Is It Possible to Detect Activated Brown Adipose Tissue in Humans Using Single-Time-Point Infrared Thermography under Thermoneutral Conditions? Impact of BMI and Subcutaneous Adipose Tissue Thickness

Abstract

Purpose: To evaluate the feasibility to detect activated brown adipose tissue (BAT) using single-time-point infrared thermography of the supraclavicular skin region under thermoneutral conditions. To this end, infrared thermography was compared with 18-F-FDG PET, the current reference standard for the detection of activated BAT.

Methods: 120 patients were enrolled in this study. After exclusion of 18 patients, 102 patients (44 female, 58 male, mean age 58±17 years) were included for final analysis. All patients underwent a clinically indicated 18F-FDG-PET/CT examination. Immediately prior to tracer injection skin temperatures of the supraclavicular, presternal and jugular regions were measured using spatially resolved infrared thermography at room temperature. The presence of activated BAT was determined in PET by typical FDG uptake within the supraclavicular adipose tissue compartments. Local thickness of supraclavicular subcutaneous adipose tissue (SCAT) was measured on CT. Measured skin temperatures were statistically correlated with the presence of activated BAT and anthropometric data.

Results: Activated BAT was detected in 9 of 102 patients (8.8%). Local skin temperature of the supraclavicular region was significantly higher in individuals with active BAT compared to individuals without active BAT. However, after statistical correction for the influence of BMI, no predictive value of activated BAT on skin temperature of the supraclavicular region could be observed. Supraclavicular skin temperature was significantly negatively correlated with supraclavicular SCAT thickness.

Conclusion: We conclude that supraclavicular SCAT thickness influences supraclavicular skin temperature and thus makes a specific detection of activated BAT using single-time-point thermography difficult. Further studies are necessary to evaluate the possibility of BAT detection using alternative thermographic methods, e.g. dynamic thermography or MR-based thermometry taking into account BMI as a confounding factor.

Fig 1. Illustration of performed measurements.

Left (thermographic data): Supraclavicular ROIs were of triangular shape and placed within the lateral neck triangle that is formed by the clavicle, the sternocleidomastoid muscle and the lateral neck contour. Jugular and presternal ROIs were circular. Right (CT data): As a measure for SCAT thickness of the supraclavicular regions, the minimal distance between the vascular compartment of the neck and the skin surface was measured (double arrow). Asterisks mark the left clavicle and the trachea.

Fig 2. Mean (left) and maximum (middle) skin temperature of the supraclavicular regions in patients without and with active brown adipose tissue.

Mean and maximum supraclavicular skin temperatures were significantly different between patients with and without active BAT (* indicates statistically significant differences). No significant difference was observed between BAT-positive and BAT-negative patients when mean supraclavicular skin temperature was normalized relative to the jugular skin temperature (right).

Fig 3. Correlation of skin temperatures of the supraclavicular, the presternal and the jugular regions with body mass index.

Red dots mark patients with activated BAT in FDG-PET. r denotes Pearson’s correlation coefficients.

Fig 4. Correlation between supraclavicular skin temperature and SCAT thickness (top) and between BMI and supraclavicular SCAT thickness (bottom).

r denotes Pearson’s correlation coefficient.

Fig 5. Examples illustrating the study results.

Thermographic images (top) and PET images (bottom) of patients without (left and right) and a patient with (middle) metabolically active BAT. Patients with low BMI (left and middle) display relatively high skin temperatures of the supraclavicular regions (arrows) independent of the presence of activated BAT.