelvic pain (risk ratio (RR) 1.25, 95% confidence interval (CI) 0.63 to 2.45), ovarian cysts (RR 1.57, 95% CI 0.55 to 4.43), headache (RR 1.09, 95% CI 0.49 to 2.43), migraine (RR 0.73, 95% CI 0.28 to 1.95), depression (RR 1.96, 95% CI 0.63 to 6.06), or other adverse events (RR 0.08, 95% CI 0.00 to 1.30) (all: 1 study, n = 93; very low‐quality evidence; Analysis 1.1).
1.1. Analysis.
Comparison 1: Raloxifene versus placebo, Outcome 1: Adverse events
Secondary outcomes
Quality of life (by quality of life scores)
The study described a statistically significant difference in mental health QoL by 12 months, in favour of placebo treatment (mean difference (MD) 11.1, 95% CI 0.01 to 21.19; 1 study, n = 93; very low‐quality evidence; Analysis 1.2). No detailed data were reported for other domains of QoL.
1.2. Analysis.
Comparison 1: Raloxifene versus placebo, Outcome 2: Quality of life ‐ Mental health
Recurrence rate of endometriosis (biopsy‐proven or by symptoms)
Biopsy‐proven recurrence
In the study, second laparoscopy was performed in cases of return of pain or at two years' follow‐up. Among those who had second laparoscopy, this was performed before two years in 12 of 47 women in the raloxifene group versus 4 of 46 in the placebo group. Second laparoscopy took place at two‐year follow‐up in 11 of 47 in the raloxifene group and in 14 of 46 in the placebo group.
Moreover, 16 of 23 women in the raloxifene group and 13 of 17 women in the placebo group had biopsy‐proven return of endometriosis. There was no association between biopsy‐proven recurrence of lesions and the experience of return of pain. We are uncertain whether raloxifene improves recurrence of biopsy‐proven endometriosis when compared to placebo (RR 1.20, 95% CI 0.66 to 2.21; Analysis 1.3). This suggests that if 28% of women taking placebo have biopsy‐proven recurrence of endometriosis, between 19% and 62% of those taking raloxifene will do so.
1.3. Analysis.
Comparison 1: Raloxifene versus placebo, Outcome 3: Recurrence rate of endometriosis ‐ Biopsy proven
It is important to note that 24 of 47 women (51%) in the raloxifene group and 28 of 46 (61%) in the placebo group did not have second laparoscopy for various reasons (refused surgery, lost to follow‐up, pregnant, or had ongoing follow‐up at end of study).
Symptom‐based recurrence
Because of lack of per person data, the composite score for recurrence of non‐menstrual pain, dysmenorrhoea, or dyspareunia remained unknown. Table 2 shows differences in non‐menstrual pain for raloxifene and placebo over a 12‐month period. Non‐menstrual pain shows improvement of 1.4 points on the subjective pain scale with use of raloxifene compared to 1.7 points with use of placebo. Dysmenorrhoea improved by 1.3 points with raloxifene and by 2.07 points with placebo. Dyspareunia improved by 2.7 points with raloxifene and by 1.7 points with placebo (Table 2).
1. Recurrence of symptoms.
| Symptoms |
Baseline (n = 46) |
12 months (n = 19) |
Mean differencea |
| Non‐menstrual pain | |||
| Raloxifene | 5.6 (SE 0.5) | 4.0 (SE 0.5) | ‐1.4 (SD 0.71) |
| Placebo | 4.7 (SE 0.4) | 3.0 (SE 0.4) | ‐1.7 (SD 0.57) |
| Dysmenorrhoea | |||
| Raloxifene | 6.7 (SE 0.8) | 5.4 (SE 0.7) | ‐1.3 (SD 1.06) |
| Placebo | 6.4 (SE 0.7) | 3.7 (SE 0.8) | ‐2.07 (SD 1.06) |
| Dyspareunia | |||
| Raloxifene | 5.7 (SE 0.7) | 3.0 (SE 0.5) | ‐2.7 (SD 0.86) |
| Placebo | 5.4 (SE 0.7) | 3.2 (SE 0.5) | ‐1.7 (SD 0.86) |
aDifferences between baseline and 12 months after laparoscopy.
SD: standard deviation; SE: standard error.
Fertility outcomes (only in women wishing to conceive)
Live birth
Clinical pregnancy
Fertility‐related adverse events (miscarriage, multiple birth)
Fertility outcomes were not examined within the included RCT. Women who were pregnant were excluded from the study. Included women agreed to postpone pregnancy during the study using non‐hormonal contraception. Four pregnancies occurred during the study period (three in the placebo group, one in the raloxifene group), and treatment with either placebo or raloxifene was stopped. During the study, pregnancy was recommended to be avoided, so we cannot make any statements about this. Four women became pregnant after completing the study. Study authors reported that all women had uncomplicated pregnancies and gave birth to normal infants.
