Hormones, genetic factors, and gender differences in cardiovascular disease

Time for primary review 31 days. On average, women develop heart disease some 10–15 years later than men. This raises the question of whether there is some aspect of ‘femaleness’ which reduces risk, or whether there is some aspect of ‘maleness’ that raises risk. To date, most attention has been focused on the hypothesis that endogenous estrogen is cardioprotective in women [1]. Rising rates of coronary heart disease (CHD) after the menopause, and after oophorectomy, are among the strands of evidence in humans that endogenous estrogen may prevent CHD [2]. However, upon closer examination this evidence is not persuasive, and in fact the evidence is amenable to alternative explanations. During the first 3 decades of adult life, low-density lipoprotein (LDL) cholesterol levels are lower in women than men, and this may contribute to the delayed onset of CHD in women. A more widely held explanation for the later onset of CHD in women is their higher high-density lipoprotein (HDL) cholesterol levels, attributed to higher endogenous estrogen levels in women. However, the difference in HDL cholesterol between women and men is an androgen effect, not an estrogen effect. Up to puberty, young men and women have similar HDL cholesterol levels. At puberty, concurrent with the rise in endogenous testosterone levels, the HDL cholesterol levels in young men decline to the adult level [3,4]. A 20% difference in HDL cholesterol levels predicts at least a 20% difference in CHD rates in the short term, and may predict even larger differences in CHD rates over a lifetime [5]. Thus, the entire gender difference in CHD risk may indeed be due to the lifelong difference in HDL cholesterol levels; however, this difference is a consequence of having the Y chromosome. During fetal development, the Y chromosome directs the formation of testes rather than ovaries, and the testes in turn produce testosterone and dihydrotestosterone rather than estrone and estradiol as the primary sex steroids [6]. The high levels of male sex hormones drive down the HDL cholesterol levels, with consequent higher early risk of CHD. Thus, the gender difference in CHD may well be due the most basic genetic difference between men and women, which is the presence of the Y chromosome. Some of the observational studies in adult males suggest that higher levels of testosterone are associated with higher (rather than lower) levels of HDL cholesterol, and with lower risk of CHD [7]. Intracoronary infusion of testosterone induces coronary artery dilatation and increases coronary blood flow in men with coronary artery disease [8]. These observations do not fit the concept that testosterone is harmful to the male cardiovascular system. On the other hand, androgen deprivation in men is associated with enhanced endothelium-dependent dilatation in men successfully treated for prostate cancer, while dilatation is reduced in genetic females taking high dose androgens [9,10]. In the laboratory, androgen receptor expression is greater in macrophages from male than from female donors, and lipid loading of male (but not female) macrophages was increased by dihydrotestosterone [11]. Dihydrotestosterone also increases the adhesion of monocytes to endothelial cells, increases the expression of endothelial vascular cell adhesion molecule-1, increases platelet expression of thromboxane receptors, and increases platelet aggregation [12,13]. However, there are even more gaps in our knowledge about the real effects of testosterone than there are for estrogen. As in the case of estrogen, clinical trials with hard endpoints using androgens, androgen antagonists, or selective androgen receptor modulators will ultimately be needed to resolve this issue, but it is difficult to anticipate that such trials will be done in the near future. The statement that CHD rates in women rise steeply after the age of menopause, and the corollary that this is due to lower levels of estrogen after that age, is also open to question. In actuality, there is no evidence for an increase in the year-on-year rate of increase in CHD around the age of menopause. The linear relationship between age and CHD incidence as seen on a semilogarithmic plot shows that there is a constant proportional increase in CHD incidence with age, with no inflection upward at the average age of menopause [14]. This is evidence for an age effect, and evidence against an effect of menopause. The Nurses' Health Study investigators have reported that, after controlling for age and smoking status, the natural menopause is not associated with an increased risk for CHD [2]. The same investigators have reported that, in contrast to the natural menopause, bilateral oophorectomy is associated with an increased risk for CHD in women who had never taken estrogen after menopause. However, the study had very few cases of CHD in women with oophorectomy, and the increased risk was no longer significant in the multivariate analysis. Nonetheless, the use of estrogens appeared to eliminate this increased risk. It is possible that women who have a hysterectomy and bilateral oophorectomy (often done for menorrhagia/metrorrhagia associated with endometrial hyperplasia) are at higher risk for CHD because of the co-existence of metabolic risk factors such as central obesity, high blood pressure, lipid disorders, and glucose intolerance. The finding of an apparent lower risk in women who subsequently used estrogen would be subject to the biases discussed below. In summary, the postmenopausal increase in risk is most likely due to age and not the menopause, and the increase in risk in women after oophorectomy may be due to confounding by other risk factors. By far the most persuasive evidence in favor of a protective effect for estrogen comes from the large number of cohort studies comparing CHD risk in postmenopausal women currently using estrogen to never-users. These studies have shown consistently that CHD risk is 35–50% lower in estrogen users, after adjusting for other risk factors [15,16]. The lower risk has been found in studies of estrogen alone, as well as in studies of estrogen in combination with a progestin [16]. For healthy women, the lower risk is found in those who have recently started estrogen as well as in long-term users [16]. These findings from observational epidemiology provide the rationale for clinical trials testing whether and to what degree current use of postmenopausal hormone therapy prevents a first heart attack. However, the observational epidemiology is not sufficient to prove the case, because even the best studies may be subject to a variety of systematic biases that could lead to an overestimation of benefit and an underestimation of harm from hormone therapy, hence the need for an unbiased estimate from clinical trials. These biases in observational studies include healthy user selection bias, compliance bias, surveillance bias, and survivor bias [14,17]. In combination, these biases will lead to a systematic overestimation of benefit, and an underestimation of risk in observational studies. Adjusting for baseline differences in risk factors will mitigate healthy user selection bias, but will not correct for compliance, surveillance, or survivor bias. Thus, the real benefit for CHD may be much less than predicted by the observational studies, or there may be no benefit at all [14]. The clinical trials to date have failed to show overall benefit for CHD over the short term (ongoing trials will ascertain the long term effects particularly in women without prevalent CHD) [18]. The possibility that estrogen may reduce CHD risk has stimulated a wide variety of studies that attempt to explain the presumed benefit. Because it was unexpected, fewer studies have been done to explain the apparent excess risk early in the course of treatment, found in recent clinical trials (see below). Effects of estrogen that may predict benefit include: lowered LDL cholesterol and lipoprotein (a) levels, raised HDL cholesterol levels, reduced fibrinogen levels and enhanced fibrinolysis (reduced plasminogen activator inhibitor-1 (PAI-1) and increased D-dimer levels), reduced homocysteine levels, antioxidant properties, and improved endothelial function (e.g. reduced E-selectin levels and enhanced flow-mediated dilatation) [1]. Estrogen and progesterone reduce lipid accumulation in macrophages from female, but not male, donors [19]. On the other hand, several mechanisms that might increase risk after estrogen administration have been found: triglycerides increase, some coagulation markers increase (e.g. Factor VII, prothrombin fragments 1+2, activated protein C resistance), and the inflammatory marker C-reactive protein increases [20–23]. Some observational studies have suggested that certain of the lipid markers changed by estrogen administration are associated with higher relative risks for CHD in women than in men, including HDL cholesterol and triglycerides [23,24]. However, particularly in respect of coagulation and inflammation, laboratory measurements do not predict whether the predominant effect will be favorable or unfavorable. Studies of venous thromboembolism (including clinical trials) provide unequivocal evidence that the overall effect is indeed procoagulant [25,26]. Compounding this difficulty in interpreting laboratory measurements is the fact that progestins counteract some of the estrogen effects, and that the clinical expression of metabolic changes may be time-dependent. The early excess risk for arterial disease observed in the Heart and Estrogen/Progestin Replacement Study (HERS) may have been due to an initial procoagulant or inflammatory effect on susceptible plaques, while the favorable effects in the survivors may be due to the later assertion of the generally favorable lipid effects [27]. In HERS, participants with higher levels at baseline had the largest decrease in lipoprotein (a) on treatment, and had a more favorable clinical outcome than participants with lower baseline levels [28]. The role of estrogen's direct vascular effect is unclear, since impaired endothelial function has not yet been established as a risk factor for CHD. Interestingly, current estrogen users do not appear to have a lower risk for angina, as one would expect if direct vascular effects were important [16]. On the other hand, in a clinical trial sublingual estrogen relieved exercise-induced angina [29]. Overall, the studies of mechanisms have not resolved the core issue of whether estrogens protect against CHD. It is important to realize that almost all the studies of mechanism used oral estrogen preparations, and may turn out to have little relevance towards explaining the gender difference in CHD. Ovarian estrogen directly enters the systemic circulation through the inferior vena cava, unlike oral estrogens, which enter the portal vein and undergo first pass hepatic circulation. Because of extensive metabolism in the liver, in order to achieve similar blood levels the dose of oral estrogen needs to be approximately ten times that of non-oral (e.g transdermal) estrogen. These doses of estrogens profoundly influence the hepatic metabolism of a variety of proteins, including lipid apoproteins, coagulation proteins, and (probably) C-reactive protein [20–22,30–34]. The large effects on lipids and coagulation proteins described for oral estrogens are greatly attenuated, absent, or in the opposite direction with non-oral estrogens. Non-oral estrogens have very modest effects on lowering LDL-cholesterol and lipoprotein (a), have no effect or reduce triglycerides, have no effect on HDL-cholesterol, and have a modest or no effect on levels of coagulation proteins [30–34]. Non-oral estrogens retain the ability to improve endothelial function [35]. Additional mechanistic studies, as well as epidemiologic studies and clinical trials, that focus on the role of non-oral estrogen preparations are needed. Of interest, epidemiologic studies in postmenopausal women have not shown an association of endogenous estrogen levels with CHD [36,37]. The role of the major monogenic disorders such as familial hypercholesterolemia and familial hyperhomocysteinemia in arterial disease is well established. Even within kindreds with familial hypercholesterolemia, affected females develop CHD later than in males, presumably because the same (unknown) factors that delay onset in non-affected females continue to operate in affected females [38]. Though estrogen administration lowers LDL cholesterol in women with familial hypercholesterolemia [39], this pharmacologic property does not necessarily mean that endogenous estrogen has the same effect. There is no sex difference in the elevated LDL cholesterol levels of individuals with familial hypercholesterolemia [38]. Rather, as in non-affected individuals, factors that blunt the impact of a given level of LDL cholesterol in females (or accelerate it in males) need to be sought. The importance to arterial disease of polymorphisms in the genes that code for coagulation proteins and markers of inflammation is not clear. For example, a variety of polymorphisms in the fibrinogen gene are known to affect circulating fibrinogen levels, and fibrinogen levels have been associated with arterial disease [40]. However, studies seeking to quantify the association of a fibrinogen gene polymorphism with arterial disease have generally yielded disappointing results. This may be because any individual polymorphism has a modest influence on fibrinogen levels, and the levels are simultaneously affected by both other genetic factors, and importantly, by environmental factors [40]. Hence, it is difficult to discern the direct effect of any individual polymorphism on clinical disease. In addition to fibrinogen, polymorphisms in the genes for FVII, FXIII, FV, prothrombin, thrombomodulin, tissue plasminogen activator, PAI-1, and platelet-membrane glycoproteins have been described [40]. Except for fibrinogen, the relationships of the phenotypic factors affected by these polymorphisms with arterial disease are uncertain. A possible interaction of gender with polymorphisms in the FVII gene has been described, in that plasma activity of FVII in males varies markedly according to the presence of three of these polymorphisms, but in females these polymorphisms are associated with much smaller changes in FVII activity [41]. Estrogen affects many of these phenotypic factors [20–22,32–35], but in general it is not known whether there is a protective or harmful interaction of estrogen with the polymorphisms as risk factors for arterial disease. Both males and females have α- and β-estrogen receptors in the vascular endothelium, smooth muscle, and myocardium, though receptor numbers may be higher in females because estrogen induces their expression [1,42]. One group of investigators found a gender difference for the non-genomic effects of estrogen on the response of coronary arteries to acetylcholine. Males did not show the reversal of acetylcholine-induced vasoconstriction in coronary arteries after acute estrogen administration found in females [43]. On the other hand, genetic males receiving long-term estrogen therapy show enhanced flow-mediated dilatation [44]. The presence of higher numbers of functional estrogen receptors in females may relate to the inhibition of injury-induced vascular intimal thickening by estrogen in female arteries, and the lower prevalence of left ventricular hypertrophy in women compared to men [1,45,46]. Estrogen receptor numbers are lower in female atherosclerotic arteries [46]. There is some potential for polymorphisms of the estrogen receptors themselves to affect the clinical outcome of estrogen administration. For example, the common ER Pvu II and Xbal genotypes are not associated with differences in HDL cholesterol levels, but after estrogen administration women who are heterozygous for the genotype have a greater elevation of HDL cholesterol [47]. The clinical relevance of this finding is The evidence from trials of hormone therapy to prevent coronary heart disease (CHD) to the that is On the a of early harm is from to a from primary trials. term effects of on the cardiovascular and also on the overall of postmenopausal women, are in the primary trials [18]. The findings from clinical trials of in women are with an early excess of arterial cardiovascular HERS, the and the Estrogen for and the Health In a of trials a in healthy women a excess risk for CHD for harm comes also from the early trials in men, and several observational studies of users of among women with CHD and one of healthy women in the Health study The evidence comes from and primary trials, trials using estrogens with and without and trials of estradiol and of the trials have in the case of the have not been their findings be but in the findings are the question is no longer whether the early harm is but rather what is the It possible that the lipid changes by may in long-term benefit for women who the first This question at this but for many women the may be if of the early harm are not for explaining early harm around a potential interaction of with coagulation inflammation mechanisms in a of women with to these to for certain