Several recent trials of combination hormone therapy found that the risk of coronary heart disease (CHD), stroke, venous thromboembolism, and breast cancer outweighed the benefits of therapy. The estrogen-only (conjugated equine estrogen [CEE]) arm of the Women’s Health Initiative trial was allowed to continue until it ended in February 2004. A study committee presented the primary outcomes of the estrogen-only arm of the trial.

Healthy women 50 to 79 years of age who had undergone hysterectomy were randomized to receive 0.625 mg of CEE per day or placebo. Clinical, laboratory, and radiographic data were collected at regular intervals to assess CEE’s effects on CHD, venous thromboembolism, fractures, cancer, overall health, and mortality. Women were followed for an average of 6.8 years. Although no predefined boundaries for termination of the study had been crossed, the risk of stroke almost reached the point of crossing and appeared to be similar to the risk of stroke previously reported in the combination arm of the trial. Of the 10,176 women with available data, 5.2 percent withdrew, and 53.8 percent discontinued medication. Most of the discontinuations occurred in both groups early in the trial, and the cohort remained stable thereafter.

Significant decreases in low-density lipoprotein levels and increases in high-density lipoprotein levels were noted in the CEE group, with mean systolic blood pressure increases of 1.1 mm Hg in the CEE group. CHD rates were similar in both groups, but stroke incidence was 39 percent higher in the study group. Venous thromboembolism was increased by 33 percent in the CEE group. Breast cancer was reduced by 23 percent compared with placebo, but this reduction was not statistically significant. Fracture rates also were reduced by one third. Overall health and mortality were equivalent in the two groups.

CEE reduced fractures significantly in women, with no effect on CHD, global health, or overall mortality. A trend toward breast cancer reduction was observed, which may have been due solely to chance (in other trials, no such reduction has been observed). However, it appears quite likely that CEE alone does not bear the same adverse breast cancer risk as combination therapy. Stroke and venous thromboembolism were increased in women taking estrogen. Overall, these findings give little reason for using estrogen therapy for any preventive purpose. The authors conclude that the only indication for estrogen therapy is to alleviate menopausal symptoms. In these women, estrogen replacement for a period of six to seven years appears to be safe.