Can "designer estrogens" or modern (3rd generation) selective estrogen receptor modulators (SERMs) such as raloxifene be used in coronary prevention instead of HRT?

Concerns about the potential side-effects of conventional hormonal replacement therapies have led to a search for alternatives. Monotherapy with the synthetic 19-nortestosterone derivative tibolone has attracted considerable attention since its estrogenic, gestagenic and androgenic metabolites reduce climacteric complaints and prevent osteoporosis without causing menstrual bleeding. These characteristics increase the compliance of postmenopausal women and allow the treatment of elder postmenopausal women without prior hormone replacement therapy. Tibolone lowers Lp(a), fibrinogen, PAI-1 and improves glucose tolerance, insulin sensitivity and endothelial function but lowers HDL cholesterol by more than 20%. The long-term effects of tibolone on cardiovascular disease are not known and need to be studied.
Alternative forms of hormone replacement therapy rely on selective estrogen receptor modulators (SERMs) such as tamoxifen and raloxifen or phytoestrogens. SERMs lower LDL cholesterol and Lp(a) and increase HDL cholesterol, although less effectively than estrogens, but do not raise triglycerides. A post-hoc analysis of 5 years follow-up data from 13,000 women who participated in the placebo controlled Breast Cancer Prevention Trial did not reveal any cardioprotective effects of tamoxifen. Similar to conventional hormone replacement therapy in HERS, tamoxifen rather increased the risk of thromboembolic events. A large prospective international randomised controlled trial is underway on the selective estrogen receptor modulator raloxifene. The results of this study are expected in 5-6 years.

Can plant estrogens (phytoestrogens) be used in HRT and do they reduce CHD risk?

Phytoestrogens such as isoflavone have been made responsible for the low cardiovascular morbidity in countries with a high consumption of soy products. However, only few data on the effects on risk factors and no trial data are available on these substances. Treatment of monkeys with soy-phytoestrogens reduced arteriosclerosis although to a lesser extent than estrogens.

Almost all modern HRT regimens contain progestin in addition to estrogen, and progestins may counteract the presumed beneficial effects of estrogen because they have androgenic effects. Should we develop HRT combinations based on progesterone or other completely nonandrogenic progestins?

What is the preferred progestogen in HRT?

Androgenic derivatives of 19-nortestosterone instead of medroxyprogesterone may offset the hypertriglyceridemic and prothrombotic effects of estrogens but oppose some positive effects of estrogens, e.g. on HDL cholesterol. Non-androgenic hydroxyprogesterone-derivatives such as medroxyprogesterone do not oppose the effects of estrogens on HDL cholesterol and triglycerides. Moreover, it is not clear of whether the effects on intermediary phenotypes (i.e. risk factors) allow any conclusion on final outcomes. For example, the increase in high density lipoprotein cholesterol seen with estrogens has been considered as evidence of an anti-atherogenic effect whereas the lowering effect of 19-nortestosterone-derivatives on high density lipoprotein cholesterol has been a major concern. However, data from genetic studies in men and transgenic animal studies call into question the concept that an increase in high density lipoprotein cholesterol is causally related to a reduction in cardiovascular risk. For example, patients and mice with a deficiency of hepatic lipase have high levels of high density lipoprotein cholesterol but nevertheless develop premature atherosclerosis. Interestingly, hepatic lipase is down-regulated by estrogens and up-regulated by androgens and thereby an important mediator for the effects of these substances on levels of high density lipoprotein cholesterol. Therefore it is not possible to give any conclusion on whether non-androgenic hydroxyprogesterone-derivatives are superior to androgenic 19-nortestosterone-derivatives. Nor is it recommendable to develop new HRT combinations on the basis of effects on intermediary phenotypes. It rather appears mandatory to perform controlled intervention trials to test the available regimens towards their effects on clinical outcomes.