women with a factor in the of an elevated blood or tissue or with an response to due to a genetic may in and induces a wide variety of changes in coagulation factors in healthy women. Some of these are likely to be while are likely to hence the clinical effect be predicted from laboratory studies. Because of an on a few factors, in in levels of fibrinogen and PAI-1, many that a favorable effect on As clinical trials and epidemiologic studies that the effect of is to coagulation in the venous [25,26]. There is a that coagulation factors may be more important in arterial disease than the coagulation mechanisms the Factor and the prothrombin are currently of most for both venous and arterial disease. a coagulation factor which the of the major natural activated protein Hence, the phenotypic expression be as activated protein C The prevalence of heterozygous in is about with a much lower prevalence of in and is associated with venous thromboembolism and is found in about of cases of and about of cases with presumed familial The risk for associated with is by to oral and the risk associated with is by use In the association of with CHD has not been found in men, and in women it has been In a study of women, was associated with in those with other risk factors for particularly smoking However, other studies in women failed to show an association with or an interaction with use In one study of women to a lipid for of use in the of was associated with a reduced risk for arterial while use in the presence of was associated with an increased risk Thus, the of as a risk factor for CHD is much less certain than for and one study suggested an interaction with On the other hand, there is to a possible role for in CHD in women, particularly in the of CHD associated with The prothrombin is procoagulant because it increases prothrombin levels. The prevalence is lower than found in about of As for the prevalence is much lower in and It is associated with found in of cases and of cases of In women, but not in men, the has been found to be associated with CHD A of women with found that the use was associated with risk for disease in the of the and increased risk in presence A study of in women with study of women without was found that presence of the was associated with an for of and current use of was associated with an of However, the presence of both the and use of was associated with an of This effect of the factors a large effect between and a factor in the of could explain the of early harm and benefit seen in the trial also affects several markers of inflammation, some in a direction and some in a favorable increases circulating levels of C-reactive protein and including On the other hand, circulating levels of cell adhesion adhesion molecule-1, and vascular cell adhesion molecule-1, and in the short term endothelial function hypothesis by on the effects of estrogen on to this acute administration may be and administration may be Estrogen levels and increases which in turn increases the expression of of in the are associated with and of the which is to the of the Thus, the acute effect of might be in women with On the other hand, for women who the acute or who do not have plaques, the long-term effect of elevated levels of might be since the might prevent the accumulation of proteins and improve arterial These observations and have potential relevance to women, as affect the use of estrogens in the of It is whether endogenous estrogens affect these markers of inflammation and and whether these factors have a role in explaining the gender difference in cardiovascular disease. the known and of estrogen with these markers against the of delayed onset in women, since the risk for arterial disease in women. women may have higher rates of than men, but it is not known whether endogenous estrogen or use of oral explain that The gender difference in the age of onset of CHD is as yet A natural is that the effects of the and are through sex and that these explain the gender difference in CHD. However, the of female and male sex hormones in this To date, has focused on estrogen. The evidence in favor of a protective effect of estrogen in women is question by recent clinical trial that is not and at least in women with heart disease may increase risk. Because of the increased in risk in trials, and the of clinical trial for primary is no longer for of CHD It possible that given for many years may be estrogen has effects on blood markers and cell including endothelial and Some of these effects may be and laboratory studies predict clinical trials that the effect of estrogen is The metabolic effects of oral estrogen are very from those of non-oral (e.g. transdermal) the is more likely to the effects of endogenous estrogen, and may be more about the mechanisms the gender difference in CHD. studies on the effects of non-oral estrogens are including trials with clinical It is but that in the gender difference may be due to effects of androgens in men, and more studies of androgens, and selective androgen receptor modulators are needed. Males and females not in their levels of sex the effects of those sex hormones also by It appear likely that more will that the gender difference has male as well as female In addition to effects of the sex it is possible that may influence the risk for CHD. By are as in males as in hence the be focused on that risk and with sex This is in but a of potential is the interaction between and estrogen oral and that a of in with and also in individuals with The most of the is to of the women who are at increased risk for CHD The are in this as are several other markers of coagulation and it possible to susceptible women, could be against while the women would have a higher of benefit from