Author + information
- Philip S Tsao, PhD
We appreciate the support of Drs. Girod and Brotman for our hypothesis that the increased prevalence of insulin resistance and hyperinsulinemia in cigarette smokers may play a central role in the dyslipidemia, endothelial dysfunction, and increased cardiovascular disease associated with tobacco use. However, we are not as persuaded as they seem to be that the link between cigarette smoking and insulin resistance is mediated via higher serum cortisol levels, secondary to smoking-induced visceral obesity. For example, in our initial study (1)we documented the presence of insulin resistance and compensatory hyperinsulinemia in smokers as compared with nonsmokers, matched for age, gender, family history of diabetes, alcohol consumption, level of physical activity, body mass index (BMI), and the ratio of waist-to-hip girth (WHR). Subsequent studies have also demonstrated that insulin resistance/hyperinsulinemia can be demonstrated in smokers independently of either BMI or WHR, and in two of these studies (2,3)it was shown that the enhancement of insulin sensitivity occurring with smoking cessation can be seen despite weight gain.
Conversely, in none of these studies was body fat distribution quantified by computerized tomography. Thus, the suggestion that the insulin resistance of smokers is due to a redistribution of body fat, with a relative increase in visceral obesity, cannot be dismissed. However, it should be noted that the observation of a smoking-related change in body fat distribution was made in only one ethnic group (4), and individuals of Korean ancestry may not be characteristic of the world at large. Furthermore, direct measurements of visceral obesity were not performed, and the suggestion that visceral obesity was present was based entirely on evidence that the ratio of WHR/BMI was increased in smokers. It should also be noted that the two-fold risk to have the highest WHR/BMI ratio was limited to 4.7% and 3.8% of the male and female smokers, respectively.
Finally, the relationship between visceral obesity and insulin resistance is a complex one. Perhaps the most revealing example of this is the recent report by Seppala-Lindroos et al. (5), showing with proton spectroscopy and magnetic resonance imaging in healthy volunteers that hepatic fat content was unrelated to amount of visceral fat, whereas the amount of fat in the liver, not visceral fat, was closely related to a variety of abnormalities associated with insulin resistance, including fasting hyperinsulinemia. In contrast, the formulation advanced by Girod and Brotman is certainly testable, and despite our skepticism it is worthy of experimental verification. Most importantly, the correspondents' letter should serve, along with our original comments, to indicate the need for a better understanding of the relationship between cigarette smoking and the wide variety of metabolic factors with which it is associated.
- American College of Cardiology Foundation