HRT received a lot of adverse publicity following the publication of data from two major studies - the Women's Health Initiative Trial (WHI) in 2002 followed by the Million Women Study (MWS) in 2003.
WHI reported increased risks of breast cancer, heart disease and stroke in women using oestrogen with progestogen. The trial was halted early because the risks of treatment were assessed to be greater than the benefit.
The oestrogen only arm of the WHI trial was discontinued in 2004 because of an increased risk of stroke in this group of women who had had a hysterectomy.
The MWS was based on observational data from UK women attending for routine breast screening. An increased risk of breast cancer was reported, greater when oestrogen and progestogen was used compared to oestrogen alone.
Following these findings, the Committee for the Safety of Medicines issued a statement in 2003 advising doctors to prescribe HRT in the 'lowest dose for the shortest time'.
Almost overnight the benefits of HRT were forgotten.
Many women stopped HRT or were too frightened to consider it despite severe menopausal symptoms. Some GPs limited the number of years they were prepared to prescribe HRT and others have refused to prescribe it at all. Overall prescriptions for HRT have fallen by one third, mainly because of fewer new users.
GPs need to give patients balanced information so they can make an informed choice.
To interpret risk data it is essential to understand the difference between relative risk and absolute risk. Relative risk is commonly used to report data from trials but takes no account of the incidence of a disease. Absolute risk gives the number of cases in a population.
So a relative risk of 2.0, which doubles risk, will have little impact on a rare event such as death at a football match (around one in 250,000) but a huge impact on a common event such as wearing spectacles (around one in two of adult population).
It is helpful therefore to give patients absolute numbers.
The main indication for using HRT is to control hot flushes and accompanying menopausal symptoms. Most women experience these symptoms for several years but it is relatively uncommon to use HRT for more than five years for this indication.
All the regulatory bodies have agreed that the risk/benefit balance for short-term use of HRT for the control of symptoms in women in their fifties is generally favourable. It is helpful to look at the data from WHI and MWS and interpret the risk in terms of five years' use for women in their fifties.
Breast cancer causes our patients more worry than most conditions and they are aware that HRT increases this risk. However, the absolute number of extra cases of breast cancer reported by WHI was four in 1,000 women using oestrogen with progestogen for five years.
This is the same as saying a woman in her fifties using HRT for five years has a 0.4 per cent extra risk of developing breast cancer during that time compared to the background risk of 1.5 per cent in non-HRT users.
However, obesity and alcohol excess carry similar increased risks of breast cancer - many women are not aware of this.
There were no extra cases of breast cancer in the WHI oestrogen-only group and 1.5 extra cases reported by MWS.
According to WHI data the risk of ischaemic stroke in HRT users (both oestrogen with progestogen and oestrogen only) is increased by about 30 per cent. The background risk of stroke is age-dependent and approximately doubles each decade. For women in their fifties the absolute risk is very small and a 30 per cent increase in risk is therefore still small compared to women in older age groups.
The risk of stroke increases in the presence of other known risk factors such as obesity, smoking, and hypertension - HRT users in these groups will carry more risk. Overall, WHI reported four extra strokes for 1,000 women using five years of HRT, but for women in their fifties one or two extra strokes could be expected.
HRT and cardiovascular risk still remains an area of considerable uncertainty.
Observational data had shown reduced risk of coronary events in HRT users, but WHI reported an extra four coronary events for 1,000 women in five years in oestrogen/progestogen users.
Sub-analysis of the figures did not show this increase in women within 10 years of their menopause. Neither was there an increased risk in oestrogen-only users.
It seems that starting HRT in older women will carry more risk than those in early postmenopausal years.
It has been well-established that HRT increases venous thromboembolism (VTE) risk two- or three-fold. As with stroke, the impact of this risk is greater with increasing age and in those with other risk factors.
WHI reported four extra VTE events for 1,000 women in their fifties using oestrogen with progestogen for five years.
We need to consider the type of therapy women will use. For example, oestrogen-only preparations have shown little or no extra risk of breast cancer and no increased cardiovascular risk in comparison to oestrogen with progestogen.
MWS reported tibolone to have less breast risk than oestrogen/progestogen preparations but to carry greater endometrial risk. Further studies are in progress that should clarify the position of this preparation. Transdermal preparations are believed to be safer in relation to clotting.
It is interesting when considering HRT risks to make comparisons with the risks associated with other commonly used therapies. A recent report on SSRIs and GI bleeding (requiring admission to hospital) quoted a risk of one in 80 patient years when used together with NSAIDs.
The WHI data concerning breast cancer risk equates to less than one extra breast cancer in 1,000 women years, which is considerably less. Also one in two smokers will die from a smoking-related disease.
Women deserve accurate information about HRT risks to make an informed choice.
- Dr Hodson is a GP in Stratford upon Avon, Warwickshire, with a special interest in menopause and HRT
- Writing Group for the WHI Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAMA 2002; 288: 321-33
- Million Women Study Collaborators. Breast cancer and hormone replacement therapy in the Million Women Study Lancet 2003; 362:419-27
- The WHI Steering Committee. Effects of conjugated equine oestrogen in postmenopausal women with hysterectomy. JAMA 2004; 291: 1,701-71
- Paton C, Ferrier N. SSRIs and GI bleeding. Editorial BMJ 2005; 331: 529-30
- HRT causes an extra four cases of breast cancer per 1,000 women taking oestrogen with progestogen.
- Up to four extra cases of stroke are found per 1,000 women taking all types of HRT for five years.