The risks and benefits of taking hormone replacement therapy (HRT) for more than five years are uncertain. There are no large long-term randomised placebo-controlled trials to guide the duration of therapy. Large trials are expensive, difficult to run and can only examine a limited number of regimens and routes of HRT.
Until the results of large trials (see box) are reported we can only guess at the potential primary effect of HRT on major postmenopausal morbidity and mortality from the much weaker designs of observational studies (cohort and case control). These trials are not randomised and are open to bias and confounding. When the results of observational trials are expressed as relative risk, there is generally less chance of a result being due to bias or confounding if the relative risk is more than halved (<0.5) or more than doubled (>2.0). The increased relative risk of lung cancer in smokers is 43.0 and thus the association is very strong. Hot flushes are reduced by HRT to a relative risk of 0.2. Both results suggest the findings are likely to be causally related. A relative risk of 1.0 suggests no effect either way.
A reanalysis of the data from 51 observational studies showed a relative risk of 1.023 (95% CI 1.01-1.04) for each year of HRT use. The relative risk for HRT use of five years or longer (average 11 years) was 1.35 (95% CI 1.21-1.49).1 These small increases in relative risk (i.e. under 2.0) could be due to detection bias, for example the HRT users could have had more breast examinations and mammograms. HRT users also have more independent risk factors, such as higher social class, Western diet, fewer pregnancies, later first pregnancy, and a higher alcohol intake, which could also account for small changes in the relative risk of breast cancer. However, if the increased risk is real it would equate to an extra two detected breast cancers per 1000 women who use HRT for five years. Paradoxically, most observational studies show a significant reduction in deaths from breast cancer in HRT users. Selection and detection bias can confound all these results.
There is no good evidence that HRT users with a family history of breast cancer further increase their risk compared to non-users with a similar history. Women taking HRT do not need to have mammography more often than other women.
Recent observational studies on the role of added cyclical and continuous progestogens given with oestrogen therapy have not clarified if these regimens have any effect on breast cancer rates.2,3 The relative risks were again small in both studies with overlap of the confidence intervals for oestrogen alone versus oestrogen/progestogen regimens. Thus, no recommendations can yet be made as to whether added progestogens influence breast cancer risk.
The most recent meta-analysis of 23 observational trials suggests that in postmenopausal women who have ever taken HRT the relative risk is 0.80 (95% CI 0.72-0.92).4 This is a 20% reduction in colorectal cancer, however this result is open to the biases of non-randomisation. The effect needs to be confirmed in long-term randomised placebo-controlled trials.
The relative risk that taking unopposed oestrogen for 10 years causes endometrial cancer is 9.5 (95% CI 7.40-12.30) and so this is likely to be a true effect. Additional progestogen greatly reduces this risk, but the degree of this protection is not accurately known. Currently the recommended regimen for women with a uterus is to take oestrogen and cyclical progestogens before the menopause and for 1-2 years after menopause. Continuous combination therapy can be introduced after four years of cyclical therapy or when the woman is definitely postmenopausal. This reduces the amount of initial bleeding that is normally seen for several months when commencing a combined continuous HRT regimen.
Current observational studies suggest that although this is a relatively rare potential complication, the absolute risk rises from 1 in 10 000 to 3 in 10 000 in HRT users (relative risks in four studies ranged from 2.1- 6.9). A past history of thromboembolism before the menopause is not an absolute contraindication to postmenopausal HRT, but might prompt the prescriber to consider testing for thrombophilia.5
Observational and animal studies suggest a potential benefit for HRT. A recent meta-analysis of these epidemiological studies reports a relative risk of 0.70 (95% CI 0.65-0.75) for oestrogen therapy alone and 0.66 (0.53-0.84) for combined HRT.6 However, many researchers argue that the studies have the potential bias of a 'healthy user' effect. A three-year randomised placebo-controlled trial (PEPI)7 has suggested a potential benefit in the primary prevention of ischaemic heart disease. Secondary prevention studies do not suggest that HRT can reverse the early risk of established ischaemic heart disease.8 Currently HRT may be offered to women with risk factors as a potential (but not established) primary cardio protective agent to complement other established drug therapies and lifestyle changes.
HRT was not consistently associated with a change in the relative risk of stroke in observational studies.
