This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wu, F. C. W. Articles by von Eckardstein, A. Search for Related Content PubMed PubMed Citation Articles by Wu, F. C. W. Articles by von Eckardstein, A.

Androgens and Coronary Artery Disease

Department of Endocrinology (F.C.W.W.), Manchester Royal Infirmary, University of Manchester, Manchester M13 9WL, United Kingdom; and Institute of Clinical Chemistry (A.v.E.), University of Zurich and University Hospital of Zurich, CH-8091 Zurich, Switzerland

Correspondence: Address all correspondence and requests for reprints to: Dr. F. C. W. Wu, Department of Endocrinology, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, England, United Kingdom. E-mail: frederick.wu{at}man.ac.uk

	Abstract

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

 
A significant and independent association between endogenous testosterone (T) levels and coronary events in men and women has not been confirmed in large prospective studies, although cross-sectional data have suggested coronary heart disease can be associated with low T in men. Hypoandrogenemia in men and hyperandrogenemia in women are associated with visceral obesity; insulin resistance; low high-density lipoprotein (HDL) cholesterol (HDL-C); and elevated triglycerides, low-density lipoprotein cholesterol, and plasminogen activator type 1. These gender differences and confounders render the precise role of endogenous T in atherosclerosis unclear. Observational studies do not support the hypothesis that dehydroepiandrosterone sulfate deficiency is a risk factor for coronary artery disease.

The effects of exogenous T on cardiovascular mortality or morbidity have not been extensively investigated in prospective controlled studies; preliminary data suggest there may be short-term improvements in electrocardiographic changes in men with coronary artery disease. In the majority of animal experiments, exogenous T exerts either neutral or beneficial effects on the development of atherosclerosis. Exogenous androgens induce both apparently beneficial and deleterious effects on cardiovascular risk factors by decreasing serum levels of HDL-C, plasminogen activator type 1 (apparently deleterious), lipoprotein (a), fibrinogen, insulin, leptin, and visceral fat mass (apparently beneficial) in men as well as women. However, androgen-induced declines in circulating HDL-C should not automatically be assumed to be proatherogenic, because these declines may instead reflect accelerated reverse cholesterol transport. Supraphysiological concentrations of T stimulate vasorelaxation; but at physiological concentrations, beneficial, neutral, and detrimental effects on vascular reactivity have been observed. T exerts proatherogenic effects on macrophage function by facilitating the uptake of modified lipoproteins and an antiatherogenic effect by stimulating efflux of cellular cholesterol to HDL.

In conclusion, the inconsistent data, which can only be partly explained by differences in dose and source of androgens, militate against a meaningful assessment of the net effect of T on atherosclerosis. Based on current evidence, the therapeutic use of T in men need not be restricted by concerns regarding cardiovascular side effects. Available data also do not justify the uncontrolled use of T or dehydroepiandrosterone for the prevention or treatment of coronary heart disease.

I. Introduction

II. The Gender Difference in Coronary Artery Disease

III. Relationships between Serum Levels of T and CAD—Observational Studies

A. T and CAD in men

B. T and CAD in women

IV. Relationships between Serum Levels of T and CAD—Interventional Clinical Studies

A. Endogenous androgen deprivation

B. Androgen excess from anabolic steroid abuse

C. Exogenous T treatment in men with CAD

D. Exogenous T treatment in women

V. Relationships between Serum Levels of T and CAD—Animal Studies

VI. Effects of T on Cardiovascular Risk Factors

A. Associations between endogenous T and cardiovascular risk factors: role of adipose tissue and insulin

B. Effects of puberty on cardiovascular risk factors

C. Effects of exogenous T on cardiovascular risk factors

VII. Effects of T on Cells of the Arterial Wall and Vascular Function

A. Vascular expression of sex hormone receptors and T converting enzymes

B. Effects of T on vascular reactivity

C. Effects of T on macrophage functions

D. Effects of T on arterial smooth muscle functions

E. Effects of T on platelet functions

VIII. DHEA(S) and CAD in Men and Women

IX. Estrogens and Cardiovascular Disease in Men

X. Summary and Conclusion

XI. Clinical Implications

	I. Introduction

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

 
ANDROGEN REPLACEMENT THERAPY has been used for over 60 yr to treat, with proven efficacy and safety, a relatively small number (estimated to be <0.5% of adult male population) of patients with male hypogonadal disorders and/or failure of sexual development. However, in the last 10 yr, evidence has accumulated to support a wider therapeutic role of androgens for nonclassical indications (1 ). These include male contraception; aplastic anemia; and sarcopenic, osteopenic, and depressive states frequently associated with an expanding variety of chronic systemic conditions (characterized by reduced circulating testosterone, T) such as AIDS, rheumatoid arthritis, chronic renal failure, chronic obstructive airways disease, and physiological aging. Androgens are also being investigated as an additional component of hormone replacement therapy, in conjunction with estrogens, in postmenopausal women, especially in those who have had bilateral oophorectomy (2 ). We are poised on the threshold of witnessing a greatly expanded population of patients of all ages who may potentially benefit from the biological actions of T or related androgens.

Despite substantial reductions in mortality over the past 30 yr, heart disease remains the leading cause of death, claiming a total of 6.3 million lives worldwide in 1990. Ischemic heart disease, fifth in the rank order of disabilities in 1990, is predicted to become the leading global cause of disease burden by 2020 (3 ). It is well known that the age-adjusted morbidity and mortality rates from coronary heart disease (CAD) are 2.5- to 4.5-fold higher in men than in women and that the gender gap narrows after the menopause (4 ). The lifetime risk of CAD at the age of 40 yr is 1 in 2 for men and 1 in 3 for women (5 ). This male preponderance is remarkably consistent across 52 countries with hugely divergent rates of CAD mortality and lifestyles (6 ). The universality of gender disparity makes it likely that there is an intrinsic sexual dimorphism in susceptibility to CAD that may involve genetic, hormonal, lifestyle, or aging factors. The most popular explanation for this male preponderance in CAD is that adult male levels of T are proatherogenic, and/or there is a lack of the cardioprotective effects of estrogens in men. With the prospects of much wider therapeutic applications of androgens (for nonclassical indications), especially in the older age groups, an important clinical question is whether androgen treatment might increase the risk or severity of CAD. Being the most common cause of mortality and morbidity in men, even a tiny increase in the risk of CAD will not only negate any personal therapeutic benefits from androgen treatment but will also impose an unacceptable extra burden on healthcare resources. This concern has become a major safety issue for androgen therapy.

