(Circulation. 2000;102:2228.)
© 2000 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Department of Medicine (J.H.), George Washington University, Washington, DC; the Departments of General Internal Medicine, San Francisco VA Medical Center (J.A.S.) and Epidemiology and Biostatistics (J.A.S., F.L.), University of California, San Francisco; the Department of Medicine (W.B.A.), Wake Forest University, Winston-Salem, NC; the Department of Orthopaedic Surgery (M.T.V.), University of Pittsburgh, Pittsburgh, Pa; the Heart Disease Prevention Clinic (D.H.), University of Minnesota, Minneapolis; and Wyeth-Ayerst Research (M.C.), Radnor, Pa.
Correspondence to Dr Judith Hsia, 2150 Pennsylvania Ave NW #4-414, Washington, DC 20037. E-mail domjah{at}gwumc.edu
| Abstract |
|---|
|
|
|---|
Methods and ResultsPostmenopausal women with coronary heart disease (CHD) and an intact uterus (n=2763) were randomly assigned to conjugated equine estrogens (0.625 mg) combined with medroxyprogesterone acetate (2.5 mg) daily or to placebo in a secondary CHD prevention trial. This analysis focuses on incident peripheral arterial procedures and deaths in the 2 treatment groups; peripheral vascular disease was a predefined secondary outcome. During a mean of 4.1 years of follow-up, 311 peripheral arterial events were reported in 213 women, an annual incidence of 2.9%. The number of women who had peripheral arterial events was 99 among those assigned to active estrogen/progestin and 114 among those assigned to placebo, a nonsignificant difference (relative hazard 0.87, 95% CI 0.66 to 1.14). In the placebo group, hypertension and diabetes mellitus were independently associated with higher rates of peripheral arterial events, and plasma HDL cholesterol and body mass index were associated with lower rates of peripheral arterial events. In the estrogen/progestin group, current smoking and diabetes were independent predictors of peripheral arterial events. Incident peripheral arterial disease was not a significant predictor of coronary, cardiovascular, or total mortality.
ConclusionsTreatment with oral conjugated estrogen plus medroxyprogesterone acetate was not associated with a significant reduction in incident peripheral arterial events in postmenopausal women with preexisting CHD.
Key Words: hormones peripheral vascular disease coronary disease women
| Introduction |
|---|
|
|
|---|
Pharmacological treatment for symptomatic peripheral arterial disease is limited and includes pentoxifylline, clopidogrel, and cilostazol. In recent trials, cilostazol reduced symptoms of intermittent claudication,12 whereas pentoxifylline was not significantly better than placebo.13 Clopidogrel has not been shown to affect claudication and did reduce the risk of cardiovascular events (myocardial infarction, stroke, and vascular death) in patients with symptomatic peripheral arterial disease.14
Although estrogen therapy in postmenopausal women has been associated with reduced carotid atherosclerosis in observational studies,15 16 the effect of estrogen on other peripheral arterial beds has not been evaluated. This report examines the effect of combined continuous conjugated estrogen and medroxyprogesterone acetate on incident peripheral arterial disease over a mean of 4.1 years among 2763 elderly postmenopausal women enrolled in the Heart and Estrogen/Progestin Replacement Study (HERS),17 a secondary coronary heart disease (CHD) prevention trial.
| Methods |
|---|
|
|
|---|
50%) were randomly assigned to treatment with
conjugated estrogens (0.625 mg) plus
medroxyprogesterone acetate (2.5 mg) daily or
treatment with placebo. The protocol and consent forms were approved by
each clinical sites institutional review board; study participants
provided informed consent before enrollment. Participants completed
baseline questionnaires that ascertained demographic characteristics,
risk factors, use of medications, and general health history, including
self-reported diabetes, smoking, and alcohol consumption. Data on use
of lipid-lowering medications, aspirin, and antioxidant supplements
(vitamin C, vitamin E, ß-carotene, and selenium) were also collected.
Body mass index (kg/m2) was calculated from
height and weight measurements performed at baseline. All participants
provided a fasting blood sample for serum chemistries that included
lipid and lipoprotein measurements. Baseline characteristics of the
placebo and estrogen/progestin treatment groups did not differ
significantly.17 Women were followed for hospitalizations, including admissions for peripheral arterial procedures at 4-month intervals. Events were confirmed by review of supporting documentation by coordinating center adjudicators blinded to treatment assignment. Peripheral arterial disease was a predefined secondary outcome of HERS.
