American Women’s Health Initiative
A summary by Professor Henry Burger, Emeritus Director, Prince Henry's Institute of Medical Research and Chairman of the Research Committee at the Jean Hailes Foundation for Women’s Health.
This summary is intended for medical and health professionals.
In July 2002 the announcement of the results of the combined hormone treatment trial within the American Women’s Health Initiative (WHI) was greeted with alarm, primarily because of the manner in which the breast cancer results were presented.
In fact, among the 75 per cent of participants (approximately 6000 in each arm of the study) who had never previously had hormone therapy (HT), there was no significant increase in breast cancer risk when the trial was ceased after 5.2 years.
In addition to investigating breast cancer, the main aim of the study was to determine whether hormone therapy had a role in the prevention of coronary heart disease.
Observational studies of women starting their treatment for symptoms around menopause (the main one being the American Nurses Health Study) had indicated that hormone therapy significantly lowered risk. However, there was no randomised trial to provide more rigorous evidence. When the WHI results for the combined treatment arm were announced, there was actually an increase in cardiovascular risk, particularly in the first year of treatment, which was somewhat unexpected.
There was a suggestion that women who were within ten years of menopause did not show any increase in risk, but those further from menopause did. It should be noted that the age range of participants was 50-79, average age 63, and many had significant risk factors for cardiovascular disease and would likely have had damaged arteries. A subsequent report described the results of giving oestrogen alone to hysterectomised women of the same age range. In them there was no significant benefit or risk for heart disease overall.
In the oestrogen only arm the rate of breast cancer was actually lower in the oestrogen treated women than in those receiving placebo.
Timing of hormone therapy?
Many writers have subsequently raised the issue of whether the timing of hormone therapy was the explanation for the failure of WHI to show benefit. Simply put, it was proposed that hormones given to women before they had significant arterial disease would reduce the risk of it developing, as had been shown in animal studies, while giving it to women who had already developed arterial disease, could not and would not be protective.
Early in 2006, two papers from investigators involved in the WHI have clarified this issue to some extent.
In the first paper, published in the Journal Of Women’s Health (Vol 15: 35-44, 2006), Grodstein (a major investigator in the Nurses Health Study), Manson (the lead heart disease investigator in WHI) and Stampfer (NHS) confirmed that in the Nurses Health Study women beginning HT near menopause had a 34 per cent reduction in risk of coronary heart disease when taking oestrogen alone and a 28 per cent reduction for combined therapy.
In contrast, in women who had started HT more than ten years postmenopausally, there was no relation between hormone therapy and coronary heart disease. The authors stated that in further analyses of the WHI data, there was a greater risk with initiation of therapy in women who were long past menopause – 11 per cent risk reduction in those less than ten years postmenopausal, 22 per cent increase in those 10-19 years after menopause and a 71 per cent increase for those 20 years or more past the menopause.
Similarly for oestrogen alone, there was 44 per cent protection (not statistically significant) with initiation at ages 50-59, whilst there was an eight per cent increase in risk 60-69 and a four per cent increase 70-79.
Timing is crucial
The second paper, by Hsia et al., (Archives of Internal Medicine, 166: 357-365, 2006), commenting on the oestrogen only arm, stated that there was a non-significant 27 per cent protection in women aged 50-59, but if heart attack was combined with coronary revascularisation procedures, there was a significant 45 per cent reduction in risk.
These two papers support the notion that the timing of hormone therapy is crucial in whether hormones may have a protective role, or alternatively may increase the risk of coronary heart disease.
There is no evidence for an adverse effect in women close to menopause and in fact in such women hormone therapy overall appears to be cardioprotective. It is only when it is started many years later in women who likely have arterial disease that problems could arise.
Thus, for neither the breast cancer outcome, nor the heart disease outcome, is there justification for anxiety regarding three to five years hormone therapy for symptoms in women around the time of menopause.
Further reading on this topic
Professor Burger recommends:
The first volume of the journal Menopause for 2006 Vol. 13. On pages 1-3 there is an Editorial stating "Age or time since menopause may importantly influence the benefit-risk ratio associated with HT, especially with respect to coronary disease outcomes".
Then an article pages 139-147 entitled "Postmenopausal hormone therapy: new questions and the case for new clinical trials" comes to precisely the same conclusion and in the Abstract it is stated "There is now a critical mass of data to support the hypothesis that age or time since menopause may importantly influence the benefit-risk ratio associated with HT, especially with respect to cardiovascular outcomes".
The first author in the latter article is JoAnn Manson who is also a co-author of the Editorial and who was the lead cardiovascular investigator for WHI.
Content updated May 08, 2008