The aim of this review is to summarize disparate and often conflicting data from a variety of disciplines into a global assessment of the relationship between androgens and CAD. It is based on MEDLINE searches up to April 30, 2002, using the following keywords: androgens, testosterone, dehydroepiandrosterone (DHEA), oestrogens (estrogens), androgen receptor, oestrogen (estrogen) receptor (ER), aromatase, 5 reductase, polycystic ovary syndrome, hypogonadism, or hyperandrogenism in combination with cardiovascular disease, coronary heart (artery) disease, atherosclerosis, arteriosclerosis, diabetes mellitus, obesity, lipids, lipoproteins, hemostasis, coagulation, vascular reactivity, macrophage, endothelium, endothelial cell (EC), smooth muscle cell (SMC), or platelets. Only full published papers, but not conference abstracts, were included. No minimum criteria for inclusion of individual studies have been imposed; the intention was to achieve comprehensive literature coverage. The relative merits and limitations of quoted information will be critically discussed in the text.

	II. The Gender Difference in Coronary Artery Disease

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

 
Male gender is one of the classic risk factors for CAD (7 ), and average life expectancy is some 8 yr less in males than females. Androgens or the lack of estrogens have traditionally been regarded as the proximate cause underlying this male disadvantage. However, the consistent 2.5–4.5:1 sex ratio in CAD across many countries compared with the ethnic/geographic disparity, with a 5- to 10-fold higher CAD mortality rates in eastern and northern Europe than in southern Europe and Japan (8 ), suggests that the gender effect is not as important as other risk factors that act on both men and women (Fig. 1). The narrowing of the gender gap after middle age, associated with a relative deceleration of CAD deaths in men and an absence of acceleration of CAD deaths perimenopausally in women, would also argue against a prime role for sex hormones in the pathogenesis of CAD (9 ). Nonhormonal factors may play a predominant part in the gender disparity in CAD. Interactions between a multiple genetic and environmental/lifestyle factors are important in the pathogenesis of atherosclerosis (10 ). Thus, uncommon genetic polymorphisms are responsible for a low background prevalence of CAD in both men and women. In addition, common genetic polymorphisms interact with classic risk factors to negate the protective genetic effects or enhance the deleterious actions of environmental or lifestyle variables (10 ). The gender-specific expression of candidate genes may involve diverse mechanisms ranging from in utero sex hormone imprinting on gender-specific behavior patterns and distribution of visceral body fat to vascular and myocardial structural and functional adaptation to aging, pressure overload, and disease (11 ). Gender differences are detectable in vascular endothelial functions (12 13 ), lipid loading in human monocyte-derived macrophages (14 ), and abdominal visceral fat deposition (15 ). These mechanisms/factors will be discussed in more detail in the ensuing sections.

View larger version (30K): [in this window] [in a new window]   Figure 1. Age-adjusted mortality rates for CAD by country and sex (age 35–74 yr). Note the much higher difference in mortality between countries than between genders. A woman living in Scotland has a higher chance of dying from CAD than a man living in France (430 ).

	III. Relationships between Serum Levels of T and CAD—Observational Studies

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

A. T and CAD in men
Table 1 summarizes 39 studies (19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 ) of the relationships between circulating T and CAD in men.

2. Prospective cohort or nested case-control studies.
Table 1 also summarizes the seven non-cross-sectional studies (19 43 45 46 48 51 52 ). None of these studies showed T to have any significant relationship or predictive value for incident CAD. The three prospective cohort studies followed 1009 Californian (Rancho Bernardo) men aged 40–79 yr over a 12-yr period (45 ), 2512 men aged 45–59 yr in the United Kingdom (Caerphilly) for a 5-yr period (51 ), and 890 largely middle-class and 87% Caucasian (Baltimore) men aged 53.8 ± 16 yr for a period up to 31 yr (19 ). There was no correlation between baseline T levels and subsequent development of fatal or nonfatal CAD, stroke, or heart failure after adjusting for relevant confounders. Despite the concern that only a single hormone measurement at recruitment was undertaken and possible storage artifact, the relatively large size and long follow-up period of these three cohort studies go a long way toward confirming that T is not an independent risk factor for CAD in men.

In the four nested case-control studies, baseline T levels in cases of CAD and matched controls from the Honolulu Heart Program (43 ), Multiple Risk Factors Interventional Trial (46 ), Baltimore Longitudinal Study of Ageing (Ref. 48 , the earlier and shorter version of Ref. 19 ), and the Helsinki Heart Study (52 ) did not predict CAD events during observation periods of 6–8, 19–20, 9.5, and 5 yr, respectively.

In summary, the seven prospective studies provide a consistent and convincing data set that shows the lack of a relationship between circulating T and incident or existing CAD in men. There is a suggestion, only from cross-sectional studies, that patients with CAD may have lower T levels; the nature of this relationship is unclear. None of the 39 studies in the literature showed a positive relationship between T and CAD to suggest that high levels of this androgen may be a risk factor.

B. T and CAD in women
There are relatively few studies that investigated the relationship between endogenous levels of androgens and CAD in women (Table 2A). Age-adjusted concentrations of T, bioavailable T, and androstenedione did not differ significantly in 651 postmenopausal women, from the Rancho Bernardo study, with and without a history of heart disease at baseline and did not predict cardiovascular death or death from ischemic heart disease during the subsequent 19 yr (59 ). In contrast, in a cross-sectional angiographic study of 109 postmenopausal women with chest pain, serum levels of free T were correlated with the maximum percentage reduction of the luminal diameter of coronary arteries. This correlation was independent of age, body mass index (BMI), systolic blood pressure, smoking, or levels of cholesterol, insulin, and estradiol (60 ). However, higher free T and androstenedione within the physiological range had also been correlated with less carotid artery atherosclerosis in premenopausal and postmenopausal women (61 ).

Wild et al. (72 ) assessed the waist-hip ratio (WHR) and previous history of symptomatic androgen excess (hirsutism and acne) in 102 consecutive women undergoing cardiac catheterization. A positive correlation between angiographic evidence of coronary artery disease and clinical evidence of hyperandrogenism was found (Table 2B). In a combined angiography and pelvic ultrasound study of 143 women aged 60 yr or less who were referred because of chest pain or valvular heart disease, the presence of polycystic ovaries (in 42% of patients) was associated with an increased number of stenosed coronary arteries (73 ). Moreover, the presence of CAD and a family history of MI as well as elevated levels of insulin and triglycerides and lower levels of high-density lipoprotein (HDL)-cholesterol (HDL-C) were independent predictors of polycystic ovaries. The prevalence of CAD (history of chest pain, MI, angioplasty, or coronary artery bypass grafts) was found to be significantly higher in 28 women (45–59 yr old) who had undergone ovarian wedge resection over 18 yr ago compared with 752 aged-matched controls (76 ). The low response rate in the cases (<50%) and the uncertain diagnosis of CAD based on history of possible angina or MI make the data in this study difficult to interpret. In cross-sectional studies using B-mode ultrasound, significantly increased carotid artery intima-media thickness was found in patients with PCOS compared with age-matched controls (74 75 ). This was not entirely explained by BMI, fat distribution, and other risk factors and may be regarded as evidence in support of subclinical premature atherosclerosis in middle-aged (>45 yr) women independently related to the increased T in PCOS. Similarly, a recent study (77 ) demonstrated an increased prevalence of coronary artery calcification (which correlates with atherosclerosis) in 32 premenopausal (30–45 yr old) women with PCOS compared with 52 controls using electron beam computed tomography. These three studies employed noninvasive markers of early atherosclerosis to demonstrate an excessive risk for subclinical cardiovascular disease in relatively young PCOS patients. The data require confirmation with larger numbers and prospective follow-up. However, despite marked differences in glucose/insulin ratio and free androgen index in 18 healthy, obese, young women (32.7 ± 1.9 yr) with PCOS, insulin resistance, hyperandrogenism, and endothelium-dependent and -independent vascular responses were normal compared with age-matched controls (78 ).

In terms of actual cardiovascular disease events associated with PCOS, there is information from only two long-term longitudinal studies. Mortality and morbidity over an average 30 yr in 786 of 1028 women (over 45 yr of age) diagnosed to have PCOS on histopathological and hospital in-patient diagnostic records between 1930 and 1979, most of whom underwent ovarian wedge resection, were compared retrospectively with 1060 age-matched control women. Despite the significantly increased diabetes, hypertension, cholesterol, and nonfatal cerebrovascular disease, the standardized mortality ratio for CAD of 1.4 (95% CI, 0.8–2.4) and OR for a history of CAD of 1.2 (95% CI, 0.5–2.6) were not significantly raised (79 80 ). In a recent Dutch cohort of 346 nonobese patients aged 30.3–55.7 yr diagnosed to have PCOS in a specialized clinic 12 yr (range 1.2–31.6) previously, the prevalence of cardiac complaints (serious heart disease or cardiac arrest) ascertained by telephone questionnaire was not significantly different from that in 8950 age-matched females in the general population, despite the higher prevalence of both diabetes and hypertension (81 ). This suggests that previous estimates of CAD risk in PCOS may have been somewhat excessive. However, both these studies suffer from methodological drawbacks such as underascertainment of PCOS (79 80 ) and the relative young age of the smaller cohort (81 ).

Endogenous T is unlikely to have a causal or protective role for CAD in women. On the other hand, there is little doubt that PCOS patients (younger women of reproductive age) have an adverse risk profile for cardiovascular disease. However, whether this leads to increased, premature heart disease and, if so, whether this is causally related to chronic hyperandrogenemia per se, as opposed to associated variables, remain unresolved questions. Nevertheless, it is important not to dismiss the possibility of an association between PCOS and CAD events (probably independent of T). Given the high prevalence of PCOS in the female population, this should remain a high priority target for future research.

	IV. Relationships between Serum Levels of T and CAD—Interventional Clinical Studies

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

 
A. Endogenous androgen deprivation
A frequently cited study (82 ) compared the life span of 297 castrated inmates with 735 intact inmates (white males) in a single state institution for the mentally retarded in Kansas between 1895 and 1950. The reasons for castration were unclear. Castrated males lived an average of 13.6 yr longer than intact controls. However, the excess mortality in intact inmates was due to infections with no difference in cardiovascular disease mortality between the two groups. The authors concluded that postpubertal castration did not decrease the frequency of deaths due to cardiovascular disease. In a historical review (83 ), the life span of 50 castrated singers (prepubertal castrates) born between 1581 and 1858 in Europe was 65.5 ± 13.8 yr compared with 64.3 ± 14.1 yr in 50 noncastrated singers. In another historical survey of castration, Wilson and Roehrborn (84 ) also concluded that there are no valid data indicating that castration has any effect on life span of men. Doubts about ascertainment accuracy and the small size of these historical studies make it difficult to draw clear conclusions. The findings are, however, consistent with findings from cross-gender sex-hormone treatment in 816 male-to-female transsexuals aged 18–86 yr (mean, 41 yr; Ref. 85 ). Administration of ethinylestradiol (100 µg/d) and cyproterone acetate (100 mg/d) for 7734 patient-years was not associated with any significant difference in cardiovascular mortality or morbidity compared with the general male population, despite a 20-fold increase in venous thromboembolic complications.

