The majority of women in the UK use HRT for the relief of menopausal symptoms that may be very unpleasant and affect quality of life and wellbeing. HRT tends to be used for relatively short durations in the perimenopausal period.¹⁰² It is generally prescribed for women at an age when fracture risk is low.

5.1 Use of HRT in the management of osteoporosis

Primary prevention of osteoporosis is outwith the scope of this guideline, but the Women's Health Initiative Study (WHI) has added greatly to knowledge in this area.¹⁰³ It is simply not known whether data from primary prevention studies can be extrapolated to treatment of osteoporosis. The WHI study looked at 16,608 normal, healthy women aged 50 to 79 taking a combination of conjugated equine oestrogens 0.625 mg and medroxyprogesterone acetate 2.5 mg. A secondary end point of the study looked at incidence of fractures. This confirmed the long held assumption that HRT prevents osteoporotic fractures. (Table 3).

Table 3: Reduction in osteoporotic fractures with HRT: WHI study data¹⁰³

There are limited data available from other randomised controlled trials on the use of HRT to prevent fractures. Recent meta-analyses^4,104 have been mainly influenced by two large trials, one of which (the HERS study) produced a negative result.¹⁰⁵

When assessing the role of HRT in the treatment of osteoporosis, there is only one small (75 patients) double blind randomised placebo-controlled trial assessing the efficacy of HRT (transdermal oestrogen with progestogen) in the secondary prevention of (vertebral) fractures.¹⁰⁶ When analysed on the basis of the number of vertebral fractures that occurred in the two groups, the oestrogen treated group had a significant reduction in vertebral fracture incidence. However, analysis based on the numbers of women with new vertebral fractures (the usual end point of more recent studies) did not show a statistically significant reduction.

A number of relatively small randomised controlled trials gauging the efficacy of oestrogen replacement (HRT) in treating low BMD have been conducted. One study¹⁰⁷ has compared the BMD benefits of HRT against alendronate and placebo and shows that with use of HRT over two to three years significant BMD gains occur at the lumbar spine and femur and are at least as great as those seen with bisphosphonates. These studies have not been of sufficient power to provide an insight into the antifracture efficacy of HRT. In the light of the WHI data on normal women where bone density was not known but HRT demonstrated overall fracture reduction, it seems highly likely that current use of HRT will also reduce fracture risk in women with known low BMD.

5.2 Benefits versus risks of hormone replacement therapy

The risks and benefits of HRT are complex and require the individual assessment of each woman considering taking HRT, especially for more than five years. The increase in risk of breast cancer associated with HRT is small, but related to duration of use and is the major reason why most women opt not to continue HRT in the long term. The large re-analysis by Beral and colleagues of the world wide data on HRT and breast cancer estimated that 45 women in every 1,000 who do not use HRT will have breast cancer diagnosed between the ages of 50 and 70.¹⁰⁸ For women who start to use HRT at age 50, the extra number of breast cancers that are diagnosed has been estimated as follows:

Table 4: Breast cancer incidence in women starting HRT at age 50

Other considerations in the risk benefit analysis include an increased risk of venous thrombo-embolism, with a relative risk between two and four with an absolute risk of around three per 10,000 users per year.¹⁰⁹ Raloxifene (a Selective Estrogen Receptor Modulator or SERM) carries a similar increased risk of venous thromboembolism (VTE).¹¹⁰ Previous history of VTE contraindicates oral HRT or raloxifene.¹¹¹

Use of unopposed oestrogen in women increases the risk of endometrial cancer by around six-fold after more than five years of use.¹¹² Progestogens should be added to reduce risk of endometrial cancer. Recent data make it clear that an increase in risk of endometrial cancer still remains with longer term use of sequential combined HRT (RR 1.5) and prescribers should be aware of this possibility after more than five years of therapy.¹¹³ This increasing risk is not found in women using continuous combined oestrogen and progestogen regimens.¹³⁹

Recently conducted randomised controlled trials have failed to show any benefit in coronary heart disease (CHD).^103,109,114 The WHI data confirmed this and, in addition, demonstrated excess risk of both myocardial infarction and stroke in HRT users (Table 5). A more recent analysis of evidence from RCTs on the long term effects of HRT provides further support for the lack of effect on CHD and increased risk of stroke.¹¹⁵

Further data from the WHI study¹⁰³ confirm increased risks of breast cancer and thromboembolism with HRT (Table 5). As yet, there are no data from the oestrogen-only versus placebo arm of the same study. Endometrial cancer rates were not affected, in keeping with previous data. Colorectal cancers appear to have a lower incidence. Overall there were no differences in mortality between the HRT and placebo groups.

Table 5: Absolute risk reduction/excess risk attributed to continuous combined HRT in 10,000 users over one year¹⁰³

Use of HRT can be considered as a treatment option for osteoporosis but the risks and benefits should be discussed with each individual woman before starting treatment.

Perimenopausal women at risk of osteoporosis could consider taking HRT and can be reassured that current usage of HRT reduces risk of osteoporotic fracture. Duration of usage should be based on regular reassessment of the risks and benefits of continuing HRT for each individual woman.