Thursday, August 16, 2007

Too weak evidence against body fat

fat and smoking
While I have been scanning new PubMed entries recently, this study jumped into my eyes: Smoking status and adiponectin in healthy Japanese men and women. It was adipo that made me click into the abstract of this study. The prefix adipo means fat-related, and as you may know from my earlier posts I am looking for any possible evidence of a direct and negative body fat effect on health.

Adiponectin is a hormone produced by human fat cells. It is thought to have various positive effects on health, reducing the risk of atherosclerosis, coronary heart disease and type 2 diabetes. The Japanese study shows that smoking reduces the level of this healthy hormone. In those who have quit smoking for more than twenty years, the hormone levels are the same as in non-smokers. Thus, smoking is likely to be the cause of the lower hormone level. A lower adiponectin level may be one negative effect of smoking on health, but surely it is not the only one.

Despite being produced by fat cells, adiponectin is lower in people with a higher fat mass. The more body fat, the less of the good hormone adiponectin. So it seems we have found a direct negative health effect of body fat.

Weak multiplied by weak equals (nearly) zero

The adiponectin case is too complicated to be discussed in full details. So I must make some simplifications here. Let me assume that a higher body fat mass is the ultimate cause of a lower adiponectin level and that this lower level is the cause of a higher heart disease risk. This is not necessarily so, but just assumed, it would be in favour of the bad body fat theory. But even this is not sufficient, as we just will see.

adiponectin graphBody fat, adiponectin and heart disease are not pieces of a domino where one falling piece is sufficient to make the next piece fall. In contrary, the effects are only weak. In obese people, the adiponectin levels are not zero but only about twenty or thirty percent lower than in slim people (see graph).

And in people with atherosclerosis (a risk factor for heart disease), this difference against healthy people may even be smaller, for instance it has been found not statistically significant in a Polish adiponectin heart risk study.

Take home message: A higher mass of body fat may lower the level of a healthy hormone to a small amount and this healthy hormone may have a very small influence on the risk of a heart attack. The cascade of these two weak effects is most likely to be smaller than detectable.

Photo credit:
Graph credit: Obesity Research

No comments: