Hormone replacement therapy for perimenopausal and postmenopausal women
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
Our objectives are to compare the effect of hormone replacement therapy versus placebo in peri‐ or postmenopausal women on mortality, heart disease, venous thromboembolism, stroke, breast cancer, colorectal cancer and reduction of fractures.
HRT will be defined as oestrogen therapy alone, or oestrogen therapy with combined, cyclic or continuous progestogen therapy, delivered by oral, transdermal, subcutaneous or intranasal routes.
Background
Menopause means the time when menstruation ceases and typically occurs between 45‐55 years of age. Women are said to be post menopausal when menstruation has ceased for 12 months. The menopause may result in a variety of symptoms that affect the well being and health of women; hot flushes, insomnia, increased risk of osteoporosis, increased risk of heart disease and urogenital atrophy are all reported. As the population continues to grow older, there has been an increased focus on the effects of aging. Hormone replacement therapy (HRT) has been utilised for the treatment of menopausal symptoms, particularly hot flushes and vaginal dryness.
HRT includes either unopposed oestrogen or combined therapy with oestrogen and progestogen administered either continuously or sequentially. These different types of HRT have differential effects. Different doses, modes of administration and duration of the same therapy can also cause variable effects in women. However, extrapolations from specific data to form general conclusions on effects may be more helpful for clinical practice.
Menopause symptoms are not life‐threatening, but may be associated with intense discomfort and may substantially influence the postmenopausal woman's physical, emotional and social functioning. Although they are most pronounced during the first years after the menopause, up to 40% of symptomatic women may still have significant symptoms, such as hot flushes more than ten years after the menopause (Nordin 1987).
Observational data from studies comparing coronary heart disease (CHD) outcomes in women taking and not taking long term HRT appear to provide consistent and convincing evidence of a protective effect for coronary heart disease and stroke. In a recent systematic review, a total of 43 observational studies of primary prevention of CHD or cardiovascular disease were reviewed (Nelson 2002). Pooling of those studies of good and fair quality (which comprised only one randomised controlled trial) showed a strong protective effect on CHD incidence (RR 0.80, 95% CI 0.68 to 0.95) and mortality (RR 0.62, 95% CI 0.40 to 0.90). It is upon these findings that the notion that HRT is "cardioprotective" is based. However, in the studies that controlled for socio‐economic confounding the relative risk attenuated towards unity (0.97 9%% CI 0.82 to 1.16) (Nelson 2002). These findings have been confirmed in another systematic review (Humphrey 2002). Adequate control of all known important confounders, which include, in addition to social class, other factors such as smoking, physical inactivity and poor diet might indicate more precisely the risk benefit ratio of HRT in the observational studies.
More recently findings from three randomised comparisons of the effects of HRT among women with pre‐existing cardiovascular disease (CVD) (heart & estrogen/progestagen replacement study ‐ HERS 1998; women's estrogen for stroke trial ‐ WEST 2001) and without (Women's Health Initiative study ‐ WHI 2002) have failed to demonstrate the marked benefits for CHD seen in observational studies. These publications have raised concerns about the risks and benefits associated with the use of HRT, although as yet the message for both clinicians and women remains unclear (Dixon 2002; Fletcher 2002; Stevenson 2002). A meta‐analysis pooled these three studies although the interventions included unopposed oestrogen and combined oestrogen/progestogen therapy (Beral 2002). No significant excess or reduction in the relative risk of CHD was reported, RR 1.11 (95% CI 0.96 to 1.30) (Beral 2002). However, the findings certainly exclude an effect as large as that seen in the earlier observational studies. A fourth trial, (EVTET 2000) examined the effects of HRT on pulmonary embolism and reported an excess of these events.
A much earlier and systematic review of the available randomised evidence reported an odds ratio for cardiovascular disease of 1.39 (95% CI 0.48 to 3.95) among those taking HRT but was based on only 12 events in those allocated HRT and 5 in the control groups. (Hemminki 1997). However, the evidence in this review was insufficient to exclude substantial benefits from HRT. Excess risks of stroke (RR 1.27, 95% CI 1.06 to 1.51) and pulmonary embolus (RR 2.16, 95% CI 1.47 to 3.18) were also demonstrated in Beral's review (Beral 2002).
Cancer and other risks from HRT were also considered in Beral's review (Beral 2002). An increased breast cancer risk was reported (RR 1.24, 95% CI 1.03 to 1.56), but a reduced colorectal cancer risk (RR 0.64, 95% CI 0.45 to 0.92). The findings for endometrial cancer risk were consistent with either benefit or harm (RR 0.76, 95% CI 0.45 to 1.31). A reduced risk of fractured neck of femur was found (RR 0.72, 95% CI 0.52 to 0.98). New data are awaited which may permit more precise estimates of the effects of HRT on these different outcomes to be made.