Economic outcomes (costs, time taken for treatment)
Costs were not reported in the included study.
Discussion
Summary of main results
Evidence presented in this review is limited because of the inclusion of only one study, which included few participants. The included study provided low‐quality evidence due to very sparse data, which included only women after surgical treatment. Potential future studies for inclusion might look into selective oestrogen receptor modulators (SERMs) as a primary treatment option, without prior surgery. A recent Cochrane Review shows that women who receive postsurgical medical therapy compared with no medical therapy or placebo may experience benefit in terms of pain recurrence, disease recurrence, and pregnancy (Chen 2020). In the included study, raloxifene was given as postoperative therapy to 47% of the women with endometriosis who had already undergone one or more laparoscopies. Moreover, 31% had already undergone two or more laparoscopies.
Moreover, in the single included RCT, the primary review outcome ‐ pain relief ‐ was not reported. This study reported on recurrence of pelvic pain based on a subjective pain rating scale. However, because of lack of per person data, the composite score for recurrence of non‐menstrual pain, dysmenorrhoea, or dyspareunia remains unknown.
Also, no usable data were provided on the secondary review outcomes of economic costs, fertility, and quality of life (QoL) other than mental health QoL.
Adverse events reported in the included randomised controlled trial (RCT) show no clear differences between raloxifene and placebo. However, adverse events named in the protocol differed from those reported in the included RCT. It is unknown whether adverse events did not occur or were not reported.
Risk of bias in the included RCT is high because of high loss to follow‐up or deviation from the protocol. Prematurely stopping study medication or reducing dose frequently occurred. In all, 9.7% of included women refused to undergo second laparoscopy, which potentially biased measurement for recurrence of endometriosis.
Based on this study, we remain uncertain whether use of SERMs improves recurrence of endometriosis. On the contrary, the only included study comparing raloxifene and placebo showed an even more rapid return of pelvic pain in the raloxifene group (P = 0.03), after which the study was prematurely stopped.
Overall completeness and applicability of evidence
In this review, only raloxifene was studied; no other types of SERMs were investigated.
The identified study only partially addressed the objectives of this review. The included RCT included only women with endometriosis who had undergone previous surgery and treatment of endometriosis lesions and had significant postoperative pelvic pain reduction. Women without pelvic pain reduction were not eligible for inclusion. Both the highly selected group and the large number of participants lost to follow‐up make these data not generalisable to the general population of women with endometriosis.
Quality of the evidence
This systematic review included only one RCT, which randomised 93 women for the comparison of raloxifene versus placebo. Therefore, the major risks of bias are imprecision, attrition bias, and selective reporting bias. Hence, evidence yielded by this study is of 'very low' quality according to GRADE criteria.
Potential biases in the review process
To prevent bias in the review process, the search was guided and developed by the Cochrane Gynaecology and Fertility Group. No limitations (such as language restriction) were applied to the search. Study selection, 'risk of bias' assessment, and data collection were conducted independently by four review authors without blinding. Disagreements were resolved by discussion with the third review author. The included study has an extremely small sample, from which any potential treatment effect could not be determined. The review authors have no conflicts of interest.
Agreements and disagreements with other studies or reviews
Even after an updated search in 2020, review authors identified no clinical trials on SERM drugs for endometriosis other than the single study included in our review. No other systematic reviews on SERM drugs for endometriosis have been conducted. The absence of good clinical trials may suggest the need to find a new focus of research in the treatment of endometriosis (Guo 2018).
Authors' conclusions
Implications for practice.
Only very limited evidence is available for one type of SERM (raloxifene) compared to placebo for management of endometriosis. The primary outcome of pain relief was not reported. Due to the very low quality of evidence, we are uncertain of the effects of raloxifene on adverse events and endometriosis recurrence compared to placebo.
Implications for research.
Future research should describe the extent of endometriosis in participating women according to international classifications. Well‐designed and well‐conducted RCTs with double blinding should be conducted to reduce the placebo effect when novel treatments for pain are evaluated. The included study compared raloxifene with placebo after surgery; this highlights the need for additional types of comparisons, such as raloxifene compared with other medical treatments or surgery, and for inclusion of other types of SERMs. Also, the effects of SERMs without prior surgery should be explored. Lack of data on economic evaluations highlights the need for new research that incorporates these outcomes.
History
Protocol first published: Issue 6, 2014 Review first published: Issue 5, 2021
Acknowledgements
We would like to thank the people who contributed to this protocol. We thank Helen Nagels, Li Wang (teacher in the Chinese‐Evidence Base Center), and the editorial board of the Cochrane Gynaecology and Fertility Group (CGF) for significant assistance in developing this protocol. We thank Marian Showell for designing the search strings for us. We thank Jack Wilkinson, Andrew Watson, James Duffy, and Co‐ordinating Edtor Madelon van Wely for their valuable peer review comments.
