TESTOSTERONE AND INSULIN RESISTANCE IN MEN: EVIDENCE FOR A COMPLEX BI-DIRECTIONAL RELATIONSHIP

ABSTRACT

Given that low testosterone levels predict type 2 diabetes mellitus (T2DM) in men, we sought to dissect the relationship between testosterone and insulin sensitivity in men. We showed a positive correlation between serum testosterone levels and insulin sensitivity in men across the full spectrum of glucose tolerance. Men with hypogonadal testosterone levels were twice as insulin resistant as eugonadal controls. By dissecting the hypothalamic-pituitary-gonadal axis, we demonstrated that the cause of low testosterone was a decrease in testicular responsiveness to luteinizing hormone (LH). Using functional and genetic studies, we provided evidence that hypogonadism may induce insulin resistance by causing mitochondrial dysfunction. We also showed that weight loss can increase testosterone levels and reverse hypogonadism in over 50% of obese men with impaired glucose tolerance. We concluded that the relationship between testosterone and insulin resistance in men is bidirectional. Our data emphasized the importance of lifestyle modification as a key therapeutic step in the management of the hypogonadal male.

INCREASE IN USE OF TESTOSTERONE

The onset of the millennium was accompanied by a significant surge in both testosterone testing and initiation of testosterone replacement, especially in men without clear indications (1). When prescribing trends are compared between countries, it is evident that this increase in testosterone use was most marked in the United States, with relatively little change in the United Kingdom (1). The resurgence of interest in a therapy that first entered the clinical arena more than 60 years earlier most likely reflects the following factors: direct-to-consumer advertising, development of more patient-friendly testosterone formulations, and an expansion of the clinical indications for testosterone replacement.

In 1997, the U.S. Food and Drug Administration decided to approve televised direct-to-consumer advertising for prescription meds, a practice not permitted in any other country than New Zealand. While increasing awareness of and demand for a medication may be beneficial if the risk-benefit ratio of the drug is favorable, that assumption is questionable for many men taking testosterone for nonspecific symptoms without a clear diagnosis. In a study of 75 designated U.S. marketing areas, a significant increase in televised direct-to-consumer ads for testosterone therapies was observed between 2009 and 2013, and both branded and condition awareness advertisements were, in turn, associated with increased testosterone testing and initiation (2).

A second important factor that contributed to the more widespread use of testosterone was the development of more patient-friendly formulations. For decades, the only method of delivering testosterone was through a deep, painful intramuscular injection every two weeks. However, the approval of a transdermal testosterone delivery system in the form of a gel meant that testosterone could now be administered easily, conveniently, and in a pain-free manner (3). The demonstration that six months of treatment with this testosterone gel formulation improved sexual function, increased lean body mass and muscle strength, and decreased fat mass (4) caught the attention of the lay press and was featured on the cover of TIME magazine as a means of potentially giving men back their “edge” as they age.

TESTOSTERONE AND THE AGING MALE

The normal aging process is accompanied by physiological changes in target organs that are sites of androgen action. Given the decline in androgen levels with aging that has been documented in epidemiological studies, it was postulated that some features of aging such as frailty and loss of vigor could be attributed to androgen deficiency. As a result, the clinical indications for androgen replacement were expanded to include the aging male.

As the idea took hold that testosterone therapy might slow the progression or even reverse some features of aging, the recognition of the huge market potential led to the sprouting up of men’s health clinics across the country. What had been a small niche product suddenly exploded into a $1.6 billion market. Magazine and TV advertisements encouraged men to have their testosterone levels checked if they felt in any way below par and touted the myriad benefits of testosterone replacement. However, as I and others pointed out, there were insufficient data from clinical trials at that time to permit any major conclusions about the role of androgen replacement in the treatment of age-related physiological changes (5). In addition, it is important to appreciate that while the term “andropause” was introduced to draw an analogy between age-related changes in testosterone in men and that of menopause in women, there are clear and important differences. All women ultimately experience a drop in estradiol levels into the menopausal range. By contrast, men begin to experience a decline in testosterone levels in the 4th decade; given that the rate of decline is only about 2% per year, any decline remains within the normal range for many men (6). In fact, a large epidemiological study of 3,369 community dwelling men in Europe found that the prevalence of hypogonadism when defined by a combination of clinical and biochemical criteria was only 2% although it increased with age, body mass index (BMI), and number of comorbid illnesses (7).