Hormone replacement therapy has been used clinically for over 60 years for the treatment of symptoms around the menopause and in the last 20 years for the prevention of long‐term conditions such as heart disease and osteoporosis. The effectiveness of any therapy must be balanced against possible harm. Side‐effects associated with HRT therapy include atypical bleeding, breast cancer, venous thrombo embolism and stroke.
HRT, while beneficial for the control of specific menopausal symptoms and prevention of osteoporosis, may affect cardiovascular disease risk and some cancers, thus making the risk‐benefit ratio uncertain. A systematic review examining all relevant outcomes is needed to enable women and their physicians to make informed choices about use of HRT.
Objectives
Our objectives are to compare the effect of hormone replacement therapy versus placebo in peri‐ or postmenopausal women on mortality, heart disease, venous thromboembolism, stroke, breast cancer, colorectal cancer and reduction of fractures.
HRT will be defined as oestrogen therapy alone, or oestrogen therapy with combined, cyclic or continuous progestogen therapy, delivered by oral, transdermal, subcutaneous or intranasal routes.
Methods
Criteria for considering studies for this review
Types of studies
Only randomised, double‐blind trials that include at least one HRT therapy group and one placebo group that report at least one of the above outcomes will be considered for inclusion in this review.
Types of participants
Suitable participants will be peri‐ or postmenopausal women recruited from any health care setting or a population based sample that may have had either spontaneous menopause or bilateral oophorectomy (removal of both ovaries).
Peri‐menopausal women will be defined as women with spontaneous menopause who have menstruated irregularly within the last 12 months.
Post‐menopausal women will be defined as women with surgical menopause or women with spontaneous menopause and amenorrhea for more than 12 months. Women with and without prior history of cardiovascular disease and fracture or a diagnosis of osteoporosis will be included.
Types of interventions
All oestrogens, with and without progestogens, administered by oral, transdermal, subcutaneous or intranasal routes and given as peri‐or postmenopausal therapy for any reason for twelve months or more.
Exclusion Criteria:
Co‐interventions that may potentially affect the outcomes being measured.
Topical oestrogen creams, tablets or rings are excluded as this intervention is covered in another Cochrane review (Lethaby 1999a).
Types of outcome measures
Mortality
·Cause specific mortality
·Total mortality
Cardiovascular events
Venous thromboembolism
Stroke
Invasive breast cancer
Colorectal cancer
Ovarian cancer
Dementia (including Alzheimer's disease)
Incidence of fractures, both vertebral and non vertebral (clinically diagnosed)
Adherence to treatment
Vasomotor symptoms (hot flushes), general menopausal symptoms and early onset side‐effects will not be included as these are the subject of other reviews (Lester 2001; MacLennan 2001). Endometrial hyperplasia and carcinoma will also be excluded as outcomes as these are the subject of another review (Lethaby 1999b). Bone mineral density (BMD) is also covered by other reviews so will not be included (Prentice 2003; Tugwell 2003; Wells 2002). Finally dementia and cognitive function will not be included as outcomes as they are also covered by other reviews ( Hogervorst 2002a; Hogervorst 2002b).
Search methods for identification of studies
See: Collaborative Review Group Search Strategy
Publications that describe randomised double‐blind trials for HRT therapy, with a placebo group and a minimum of 12 months therapy duration will be obtained using the strategy developed by the Menstrual Disorders and Subfertility Group. Electronic searches will be performed of the Trials Register of the Cochrane Menstrual Disorders and Subfertility Group, the Cochrane Controlled Trials Register, MEDLINE, EMBASE, Current Contents, Biological Abstracts and hand searches and hand searching of relevant journals and conference abstracts will be performed.
The National Research Register (NRR), a register of ongoing and recently completed research projects funded by, or of interest to, the United Kingdom's National Health Service, as well as entries from the Medical Research Council's Clinical Trials Register, and details on reviews in progress collected by the NHS Centre for Reviews and Dissemination, will also be searched.
All pharmaceutical companies manufacturing HRT products will be contacted to request data from all randomised controlled trials (RCT's) in their files, including unpublished trials.
Data collection and analysis
Selection of trials:
One reviewer will screen the abstracts of all publications that are obtained by the search strategy for eligible trials. Where the screened abstract suggests the trial is potentially eligible for inclusion, the full article will be obtained. Each study will be assessed against the inclusion criteria by one of the reviewers. This assessment will be performed unblinded. Where there is uncertainty regarding eligibility, a second reviewer will also assess the study and a decision will be reached through discussion. Where necessary, additional information will be sought from the corresponding author of the study.
