Is there a link between obesity, diabetes, erectile dysfunction, and cardiovascular disease?
Some of my disclosure information.
Now that is a cross-sectional view of a coronary artery and this is the disease process that as a cardiologist I spend most of my time dealing with the consequences of. This lipid rich plaque here in the servant intramolaria. We do not know much about why atherosclerosis occurs but we do know it is associated with a number of established risk factors.
And we know that the risk factors are important in a fixed way, this is a relative risk of the importance of these risk factors and we know that some are modifiable, and some are not modifiable. Difficult to modify age, male sex, although there are some methods of doing that, and angiogenetic profile.
Another thing we know about risk factors though is that they are far more important when they are combined.
This is data from the inter-heart study and these are the relative risks of individual risk factors down here, but when we start combining three of them together, and four and five, then this is an exponential scale here, logo rhythmic scale, that risk really starts to increase. So combining risk factors is a bad thing.
This process of metabolic syndrome is where, in a model if you like, they are combined, associated with trunkal obesity, that is a man, but it could be a woman, it really doesn’t matter. But that links with hypertension, dyslipidaemia, and type II diabetes. These three risk factors are some of the most powerful and important risk factors for the premature development of vascular disease.
This is a conference about aging, so what are the effects of aging?
What about vascular risk? Well, vascular risk increases with aging. Men seem to have it earlier than women; there is a ten-year difference roughly between the age-related increase in death from vascular disease compared to women. But it increases inexorably with age.
So do risk factors, blood pressure increases with age
So whatever arbitrary cut off you aim to define hypertension as, that increases with age as well. Remember hypertension is simply an arbitrary definition. It is changing all the time. We are lowering the level of blood pressure that we think is important.
Cholesterol increases with age both in men and in women, as do triglycerides so all our risk factors are getting worse with age.
Diabetes as a risk factor associated with coronary disease, increases with age too. This is data for men but data for women are the same.
So it would not be surprising to learn that metabolic syndrome increases with age. The striking thing though about this is the relationship between the sexes. Metabolic syndrome increases with age both in men and women.
Yet the effects, in terms of the event rate, are much higher in men. Whether you look at coronary heart disease and deaths from vascular disease in a low frequency population down in Japan, or high frequency population in the eastern states of Europe, then there is a disadvantage in being a male. At least twice the risk of death compared to being a woman.
Now one of the things that happens to men as we age is we change our shape. It seems to be an inexorable consequence of aging. And we develop what is a pro-atherosclerotic shape; we develop this trunkal obesity here. Most men seem to do that.
They do that in association with this phenomenon, which is a falling level of testosterone. The difficulty here is what comes first, because adipose tissue can convert testosterone to oestrogen, does the changing body shape of a man with trunkal obesity cause the low testosterone or is it vice versa?
I do not need to tell you much about testosterone physiology, I am a simple cardiologist there are far more people in this room I’m sure who know far more about testosterone than I ever will. But it is important when you look at the epidemiology to understand that total testosterone is not always the best marker of androgenicity. Many of the large epidemiological studies have used total testosterone and not shown much effect in terms of a relationship with coronary disease. We measured bioavailable testosterone we measure it directly although we can estimate it as well. And this is the element, the moiety of testosterone, which is we think most important biologically.
So the first question we asked ourselves about ten years ago now was if low testosterone is associated with an adverse coronary risk profile, is there evidence that men with coronary disease have lower than normal testosterone? And in this case control study using coronary angiogram as the gold standard of whether there is coronary atherosclerosis or not. We did a study of 60 men with coronary disease and 30 without, and we excluded them if they had any other reason to have a low testosterone, we took early morning blood samples.
We showed that bioavailable testosterone was significantly lower in our men with coronary disease. We checked that this was not related to body mass index or age. We corrected for that and the relationship persisted.
We have done a much bigger study looking at a much larger cohort of men with coronary disease and about a quarter have really very low levels of testosterone. Total testosterone or bioavailable testosterone, and really quite low hypogonadal ranges.