Peripheral arterial events were categorized as aortic, carotid, cerebrovascular, upper/lower extremity, or mesenteric/renal. Aortic events included abdominal aneurysm repair (n=15), aortic endarterectomy, (n=1) angioplasty (n=3), stent (n=2), and arch replacement (n=1). Carotid events included endarterectomy (n=116), angioplasty (n=2), and stent (n=1). Cerebral vascular events included aneurysm clipping (n=3), aneurysm embolization (n=1), acute retinal artery occlusion (n=1), and middle cerebral artery angioplasty (n=1). Events affecting the extremities included amputations (n=9 each for above and below the knee amputation and n=4 for toe amputation), percutaneous revascularization of the lower extremity (n=62), surgical revascularization of the lower extremity (n=55), and percutaneous subclavian angioplasty and stenting (n=4). Mesenteric/renal vascular events included renal arterial revascularization procedures (n=20) and fatal attempted mesenteric revascularization (n=1).
Women were not specifically asked about claudication in HERS but volunteered this diagnosis when asked about new medical conditions. Consequently, claudication was reported by only 10 participants: 6 assigned to estrogen/progestin and 4 to placebo. Five women (3 in the active treatment group and 2 assigned to placebo) went on to lower extremity revascularization. Because claudication was so infrequently reported, it was not included as an indicator of peripheral arterial disease in this analysis.
Statistical Analysis
All variables were treated categorically except age at
randomization, plasma lipid and lipoprotein levels, and body mass
index. For dichotomous variables, the referents were as follows:
being less than a high school graduate (for the education
variable); having no hypertension or diabetes mellitus; being a
nonsmoker at baseline; using no aspirin, HMG-CoA reductase
inhibitor, or antioxidant at baseline; and randomization to
placebo. We used unpaired 2-tailed t tests to compare
continuous variables and
2 tests to
compare categorical variables. Treatment group was based on
randomization assignment.
At the end of the first year, 82% of women in the hormone group and
91% in the placebo group reported taking study medication; at the end
of year 3, these proportions had fallen to 75% and
81%.18 In Table 2
, multivariate
Cox models were used to determine relative risk and included all the
variables listed in Table 1
except for active
estrogen/progestin treatment. In Table 3
, 2 separate Cox models
included all variables in Table 2
. SAS version 6.12 system
software was used for all analyses.
|
|
|
| Results |
|---|
|
|
|---|
During 10 681 woman-years of follow up, 311 peripheral
arterial events were reported in 213 subjects (Table 2
), representing an annual
incidence of 2.9%. The proportion of peripheral
arterial procedures and deaths was similar among women
assigned to placebo and to estrogen/progestin after adjustment for age
alone or for multiple variables. HERS reported more
coronary events during the first year of treatment among women
assigned to active estrogen/progestin.18 During the 4
months after randomization, 8 peripheral vascular events
were reported in women assigned to active treatment, and 7 events were
reported in women assigned to placebo. During the 12 months after
randomization, 36 peripheral vascular events were reported
in the active treatment group, and 32 events were reported in the
placebo group (P=0.3).
Table 3
examines the determinants of
incident peripheral arterial events within the
treatment group. In multivariate analyses,
independent predictors of peripheral arterial
events in the placebo group included self-reported hypertension
(P<0.001) and diabetes mellitus (P=0.003);
negative predictors included HDL cholesterol
(P=0.02) and body mass index (P=0.002). Among
women randomized to active treatment with estrogen/progestin,
independent predictors of peripheral arterial
events included diabetes mellitus (P=0.004) and history of
smoking at baseline (P<0.001).
Incident peripheral arterial disease was not a statistically significant determinant of cardiovascular death, coronary death, or total mortality (data not shown).
| Discussion |
|---|
|
|
|---|
The incidence of peripheral vascular events in women with CHD has not been previously described. In 1989, the annual age- and sex-adjusted rate of lower extremity procedures (peripheral angioplasty, bypass surgery, and amputation) in the state of Maryland was 119 per 100 000.19 In contrast, the annual rate of lower extremity procedures in the HERS cohort was 1301 per 100 000. Thus, individuals with preexisting coronary atherosclerosis appear to have a roughly 10-fold increased risk of lower extremity arterial disease requiring revascularization or amputation.
