(Circulation. 1995;91:1952-1958.)
© 1995 American Heart Association, Inc.
Articles |
From the Institute for Prevention of Cardiovascular Disease, Cardiovascular Division, Deaconess Hospital, and Harvard Medical School, Boston, Mass (O.C.E.G., M.A.M., P.S., I.L., T.M., P.X., F.W., J.E.M., G.H.T.), and the Framingham Heart Study, Framingham, Mass (P.W.F.W., D.L.).
Correspondence to Geoffrey H. Tofler, MD, Institute for Prevention of Cardiovascular Disease, Deaconess Hospital, One Autumn St, 5th Floor, Boston, MA 02215.
| Abstract |
|---|
|
|
|---|
Methods and Results We determined levels of plasminogen activator inhibitor (PAI-1) antigen and tissue plasminogen activator (TPA) antigen in 1431 subjects from the Framingham Offspring Study. Fibrinolytic potential was compared between subjects with high estrogen status (premenopausal women and postmenopausal women receiving hormone replacement therapy) and low estrogen status (men and postmenopausal women not receiving hormone replacement therapy). In all comparisons, subjects with high estrogen status had greater fibrinolytic potential (lower PAI-1 levels) than subjects with low estrogen status. First, postmenopausal women receiving estrogen replacement therapy had lower levels of PAI-1 than those not receiving therapy (13.0±0.5 versus 19.5±1.0 ng/mL, P<.001). Second, premenopausal women had lower levels of PAI-1 than men of a similar age (14.8±0.6 versus 20.3±0.8 ng/mL, P<.001); this sex difference diminished when postmenopausal women not receiving hormone replacement therapy were compared with men of a similar age (19.6±0.7 versus 21.1±0.7 ng/mL, P=.089). Third, premenopausal women had markedly lower levels of PAI-1 antigen than postmenopausal women not receiving estrogen therapy (14.8±0.6 versus 19.5±1.0 ng/mL, P<.001). The between-group differences observed for TPA antigen were similar to those for PAI-1 antigen.
Conclusions Each of these comparisons indicates that the cardioprotective effect of estrogen may be mediated, in part, by an increase in fibrinolytic potential. These findings might provide at least a partial explanation for the protection against cardiovascular disease experienced by premenopausal women, and the loss of that protection following menopause.
Key Words: hormones fibrinolysis cardiovascular diseases
| Introduction |
|---|
|
|
|---|
Given the importance of thrombosis in causation of acute coronary syndromes, it is also possible that estrogen exerts a beneficial effect on risk by lowering thrombotic tendency. Recent reports demonstrated that high estrogen status is associated with a favorable decrease in fibrinogen and factors promoting coagulation,15 16 but the effect on the countervailing portion of the hemostatic balance, the fibrinolytic system, has not been clarified. The importance of determining the effect of estrogen on the fibrinolytic system is increased by recent studies demonstrating that impaired fibrinolytic capacity is associated with increased cardiovascular risk.17 18 19 20 Most studies of fibrinolytic capacity have measured levels of plasminogen activator inhibitor (PAI-1),21 22 23 24 25 a plasma protein controlling activity of tissue plasminogen activator (TPA) and the fibrinolytic system. High levels of PAI-121 22 23 24 25 26 or TPA antigen27 28 have been associated with increased risk of coronary artery disease. While high levels of TPA activity indicate increased fibrinolytic potential, antigen levels reflect both free TPA and TPA bound to PAI-1.
We measured PAI-1 and TPA antigen levels in 1431 participants in the Framingham Offspring Study to examine the relation between estrogen status and fibrinolytic potential.
| Methods |
|---|
|
|
|---|
Data were collected from
each subject during a visit to the Framingham
Heart Study clinic. Menopausal status was determined by a standardized
medical history questionnaire administered by the examining physician.
Menopause was defined as the cessation of menses for at least 1 year.
The postmenopausal women were further classified according to whether
menopause was natural or surgically induced (bilateral oophorectomy).
Women who had undergone hysterectomy but who had at least one ovary
intact and were
51 years of age were classified as having undergone
natural menopause. The use of hormone replacement therapy (estrogen
with or without progestin) or oral contraceptives was ascertained by
the examining physician. Individuals who regularly smoked at least one
cigarette per day during the year before the index examination were
classified as current smokers. Alcohol consumption was assessed as the
reported number of drinks per day of beer (12 oz), wine (4 oz), and
spirits (1 oz). Body mass index was computed by dividing the weight
(kilograms) by the square of the height (meters).
Subjects were excluded from the analysis for the following reasons: (1) history of coronary artery disease or stroke, as previously described29 (n=127), (2) current use of oral contraceptives (n=10), (3) missing information regarding menopausal status or hormone therapy (n=22), and (4) missing data for PAI-1 or TPA (n=34). After these exclusions, a total of 1431 subjects (749 women, 682 men) met entry criteria and form the basis for the present report.
Blood Sampling and Analysis
Blood samples were collected from
an antecubital vein between
8:00 and 9:00 AM, with subjects in the supine position
after an overnight fast. For determination of plasma levels of PAI-1
and TPA antigen, blood was anticoagulated with 3.8% trisodium citrate
(9:1, vol/vol) and kept on crushed ice until centrifugation. Plasma was
separated by centrifugation at 2500g for 30 minutes at
4°C. Plasma aliquots were quickly frozen and stored at -70°C for
subsequent analysis. PAI-1 antigen levels were determined by a
commercially available sandwich enzyme-linked immunosorbent assay
according to the description of Declerck et al30
(TintElize PAI-1, Biopool AB). Levels of TPA antigen also were obtained
using an enzyme-linked immunosorbent assay (TintElize TPA, Biopool),
following a procedure described by Ranby et al.31 The
intra-assay coefficient of variation in our laboratory was 8.1% for
PAI-1 and 5.5% for TPA.
Lipid Analysis
For determination of lipids, blood was
anticoagulated with EDTA
at a final concentration of 1 mg/mL. Plasma was separated by
centrifugation at 2500g for 30 minutes at 4°C, and lipid
measurements were made in fresh specimens. HDL cholesterol was measured
after precipitation of LDL and very- low-density lipoprotein (VLDL)
cholesterol with dextran-magnesium.32 Plasma levels of
total cholesterol, HDL cholesterol, and triglycerides were measured by
automated enzymatic methods with an Abbot Diagnostics ABA-200
bichromatic analyzer and Abbot A-Gent enzymatic
reagents.33 The level of LDL cholesterol was calculated
with the Friedewald equation34 in all cases with
triglyceride levels <500 mg/dL. The laboratory participates in the
Centers for Disease Control (Atlanta, Ga) lipid standardization
program.
Statistical Analysis
For variables that were not normally
distributed (PAI-1, TPA,
and triglyceride levels), logarithmic transformation was performed and
geometric means were presented. Linear regression models were used
to evaluate PAI-1 levels for the following comparisons: (1) hormone
users versus nonusers among the postmenopausal women, (2) men versus
women, (3) premenopausal women versus postmenopausal women not
receiving hormone therapy, and (4) users of estrogen alone versus users
of estrogen-progestin combinations. For a comparison between sexes, the
male population was divided into individuals <50 and
50 years of age
since 50 was the median age of menopause of the women in the
study.
Results are presented first as unadjusted comparisons between
groups and second, after adjustment for the following covariates: age,
body mass index, number of alcohol-containing drinks per day, use of
antihypertensive medication, diabetes mellitus, and smoking. Adjustment
for age was included in all comparisons except for the comparison
between premenopausal and postmenopausal women because in this
comparison, there was little overlap in the age distributions. Since
increased plasma triglyceride levels have been associated previously
with increased PAI-1 levels,35 36 a final analysis
was
performed adjusting for triglyceride levels in addition to the other
covariates. Baseline clinical characteristics were compared using the
unpaired t test and Pearson's
2 test
for continuous and discrete variables, respectively. All data are
presented as mean±SEM. Two-tailed P values <.05 were
considered statistically significant.
| Results |
|---|
|
|
|---|
|
Estrogen Use Versus NonEstrogen Use in Postmenopausal Women
Fig 1
shows the PAI-1 antigen levels for
postmenopausal women stratified by estrogen therapy. The women
receiving therapy had 35% lower levels of PAI-1 than postmenopausal
women not receiving hormones, with an unadjusted difference of 6.7
ng/mL (95% confidence interval from 4.8 to 8.5 ng/mL,
P<.001) and a difference of 3.7 ng/mL (95% confidence
interval from 1.9 to 5.5 ng/mL, P<.001) after adjustment
for the covariates.
|
Women receiving estrogen alone and those receiving
an
estrogen-progestin combination did not differ with regard to baseline
characteristics except for type of menopause (Table 2
).
Unadjusted PAI-1 levels also did not differ between the two groups
(12.1±1.0 versus 14.3±1.8 ng/mL, estrogen alone and combined
therapy,
respectively, P=.241), although the difference between the
two groups was of borderline significance after adjustment for the
covariates (11.3±1.2 versus 15.4±2.1 ng/mL, estrogen and
combined
therapy, respectively, P=.041), suggesting that combination
therapy may not be associated with as great an increase in fibrinolytic
potential as estrogen therapy alone.
|
Comparison of Women Versus Men
Table 3
shows
the differences in clinical
characteristics between men and women of similar age. Premenopausal
women had a lower mean level of PAI-1 than men of comparable age
(14.8±0.6 versus 20.3±0.8 ng/mL unadjusted, P<.001,
and
14.1±0.5 versus 17.1±0.8 ng/mL after adjustment,
P=.001).
In contrast to the findings in the younger age group, PAI-1 levels in
postmenopausal women not receiving hormone replacement therapy did not
differ from those observed in older men (19.6±0.7 versus
21.1±0.7
ng/mL unadjusted, P=.089, and 17.1±0.8 versus
17.1±0.6
ng/mL adjusted, P=.678).
|
Premenopausal Women Versus Postmenopausal Women
While the
PAI-1 antigen levels were not significantly different
between the younger and older men, premenopausal women had 25% lower
levels than postmenopausal women not receiving hormones, with an
average unadjusted difference of 4.8 ng/mL (95% confidence interval
from 3.6 to 5.9 ng/mL, P<.001) (Fig 2
). This
difference was attenuated to 2.9 ng/mL (95% confidence interval from
1.8 to 4.0 ng/mL) after adjustment for the covariates
(P<.001). PAI-1 levels were similar for women who had
natural or surgical menopause (19.6±0.8 versus 21.2±1.4 ng/mL,
respectively, P=.324).
|
TPA Antigen
The between-group differences observed for TPA
antigen were
similar to those for PAI-1 antigen (Table 4
).
Postmenopausal women receiving hormones had lower levels than those not
receiving hormones. Premenopausal women had lower levels of TPA than
men of similar age. Premenopausal women had lower levels of TPA than
postmenopausal women not receiving hormones. Adjustment for the
covariates did not account for the differences between these groups.
Although postmenopausal women had lower unadjusted levels of TPA
antigen than men in the same age group (P=.001), this
difference became nonsignificant after adjustment for covariates
(P=.104). Postmenopausal women receiving estrogen alone did
not differ from those receiving combined estrogen-progestin (5.9±0.4
versus 5.9±0.5 ng/mL, respectively, P=.964). TPA
antigen
levels were significantly correlated with PAI-1 levels
(r=.66, P<.001).
|
Plasma triglyceride levels
were significantly correlated with PAI-1
(r=.47, P<.001) and TPA antigen
(r=.49, P<.001) levels; however, addition of
plasma triglyceride levels to the regression model did not
significantly change the results for either PAI-1 or TPA antigen. Fig
3
summarizes findings that subjects with high estrogen
status had higher fibrinolytic potential (lower PAI-1 levels) than
subjects with low estrogen status.