Assessment of Methodological Quality
The quality of all studies that are deemed eligible for the review will be assessed independently by two reviewers. Any discrepancies will be resolved by discussion.
The standard checklist created by the Menstrual Disorders and Subfertility Group (MDSG) will be used.
Section i: Internal Validity
1)Was the assigned treatment adequately concealed prior to allocation?
2)Were the outcomes of participants who withdrew or were excluded after allocation described and included in an "intention to treat" analysis?
3)Were the outcome assessors blind to assignment status?
4)Were the treatment and control group comparable at entry?
5)Were the participants blind to assignment status following allocation?
6)Were the treatment providers blind to assignment status?
7)Were the care programmes, other than the trial options, identical?
8)Were the withdrawals <10% of the study population?
9) Did any of the control group receive the intervention assigned to the treatment group?
10) Did any of the treatment group receive the intervention assigned to the control group?
Section ii: External Validity
9)Were the inclusion and exclusion criteria for entry clearly defined?
10)Were the outcome measures used clearly defined?
11)Were the accuracy, precision, and observer variation of the outcome measures adequate?
12)Was the timing of the outcome measures appropriate?
It is intended to use this information in investigation of any heterogeneity and in sensitivity analyses. Other aspects of study quality, including the extent of blinding (if appropriate), whether groups were comparable at baseline, the extent of losses to follow‐up, participation levels, whether the outcome assessment standardised, and whether an "intention to treat" analysis was undertaken, will also be assessed. This information will be presented in a table describing the included studies and will provide a context for discussing the reliability of the results.
Data Extraction
Data will be extracted independently by two reviewers, using a proforma, and checked for agreement. Any disagreements will be resolved by discussion. The following quality criteria, methodological details and data will be extracted:
Trial Characteristics
1.Method of allocation concealment
2.Method of randomisation.
3.Adequacy of double blinding (i.e. an explicit statement that therapies could not be distinguished by appearance and/or administration route)
4.Number of participants randomised, excluded, or lost to follow‐up. (not including those who withdrew from the trial but were followed up)
5.Whether an "intention to treat" analysis was done and what type of ITT analysis.
6.The use of a power calculation to estimate sample size
7.Duration, timing and location of the study.
8.Study design (e.g. parallel or crossover, single centre or multi‐centre)
9.Sources of any funding.
Characteristics of the Study Participants
1.Age and any other recorded characteristics of women in the study
2.Baseline equality of treatment groups with respect to symptom and quality of life score outcomes, age and assessment of menopausal status
3.Menopausal status (i.e. peri‐ or postmenopausal and how status was defined, surgical or natural menopause) of the women in the study
4.Inclusion/exclusion criteria
Interventions Used
1.Type, dosage and administration route of oestrogens and progestogens
2.Duration of therapy, minimum of six months
Outcomes
1.How measured and defined
For cross‐over trials, only results from the end of the first phase (before the treatment cross‐over) will be used because of the potential carry‐over effect of HRT therapy from the first treatment phase.
Analysis
Statistical analyses will be undertaken following the guidelines of the Handbook of the Cochrane Collaboration. Analysis of treatment effects will compare outcomes for each therapy group measured at the end of therapy. Measures of the change from baseline will not be analysed (except for evaluation of the placebo effect) as this measure may overcorrect for chance baseline imbalances between treatment groups (Frison 1997). Random effects estimates will be employed, although heterogeneity from chi‐square tests will also be reported.
For dichotomous data, two by two tables will be generated for each trial and expressed as an odds ratio (OR) with 95% confidence intervals (CI). This data will be combined for meta‐analysis with RevMan software ‐using the Peto‐modified Mantel‐Haenszel method and a fixed and random effects model.
Quality of life scores, although measured as ordinal variables, are considered to be drawn from an underlying continuous distribution, and will be analysed as continuous outcomes. These will be adjusted to a common numerical scale for combined data analysis by using standardised mean differences. However, weighted and unweighted single symptom scores will not be combined and symptom cluster scores will be treated as separate outcomes. Attempts will be made to obtain a recalculated score (unweighted, with the vasomotor component excluded) from the corresponding author of each trial, which will be analysed as a separate outcome.
The analysis will compare HRT therapy with placebo, and subgroup analysis will be performed according to the route of HRT administration. Sensitivity analysis of results to influential individual trials will be performed by the method of Baujat 1999. Sensitivity analysis to quality assessment criteria and heterogeneity analysis will be performed. Quality assessment criteria will be adequate allocation concealment, intention‐to‐treat analysis, and losses less than 10%. A priori, sensitivity analysis will include the outcome level in the control group, duration of therapy and oestrogen alone compared to oestrogen combined with progestogen therapy.