ENDOCRINE SOCIETY CLINICAL PRACTICE GUIDELINES

As prescriptions for testosterone skyrocketed, safety concerns began to emerge. A number of papers reported an increased risk of cardiovascular events (8,9) calling into question the safety of testosterone and the appropriateness of its use, especially in the aging male. Given the uncertainty and controversy about the indications for testosterone use, the Endocrine Society convened a panel of experts in the field to provide guidance to physicians. In this clinical practice guideline, we outlined the challenges to making a diagnosis of hypogonadism, highlighting that there is debate as to the threshold testosterone level for diagnosing hypogonadism and whether it is the same for all target tissues (10). Many factors can influence the level of testosterone at any given time including circadian rhythms, medications, and concomitant illness. To avoid over-treatment, it is important to recognize that approximately one-third of men with low testosterone levels have normal levels on repeat testing. Therefore, a low testosterone level must always be confirmed by repeat testing before a diagnosis of hypogonadism can be established. In addition, data from our group led to the recommendation that testosterone be measured in the fasting state. In a study of 74 men with a spectrum of glucose tolerance, we showed that administration of a standard 75 g glucose load caused a 25% decrease in testosterone levels 60 minutes after ingestion (11). Importantly, in 15% of men, testosterone levels declined into the hypogonadal range, indicating the potential to make an erroneous diagnosis of hypogonadism if testosterone is measured after eating.

Impact of Testosterone Replacement

Over the past decade, a number of randomized, placebo-controlled clinical trials have been conducted to determine the impact of testosterone replacement on a variety of clinically important endpoints in middle-aged and older men (12–15). Details of the studies are summarized in Table 1. Three of the trials used a testosterone gel to restore testosterone levels to the normal range, while the fourth combined a testosterone injection with lifestyle changes. Duration of treatment ranged from one to four years. When evidence from all the trials was analyzed, testosterone proved to have positive, negative, and neutral effects (Table 2). For some parameters like prostate cancer, the studies were not adequately powered and/or of sufficient duration to rule out an effect.

Table 1.

Clinical Trials Examining Impact of Testosterone Therapy in Middle-Aged and Older Men

Trial	Participants	Intervention	Primary Outcome
TEEAM	308 men >60 years	Testosterone gel vs. placebo for 3 years	Carotid artery intimal thickness and coronary artery calcium score
T trials	790 men >65 years	Testosterone gel or placebo for 1 year	Sexual activity, physical function, vitality, cognition, hemoglobin, bone mineral density, noncalcified plaque
TRAVERSE	5,246 men 45–80 years	Testosterone gel vs. placebo for up to 4 years	Major adverse cardiovascular event (nonfatal myocardial infarction, nonfatal stroke, cardiovascular disease death)
T4DM	1,007 men 50–74 years, impaired glucose tolerance or newly diagnosed type 2 diabetes	Intramuscular testosterone undecanoate with lifestyle vs. placebo for 2 years	Prevention or reversal of type 2 diabetes

Table 2.

Summary of Impact of Testosterone in Middle-Aged and Older Men

The most striking benefits were seen for libido, body composition, and correction of anemia. While bone density improved, there was a surprising increase in fracture rate although not the typical fragility fractures of hip and spine typically seen in men with osteoporosis. Reassuringly, there was no evidence of an increased risk of cardiovascular disease, although a slight increased risk of venous thromboembolism and atrial fibrillation was observed.

Relationship Between Testosterone and Insulin Resistance

We and others have shown that testosterone levels are significantly lower in men with impaired glucose tolerance and T2DM than in normoglycemic controls (16). In a study of men with a spectrum of glucose tolerance, we showed that increasing insulin resistance assessed by a glucose clamp is associated with a decrease in the Leydig cell response to stimulation with human chorionic gonadotropin (hCG) (Figure 1). Potential mechanisms include resistance to the normal stimulatory effects of insulin on the hypothalamic-pituitary-gonadal axis, suppression by pro-inflammatory cytokines, and high leptin levels.

Fig. 1.

Positive correlation between insulin sensitivity (M) and increase in testosterone post stimulation of the testes with hCG.

In terms of addressing causality, it has been shown that improving insulin sensitivity by losing weight, whether by lifestyle changes or bariatric surgery, results in a significant increase in total testosterone levels (17). While a weight loss of 5% body weight is sufficient to increase total testosterone levels, a 15% weight loss is needed to increase free testosterone levels due to concomitant changes in sex hormone binding globulin levels (18).

