Subcutaneous fat of extremities independent of abdominal fat does not promote insulin resistance and cardiovascular risk in Japanese type 2 diabetic patients

Thigh subcutaneous fat has recently been advocated to have a favorable association with insulin resistance or cardiovascular disease (CVD) risk, in contrast to abdominal and visceral fat. However, because individual regional fat components are strongly correlated with each other, multiple regression analysis of these independent variables in the same model is not valid due to multicollinearity, even though most previous reports regarding this issue have applied this invalid statistical analysis. In the present study, we applied another statistical analysis instead of this widely applied multiple regression analysis. In 570 type 2 diabetic patients, we measured the amount of fat in each region using bioelectrical impedance analysis (INBODY720) and drew a regression line between trunk fat percentage (F%) on the x-axis and logarithm of leg or arm F% on the y-axis. Thereafter, the residual from this line on the y-axis was calculated and defined as the trunk fat-adjusted leg or arm F% (residual leg or arm F%), from which the effect of abdominal adiposity had been eliminated. Total and regional F% values were all positively correlated to a log-transformed homeostasis model assessment of insulin resistance (HOMA-IR) and triglyceride (TG)/HDL-cholesterol (HDL-C). However, residual leg or arm F% was not related to log(HOMA-IR) or log(TG/HDL-C) in either male or female patients, with the exception of a negative correlation between residual leg F% and log(TG/HDL-C) in male patients. On the other hand, while trunk F% showed a strong positive relation to the number of CVD risk factors, residual arm or leg F% did not show any relationship to this parameter in either gender. These results confirmed that subcutaneous fat in extremities does not promote insulin resistance or CVD risk per se, in contrast to abdominal visceral fat. Furthermore, leg subcutaneous fat may exert a beneficial effect on insulin resistance in male diabetic patients. This is the first report to have investigated the effects of leg or arm fat independent of abdominal fat using a unique residual method.