| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Division of Cardiology (C.N.B.M., G.D.B., M.P.-L., G.H.), Department of Medicine, Cedars-Sinai Research Institute, Cedars-Sinai Medical Center, Los Angeles, California 90048; Department of Epidemiology (B.D.J., S.F.K.), Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania 15261; Division of Cardiovascular Medicine (C.J.P.), University of Florida, Gainesville, Florida 32611; Cardiovascular Institute (S.E.R.), University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania 15213; Rhode Island Hospital (B.L.S.), Providence, Rhode Island 02903; Division of Cardiology (V.B.), Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama 35294; and Division of Heart and Vascular Diseases (G.S.), National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: C. Noel Bairey Merz, M.D., c/o WISE Coordinating Center, University of Pittsburgh, 127 Parran Hall, Graduate School of Public Health, 130 DeSoto Street, Pittsburgh, Pennsylvania 15261.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background: Evidence for a beneficial role and the potential mechanism(s) of plant estrogens (phytoestrogens) on blood lipoproteins in humans is controversial.
Methods: We evaluated 483 women enrolled in the National Heart, Lung, and Blood Institute-sponsored Women’s Ischemia Syndrome Evaluation with coronary risk factors undergoing coronary angiography for evaluation for suspected ischemia for blood phytoestrogen levels (daidzein and genistein), lipoprotein levels [total cholesterol, triglycerides, low-density lipoprotein-cholesterol, high-density lipoprotein-cholesterol (HDL-C)], estrogen levels (estradiol, bioavailable estradiol, estrone), and angiographic coronary artery disease using core laboratories.
Results: Higher blood levels of the phytoestrogen daidzein were associated with lower triglycerides (P = 0.01), higher HDL-C (P = 0.05) levels, and a beneficial total cholesterol to HDL-C ratio (P = 0.02). This beneficial association was evident among the subgroup of women with low [<184 pmol/liter (<50 pg/ml)] blood estradiol levels, regardless of age and lipoprotein levels. The phytoestrogen associations with lipoproteins were incrementally related to the magnitude of daidzein level and independent of other lipoprotein modulators. There were no detectable relationships between the phytoestrogen levels and angiographic coronary artery disease.
Conclusions: Higher blood phytoestrogen daidzein levels are associated with beneficial lipoprotein levels in women with low estrogen levels, possibly by an estrogen receptor mechanism. These results suggest a potential explanation for the variable lipoprotein results observed in prior randomized controlled trials and call for investigation regarding subgroups of subjects who may preferentially benefit from dietary intake of food products, such as soy.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The Women’s Ischemia Syndrome Evaluation (WISE) is a multicenter National Heart, Lung, and Blood Institute-sponsored study designed to explore sex-specific ischemic heart disease pathophysiology (12). We hypothesized that higher blood levels of the phytoestrogen daidzein and genistein would be associated with a beneficial blood lipoprotein pattern in the WISE women.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Lipoprotein, reproductive hormone, and phytoestrogen analyses
All determinations were performed at core laboratories. Total plasma cholesterol, triglycerides, and high-density lipoprotein (HDL)-cholesterol were determined by enzymatic assay, and LDL-C was calculated using the Friedewald formula (12). Validated steroid and protein assay methods were used to determine levels of total estradiol, progesterone, estrone, FSH, and LH in addition to bioavailable estradiol (12). Daidzein, dihydrodaidzein, glycitein, ethyl phenol, equol, genistein, and o-desmethylangolensin (o-DMA) were determined by a core laboratory (Ralston Analytical Laboratories, St. Louis, MO) using liquid chromatography by a modification of the technique of Coward et al. (13) with an HPLC coupled to single and array electrochemical detection and validated using samples from a prior controlled phytoestrogen intervention study (14).
Measurement of coronary angiography
Coronary angiography was assessed by a core laboratory used in previous multicenter trials with angiographic outcomes (15). Measurements included quantitative assessment of the presence, severity, and complexity of epicardial coronary artery stenosis, using previously published methods, and a coronary severity score (15).
Statistical methods
Spearman rank correlations were used to assess the bivariate association between variables. Distributions of continuous measures such as lipoproteins were compared among phytoestrogen terciles using the Jonckheere-Terpstra test. Multivariate modeling of continuous variables was performed by analysis of covariance, with phytoestrogen terciles entered as class variables and other variables entered as continuous or dichotomous covariates. Stepwise logistic regression analysis was used to model the independent contribution of daidzein to the lipoprotein levels. Where standard methods identified multicollinearity among predictors, we fit several models containing unrelated covariates. We also performed multiple linear regression to fit a phytoestrogen-estradiol interaction term. Dichotomous variables such as the presence or absence of significant coronary artery disease were modeled in a bivariate context by the Wilcoxon rank-sums test and a multivariate context using logistic regression. Probability values of 0.05 or less were considered statistically significant. All analyses were performed using the SAS 6.12 software (SAS Institute, Cary, NC).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mean and median phytoestrogen blood levels are shown in Table 1
. Initial evaluation demonstrated moderate but significant correlations for daidzein and triglycerides (r = –0.13, P = 0.004) and HDL-C (r = 0.09, P = 0.05), whereas no significant relationship was apparent for total cholesterol (r = –0.07, P = 0.13) or LDL cholesterol (r = –0.06, P = 0.24). The correlations were higher among the subgroup of 372 women with low (<50 pg/ml) blood estradiol levels (daidzein and triglycerides, r = –0.16, P = 0.002; HDL cholesterol, r = 0.10, P = 0.06; total cholesterol, r = –0.09, P = 0.07; and LDL cholesterol, r = –0.10, P = 0.07), whereas no relationship was apparent among the 111 women with high [
184 pmol/liter (50 pg/ml)] blood estradiol level. There were no significant correlations between the other phytoestrogen and lipoprotein levels, despite significant correlations among daidzein and genistein level (r = 0.14, P = 0.002), glycitein (r = 0.24, P = 0.0001), and o-DMA (r = 0.30, P = 0.0001). Daidzein did not correlate with age, estradiol, or bioavailable estradiol.
|
). Notably, lower levels of triglycerides and higher levels of HDL cholesterol were evident with increasing tercile of daidzein level. In the subgroup of women with low [<184 pmol/liter (<50 pg/ml)] blood estradiol levels, the associations between daidzein and the lipoproteins remained beneficial, whereas among the women with high estradiol, there were no beneficial associations observed (Table 1
). All of these relationships were unaffected by controlling for age.
|
We next assessed the independent contribution of the daidzein level to the lipoprotein levels in multivariate analyses including estradiol level, hormone therapy, lipid-lowering medication, ß-blocker and diuretic use, race, physical activity, alcohol intake, body mass index, and cigarette smoking. The results demonstrated that inclusion in the multivariate models did not diminish the relationships between daidzein and lipoproteins.
Further analysis of interaction terms demonstrated significant differences between the high vs. low estradiol subgroups (Fig. 1). Specifically, significant relationships between daidzein and lipoproteins were found only among women with low estradiol [<184 pmol/liter (<50 pg/ml)] levels and included LDL-cholesterol (P = 0.03) and total cholesterol (P = 0.06).
|
Multivariate analysis demonstrated no relationship between daidzein or the other phytoestrogens with angiographic coronary disease, even when controlling for other significant variables, or exploring interactions.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
These results are consistent with previous literature regarding phytoestrogens and lipoprotein levels. Previous work in female primates demonstrated significantly lowered total and LDL-cholesterol and triglycerides as well as higher HDL-cholesterol when fed a soy protein diet, compared with controls (2). The magnitude of beneficial lipoprotein association observed in the current study, ranging from 5 to 17% for the various lipoproteins, compare favorably to the 5 to 20% observed in the prior human soy protein dietary studies (3, 4, 5, 10), and the 15 to 40% observed in the primate study (2). The positive results here are not consistent with a recent large dietary trial in women (11), possibly due to population differences in regard to blood cholesterol and estrogen levels as well as unmeasured factors, such as diet, genetics, and underlying disease burden.