In prospective studies, low testosterone levels have been shown to predict future development of T2DM. Similarly, use of androgen deprivation in men with prostate cancer significantly increases the risk of T2DM with a hazard ratio (HR) of 1.4 (19). We therefore sought to dissect the relationship between testosterone and insulin resistance in men by performing a detailed hormonal and metabolic evaluation in 60 men with a spectrum of glucose tolerance. Insulin sensitivity was measured using a hyperinsulinemic-euglycemic clamp. Mitochondrial function was assessed by measuring maximal aerobic capacity (VO₂ max) and expression of oxidative phosphorylation genes in skeletal muscle.

We showed a positive correlation between serum testosterone levels and insulin sensitivity in men across the full spectrum of glucose tolerance (Figure 2A). Men with hypogonadal testosterone levels were twice as insulin resistant as eugonadal controls. We also showed a positive correlation between physiological markers of mitochondrial function such as maximal aerobic capacity (VO₂ max; Figure 2C) as well as genes involved in oxidative phosphorylation (UQCRB, Figure 2D). These data indicate that low serum testosterone levels are associated with an adverse metabolic profile and suggest a novel unifying mechanism for the previously independent observations that low testosterone levels and impaired mitochondrial function promote insulin resistance in men.

Fig. 2.

Correlation between insulin sensitivity (M) and serum testosterone (T) levels (A) and SHBG levels (B) in 60 men; 27 had NGT (□), 12 had IGT (△), and 21 had type 2 diabetes (•). Shaded area represents values for subjects with hypogonadal testosterone levels (i.e., <9.7 nmol/L). In a subset of 42 men, serum testosterone levels were corelated with maximal aerobic capacity (VO₂ max) (C) and expression of UQCRB in skeletal muscle (D). Seventeen men had NGT (□), 7 had IGT (△), and 18 had type 2 diabetes (•).

While epidemiologic studies point to a bidirectional relationship between testosterone and insulin resistance, a review of the literature on the impact of testosterone administration on glucose homeostasis yielded disappointing results with only modest changes seen in insulin sensitivity and no consistent change in markers of glycemic control such as HbA1c (20). A singe Australian study (known as T4DM for Testosterone for the prevention of Diabetes Mellitus) of men with impaired glucose tolerance or newly diagnosed type diabetes showed that the combination of testosterone and lifestyle changes reduced absolute risk of T2DM in men at high risk compared to those randomized to lifestyle changes alone, 12.4% versus 21% (13).

CONCLUSIONS

Incidence of hypogonadism in men increases with age but, unlike menopause, is not inevitable. Recent clinical trials highlight that testosterone replacement in older men has both benefits and risks. The decision to prescribe should be guided by severity of testosterone deficiency, burden of symptoms, and presence of comorbid illness. Given the bidirectional relationship between testosterone and both obesity and insulin resistance, a holistic approach focused on healthy lifestyle behaviors should always be encouraged when managing a hypogonadal male.

DISCUSSION

Liedtke, Durham: Your talk reminded me of another medical folly that happened over the prescription of opioids. Remarkably, it happened in the United States but not in neighboring Canada or Mexico or Europe. I was trained on the use of opioids and when I first learned about it, I said, “Huh?” However, it has been a striking reality that reverberated through the states. In thinking about psychological mechanisms that led to this disaster, how about the handing out of testosterone mostly to men who don’t have hypogonadism? That is clearly what happened from the data you’ve shown, or am I mistaken? Are there psychological or weird market mechanisms at play that share a feature with what happened with opioids and what is now happening with this testosterone prescription that is not matched anywhere else in the world?

Hayes, Boston: Good point! I think part of it is the direct-to-consumer advertising that leads patients to believe that their nonspecific symptoms can be corrected with a simple prescription. If you don’t do the testing under appropriate conditions and with the right assays, it’s easy to misclassify somebody as being hypogonadal. I think that’s part of the difference between the United States and other parts of the world.

Stevenson, Nashville: Thank you so much. I’m very, very grateful for this. This information will be very helpful in a clinic, and I was scribbling frantically during your presentation. We clearly are seeing many more people who are taking high levels of testosterone in gel form because they tell themselves it’s one dose and they apply it all over. This is both people for whom it has been prescribed and people who are using it for bodybuilding. We clearly see an increase in hypertrophic cardiomyopathy, and we see patients who have decreased ejection fraction without much dilatation and often, as you mentioned, with the high hematocrit, which I’m sure is also deleterious. This is something that is not widely recognized, but we are seeing an increase in my opinion.

Hayes, Boston: There is a distinction between testosterone per se and the androgenic anabolic steroids that are certainly more detrimental in terms of hypertrophy of the heart. You don’t get aromatization to estrogen, and they also have much more negative effects on the lipid profile with very low HDL levels.