Appendices
Appendix 1. Cochrane Gynaecology and Fertility Group Specialised Register search strategy
ProCite platform
Searched 28 May 2020
Keywords CONTAINS "Endometriosis" or "endometriosis‐outcome" or "endometriosis scores" or "Endometriosis‐Symptoms" or "endometriotic cysts" or "endometrioma" or Title CONTAINS "Endometriosis" or "endometriosis‐outcome" or "endometriosis scores" or "Endometriosis‐Symptoms" or "endometriotic cysts" or "endometrioma"
AND
Keywords CONTAINS "SERM" or "selective estrogen receptor modulator" or"*Raloxifene" or "Tamoxifen" or "toremifen" or "toremifene" or "Evista" or Title CONTAINS "SERM" or "selective estrogen receptor modulator" or"*Raloxifene" or "Tamoxifen" or "toremifen" or "toremifene" or "Evista"
(4 records)
Appendix 2. CENTRAL via Cochrane Register of Studies Online (CRSO) search strategy
Web platform
From inception to 28 May 2020
#1 MESH DESCRIPTOR Selective Estrogen Receptor Modulators EXPLODE ALL TREES 2946
#2 MESH DESCRIPTOR Raloxifene Hydrochloride EXPLODE ALL TREES 474
#3 MESH DESCRIPTOR Tamoxifen EXPLODE ALL TREES 2253
#4 MESH DESCRIPTOR Toremifene EXPLODE ALL TREES 95
#5 (selective oestrogen receptor modulator*):TI,AB,KY 36
#6 (selective estrogen receptor modulator*):TI,AB,KY 718
#7 SERM*:TI,AB,KY or SR‐16234*:TI,AB,KY 327
#8 raloxifene:TI,AB,KY 930
#9 tamoxifen:TI,AB,KY 4697
#10 toremifene:TI,AB,KY 187
#11 Evista*:TI,AB,KY 47
#12 Nolvadex:TI,AB,KY 89
#13 Fareston:TI,AB,KY 19
#14 #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 6509
#15 MESH DESCRIPTOR Endometriosis EXPLODE ALL TREES 791
#16 Endometrio*:TI,AB,KY 2545
#17 #15 OR #16 2545
#18 #14 AND #17 26
Appendix 3. MEDLINE search strategy
Ovid platform
From 1946 to 28 May 2020
1 exp selective estrogen receptor modulators/ or exp raloxifene/ or exp tamoxifen/ or exp toremifene/ (27755) 2 selective oestrogen receptor modulator$.tw. (308) 3 selective estrogen receptor modulator$.tw. (2981) 4 SERM$.tw. (2473) 5 raloxifene.tw. (3267) 6 tamoxifen.tw. (22703) 7 toremifene.tw. (671) 8 Evista$.tw. (67) 9 Nolvadex$.tw. (140) 10 Fareston$.tw. (27) 11 SR‐16234*.tw. (4) 12 or/1‐11 (37915) 13 exp Endometriosis/ (21609) 14 Endometrio$.tw. (30052) 15 or/13‐14 (33820) 16 12 and 15 (352) 17 randomized controlled trial.pt. (506268) 18 controlled clinical trial.pt. (93688) 19 randomized.ab. (480632) 20 placebo.tw. (213714) 21 clinical trials as topic.sh. (191315) 22 randomly.ab. (333912) 23 trial.ti. (218930) 24 (crossover or cross‐over or cross over).tw. (84731) 25 or/17‐24 (1320949) 26 exp animals/ not humans.sh. (4701352) 27 25 not 26 (1214297) 28 16 and 27 (34)
Appendix 4. Embase search strategy
Ovid platform
From 1980 to 28 May 2020
1 exp selective estrogen receptor modulator/ (7794) 2 exp raloxifene/ (11231) 3 exp tamoxifen citrate/ or exp tamoxifen/ (62528) 4 exp toremifene/ (2154) 5 selective oestrogen receptor modulator$.tw. (399) 6 selective estrogen receptor modulator$.tw. (4051) 7 SERM$.tw. (3850) 8 raloxifene.tw. (4604) 9 tamoxifen.tw. (33225) 10 toremifene.tw. (804) 11 Evista$.tw. (1022) 12 Nolvadex$.tw. (1389) 13 Fareston$.tw. (173) 14 SR‐16234*.tw. (16) 15 or/1‐14 (78292) 16 exp endometriosis/ (35604) 17 Endometrio$.tw. (43331) 18 or/16‐17 (49461) 19 15 and 18 (745) 20 Clinical Trial/ (963034) 21 Randomized Controlled Trial/ (598954) 22 exp randomization/ (86801) 23 Single Blind Procedure/ (38819) 24 Double Blind Procedure/ (169202) 25 Crossover Procedure/ (62950) 26 Placebo/ (335995) 27 Randomi?ed controlled trial$.tw. (227639) 28 Rct.tw. (36951) 29 random allocation.tw. (1997) 30 randomly allocated.tw. (34904) 31 allocated randomly.tw. (2533) 32 (allocated adj2 random).tw. (812) 33 Single blind$.tw. (24525) 34 Double blind$.tw. (201612) 35 ((treble or triple) adj blind$).tw. (1133) 36 placebo$.tw. (301171) 37 prospective study/ (598878) 38 or/20‐37 (2176633) 39 case study/ (68827) 40 case report.tw. (400435) 41 abstract report/ or letter/ (1092786) 42 or/39‐41 (1551618) 43 38 not 42 (2123502) 44 19 and 43 (182)
