Testosterone and the Metabolic Syndrome

Testosterone and the metabolic syndrome

Stefan Arver, MD, PhD, Reproductive Medicine Center, Karolinska Institute, Stockholm, Sweden

There are a number of variations on the definitions of the metabolic syndrome but the key points are listed here from the WHO statement.

It’s a state where it’s a combination of hyperinsulinemia, reflecting resistance toward the action of insulin, in combination with abdominal obesity which can be defined as the ratio between the waist and the hip circumference; or solely by the waist girth. It is also a disarrangement of the lipoprotein metabolase and increased levels of triglycerides and cholesterol. It’s also associated to hypertension. This is a fairly malignant syndrome that covers a large morbidity and mortality. The prevalence reaches staggering numbers in the general population in northern Europe around 20%, in the US, maybe 25%, and in certain ethnic groups in specific geographic areas, the prevalence numbers up to 60% and 80%. There is a tremendous increase in the prevalence of this condition and it causes a lot of health concerns and, of course, also a burden on the health economy of our countries. The numerous aspects of the metabolic syndrome has been addressed in various ways and you can enter into studying different specifics, but in light of being a generalised metabolic derangement, one may seek for a common ground in development, or a pathway that changes the regulation of the system. One of those entries into the metabolic syndrome are studies of testosterone, and changes in testosterone levels, that might be a crucial factor in regulating the driving force, as I see it, behind the metabolic syndrome, and that is the accretion of abdominal fat.

Going back to reality this is a patient you are pretty familiar with. It’s a middle-aged male, 56 years old, to the left. He has a low testosterone level. He has a conspicuous accumulation of intraabdominal fat, he could actually be a suspicious candidate for a Cushing’s Syndrome, with a relatively poorly developed muscular mass, and this increase in abdominal fat. The reason for his drop in testosterone lays in an orchitis he had around 35 years of age that knocked down, long-term his testosterone levels. When he presented he had a testosterone level of 3 nanomoles per litre, which is a little bit above the normal reference range for women. He was put on testosterone and the second picture to the right shows you a decrease in his truncal fat mass. This was also combined with changes in his clinical presentation in terms of lowering his blood pressure, and a better control of his type 2 diabetes. The central point here to start this presentation is what is happening with the intraabdominal fat mass? And the regulation of the accumulation of fat in the abdomen? Whether or not this metabolic part of the total adipose tissue mass in the body differs, or not, from other pools is a little bit a matter of controversy. But it is a pool of fat that has a specific metabolic impact because its localisation. Releasing, or mobilising, fat from the intra-abdominal fat pool ends up in the liver, so the fatty acids directly exposes the liver, causing a change in liver metabolism, and also an accumulation of fat in the liver, which we recognize as liver steatosis, and as we often see in our laboratory program, elevation of enzymes relating to liver function.

The changes in metabolism in the periphery will reflect on the glucose levels in the periphery, the insulin levels and the lipid levels. And this part is where the endocrinology comes in. The macro-endocrinology of the intraabdominal fat mass is under the influence of cortisol and our main 2 anabolic hormones, growth hormone mostly exerted via the IGF1 mechanism that suppresses the fat accumulation in the abdomen, as contrary to cortisone that stimulates fat accumulation.

Men with type 2 diabetes and a Metabolic syndrome have lower testosterone levels than normal controls.

This is data from Barret-Connor’s group case control study of men with diabetes and matched controls. As you can see here, there is a clear difference in total testosterone levels but also in biologically active testosterone, which is non SHBG bound testosterone. This is a tricky part of the testosterone diagnosis, as changes in lipid mass also reflects in changes on the binding protein, SHBG and there is a simultaneous drop in SHBG the more fat you accumulate in the body. But at the end we have a drop in both testosterone, secretion testosterone levels and SHBG resulting in a decrease in the bioavailable part of testosterone.

The drop in testosterone has an impact on the risk for development of type 2 diabetes. These are age adjusted correlation coefficients between testosterone and some variables linked to the Metabolic syndrome. We can see that there is a negative correlation to BMI, which usually reflects the amount of fat in the body, except for extreme athletes. The waist circumference and triglyceride levels, fasting insulin levels, pulse challenged insulin levels and homocystein levels; these things are thus increasing while testosterone decreases, or if testosterone decreases, these variables increase. Contrary to this, HDL levels are positively correlated to testosterone in this type of setting. This also is true for intervention studies when you increase or normalise the testosterone in this group of men, that you have a positive correlation, meaning that the higher the testosterone level, the higher the HDL levels.

This is longitudinal data summarised in looking at the risk factor, or a variety of risk factors, for development of type 2 diabetes in the Metabolic syndrome. If you look at free testosterone, the biologically active part, a significant over-risk for development, as well the drop in SHBG levels picked up, which is linked then to the increase in fat mass. Another factor that is a risk factor for the development of type 2 diabetes is depression. This may also in the end be a part of the overall scheme that affects the neuro-endocrine system, and maybe a part of the mechanism behind the drop in testosterone levels.

