Nurses' Health Study Finds Increased Risk of Stroke With Hormone Therapy

Robert A. Wild, MD, PhD, MPH


June 27, 2008

Grodstein F, Manson JE, Stampfer MJ, Rexrode K. Postmenopausal hormone therapy and stroke: role of time since menopause and age at initiation of hormone therapy. Arch Intern Med 2008;168:861-868.

Hormone therapy (HT)—including estrogen alone (ET) and estrogen plus progestin (EPT)—is associated with increased risk of stroke whether HT is initiated near menopause or after years have elapsed. Risk also increases with increasing estrogen dose, reports the prospective, observational Nurses' Health Study (NHS). Risk is low in younger women, especially if estrogen dose is also low. The investigators used information obtained through biennial questionnaires completed from 1976 to 2004 by 121,700 female nurses (aged 30-55 y) about menopause, post-menopausal HT use, and cardiovascular disease. The study evaluated stroke risk associated with initiating oral HT at different ages and at varying intervals since menopause. It also examined risk associated with varying oral HT doses, and analyzed risks by stroke type. Multivariate relative risks (RR) and 95% confidence intervals (CI) were calculated for 485,987 person-years of follow-up for women who had never taken HT and 409,629 person-years for current HT users.

After adjusting for stroke risk factors, the RR of total stroke for current HT users was 1.39 for ET (95% CI, 1.18-1.63) and 1.27 for EPT (95% CI, 1.04-1.56). There was no clear difference in risk for women initiating therapy at younger versus older ages, but short-term use (<5 y) at a young age was not shown to increase risk. The RR for ET initiated near menopause was 1.20 (95% CI, 1.06-1.58) and for initiation following 10 years or more after menopause, RR was 1.31 (95% CI, 1.06-1.63). The risk was similar with EPT initiated near menopause (RR, 1.22; 95% CI, 0.95-1.55) and EPT initiated 10 years or more after menopause (RR, 1.18; 95% CI, 0.87-1.60). There was a strong trend of increasing risk with increasing estrogen dose, with RRs of 0.93, 1.54, and 1.62 for doses of 0.3 mg, 0.625 mg, and 1.25 mg, respectively (P for trend, <0.001).

Results from this well-done observational study corroborate those from the Women's Health Initiative (WHI) regarding effects of ET and EPT in women of different ages and different proximity to menopause: risk of stroke increases with age and dose of HT used. The NHS results are more generalizable to the population at large than the WHI because NHS participants provided information outside of a clinical trial format and rates from clinical trials are expected to be lower than those in the general population. Because NHS participants are educated healthcare professionals, their stroke rates and risks are likely lower than those in the general population (the rates of stroke reported here are lower than those reported by the American Heart Association by age group). Women with obvious comorbidities were excluded from NHS analysis because their stroke risk would be higher.

We also now commonly prescribe lower doses of HT than studied here; women in the NHS using the lowest doses had the lowest risk. Neither the age-adjusted nor the multivariate-adjusted risk of stroke was significantly increased for hemorrhagic stroke; risk increased for ischemic stroke only. The increased risk in the NHS is probably caused by systemic HT-induced liver changes of clotting and anticlotting factors, creating a slightly greater risk of thrombosis. Risk of venous thromboembolism (VTE) and pulmonary embolism (PE) is increased with both oral and transdermal HT use, with greater risk with oral HT use by overweight or obese women. The data presented here are not relevant to other forms of estrogen or other routes of delivery because women using other agents or routes were excluded from the study. Note, too, that the Estrogen and Thromboembolism Risk observational study suggested that both transdermal delivery and normal body weight are associated with less risk for VTE or PE.

The NHS results do not change my prescribing practices, but the results are very useful for discussion of HT risks and benefits with my patients. The most important message is that the risk is extremely small; incidence of stroke was relatively low in younger women; and the risk difference in women on HT aged 50 to 54 years was an additional 2 cases of stroke per 10,000 women per year. These NHS observational data are consistent with clinical trial data and help us make a reasonable estimate of excess risk for any individual woman. The authors suggest that their data are consistent with the practice of using the lowest dose of HT for the shortest duration.

When HT use is discussed with a patient, a window of opportunity is opened that could reduce risk of stroke through education and examination (of her waist measurement, blood pressure, fasting glucose, family history of stroke, and cholesterol targets). The major cause of stroke is uncontrolled hypertension, and the chance of having a stroke doubles for each decade of life after age 55. Prior stroke, transient ischemic attacks, heart attack, cigarette smoking, diabetes mellitus, vascular disease, high cholesterol, poor diet, physical inactivity, living in the southeast United States, being of lower socioeconomic status, abusing alcohol, and drug abuse are all substantial risk factors for stroke. Each of these carries more risk than using oral HT, even at the doses in this study. As menopause practitioners, we are in the optimal position to make a difference in a woman's heart health.

From the NAMS First to Know e-newsletter released May 27, 2008

For more, please visit


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.