About 50 percent of men with coronary disease have a total testosterone of less than 11. And we found the obvious negative correlations between body mass index and age.
We are not the only ones who have shown a relationship between testosterone and atherosclerosis. These Finnish authors have looked at the carotid artery as another vascular territory prone to development of atherosclerosis. They have used a non-invasive technique called Doppler Intima Media Thickness. This is a non-invasive ultrasound technique to look at the sub-endothelial layer and look to see if there is thickening there, which we believe is likely to be atherosclerosis. And there is a clear negative correlation here. So the men with the lowest levels of testosterone have the highest level of intima media thickness.
A Dutch group has done the same technique but looked four years later to see whether those men with low testosterone had more progression of carotid disease and they found that the men with the lowest total testosterone level had the highest increase in intima media thickness over a four-year period. You can see it is particularly important in the free testosterone data.
Another Dutch group looked at yet another vascular territory; this is abdominal aortic atherosclerosis as measured by calcification on lateral abdominal x-rays. And they looked eight years, seven or eight years apart to look at progression of abdominal aortic calcification, and they found again that men with the lowest levels of total testosterone and the lowest levels of bioavailable testosterone had the biggest increase in aortic atherosclerosis.
So, there are epidemiological data to suggest that low testosterone is associated with atherosclerosis, and with the development, the progression of atherosclerosis. How might that be? What is the mechanism?
Well, the mechanism may be through the metabolic changes that occur in diabetes. We do not know why diabetes is associated with coronary disease. We know it is the commonest cause of death in type II diabetes. And we know that hypogonadism increases with age and increases particularly in the presence of diabetes.
We know that low testosterone is a feature of the metabolic syndrome. These are data to show that for every factor in the metabolic syndrome you have, the lower the level of testosterone that you have. So when you have the full-blown metabolic syndrome you tend to have a lower level of testosterone. But it is chicken or egg, what comes first? And what is pathogenic here?
Is it something to do with the endothelium itself? We know that erectile dysfunction is another marker of low testosterone; it is also another marker of coronary disease. We know that fat men, obese men have a higher risk of erectile dysfunction, and we know that erectile dysfunction is associated with endothelial dysfunction. But again, what comes first, chicken or egg?
We have developed an interesting model, which we have been looking at over the last few years now. And this is a testicular feminised mouse. This mouse has a genetic absence of the nuclear androgen receptor. So, it is a male mouse, but it cannot develop the male characteristics because of this receptor lack. But it has got more than that. It has got also an enzyme deficiency, so in fact, it doesn’t have a high level of testosterone it has a low level of testosterone. So this mouse has a level of testosterone that is 10 percent the level of its littermate control, and no nuclear androgen receptor.
This is what happens when you feed these mice with a cholesterol rich diet. Here is our TFM mouse, and its littermate control, and both have been fed on a cholesterol rich diet. You can see that the TFM mouse with the low level of testosterone has a dramatic increase in weight gain compared to its littermate control.
When you look at the vessels, these are cross-sectional views of the aorta you can see that significant levels of lipid are deposited in the vessels in our fat TFM mouse.
So we have a model of effectively hypogonadism, and this model tells us two things. It tells us first of all that low testosterone occurred first before the development of obesity, and that this effect was not mediated by the nuclear androgen receptor.
What about testosterone replacement? Much of our other work that we have been looking at, mechanisms of atherosclerosis been in models of hypogonadism, human and animal, where we have looked at change when we have given testosterone replacement.
First of all, our mouse model when you replace testosterone physiologically in our mouse model then we get the same level of aortic deposition as littermate controls. So we have totally reversed the deposition of the lipid in the aorta by normalizing the testosterone level. When we give too much, when we give super physiological testosterone, we do not get as good an effect. But in the normal mouse super physiological testosterone did not really do very much. In the animal model we can melt away the lipid deposition in the aorta by simply replacing the testosterone.
