| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Cardiovascular Research Group (H.J.T., D.K., Y.-L.L., B.P.M.), Department of Medicine, Monash University, Clayton; and International Health and Development Unit (F.S.D.), Monash University, Monash Medical School, The Alfred Hospital, Prahran and Jean Hailes Foundation (S.D.), Clayton, Australia
Address all correspondence to: Dr. Helena Teede, Vascular Medicine Unit, Dandenong Hospital, P.O. Box 478, Dandenong, Victoria, 3175, Australia. E-mail: h.teede{at}southernhealth.org.au
Abstract
To address the cardiovascular effects of dietary soy containing phytoestrogens, we measured blood pressure (BP), lipids, vascular function (systemic arterial compliance and pulse wave velocity), and endothelial function (flow-mediated vasodilation) in a randomized, double-blind trial.
Two hundred thirteen healthy subjects (108 men and 105 postmenopausal women), 50–75 yr old, received either soy protein isolate (40 g soy protein, 118 mg isoflavones) or casein placebo for 3 months.
There were 34 withdrawals (16%), with 179 subjects (96 men and 83 women) completing the protocol. After intervention in the soy group, compared with casein placebo, urinary phytoestrogens increased, accompanied by a significant fall in BP reflected by the BP model (P < 0.01) encompassing mean change (±SEM) in systolic (-7.5 ± 1.2 vs. -3.6 ± 1.1 mm Hg, P < 0.05), diastolic (-4.3 ± 0.8 vs. -1.9 ± 0.7 mm Hg, P < 0.05), and mean BP (-5.5 ± 1 vs. -0.9 ± 1 mm Hg, P < 0.008). In the lipid model, soy induced greater changes, compared with placebo (P < 0.001). On individual analysis, significant contributors included a reduction in the low- to high-density lipoprotein ratio (-0.33 ± 0.1 vs. 0.04 ± 0.1 mmol/L, P < 0.05) and triglycerides (-0.2 ± 0.05 vs. -0.01 ± 0.05 mol/L, P < 0.05) and an increase in Lp(a) lipoprotein (± 95% confidence interval) [42 (range, 17–67) vs. 4 (range, -22–31) mg/L, P < 0.05], whereas total, low-density lipoprotein, and high-density lipoprotein cholesterol improved in both groups; but no treatment effect was demonstrated. The arterial functional model demonstrated no difference between groups; although again, overall function improved in both groups. On individual analysis, peripheral PWV (reflecting peripheral vascular resistance) improved with soy (P < 0.01), whereas flow-mediated vasodilation (reflecting endothelial function) declined (in males only), compared with casein placebo (P < 0.02). No effect of treatment on the hypothalamic-pituitary-gonadal axis was noted in males or females.
In normotensive men and postmenopausal women, soy improved BP and lipids but, overall, did not improve vascular function. Potential adverse effects were noted, with a decline in endothelial function (in males only) and an increase in Lp(a). Further research in hypertensive and hyperlipidemic populations is needed.
PHYTOESTROGENS ARE UBIQUITOUS,
nonsteroidal, plant-derived compounds reported to exhibit both estrogen
agonist and antagonist activities (1). The soybean is a
rich source of the isoflavone phytoestrogens genistein and daidzein
(2), ligands for both estrogen receptors ER
and ERß
(3), with greater affinity demonstrated for ERß. ERß
has been shown to be the primary estrogen receptor in the vessel wall
and is up-regulated in response to vascular injury (4, 5).
High soy diets are associated with low cardiovascular disease risk
(6). Interventional studies, primarily focusing on the
effects of adding soy protein supplements to existing diets, indicate
that potential cardiovascular benefits may be partially attributable to
lipoprotein effects (7).
Intact soy protein supplementation lowered total and low-density lipoprotein (LDL) cholesterol and triglycerides in some, but not all, studies (7), whereas high-density lipoprotein (HDL) cholesterol was unchanged (7). The effects on Lp(a) lipoprotein are yet to be resolved. Furthermore, it is not known whether reported lipoprotein effects are attributable to the soy isoflavones or to other nonphytoestrogen components.
Preliminary studies, in women only, suggest that phytoestrogen supplementation reduces diastolic blood pressure (BP) (8) and has direct effects on the vessel wall (9). The vasculoprotective effects of soy resemble those of exogenous estrogen (4, 10). Estrogen improves large artery function in postmenopausal women (11, 12, 13) and male-to-female transexuals (14). Hence, we hypothesize that dietary soy may influence large artery function via biological estrogenic mechanisms.
