Overall body fat and its distribution in different regions are important predispositions to known aberrations in lipid and glucose metabolism. The accuracy of MR imaging in estimating overall body fatness and regional fat distribution at individual landmarks was determined by comparing it with well-accepted measures by deuterium-oxide (D2O) dilution and bioimpedance analysis. Fourteen normal young women (athletes and control subjects) were studied. A total of 308 axial, T1-weighted, spin-echo MR images over a specific region in the trunk (21-24 scans per subject) were obtained. Morphometric computer image analysis was performed to determine the subcutaneous, internal, and total fat volumes in each image. The data were analyzed in two ways: data from all slices were summed to assess overall body fatness, and six anatomic landmarks were chosen for regional comparisons. MR-determined estimates of overall body fatness strongly correlated with total body fat measures by D2O dilution in both total fat (r = .91) and subcutaneous fat (r = .92) determinations. Athletes in both the low- and high-intensity training phases had significantly lower values of MR-determined total body fatness than did control subjects. Parallel to total body fatness, athletes had significantly lower MR-determined ratios of total fat/total volume in four of six individual landmarks compared with control subjects. Our experience suggests that MR is an accurate method to quantify overall body fatness, when compared with D2O dilution and bioimpedance analysis. MR could also discriminate regional components of subcutaneous and internal body fat at individual landmarks.
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging