Introduction

For many years, the higher cardiovascular (CV) risk in men has been wrongly attributed to the negative systemic effects of testosterone (T); however, nowadays, a large amount of evidence suggests that, indeed, in males the low levels of T could be associated with morbidity and mortality. Also, the high CV mortality could be due to some physical and behavioral characteristics typical of male sex, such as the central distribution of fat adiposity or the smoking or alcohol consumption habit [1]. In any case T deficiency gains relevance because, besides occurring as a consequence of rare pathological conditions which are almost completely of endocrinological competence, can also take place with natural aging, causing a “male menopause” state, also defined as “late-onset hypogonadism” (LOH). This condition is generally associated with metabolic disturbances and increased CV risk, and often requires testosterone replacement therapy (TRT).

In this short review, we aimed to summarize the present state of the art concerning the association between T deficit and CV disease.

Hypogonadism

Male hypogonadism diagnosis requires unequivocally low levels of serum T (total T < 8 nmol/l and free T < 225 pmol/l) together with signs and symptoms consistent with hypogonadism (e.g., decreased libido, erectile dysfunction (ED), decrease of muscle mass or strength, increase of body fat, low bone mineral density or osteoporosis or decreased vitality) [2]. Free T calculation according to Vermeulen’s formula (at http://www.issam.ch/freetesto.htm) can be helpful in clinical states in which sex hormone-binding globulin might be decreased (e.g., obesity, acromegaly, andhypothyroidism) or increased (e.g., hepatic illness, hyperthyroidism, and use of anticonvulsants) [2]. As far as symptomatology is concerned, all the andrology society guidelines recognize that sexual symptoms are important in defining symptomatic hypogonadism and that the presence of at least one of the hypogonadism-related symptoms is an essential element to identify a clinically significant hypogonadism; among the relevant symptoms, hypoactive sexual desire is considered the one which is most commonly associated with male hypogonadism, while ED is the one that more frequently leads the patient to a medical consultation [2].

In cases of T deficiency, TRT is usually warranted when, in symptomatic men, circulating total T is below 8 nmol/l. When total T is repetitively between 8 and 12 nmol/l, in the presence of typical hypogonadal symptoms, a T treatment trial should be considered. Conversely, when total T is above 12 nmol/l, TRT should not be used [2].

Protective effects of T on CV system

T seems to have several beneficial effects on CV system (Table 1). First of all, a large body of evidence suggests a direct vasodilator effect of this hormone on many vascular districts, including the coronaric one, in addition to a role in endothelium repair [1]. Second, studies carried out in hypogonadal subjects demonstrated that TRT is able to decrease the production of inflammatory cytokines, such as tumor necrosis factor-α, interleukin-1β and interleukin-6, and to increase the level of the antiaterogenic interleukin-10, while, in animal models of hypogonadism, it has been demonstrated that TRT can decrease the occurrence of typical atherosclerotic processes such as neointima formation, fatty streak accumulation and smooth muscle cells proliferation in the intima layer [1]. Third, in vitro studies and evidence collected in patients with low levels of T before and after TRT suggested that androgens could decrease arterial stiffness probably by inhibiting vascular smooth muscle calcification [1]. Finally, T seems to have a weak anticoagulant activity that could consist in stimulating tissue factor pathway inhibitor and tissue plasminogen activator expression and in inhibiting plasminogen activator inhibitor type 1 secretion by the endothelium [1].

Table 1 Protective effect of T on CV system and association between T deficiency and CV events
Full size table

Relationship between T and metabolic syndrome

A large amount of evidence suggests a connection between hypogonadism and metabolic syndrome. In this regard, in a recent meta-analysis about the cross-sectional studies evaluating the association between metabolic syndrome and hypogonadism, it has been reported that patients affected by metabolic syndrome showed significantly lower T plasma levels in comparison to age-matched subjects [3]; this relationship was independent from the ED and from the various diagnostic criteria of the metabolic syndrome. Link between hypogonadism and metabolic syndrome/CV complications could be bidirectional: in fact, while some studies suggest that in patients affected by prostate cancer (and, therefore, treated with androgens ablation) an increased prevalence of CV disease and metabolic syndrome occurs, other works show that several chronic diseases, including obesity, metabolic syndrome and type 2 diabetes, exert a suppressing effect on T [1]. The T decrease could entail, through the androgen receptor mainly expressed in visceral fat, a lack of negative regulation both of the differentiation of preadipocytes into mature adipocytes and of the differentiation of mesenchymal stem cells to adipocytes [2]. The increase of visceral fat, a well-known CV risk factor [4–6], could, in turn, provoke the worsening of hypogonadism through several mechanisms that may be represented by the augmentation of insulin-resistance, of adipose tissue aromatasic activity, of leptin or tumor necrosis factor-α secretion together with that of other adipokines [2]. As far as insulin-resistance is concerned, it has been demonstrated that, physiologically, insulin may act both at the hypothalamic and at the testicular level, increasing, in the first case, the synthesis and release of gonadotropin-releasing hormone and, in the second case, directly stimulating T secretion. These effects could be greatly blunted in central obesity, a condition characterized by insulin-resistance [2]. Furthermore, luteinizing hormone (LH) secretion could be inhibited by a mechanism of negative feedback exerted by the increased estrogenic synthesis that occurs as a consequence of the hyperexpression of P450 aromatase in adipose tissue [2]. It is also worth considering that leptin, an adipocyte-derived adipokine that circulates in plasma at concentrations that parallel the amount of fat reserves, has been demonstrated to contribute, in case of excess, to the development of hypogonadism in males [2]. Finally, obesity is also associated with increased circulating levels of tumor necrosis factor-α, a cytokine whose intratesticular augmentation has been associated with blunted T response to the human chorionic gonadotropin stimulation and with decrease of LH release in consequence of its activity at pituitary or hypothalamic level [2].

