Abdominal adipose tissue and serum HDL-cholesterol: association independent from obesity and serum triglyceride concentration.

It has been recently shown that an excess of abdominal fat was associated with reduced serum HDL-cholesterol level, supporting previous studies that have reported a relationship between fat distribution and cardiovascular risk factors. Since a negative relationship has been observed between serum triglyceride and HDL-cholesterol concentrations, the associations between body fat distribution and HDL-cholesterol level was studied with control over serum triglycerides in a sample of 429 healthy adult men. The relative distribution of subcutaneous fat, as reflected by the trunk to extremity skinfolds (T/E) ratio and the absolute amount of subcutaneous abdominal fat, obtained by the measurement of the abdominal skinfold thickness, were significantly correlated with serum triglycerides (r = 0.27 and 0.35 respectively, P less than 0.0001), with serum HDL-cholesterol (r = -0.14, P less than 0.01, and -0.26, P less than 0.0001) and with the serum HDL-cholesterol/total cholesterol ratio (r = -0.25 and -0.39, P less than 0.0001). Analysis of variance on two factors, the T/E ratio and the body mass index (BMI), revealed significant and independent effects of adiposity and relative distribution of subcutaneous fat on serum lipids and HDL-cholesterol (0.05 greater than P less than 0.001). However, when comparable analyses of variance were performed to study the respective contributions of obesity (as measured by the BMI) and the absolute amount of abdominal fat (as measured by the abdominal skinfold), most of the variance in serum triglycerides, cholesterol and HDL-cholesterol was explained by abdominal fat alone and not by the BMI. Therefore, it appeared that with the exception of the HDL-cholesterol/total cholesterol ratio, the association between obesity and serum lipids and HDL-cholesterol was, in the present sample, primarily explained by the amount of abdominal fat. As reported by others, serum triglyceride level was negatively correlated with HDL-cholesterol (r = -0.33, P less than 0.0001) and the HDL-cholesterol/total cholesterol ratio (r = -0.54, P less than 0.0001). After cholesterol and HDL-cholesterol scores were adjusted for the effect of triglycerides, the T/E ratio was no longer associated with serum HDL-cholesterol whereas the abdominal skinfold remained significantly correlated with serum HDL-cholesterol concentration (r = -0.16, P less than 0.01). These results suggest that a portion of the association between body fat topography and serum HDL-cholesterol is mediated by the effect of fat distribution on serum triglycerides.(ABSTRACT TRUNCATED AT 400 WORDS)