|
| Discussion |
|---|
|
|
|---|
Estrogen and Coronary Heart Disease
Epidemiological data
suggest that postmenopausal women receiving
estrogen replacement therapy experience a cardiovascular risk reduction
of approximately 40% to
50%.3 4 5 6 7 38
Potential mechanisms
for this marked clinical benefit include a favorable reduction in total
and LDL cholesterol, an increase of HDL
cholesterol,8 9 10
and a decrease in LDL uptake by the arterial wall.14
Estrogen also has beneficial effects on carbohydrate metabolism,
decreasing glucose and insulin levels and improving body fat
distribution.11 12 In addition, a reduction of plasma
fibrinogen levels15 and improvement of
endothelium-dependent vasomotion via a direct vascular
effect4 13 may lessen the risk of thrombosis and
coronary
occlusion. However, these multiple potential mechanisms do not fully
explain the marked protective effect associated with high estrogen
levels.
Impaired Fibrinolysis and Coronary Heart Disease
The
observation that coronary artery thrombosis and myocardial
infarction often occur in the absence of severe coronary stenosis has
stimulated interest in the role of impaired fibrinolysis in onset of
coronary heart disease.39 The activity of the endogenous
fibrinolytic system is dependent on the balance between plasminogen
activators (mainly TPA and urokinase-type plasminogen activators) and
plasminogen activator inhibitors, of which PAI-1 is considered the most
important. Since most of the TPA antigen measured is complexed with
PAI-1 and is inactive, high levels of TPA antigen, in association with
high levels of PAI-1, reflect impairment of the fibrinolytic
system.40 41 42 43
Elevated PAI-1 levels are associated with thromboembolic disease, although a cause-and-effect relation remains to be proven. Most but not all35 44 45 cross-sectional studies of patients with angina pectoris or previous myocardial infarction have demonstrated decreased fibrinolytic activity in patients compared with control subjects.17 18 19 20 40 Increased levels of PAI-1 have been found in patients with angina and myocardial infarction.21 22 23 24 Additionally, there is a temporal association between the increased morning risk of myocardial infarction46 and high PAI-1 levels.47 Supportive prospective data that increased plasma PAI-1 levels are important in the development of myocardial infarction come from a study in which high PAI-1 levels after infarction were associated with an increased risk of reinfarction.25 In addition to these findings for PAI-1, recent evidence links elevated levels of TPA antigen to increased risk of future myocardial infarction in asymptomatic men27 and increased mortality in patients with known coronary artery disease.28
Estrogen and Fibrinolysis
Our study shows that the presence
of estrogen, either naturally in
premenopausal women or as a result of replacement therapy in
postmenopausal women, was associated with lower levels of PAI-1 and TPA
antigen compared with those observed in subjects expected to have low
estrogen levels. Prior studies of small numbers of postmenopausal women
receiving estrogen48 49 or a combination of estrogen
and
progestin16 have demonstrated improvement in fibrinolytic
potential consistent with our findings. The present report is the
first, to our knowledge, to examine the relation between estrogen
status and fibrinolytic potential in a comprehensive manner in a large
population-based study.
Activation of the coagulation system has been reported in postmenopausal women receiving estrogen therapy50 and in premenopausal women using oral contraceptives,51 suggesting that an increase in fibrinolytic potential may counterbalance a harmful effect on coagulation. Findings that current estrogen users had a greater cardiovascular risk reduction than former users38 are consistent with an acute effect of estrogen on coagulation, in addition to a possible long-term effect on atherosclerosis.