If part of the protective effect of estrogens is due to their antioxidant properties, might it be possible to introduce fat?soluble estrogen derivatives which might become incorporated into LDL particles offering antioxidant protection? Most of LDL oxidation occurs in the artery wall suggesting that antioxidants embedded into the lipoprotein structure could retain protection when LDL particles have entered the arterial intima and lose the protection by water?soluble antioxidant in the circulation.

An interesting concept. No data are available as yet. Such drugs need to be tested in animal experiments first. The current experience with studies on the antioxidant vitamin E however suggest that controlled clinical trials are inevitable to demonstrate that anti-oxidative activities found in vitro can be extrapolated to anti-atherogenicity in vivo.

All observational studies indicating CHD risk reduction by HRT have employed oral estrogens. Does transdermal administration of estrogen (no first pass effect via the liver and therefore, perhaps, less effect on levels of clotting factors) have the same effect on CHD risk?

Dermal application reduces the stimulatory effects of estrogens on the hepatic production of very low density lipoprotein and coagulation factors, thus reducing the adverse effects of hypertriglyceridemia and a prothrombotic state but also the wanted effects on HDL and Lp(a). Again, controlled intervention trials are needed to assess the final effect on clinical outcome.

Could the presumably adverse androgenic effects of progestins be abolished by: a) transdermal administration of progestin or b) local intrauterine delivery of progestin (IUD)?

See answer to questions 3 and 4

The only placebo?controlled randomized CHD prevention study (HERS) did not result in CHD reduction in the treatment group. HERS employed conjugated equine estrogen. Should future studies also test estradiol, which is the most commonly used drug in many European countries?

Yes. Nevertheless, controlled intervention trials are needed to assess the final effect on clinical outcome.

As the evidence for CHD prevention with HRT is based on observational trials and the only adequate trial (HERS) was negative, is there any evidence?based justification for advocating HRT for CHD prevention at the present time?

Two large primary prevention trials of HRT are now underway (WISDOM in the UK and WHI in the US). Should we await the results of these trials before advising HRT for prevention of CHD?

The role of hormone replacement therapy in the prevention of coronary atherosclerosis in women after the menopause clearly requires further study. Long-term follow-up of the HERS patient group is proceeding. In addition, several large-scale randomized trials of hormone replacement are currently underway. These include the U.S. Women's Health Initiative, which is scheduled to yield results by 2005, the WELL-HART, a randomized trial of hormones in women with established coronary artery disease, and the British Women's International Study of Long Duration Oestrogen after Menopause study. The Women´s Health Initiative recently made data available to the Data and Safety Monitoring Board (DSMB). According to this the women in this study were informed that a "small increase" in heart attacks, strokes and venous thromboses was observed in the hormone-treated group. The DSMB has recommended, however, that the study should continue. The results of a small multicenter trial, the Estrogen Replacement and Atherosclerosis Study (ERA) were presented by Herrington and colleagues at the 49th meeting of the American College of Cardiology on March 13, 2000. In this trial, 309 women with a mean age of 65.8 years and documented coronary heart disease were randomly assigned to receive 0.625 mg conjugated estrogen per day, a daily combination of 0.625 mg of estrogen plus 2.5 mg medroxyprogesterone acetate (PremPro), or placebo. Over 3.5 years of treatment, coronary angiography did not detect any differences in disease progression between any of the groups. There were also no reported increases in cardiac events in any of the three treatment groups. The absence of a difference between the estrogen-only cohort and the PremPro cohort of this study argues against a proatherogenic effect due to daily administration of medroxyprogesteron acetate, a concern which had been raised based on studies in monkeys. The preliminary evidence from these first prospective trials of the effects of hormone replacement on CHD in postmenopausal women indicates that the early rise in events on hormone replacement may be due to thrombotic events in the coronary artery occurring within the first year of therapy. Such thrombotic events are thought to occur in areas of "active" coronary atherosclerosis which are characterized by the presence of unstable plaques. One therapeutic approach to this problem in the future might be the use of statin therapy for a period before hormone replacement therapy is introduced in order to increase plaque stability.

Should we go for a clinical selection of women who would gain from HRT? Who are the ideal candidates?

Since hormones and synthetic steroids used for HRT vary from each other in their anti- and pro-atherogenic effects it is appealing to stratify postmenopausal women for the indication of HRT in general and for the choice of a specific HRT regimen. A recent subgroup analysis of HERS showed that the increase in coronary events associated with estrogen-progestin treatment in the first year of the HERS trial was mainly seen in women with low levels of lipoportein(a) [Lp(a)]. Moreover, only in the subgroup in which Lp(a) was lowered, there was a significant reduction in CHD events. However, such post-hoc analyses can serve only as indicators for the design of future trials and cannot be used as definitive support for the notion that women with high Lp(a) levels do indeed benefit from HRT in terms of reduced coronary risk. However, this study shows the feasibility of the approach. Further candidate markers which will influence the choice of the specific HRT regimen are: Lp(a), HDL cholesterol, triglycerides, thrombophilic risk markers, personal and family history of CHD, osteoporosis, thrombosis or breast cancer.

If HRT is to be prescribed after the menopause for prevention of CHD, for how long should this prescription continue?

If trials can identify a safe HRT regimen effective in preventing CHD, this regimen should be taken lifelong.

What categories of women should be excluded from HRT (breast cancer families, BRCA genes)?

There is consensus that women with a personal or family history of breast cancer should not use HRT. BRCA screening is a matter of controversial debate. Further exclusion criteria: a personal or family history of venous thromboembolism and gallstones.

What checks for thrombogenic conditions should be performed before starting HRT?

A personal and family history of venous thromboembolism should be considered as contraindication of HRT. How certain thrombophilic risk factors affect the relative risks for thromboembolic events of women with certain HRT regimen needs to be assessed. Candidate genes are prothrombin, factor V, factor VIII, protein C and protein S, antithrombin, and MTHFR. In addition, autoantibodies against phospholipids and cardiolipin may be useful markers.

The CHD risk factors appear to be somewhat different than in men (HDL a better predictor and triglyceride a more powerful risk factor). Could this be clarified in the analysis of MONICA or some other epidemiologic database? Should preventive measures take into account sex difference in risk factors?

Meta-analyses of prospective population studies have shown that triglycerides have a closer association with CHD in women than in men. Data from the Framingham study also suggest that HDL cholesterol is a more important risk factor for CHD in women than in men. However this needs confirmation in larger studies. Since cardiovascular risk is best assessed by mathematic algorithms which combine the information of several risk factors (e.g. the PROCAM algorithm with age, smoking, total or LDL cholesterol, HDL-cholesterol, triglycerides, diabetes, blood pressure, family history of myocardial infarction) it is necessary to have epidemiological data which allow sex-specific global risk assessment. These data are as yet missing.

Different populations (or different sections of a population, i.e. specific social class) have different risk of fatal and non fatal IHD. Thus the balance between IHD vs. breast cancer is specific to the population observed. Should this be stressed in international guidelines?

In principal yes. But is this feasible? Likewise every nation would need its own epidemiological data base for decision making.

Are obesity and insulin resistance different in their importance for risk factors in men and women?