Endothelial dysfunction, an early marker of vascular disease, can be assessed in vivo with flow-mediated vasodilation (FMD) (15). Brachial artery FMD correlates with coronary endothelial function (16) and with cardiovascular risk factors (15, 17, 18, 19). It deteriorates after the menopause (19) and improves with estrogen therapy (20). It is not yet established whether soy can similarly influence endothelial function.
To investigate the cardiovascular effects of soy, we have conducted a double-blind, randomized, placebo-controlled study of 3 months of addition of dietary soy protein supplementation in healthy normotensive men and postmenopausal women. No other dietary changes were made during the study. Endpoints included lipid parameters, BP, arterial compliance, and endothelial function measured by noninvasive Doppler and tonometry ultrasound techniques.
Materials and Methods
Study design
Two hundred thirteen healthy participants (108 men and 105 postmenopausal women) participated in a 3-month randomized, double-blind, placebo-controlled design. Participants (50–75 yr old) were recruited from community advertisements. They had not consumed antibiotics, soy products, or supplements (for 3 months), nor had they taken estrogen therapy (for 12 months) before entry. Postmenopausal status was defined as: 12 months of amenorrhea, and FSH more than 20 IU/L. Exclusion criteria included: moderate-to-severe menopausal symptoms, smoking (last 10 yr), diabetes, alcohol consumption more than 30 g/day, hypertension, abnormal uterine bleeding, cervical cytology or mammogram, and coexistent major illness. The Monash Medical Center Human Research and Ethics Committee approved the study, and all participants gave written informed consent. Randomization was performed independently using computer-generated random numbers, with 105 participants allocated to soy, (50 women and 55 men) and 108 to casein placebo (55 women and 53 men).
Participants were screened for dietary phytoestrogen intake at baseline based on history, dietary questionnaire, and urinary phytoestrogen excretion. Supplements were presented in identical unmarked powder sachets. They were consumed twice daily after mixing into beverage form and were taken in addition to the usual diet, with no other changes in daily dietary intake. The soy isolate powder was prepared by dehulling and defatting soybeans and then blending for 45 min. The product was then tested before packaging to ensure content uniformity for protein and isoflavones and was provided in a single batch, by Protein Technologies International (St. Louis, MO) (AB1.2 HG 70CA 29, lot A161–7). Each sachet contained 28 g powdered soy protein isolate, of which 71% was protein. This product was tested for isoflavone content using high-performance liquid chromatography (HPLC) with ultraviolet detection and was determined to contain 2.11 mg total isoflavones per g of protein. This was composed specifically of genestein 1.35, daidzein 0.66, and glycitein 0.09 (expressed as basic compound + respective glycosides, in milligrams per g of product). Overall, this provided 40 g soy protein per day, with a total of 118 mg isoflavones daily. This intake was selected because it provides a level of phytoestrogens between that of Pacific Rim nations (25–45 mg) and some areas in Japan (200 mg).
A baseline medical assessment was conducted, and a National Heart Foundation of Australia cardiovascular risk factor questionnaire was administered at both baseline and after 3 months (21). Exercise information was based on specific physical activities, not including household chores.
Dietary adherence was assessed by measurement of spot urine phytoestrogen concentrations at baseline and at the end of 3 months of therapy. Urine phytoestrogens were normalized for creatinine (Cr), a standard methodology applied in previous work both from our group and that of other international groups involved in human interventional studies using phytoestrogens (22, 23). Height, weight, waist-hip ratio, heart rate, and BPs were measured. Six brachial arterial BP readings were recorded, over 15 min, on a Dinamap (CRITIKON 1846 SX; Johnson and Johnson, Sydney, Australia), in a recumbent position, after 10 min of rest. The first reading was disregarded, the average of the subsequent readings being recorded.
Phytoestrogen assays
Urinary phytoestrogen excretion was analyzed using HPLC. The principal analytes were daidzein and genistein, as previously described (24). Phytoestrogen isolation was by reverse-phase HPLC (Shimadzu system LC 10A), following a method developed by Eldridge (25). Detection was by dual-wavelength ultraviolet absorbance (250 nm for daidzein, 262nm for genistein) and quantification by comparing the area under the curve with reference standards, genistein (Sigma, St. Louis, MO) and daidzein (ICN Biomedicals, Inc., Aurora, CO). Samples for each subject were processed in one batch.
Hormone, lipid, and lipoprotein assays
Fasting morning blood samples were collected at baseline and at 3 months, for testosterone (in males), estradiol (in females), gonadotropins, and lipid profiles and were centrifuged at 2500 x g for 12 min. Plasma was stored at -80O C and thawed immediately before analysis. Gonadotropins were measured by RIA performed on an automated MEIA using the AxSYM Immunoassay Analyzer (Abbott Laboratories Diagnostics Division, North Chicago, IL). Standardization was against the WHO 2nd International Reference Preparation (78/549) and (80/552) for FSH and LH, respectively. Total testosterone and estradiol were measured by standard automated immunoassay ACS180 (Bayer, Melbourne, Australia). Total cholesterol and triglycerides were measured using enzymatic reagents (DADE Diagnostics, Brisbane, Australia); HDL cholesterol was measured by homogeneous assay techniques (HDLC-Plus, DADE Diagnostics) adapted to a DADE Dimension RXL chemistry analyzer (DADE Diagnostics). LDL cholesterol was calculated using the Friedewald equation [LDL cholesterol = (total cholesterol - HDL cholesterol) - (triglycerides x 0.20)], adapted to S.I. units. Lp(a) lipoprotein was measured by an immunonephelometric assay on a nephlometric Analyzer (Behring, Marburg, Germany) using antiserum (DAKO Corp., Glostrup, Denmark).
Vascular parameters
All vascular parameters were measured by an experienced research assistant (D. Kotsopoulos). Studies were performed after an 8-h fast (including avoidance of caffeine-containing drinks), in a dark, quiet, air-conditioned clinical laboratory, after 10 min of rest in the supine position.
Total systemic arterial compliance (SAC). Noninvasive determination of blood flow and pressure waveforms were applied to determine SAC, as previously described (26). Aortic volumetric blood flow was measured from a hand-held 3.5-MHz continuous-wave Doppler flow velocimeter (Multidoplex M.D.1, Huntleigh Technology, Cardiff, UK) at the suprasternal notch. Simultaneous driving pressure was ascertained by applanation tonometry with a pressure transducer (Millar Mikro-tip, Millar Instruments Houston, TX) over the carotid artery, with pressures calibrated against Dinamap brachial artery pressures (CRITIKON 1846 SX). Given that SAC varies with changes in BP, we corrected for this by standardizing SAC values at the mean BP level for each individual. Compliance, over the total systemic arterial tree, was calculated by the following formula according to the method of Liu et al. (27): SAC = Ad ÷ R(Ps - Pd), where Ad is the area under the BP diastolic decay curve from end-systole to end-diastole; Ps is end-systolic BP; Pd is end-diastolic BP (carotid); and R is total peripheral resistance derived from BP and aortic root blood flow.
Pulse wave velocity (PWV). PWV was determined from recorded
pressure waveforms over both the aorto-femoral (AF) and the
femoro-dorsalis (FD) arterial segments (26). Pulse transit
time was defined as the time between the foot of simultaneously
recorded pressure waves, occurring at the end of diastole and the
beginning of systole, averaged over 10 cardiac cycles. Velocity was
derived from computer-generated pulse transit times and measured
distances between the two recording sites, as previously described
(26). PWV was calculated based on the formula:
PWV = D ÷ 
t (m/sec), where D is distance and t is the
time interval.
Brachial artery FMD. The brachial artery diameter was measured from B-mode ultrasound images captured on a Diasonics DRF-400 machine using a 10-MHz transducer, while an ECG trace was simultaneously recorded. Longitudinal scanning identified the clearest image of the brachial artery above the elbow, with continuous scanning held for 30 sec before and 4 min after ischemia, induced via a pneumatic tourniquet inflated (around the upper arm) to 40 mm Hg above systolic pressure, for 4 min. Vessel diameter was measured during systole and diastole and averaged over 5 cardiac cycles. FMD was determined as the percentage change, from baseline to 60 sec post ischemia (the point of maximal dilation) (26).
Repeatability. A repeatability study of vascular parameters was completed concurrently by our group (26). Bland-Altman plots showed satisfactory repeatability for SAC, PWV(AF), and PWV(FD), with repeatability coefficients of 9.2%, 3.2%, and 5.0%, respectively (26). FMD curves were not different between visits, with maximal diameter changes observed 60 sec post ischemia. Based on the sample size in this study, changes of 4% in SAC, 3% in PWV(AF), 2% in PWV(FD), and 10% in FMD were detectable, less than noted with estrogen therapy (11, 20). Also, a repeatability substudy was completed, before randomization, in 31 of the study participants who had 2 consecutive baseline evaluations. This demonstrated no significant differences in arterial compliance or PWV between visits.
Statistical analysis
The values for Lp(a) lipoprotein, urinary phytoestrogens, and
SAC were positively skewed. This was corrected by log transformation,
and the data was summarized as geometric means with 95% confidence
intervals (CI). The other variables were summarized as arithmetic
means ± 1 SE. Data from all participants who
completed the study protocol were included in the analyses. The levels
of significance was accepted at P 
0.05. Statistical
calculations were performed using the SPSS, Inc.
statistical package, version 9 (SPSS, Inc., Chicago,
IL).