Appendix 5. PsycINFO search strategy
Ovid platform
From 1806 to 28 May 2020
1 selective oestrogen receptor modulator$.tw. (21) 2 selective estrogen receptor modulator$.tw. (172) 3 SERM$.tw. (702) 4 raloxifene.tw. (142) 5 tamoxifen.tw. (559) 6 toremifene.tw. (5) 7 Evista$.tw. (1) 8 Nolvadex$.tw. (2) 9 Fareston$.tw. (0) 10 SR‐16234*.tw. (0) 11 or/1‐10 (1346) 12 exp Gynecological Disorders/ (1812) 13 Endometrio$.tw. (291) 14 12 or 13 (2011) 15 11 and 14 (3)
Appendix 6. CINAHL search strategy
EBSCO platform
Searched from 1982 to 28 May 2020
| # | Query | Results |
| S29 | S16 AND S28 | 9 |
| S28 | S17 OR S18 OR S19 OR S20 OR S21 OR S22 OR S23 OR S24 OR S25 OR S26 OR S27 | 1,601,214 |
| S27 | TX allocat* random* | 13,292 |
| S26 | (MH "Quantitative Studies") | 30,564 |
| S25 | (MH "Placebos") | 13,715 |
| S24 | TX placebo* | 71,386 |
| S23 | TX random* allocat* | 13,292 |
| S22 | (MH "Random Assignment") | 68,268 |
| S21 | TX randomi* control* trial* | 221,740 |
| S20 | TX ( (singl* n1 blind*) or (singl* n1 mask*) ) or TX ( (doubl* n1 blind*) or (doubl* n1 mask*) ) or TX ( (tripl* n1 blind*) or (tripl* n1 mask*) ) or TX ( (trebl* n1 blind*) or (trebl* n1 mask*) ) | 1,217,999 |
| S19 | TX clinic* n1 trial* | 295,057 |
| S18 | PT Clinical trial | 110,789 |
| S17 | (MH "Clinical Trials+") | 319,652 |
| S16 | S3 AND S15 | 49 |
| S15 | S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14 | 8,535 |
| S14 | TX Fareston | 2 |
| S13 | TX Evista | 64 |
| S12 | TX toremifene | 73 |
| S11 | TX tamoxifen | 5,448 |
| S10 | TX raloxifene | 1,286 |
| S9 | TX SERM* | 1,472 |
| S8 | TX selective estrogen receptor modulator* | 1,412 |
| S7 | TX selective oestrogen receptor modulator* | 1,412 |
| S6 | (MM "Tamoxifen+") | 2,436 |
| S5 | (MM "Raloxifene") | 407 |
| S4 | (MM "Selective Estrogen Receptor Modulators+") | 3,045 |
| S3 | S1 OR S2 | 7,768 |
| S2 | TX Endometrio* | 7,768 |
| S1 | (MM "Endometriosis") | 3,862 |
Data and analyses
Comparison 1. Raloxifene versus placebo.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1.1 Adverse events | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1.1 Pelvic pain | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.25 [0.63, 2.45] |
| 1.1.2 Ovarian cyst | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.57 [0.55, 4.43] |
| 1.1.3 Headache | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.09 [0.49, 2.43] |
| 1.1.4 Migraine | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.73 [0.28, 1.95] |
| 1.1.5 Depression | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.96 [0.63, 6.06] |
| 1.1.6 Other | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.08 [0.00, 1.30] |
| 1.2 Quality of life ‐ Mental health | 1 | 93 | Mean Difference (IV, Fixed, 95% CI) | 11.10 [0.01, 22.19] |
| 1.3 Recurrence rate of endometriosis ‐ Biopsy proven | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.3.1 Biopsy‐proven endometriosis | 1 | 93 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.20 [0.66, 2.21] |
Characteristics of studies
Characteristics of included studies [ordered by study ID]
Stratton 2008.