However, it cannot even be ruled out that reduction of T, as evident in visceral obesity and in several chronic diseases, could represent a protective mechanism in order to temporarily inhibit some of the T dependent functions like reproduction and physical labor that are not desirable when the general physical state is debilitated.

Relationship between T and CV events

Low levels of T have been found to be associated with different types of CV diseases including coronary artery disease (CAD), heart failure (HF) and stroke (Table 1).

Regarding CAD, Phillips et al. [7] have reported an inverse relationship between free T levels and the degree of CAD in 61 men. In addition English et al. [8] have found lower values of T in men with CAD compared with matched controls with normal coronary angiography. These findings have been further confirmed by Rosano et al. [9] who, evaluating the relationship between plasma sex hormone levels and presence/degree of CAD in patients undergoing coronary angiography and in matched controls, found that subjects with CAD had lower T and estradiol levels than healthy controls and that the levels of these hormones were inversely correlated to the degree of CAD.

Concerning HF, some authors reported that, in men affected by this pathology, total and free T are reduced proportionally to the HF severity and, in this condition, the prevalence of male hypogonadism is higher than 40 % [1]. It is also worth noting that low levels of T could play a role in favoring the onset of some peculiar features of the advanced HF like decreased exercise capacity, diminished muscle mass and energy handling, fatigue dyspnea and final cachexia [1].

Production of total T could be linked to the incidence or the severity of stroke or transient ischemic attack, to the 6-month mortality and to the infarct size [10, 11]. This hormonal fall could promote the thickening of carotid artery intima-media, the formation of abdominal aortic aneurism and, even, the onset of lone atrial fibrillation [1].

Furthermore, some epidemiological studies point out a relationship between T decrease and mortality [1]. In this regard, it is important to highlight that in the “Health in men” study, which analyzed 3,443 men older than 70 years, low total and free T concentrations were associated with reduced event-free survival and lower total T predicted increased incidence of stroke or transient ischemic attack after adjustment for many confounding factors like age, waist-hip ratio, waist circumference, smoking, hypertension, dyslipidemia, and medical comorbidity [11]. Furthermore, in a prospective study (European Prospective Investigation Into Cancer in Norfolk) examining the prospective relationship between endogenous T concentrations and mortality in 11,606 men, it has been reported that endogenous T concentrations at baseline were inversely related to mortality due to all causes, CV disease, and cancer [12]. In addition, in patients suffering from ED (a condition strongly associated not only with hypogonadism but even with CV risk as the smaller penile arteries reach critical narrowing with insufficient blood flow earlier than larger vessels) low levels of T have been found to be associated with higher mortality from major adverse CV events [13]. However, recent meta-analyses, although confirming the association between low levels of T with the increased risk of overall and CV mortality, did not find a relationship between low T and incident CV disease, suggesting, therefore, that T decrease could only represent a marker of poor general health [14, 15].

Effects of TRT on CV disease and metabolic syndrome

The beneficial effect of TRT on CV risk is a very controversial point. In this regard, although a recent meta-analysis on the effects of TRT on metabolic syndrome found that TRT was associated with a significant reduction of fasting plasma glucose, homeostatic model assessment index, triglycerides, waist circumference and with an increase of high-density lipoprotein cholesterol [3], three meta-analyses failed to find any differences between the group of subjects treated with T and the placebo group either regarding all CV events or a specific type of event [1]. However, Caminiti et al. [16] who analyzed the efficacy of TRT in patients affected by HF, found that TRT was able to improve exercise capacity, muscle strength and baro-reflex sensitivity. The issue remains undefined especially because the trials are often limited by the inadequate design or the insufficient statistical power.

Conclusions

Despite many, in vitro and in vivo, findings indicate that T plays a protective effect on CV system and that, therefore, its decrease could entail an increase of CV risk, several works did not support a relationship either between hypogonadism and CV disease or between TRT and CV events. However, low values of T seem to be associated with increased risk of CV death in community-dwelling men and in men with ED. Great uncertainty remains about the possibility that T deficiency could be the cause or the consequence of the increased risk of overall and CV lethality. Further and larger studies evaluating the impact of TRT on mortality are therefore needed.