Reported risk factors for peripheral arterial disease include cigarette smoking, hypertension, diabetes mellitus, low plasma HDL cholesterol, and elevated LDL cholesterol and triglycerides.20 21 This analysis differs from previous descriptions of factors contributing to peripheral arterial disease in several ways. The cohort includes only women with known CHD, and the analysis defines peripheral arterial events as vascular procedures rather than intermittent claudication or noninvasive measures of peripheral arterial atherosclerosis. Peripheral arterial events in HERS represent a mixture of thrombotic and atherosclerotic events; consequently, some variation in identifiable risk factors is to be expected. The observation that plasma lipids, except for triglycerides, were not associated with peripheral arterial events is presumably due to confounding and the fact that all women in the present study had advanced atherosclerosis; 36% of women were taking HMG-CoA reductase inhibitors at the time of baseline lipid measurement, which multivariate adjustment can only partly address. Furthermore, results of lipid profiles were provided to womens usual care providers, some of whom subsequently initiated lipid-lowering medications. Overall, 903 (65%) of 1383 women randomized to placebo and 809 (59%) of 1380 women randomized to estrogen/progestin used HMG-CoA reductase inhibitors at some time during the trial.
A second unexpected observation is the inverse association between body
mass index and risk of peripheral arterial
events in this analysis; ie, body mass index was lower among
women with peripheral vascular events. The basis for this
inverse relationship is somewhat mysterious, particularly because
atherosclerotic processes generally are more pronounced in obese
individuals.22 In earlier reports of risk factors for
peripheral arterial disease,20 21
which did not identify body mass index as a risk factor, mean body mass
index was
25 kg/m2. In contrast, mean body
mass index of HERS women exceeded 28 kg/m2. One
possibility is that higher estrogen levels in obese women may provide
some protection during the 40 years preceding menopause. Inasmuch as
the longitudinal weight history and serum and tissue estrogen levels
are not known in HERS, this remains a speculative hypothesis. Another
possible explanation is residual confounding from cigarette
smoking.
Independent determinants of peripheral arterial events differed in the placebo and estrogen/progestin groups. In the placebo group, but not the active treatment group, plasma HDL cholesterol and body mass index were associated with lower risk of peripheral arterial events, possible because estrogen raises HDL cholesterol23 and women with higher body mass indexes generally have higher endogenous estrogen levels.24 Estrogen/progestin appeared to attenuate the risk associated with hypertension, but the interaction was not statistically significant. Although estrogen affects vasomotor function,25 it did not alter blood pressure in the Postmenopausal Estrogen/Progestin Interventions trial,23 consistent with our findings. Estrogen/progestin did potentiate the adverse impact of smoking on peripheral arterial events (P=0.03 for the interaction term). This relationship between smoking and exogenous estrogen may parallel the observation that myocardial infarction is 10 times more common among oral contraceptive users who smoke compared with those who do not26 and provides an additional argument against smoking for women taking hormone replacement.
Plasma LDL cholesterol at baseline did not independently predict peripheral arterial events in either the placebo or estrogen/progestin groups. In the placebo group, 37% took HMG-CoA reductase inhibitors at baseline, and 28% began therapy during the trial; in the estrogen/progestin group, 35% took HMG-CoA reductase inhibitors at baseline, and 23% started therapy during the trial. Women randomized to placebo were significantly more likely (P<0.001) to have taken HMG-CoA reductase inhibitors during the trial, presumably at least in part because they were not receiving the LDL-lowering benefit from estrogen/progestin. This difference in use of concurrent lipid-lowering medication may have contributed to the apparent inability of estrogen/progestin to protect against peripheral arterial events.
