BACKGROUND:

The trial was a logical extension of preceding observational and cross-sectional studies on estrogen replacement therapy. Overall, the studies suggested a 50 percent reduction in risk of coronary heart disease in current estrogen users compared to non-users. In spite of such striking findings, most studies had been prone to a number of biases. One major criticism of observational studies had been that women receiving estrogen were generally healthier and more compliant than non-estrogen users.

There was a very large body of observational data suggesting that the use of estrogen in postmenopausal women reduced coronary heart disease mortality by approximately 45 percent. At the same time, there had been some concern that replacement therapy increased the likelihood of uterine cancer and perhaps breast cancer as well, although it was generally accepted that this risk was probably significantly less than the benefits obtained from the reduction of coronary heart disease mortality.

DESIGN NARRATIVE:

Randomized, double-blind, placebo-controlled. After baseline angiograms, patients were randomized to one of three arms: micronized 17-beta estradiol, 1 milligram per day; 17-beta estradiol plus medroxyprogesterone, 5 milligrams per day for twelve days per month; and placebo. Subjects in all three arms received lipid-lowering therapy, low fat/low cholesterol diet, and the HMG-CoA reductase inhibitor, pravastatin, in sufficient dosage to reduce low density lipoprotein (LDL) cholesterol levels below 130 mg/dl. The primary endpoint was progression/regression of coronary obstructive disease as measured by angiography, including the expert human panel and quantitative computer analysis. The secondary endpoint was carotid media-intima thickness determined by ultrasound. Clinical measures included lipids, lipoproteins, apolipoproteins, estradiol and medroxyprogesterone levels, urinary prostanoid metabolites, and insulin/glucose metabolism. Subjects were recruited at three centers with active coronary angiography units. Several core facilities supported the study: a Core Lipid Lab, a Reproductive Endocrine Lab, the Biostatistics Lab (Data Coordinating Center) and the Angiographic Imaging Laboratory.