So what happens in the human model? This is a study we did in hypogonadal men, many of whom had coronary disease identified in our prevalence work that I showed you earlier. It is a crossover study of testosterone or placebo. The testosterone effects are shown in the dark colours, dark grey, and the placebo effects in the pale grey. You can see that when we replace testosterone in hypogonadal men over a three-month period then we get significant reductions in TNF-a, and some reduction in IL1-b . These cytokines are inflammatory cytokines and thought to be important in the inflammatory milieu associated with the development of atherosclerosis. IL-10 is an anti-inflammatory cytokine and that is increased by testosterone replacement. So we alter the immunology, we alter the inflammatory status of men, hypogonadal men, by simply replacing testosterone. Now this implies that men with low testosterone are inflammatory activated if you like.
In the same study we showed significant reductions in cholesterol, about 0.4 nmol/L reduction in cholesterol. That is simply with testosterone replacement, and some other effects, which weren’t significant. Now, these are not huge numbers of men. This was a cohort of about 30 men that we showed this in. So these are significant effects in small numbers of individuals. We believe therefore, probably a biological significance.
We have done some work in diabetes and you know diabetes we have taught that diabetes is an important disease association with coronary artery disease. We know that in type II diabetes particularly, low testosterone occurs very commonly. There have been two studies now, Boyanov study was the first, but we have done a study with my colleague, Hugh Jones, in Barnsley, looking at testosterone replacement therapy in men with type II diabetes who started with a low level of testosterone. And we found significant improvements in glycaemic control, insulin resistance another important phenomenon, and thought to be pathophysiologically important, in the development of atherosclerosis and reductions in waist-hip ratio, and body weight. So, simply normalising the hormone profile, the male sexual hormone profile in diabetes seems to be a good thing.
Now people have been very worried about the use of testosterone in men with vascular disease, and another part of the work that we have been doing is demonstrating that in fact testosterone is useful, it helps men with established vascular disease, not now trying to prevent atherosclerosis and advance, but actually men with established vascular disease. Because acute administration of testosterone is a vasodilator and it works differently from oestrogen it is an endothelial independent vasodilator. It is not at all mediated by the nuclear androgen receptor our testicular feminised mouse model still vasodilates with testosterone. And in fact, we believe it blocks an L type calcium channel blocker, which is the same calcium channel blocker that nifedipine blocks.
We, and others, have looked at its effect on angina and in fact, it was first used in angina many, many years ago in the ‘40s and it was shown to improve symptoms of angina. A vasodilator like nifedipine would improve symptoms of angina.
All these studies are short-term studies but all of them have been positive apart from the bottom one here, Thompson’s study. We do not understand why that was not positive, but all the studies so far have been published as to the use of testosterone replacement therapy and angina have been positive. It prolongs time to ischaemia on the treadmill and it reduces symptoms. Its effect is particularly obvious in men with low starting levels of testosterone. Replacing or giving patients with a normal level of testosterone, testosterone replacement therapy has very little effect and we now don’t do any studies in men who have normal levels of testosterone because it really doesn’t do very much at all. But if you have a low level of testosterone then we are increasingly using testosterone to normalise the endocrine profile in patients with angina.
Heart failure again is probably going to be over the next decade or so the biggest workload for cardiologists. We have many, many strategies to help men with angina. We have coronary intervention of one kind or another, which are great symptomatic treatments for angina, so men are living longer and are having fewer acute events, fewer myocardial infarctions. They are living longer and they are developing heart failure. Heart failure has a clinical consequence, a natural history worse than many cancers. Fifty percent of men with heart failure will be dead in five years. The treatments we have are symptomatic largely. We have shown that hormone replacement therapy, testosterone physiological levels given to men acutely vasodilate in men with heart failure. They reduce peripheral resistance and increase cardiac index, but only in the men with low levels of testosterone. We’ve done a further one-year study looking at hormone replacement therapy in men with heart failure over 12 months and shown that exercise time improves, this was a placebo randomised controlled trial, exercise time improves and men felt better. Of course, it is a small study; it’s a proof of concept study. We have no data on longevity whether hormone replacement therapy will make men live longer, we don’t know.
