Women have a lower risk of cardiovascular mortality than men, and female sex hormones have been considered to be a protective factor. Support for this advantage includes the low rate of coronary heart disease in pre-menopausal women, the increased occurrence of heart disease in women who have premature menopause or bilateral oophorectomy, and studies showing less heart disease in post-menopausal women who use hormone therapy. Other studies have disputed these results. Studies confirming that postmenopausal hormone therapy reduces cardiovascular deaths were based on limited observational data that may have been confounded by population biases. Based on an influential meta-analysis published in 1992, the American College of Physicians and other medical organizations recommended that all postmenopausal women be offered hormone therapy to prevent heart disease. Barrett-Connor looked at the clinical trials of estrogen used to decrease the risk of heart disease.

The Coronary Drug Project, an early prospective trial comparing estrogen use with placebo in the secondary prevention of coronary events in men, was concluded early because of an increase in thromboembolic events and myocardial infarctions in patients receiving hormone therapy. No further studies were initiated until the Heart and Estrogen/Progestin Replacement Study started in 1993. This secondary prevention trial studied unopposed estrogen and estrogen plus progestin in women who already had heart disease. There was no difference demonstrated in the incidence of nonfatal myocardial infarction or death from heart disease in the treated women. During the first year, there was a significant increase in cardiac events in the hormone therapy group. Multiple other small secondary prevention trials using various hormone regimens also have shown no benefit with therapy and possibly some harm.

Primary prevention trials, including the Women's Health Initiative (WHI), compared prevention strategies including hormone therapy in postmenopausal women. No benefit was demonstrated, and small excess risks of breast cancer, stroke, and pulmonary emboli were noted in the treatment group.

The author uses an epidemiologic review to conclude that recent clinical trials demonstrate that postmenopausal hormone therapy is unlikely to prevent heart disease and should not be included in heart protection plans for women. The early harm noted in these studies may be the result of a thrombotic state or inflammation.

In an accompanying editorial, Prestwood points out that lower dosages of estrogen (0.25 mg per day) and progesterone appear to be effective in preventing bone loss and reducing hot flushes in postmenopausal women, possibly without the adverse effects that may accompany WHI.

editor's note: The U.S. Preventive Services Task Force guidelines¹ on the use of post-menopausal hormone therapy for the primary prevention of chronic conditions, revised in 2002, recommend against the routine use of estrogen and progestin for this purpose. Benefits of combination hormone therapy appear to be increased bone mineral density, reduced risk for fracture, and, possibly, reduced incidence of colon cancer. Potential harms include increased risk for breast cancer, venous thromboembolism, coronary heart disease, and, possibly, stroke and cholecystitis.

Existing data are inadequate to support conclusions about the effect of hormone therapy on dementia and cognitive function, ovarian cancer, mortality from breast cancer or cardiovascular disease, or all-cause mortality. Based on this analysis, the harmful effects of estrogen and progestin probably exceed the chronic disease prevention value. Final decisions about postmenopausal hormone therapy should be made by patient preference, analysis of individual risk for specific chronic diseases, and the occurrence of post-menopausal symptoms. There is currently insufficient evidence to determine the balance of the benefits and harms of unopposed estrogen therapy in women who have had a hysterectomy. More data are anticipated with completion of ongoing studies.—R.S.