For the purpose of statistical analysis, the outcomes were defined as the arithmetic difference (change) between baseline observations and those made after 3 months of treatment. The outcome variables were divided into three sets: 1) descriptors of BP (SBP, DBP, and MBP); 2) descriptors of lipid profiles [total, LDL, HDL, and LDL/HDL cholesterol ratio; triglycerides; and Lp(a) lipoprotein]; and 3) descriptors of vascular function [SAC, PWV(AF), PWV(FD)]. The outcome variables within each set were entered as dependent variables in multivariate ANOVA (MANOVA). For all sets, the categorical, independent variables were: treatment group (soy or casein placebo), sex (male or female), and age, as well as their treatment/sex and treatment/age interactions. After initial MANOVA, univariate ANOVAs were completed for each of the independent variables. A process of backward elimination was then followed, in which the independent variables were successively eliminated if P was more than 0.05, from univariate analyses. In the case of descriptors of BP and vascular function, the final MANOVA model contained only treatment, with one exception (in the case of FMD, the treatment/sex interaction, on univariate analysis, gave P = 0.02). Therefore, separate analyses by sex were performed.
For the final MANOVA models, the subsequent univariate analyses involved testing multiple hypotheses, thus increasing the risk of type I error (false-positive inference). This was controlled by employing the Ryan-Holm step-down Bonferonni procedure (28). This technique was also applied when analyzing differences between groups at baseline.
Results
In total, 34 participants withdrew (19 active and 15 casein
placebo), with 83 women and 96 men completing the study protocol. The 2
groups were well matched for baseline characteristics, including BP
(Table 1
), lipids (Table 2), vascular functional parameters (Table 3), age (61 ± 1 vs.
60 ± 1 yr), height (1.7 ± 0.01 vs. 1.7 ±
0.01 m), weight (72 ± 1 vs. 74 ± 1 kg),
body mass index (25 ± 0.4 ± 26 ± 0.04
kg/m2), waist-hip ratio (0.9 ± 0.01
vs. 0.9 ± 0.01), heart rate (61 ± 1
vs. 62 ± 1 beats/min), and exercise frequency
(5.2 ± 0.5 vs. 5.2 ± 0.7 h/fortnight). We have
previously reported that, based on a validated questionnaire, there was
a 57% incidence of hot flashes of mild severity and a high incidence
(80%) of very mild generalized menopausal symptoms in this
older-community-dwelling population of postmenopausal women
(29). These symptoms were similar in the active and
placebo groups (29).
|
|
|
Phytoestrogen levels
Baseline phytoestrogen excretion was similar in the two groups. After therapy, genistein increased 12-fold (15 ± 3 to 186 ± 23 ng/µmol Cr, P < 0.001) and daidzein increased 7-fold (44 ± 8 to 316 ± 37 ng/µmol Cr, P < 0.001) in the active group, whereas both genestein (10 ± 2 to 9 ± 2 ng/µmol Cr) and daidzein (23 ± 5 to 22 ± 5 ng/µmol Cr) remained unchanged in the casein placebo group.
Blood pressure and lipid parameters
Significantly greater falls in systolic, mean, and diastolic BP in
the active (compared with the casein placebo) group were seen on both
MANOVA (P = 0.014) and univariate ANOVA (Table 1
). The
multivariate model incorporating changes in total, LDL, HDL, and
LDL/HDL cholesterol ratio; Lp(a) lipoprotein; and triglycerides (Table 2) demonstrated significantly greater changes in the active (compared
with the casein placebo) group (P = 0.006). Only the
improvements in triglycerides and LDL/HDL cholesterol ratio and the
increase in Lp(a) lipoprotein contributed significantly to the
multivariate model. Changes in these parameters remained significant
(P always less than 0.05) on corrected ANOVA in the soy
group, compared with the casein placebo group (Table 2). Total, LDL,
and HDL cholesterol improved in both groups; yet, no effect of
treatment status was noted (Table 2). No significant change in body
weight was observed in either group (0.03 ± 0.03 vs.
0.15 ± 0.02 kg, P = 0.6).
Vascular compliance and endothelial function
Table 3 demonstrates mean (95% CI) SAC and mean (±
SEM) PWV [PWV(AF) and PWV(FD)]. Soy supplementation had
no significant effect on arterial compliance as assessed by MANOVA
incorporating SAC, PWV(AF), and PWV(FD) (P = 0.1). On
univariate ANOVA, only PWV(FD) improved significantly, compared with
casein placebo (P = 0.02). Correction of SAC for BP
changes did not alter these observations. There was no significant
effect of treatment on resting brachial artery diameter in either sex.
