[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
July 5, 2000

Early Risks of Hormone Therapy in Patients With Coronary Heart Disease

Author Affiliations

Stephen J.LurieMD, PhD, Contributing EditorIndividualAuthor

JAMA. 2000;284(1):41-43. doi:10.1001/jama.284.1.39

To the Editor: An early increase in coronary death and nonfatal myocardial infarction (MI) in postmenopausal women with coronary heart disease (CHD) taking hormone therapy compared with placebo was an unanticipated adverse outcome in the Heart and Estrogen/progestin Replacement Study (HERS).1 The time trend for initial risk and later benefit was statistically significant. Two of us (N.K.W. and G.L.K.) involved in the HERS study recognized some striking similarities between the results of this clinical trial in postmenopausal women with those of the Coronary Drug Project (CDP)2 (with which all 3 of us were involved), conducted almost 3 decades earlier in men. HERS and CDP are the only large trials of hormone therapy in populations with CHD, and both data sets could be analyzed further owing to our personal involvement. This report compares the primary outcomes of these trials with particular focus on early excess events following initiation of hormone therapy.


HERS was a randomized placebo-controlled trial involving postmenopausal women aged 44 to 79 years (mean age, 67 years) with documented CHD who were randomly assigned to receive 0.625 mg of conjugated equine estrogen plus 2.5 mg of medroxyprogesterone acetate (n=1380) vs identical placebo (n=1383).1 Most subjects had had either coronary angioplasty or coronary artery bypass graft surgery; 17% qualified with Q-wave MI. HERS subjects were followed up for a mean of 4.1 years between 1993 and 1998.

In 1966-1974, the CDP compared 2 estrogen regimens (2.5 mg/d and 5.0 mg/d) (n=1101 and n=1119, respectively) vs placebo (n=2789) in men aged 30 to 64 years (mean age, 52 years) with electrocardiographically documented MI.2 Most CDP subjects (78%) qualified with Q-wave MI, and 22% had other electrocardiographic abnormalities. In 1970, with a mean follow-up of 1.5 years, the 5.0-mg/d estrogen treatment arm was discontinued because of excess mortality and nonfatal MI.3 In 1973, with a mean follow-up of 4.7 years, the 2.5-mg/d estrogen treatment arm was discontinued because of adverse treatment effects and absence of benefit for total mortality.4 Estrogen dosages in the CDP were 4- to 8-fold greater than in HERS.

Time trends of events were not standard statistical analyses at the time of the CDP. Thus, we performed new analyses of the CDP data according to period of follow-up. At the same time, we reanalyzed treatment effects in the first 4 months of follow-up in HERS.


Rates of combined CHD deaths and nonfatal MI were nearly identical in the HERS hormone therapy (13.0%) and placebo (13.2%) groups1,5 and in the 2.5-mg/d estrogen therapy (23.3%) and placebo (23.4%) groups in the CDP (Table 1).4 For all-cause mortality, treatment effects in HERS and the CDP were nearly identical, with relative hazards (RHs) of 1.07 in HERS and 1.06 in CDP (Table 1). In both trials, hormone therapy groups had substantially higher incidences of thromboembolic events and gallbladder disease than placebo groups (Table 1).1,4-6

Table 1. Mortality and Morbidity Outcomes
by Treatment Group*
Table 1. Mortality and Morbidity Outcomes by Treatment Group*

In both trials, an early adverse effect of hormone therapy was followed by a subsequent modest benefit. For months 0-4 in HERS, 3.5 women per 100 women per year in the hormone therapy group had primary CHD events compared with 1.5 women in the placebo group (RH, 2.29) (Table 2). For months 13-60, the corresponding rates were 3.1 and 3.6 women per 100 women per year for hormone therapy and placebo groups, respectively (RH, 0.85; P=.04 compared with months 0-4).1,5 Similarly, in the CDP the RH was 1.58 for months 0-4 compared with 0.96 for months 13-60 (P=.03; Table 2). In the CDP 5.0-mg/d estrogen therapy group, early adverse effects lasted 2 years and were 1 of the reasons for early discontinuation of that treatment arm.3

Table 2. Rates per 100 Persons per Year of
Primary CHD Events*
Table 2. Rates per 100 Persons per Year of Primary CHD Events*


These findings support the conclusion7 that starting estrogen therapy for cardioprotection in postmenopausal women with CHD is not advisable. For instance, hormone use for secondary prevention in the Nurses' Health Study showed an early increase in risk followed by a later decrease in risk.7 For women receiving estrogen therapy for 1 year or more, however, these data from HERS and CDP indicate a possible cardiovascular benefit of continuing therapy. The concordance of these findings is striking given that the earlier trial involved only men with CHD while the latter trial involved only postmenopausal women, also with CHD.

Box Section Ref ID

Financial Disclosures: Dr Wenger has received research grants/contracts from Wyeth Ayerst, Solvay Pharmaceuticals Inc, and Eli Lilly & Co; has served as a consultant on the Eli Lilly Raloxifene Advisory Committee; and has served on speakers bureaus for Wyeth Ayerst and Eli Lilly & Co. Dr Knatterud has received research support from Boston Scientific, Centacor, Cor Therapeutics Inc, Guidant Corp, Medtronic, Merck & Co Inc, General Electric Co, Critikon, and EI DuPont Co; has served on the data safety and monitoring board for HERS; and has received honoraria from Wyeth Ayerst. Dr Canner has served on the data and safety monitoring board for the Wegener's Granulomatosis Etanercept Trial, funded by the National Institute of Arthritis and Musculoskeletal and Skin Diseases and Immunex Corp.

Note: The final HERS data on which these results are based are available on the Internet.5

Acknowledgment: We thank Eric Vittinghoff, PhD, for assistance with the statistical analysis.

Hulley  SGrady  DBush  T  et al.  Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women.  JAMA. 1998;280:605-613.Google Scholar
Coronary Drug Project Research Group, The Coronary Drug Project: design, methods, and baseline results.  Circulation. 1973;47(suppl 3):I1-I50.Google Scholar
Coronary Drug Project Research Group, The Coronary Drug Project: initial findings leading to modifications of its research protocol.  JAMA. 1970;214:1303-1313.Google Scholar
Coronary Drug Project Research Group, The Coronary Drug Project: findings leading to discontinuation of the 2.5-mg/day estrogen group.  JAMA. 1973;226:652-657.Google Scholar
 Final results of the HERS main trial. Available at: http://www.keeptrack.ucsf.edu/hers2/HERSfindat.htm. Accessed March 3, 1999.
Coronary Drug Project Research Group, Gallbladder disease as a side effect of drugs influencing lipid metabolism: experience in the Coronary Drug Project.  N Engl J Med. 1977;296:1185-1190.Google Scholar
Grodstein  FManson  JEStampfer  MJ Postmenopausal hormones and recurrence of coronary events in the Nurses' Health Study.  Circulation. 1999;100(suppl 18):I871.Google Scholar