Mayo Clinic and Mayo Foundation Rochester, Minnesota 55905
Address correspondence to: Lorraine A. Fitzpatrick, M.D., Director, Womens Health Fellowship, Department of Internal Medicine, Endocrine Research Unit, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905.
![]() |
Introduction |
---|
![]() ![]() ![]() |
---|
Studies of the relationship between physiological or pharmaceutical estrogen levels and protection from cardiac disease remain problematic. Observational studies in both men and women have not unequivocally demonstrated a link between endogenous estrogen levels and the risk of CAD. Estimates of the relative risk of myocardial infarction in users of estrogen or estrogen-progestogen combination have varied widely. Many studies have presented relative risks ranging from 0.50.65 (2); contradictory literature has also been published (odds ratio 0.96, 95% CI 0.661.40 in current users) (3). Our lack of basic understanding of hormone interactions and bioavailability of different estrogens prevents definitive interpretation of these data.
Many prospective studies have indicated that administration of oral estrogen is associated with a reduced relative risk for the development of coronary artery disease. The Nurses Health Study, which followed 48,470 postmenopausal women, indicated that estrogen-treated women had half the risk of developing CAD compared with untreated women (4). Case-controlled studies that used angiographically-defined CAD in postmenopausal women found a greater than 50% reduction among estrogen users compared with nonusers. Survival has also been documented to be improved in those women taking estrogen replacement therapy (ERT) with angiographically proven disease.
Estrogen replacement therapy is commonly prescribed to prevent bone loss, to prevent the development of atherosclerosis, and most recently, to improve cognition. Undesirable side effects such as breast tenderness and resumption of menses have dampened the enthusiasm for some of the beneficial effects offered by estrogen replacement therapy. The designer estrogens, such as raloxifene, are compounds with biological actions that are mediated through binding to the estrogen receptor. Differential expression of estrogen-related genes mediated by separate response elements occurs, resulting in diverse actions in target tissues (5).
Raloxifene has been evaluated in 276 postmenopausal women in clinical pharmacologic trials and in 1300 postmenopausal women in selected raloxifene trials. Pharmacokinetic parameters exhibit high within-subject variability with a 30% coefficient of variation. In clinical trials, raloxifene has estrogen-like effects on bone and lipid metabolism (6). Newly acquired data suggest a lack of estrogen-like action on uterine and breast tissue. Distinct differences among raloxifene and various estrogens are clearly documented in animal and human studies. Raloxifene prevented increased bone loss after ovariectomy in rats, and it increased bone mineral density (BMD) in cynomolgus monkeys. These effects were similar to the findings in estrogen-treated animals. In postmenopausal women, raloxifene preserved bone mass and increased BMD at all sites (hip, spine, total body) as compared with calcium-treated control subjects. Three randomized, placebo-controlled, double-blind osteoporosis prevention trials had similar results. Thus, raloxifene is an estrogen agonist on skeletal tissue.
Uterine stimulation is a frequent side effect that results in discontinuation of estrogenic compounds. In the osteoporosis preventive trials, endometrial thickness was evaluated by transvaginal ultrasound every 6 months. No differences in endometrial thickness or incidence of vaginal bleeding were noted in raloxifene compared with placebo-treated groups. These findings are in contrast to the uterine effects noted in an unopposed estrogen-treated group, where 13 of 37 developed proliferative endometrium compared with none of the 43 raloxifene-treated women. The frequency and severity of breast pain and tenderness was indistinguishable in the raloxifene vs. the placebo-treated groups.
What of the specific biological response of cardiovascular tissue? Heart disease is the leading cause of death in women, and market research indicates that it is among the top three leading health concerns of women. One issue is the complicated pathophysiology of cardiovascular disease. The mechanism of estrogen action on the cardiovascular system encompasses direct and indirect effects: favorable alterations of the balance between fibrinolysis and coagulation, changes in lipoprotein levels, and direct effects on relaxation/contraction of arterial smooth muscle. Exercise-induced myocardial ischemia is reduced by pharmacological doses of estrogen (7), and estrogen reverses the acetylcholine-induced vasoconstriction of atherosclerotic coronary arteries (8). Other investigators have demonstrated effects of estrogen that are mediated through intracellular calcium changes in vascular smooth muscle cells and relate these cellular pertubations to alterations in calcium channel activation, nitric oxide levels, or the influence of endothelin. Although many investigators have described a substantive relationship between estrogen replacement therapy and coronary artery disease, long-term effects are not well studied. Estrogen favorably alters lipid levels, but this change only accounts for approximately 30% of the cardioprotective effect. Other actions of estrogen such as antioxidant potency, changes in endothelial function, alterations in apoprotein levels, and vasodilation/constriction are unable to be fully accounted for when assessing the risk-to-benefit ratio for estrogen.
What about the primate model? Cynomolgus monkeys are considered a well-accepted model to follow progression of atherosclerosis over time. These animals respond to an atherogenic diet with the formation of plaque, and treatment with estrogen/progestin combination prevents the development of disease. So where do the new "designer estrogens" fit into the picture with regard to coronary artery disease? Data in humans has been published indicating some differences in the lipid profile response to raloxifene. Conjugated and esterified estrogens will raise high density lipoprotein HDL and lower low density lipoprotein LDL cholesterol; however, raloxifene reduced LDL and had no significant effect on HDL cholesterol (6). In the article by Clarkson et al. (9) (see page 721), ovariectomized cynomolgus monkeys were fed an atherogenic diet and treated with placebo, raloxifene (1 mg/kg/day), raloxifene (5 mg/kg/day), or conjugated equine estrogen (CEE) at equivalent replacement doses. Comparable lipid profile changes were noted in the raloxifene and CEE-treated monkeys as compared with published data in humans. However, as it is difficult to find human volunteers to provide coronary arteries for pathological evaluation, the "gold standard" of plaque formation was assessed in these animals. Surprisingly, there was significant plaque formation in the raloxifene-treated groups compared with those taking CEE.
All of these studies have lead to additional questions. What is the mechanism of the action of raloxifene on vascular smooth muscle cells, endothelium, and natural history of plaque formation? How do these questions translate into the clinical paradigms we face daily? What will happen to the woman at risk for coronary artery disease in terms of decisions regarding treatment options? How does one advise the osteoporotic patient with no risk factors for coronary artery disease? The patient with a family history of Alzheimers? The patient with a past history of breast or uterine cancer?
Overall, this novel class of compounds provides new therapeutic options for treatment of osteoporosis. Long-term evaluation is essential to provide mechanistic answers to questions regarding the effects of designer estrogens on the cardiovascular system. Careful consideration, deliberation and thoughtfulness will be necessary to evaluate each case in order to provide the best treatment course for the postmenopausal woman.
Received January 6, 1998.
Accepted January 6, 1998.
![]() |
References |
---|
![]() ![]() ![]() |
---|