Brachial artery FMD was significantly reduced during soy treatment in
men (P < 0.05) but was unaffected in women (Fig. 1).
|
Baseline levels were similar in the active and casein placebo
groups for FSH and LH and serum testosterone (Table 4). Serum estrogen levels were less than
20 mmol/L (because of the assay sensitivity constraints); and
therefore, average levels were not given. After 3 months of therapy,
there were no significant changes in any hormonal levels in males or
females in either group (Table 4).
|
Three months of soy protein supplementation in normotensive subjects significantly reduced systolic, diastolic, and mean BP and also lowered triglycerides and LDL/HDL cholesterol ratio in men and postmenopausal women but did not affect arterial compliance. The hypothalamic-pituitary- gonadal axis was unaffected by treatment in both males and females. Not all effects of soy treatment were beneficial, given that Lp(a) lipoprotein rose by 15% and a reduction in mean brachial artery FMD was noted in males.
In the current study, though baseline levels of isoflavone excretion were similarly low in both groups at baseline, 3months of soy dietary supplementation resulted in significant increases in urinary genistein (12-fold) and daidzein (7-fold) levels, whereas no change was noted with casein placebo. Despite some variability, demonstrated with equivalent phytoestrogen intakes (30, 31), urinary phytoestrogen excretion is considered a reliable reflection of dietary phytoestrogen intake.
We have observed a reduction in systolic, diastolic, and mean BP in a healthy normotensive population, a novel and potentially important effect of soy. Dietary soy supplementation resulted in placebo-adjusted falls of 4.2 mm Hg in mean, 2.4 mm Hg in systolic, and 3.9 mm Hg in diastolic pressure. These changes are greater than those noted in the DASH study, in normotensive individuals, with dietary intervention alone (32). Our findings are consistent with the 4.9 mm Hg reduction in diastolic BP noted in a study of soy therapy in 51 normotensive women (8). However, in that study, differences in ion concentrations between active and placebo supplements may have contributed to the BP effects (8). In contrast, Hodgson et al. reported that 8 weeks of concentrated phytoestrogens (55 mg/day), in tablet form, did not alter supine, erect, or ambulatory BP in normotensive subjects (33). Potentially, this discrepancy is attributable to several factors, including comparatively lower phytoestrogen doses (33), the necessity for phytoestrogens to be ingested in combination with other soy components, or alternatively, the fact that hypotensive effects of soy may not be mediated by estrogenic mechanisms. In the present study, the observation that distal PWV [PWV(FD)], which reflects vasoconstriction in peripheral resistance vessels, improved significantly with soy therapy is consistent with the observed fall in BP.
The most consistently demonstrated beneficial effect of soy has been on lipids (7). Federal Drug Administration approval for consumer information labeling of soy foods has given in the United States, stating that "included in the daily diet, they may reduce the risk of heart disease". This is based primarily on the lipid benefits. A metaanalysis of 38 published controlled human clinical trials, of an average of 47g daily soy protein consumption (primarily in subjects with hyperlipidemia), noted significant reductions in total [9.3% decrease, 95% CI (0.35–0.85 mmol/L)] and LDL cholesterol [12.9% decrease, 95% CI (0.30–0.82 mmol/L)] and triglycerides [10.5% decrease, 95% CI (0.003–0.29 mmol/L)], with the hypocholesterolemic effect significantly related to pretreatment cholesterol levels (7).
In the present study, we observed improvements in the overall lipid
profile, the LDL/HDL cholesterol ratio, and triglycerides in both men
and postmenopausal women (Table 2). The fall in triglycerides is
encouraging, because there is increasing epidemiological evidence that
elevated triglycerides independently increase vascular risk and
emerging interventional data that suggest that reduction in
triglyceride levels reduces vascular risk (34). However,
the significance of the improvement in the LDL/HDL ratio is more
difficult to interpret, with epidemiological data consistently
suggesting that it is a predictor of vascular risk; yet, it has not
been shown to be an independent factor in intervention studies. In the
current study, total and LDL cholesterol levels fell in both groups;
yet, despite greater falls with soy, the differences between groups
were not significant. We would suggest that this is potentially
attributable to the normolipidemic population studied, with only small
changes in lipid levels expected. The fall in cholesterol levels in the
placebo group is consistent with the effects of involvement in clinical
intervention trials. Dietary casein protein has been noted to increase
cholesterol levels (35). However, given the observed
improvements in lipid profiles in both groups, this effect of casein is
unlikely to have contributed to the differences observed between the
soy and casein groups. As in previous literature, HDL cholesterol did
not change with soy therapy (7). The mechanism of soy
effects on lipids remains unresolved; however, the response pattern
does not parallel that of estrogen. Also, isolated phytoestrogen
supplements do not seem to effect lipid profiles (36).
Potentially, the saponins, polyunsaturated oils, or vegetable proteins
in soy (either individually or in a combined way) may contribute to the
beneficial reduction in triglycerides.
The rise in Lp(a) observed here with dietary soy supplementation is consistent with the results of a previous report of an unblinded cross-over study in nine normal men (37). This is potentially an adverse effect because, although Lp(a) is recognized as a risk factor for cardiovascular disease, it may be a cofactor rather than an independent risk factor (38). The pathophysiological significance of this, the most complex and polymorphic of the lipoprotein particles, remains unclear (38, 39). In contrast to the observed increase in Lp(a) with dietary soy, HRT has been reported to significantly reduce Lp(a) levels (40).
Both estrogen and phytoestrogens seem to have direct effects on the vessel wall. Animal data have shown similar dose-dependent vasculoprotective effects in animal carotid injury models, with both ligands equally inhibiting replication and migration of smooth muscle cells in vitro (4). In female monkeys, estrogen treatment significantly reduced aortic cholesteryl ester content, with a similar trend (P = 0.14) noted with soy protein, compared with casein (41). Also improved acetylcholine-mediated coronary artery vasodilation has been documented with estrogen (42), intact soy, and iv genistein administration (43). In male monkeys, the effects may be more complex because, although reduced atherosclerosis has been noted with intact soy (44), improved acetylcholine-mediated coronary artery vasodilation was not observed (43).
No beneficial effect of soy on arterial compliance or central PWV [PWV(AF)] was observed in the current study, despite a sample size adequate to detect improvements less than those noted with estrogen therapy. In contrast, two small studies, using lower doses of concentrated phytoestrogens rather than soy, reported improved SAC (9, 45). They involved both peri- and postmenopausal women and had no male subjects (9, 45). We have previously demonstrated a 22–39% greater compliance in healthy postmenopausal HRT users, compared with nonusers (11, 12, 13). Also, cessation of HRT results in an acute, significant (23%) fall in SAC (11). However, there are no reported long-term placebo- controlled studies of the effects of HRT on arterial function. Peripheral PWV [PWV(FD)] did improve with soy therapy, reflecting a reduction in the degree of vasoconstriction in peripheral vessels, most likely related to the observed changes in BP. However, the significance of isolated changes in this marker of vascular function, without accompanying changes in the more robust and predictive measure [PWV(AF)], must be interpreted with caution, with further study required.
Brachial FMD, mediated by nitric oxide release (46), has a 95% positive predictive value for coronary artery endothelial dysfunction (16). Impaired FMD is positively associated with cardiovascular risk factors (15, 17, 18, 19). Estrogen seems to improve endothelial function in normal healthy postmenopausal women (20, 47), in women with cardiovascular disease (47), and in male-to-female transexuals (14). An aim of the present study was to determine whether dietary soy had estrogen-like effects on endothelial function, as assessed by brachial artery FMD. It is recognized that this technique has its limitations, with less than ideal repeatability, significant operator dependence, and the influence of external factors such as exercise and food ingestion (15). In this study, a single experienced operator performed FMD under controlled conditions, with the study powered to detect a 10% change in FMD (26).
Compared with casein placebo, soy supplementation had no effect on FMD in healthy postmenopausal women, and it reduced FMD responses in males. A recent double-blind placebo-controlled cross-over study in 20 postmenopausal women has also noted no improvement of FMD; however, the intervention used was 80 mg isolated phytoestrogens in tablet form (36). The mechanism for this apparent deterioration in endothelial function in males is unclear, although previous studies have shown that acute estrogen administration did not improve FMD in human healthy males (48), nor did it induce endothelial dependent vasodilation in response to acetylcholine in the male monkey model (43).
The incidence of side effects was similar in both groups. Importantly,
this is the first report on the impact of estrogenic soy components on
the hypothalamic-pituitary-testicular axis in men and women. In men,
lowered serum testosterone and altered sexual behavior have been
demonstrated with herbal remedies containing phytoestrogens
(49), although soy therapy in male monkeys did not affect
testicular weights (44). Reassuringly, in males, we have
demonstrated no impact of soy on gonadotropins or serum testosterone
levels. In females, soy also did not affect the reproductive axis. This
was expected, however, because (by definition) these women are
menopausal and therefore in primary ovarian failure. Literature
suggests that menopause is associated with feedback insensitivity to
more potent estrogens; therefore, it was expected that weakly
biologically active phytoestrogens would not actively modulate this
reproductive axis (50). The mechanisms of phytoestrogen
action are varied and complex. At ER and ERß, phytoestrogens can
act as pure agonists or antagonists or have partial or selective
agonist/antagonist activity, further modified by variable affinity for
the ER subtypes (3) and by coactivators and corepressors.
This complexity significantly limits our understanding of the
mechanisms and sites of action of phytoestrogens (1). It
is clear, however, from both the current findings and those of animal
studies (41), that the effects of soy containing
phytoestrogens do not parallel those of estrogen, with respect to BP
responses, lipoprotein changes, or effects on vascular function or on
the hypothalamic-pituitary-gonadal axis.
Conclusion
Dietary soy protein supplement significantly reduced BP and improved LDL/HDL cholesterol ratio and triglycerides in healthy men and postmenopausal women, without improving indices of arterial function. Not all effects of soy treatment were beneficial. Lp(a) lipoprotein increased overall, and there was a reduction in mean brachial artery FMD in males. Soy consumption did not suppress the hypothalamic-pituitary-gonadal axis in males. Thus, it seems that the vascular, lipid, and hormonal effects of soy and estrogen differ. Further research is required to elucidate the active component(s) of soy and its mechanisms of action. The results of the present study suggest that soy may have a role to play in the prevention of cardiovascular disease, by improving both BP and lipid profiles; however, more research is needed, specifically in hyperlipidemic and hypertensive populations, before public health recommendations can be made.
Acknowledgments
We thank Prof. John Ludbrook, from the Biomedical Statistical Consulting Service, for his contribution to the data analysis; and Dr. N. Balazs and the Department of Clinical Biochemistry, Southern Health Care Network, for the plasma lipid measurements and hormone profiles. The methodology for SAC measurements was developed by Dr. James Cameron, who provided the computer software program and technical support for the study.
Footnotes
1 Supported by Grant-in-Aid GM98MO139 from the National Heart
Foundation of Australia. 
Received August 25, 2000.
Revised January 9, 2001.
Revised March 1, 2001.
Accepted March 13, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  Y. Kokubo, H. Iso, J. Ishihara, K. Okada, M. Inoue, S. Tsugane, and for the JPHC Study Group Association of Dietary Intake of Soy, Beans, and Isoflavones With Risk of Cerebral and Myocardial Infarctions in Japanese Populations: The Japan Public Health Center Based (JPHC) Study Cohort I Circulation, November 27, 2007; 116(22): 2553 - 2562. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-H. Chan, K.-K. Lau, K.-H. Yiu, S.-W. Li, H.-T. Chan, S. Tam, X.-O. Shu, C.-P. Lau, and H.-F. Tse Isoflavone intake in persons at high risk of cardiovascular events: implications for vascular endothelial function and the carotid atherosclerotic burden Am. J. Clinical Nutrition, October 1, 2007; 86(4): 938 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Clerici, K. D. R. Setchell, P. M. Battezzati, M. Pirro, V. Giuliano, S. Asciutti, D. Castellani, E. Nardi, G. Sabatino, S. Orlandi, et al. Pasta Naturally Enriched with Isoflavone Aglycons from Soy Germ Reduces Serum Lipids and Improves Markers of Cardiovascular Risk J. Nutr., October 1, 2007; 137(10): 2270 - 2278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Mann, D. J. Rowlands, F. Y.L. Li, P. de Winter, and R. C.M. Siow Activation of endothelial nitric oxide synthase by dietary isoflavones: Role of NO in Nrf2-mediated antioxidant gene expression Cardiovasc Res, July 15, 2007; 75(2): 261 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Hamilton-Reeves, S. A. Rebello, W. Thomas, J. W. Slaton, and M. S. Kurzer Isoflavone-Rich Soy Protein Isolate Suppresses Androgen Receptor Expression without Altering Estrogen Receptor-{beta} Expression or Serum Hormonal Profiles in Men at High Risk of Prostate Cancer J. Nutr., July 1, 2007; 137(7): 1769 - 1775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gonzalez, V. Jayagopal, E. S. Kilpatrick, T. Chapman, and S. L. Atkin Effects of Isoflavone Dietary Supplementation on Cardiovascular Risk Factors in Type 2 Diabetes Diabetes Care, July 1, 2007; 30(7): 1871 - 1873. [Full Text] [PDF]
|
|
|
|
|
|
