Purpose: Increasing evidence suggests that abdominal obesity may be a better predictor of disease risk than total fatness. This study sought to determine how obesity and fat distribution measured by readily available anthropometric and dual-energy x-ray absorptiometry (DXA) methods is related to abdominal obesity assessed by magnetic resonance imaging (MRI).
Methods: Men (n = 43) and women (n = 47), ages 55 to 75 years, were assessed for body mass index, waist-to-hip ratio, waist circumference, and skin folds by anthropometric methods; for percentage of body fat by DXA; and for abdominal total, subcutaneous, and visceral fat by MRI.
Results: In stepwise regression models, the waist-to-hip ratio explained 50% of the variance in abdominal visceral fat among men (P <.01), and body mass index explained an additional 6% of the variance (P <.01). Among women, waist circumference was the only independent correlate of abdominal visceral fat, accounting for 52% of the variance (P <.01). Among men, the percentage of body fat was the only independent correlate of abdominal subcutaneous fat, explaining 65% of the variance (P <.01). Among women, the percentage of body fat explained 77% of the variance in abdominal subcutaneous fat and body mass index explained an additional 3% (P <.01).
Conclusions: Obesity and body composition obtained by readily available anthropometric methods and DXA provide informative estimates of abdominal obesity assessed by MRI imaging.