Discrimination between brown and white adipose tissue using a 2-point dixon water-fat separation method in simultaneous pet/MRI

Abstract

The purpose of the study was to evaluate signal-fat-fraction (SFF) analysis based on a 2-point-Dixon water-fat separation method in whole-body simultaneous PET/MR imaging for identifying brown adipose tissue (BAT) and discriminating it from white adipose tissue (WAT) using cross-validation via PET. Methods: This retrospective, internal review board-approved study evaluated 66 PET/MR imaging examinations of 33 pediatric patients (mean age, 14.7 y; range, 7.4-21.4 y). Eleven elderly patients were evaluated as controls (mean age, 79.9 y; range, 76.3-88.6 y). Pediatric patients were divided into 2 groups: with and without metabolically active supraclavicular BAT. The standard of reference for the presence of BAT was at least 1 PET examination showing 18F-FDG uptake. PET/MR imaging included a 2-point Dixon water-fat separation method. Signal intensities in regions of interest on fat and water images and mean standardized uptake values (SUVmean) were determined bilaterally in supraclavicular and gluteal fat depots. SFF was calculated from the ratio of fat signal over summed water and fat signal. Statistical analysis was conducted using the Student t test and correlation analysis. Results: SFF was significantly lower (P < 0.0001) in supraclavicular BAT than gluteal WAT in all pediatric subjects. Supraclavicular SFF was significantly higher in the control than in the pediatric group (P < 0.0001). In PET-positive patients with multiple examinations, SFF stayed stable whereas SUVmean fluctuated (median intraindividual change, 5% vs. 91%). No significant correlation between SUVmean and SFF could be observed for BAT. Conclusion: The results demonstrate that MR imaging-SFF analysis is a reproducible imaging modality for the detection of human BAT and discrimination from WAT. SFF values of BAT are independent from its metabolic activity, making SFF a more reliable parameter for BAT than the commonly used PET signal. However, with the intent to investigate both the composition of BAT and its activation status, hybrid PET/MR imaging might provide supplemental information.