The mechanism by which estrogen may decrease PAI-1 is not clear. Despite the previously described correlation between PAI-1 and triglyceride levels,35 36 lowering of triglyceride levels cannot explain the effect of estrogen on PAI-1 because the women receiving estrogen therapy had higher triglyceride levels after age adjustment but lower levels of PAI-1. Estrogen may directly decrease PAI-1 biosynthesis and secretion, since the production of PAI-1 is influenced by several hormones.52 However, incubation of endothelial cells with estrogen did not decrease PAI-1 production.51 53 Our finding that menopausal women receiving estrogen alone tended to have lower levels of PAI-1 than women receiving combined estrogen-progestin suggests that progestin may diminish the beneficial effect of estrogen on fibrinolysis. However, the number of women available for this comparison in our study was small, and further investigation on this issue is necessary.
Study Limitations
A limitation of this cross-sectional study
is that confounding
caused by unrecognized factors may remain despite statistical
adjustment for known confounders. Randomized clinical trials, such as
the Women's Health Initiative, in which women are randomly assigned to
estrogen replacement or placebo, are required to demonstrate with
certainty that estrogen therapy decreases PAI-1 levels and that this
effect translates into decreased cardiovascular risk. Since studies of
the role of PAI-1 in the development of coronary heart disease have
been conducted almost exclusively in men, the importance of PAI-1 as a
potential risk factor in women requires further investigation. Our
sample size, while adequate for the primary analysis, was too small
to analyze the effect of different types, dosage, and duration of use
of hormone preparations on PAI-1 levels. As noted, we were unable to
standardize the timing of blood sampling during the menstrual cycle for
the premenopausal women or the pill-taking cycle for the postmenopausal
women, and menopausal status was determined by history rather than by
measurement of hormonal levels. However, any cycle-related variability
would tend to increase the observed variability within groups and
decrease the power to detect a true difference between groups.
Misclassification of menopausal status would have a similar effect.
Conclusions
Our finding that high estrogen status is
associated with increased
fibrinolytic potential may provide a partial explanation for the
cardioprotective effects of estrogen. In addition, the finding provides
a stimulus to the design of agents that can reduce PAI-1 without
causing other undesirable estrogen effects. This possibility may lead
to new approaches to reduce PAI-1 that may be of value not only for
postmenopausal women but for men as well.
| Acknowledgments |
|---|
Received July 21, 1994; revision received November 1, 1994; accepted November 14, 1994.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M.S.V. Elkind, S. Prabhakaran, J. Pittman, W. Koroshetz, M. Jacoby, K. C. Johnston, and for the GAIN Americas Investigators Sex as a predictor of outcomes in patients treated with thrombolysis for acute stroke Neurology, March 13, 2007; 68(11): 842 - 848. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. D. Bushnell, P. Hurn, C. Colton, V. M. Miller, G. del Zoppo, M. S.V. Elkind, B. Stern, D. Herrington, G. Ford-Lynch, P. Gorelick, et al. Advancing the Study of Stroke in Women: Summary and Recommendations for Future Research From an NINDS-Sponsored Multidisciplinary Working Group Stroke, September 1, 2006; 37(9): 2387 - 2399. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. K. Koh and B.-K. Yoon Controversies regarding hormone therapy: Insights from inflammation and hemostasis Cardiovasc Res, April 1, 2006; 70(1): 22 - 30. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Reimann, J. Dierkes, A. Carlsohn, D. Talbot, M. Ferrari, J. Hallund, W. L. Hall, K. Vafeiadou, U. Huebner, F. Branca, et al. Consumption of Soy Isoflavones Does Not Affect Plasma Total Homocysteine or Asymmetric Dimethylarginine Concentrations in Healthy Postmenopausal Women J. Nutr., January 1, 2006; 136(1): 100 - 105. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. R. Sowers, K. A. Matthews, M. Jannausch, J. F. Randolph, D. McConnell, K. Sutton-Tyrrell, R. Little, B. Lasley, and R. Pasternak Hemostatic Factors and Estrogen during the Menopausal Transition J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 5942 - 5948. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Jayachandran, A. Sanzo, W. G. Owen, and V. M. Miller Estrogenic regulation of tissue factor and tissue factor pathway inhibitor in platelets Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H1908 - H1916. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. I. Savitz, G. Schlaug, L. Caplan, and M. Selim Arterial Occlusive Lesions Recanalize More Frequently in Women Than in Men After Intravenous Tissue Plasminogen Activator Administration for Acute Stroke Stroke, July 1, 2005; 36(7): 1447 - 1451. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. M. Szymanski, C. M. Kessler, and B. Fernhall Relationship of physical fitness, hormone replacement therapy, and hemostatic risk factors in postmenopausal women J Appl Physiol, April 1, 2005; 98(4): 1341 - 1348. [Abstract] [Full Text] [PDF] |
||||
![]() |
E.-G. V. Giardina, H. J. Chen, R. R. Sciacca, and L. E. Rabbani Dynamic Variability of Hemostatic and Fibrinolytic Factors in Young Women J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 6179 - 6184. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. D. Pradhan, A. Z. LaCroix, R. D. Langer, M. Trevisan, C. E. Lewis, J. A. Hsia, A. Oberman, J. M. Kotchen, and P. M Ridker Tissue Plasminogen Activator Antigen and D-Dimer as Markers for Atherothrombotic Risk Among Healthy Postmenopausal Women Circulation, July 20, 2004; 110(3): 292 - 300. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. B. Kannel and D. Levy Menopause, Hormones, and Cardiovascular Vulnerability in Women Arch Intern Med, March 8, 2004; 164(5): 479 - 481. [Full Text] [PDF] |
||||
![]() |
J. K. Ghali, H. J. Krause-Steinrauf, K. F. Adams Jr, S. S. Khan, Y. D. Rosenberg, C. W. Yancy Jr, J. B. Young, S. Goldman, M. A. Peberdy, and J. Lindenfeld Gender differences in advanced heart failure: insights from the BEST study J. Am. Coll. Cardiol., December 17, 2003; 42(12): 2128 - 2134. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Sowers, C. Derby, M. L. Jannausch, J. I. Torrens, and R. Pasternak Insulin Resistance, Hemostatic Factors, and Hormone Interactions in Pre- and Perimenopausal Women: SWAN J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 4904 - 4910. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. C. Carr The Emergence of the Metabolic Syndrome with Menopause J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2404 - 2411. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. E. Rabbani, N. A. Seminario, R. R. Sciacca, H. J. Chen, and E.-G. V. Giardina Oral conjugated equine estrogen increases plasma von Willebrand factor in postmenopausal women J. Am. Coll. Cardiol., December 4, 2002; 40(11): 1991 - 1999. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Stangl, G. Baumann, and K. Stangl Coronary atherogenic risk factors in women Eur. Heart J., November 2, 2002; 23(22): 1738 - 1752. [Full Text] [PDF] |
||||
![]() |
H. Hognert, B. Ghanoum, H. Gustafsson, I. Milsom, and K. Manhem Acute Effects of Transdermal 17{beta}-Estradiol on Hemostatic Variables after 24-hour Treatment Clinical and Applied Thrombosis/Hemostasis, July 1, 2002; 8(3): 239 - 243. [Abstract] [PDF] |
||||
![]() |
J. B. Braunstein, D. W. Kershner, P. Bray, G. Gerstenblith, S. P. Schulman, W. S. Post, and R. S. Blumenthal Interaction of Hemostatic Genetics With Hormone Therapy : New Insights To Explain Arterial Thrombosis in Postmenopausal Women Chest, March 1, 2002; 121(3): 906 - 920. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. K. Dubey, S. Oparil, B. Imthurn, and E. K. Jackson Sex hormones and hypertension Cardiovasc Res, February 15, 2002; 53(3): 688 - 708. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. G. Shlipak, B. G. Angeja, A. S. Go, P. D. Frederick, J. G. Canto, and D. Grady Hormone Therapy and In-Hospital Survival After Myocardial Infarction in Postmenopausal Women Circulation, November 6, 2001; |