Potential limitations of the present study design include sample size, choice of estrogen/progestin regimen, confounding due to prevalent use of lipid-lowering medications (particularly among women randomized to placebo), and the inclusion only of women with known CHD. The sample size for HERS was chosen to address the primary CHD outcome but not the therapeutic effect of estrogen/progestin on peripheral arterial disease. Thus, it is conceivable that a larger study might have demonstrated benefit, although the 95% CIs for treatment effect on total peripheral arterial events suggest that a considerably larger sample would be required, and the point estimate of the relative hazard (0.87) suggests that the benefit would be modest. We cannot exclude the possibility that a different hormone replacement regimen might produce different results or the possibility that this regimen might be efficacious in healthy women. The Womens Health Initiative,27 which has randomized 27 348 predominantly healthy postmenopausal women in an exogenous hormone trial, will examine the incidence of peripheral vascular outcomes categorized as carotid, aortic, and lower extremity atherosclerotic disease requiring hospitalization or intervention. Thus, by 2005, additional data regarding the efficacy of estrogen replacement for primary prevention of peripheral arterial disease should become available.
Our finding that daily use of conjugated estrogens with medroxyprogesterone did not significantly reduce the risk of peripheral arterial events is consistent with the primary HERS outcome that estrogen/progestin did not reduce coronary events in this cohort of elderly women. Taken together, these observations do not support use of this regimen for the purpose of preventing coronary or peripheral vascular events in women with known CHD.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received March 27, 2000; revision received June 13, 2000; accepted June 14, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
V. M. Miller and S. P. Duckles Vascular Actions of Estrogens: Functional Implications Pharmacol. Rev., June 1, 2008; 60(2): 210 - 241. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. D. Pradhan, S. Shrivastava, N. R. Cook, N. Rifai, M. A. Creager, and P. M Ridker Symptomatic Peripheral Arterial Disease in Women: Nontraditional Biomarkers of Elevated Risk Circulation, February 12, 2008; 117(6): 823 - 831. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Voci, F. Pizzuto, and F. Romeo Coronary flow: a new asset for the echo lab? Eur. Heart J., November 1, 2004; 25(21): 1867 - 1879. [Full Text] [PDF] |
||||
![]() |
R. Hultgren, P. Olofsson, and E. Wahlberg Reproductive History in Women with Lower Limb Ischemia Angiology, July 1, 2004; 55(4): 373 - 383. [Abstract] [PDF] |
||||
![]() |
A. M. O'Hare, E. Vittinghoff, J. Hsia, and M. G. Shlipak Renal Insufficiency and the Risk of Lower Extremity Peripheral Arterial Disease: Results from the Heart and Estrogen/Progestin Replacement Study (HERS) J. Am. Soc. Nephrol., April 1, 2004; 15(4): 1046 - 1051. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Hsia, M. H. Criqui, R. J. Rodabough, R. D. Langer, H. E. Resnick, L. S. Phillips, M. Allison, D. E. Bonds, K. Masaki, P. Caralis, et al. Estrogen Plus Progestin and the Risk of Peripheral Arterial Disease: The Women's Health Initiative Circulation, February 10, 2004; 109(5): 620 - 626. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. P. Korovkina, A. M. Brainard, P. Ismail, T. J. Schmidt, and S. K. England Estradiol Binding to Maxi-K Channels Induces Their Down-regulation via Proteasomal Degradation J. Biol. Chem., January 9, 2004; 279(2): 1217 - 1223. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Barrett-Connor An Epidemiologist Looks at Hormones and Heart Disease in Women J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4031 - 4042. [Full Text] [PDF] |
||||
![]() |
D. Grady, D. Herrington, V. Bittner, R. Blumenthal, M. Davidson, M. Hlatky, J. Hsia, S. Hulley, A. Herd, S. Khan, et al. Cardiovascular Disease Outcomes During 6.8 Years of Hormone Therapy: Heart and Estrogen/Progestin Replacement Study Follow-up (HERS II) JAMA, July 3, 2002; 288(1): 49 - 57. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Hlatky, D. Boothroyd, E. Vittinghoff, P. Sharp, M. A. Whooley, and for the HERS Research Group Quality-of-Life and Depressive Symptoms in Postmenopausal Women After Receiving Hormone Therapy: Results From the Heart and Estrogen/Progestin Replacement Study (HERS) Trial JAMA, February 6, 2002; 287(5): 591 - 597. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Voci and F. Pizzuto Coronary flow: how far can we go with echocardiography? J. Am. Coll. Cardiol., December 1, 2001; 38(7): 1885 - 1887. [Full Text] [PDF] |
||||
![]() |
W. R. Hiatt Medical Treatment of Peripheral Arterial Disease and Claudication N. Engl. J. Med., May 24, 2001; 344(21): 1608 - 1621. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |