International Task Force for Prevention of
Coronary Heart Disease

CORONARY HEART DISEASE: REDUCING THE RISK

3.9 CHD in post-menopausal women and hormone replacement therapy

CHD rates are lower in premenopausal women than in men of similar age. Thereafter, the male and female rates converge, though they do not meet until the eighth decade. Although CHD mortality rates increase exponentially with age, there is no abrupt increase in the slope of the rates at the age of the menopause when plotted on a logarithmic scale. If reduced plasma oestrogen levels due to cessation of ovarian function at the natural menopause directly influenced CHD mortality, the slope would be expected to increase after the age of 50 years. Nevertheless, there is evidence indicating that women who have had a surgical menopause are at at increased risk of CHD; this implies that postmenopausal oestrogen deficiency may increase risk directly.

Of the limited number of epidemiological studies have been carried out on CHD risk factors in women, most indicate a direct relationship between plasma cholesterol levels and CHD risk as is seen in men. These studies have also established that HDL-cholesterol levels are strongly and inversely related to risk in women. The direct relationship of triglyceride level to CHD risk is also more evident in women than in men. Cigarette smoking and hypertension are CHD risk factors in women. Risk is threefold greater in women who smoke 20 or more cigarettes per day than in nonsmokers, and even one to four cigarettes per day is accompanied by a two-fold excess risk; in women who stop smoking, CHD risk declines rapidly, reaching that of women who have never smoked within three to five years. Diabetes mellitus is a particularly potent source of risk, reducing the sex difference in CHD incidence.

The effects of oestrogen on arterial tone have been studied intensively. As discussed in section 2.8.2 (Endothelial function), one of the potent factors causing vasodilatation is nitric oxide, produced locally in response to physiological need, by the endothelium. This response is attenuated by the menopause, and is restored by oestrogen replacement. Though proof is awaited. it is plausible that this effect of oestrogen is a mechanism that could contribute importantly to reduction of CHD risk in menopausal women through its influence on coronary blood flow. Preliminary studies have indicated that oestrogens induce favourable changes in the carotid artery pulsatility index (an index of the presence of atherosclerotic lesions). In acute experiments oestrogens have a beneficial effect on exercise-induced myocardial ischaemia in menopausal women.

Premenopausal women are generally at low absolute risk of CHD. However, the use of combined oral contraceptive preparations in cigarette smokers increases risk materially.

There is comparatively little clinical trial data on the effects of cholesterol lowering in women. A trial of lipid-lowering drugs employing coronary angiographic end-points in persons of both sexes with familial hypercholesterolaemia showed a significant and more favourable outcome in women than in men. Two secondary prevention trials using clinical end-points have provided results consistent with this. In the (comparatively small) subsets of women the outcomes were, as in men, highly favourable. In addition, a meta-analysis of published observational studies has indicated that the overall CHD risk was reduced by almost 50% in women who had received oestrogen therapy at any time. As selection factors cannot be controlled in observational studies, this protective effect may be an overestimate. Ongoing randomized controlled clinical trials will clarify this question.

In addition to its value in the symptomatic relief of such menopausal symptoms as hot flushes and night sweats, hormone replacement therapy has long-term benefits. One is the reduction in menopausal bone loss. Extensive studies have established that oestrogens lower LDL-cholesterol and increase HDL-cholesterol; used as menopausal oestrogen replacement, the effect on lipoproteins is considerable, HDL-cholesterol increasing by 10% or more and LDL-cholesterol decreasing by 5-10%. Oestrogen replacement might therefore be expected, on theoretical grounds, to prevent or attenuate the adverse trends in cardiovascular diseases after the menopause.

Progestogens, given as part of hormone replacement regimens to offset the effect of unopposed oestrogen in increasing the incidence of carcinoma of the uterus, can, however, attenuate the potentially favourable effect of oestrogens on lipoproteins. Some progestogens (those with 19-nortestosterone configuration) have some androgenic activity and given alone can lower HDL-cholesterol as much as 30%, while somewhat increasing LDL-cholesterol. Those progestogens with 17-a configuration e.g. medroxyprogesterone are preferable in this respect and show smaller adverse effects on liporproteins. Conjugated equine oestrogen preparations elevate plasma triglyceride levels slightly or moderately; this effect is usually not observed in women taking estradiol-containing preparations. In hypertriglyceridaemic women, however, exogenous oestrogen has on occasion severely exacerbated the lipid abnormality, and has led to acute pancreatitis. Preliminary evidence suggests that elevated levels of lipoprotein(a) may be reduced by oestrogen either alone or in combination with progesterone.

A more recent form of hormone therapy is transdermal administration of oestradiol alone or in combination with transdermally administered norethisterone. Only limited data on transdermal administration are available, suggesting that effects on plasma lipids are weaker than with oral administration.

The findings in more than 30 observational studies suggest that postmenopausal oestrogen replacement may decrease CHD incidence by as much as 44%. Angiographic studies indicate a lower incidence of coronary occlusions. Findings for stroke incidence are less consistent, with no clear downward trend. Few data are available on CHD event rates associated with combined replacement therapy with oestrogen and progestogen, but it is evident that favourable lipoprotein effects are seen.

Unfortunately, the findings in observational studies, though highly suggestive of benefit, are not conclusive. Because of self-selection and other sources of selection, the value of hormone replacement in CHD risk reduction, and of its possible untoward effects, will only become apparent after completion of formal randomised controlled trials. One such trial, the Women's Health Initiative, will be completed in about 8 years and will, it is hoped, clarify the effects of hormone replacement regimes and of a low fat diet. Such trials will also clarify the extent of any effect of oestrogen replacement in increasing the risk of breast cancer.