N. R Matthan, S. M Jalbert, L. M Ausman, J. T Kuvin, R. H Karas, and A. H Lichtenstein Effect of soy protein from differently processed products on cardiovascular disease risk factors and vascular endothelial function in hypercholesterolemic subjects Am. J. Clinical Nutrition, April 1, 2007; 85(4): 960 - 966. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Goodin, F. Shen, W. J. Shih, N. Dave, M. P. Kane, P. Medina, G. H. Lambert, J. Aisner, M. Gallo, and R. S. DiPaola Clinical and Biological Activity of Soy Protein Powder Supplementation in Healthy Male Volunteers Cancer Epidemiol. Biomarkers Prev., April 1, 2007; 16(4): 829 - 833. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cupisti, L. Ghiadoni, C. D'Alessandro, I. Kardasz, E. Morelli, V. Panichi, D. Locati, S. Morandi, A. Saba, G. Barsotti, et al. Soy protein diet improves endothelial dysfunction in renal transplant patients Nephrol. Dial. Transplant., January 1, 2007; 22(1): 229 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Teede, D. Giannopoulos, F. S. Dalais, J. Hodgson, and B. P. McGrath Randomised, Controlled, Cross-Over Trial of Soy Protein with Isoflavones on Blood Pressure and Arterial Function in Hypertensive Subjects J. Am. Coll. Nutr., December 1, 2006; 25(6): 533 - 540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. K Nies, A. A Cymbala, S. L Kasten, D. G Lamprecht, and K. L Olson Complementary and Alternative Therapies for the Management of Dyslipidemia Ann. Pharmacother., November 1, 2006; 40(11): 1984 - 1992. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. Sacks, A. Lichtenstein, L. Van Horn, W. Harris, P. Kris-Etherton, M. Winston, and for the AHA Nutrition Committee Soy protein, isoflavones, and cardiovascular health: a summary of a statement for professionals from the american heart association nutrition committee. Arterioscler. Thromb. Vasc. Biol., August 1, 2006; 26(8): 1689 - 1692. [Full Text] [PDF]
|
|
|
|
 |
|
 C. N. Bairey Merz, B. D. Johnson, G. D. Braunstein, C. J. Pepine, S. E. Reis, M. Paul-Labrador, G. Hale, B. L. Sharaf, V. Bittner, G. Sopko, et al. Phytoestrogens and Lipoproteins in Women J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2209 - 2213. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M Hodgson, V. Burke, L. J Beilin, and I. B Puddey Partial substitution of carbohydrate intake with protein intake from lean red meat lowers blood pressure in hypertensive persons. Am. J. Clinical Nutrition, April 1, 2006; 83(4): 780 - 787. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. L Hall, K. Vafeiadou, J. Hallund, S. Bugel, M. Reimann, C. Koebnick, H-J F. Zunft, M. Ferrari, F. Branca, T. Dadd, et al. Soy-isoflavone-enriched foods and markers of lipid and glucose metabolism in postmenopausal women: interactions with genotype and equol production Am. J. Clinical Nutrition, March 1, 2006; 83(3): 592 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Dewell, P. L. W. Hollenbeck, and C. B. Hollenbeck A Critical Evaluation of the Role of Soy Protein and Isoflavone Supplementation in the Control of Plasma Cholesterol Concentrations J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 772 - 780. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Sacks, A. Lichtenstein, L. Van Horn, W. Harris, P. Kris-Etherton, M. Winston, and for the American Heart Association Nutrition Commi Soy Protein, Isoflavones, and Cardiovascular Health: An American Heart Association Science Advisory for Professionals From the Nutrition Committee Circulation, February 21, 2006; 113(7): 1034 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L McVeigh, B. L Dillingham, J. W Lampe, and A. M Duncan Effect of soy protein varying in isoflavone content on serum lipids in healthy young men Am. J. Clinical Nutrition, February 1, 2006; 83(2): 244 - 251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Appel, M. W. Brands, S. R. Daniels, N. Karanja, P. J. Elmer, and F. M. Sacks Dietary Approaches to Prevent and Treat Hypertension: A Scientific Statement From the American Heart Association Hypertension, February 1, 2006; 47(2): 296 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Adams, D. L. Golden, J. K. Williams, A. A. Franke, T. C. Register, and J. R. Kaplan Soy Protein Containing Isoflavones Reduces the Size of Atherosclerotic Plaques without Affecting Coronary Artery Reactivity in Adult Male Monkeys J. Nutr., December 1, 2005; 135(12): 2852 - 2856. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kublickiene, E. Svedas, B.-M. Landgren, M. Crisby, N. Nahar, H. Nisell, and L. Poston Small Artery Endothelial Dysfunction in Postmenopausal Women: In Vitro Function, Morphology, and Modification by Estrogen and Selective Estrogen Receptor Modulators J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 6113 - 6122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-L. Low, J. I. Taylor, P. B. Grace, M. Dow |
