Expert Commentary

Do the risks of hormone therapy persist after discontinuation?

Author and Disclosure Information


Yes, a few. Although the higher cardiovascular, thromboembolic, and stroke risks observed with conjugated equine estrogens (CEE) and medroxyprogesterone acetate (MPA) in the Women’s Health Initiative (WHI) disappeared after discontinuation, other risks remained elevated. For example, women who discontinued CEE plus MPA continued to have a greater risk of invasive breast cancer than women who had never taken hormones, although the difference was not statistically significant and risk declined slightly after discontinuation. All-cause mortality also remained higher in women who had taken hormones.


In Summer 2002, the WHI randomized trial of CEE plus MPA was stopped after the WHI Writing Group concluded that data showed more risks than benefits from therapy. The risks included venous thromboembolism and stroke, cardiovascular disease, coronary heart disease, and invasive breast cancer, while benefits were reduced risk of fracture and colorectal cancer. In this latest update, WHI investigators report on outcomes 3 years after discontinuation of the study medication.

Risk–benefit ratio improved

In contrast to findings at the time that hormone therapy was discontinued in the WHI, no increased risk of thrombosis, coronary heart disease, or stroke was observed during the subsequent 3 years in women who had taken CEE plus MPA. Furthermore, neither a statistically significant increase in the risk of invasive breast cancer nor a lower risk of fracture or colorectal malignancy was observed after discontinuation.

The global risk index for the entire 8 years of follow-up (hazard ratio [HR], 1.12; 95% confidence interval [CI], 1.03–1.21), which was elevated at the time the trial was stopped, was noted to be 1.11 (95% CI, 0.99–1.27) during the 3 years after discontinuation. All-cause death rate was higher after discontinuation than during overall follow-up, although this difference was not statistically significant (HR, 1.15; 95% CI, 0.95–1.39 and HR, 1.04; 95% CI, 0.91–1.18, respectively).

Three years after the trial was discontinued, risk of any diagnosis of cancer was modestly higher than during active trial in women who had taken hormones (HR, 1.24; 95% CI, 1.04–1.48). An increased incidence of cancer in these women appeared to reflect a higher risk of a diagnosis of cancer other than ones prespecified as outcomes (especially lung).

What this evidence means for clinical practice

The fact that the increased risk of cardiovascular outcomes and invasive breast cancer, and the reduced risk of fracture and colorectal cancer, did not persist after discontinuation of hormones is the take-home message from this follow-up analysis. At the same time, the overall increase in risk of malignancy justifies continued surveillance after combination hormone therapy is stopped.

Recent WHI reports have suggested that hormone therapy—particularly estrogen-only therapy—may be associated with a lower risk of coronary heart disease in recently menopausal women and menopausal women in their 50s. Accordingly, we look forward to the WHI age-specific follow-up analyses for both combination and estrogen-only therapy.—Andrew M. Kaunitz, MD

Next Article: