Are dual-energy X-ray absorptiometry regional estimates associated with visceral adipose tissue mass?

OBJECTIVE: Recent studies support the possibility of estimating abdominal fat using a region of interest (ROI) selected by conventional whole body dual-energy X-ray absorptiometry (DXA). This is an important observation as DXA ROI estimates have some advantages over waist circumference or computed tomography/magnetic resonance imaging (MRI) as a means of assessing visceral adipose tissue (VAT) and adipose tissue distribution. The aim of this study was to evaluate the usefulness of DXA abdominal ROI estimates in assessing VAT among non-obese men. DESIGN: Observational, cross-sectional study comparing correlations between MRI-measured total VAT and surrogate measures including DXA ROIs. A stepwise multiple regression model was applied to derive a predictive equation with total VAT mass. SUBJECTS: Ninety non-obese healthy men between the ages of 18 and 44y with BMI < 30 kg/m(2). MEASUREMENTS: Abdominal adipose tissue and total VAT were measured by whole body MRI; VAT area by single-slice MRI at the L4-5 level; specific DXA ROIs for abdominal regional fat defined as ROI A (L2-4), B (L2-upper iliac), C (lower costal-upper iliac), and D (ROI C excluding spine); and simple anthropometric measures. RESULTS: Correlations between total VAT and ROIs A (r= 0.85) and B (r= 0.84) were not significantly different from that of VAT area at L4-5 (r= 0.87), but significantly higher (P < 0.01) than that of waist circumference (r= 0.77). The highest correlations with total abdominal adipose tissue were for DXA ROIs and conventional DXA trunk fat (r=0.95-0.97). A stepwise multiple regression analysis revealed that 86% of the variance in total VAT was predicted by VAT area at L4-5, ROI A, and waist-hip ratio. CONCLUSIONS: DXA ROIs (L2-4, L2-upper iliac) were associated with total VAT as well as MRI-derived VAT area at L4-5 in non-obese men. DXA ROI fat distribution estimates may be useful in the early detection of men with abdominal/visceral obesity.