If we look at obesity as such, there is a difference between obese males and lean controls in testosterone levels, free-testosterone levels and also in fasting insulin levels, indicating an insulin resistance in obese men, which is generally a known fact. But, this is also linked to the depression of testosterone.

Even in intervention studies where we have a group of hypogonadal men, and we treat them with testosterone, we also pick up this positive correlation between HDL levels and testosterone levels. So, both in the spontaneous situation and in the intervention situation, we have this relation, which indicates that testosterone, which for a number of years has been held as one of the key elements of premature development of cardiovascular disease, may on the contrary, contribute to the prevention of cardio-vascular disease, looking at risk factors that are impacting.

Changes in lipid panel, and these are data from Ron Swerdloff’s group, looking at hypogonadal men treated with testosterone gel and the changes in cholesterol levels. This was separated into younger men and older men. As you can see, during therapy there is a drop in total cholesterol, LDL cholesterol and in triglycerides while the HDL levels are not affected.

We will now go to the general population and look at epidemiological data, and these are data retrieved from the Tromso study and I’ve borrowed them from Johan Svartberg recently presented his thesis in Tromso. This is a population study looking at testosterone and BMI and waist circumference. As you can see from this three-dimensional graph, it’s not made three-dimensional because it’s to demonstrate something very interesting. And that is that the higher, or wider the girth is, the girth of the waist is, the lower the testosterone, despite a low BMI. So the worst case scenario is a fairly lean man with an increase in abdominal fat. If you want to pick out men with low testosterone and with a high chance of finding hypogonadal men, this is the group to go for. The wider the belly points out, the lower the testosterone levels are, and if that is combined with a normal BMI, it is extremely low. This is in fact a graph that superimposes with a risk for cardiovascular morbidity in another population study that was performed in Gothenburg a few years ago, demonstrating that the increase in abdominal diameter coincides with the highest risk for cardiovascular morbidity and mortality.

Tromso is a city located above the polar circle and has extreme seasonal variations in light and temperature. The Tromso group studied the variation in testosterone levels throughout the year, and the changes in testosterone levels had a pattern with increased levels during the late autumn and winter months. And what more of interest here is that they also followed the waist circumference, and the waist circumference showed a pattern with an increase in waist circumference coinciding with a lower level of testosterone seen during the summer period. So, by using graphic help, changing the scale, flipping it upside down, this is the waist, hip ratio, so the lower down the larger belly, and you can see that there is a superimposable pattern of changes in abdominal fat mass, with the annual variation in testosterone, which I think is a clear indication that these two things follow each other. The spontaneous changes in testosterone levels in this range in a normal situation due to annual changes, also impacts on the abdominal fat mass.

If you take men and expose them to different levels of testosterone, which was done in this study performed in Los Angeles by Bhasin’s group, this is a group of men given different testosterone regimens. These are milligrams of testosterone and are taken per week. Going from a low level up to extremely high superficial logical levels, you can see that the fat-free mass increases almost linearly with the testosterone dose, and that the fat mass decreases with the increasing levels of testosterone. So, overall we have a body composition regulation, exerted by testosterone, and ending up in increasing the lean body or the muscle mass of the body, and decreasing the fat mass. There’s a general regulation of the fat mass exerted by testosterone.

The Bhasin’s group went further with this, and looked at stem cells in the tissue that can be recruited into becoming adipocytes or muscular cells. Expose these stem cells to androgenic stimulation and with BIC they inhibited which is an anti-androgen. And as you can see here, the number of fat cells decreased while they were exposed to an increased level of androgenic stimulation, and by gradually blocking the androgenic activity the number of fat cells increased. Meaning that testosterone seems to have a direct role on the path taken by the stem cells into development of fat cells.

Looking at the other side of the coin, the development in muscle cell direction, the reverse is true. The more testosterone or the more androgenic stimulation the more muscle cells developed. So, not only the acute regulation of the lipid mass. but also the recruitment of cells building the tissue, the adipose tissue and the muscle tissue is, in part, controlled by androgenic stimulation.

So, let’s look at these three icons. We have the abdominally fat person, we have the generally obese man, and we have the lean man. These situations also carry a reflection in androgenic status. The abdominal fat man is probably the one with the lowest testosterone levels, while the generally obese man may maintain his testosterone level up to a certain point, is my guess, when the fat mass impacts on the neuro-endocrine regulation of testosterone secretion, and flips over into a state where the testosterone levels no longer can keep path with the increased fat mass.

Using testosterone therapy may reverse some aspects of this. If hypogonadal men receive testosterone in the standard dose, there is a decrease in the percentile body fat, which was demonstrated in this study by Katznelson.

We performed a twelve month placebo control study with testosterone in men with the metabolic syndrome including type 2 diabetes. So we have this group of men who are treated with testosterone, and this group was treated with placebo. Looking at different parts of the body, in the body composition we can see that we had the expected increase in fat-free mass or lean body mass. We had a decrease in the adipose tissue area in the lumbar region, which is the abdominal fat. That was significant. And simultaneously we had an increase in the muscle area, and if we combine both the abdominal part and the thigh, there clearly is significant increase in the muscle area. Looking at the subcutaneous adipose tissue there was also a decrease. So the testosterone intervention in this group of men reduces the fat mass generally in the body, both subcutaneous and in the intraabdominal pool.

Per Mårin and Bjorntorp’s group pioneered this field in the beginning of the ‘90s. This is from one of their early studies, where they picked men with abdominal obesity and treated them with testosterone, and demonstrated the conspicuous decrease in waist circumference that was more pronounced than the changes in hip circumference. Their conclusion, from additional studies, was that there was a significant difference in sensitivity to testosterone between the abdominal fat, and the subcutaneous fat pool.

In general, treating men with type 2 diabetes, and with the Metabolic syndrome with testosterone, induces changes, and what I want to demonstrate here and direct your attention to is the drop in LH levels, even if we have a fairly limited increase in testosterone.

In this group of men that we treated, we looked at the changes in testosterone, lowering for those who increase less in testosterone, compared to those who have the largest increase in testosterone. We can see that the group that increased most in testosterone had the much larger increase in glucose disposal rate, meaning that they became more insulin sensitive. They also had an increase in liver attenuation, meaning that the fat accumulation in the liver decreased.

This was also seen in the group that Mårin studied, that we have a testosterone induced increase in insulin sensitivity, depending on the changes in testosterone.

If we take away testosterone from men, castrate them, there is a decreased insulin sensitivity reflecting in increasing levels of fasting insulin as well as the correlation between changes in insulin and fat mass.

Testosterone levels in experimental data show a biphasic change in insulin sensitivity, so with castration, and very high levels, there is an insulin resistance while low testosterone levels increase the insulin sensitivity. So when you give it back testosterone it’s normalised.

DHT does not exert this impact on insulin sensitivity, and in a study with DHT there was no change in glucose disposal rates. So international studies indicate that testosterone therapy in men with Metabolic syndrome, and overt diabetes, decreased the abdominal fat mass, increase fat-free mass, and increase insulin sensitivity.

Just one last thing here that I want to point out. Where does this happen? Is this a central deactivation of testosterone secretion, or a peripheral? In our study we can see that testosterone therapy, increasing testosterone from an average 10 to 14 nanomoles per litre, caused a significant suppression of LH levels, meaning that testosterone was early suppressed in men with the Metabolic syndrome. So they have a low set point. The internal secretion is stopped at an early point. But there are also indications that the testicular response to ACD or LH stimulation is suppressed in obese men, or men with metabolic syndrome, indicating that there is also peripheral change in the response towards testosterone.

Putting this together, we can look at the metabolic syndrome from one aspect, being a stress related increase in the activity of the cortisone access, which suppresses both the GnRH and the GnRH pathways, rendering the visceral abdominal fat mass into a state of accumulation. We have these situations where we have the balance between the cortisone and insulin and in Cushing’s syndrome with the increment in cortisone, and in situations where we have a decrease of testosterone and growth hormone, we have an overweight for the corticosteroids and the insulin access, putting this balance out of action.

So there are two entries into the metabolic syndrome that I can see from this. One is obesity that at some point hits the testosterone production capacity, and lowers testosterone into the area where we identify men with metabolic syndrome.

We have the stress pathway, which centrally suppresses the gonadal access and renders testosterone levels to decrease and then end up in the metabolic syndrome.

So the T-levels, the testosterone, predict the risk for development of abdominal obesity.

Low T-levels are associated with a number of well-known risk factors for cardiovascular disease, and obesity is associated with decreased T-levels. While T-therapy changes by the composition in a favourable way.

The increase in adipose tissue, or in the reverse, the decrease of adipose tissue by testosterone is exerted metabolically, and also by stem cell recruitment.

What I would like you to take home today from this presentation, is the fact that testosterone can inhibit abdominal fat accretion, that there is an accumulating body of evidence that suggests the beneficial role of testosterone in man at risk for, or with the manifest metabolic syndrome. And the last point, which addresses different societies to support research, is that we need long-term intervention studies to really get a grasp on the risk benefit from testosterone therapy, which would be a simple way to address this very prevalent disorder.

We would switch from using replacement therapy into using androgens as a therapeutic tool, and instituting testosterone therapy as such for a specific syndrome.