by Frances J. Hayes, MB, FRCPI

A link between testosterone and cardiovascular function has been suggested by data comparing the prevalence of cardiovascular disease in men and women.  Men have a higher prevalence of cardiovascular disease than age-matched pre-menopausal women, but this sex difference disappears after menopause (33).  It has been suggested that differing levels of sex hormones may explain the finding that as men and women get older, the risk of developing cardiovascular disease becomes similar.  A number of studies have shown that high levels of testosterone are not associated with the development of coronary artery disease in men or women (34).  Other research has shown that testosterone levels may affect other aspects of cardiovascular health.  Low testosterone is more common in patients with type 2 diabetes, a condition that itself is associated with multiple cardiovascular comorbidities (35). 

In a study of 794 healthy men 50 to 91 years of age who were studied for up to 20 years, it was shown that as total testosterone levels fell below 300ng/dL, the risk of all-cause mortality and cardiovascular disease increased compared with higher testosterone levels (36). All-cause mortality rates did not change in men with total testosterone levels over 300mg/dL (36).  The influence of testosterone on morality rates and cardiovascular disease rates may be the result of a direct effect of testosterone on coronary vasculature.  Testosterone can cause a dose-dependent increase in coronary artery diameter, resulting in an increase in coronary artery blood flow (37). 

Several studies were performed to determine the effect of testosterone on different chronic cardiovascular and metabolic parameters.  A 12-week, double-blind, randomized study of 46 men with stable angina and low testosterone levels (12-13 nmol/L) examined the effect of a daily 5 mg testosterone patch versus placebo treatment (38).  There was a significant increase in time to 1 mm ST-segment depression after 12 weeks of testosterone treatment compared with the time increase in the placebo group (P=0/02) (38).  The frequency of angina did not change in either treatment group at any point in the study (38).  In another study, 10 men with stable angina and hypogonadism (mean total testosterone level: 4.2 nmol/L) participated in a 4-week, randomized, single-blind clinical trial that examined the effect of a 100 mg biweekly intramuscular injection of testosterone versus placebo (39).  The time to 1 mm ST-segment depression increased compared with placebo (P=0.002), though the frequency of angina did not change (39). 

Most of the data relating to testosterone and the formation of atheroscierotic plaques are favorable.  In culture aortic ring segments taken from male rabbits exposed to aortic endothelial denudation as a method of inducing neointimal plaque formation, exposure to testosterone (10 and 100ng/mL) reduced plaque formation compared with the control group (40).  Similarly, another study showed that castration was used to induce arterial plaque formation in rabbits, whereas TRT (6 mg/kg/2 weeks) reduced plaque area and aortic intimal thickness.  In addition to increasing the fibrous cap thickness and collagen content (41).  However, other data using human monocyte-derived macrophages indicated that in vitro exposure to ditrydrotestosterone (40 nmol/L) was capable of upregulating multiple atherosclerosis-related genes, such as acyl coenzyme A, lysosomal acid lipase, and caveolin-2 (42).  In vivo data from the Rotterdam Study confirmed an inverse relationship between endogenous total testosterone levels and the development of aortic plaques as measured by standard x-ray in 504 men (43).  When divided into tertiles (0-280, 281-360, 361-1050 ng/dL), the data showed that as the levels of total testosterone increased, the risk of both severe aortic atherosclerosis and the progression of severe aortic atherosclerosis declined (43).  

Type 2 diabetes is associated with a much higher prevalence of hypogonadism compared with controls, regardless of age or body mass index (44).  Men with a total testosterone of less than 445 ng/dL are twice as likely to develop metabolic syndrome or diabetes than are men with higher testosterone levels (45).  A meta-analysis of cross-sectional studies that examined the relationship between type 2 diabetes and testosterone levels found that total testosterone levels were significantly lower (mean difference, -77.6 ng/dL; P<0.001) in men with type 2 diabetes versus controls (45).  Furthermore, the risk of developing type 2 diabetes was 42% lower in men with total testosterone levels between 450 and 605 ng/dL than in men who had testosterone levels between 213 and 476 ng/dL (46).  An observational study of 73,196 male Medicare patients found that androgen deprivation therapy for prostate cancer significantly increased the risk of both incident diabetes (P<0.001) and incident coronary heart disease (P<0.001) (47). 

