Current issues of ACP Journal Club are published in Annals of Internal Medicine


Coronary disease and stroke after postmenopausal estrogen

ACP J Club. 1992 Jan-Feb;116:29. doi:10.7326/ACPJC-1992-116-1-029

Source Citation

Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and cardiovascular disease. Ten-year follow-up from the Nurses' Health Study. N Engl J Med. 1991 Sep 12;325:756-62.



To ascertain whether postmenopausal estrogen use increases the risk for coronary disease and stroke.


Cohort of women followed for 10 years.


Mailed-response study of registered nurses.


48 470 postmenopausal women (age 30 to 63 years) from the sample of 121 700 women surveyed every 2 years as part of the Nurses Health Study were included. Excluded from the analysis were those who had provided no information on hormone use and those reporting diagnoses of cardiovascular disease or stroke. Follow-up was 88% and included 337 854 postmenopausal person-years.

Assessment of prognostic factors

Mail surveys, started in 1976, assessed all aspects of health and included specific questions on hormone supplementation after menopause and previous, current, and parental cardiovascular disease. Medical records were checked and telephone interviews completed with patients and next of kin for nurses who had fatal and nonfatal myocardial infarctions and strokes. These were classified as either confirmed or probable. Investigators were blinded to estrogen use of the patient during chart audit and interviews. Women were classified into 3 categories of estrogen use: current, former, or no use.

Main outcome measures

Nonfatal myocardial infarction, fatal coronary heart disease, coronary artery surgery, fatal and nonfatal stroke, cardiovascular mortality, and death from all causes. All analyses were adjusted for age.

Main results

There were 293 nonfatal myocardial infarctions (228 confirmed, 65 probable), 112 confirmed deaths from coronary artery disease, and 224 strokes (52 fatal, 172 nonfatal; 177 confirmed, 47 probable). Age-adjusted relative risk (RR) for major coronary disease for current estrogen users was 0.51 (95% CI, 0.37 to 0.70; P < 0.001) and for previous estrogen users was 0.91 (CI, 0.73 to 1.14; P > 0.2). The ageadjusted RR for stroke for current users was 0.96 (CI, 0.67 to 1.37). After adjustment for other risk factors, the RR for mortality from all causes for current and previous estrogen users was 0.89 (CI, 0.78 to 1.00); the RR for cardiovascular mortality was 0.72 (CI, 0.55 to 0.95; P = 0.02). There was no association between current estrogen use and coronary-bypass surgery (age-adjusted RR, 1.21; CI, 0.84 to 1.73).


Current estrogen use is associated with a reduction in major coronary disease and cardiovascular mortality, but not with changes in risk for stroke or coronary-bypass surgery.

Source of funding: National Institutes of Health.

Address for article reprint: Dr. M.J. Stampfer, Channing Laboratory, 180 Longwood Avenue, Boston, MA 02115.


This and many other studies (1) suggest that postmenopausal estrogen reduces the risk of a coronary event by about 50%, a biologically plausible effect because noncontraceptive estrogen lowers blood pressure and improves lipoproteins in most women. As prevention of heart disease would outweigh any known risk, why not treat nearly every postmenopausal woman with estrogen?

First, these reports are based almost entirely on the use of estrogen without progestin. Today, women with an intact uterus are usually prescribed estrogen with a progestin in order to prevent the endometrial cancer associated with unopposed estrogen. The long-term consequences of this regimen are unknown, but progestins mask some of the HDL-cholesterol elevation that is seen with unopposed oral estrogen.

Second, some of the apparent benefit may be exaggerated. Women prescribed elective replacement estrogen are apt to be more educated and healthier than untreated women, factors that would reduce their risk. Hormone-treated women need to see their physician regularly and may more often have their cholesterol and blood pressure checked and treated, thus gaining additional benefit. Studies have been based on compliant women, and compliance itself is a characteristic associated with good health.

Heart disease is not the only consideration of interest, or necessarily the primary concern, of women who seek advice about hormone replacement. Long-term estrogen, with or without progestin, may increase the risk for breast cancer by 50%. Other less well-defined associations, which may be increased or decreased depending on the estrogen preparation or the patient, include headaches, arthritis, depression, and phlebitis. On the other hand, estrogen clearly delays bone loss and reduces fracture risk.

Until clinical trial data are available, estrogen replacement intended primarily to prevent heart disease should probably be reserved for women with unfavorable lipoproteins, premature menopause, or a family history of heart disease.

Elizabeth Barrett-Connor, MD
University of California San Diego, California