| Study characteristics | ||
| Methods | Randomised controlled trial, 93 participants, single‐centre study, in USA | |
| Participants | 93 enrolled: active = 47, placebo = 46 Enrollment: 1999 to 2004 Age, years: 18 to 45. Active 31.1 ± 1.1, placebo 32.0 ± 1.1 Body mass index: active 25.3 ± 0.7, placebo 25.2 ± 0.9 Duration of pelvic pain (range), years: active 10.6 ± 0.9 (0.75 to 32), placebo 10.8 ± 1.3 (1 to 28) Menstrual cycle length, days: active 29.5 ± 1.0, placebo 27.8 ± 0.6 Bone mineral density, g/cm²: active 1.030 ± 0.016, placebo 1.027 ± 0.018 T score: active 0.37 ± 0.15, placebo 0.26 ± 0.16 Diagnosis: biopsy‐proven endometriosis at study laparoscopy, after which randomisation took place Inclusion: patients who had a 3‐month history of pelvic pain and significant postoperative pelvic pain reduction; without additional treatment for endometriosis within 6 months; complete excision of lesions. In 125 of 127 women, surgery was performed in a standardised surgical approach by 1 investigator Identically appearing capsules with lactose + raloxifene or lactose + inert placebo. The Pharmaceutical Development Service (PDS) created allocation sequence, using a table of random numbers and alternating blocks of 8 and 10, which was accessible only to the pharmacy. Treatment assignment was concealed from study staff and participants until the study ended Exclusion: chronic pelvic pain resulting only from infectious, gastrointestinal, musculoskeletal, neurological, or psychiatric causes; significant abnormalities in the physical or laboratory examination, including renal and liver function more than twice the normal range; untreated abnormal cervical cytology or other gynaecological condition; hysterectomy or bilateral salpingo‐oophorectomy; pregnant or lactating; history of venous thrombosis events, stroke, transient ischaemic attacks, manic depressive illness, or untreated major depression, except for use of antidepressants, medication for migraine and headache, and allergy medication |
|
| Interventions | Active group: raloxifene 180 mg per day for 6 months, actually 90 mg of raloxifene, blended with lactose Placebo group: lactose and inert materials |
|
| Outcomes | All outcome measures were observed for up to 18 months post treatment or for up to 2 years after first surgery Outcomes in this RCT Primary outcomes ‐ Recurrence of pain, defined as 2 consecutive months of pelvic pain equal to or more severe than at study entry Secondary outcomes ‐ Assessment of adverse events ‐ Presence and severity of pelvic pain ‐ Quality of life (defined by outcomes in the Duke Health Profile Questionnaire) ‐ Menstrual cycle length ‐ Bone mineral density Outcomes needed for the review Primary outcomes ‐ Relief of pelvic pain, on a visual analogue scale (VAS) (0 (no pain) to 10 (worst pain imaginable)), or subjective improvement ‐ Adverse events resulting from treatment with SERMs (hot flushes, endometrial cancer, vaginal bleeding, thrombosis) either during or following treatment. Significant earlier recurrence of pain, which was the reason to stop this study prematurely in December 2004
Secondary outcomes ‐ Quality of life: in this RCT, measured by the Duke Health Profile Questionnaire. No data were provided ‐ just a general statement that QoL scores were similar at baseline and during the study
‐ Recurrence rate of endometriosis (symptoms or biopsy‐proven): only general statement that the raloxifene group had earlier return to pain than the placebo group (P = 0.01)
‐ Fertility outcomes (only in women wishing to conceive): not reported in this study, as patients were asked not to become pregnant during the study
‐ Economic outcomes (costs, time taken for treatment): not reported in this study |
|
| Notes | Clinical trial registration https://clinicaltrials.gov/ct2/show/study/NCT00001848 |
|
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Computer‐generated table of random numbers in alternating blocks of 8 and 10 |
| Allocation concealment (selection bias) | Low risk | The Pharmaceutical Development Service created the allocation sequence. Treatment assignment was concealed from study staff and participants |
| Blinding of participants and personnel (performance bias) All outcomes | Low risk | Double blinding of study staff and participants to treatment assignment |
| Blinding of outcome assessment (detection bias) All outcomes | Low risk | Double blinding of study staff and participants to treatment assignment |
| Incomplete outcome data (attrition bias) All outcomes | High risk | ITT for all randomised women included in primary and safety analyses Second‐look laparoscopy was performed in only 40/93 women: 23/47 in the raloxifene group and 18/46 in the placebo group (1 had repeat laparoscopy elsewhere) Recurrence of endometriosis confirmed by second‐look laparoscopy: 16/23 in the raloxifene group and 13/17 in the placebo group Recurrence of pelvic pain as an adverse event: 14/47 in the raloxifene group and 11/45 in the placebo group It is unclear whether there was overlap between patients reporting recurrence of pelvic pain and laparoscopy finding of new lesions Lost to follow‐up: 9 in the raloxifene group and 11 in the placebo group Withdrew because of (unintended) pregnancy: 3 in the raloxifene group and 5 in the placebo group Incomplete outcome because still ongoing at the time of publication: 8 in the raloxifene group and 7 in the placebo group |
| Selective reporting (reporting bias) | High risk | Protocol did not mention any pre‐specified primary nor secondary outcomes Outcomes reported incompletely: only in a general statement on QoL; no detailed data provided |
| Other bias | High risk |
Deviations from study protocol Among all women without adverse effects, 8 (8.6%) delayed start of medication, stopped medication, or reduced recommended dosage. Of all women with adverse events, 16 (35%) in the raloxifene group and 20 (43%) in the placebo group reduced dosage or stopped study drug
This is a highly selected population, as most women had undergone 1 or more laparoscopies
|
ITT: intention‐to‐treat principle; QoL: quality of life; RCT: randomised controlled trial; SERM: selective oestrogen receptor modulator.