B. Androgen excess from anabolic steroid abuse
Excessive T exposure in men is uncommon in clinical practice. However, anabolic-androgenic steroid (AAS) abuse in the general population is said to have reached epidemic proportion, with over 1 million current and former users in the United States alone (86 87 88 ). In two reviews of the literature covering a 12-yr period from 1987–1998 (89 90 ), there was a total of 17 case reports of cardiovascular events in young male body builders using suprapharmacological doses of AAS. Invariably, multiple preparations seldom prescribed in clinical practice, including oral 17-alkylated androgens, are used in combination simultaneously. There are 11 documented cases of acute MI, 4 cardiomyopathy, and 2 strokes. It is not possible to draw firm scientific conclusions from these sporadic case reports, especially when the baseline denominator information on prevalence and extent of exposure is shrouded in uncertainty and secrecy. But with the vast increase in abuse since the 1960s (86 87 89 ), there is no clear evidence for an epidemic of cardiovascular events among likely users and ex-users of AAS. A formal case-control study of AAS abuse in younger men presenting with acute MI has not been performed. Nevertheless, it has been suggested that dose-dependent androgen-induced vasospasm, platelet aggregation, activation of coagulation cascade, atherogenic lipid profiles [increased low-density lipoprotein (LDL)-cholesterol (LDL-C) and decreased HDL-C], and abnormal left ventricular function and hypertrophy are relevant mechanisms precipitating sudden cardiac deaths in young power athletes and body builders (90 ). It must be emphasized that pathological data from men abusing exotic AASs in doses several orders of magnitude higher than those prescribed in the clinical setting should not be extrapolated to the legitimate medical therapeutic use of approved T preparations or indeed to androgen physiology.

C. Exogenous T treatment in men with CAD
There are 17 reports in the literature documenting the effects of therapeutic doses of T in men with CAD. All showed some improvement or beneficial effects. The early studies from the 1940s are of historical interest only because of the small number of patients included and the uncontrolled observations (91 92 93 94 95 96 97 98 99 100 101 ).

Webb and colleagues (102 103 ) showed that a single iv bolus of 2.3 mg of T increased the time to 1-mm ST-segment depression on ECG by 66 sec 15–117(15–117, P < 0.016) in 14 men with CAD and low plasma T. The plasma T increased from 5.2–117 nmol/liter, indicating that this is a pharmacological action on the coronary vasculature. These direct acute pharmacological effects of T have been further studied during coronary angiography. Webb and colleagues (102 103 ) infused T over 3 min into the coronary arteries of 13 men with established CAD during coronary angiography at doses of 10^–7 to 10^–10 mol/liter (8 µmol/liter to 8 nmol/liter). Coronary vessel diameter increased by 3.1–4.5% at the three higher doses but not at the physiological dose of 10^–10 mol/liter. Coronary artery blood flow increased by 12–17.4% at all four doses of T. These effects were mediated by endothelium-independent and nongenomic mechanisms. This is the first demonstration of a direct vasodilatory action of T on coronary arteries in vivo in human males. These results have been confirmed by a similar study (104 ) in 14 men with established CAD. Intravenous infusion of 2.5 mg of T prolonged time to 1-mm ST depression from 471–579 sec and increased total exercise time from 541–631 sec. Whether the acute vasodilatory action of T at pharmacological doses translates into physiological therapy remains to be determined (also see Section VII.B).

Jaffe (105 ) reported the first randomized placebo-controlled double-blind study investigating the effects of T cypionate (200 mg im weekly) in 50 men with positive exercise ECG (n = 25 in each group). The sum of ST-segment depression in leads II, V4, V5, and V6 immediately 2, 4, and 6 min after the standard two-step exercise test (16 measurements in all) decreased by 32% and 51% from baseline after 4 and 8 wk in the active group with no change in the placebo group. There was no mention of any symptomatic improvement. Wu and Weng (106 ) reported another randomized placebo-controlled crossover (but not double-blinded) study in 62 elderly men with CAD treated with oral T undecanoate or placebo for 4 wk. T increased from 17–27 nmol/liter on T undecanoate (120–40 mg daily). The response categories were established by the Chinese Ministry of Public Health and denoted as very effective, effective, ineffective, worsened, and total efficacy but were not defined further. Both subjective symptom scores and resting ECG were improved in 69% and 75% of subjects, respectively, after 4 wk of treatment. In a recent study, English et al. (107 ) investigated the effects of a physiological dose of transdermal T (5 mg daily) for 12 wk in 50 patients with symptomatic CAD in a double-blind randomized placebo-controlled add-on trial. Plasma T increased from 13.6–22.3 and 18.6 nmol/liter after 4 and 12 wk of T treatment. The time to 1-mm ST-segment depression increased from 309–343 at wk 4 and 361 sec at wk 12 in the treated and from 266–284 at wk 6 and 292 sec at wk 12 in the placebo group (P < 0.02, treated vs. placebo).

These preliminary data suggest short-term improvements in ECG changes of CAD after (maximum of 12 wk) T supplement. Whether there are real symptomatic or functional benefits or decreased mortality in the long term remain important but unanswered questions.

D. Exogenous T treatment in women
The possible physiological roles of androgens in women may include increasing libido, energy, bone mineral density, muscle mass, and strength, but the data to support these possible roles are currently limited and not entirely convincing (108 ). Although hypopituitary (109 ) and bilaterally ovariectomized females (110 ) are undoubtedly androgen deficient, circulating T is only minimally lower after the natural menopause because ovarian secretion is maintained (111 108 ). Nevertheless, there is increasing interest in the use of T as part of postmenopausal hormone replacement therapy, in particular to improve reported impaired sexual function (2 112 ). Whether the concurrent use of T will impact the effects of estrogen hormone replacement therapy on the cardiovascular system is currently unknown. In a 20-yr (1975–1994) retrospective survey of the Amsterdam Gender Dysphoria Clinic (85 ), 293 female-to-male transsexuals aged 17–70 yr (mean, 34 yr) were treated for 2 months to 41 yr (total exposure of 2418 patient-years) with oral T undecanoate (160 mg daily) or T (Sustanon; 250 mg im every 2 wk). There was no excess of cardiovascular mortality (all cause) or morbidity compared with the general female Dutch population. However, there is currently insufficient evidence to exclude harmful cardiovascular effects of T treatment in women.