Our findings contribute to an understanding of the possible mechanisms of action of the beneficial association between dietary phytoestrogens and lipoproteins, which have not been clearly established. Proposed mechanisms have included changes in T4 level (16), enhanced bile acid excretion (17), up-regulation of LDL cholesterol receptors (18), decreased zinc to copper ratios (19), and activation of the estrogen receptor (20), but few to no data exist in humans regarding any of these mechanisms. Our findings of the beneficial association between daidzein and lipoproteins being dominantly evident among women with low blood estrogen levels support the idea that daidzein may operate, at least in part, via an estrogen receptor mechanism. The differential association with triglycerides, compared with oral estrogen replacement therapy, however, suggests that additional mechanisms may be involved.
We did not find a similar association between genistein and lipoprotein levels, which is consistent with three prior studies of soy phytoestrogen extracts (5, 6) and liquid soy protein isolates (7). Soy phytoestrogen extracts contain genistein and daidzein in near equal amounts, and the relative binding affinity of genistein for the estrogen receptor-
and estrogen receptor-ß is about 20–40 times and 10–50 times higher than daidzein, respectively (20). In the primate model, genistein has been shown to have a beneficial effect on vascular reactivity, but daidzein levels appear more closely related with the beneficial lipoprotein effects (21). No prior studies have compared genistein and daidzein with respect to their effects on lipoprotein metabolism.
We did not find an association between phytoestrogen levels and angiographic coronary artery disease. It is possible that the overall prevalence of significant coronary artery disease (38%) in our population may have been too low to allow detection of a relationship, if it exists. Also, a single assessment of blood phytoestrogen levels at one point in time may not reflect chronic phytoestrogen levels and therefore may correlate better with a dynamic variable, such as blood cholesterol, compared with a chronic marker, such as angiographic disease.
Limitations
Our results are limited by the cross-sectional study design that cannot prove causality between variables. Indeed, the current results could be confounded by variables that have an impact on lipoprotein levels, such as diet, that were not measured in the current sample population. Specifically, dietary factors other than sources of phytoestrogens, such as higher fiber intake or lower fat intake, could contribute to beneficial lipoprotein levels. This analysis used multiple tests, assessing the relationships between various phytoestrogens and lipoproteins, thus increasing the chance of a type I error (or the possibility of incorrectly declaring a relationship to be statistically significant). If we adjust for this possibility using the Bonferroni correction, we would require P < 0.002 to be confident that we have a statistically significant relationship. However, with this correction, the chance of a type II error also increases (i.e. reporting significant relationship as nonsignificant). The finding of a consistent pattern of improved lipid distributions across daidzein terciles, both in the total population and among women with low estradiol, argues against a strict application of the Bonferroni correction as such a finding could not be obtained by chance alone. Further study is needed. Finally, whereas our analyses indicate that estrogen status is the primary predictor of the beneficial associations, rather than baseline lipoprotein level as suggested by prior work (1, 5, 6, 10, 18), our cross-sectional study design limits our ability to evaluate this.
Implications
The current study findings suggest that higher blood phytoestrogen daidzein levels are associated with beneficial blood lipoprotein levels in women with cardiac risk factors undergoing evaluation for suspected myocardial ischemia. Women with low blood estrogen levels demonstrate the strongest beneficial lipoprotein associations. These results require replication in future prospective and randomized trials using a dietary source of daidzein in women with low estrogen levels as well as correlative dietary studies to link them to dietary habits, including supplement use. Nevertheless, our findings support the idea that dietary consumption of phytoestrogen-rich foods, which have been shown to result in beneficial blood lipoprotein changes in humans (3, 4, 5, 10, 18), may mechanistically operate via higher daidzein levels. These and prior studies suggest that cardiovascular risk reduction strategies in women should consider dietary intake of food products, such as soy, which elevate blood daidzein levels, consistent with recent recommendations (19).
|
Footnotes |
|---|
First Published Online April 4, 2006
Abbreviations: HDL, High-density lipoprotein; LDL, low-density lipoprotein; o-DMA, o-desmethylangolensin; WISE, Women’s Ischemia Syndrome Evaluation.
Received August 16, 2005.
Accepted March 29, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
