Brown fat depots in adult humans remain static in their locations on PET/CT despite changes in seasonality

Active brown adipose tissue (BAT) in humans has been demonstrated through use of positron emission tomography with 2-deoxy-2-(fluorine-18) fluoro-D-glucose integrated with computed tomography (18F-FDG PET/CT) scans. The aim of our study was to determine whether active human BAT depots shown on 18F-FDG PET/CT scans remain static in their location over time. This was a retrospective study. Adult human subjects (n = 15) who had had 18F-FDG PET/CT imaging (n = 38 scans in total) for clinical reasons were included on the basis of 18F-FDG uptake patterns consistent with BAT activity. For each subject, 18F-FDG BAT uptake pattern on serial 18F-FDG PET/CT images was compared to an index 18F-FDG PET/CT image with the largest demonstrable BAT volume. Object-based colocalization was expressed as Mander's correlation coefficient (where 1 = 100% overlap, 0 = no overlap). Distribution of 18F-FDG BAT activity over time and across multiple 18F-FDG BAT scans was equivalent in 60% (n = 9) of the subjects. The degree of consistency in the pattern of 18F-FDG BAT uptake in each subject over time was greater than expected by chance in 87% (n = 13) of the subjects (pair-wise agreement 75-100%, Fleiss' κ 0.4-1). The degree of BAT colocalization on serial scans was greater than that expected by chance in 93% (n = 14) of the subjects (mean Mander's coefficient 0.81 ± 0.21 [95% CI]). To our knowledge, our study provides the most conclusive evidence to date to support the notion that active BAT depots in humans (volumes and activities of which were measured through use of 18F-FDG PET/CT scans) remain static in location over sustained periods.