In summary, interventional studies to decrease endogenous T or administration of T generally do not suggest a causal relationship between T exposure and the development of CAD. Although some preliminary information hints at possible beneficial effects on myocardial ischemia, prospective controlled data on cardiovascular disease endpoints (MI, angina, mortality) from large-scale interventional studies using physiological doses of androgens are currently lacking.

	V. Relationships between Serum Levels of T and CAD—Animal Studies

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

 
The influence of androgens on the development and progression of experimentally induced atherosclerosis has been investigated in six animal models with diet- or injury-induced atherosclerosis and in two genetic atherosclerosis-susceptible mouse models (Refs. 113 114 115 116 117 118 119 120 121 122 and Table 3). Larsen et al. (114 ) investigated the effects of im T enanthate in castrated male rabbits and found no difference in the cholesterol content of abdominal aorta lesion after 17 wk. A similar negative result was obtained with the anabolic steroid stanozolol (115 ). Bruck et al. (118 ) demonstrated gender-specific effects of T and estradiol in castrated male and female rabbits fed an atherogenic diet. After 12 wk, aortic arch atheroma formation was significantly inhibited by im estradiol valerate (1 mg/kg·wk) in females but not in males, by T enanthate (25 mg/kg·wk) in males but not in females, and by combined estradiol and T administration in both sexes. Interestingly, T treatment in female rabbits increased plaque sizes, but estradiol had no effect in male rabbits. The authors concluded that the antiatherogenic effects of sex steroids involve gender-specific mechanisms and are independent of changes in plasma lipids. T did not have any effect on the myointimal proliferation response to balloon injury of the carotid artery in vivo in either male or female intact or gonadectomized rats, whereas estradiol inhibited this response in both sexes (117 ). Alexandersen et al. (119 ) showed that castration per se in male rabbits resulted in a doubling of aortic atherosclerosis compared with sham-operated controls, suggesting that endogenous T has an antiatherogenic effect. This can be reversed by oral T undecanoate (80 mg daily) or DHEA (500 mg daily) via a lipid-dependent mechanism. In addition, im T enanthate (25 mg twice weekly), which raised circulating T levels by 10-fold, decreased aortic atherosclerosis by lipid-independent mechanisms. This suggests that androgens in pharmacological doses may exert antiatherogenic effects on the vasculature. In contrast, treatment of male chicks with T resulted in a dose-dependent increase in aortic atherosclerosis (113 ). Similarly, in female ovariectomized cynomolgus monkeys fed an atherogenic diet for 24 months, the extent of coronary atherosclerosis was doubled with loss of compensatory remodeling of the arterial lumen in the T-treated group compared with the intact and untreated ovariectomized controls (116 ). These effects were independent of various risk factors including lipids. However, the acetylcholine-induced atherosclerosis-related coronary artery vasoconstriction was reversed by T treatment. Thus, despite the adverse pathomorphological changes in the arterial wall, functional parameters of the endothelium nevertheless improved upon treatment with T (116 ). It should also be pointed out that the SILASTIC-brand (Dow Corning Corp., Midland, MI) T implants used failed to maintain T levels (0.6 nmol/liter) in the adult male physiological range in the ovariectomized animals. These results may therefore be more relevant to atherogenesis in androgenized females, e.g., PCOS, rather than males. With the same experimental model and design, Obasanjo et al. (120 ) showed that coronary artery atherosclerosis was significantly increased by the AAS nandrolone for 2 yr compared with the intact sham-operated group (P < 0.05) but not with the ovariectomized placebo group. The groups administered nandrolone had significantly larger arteries than the other two groups. Lumen area was significantly larger in the group given nandrolone for 1 yr (deferred start by 12 months) compared with all other groups (P < 0.05). Remodeling of the vessel wall and lumen could possibly counterbalance the increased plaque size. In view of these inconsistent results and the major gender-specific action of T (118 ), it should be emphasized that data obtained on experimentally induced atherosclerosis in female animals should not be extrapolated to males. To date, no experimental studies have been performed to investigate the effects of androgens on the mechanisms underlying atherosclerosis in male monkeys.

In summary, various animal models have highlighted the existence of many different mechanisms in the evolution of atherosclerosis that can potentially be influenced by androgens. The inconsistent and conflicting results from these in vivo studies reflect the complexity of pathogenesis, the sexually dimorphic response to atherogenic triggers, as well as the gender-specific response to sex steroids.

	VI. Effects of T on Cardiovascular Risk Factors

Top Abstract I. Introduction II. The Gender Difference... III. Relationships between Serum... IV. Relationships between Serum... V. Relationships between Serum... VI. Effects of T... VII. Effects of T... VIII. DHEA(S) and CAD... IX. Estrogens and Cardiovascular... X. Summary and Conclusion XI. Clinical Implications References

 
The effects of T on cardiovascular risk factors are contradictory depending on whether associations with endogenous T or effects of exogenous T have been investigated.

A. Associations between endogenous T and cardiovascular risk factors: role of adipose tissue and insulin
Several cross-sectional population studies found statistically significant correlations between plasma levels of T and various cardiovascular risk factors that appear to be profoundly influenced by the interrelationships between T, adipose tissue, and insulin action. Furthermore, T showed opposite relationships with risk factors in men and women.

