Author + information
- ↵*Reprint requests and correspondence:
Dr. David Waters, Division of Cardiology, Room 5G1, San Francisco General Hospital, 1001 Potrero Avenue, San Francisco, California 94110, USA.
Unstable angina and non-ST-elevation myocardial infarction (MI) accounted for 1,433,000 hospitalizations in the U.S. in 1996 (1). Nearly 60% of those with unstable angina were >65 years and 46% were women. Women comprise a larger proportion of the unstable angina population than of the MI population (2).
The outcome of patients with unstable angina has improved remarkably over the past two decades as antiplatelet therapy, antithrombotic therapy and new revascularization techniques have been introduced and refined. Nevertheless, approximately 15% of unstable angina patients will experience death or nonfatal MI within the first six months.
Reducing this risk represents a major challenge. It could be argued that further advances in antiplatelet, antithrombotic or revascularization treatments are unlikely to reducethis high event rate by more than small increments, and that new approaches should be tried. For this reason, the study by Schulman et al. (3)reported in this issue of the Journalis welcomed.
The current study
The 293 patients enrolled in this trial exhibited the typical features of women with unstable angina. Their mean age was 70 years, more than three-quarters had hypertension, nearly half had diabetes and one-fifth had had a previous episode of heart failure. For each of these characteristics, the prevalence was higher than what would be expected in men with unstable angina.
The women were randomized to a 30-min intravenous (IV) infusion of estrogen followed by oral conjugated equine estrogen for 21 days, IV estrogen followed by conjugated estrogen plus medroxyprogesterone for 21 days or placebo. The primary end point, namely ambulatory electrocardiographic ischemic episodes during the first 48 h, was not reduced by either of the active treatments. In-hospital and six-month rates of death plus nonfatal MI were also similar in the three groups.
As noted by the investigators (3), their study lacked sufficient power to detect modest differences among the treatment groups. They planned to enroll 351 patients, expected a mean of one ischemic episode per patient and assumed a 40% reduction in this end point. The trial was stopped after enrollment of 293 patients because the mean number of ischemic episodes was only 0.74 per patient, and the proportion of patients with ischemia had fallen from 16% in the first third of enrollees to 7.5% in the last third. An additional reason for stopping was the lack of any benefit from active treatment.
Potential effects of estrogen in acute coronary syndromes
The investigators stop short of concluding that hormone replacement therapy is of no value in the treatment of unstable angina. Their reticence may stem from the lack of power of their trial or from the theoretical benefits that estrogen could confer in unstable angina. Intracoronary 17β-estradiol has been shown to potentiate the acetylcholine-induced increase in coronary blood flow in postmenopausal women through a nitric oxide–dependent mechanism (4)and to decrease coronary endothelin-1 levels (5). In postmenopausal women with coronary disease, either estrogen or 17β-estradiol has been reported to reduce demand ischemia induced by exercise (6), dobutamine (7)or atrial pacing (8). These beneficial effects of estrogen might not be manifest in unstable angina patients who receive IV nitroglycerin, because it acts through similar mechanisms.
Recurrent events after acute coronary syndromes depend partly upon the outcome of the struggle between thrombotic and fibrinolytic forces at the culprit lesion. Oral estrogen, with or without medroxyprogesterone, but not transdermal estradiol, enhances fibrinolysis (9). Conversely, oral, but not transdermal, estradiol shifts markers of coagulation toward hypercoagulability (10). Estrogen increases blood levels of matrix metalloproteinase-9, interleukin-6 and C-reactive protein (11). Matrix metalloproteinases are involved in plaque disruption, and inflammatory markers reflect an increased risk of recurrent events. The prothrombotic, fibrinolytic and pro-inflammatory effects of estrogen may not be relevant in unstable angina patients receiving antiplatelet and antithrombotic therapy, just as the vasodilatory effects may not be relevant in the presence of IV nitroglycerin.
Could estrogen be harmful in acute coronary syndromes?
The stated purpose in this trial (3)was to assess the possible benefit of estrogen or combined therapy in women with unstable angina, but of course the potential for harm was also being assessed. The prothrombotic tendencies of estrogen provide a theoretical basis for this concern and data from recent clinical trials are not reassuring. In the first year of treatment in the Heart and Estrogen/progestin Replacement Study (HERS), the incidence of coronary death and nonfatal MI was higher in the active treatment group: relative risk (RR) 1.52 (95% confidence interval [CI], 1.01 to 2.29) (12). Women with unstable angina were excluded from HERS, but the HERS patients were otherwise quite similar to the women in the current study.
The early increased hazard in HERS was unexpected and could have been a chance finding. However, a similar trend has been noted in the Women’s Health Initiative (13). This trial has randomized 16,609 women either to placebo or to combined hormone treatment, and 11,739 women who had had hysterectomies either to placebo or to unopposed estrogen. The Data and Safety Monitoring Board of this trial informed participants that “a small increase” in cardiovascular events had occurred during the first two years of treatment in women randomized to active therapy.
In the Coumadin Aspirin Reinfarction Study (CARS), 8,803 patients with recent MI were randomized in a 2 × 2 factorial design to one of two doses of aspirin and one of two doses of coumadin. Among the 1,857 postmenopausal women in CARS, 111 began taking estrogen or estrogen/progesterone during follow-up (14). The adjusted RR of death, new MI or unstable angina in these new users was 1.44 (95% CI, 1.05 to 1.99), strikingly similar to the increased RR in HERS. However, in contrast to HERS, the excess events were mostly episodes of unstable angina.
Coronary events would not be expected during short-term treatment periods in small numbers of postmenopausal women with stable coronary disease. However, 2 of 11 women were withdrawn from one such study when they developed unstable angina during estrogen/progestin therapy (6). In the context of these findings in previous studies, the higher frequency of ischemic episodes in women randomized to combination therapy in the present study should arouse suspicion, even though it is not statistically different from the frequencies in the other groups.
Other estrogen trials with cardiovascular end points
In the Estrogen Replacement and Atherosclerosis (ERA) trial, 309 women with coronary disease were randomized either to estrogen, estrogen plus medroxyprogesterone or placebo (15). Coronary arteriography was done at baseline and after a mean follow-up of 3.2 years, and progression of coronary atherosclerosis was assessed by quantitative methods. No differences were found among the three groups for changes in minimum lumen diameter—the primary end point—or for any of the other angiographic or clinical end points.
Two trials have assessed the effects of 17β-estradiol on carotid intimal media thickness. In the Estrogen in the Prevention of Atherosclerosis Trial (EPAT), 222 postmenopausal women were randomized to two years of treatment with either micronized 17β-estradiol or to placebo. Although the results have not yet been published, the 17β-estradiol group was reported to have improved carotid measurements (16). In contrast, the Postmenopausal Hormone Replacement against Atherosclerosis (PHOREA) trial found no effect of 17β-estradiol combined with either of two doses of a progestin on carotid intimal media thickness in a study of 321 postmenopausal women followed for one year (17).
The Papworth HRT and Atherosclerosis Survival Enquiry (PHASE) included 255 postmenopausal women with coronary disease, randomized to transdermal 17β-estradiol, with or without cyclic norethisterone (18). The primary end point, death, MI or hospitalization for unstable angina, occurred slightly more often in the hormone group (hazard ratio 1.23, 95% CI, 0.82 to 1.86). The trial was stopped early because of the likelihood that it would not show benefit.
Adverse effects of hormone replacement
In both observational studies (19–21)and clinical trials (22), estrogen use in postmenopausal women increased the risk of venous thromboembolic disease. Hormone replacement should be discontinued in women who are immobilized by fractures, surgery or stroke. Estrogen replacement increases the risk of gall bladder disease (23). Use for longer than five years increases the risk of breast cancer (24).
Unopposed estrogen causes breakthrough bleeding and endometrial hyperplasia in the older postmenopausal age group where coronary disease is most prevalent (15), and it is not practical long-term therapy, even with frequent monitoring to detect endometrial cancer. Hormone replacement is extremely useful to control menopausal symptoms; however, older women usually do not appreciate the breast tenderness and bleeding that often occurs with initiation of therapy. Even among the selected women who participate in clinical trials, maintaining compliance with this treatment is difficult; for example, in HERS the proportion of women who discontinued hormone treatment was 18% at one year and 25% at three years (12).
Conclusions and clinical implications
The present study by Schulman et al. (3), and many of the other studies discussed above, lack sufficient power to be conclusive. Diehard estrogen advocates will continue to argue that the problems of hormone replacement relate to progesterone, or to the type of estrogen, the dose or the mode of delivery. Future clinical trials may prove that they are correct. In the meantime, based upon the currently available information, what are we to do?
The recommendation of the HERS investigators (12), that hormone replacement therapy not be begun for the purpose of secondary prevention of coronary disease, is becoming more widely accepted. Should we go further? Based upon the suspicion of potential risk, and the documented absence of any benefit, we should consider discontinuing hormone replacement in women with acute coronary events, particularly if it had been begun within the previous year, unless a compelling reason exists to continue treatment. Many other drugs improve the outcome of patients after an acute coronary event; thus, we should focus on compliance with treatment of proven benefit.
In an editorial published in 1996, Jacques Rossouw noted that two-thirds of physicians reported that they prescribed hormone replacement therapy for coronary heart disease despite the absence of conclusive evidence of benefit (25). Rossouw (25)recommended “putting the brakes on the bandwagon.” The clinical trial data that have accumulated since then have vindicated his point of view: the bandwagon is now bedraggled and the music has died. Yet hormone replacement is still prescribed by some physicians with the expectation that it will prevent coronary events. The greatest advance in our understanding of how estrogen prevents heart disease may turn out to be our realization that it does not.
↵∗ Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology
- National Center for Health Statistics
- Schulman S.P.,
- Thiemann D.R.,
- Ouyang P.,
- et al.
- Guetta V.,
- Quyyumi A.A.,
- Prasad A.,
- Panza J.A.,
- Waclawiw M.,
- Cannon R.O.
- Webb C.M.,
- Ghatei M.A.,
- McNeill J.G.,
- Collins P.
- Rosano G.M.C.,
- Webb C.M.,
- Chierchia S.,
- et al.
- Aspaslan M.,
- Shimokawa H.,
- Kuroiwa-Matsumoto M.,
- Harasawa Y.,
- Takeshita A.
- Rosano G.M.C.,
- Caixeta A.M.,
- Chierchia S.,
- et al.
- Zanger D.,
- Yang B.K.,
- Ardans J.,
- et al.
- ↵Lenfant C. Statement from Claude Lenfant, MD, NHLBI Director, on preliminary trends in the Women’s Health Initiative. Washington: NHLBI Communications Office, April 3, 2000
- Alexander K.P.,
- Newby L.K.,
- Hellkamp A.S.,
- et al.
- Hodis H.N.,
- Mack W.J.,
- Lobo R.A.,
- et al.
- Angerer P.,
- Störk S.,
- Kothny W.,
- Schmitt P.,
- von Schacky C.
- Clarke S.,
- Kelleher J.,
- Lloyd-Jones H.,
- Sharples L.,
- Slack M.,
- Schofield P.M.
- Rossouw J.E.