Initiating TRT in men who are hypogonadal can have positive metabolic and cardiovascular effects.  In a study of 24 men with hypogonadism and type 2 diabetes, the administration of testosterone (200 mg/2 weeks, intramuscular injection) improved insuling sensitivity as measured by the homeostatic model index (P<0.02) and reduced hemoglobin A1C (P<0.03) levels (48).  Another study examined the effect of TRT (80 mg of testosterone undecanoate twice a day for 8 months) in 23 obese men (body mass index >25) over the age of 45 (49).  Men who received testosterone showed a significant increase in insulin sensitivity (measured by glucose disposal rate; P<0.01) and a reduction in visceral fat (P<0.05) compared with baseline values (49).  These effects were not seen in the placebo group (49).  

In contrast to the findings that RT has a positive effect on insulin sensitivity, some studies have yielded different results.  A 6 month study of 55 men over 50 with type 2 diabetes (mean age 63.6 years) found that testosterone supplementation (150 mg/2 weeks via intramuscular [IM] injection) had no effect on glycemic control (50).  In a study of 61 healthy eugonadal men between the ages of 18 and 35 in whom hypogonadism was induced by the administration of a gonadotropin releasing hormone (GnRH) agonist, testosterone supplementation (25-600 mg/week, IM) had no effect on insulin sensitivity regardless of the dose administered (50,51).  Another study found that recombinant human chorionic gonadotropin had no effect on insulin sensitivity when administered to men over 60 with low total testosterone levels < or equal to 15 nmol/L (52). 

The mechanisms underlying the association between testosterone and metabolic and cardiovascular health have been examined.  A study of 22 patients with prostate cancer revealed that 3 months of GnRH agonist therapy increased fasting insulin levels (p-0.02), which was positively correlated with a concurrent increase in fat mass (P=0.008) (53). These data show that changes in body composition induced by testosterone play an important role in mediating its effects on insulin sensitivity.  Evidence suggests that mitochondrial dysfunction may be involved in the ability of testosterone to influence insulin sensitivity.  A study of 50 men (mean age 60.5) with varying levels of insulin sensitivity (healthy, obese and those newly diagnosed with type 2 diabetes) examined the relationship between testosterone levels, insulin sensitivity, and mitochondrial function (54). Testosterone levels were positively correlated with both insulin sensitivity measured by the euglycemic-hyperinsulinemic clamp technique (P<0.05) and expression in skeletal muscle of genes involved in mitochondrial oxidative phosphorylation (P<0.0001) (54).  

Other data show that testosterone can also directly modulate lipolysis.  In a study of 11 men over 40 with a mean body mass index of 33, the administration of testosterone (40 mg, 4 times per day, oral) reduced both lipoprotein lipase activity in abdominal adipose tissue (P<0,01) and the waist/hip circumference in 9 of 11 men (56).  In another study of 27 men with symptomatic hypogonadism, the administration of testosterone (100mg/2 weeks/injected IM) resulted in a decrease in the proinflammatory cytokine tumor necrosis factor (P<0.01) (56).  Testosterone has additional positive effects on the cardiovascular system in hypogonadal men, as indicated by a reduction in total cholesterol, low-density lipoproteins, and triglyceride levels (57).  

SUMMARY 

• Low testosterone levels in men are associated with increased cardiovascular morality, obesity, type 2 diabetes, and metabolic syndrome. 
• Small interventional studies show favorable effects of TRT on ischemia, lipid levels, proinflammatory cytokines, and possibly insulin sensitivity. 
• The effect of TRT on myocardial infarction in hypogonadal men is currently unknown.