1. Observations in men.
In men, plasma T levels showed positive correlations with HDL-C and inverse correlations with triglycerides, total cholesterol, LDL-C, fibrinogen, and plasminogen activator type 1 (PAI-1; Refs. 33 51 123 124 125 126 127 128 129 130 ). However, T levels have even stronger inverse correlations with BMI; waist circumference; WHR; amount of visceral fat; and serum levels of leptin, insulin, and free fatty acids (FFA). After adjustment for these anthropometric, radiological, or biochemical measures of obesity and insulin resistance, the correlations of the cardiovascular risk factors with T but not with visceral fat or insulin lost their statistical significance (131 132 133 ). Likewise, in a case-control study of 50 men who were matched by age and ethnic background but segregated by T levels, hypoandrogenemia was associated with significantly higher BMI, WHR, higher systolic blood pressure, higher fasting and 2-h glucose and insulin levels, and higher levels of total cholesterol, LDL-C, triglycerides, and apolipoprotein (apo)B as well as with lower levels of HDL-C and apoA-I. After adjustment for BMI and WHR, only the negative correlations of T with insulin and triglycerides remained statistically significant (134 ). These findings indicate that low T in men is a component of a plurimetabolic syndrome, which is characterized by obesity, type 2 diabetes mellitus, hypertension, hypertriglyceridemia, low HDL-C, and a procoagulatory and antifibrinolytic state.

What comes first, hypotestosteronemia, obesity, or insulin resistance? On the one hand, morbidly obese and insulin-resistant men frequently have low serum levels of T (132 135 ) that increase upon weight loss (136 137 ). Estradiol levels show the opposite changes to T, with obesity and weight loss. It has therefore been suggested that obesity causes hypotestosteronemia by increased aromatization of T to estradiol in the adipose tissue (Fig. 2). In agreement with an important role of hyperinsulinemia as an etiological factor of hypotestosteronemia in obese men is the negative regulatory effect of insulin on the production of SHBG (138 ) and the inverse correlation between serum concentrations of insulin and SHBG (139 ). Also supporting a role of insulin in the determination of T levels in men, in one study infusion of insulin during euglycemic clamp increased T levels in obese men, but not in lean men (140 ). On the other hand, hypogonadal men are frequently obese with increased levels of leptin and insulin (140 141 142 143 144 145 ). Body weight, leptin levels, and insulin levels decrease upon substitution of T in hypogonadal men (146 147 148 ). Treatment of eugonadal obese men with T led to a decrease of visceral fat mass and, in parallel, improved insulin sensitivity and corrected dyslipidemia (149 150 151 ). In the opposite experiment, suppression of T by the GnRH antagonist cetrorelix increased serum levels of leptin and insulin (152 ). These latter data indicate that, in men, the dominant action in the bidirectional relationship between T and insulin is that T reduces fat mass, especially in the abdomen, and improves insulin action (Fig. 2). Mediated by the androgen receptor in adipocytes, and further up-regulated by T, androgens activate the expression of ß-adrenergic receptors, adenylate cyclase, protein kinase A, and hormone-sensitive lipase (153 154 ). As a result, T stimulates lipolysis and thereby reduces fat storage in adipocytes (Fig. 2). Androgens elicit an antiadipogenic effect in preadipocytes in vitro, whereas estrogens behave as proadipogenic hormones, effects that are related to changes in the expression of the IGF receptor (androgens and estrogens) and peroxisome proliferator-activated receptor 2 expression (estrogens; Ref. 155 ). This may explain the reduction of fat mass after androgen treatment (132 ).

View larger version (66K): [in this window] [in a new window]   Figure 2. Model of metabolic effects of T in eugonadal nonobese (A) and hypogonadal obese (B) men. T activates hormone sensitive lipase (HSL) in adipocytes and thereby decreases body fat mass (1a). This implies little aromatization of T into estradiol (2 ), i.e., facilitates the maintenance of normal T levels in nonobese men (A). Hydrolysis of body fat by HSL produces FFA, which stimulate hepatic very (V)LDL production (3a). However, this hypertriglyceridemic effect is balanced by improved insulin sensitivity in lean individuals with the result of reduced FFA release from adipocytes (1b), inhibited VLDL production (2b), and stimulated secretion of lipoprotein lipase (LPL) by the adipose tissue (3 ). Normal VLDL production and regular lipolysis of VLDL (and chylomicrons) by LPL (4 ) lead to normotriglyceridemia and, via low cholesterol ester (CE) transfer protein (CETP)-mediated exchange of CEs and triglycerides between VLDL and HDL (6 ), to normal HDL-C levels. Taken together, normal T levels, low insulin levels, and normotriglyceridemia help to suppress PAI-1 production in the endothelium (7 ). In hypogonadal men (B), low T levels impair lipolysis in adipocytes and favor obesity (1a). Enhanced aromatization of T into estradiol in obese men (2 ) further decreases T levels. Obesity causes insulin resistance with the result of increased FFA release from adipocytes (1b), disinhibited VLDL production (3 ), and decreased LPL secretion (4 ). Both increased VLDL secretion (3 ) and decreased lipolysis of triglyceride-rich lipoproteins (5 ) cause hypertriglyceridemia, which stimulates the CETP-mediated removal of CEs from HDL and thereby causes low HDL-C (6 ). Finally, hypotestosteronemia (7a), hypertriglyceridemia (7b), and hyperinsulinemia (7c) stimulate the production of PAI-1 in endothelial cells.

Because many women with PCOS are overweight, and most, if not all, are insulin resistant, it is a matter of debate whether the dyslipidemic and procoagulatory states in women with PCOS are secondary to obesity and insulin resistance (160 162 167 168 170 171 ) or whether hyperandrogenemia itself contributes to obesity, insulin resistance, and hyperinsulinemia (71 132 153 154 172 173 174 175 176 177 178 179 ). On the one hand, insulin sensitivity appears to play an important role for the pathogenesis of hyperandrogenemia in PCOS. Insulin stimulates androgen synthesis in the ovaries via its cognate receptor and the inositolglycan pathway (Ref. 180 and Fig. 3). Because the ovaries remain sensitive to insulin when other tissues such as fat and muscle are resistant, hyperinsulinemia can augment the LH- and ACTH-dependent hyperandrogenism in insulin-resistant women with PCOS (Ref. 181 and Fig. 3). In support of this, treatment of insulin resistance in women with PCOS with metformin or the insulin-sensitizer troglitazone significantly decreased serum levels of insulin as well as T, independently of BMI or gonadotropin levels (182 183 184 185 ). Concomitantly, plasma levels of HDL-C increased, and plasma levels of PAI-1 decreased (181 182 183 ). In contrast, short-term lowering of ovarian androgens by laparoscopic ovarian cautery did not alter insulin or lipid levels (186 ).

