A Comparison of Dual-Energy X-Ray Absorptiometry Defined Android Fat with Abdominal Adipose Tissue by Magnetic Resonance Imaging in Type 2 Diabetic Adults
Abstract Number: 1842-P
Authors: SUMI THARAN, DAVID KELLEY, JOHN THORNTON, DYMPNA GALLAGHER, THE MRI ANCILLARY STUDY GROUP OF THE LOOK AHEAD RESEARCH GROUP
Institutions: New York, NY; Pittsburgh, PA
Results: Quantifying intra-abdominal fat is of increasing interest due to the known metabolic consequences of this fat depot. We sought to compare the region of interest (ROI) defined as Android by dual energy X-ray absorptiometry (DXA: GE Lunar-Prodigy) software and abdominal ROI's by magnetic resonance imaging. Subjects were adults with type 2 diabetes participating in an ancillary study of Look AHEAD (Action for Health in Diabetes), a multi-center, randomized controlled trial of weight loss intervention in obese adults with T2DM, (N=24; 58.8 ± 6.2 years; BMI 43.7 ± 10.9 kg/m2: mean ±SD) who had subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) areas measured by magnetic resonance imaging at 50 mm distances above and below L4-L5 and Android fat by DXA. The statistically significant Pearson correlation coefficients (r) between VAT, SAT, and Android fat are shown in Table 1. for levels 0, -50, and -100 in relation to L4-L5 where VAT was significantly correlated with Android. The correlations show that VAT and SAT combined are more highly correlated with Android fat than VAT or SAT alone. Using multiple regression analysis, with Android as the dependent variable, both VAT and SAT contributed significantly and approximately equally to the model. We conclude that in this small group of obese type 2 diabetic subjects in this sample, variation in SAT and in VAT were equally reflected. The DXA defined Android region was not more closely correlated with VAT than SAT and has minimal validity as a measure of VAT.
Supported by HL70298; P30-DK-46204; P30-DK-26687.
|Table 1||Android r|
|VAT + SAT 0||0.61|
|VAT + SAT -50||0.77|
|VAT + SAT -100||0.79|