Are overall adiposity and abdominal adiposity separate or redundant determinants of blood viscosity?

Abstract

In line with recent literature showing that both general adiposity and abdominal adiposity are independently associated with the risk of death, we recently reported that body mass index (BMI) and waist-to hip ratio (WHR) were independent predictors of blood viscosity, related to different determinants of viscosity (for BMI: plasma viscosity and red cell aggregation; for WHR: hematocrit). Since this report was challenged by a study showing that abdominal adiposity (as measured with waist circumference WC and not WHR) is the only independent determinant of viscosity, we re-assessed on our previous database correlations among viscosity factors, BMI, WHR and WC. Blood viscosity was correlated to BMI (r = 0.155 p = 0.004), WHR (r = 0.364; p = 0.027) and WC (r = 0.094; p = 0.05). Hematocrit was correlated to WHR (r = 0.524) but neither to BMI (r =-0.021) nor waist circumference (r = 0.053). WC was correlated with plasma viscosity (r = 0.154; p = 0.002) while WHR was not (r =-0.0102 NS). A stepwise regression analysis selected two determinants of whole blood viscosity at high shear rate: BMI (p = 0.0167) and WC (p = 0.0003) excluding WHR. Therefore, in this sample, abdominal fatness expressed by WC and whole body adiposity remain independent determinants of blood viscosity. WHR and WC have not the same meaning, WC measuring the size of abdominal fat while WHR measuring the shape of body distribution regardless the degree of fat excess. Interestingly, hematocrit is rather related to shape (even within a normal range of body size) than the extent of abdominal fatness, and is not related to whole body adiposity.