So, I hope over the last half and hour or so I have convinced you, at least I have sown the seed that maybe testosterone, low testosterone is the link between obesity, erectile dysfunction, diabetes and cardiovascular disease.
Thank you very much.
Audience Question: The heavy use of testosterone in gyms for muscle building, is there any data on cardiovascular problems? Do they die earlier? Do they have any problems, or do we not know anything about this?
Dr. Channer: You mean super physiological doses? Yes, yes. Yes, anabolic steroids are bad. There have been lots of studies, formal studies of the use of anabolic steroids in, for example, weight lifters. In the old days, before we had drugs, weight lifters got better by eating more and training hard. If you augment the eating more and training hard with an anabolic steroid you get bigger effects.
One of the things, the downside of this is that you get cardiac hypertrophy, and cardiac hypertrophy is in fact an abnormal adaptation, it is a maladaptation. Because you are born with all the cardiac myocytes you are ever going to have, the only way the heart can get stronger is by making the ones that you have bigger. But that increases apoptosis and so cardiac hypertrophy tends to shorten cell life, so that is why it is bad. And it is bad in all the other models of hypertrophy we have, hypertension, aortic stenosis and so on.
There is also some concern over sudden deaths in men taking high doses of anabolic steroids and in high doses liver disease can occur, and clotting abnormalities can occur, and myocardial infarction can occur. But there are no concerns about physiological replacement therapy that I am aware of. This is why in the literature when we started using normal levels, physiological levels of testosterone in men, there was so much concern that we were actually going to do some harm.
One of the things we do not understand so far in our heart failure patients is why they got better. And we are doing much more work on why giving testosterone replacement therapy to men with heart failure they would get better. We know echocardiographically the hearts did not get bigger. But echocardiographically the heart changes shape with testosterone replacement therapy. In heart failure the heart becomes globular. Its long axis shortens. It becomes more like a sphere and that is less physiologically efficient as a pump. We saw improved long axis function for the heart got bigger, longer with testosterone replacement therapy, but not, we didn’t, induce LVH, because if we had induced left ventricular hypertrophy that may be a bad thing. But we are looking much more in detail at that kind of thing with MRI and other things.
Audience Question: I wonder what in your practice for patients with angina and heart failure, whether you use testosterone and to what extent these patients, if they do use testosterone, continue with treatment? And whether you have a feel for the long-term effects or outcome of testosterone in heart patients?
Dr. Channer: Thank you. As you know none of the testosterone preparations is licensed in cardiac disease. But we are identifying men now who have low levels of testosterone and they go through the usual replacement programs. So we have a very much more active androgen clinic now. Our men in our randomised control trials, those men who had testosterone actually felt much better than the men who didn’t have testosterone. And the ones who we recruited who didn’t have testosterone now have open label testosterone.
In terms of safety, we have looked very carefully, obviously, at the effects of prostate using PSA and other things and show no difference in the short-term. But our studies are limited to one year, although men on normal hormone replacement therapy will have the usual safety measures throughout. We are not aware of any concerns with our patients.
Audience Question: I was not asking about concerns for safety but the efficacy. Say beyond your trials do the patients continue with testosterone, and whether you have any feel for whether they have any longer-term benefits?
Dr. Channer: Right. The only data we have is up to 12 months. We have not followed any patients longer than 12 months. The 12 months in the heart failure study there was some tailing off of efficacy. We don’t know whether that was because we had a large number of drop-outs because we used the plaster kind of preparation, a lot of side effects with that. So, I don’t have any feel I’m afraid.
Audience Question: You mention at several points during your talk of the importance of normalising testosterone in aging men. How do you ascertain normal in men with a downwardly shifting target?
Dr. Channer: Thank you. The trouble is that the normal range is quite wide. So, when I say normalising what I mean is that if patients had a lower than, all our treatment studies now have to have lower than 12 or 11, and we aim to get them to 20 or less, so we are only putting people in to the normal range. But it is a range. And we don’t know what the normal range is for a healthy living eight year old.
Copyright © 2006 ISSAM