Short-term randomised controlled trials consistently show that HRT improves low bone density, and when used prophylactically it inhibits loss of bone after the menopause. However, long-term randomised trials are still needed to show that improved bone density results in a major reduction in osteoporotic fractures, particularly at the hip. Improvements in surrogate end-points suggest that a reduced risk of fractures will be one of the main benefits of taking oestrogen for many years.
All therapies for osteoporosis require long-term compliance to achieve their effect. In South Australia in 1997 the median length of use in all women on HRT was five years with 70% continuance rate at five years. In women with a diagnosis of osteoporosis the median length of use was six years.9
A meta-analysis of 10 observational studies showed a reduction of this dementia in HRT users. The relative risk was 0.71 (CI 0.52-0.98).10 Although there are plausible neuro protective mechanisms for HRT, long-term randomised placebo-controlled trials are awaited to see if HRT has a primary preventative role in this disease which is becoming more common with increasing longevity. A recent secondary prevention trial does not suggest that HRT can reverse established disease.11
Other risks and benefits
Potential long-term risks still need to be defined by long-term randomised controlled trials. They include increased risks of gall bladder and uterine surgery.
Other potential benefits may include a reduction in tooth loss, dry eyes, dry skin, arthritic symptoms, urge incontinence, frequency, nocturia, urinary tract infections, dry vagina, dyspareunia, memory loss and possibly some types of depression. All of these need to be assessed in large trials, but if a woman experiences sustained symptom relief from HRT then long-term therapy may be appropriate to maintain her quality of life.
Current options for length of HRT use
With all the caveats about the weaknesses of observational data, these data are all we can use when advising a woman about the potential risks and benefits of long-term HRT. Until the results of the Women's Health Initiative and WISDOM are available it is not possible to make general recommendations for the duration of treatment. Probably, for menopausal symptom control, up to five years therapy is appropriate, with the option of another five years if still symptomatic when weaned off HRT. Longer therapy would be necessary for other potential indications such as the prevention of cardiovascular disease, dementia, some urological problems and osteoporosis. Phytoestrogens have yet to be shown in published rigorous scientific trials to be of greater benefit for menopausal symptoms than a placebo or to prevent osteoporotic fractures and cardiovascular events.
A woman's informed choice for long-term HRT should be based on unbiased information, explanation of the current data and its limitations and an understanding of her individual needs, risks and preferences. It should not be based on myth, selected information, vested interests in HRT or other products for the menopause and especially not on the lack of skill or knowledge of the adviser.
- Beral V, Bull D, Doll R, Key T, Peto R, Reeves G, et al. Breast cancer and hormone replacement therapy: Collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet 1997;350:1047-59.
- Schairer C, Lubin J, Troisi R, Sturgeon S, Brinton L, Hoover R. Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA 2000;283:485-91.
- Ross RK, Paganini-Hill A, Wan PC, Pike MC. Effect of hormone replacement therapy on breast cancer risk: Estrogen versus estrogen plus progestin. J Natl Cancer Inst 2000;92:328-32.
- Grodstein F, Newcomb PA, Stampfer MJ. Postmenopausal hormone therapy and the risk of colorectal cancer: A review and meta-analysis. Am J Med 1999;106:574-82.
- Greer IA, Walker ID. Hormone replacement therapy and venous thromboembolism. Climacteric 1999;2:224-31.
- Barrett-Connor E, Grady D. Hormone replacement therapy, heart disease, and other considerations. Annu Rev Public Health 1998;19:55-72.
- Writing Group for the PEPI Trial. Effects of estrogen orestrogen/progestin regimens on heart disease risk factors in postmenopausal women [published erratum appears in JAMA 1995;274:1676]. JAMA 1995;273:199-208.
- Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605-13.
- MacLennan AH, Wilson DH, Taylor AW. Hormone replacement therapies in women at risk of cardiovascular disease and osteoporosis in South Australia in 1997. Med J Aust 1999;170:524-7.
- Yaffe K, Sawaya G, Lieberburg I, Grady D. Estrogen therapy in postmenopausal women: effects on cognitive function and dementia. JAMA 1998;279:688-95.
- Mulnard RA, Cotman CW, Kawas C, van Dyck CH, Sano M, Doody R, et al. Estrogen replacement therapy for treatment of mild to moderate Alzheimer disease: a randomized controlled trial. JAMA 2000;283:1007-15.