Sunday, August 1, 2010

Growth hormone, insulin resistance, body fat accumulation, and glycogen depletion: Making sense of a mysterious hormone replacement therapy outcome

Hormone replacement therapies are prescribed in some cases, for medical reasons. They usually carry some risks. The risks come in part from the body down-regulating its own production of hormones when hormones are taken orally or injected. This could be seen as a form of compensatory adaptation, as the body tries to protect itself from abnormally high hormone levels.

More often than not the down-regulation can be reversed by interrupting the therapy. In some cases, the down-regulation becomes permanent, leading to significant health deterioration over the long run. One can seriously regret having started the hormone replacement therapy in the first place. The same is true (if not more) for hormone supplementation for performance enhancement, where normal hormone secretion levels are increased to enhance (mostly) athletic performance.

Rosenfalck and colleagues (1999) conducted an interesting study linking growth hormone (GH) replacement therapy with insulin resistance. Their conclusions are not very controversial. What I find interesting is what their data analysis unveiled and was not included in their conclusions. Also, they explain their main findings by claiming that there was a deterioration of beta cell function. (Beta cells are located in the pancreas, and secrete insulin.) While they may be correct, their explanation is not very plausible, as you will see below.

Let us take a quick look at what past research says about GH therapy and insulin resistance. One frequent finding is a significant but temporary impairment of insulin sensitivity, which usually normalizes after a period of a few months (e.g., 6 months). Another not so frequent finding is a significant and permanent impairment of insulin sensitivity; this is not as frequent in healthy individuals.

The researchers did a good job at reviewing this literature, and concluded that in many cases GH therapy is not worth the risk. They also studied 24 GH-deficient adults (18 males and 6 females). All of them had known pituitary pathology, which caused the low GH levels. The participants were randomly assigned to two groups. One received 4 months treatment with biosynthetic GH daily (n=13); the other received a placebo (n=11).

The table below (click on it to enlarge) shows various measures before and after treatment. Note the significant reduction in abdominal fat mass in the GH group. Also note that, prior to the treatment, the GH group folks (who were GH-deficient) were overall much heavier and much fatter, particular at the abdominal area, than the folks in the placebo (or control) group.

From the measures above one could say that the treatment was a success. But the researchers point out that it was not, because insulin sensitivity was significantly impaired. They show some graphs (below), and that is where things get really interesting, but not in the way intended by the researchers.

On the figure above, the graphs on the left refer to the placebo group, and on the right to the GH group. The solid lines reflect pre-treatment numbers and dotted lines post-treatment numbers. Indeed, GH therapy is making the GH-deficient folks significantly more insulin resistant.

But look carefully. The GH folks are more insulin sensitive than the controls prior to the treatment, even though they are much fatter, particularly in terms of abdominal fat. The glucose response is significantly lower for the GH-deficient folks, and that is not due to them secreting more insulin. The insulin response is also significantly lower. This is confirmed by glucose and insulin “area under the curve” measures provided by the researchers.

In fact, after treatment both groups seem to have generally the same insulin and glucose responses. This means that the GH treatment made insulin-sensitive folks a bit more like their normal counterparts in the placebo group. But obviously the change for the worse occurred only in the GH group, which is what the researchers concluded.

Now to the really interesting question, at least in my mind: What could have improved insulin sensitivity in the GH-deficient group prior to the treatment?

The GH-deficient folks had more body fat, particularly around the abdominal area. High serum GH is usually associated with low body fat, particularly around the abdominal area, because high GH folks burn it easily. So, looking at it from a different perspective, the GH-deficient folks seem to have been more effective at making body fat, and less effective at burning it.

Often we talk about insulin sensitivity as though there was only one type. But there is more than one type of insulin sensitivity. Insulin signals to the liver to take up glucose from the blood and turn it into glycogen or fat. Insulin also signals to body fat tissue to take up glucose from the blood and make fat with it. (GLUT 4 is an insulin-sensitive glucose transporter present in both fat and muscle cells.)

Therefore, it is reasonable to assume that folks with fat cells that are particularly insulin-sensitive would tend to make body fat quite easily based on glucose. While this is a type of insulin sensitivity that most people probably do not like to have, it may play an important role in reducing blood glucose levels under certain conditions. This appears to be true in the short term. Down the road, having very insulin-sensitive fat cells seems to lead to obesity, the metabolic syndrome, and diabetes.

In fact, in individuals without pituitary pathology, increased insulin sensitivity in fat cells could be a compensatory adaptation in response to a possible decrease in liver and muscle glucose uptake. Lack of exercise will shift the burden of glucose clearance to tissues other than liver and muscle, because with glycogen stores full both liver and muscle will usually take up much less blood glucose than they would otherwise.

I am speculating here, but I think that in individuals without pituitary pathology, an involuntary decrease in endogenous GH secretion may actually be at the core of this compensatory adaptation mechanism. In these individuals, low GH levels may be an outcome, not a cause of problems. This would explain two apparently contradictory findings: (a) GH levels drop dramatically in the 40s, particularly for men; and (b) several people in their 50s and 60s, including men, have much higher levels of circulating GH than people in their 40s, and even than much younger folks.

Vigorous exercise increases blood glucose uptake, inside and outside the exercise window; this is an almost universal effect among humans. Exercise depletes muscle and liver glycogen. (Fasting and low-carbohydrate dieting alone deplete liver, but not muscle, glycogen.) As glycogen stores become depleted, the activity of glycogen synthase (an enzyme involved in the conversion of glucose to glycogen) increases acutely. This activity remains elevated for several days in muscle tissue; the liver replenishes its glycogen in a matter of hours. With glycogen synthase activity elevated, glucose is quickly used to replenish glycogen stores, and not to make fat.

Depleting glycogen stores on a regular basis (e.g., once every few days) may over time reverse the adaptations that made fat cells particularly insulin-sensitive in the first place. Those adaptations become a protection that is not only no longer needed but also detrimental to health, since they lead to obesity. This could be the reason why many people initially find it difficult to lower their body fat set point, but once they lose body fat and stay lean for a while, they seem to become able to maintain their leanness without much effort.

Well, perhaps glycogen-depleting exercise is more important than many people think. It can help make you thin, but through a circuitous path.

And, incidentally, glycogen-depleting exercise causes a temporary but dramatic spike in GH secretion. This natural increase in GH secretion does not seem to be associated with any significant impairment in overall insulin sensitivity, even though glycogen-depleting exercise increases blood glucose levels a lot during the exercise window. This is a temporary and physiological, not pathological, phenomenon.


Rosenfalck A.M., Fisker, S., Hilsted, J., Dinesen, B., Vølund, A., Jørgensen, J.O., Christiansen, J.S., & Madsbad, S. (1999). The effect of the deterioration of insulin sensitivity on beta-cell function in growth-hormone-deficient adults following 4-month growth hormone replacement therapy. Growth Hormone & IGF Research, 9(2), 96–105.