View larger version (65K): [in this window] [in a new window]   Figure 3. Model of metabolic effects of T in women. In nonobese women with normal insulin sensitivity (A), adipocytes release limited amounts of FFA (1 ) and regular amounts of lipoprotein lipase (LPL; 2). VLDLs are secreted at regular amounts by the liver (3 ) and properly hydrolyzed by LPL (4 ). Normotriglyceridemia is associated with a low exchange of triglycerides and CE between HDL and VLDL (5 ) so that HDL-C levels stay normal. Normotriglyceridemia (6a) and low insulin levels (6b) inhibit the release of PAI-1 from endothelial cells. Low insulin levels also limit the ovarian production of T (7 ). Low T levels support insulin in suppressing FFA release from adipocytes (1b), and thereby hepatic VLDL production, as well as in inhibiting PAI-1 release from endothelial cells (6c). Obese women, by contrast (B), have insulin resistance and hyperinsulinemia. In the adipose tissue, insulin resistance increases the production of FFA (1 ) and inhibits LPL secretion (2 ). Enhanced hepatic VLDL secretion (3 ) and low LPL activity (4 ) cause hypertriglyceridemia and, indirectly via enhanced CE/triglycerides exchange between VLDL and HDL (5 ), low HDL-C. Hypertriglyceridemia (6a) and hyperinsulinemia (6b) both stimulate PAI-1 secretion from endothelial cells. Hyperinsulinemia also stimulates the production of T in the ovary (7 ). Hyperandrogenemia then aggravates the detrimental effects of insulin resistance on FFA release from adipocytes (1b) and thereby on hepatic VLDL production.

In summary, the observational studies do not allow any clear conclusions on the role of T in determining cardiovascular risks because the associations between serum levels of T and cardiovascular risk factors are in opposite directions for men and women. These gender-specific correlations are also confounded by the bidirectional relationships between T, adipose tissue, and insulin sensitivity. However, the weight of current experimental evidence would suggest that low endogenous T may be the driving etiological factor for obesity, insulin resistance, and the occurrence of multiple cardiovascular risk factors in men, whereas in women defective insulin action appears to be critical for the development of the hyperandrogenemia associated with polycystic ovaries.

B. Effects of puberty on cardiovascular risk factors
Longitudinal studies were used to study the effect of puberty and hence endogenous sex hormones on cardiovascular risk factors in children. Prepubertal boys and girls do not differ significantly in their serum lipid and lipoprotein levels. In contrast to girls, in whom levels of HDL-C and LDL-C change little with puberty, sexually maturing boys experience a decrease in HDL-C and increases in LDL-C and triglycerides (200 ). However, these changes may not reflect effects of sex hormones only because they are confounded by other endocrine changes, for example in the GH/IGF-I axis, which also regulate lipoprotein metabolism.

C. Effects of exogenous T on cardiovascular risk factors
In clinical studies, the effects of exogenous T on cardiovascular risk factors differed considerably depending on the dose, route of administration, and duration of treatment, as well as the age, gender, and conditions of the recipients (Table 4). The most consistent findings were decreases in plasma levels of HDL-C, lipoprotein(a) [Lp(a)] and fibrinogen, which are accompanied by much less prominent declines of LDL-C and triglycerides.

Substitution of T in hypogonadal men or in elderly men with low to normal T or elevated gonadotropins led to minor or no decrease in HDL-C (Table 4 and Refs. 148 200 202 203 204 221 225 226 and 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 ). In a recent meta-analysis of 19 studies published between 1987 and 1999, Whitsel et al. (257 ) calculated that im administration of an average dosage of 179 ± 13 mg of T ester every 16 ± 1 d for 6 ± 1 months was associated with a decrease of 2–5 mg/dl HDL-C. The older the treated men and the longer the treatment, this decrease of HDL-C appeared to become less prominent. T substitution for up to 3 yr in men over the age of 50 yr did not produce any consistent changes in circulating lipid levels (247 258 ). Moreover, an international multicenter male contraception study found a significant decrease in HDL-C in non-Chinese but not in Chinese volunteers (259 ). Transdermal application of T or dihydrotestosterone also exerted less effect on HDL-C than im application (237 248 249 250 ).

Lowering of HDL-C by T is considered to increase cardiovascular risks because HDL-C exerts several potentially antiatherogenic actions. However, in transgenic animal models, only increases of HDL-C induced by apoA-I overproduction, but not by inhibition of HDL catabolism, were consistently found to prevent atherosclerosis (260 ). Therefore, the mechanism of HDL modification and, by inference, changes in metabolism of HDL-C rather than changes in levels of HDL-C per se appear to determine the (anti-)atherogenicity of HDL (260 261 ). Unfortunately, the mechanism and target genes by which T regulates HDL metabolism are not well understood at present. Figure 4 summarizes important steps in HDL metabolism. The production rate of HDL is determined by the hepatic and, to lesser degree, intestinal synthesis of apoA-1, the main protein constituent of HDL (262 ). The effects of T on apoA-1 production in man are not known. In mice, T increases the synthesis of apoA-1 (263 ), which at first sight is counter to the HDL-lowering effect of exogenous T. However, in mice, as opposed to man, HDL-C is increased by T and decreased by estradiol (121 ). At the catabolic site, two genes are likely to be regulated by T, namely hepatic lipase (HL) and scavenger receptor B1 (SR-B1). Regulated by corticotropin and gonadotropins, SR-B1 mediates the selective uptake of HDL lipids into hepatocytes and steroidogenic cells including Sertoli and Leydig cells of the testes as well as cholesterol efflux from peripheral cells including macrophages (260 264 ). T up-regulates SR-B1 in the human hepatocyte cell line HepG2 and in macrophages and thereby stimulates selective cholesterol uptake and cholesterol efflux, respectively (264A ). HL hydrolyzes phospholipids on the surface of HDL, thereby facilitating the selective uptake of HDL core lipids by SR-B1 (260 227 ). The activity of HL in postheparin plasma is increased after administration of exogenous T (225 226 229 238 265 ) and slightly decreased by suppression of T after GnRH antagonist treatment (152 ). However, castration of male rats did not cause significant changes in postheparin plasma activity of HL or in HL mRNA levels in the liver. Subsequent substitution of T raised HL activity without changing HL mRNA expression (266 ). This raises the possibility that T does not directly regulate the HL gene. In agreement with this, we did not observe any change of HL activity in the supernatants of HepG2 cells that were incubated with T (264A ). The increase in both SR-B1 and HL activities is consistent with the HDL-lowering effect of T. Up-regulation of HL and SR-BI also explains why T induces the most prominent changes in HDL subclasses HDL₂ and LpA-I, because these particles are preferred substrates of HL and SR-BI over small HDL₃ and apoA-II-containing HDL. Interestingly, in transgenic mice, overexpression of HL caused a dramatic fall in HDL-C but inhibited rather than enhanced atherosclerosis (260 264 227 ). This again demonstrates the difficulty in extrapolating the HDL-lowering effect of T to increased cardiovascular risk.

View larger version (42K): [in this window] [in a new window]   Figure 4. Pathways of HDL metabolism and regulation by T and estradiol (E2). Mature HDL3 and HDL2 are generated from lipid-free apoA-I or lipid-poor preß1-HDL as the precursors. These precursors are produced as nascent HDL by the liver or intestine or are released from lipolyzed VLDL and chlyomicrons, or by interconversion of HDL3 and HDL2. ATP-binding cassette 1 (ABCA1)-mediated lipid efflux from cells is important for initial lipidation; lecithin cholesterol acyl transferase (LCAT)-mediated esterification of cholesterol generates spherical particles, which continue to grow upon ongoing cholesterol esterification, and phospholipid transfer protein (PLTP)-mediated particle fusion and surface remnant transfer. These mature HDL particles also continue to accept cellular cholesterol by processes that are facilitated by SR-BI and LCAT. Larger HDL2 are converted into smaller HDL3 upon CETP-mediated export of CEs from HDL onto apoB-containing lipoproteins, SR-B1-mediated selective uptake of CEs into liver and steroidogenic organs, and HL-mediated hydrolysis of phospholipids. HDL lipids are catabolized either separately from HDL proteins, i.e., by selective uptake or via CETP-transfer, or together with HDL proteins, i.e., via uptake through as-yet unknown HDL receptors or apoE receptors. Both the conversion of HDL2 into HDL3 and the PLTP-mediated conversion of HDL3 into HDL2 liberate lipid-free or poorly lipidated apoA-I, which is either reused for the formation of mature HDL or is filtrated into the kidney. Gray arrows represent lipid transfer processes, and black arrows represent protein transfer processes. The hepatic expression and activity of both HL and SR-B1 was shown to be up-regulated by T and down-regulated by estradiol. In addition estradiol up-regulates the hepatic expression and secretion of apoA-I. These actions of T and estradiol are in good agreement with their lowering and increasing effect on HDL-C, respectively. In addition, both T and estradiol stimulate SR-BI expression in macrophages and thereby cholesterol efflux from these cells onto lipidated HDL. [Modified from Refs. 260 and 261 .]

Lp(a) levels are generally assumed to remain stable throughout life. However, estrogens, progestins, GH, and T₄ can lower Lp(a) levels (273 274 ). Likewise, administration of R to orchidectomized patients with prostate cancer (275 276 ), as well as administration of supraphysiological doses of T enanthate to healthy men, decreased serum levels of Lp(a) significantly by 25–59% (203 218 236 252 277 ). Lp(a) levels were increased by 40–60% in controls and in patients in whom endogenous T was suppressed by treatment with the GnRH antagonist cetrorelix or the GnRH agonist buserelin (219 275 276 278 279 ). The Lp(a)-lowering effect of T is independent of estradiol because Lp(a) levels were also lowered when T was administered in combination with an aromatase inhibitor, testolactone (277 ). Animal studies have also provided evidence for the involvement of T in the regulation of Lp(a) levels. Frazer et al. (280 ) observed that Lp(a) levels decrease in male, but not in female, apo(a)-transgenic mice after sexual maturation. Castration of male animals restored initial Lp(a) levels, which decreased again upon application of dihydrotestosterone. However, it is also important to note that, similar to the changes observed in HDL-C, treatment of hypogonadal men with physiological dosages of T did not cause large changes in Lp(a) levels (Table 4).

It is not known how T regulates Lp(a). Turnover studies have shown that Lp(a) levels are mainly determined by production. The majority of newly synthesized apo(a) is degraded intracellularly before secretion. The larger the apo(a) isoforms, the more they are degraded intracellularly and, hence, the less is secreted. Interestingly, estradiol decreases Lp(a) production, but it is not known whether estradiol regulates the transcription of the apo(a) gene or the posttranslational processing of apo(a) (267 281 ).

It is also not known whether changes in Lp(a) induced by T will affect cardiovascular risk. Interestingly, however, in the Heart and Estrogen/Progestin Replacement Study (HERS; Ref. 282 ), postmenopausal hormone replacement therapy prevented coronary events only in those women who had elevated Lp(a) at baseline and experienced a decrease of Lp(a) levels by treatment with conjugated equine estradiol and medroxyprogesterone.

3. The hemostatic system.
In agreement with an important role of thrombus formation in the pathogenesis of acute coronary events and stroke, prospective studies have identified various hemostatic variables as cardiovascular risk factors (283 ). The risk of MI increases with plasma levels of the thrombogenic factors fibrinogen and factor VII, as well as with plasma levels of the fibrinolysis inhibitor PAI-1 or tissue plasminogen activator antigen, which represents the inactivated form (283 ). Platelet aggregability is another important factor that determines thrombogenicity and, thereby, cardiovascular risk. T was shown to regulate plasma levels of fibrinogen and PAI-1. Administration of supraphysiological dosages of T