Especially in elder women, obesity, diabetes mellitus and hypertension are more prevalent than in men and therefore probably play an important role for the manifestation of CHD. A representative survey of German women performed by the Federal Ministry of Health showed that in women above the age of 50, about a third were overweight, defined as a body mass index of at least 30 kg/m2, more than a third had a systolic blood pressure of at least 160 mmHg and/or a diastolic blood pressure of at least 95 mmHg, while about 10% suffered from diabetes mellitus. In addition, physical inactivity was very common in women in this age group, with more than two-thirds reporting less than one hour of moderately strenuous physical exercise per week. Very similar data have been reported for American women, and this pattern of risk factors probably applies to older women in most developed countries.
A risk factor which requires special mention in women is diabetes mellitus. Prior to the menopause, the effect of diabetes mellitus on coronary heart disease risk in women is such as to abolish the specific advantage of female sex, while after the menopause diabetes increases coronary risk 3-fold to 7-fold in women compared to a 2-fold to 3-fold increase in men. These effects may be due to a particularly deleterious effect of diabetes on lipids and blood pressure in women. Type 2 diabetes mellitus, and the period of insulin resistance which usually precedes it for many years, are characteristically accompanied by high triglyceride levels and low levels of high density lipoprotein cholesterol. Low levels of high density lipoprotein cholesterol appear to be a stronger predictor of coronary heart disease in women above the age of 65 than in men of this age.
For all these reasons, conservative measures which interfere with the insulin resistance syndrome are a mainstay to reduction of coronary heart disease risk in women after the menopause: normalization of body weight and reduction of dietary fat. One of the most important conservative measures, and one which has not until now received sufficient attention, is the need for regular physical exercise. This means at least 30 minutes of moderately strenuous activity such as brisk walking every day. Women who have suffered a myocardial infarction should be enrolled in a cardiac rehabilitation program.

Is estrogen deficiency responsible for a specific cardiovascular risk factor profile in elderly women?

Most of the traditional and the novel risk factors become more prevalent in aging women: hypercholesterolemia, hypertension, diabetes mellitus, hypertriglyceridemia, hyperhomocysteinemia, hyperinsulinemia, hyperfibrinogenemia. Some but not all of these changes in risk factors are attributable to menopause. They are also the effect of other changes e.g. in the endocrine system which are caused by aging: somatopause, adrenopause, hypothyreoidism, and insulin resistance.

What is the role of MRI, EBCT and PET in risk evaluation in postmenopausal women?

Non-invasive imaging of coronary heart disease will probably become an important tool for the prevention of coronary heart disease in both men and women. The documentation of CHD by these methods will determine whether and which therapeutic intervention will be chosen to interfere with the occurrence of clinical events.

How would a trial to compare hypocholesterolemic drugs with HRT in the prevention of CHD in postmenopausal women be designed?

In contrast to the disappointing results of the HERS study, several of the large prospective trials of the statin drugs have clearly shown that lowering of low density lipoprotein cholesterol reduces the risk of coronary artery disease in women just as effectively as in men. The Air Force/Texas Armed Forces Coronary Atherosclerosis Prevention Study was performed in persons without coronary heart disease while the Scandinavian Simvastatin Survival Study and the Cholesterol and Recurrent Events Study included persons with a prior history of coronary atherosclerosis. All three studies contained substantial numbers of women, most of whom were post-menopausal. In all three trials, the benefits in women were as great as those in men with statins reducing the risk of a fatal or non-fatal myocardial infarction by about a third. In view of these results both medical and ethical issues need consideration for the design of studies which compare HRT versus statin intervention. In secondary prevention, it is not acceptable to exclude women with an LDL-cholesterol > 100 mg/dl from statin treatment. Insofar only a comparison of statin alone with statin plus HRT appears feasible. In primary prevention the risk factor profile of women must be taken into consideration. A direct comparison of statin alone with HRT alone appears feasible only in women with an LDL-cholesterol below a threshold value (e.g. <160 mg/dl). Above this threshold value, only comparisons of statin monotherapy with combined statin-HRT therapy appears feasible. A comparison of fibrates with HRT raises less ethical concerns since the preventive effects of fibrates are also still a matter of controversy.

In Eastern and Central Europe, widespread prescription of HRT or statins is not possible for cost reasons. What non?pharmacological intervention should be advised instead?

In contrast to the fall in CHD incidence seen in developed countries in recent years, the incidence of coronary heart disease has increased sharply in many of the former socialist countries of central and eastern Europe. This incidence has been seen in both women and men and has not yet been completely explained. However, the increase in cigarette smoking may carry some of the blame. Therefore public campaigns which advocate the cessation of smoking as well as high taxation of cigarettes are important interventions.
See also answer to question 17 for the importance of dietary and hygienic interventions.