Objectives: Brown fat is a potential source of false-positive findings on FDG PET. The purpose of this study was to show the variability in body distribution of brown fat, the degree of FDG uptake, the changes on dual-time-point scanning, and determine if dual-time-point scanning can help in differentiating brown fat from malignant lesions.
Methods: Thirty-two patients were included in this retrospective study (14 male, 18 female, age range: 8-72 years). All patients had hypermetabolic brown fat activity on FDG PET imaging confirmed by computed tomography (CT) scanning. All patients underwent 2 sequential FDG PET scanning (dual-time-point imaging). The average percent change in maximum standardized uptake value (SUVmax) for 120 brown fat spots between time point 1 and time point 2 was calculated.
Results: Body distribution of hypermetabolic brown fat in the 32 patients included supraclavicular area (n = 7); cervical and supraclavicular (n = 5); cervical, supraclavicular, and axillae (n = 5); cervical area, supraclavicular, axillae, and paravertebral (n = 8); supraclavicular, cervical, axillae, paravertebral, and mediastinum (n = 4); supraclavicular, cervical, axillae, paravertebral, and upper abdomen (n = 2); and supraclavicular, cervical, axillae, paravertebral, mediastinum, and intercostals (n = 1). SUVmax for brown fat spots ranged from 0.8 to 12.4 and mean SUV was 4.6 + 1.6. On dual-time-point imaging, 91 (76%) of the brown fat spots demonstrated an increase in SUVmax that ranged from 12% to 192% and mean value of 42%, whereas 16 (13%) brown fat spots did not show any change and 11 (11%) spots underwent a drop in SUVmax by 4% to 12%. There was an increase in the number of active brown fat spots in 3 patients on the second time images.
Conclusions: Brown fat is a potential source of false positives, which has wide variability in distribution and degree of FDG uptake. On dual-time-point scanning, there is a progressive increase in FDG uptake within most of the hypermetabolic brown fat areas that mimic malignant lesions.