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
|---|---|
| Barra 2019 | Prospective study, single arm, not an RCT |
| Harada 2018 | Single arm, not an RCT |
| Li 1999 | No evidence that this Chinese herbal medicine ‐ Yiqi Huoxue Huayu Tongfu Principle ‐ is a specific SERM drug |
| Parker 2006 | This is a prospective clinical trial, not an RCT |
RCT: randomised controlled trial; SERM: selective oestrogen receptor modulator.
Differences between protocol and review
OR had changed to RR between protocol and review. The first author's affiliation address has been changed. We added an affiliation and changed SU to Sichuan University.
We added a primary outcome post hoc: pain measures as reported in the primary study. The rationale was that our included study reported measures of pain that we had not considered a priori, and that this information would be useful for decision‐makers in the absence of specific data on pain relief.
We broadened inclusion to studies of women who had clinically diagnosed endometriosis but without visual or laparoscopic confirmation.
We had used a fixed model to assess heterogeneity.
For the 2021 update, we added 'other' adverse events (e.g. headache, migraine, ovarian cyst).
Contributions of authors
Maaike van Hoesel: selection of trials for the updated 2020 search; drafting of the review.
Selma Mourad: selection of trials for the updated 2020 search; drafting of the review; assessment of risk of bias.
Yali Chen: selection of trials for inclusion or exclusion; extraction of data; drafting of protocol and review; search for trials; data entry into RevMan.
Qi Wan: selection of trials for inclusion or exclusion; obtaining copies of trial reports; extraction of data; assessment of risks of bias.
Ai Zheng: all correspondence; drafting of protocol and review; interpretation of results.
Sources of support
Internal sources
No sources of support provided
External sources
No sources of support provided
Declarations of interest
MvH: no commercial interests to disclose.
SM: no commercial interests to disclose.
YC: none known.
QW: none known.
AZ: none known.
New
References
References to studies included in this review
Stratton 2008 {published data only}
- Stratton E, Sinaii N, Segars J, Koziol D, Wesley R, Zimmer C, et al. Raloxifene accelerates the return of chronic pelvic pain from endometriosis after surgical treatment. European Journal of Obstetrics & Gynecology and Reproductive Biology 2005;123(Suppl 1):S22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stratton P, Sinaii N, Segars J, Koziol D, Wesley R, Zimmer C, et al. Return of chronic pelvic pain from endometriosis after raloxifene treatment: a randomized controlled trial. Obstetrics & Gynecology 2008;111(1):88-96. [DOI] [PMC free article] [PubMed] [Google Scholar]
References to studies excluded from this review
Barra 2019 {published data only}
- Barra F, Evangelisti G, Tantari M, Ferrero S. Effect of ospemifene on patients with symptomatic genitourinary syndrome of menopause affected by deep endometriosis. BJOG: an International Journal of Obstetrics and Gynaecology. 2019;Conference: 2019 World Congress of the Royal College of Obstretricians and Gynaecologists, RCOG 2019. United Kingdom(126 (Suppl 2)):188.
Harada 2018 {published data only}
- Harada T, Ohta I, Endo Y, Sunada H, Noma H, Taniguchi F. SR-16234, a novel selective estrogen receptor modulator for pain symptoms with endometriosis: an open-label clinical trial. Yonago Acta Medica 2018; (60):227-33. [PMID: doi: 10.24563/yam.2017.12.003. eCollection 2017 Dec.] [DOI] [PMC free article] [PubMed] [Google Scholar]
Li 1999 {published data only}
- Li J, Zheng J, Wang D. Clinical observation on treatment of endometriosis by tonifying qi and promoting blood circulation to remove stasis and purgation principle. Zhong guo Zhong Xi Yi Jie He Za Zhi 1999;19(9):533-5. [PubMed] [Google Scholar]
Parker 2006 {published data only}
- Parker JD, Leondires M, Sinaii N, Premkumar A, Nieman LK, Stratton P. Persistence of dysmenorrhea and nonmenstrual pain after optimal endometriosis surgery may indicate adenomyosis. Fertility and Sterility 2006;86(3):711-5. [DOI] [PubMed] [Google Scholar]
Additional references
Atkins 2004
- Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al, GRADE Working Group. Grading quality of evidence and strength of recommendations. BMJ 2004;328(7454):1490. [DOI] [PMC free article] [PubMed] [Google Scholar]
Becker 2017
- Becker CM, Gattrell WT, Gude K, Singh SS. Reevaluating response and failure of medical treatment of endometriosis: a systematic review. Fertility and Sterility 2017;108(1):125-36. [DOI: 10.1016/j.fertnstert.2017.05.004] [DOI] [PMC free article] [PubMed] [Google Scholar]
Bianconi 2007
- Bianconi L, Hummelshoj L, Coccia ME, Vigano P, Vittori G, Veit J, et al. Recognizing endometriosis as a social disease: the European Union-encouraged Italian Senate approach. Fertility and Sterility 2007;88(5):1285-7. [DOI] [PubMed] [Google Scholar]
Brosens 1993
- Brosens I, Donnez J, Benagiano G. Improving the classification of endometriosis. Human Reproduction 1993;8:1792-5. [DOI] [PubMed] [Google Scholar]
Brown 2010
- Brown J, Pan A, Hart RJ. Gonadotrophin-releasing hormone analogues for pain associated with endometriosis. Cochrane Database of Systematic Reviews 2010, Issue 12. Art. No: CD008475. [DOI: 10.1002/14651858.CD008475.pub2] [DOI] [PMC free article] [PubMed] [Google Scholar]
Brown 2012
- Brown J, Kives S, Akhtar MP. Progestogens and anti-progestogens for pain associated with endometriosis. Cochrane Database of Systematic Reviews 2012, Issue 3. Art. No: CD002122. [DOI: 10.1002/14651858.CD002122.pub2] [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen 2020
- Chen I, Veth VB, Choudhry AJ, Murji A, Zakhari A, Black AY, et al. Pre‐ and postsurgical medical therapy for endometriosis surgery. Cochrane Database of Systematic Reviews 2020, Issue 11. Art. No: CD003678. [DOI: 10.1002/14651858.CD003678.pub3] [DOI] [PMC free article] [PubMed] [Google Scholar]
Delmas 1997
- Delmas PD, Bjarnason NH, Mitlak BH, Ravoux AC, Shah AS, Huster WJ, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. New England Journal of Medicine 1997;337:1641-7. [DOI] [PubMed] [Google Scholar]
Dunselman 2014
- Dunselman GAJ, Vermeulen N, Becker C, Calhaz-Jorge C, D’Hooghe T, et al. ESHRE guideline: management of women with endometriosis. Human Reproduction 2014;29(3):400-12. [DOI] [PubMed] [Google Scholar]
Egger 1997
- Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
Eng ‐Wong 2004
- Eng-Wong J, Zujewski JA. Raloxifene and its role in breast cancer prevention. Expert Review of Anticancer Therapy 2004;4:523–32. [DOI] [PubMed] [Google Scholar]
ESHRE Guideline 2014
- Dunselman GAJ, Vermeulen N, Becker C, Calhaz-Jorge C, D'Hooghe T, De Bie B, et al. ESHRE guideline: management of women with endometriosis. Human Reproduction 2014;29(3):400-12. [DOI: 10.1093/humrep/det457] [DOI] [PubMed] [Google Scholar]
Eskenazi and Warner 1997
- Eskenazi B, Warner ML. Epidemiology of endometriosis. Obstetrics and Gynecology Clinics of North America 1997;24(2):235-58. [DOI: 10.1016/s0889-8545(05)70302-8] [DOI] [PubMed] [Google Scholar]
Francucci 2005
- Francucci CM, Romagni P, Boscaro M. Raloxifene: bone and cardiovascular effects. Journal of Endocrinological Investigation 2005;28(10 Suppl):85-90. [PubMed] [Google Scholar]
GRADEpro GDT 2014 [Computer program]
- GRADE Working Group, McMaster University GRADEpro GDT. Version accessed before November 2016. Hamilton (ON): GRADE Working Group, McMaster University, 2014.
Guo 2018
- Guo SW, Groothuis PG. Is it time for a paradigm shift in drug research and development in endometriosis/adenomyosis? Human Reproduction Update 2018;24(5):577-98. [DOI: 10.1093/humupd/dmy020] [DOI] [PubMed] [Google Scholar]
Higgins 2011
- Higgins JP, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from handbook.cochrane.org.
Jones 2002
- Jones GL, Kennedy SH, Jenkinson C. Health-related quality of life measurement in women with common benign gynecologic conditions: a systematic review. American Journal of Obstetrics and Gynecology 2002;187:501–11. [DOI] [PubMed] [Google Scholar]
Kondo 2011
- Kondo W, Bourdel N, Tamburro S, Cavoli D, Jardon K, Rabischong B, et al. Complications after surgery for deeply infiltrating pelvic endometriosis. BJOG 2011;118(3):292-8. [DOI: 10.1111/j.1471-0528.2010.02774.x] [DOI] [PubMed] [Google Scholar]
Marques 2004
- Marques A, Bahamondes L, Aldrighi JM, Petta CA. Quality of life in Brazilian women with endometriosis assessed through a medical outcome questionnaire. Journal of Reproductive Medicine 2004;49:115–20. [PubMed] [Google Scholar]
Maximov 2013
- Maximov PY, Lee TM, Jordan VC. The discovery and development of selective estrogen receptor modulators (SERMs) for clinical practice. Current Clinical Pharmacology 2013;8(2):135-55. [DOI: 10.2174/1574884711308020006.] [DOI] [PMC free article] [PubMed] [Google Scholar]
Meuleman 2009
- Meuleman C, Vandenabeele B, Fieuws S, Spiessens C, Timmerman D, D'Hooghe T. High prevalence of endometriosis in infertile women with normal ovulation and normospermic partners. Fertility and Sterility 2009;92(1):68-74. [DOI: 10.1016/j.fertnstert.2008.04.056] [DOI] [PubMed] [Google Scholar]
Moher 2009
- Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;339:2535. [PMC free article] [PubMed] [Google Scholar]
Parkerson 1990
- Parkerson GR Jr, Broadhead WE, Tse CK. The Duke Health Profile: a 17-item measure of health and dysfunction. Medical Care 1990;28(11):1056-72. [PMID: doi: 10.1097/00005650-199011000-00007.The Duke Health Profile.] [DOI] [PubMed] [Google Scholar]
Rafique 2017
- Rafique S, Decherney AH. Medical management of endometriosis. Clinical Obstetrics and Gynecology 2017;60(3):485-96. [DOI: 10.1097/GRF.0000000000000292] [DOI] [PMC free article] [PubMed] [Google Scholar]
RCOG 2006
- Royal College of Obstetricians and Gynaecologists (RCOG). The investigation and management of endometriosis. London (UK): Royal College of Obstetricians and Gynaecologists (RCOG) 2006;24:1-14. [Google Scholar]
Review Manager 2014 [Computer program]
- Nordic Cochrane Centre, The Cochrane Collaboration Review Manager 5 (RevMan5). Version 5.3. Copenhagen: Nordic Cochrane Centre, The Cochrane Collaboration, 2014.
Scott 1990
- Scott JA, da Camara CC, Early JE. Raloxifene: a selective estrogen receptor modulator. American Family Physician 1990;60:1131-9. [PubMed] [Google Scholar]
Silfen 1999
- Silfen SL, Ciaccia AV, Bryant HU. Selective estrogen receptor modulators: tissue selectivity and differential uterine effects. Climacteric 1999;2:268–83. [DOI] [PubMed] [Google Scholar]
Simsa 2007
- Simsa P, Mihalyi A, Kyama CM, Mwenda JM, Fülöp V, D'Hooghe TM. Selective estrogen-receptor modulators and aromatase inhibitors: promising new medical therapies for endometriosis? Women's Health 2007;3(5):617-28. [DOI] [PubMed] [Google Scholar]
Thiebaud 2001
- Thiebaud D, Secrest RJ. Selective estrogen receptor modulators: mechanism of action and clinical experience. Focus on raloxifene. Reproduction Fertility and Development 2001;13:331–6. [DOI] [PubMed] [Google Scholar]
Vercellini 1993
- Vercellini P, Trespidi L, Colombo A, Vendola N, Marchini M, Crosignani PG. A gonadotropin-releasing hormone agonist versus a low-dose oral contraceptive for pelvic pain associated with endometriosis. Fertility and Sterility 1993;60(1):75-9. [PMID: ] [PubMed] [Google Scholar]
Vitale 2016
- Vitale SG, La Rosa VL, Rapisarda AMC, Laganà AS. Impact of endometriosis on quality of life and psychological well-being. Journal of Psychosomatic Obstetrics & Gynecology 2017;38(4):317-9. [DOI: 10.1080/0167482X.2016.1244185] [DOI] [PubMed] [Google Scholar]