Imaging techniques for measuring adipose-tissue distribution in the abdomen: a comparison between computed tomography and 1.5-tesla magnetic resonance spin-echo imaging.

Eight subjects were examined both by abdominal X-ray computed transverse axial tomography (CT) and magnetic resonance imaging (MRI) (SE) (TR/TE, 200 ms/15 ms); another eight volunteers were subjected to three MRI scans to test the reliability of repeated measures. Correlations between fat area measures obtained by CT and by MRI for subcutaneous fat, total fat, and visceral vs. subcutaneous-fat ratio were highly significant (r = 0.93, 0.91, and 0.94, respectively; p < 0.01), and the standard errors of estimation were 9.99, 23.87, and 0.0047. The average errors of the method for different fat areas were 2.20 cm2 (intra-examination variance) and 3.75 cm2 (inter-examination variance) for visceral and 0.82 cm2 (intra-examination variance) and 1.29 cm2 (inter-examination variance) for subcutaneous fat areas, respectively. These results suggest that SE MRI is a practical approach to evaluate body fat distribution without the exposure to radiation. The reproducibility of SE MRI for the determination of fat areas is high; variation is small and acceptable. However, it is difficult to determine which estimate of fat area should be accepted when there is a discrepancy between MRI and CT measures.