Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 May 1.
Published in final edited form as: Obesity (Silver Spring). 2020 Mar 13;28(5):916–923. doi: 10.1002/oby.22767

Cold-Activated Brown Adipose Tissue is Associated with Less Cardiometabolic Dysfunction in Young Adults with Obesity

Nicole L Mihalopoulos 1, Jeffrey T Yap 2,3, Britney Beardmore 2, Richard Holubkov 4, M Nazeem Nanjee 5, John M Hoffman 2,3
PMCID: PMC7180112  NIHMSID: NIHMS1556339  PMID: 32170839

Abstract

Objective:

To test the hypothesis that young adults with obesity and cold-activated brown adipose tissue (BAT) are less likely to have metabolic dysfunction (dyslipidemia, insulin resistance, hypertension) than those without cold-activated BAT. Previous studies have noted a potentially protective effect of BAT and higher adiponectin/leptin ratios, but they acknowledged that the clinical implications of these findings remain uncertain.

Methods:

We enrolled 21 females and 23 males with obesity (BMI ≥ 30kg/m2) who underwent a 2-hour cooling protocol before FDG-PET/CT scan, to determine the prevalence, volume and FDG uptake of cold-activated BAT.

Results:

Cold-activated BAT was identified in 43% of participants (11F, 8M); females had greater FDG uptake. Those with cold-activated BAT had lesser degree of metabolic dysfunction. Cold-activated BAT volume correlated with triglycerides (inversely) and adiponectin (concordantly). Body mass adjusted cold-activated BAT activity correlated with HDL cholesterol (concordantly). Males with cold-activated BAT had lower leptin and higher adiponectin/leptin ratios.

Conclusions:

We report a high prevalence of cold-activated BAT in our study participants. We suggest that BAT could be important in decreasing metabolic dysfunction among young adults with obesity, making it a potential target for treating metabolically unhealthy obesity.

Clinical Trial Registration:

Trial registry name is STAGES Trial: Study of Adiposity, Growth and Endocrine Stages (STAGES). The clinical trial registration number is NCT01460784. https://clinicaltrials.gov/ct2/show/NCT01460784?term=obesity&cond=brown&draw=8&rank=54

Keywords: brown adipose tissue, obesity, metabolic syndrome

Introduction

Obesity increases risk for dyslipidemia, insulin resistance/impaired glucose tolerance and elevated blood pressure, which are components of the metabolic syndrome.1,2 However, not all individuals with obesity have metabolic dysfunction (defined as an abnormality of any of the components of metabolic syndrome). Previous studies of brown adipose tissue (BAT), conducted primarily with normal weight, healthy men, have reported an association between cold-activated BAT via 2-[18F]fluoro-2-deoxyglucose positron emission tomography and X-ray computed tomography (FDG-PET/CT) and lower body mass index (BMI).3–5 Those with cold-activated BAT have healthier metabolic profiles (improved glucose metabolism and lipids profiles) compared to those without cold-activated BAT.6–8 Whether individuals with obesity who have cold-activated BAT via FDG-PET/CT differ in any clinically meaningful way from individuals with obesity who lack FDG-avid BAT is unknown.9,10

Adipose tissue functions as an endocrine organ.11 Both white adipose tissue (WAT) and BAT secrete adiponectin and leptin; concentrations of adiponectin decrease and leptin increase with increasing WAT.11,12 The ratio of adiponectin to leptin (A/L) is associated with insulin sensitivity and lipid metabolism.13,14 Animal studies suggest a protective effect of BAT in the development of cardiovascular disease.15,16 It is possible that BAT and higher A/L account for this protective effect in humans.17

Activation of BAT in humans has been achieved primarily by using various cooling methods.4,5 These range from placing participants in a temperature-regulated cooling tent or room to submerging one foot in an ice bath.5,18

We tested the hypothesis that young adults with obesity and with a considerable volume of cold-activated BAT would have less metabolic dysfunction (specifically lower lipids, serum glucose and blood pressure) than those without cold-activated BAT; and that this relationship would be associated with a favorable A/L ratio. We report on the prevalence of cold-activated BAT using FDG-PET/CT in otherwise healthy, young adults with obesity undergoing a basic cooling protocol, as well as on the relationships between volumes and activities of cold-activated BAT with metabolic dysfunction criteria, lipids and adipokine concentrations.

Methods

We recruited men and women, 18–30 years of age, with obesity (BMI ≥ 30kg/m2). Participants were excluded if they had known thyroid disease, diabetes mellitus, familial dyslipidemia, or were taking second-generation anti-psychotics or prednisone. To avoid unnecessary radiation exposure, women who were pregnant at the time of enrollment were excluded. In addition, women were excluded if they had ever been pregnant, since changes in glucose and lipid metabolism vary significantly depending upon the duration of breastfeeding.19 Participants completed a questionnaire to provide demographic data. All data collection was performed from November 1 to April 15 (2010–2013), when the average temperature ranged from −7⁰C - 12.3⁰C.(National Weather Service, https://w2.weather.gov/climate/xmacis.php?wfo=slc)

The study was approved by the Institutional Review Board at the University of Utah and was in accordance with the principles set out in the Declaration of Helsinki. All participants provided written informed consent before enrollment.

Anthropometrics

Weight was obtained using a digital scale (Model 5002, Scale Tronix Inc., White Plains, NY). Height was measured to the nearest 0.1cm with a stadiometer (Model Hite-Rite 226, Seca, Chino, CA). Waist circumference was measured to the nearest 0.1cm with a measuring tape at the level of the umbilicus. BMI was calculated as kg/m2. Blood pressure was measured two times, usually in the right arm, using an automated sphygmomanometer (Model DINAMAP ProCare 400 Monitor, GE Healthcare, Fairfield, CT). If the difference between the measurements was greater than 5mmHg, a third measurement was performed and an average of the measurements was used.

Biomarkers

Participants fasted for at least 12 hours overnight before venipuncture. Serum adiponectin, leptin, and insulin concentrations were quantified in duplicate using commercial enzyme-linked immunosorbent assays (cat. nos: 80-ADPHU-E01, 11-LEPHU-E01, and 80-INSHU-E01.1, respectively, Alpco Diagnostics, Salem, NH; intra-assay precision: 1.0–7.4%, inter-assay precision 2.4–8.4%; sensitivity: 1.5 ng/mL; accuracy: 92–100%). Calibrators and quality controls were always included on the sample plate as unknowns. Standard curves had R2 values of ≥0.98. Samples exceeding the high-end detection limit were diluted 1:2 and re-assayed. Enzymatic colorimetric assays were used for quantification of total cholesterol, high-density lipoprotein cholesterol (HDL-C), triglycerides, and glucose (Sekisui Diagnostics, Burlington, MA; cat. nos: 234–60, 236–60, 220–32 and SE-035, [calibrator], respectively); HDL-C was quantified in the supernatants after precipitation of non-HDLs with dextran-sulfate/Mg2+.20 All specimens were assayed in duplicate with intra-assay coefficients of variation (CV) of <2% for total cholesterol, triglycerides and glucose or <5% for HDL-C.21 Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald equation.22

Metabolic Dysfunction

Our definition of metabolic dysfunction is consistent with that of Alberti et al.1 regarding the presence of at least one abnormal measurement of HDL-C, LDL-C, triglycerides, insulin resistance index (estimated using HOMA-IR=[glucose(mg/dL) x insulin(mIU/L)]/405), blood pressure, or waist circumference. Specific cutoffs were: HDL-C<40mg/dL (males), <50mg/dL (females), LDL-C≥130mg/dL, triglycerides≥150mg/dL, HOMA-IR≥3.16, blood pressure≥130/85mmHg, and waist circumference >94cm (males), >80cm (females).

Thermometry

Core body temperature was measured using a urine thermometer (cat. #ATC4030ADP, Medline Industries Inc., Salt Lake City, UT) before cooling, 2 hours after cooling prior to FDG-PET/CT scan, and 30 minutes after completion of FDG-PET/CT scan. Temporal artery temperature was measured using a Model TAT-5000 Temporal Artery Thermometer (Exergen Corporation, Watertown, MA), both before cooling and after FDG-PET/CT imaging.

FDG-PET/CT Scan

Participants avoided exercise and rigorous physical activity the day before FDG-PET/CT imaging, and fasted at least 12 hours prior to their scheduled exam. FDG was dosed before 12:01pm. Participants wore lightweight clothing (t-shirt, thin cotton pants, socks) and were seated for 2 hours in a semi-supine position in a hemodialysis chair, in a room cooled to 16ºC. Participants were instructed to inform research personnel if they felt too cold and were near shivering, at which point a thin cotton sheet was placed over the participant. Two participants requested this modification.

At the beginning of the second hour, approximately 15mCi of FDG was injected intravenously, and patients remained in a sitting position for the remaining hour. FDG-PET/CT imaging was performed using a standardized protocol,23 which included FDG uptake phase of 60 minutes following the injection of FDG and emission imaging for 5 minutes/bed position. The typical eyes-to-thighs imaging required 5 to 6 bed positions. The CT imaging was performed immediately prior to the FDG-PET scan using moderate dose technique. CT was also used to perform attenuation correction of emission FDG-PET images and to assist with anatomical localization. The FDG-PET images were reconstructed as a standardized uptake value (SUVbw) image normalized for body weight.

Image Analysis

The FDG-PET/CT images were interpreted by a board-certified nuclear medicine physician/expert reviewer (J.H.) to determine the presence or absence of cold-activated BAT, based on hypermetabolic FDG uptake in regions of BAT relative to surrounding tissue, as well as potential altered biodistribution due to muscle shivering. For each participant, those regions that were identified as visually positive for the presence of cold-activated BAT were then further analyzed quantitatively by another member of the research team (B.B.). For 2 subjects without visible BAT, we performed quantitative analysis to confirm the absence of cold-activated BAT. A 3D multi-modality method was developed to quantify 3 regions of cold-activated BAT. Inveon™ Research Workplace (Siemens Medical Solutions USA, Malvern, PA) software was used to quantitatively assess BAT by applying a series of threshold values based on both PET and CT segmentation. The quantitative Hounsfield Unit (HU) of CT was used initially to identify CT voxels containing fat and/or soft tissue by applying a threshold in the range of −300 to +200 HU in order to exclude voxels corresponding to air (HU<−300) and bone (HU>200) (Figure 1). The 3-dimensional location of the thresholded CT image voxels that were identified was saved. Since the FDG-PET and CT images are fused, the initially identified voxels were mapped to the FDG-PET SUVbw image set within the areas of interest. The initial FDG-PET images included WAT, BAT and cold-activated BAT. The FDG-PET SUVbw images were then further segmented to identify cold-activated BAT. Initially, PET segmentation used a fixed lower threshold of 1.5 SUVbw (Figure 2), based upon previously reported methods.24 In order to exclude normal hypermetabolic structures such as the brain or heart, manual editing of each axial plane was performed to exclude these structures from the region of interest (ROI). A final 3D volume of interest (VOI) of cold-activated BAT was formed by combining all of the relevant 2D ROIs. From these VOIs, the volume, minimum, maximum, and mean SUVbw were calculated from the PET images, as well as the same statistics for the HU values based on CT images. While the fixed SUVbw threshold of 1.5 has been previously reported,24 our preliminary analysis revealed that this threshold overestimated the BAT segmentation value (Figure 2), and that the results varied among participants. In order to obtain an optimal threshold that was participant-specific and applicable in this population with obesity, we repeated the FDG-PET segmentation, allowing the research team to interactively define a variable threshold for segmenting cold-activated BAT based on the optimal visual correlation of hypermetabolic foci within the appropriate anatomical structure of the VOI (Figure 3). The threshold was manually selected by the expert reviewers to include all of the increased FDG uptake above background in the PET image contained within the anatomic structure of BAT seen on the CT image. The fixed threshold method included voxels that were not within BAT (Figure 3). The total cold-activated BAT FDG uptake was calculated for both the fixed and variable threshold methods by multiplying the segmented volume (cc) by the SUVmean (unitless).

Figure 1.

Figure 1.

Open in a new tab

Coronal (left) and sagittal (right) images show green outline of mask based on CT Hounsfield Units used to exclude air (HU < −300) and bone (HU > 200) voxels prior to the segmentation of cold-activated BAT in cervical region. (Figures 1-3 depict the same male participant).

Figure 2.

Figure 2.

Open in a new tab

Coronal image showing segmented cold-activated BAT, which is overestimated using the fixed threshold of SUVbw > 1.5. (van der Lans et al. 2014).

Figure 3.

Figure 3.

Open in a new tab

Coronal imaging showing segmented cold-activated BAT based on subject-specific variable SUV threshold.

All data collection was completed prior to publication of the Brown Adipose Reporting Criteria in Imaging STudies (BARCIST 1.0).25 We have reported as many of the recommended parameters as possible. The FDG-PET/CT acquisition was compliant with the National Cancer Institute consensus guidelines for FDG-PET and FDG-PET/CT UPICT protocol,23 which were contributing standards cited in the later BARCIST criteria. In addition, a retrospective analysis was performed to evaluate the impact of Fat Free Mass (FFM) corrections to estimate SUVlean and the proposed thresholds for segmenting BAT. The FFM was estimated from the weight and BMI using the sex-specific formulae.26 The SUVlean was calculated as: SUVbw x (FFM/body weight). The optimal variable thresholds for SUVbw in the original analysis were compared to the patient-specific thresholds suggested by BARCIST based on the (lean body mass)/(body mass) [LBM/BM] correction.25

Statistical Analyses

Demographic data were reported as means ± standard deviation. The Wilcoxon rank-sum test was used to compare continuous variables of metabolic dysfunction in those with cold-activated BAT and those without cold-activated BAT. Pearson and Spearman correlation coefficients were used to assess the associations of BAT volume (cc) and metabolic activity (mean SUVbw and SUVlean) with anthropometric and blood biomarkers. Statistical analyses were performed using SAS (version 9.4; SAS Institute, Cary, NC). P<.05 was considered statistically significant.

Results

We enrolled 44 participants, 21 females and 23 males. The means (and ranges) for age, BMI, and waist circumference were 25.0 years (18.6–31.5), 35.4kg/m2 (29.2–54.4) and 116.0cm (84.4–155.7), respectively. One participant had BMI 30.0 at initial screening but had lost weight and BMI was 29.3 at the time of study procedures. All but one participant had a large waist circumference that met the criteria for metabolic dysfunction (Table). Most of the participants were non-Hispanic white (n=36), 4 Hispanic, two Pacific Islander (males), one was Hispanic black (female), and one did not answer. Neither core body temperature nor temporal artery temperature changed significantly from baseline to 2 hours after completion of the cooling protocol, or to 30 minutes after completion of PET/CT scan.

Table.

Comparison of Demographics, Serum Biomarkers and Skin Temperatures in Participants with Obesity and With (11 female, 8 male) and Without (10 female, 15 male) Cold-Activated Brown Adipose Tissue

Variable Cold-Activated Brown Adipose Tissue Significance
Present Absent
Female 11 10 NS
Male 8 15 NS
Age (y) 23.6 (4.1) 26.4 (3.6) P = .01
Temperature (⁰C)
Pre-cooling 36.8 (0.4) 36.9 (0.5) NS
2 hours after cooling 36.6 (0.5) 36.6 (0.4) NS
After 18FDG-PET/CT 36.6 (0.5) 36.6 (0.4) NS
BMI (kg/m2) 36.0 (6.3) 37.4 (6.0) NS
BMI 30 - <40 (n) 14 18 NS
BMI ≥ 40 (n) 5 7 NS
WC (cm) 113.0 (17.9) 118.9 (17.9) NS
Systolic blood pressure (mmHg) 118 (11.4) 121 (10.8) NS
Diastolic blood pressure (mmHg) 71 (8.6) 72 (7.6) NS
Total cholesterol (mg/dL) 169 (27.2) 184 (29.5) NS
Triglycerides (mg/dL) 108 (45.7) 185 (113.4) P = .01
HDL-C (mg/dL) 44 (9.8) 41 (12.6) NS
LDL-C (mg/dL) 104 (21.9) 108 (32.5) NS
Glucose (mg/dL) 92 (5.0) 95 (8.0) NS
Insulin (mIU/L) 19 (17.7) 20 (15.1) NS
HOMA-IR (unitless) 4.4 (4.2) 4.8 (3.6) NS
Number of Criteria of Metabolic Dysfunction (unitless) 2.4 (0.8) 3.1 (1.3) P = .02
Adiponectin (μg/mL) 8 (4.1) 8 (7.9) NS
Leptin (ng/mL) 72 (61.1) 56 (42.1) NS
Adiponectin/Leptin 262 (294) 348 (340) NS
Open in a new tab

Results are presented as mean (standard deviation). y = years; 18FDG-PET/CT = 18F-fluorodeoxyglucose positron emission topography/computed cosmography; BMI = body mass index; WC = waist circumference; HDL-C = high density lipoprotein cholesterol; LDL-C = low density lipoprotein cholesterol; HOMA-IR = homeostatic model assessment of insulin resistance; NS = not significant

Total cold-activated BAT was detected in 19 (43%) participants (11 females, 8 males; 5 [1 male] with BMI≥40kg/m2); this study was insufficiently powered to allow comparisons of cervical vs thoracic vs paranephric BAT. Biomarker data were missing for one participant with cold-activated BAT and one without cold-activated BAT. Those with cold-activated BAT were younger (23.6 years [range:18.8–30.1] vs 26.4 [range:18.6–31.5]; P=0.01) and, importantly, had fewer criteria of metabolic dysfunction (2.4 [range:1–4] vs 3.1 [range:1–6]; P=0.02) (Table). Serum triglycerides were on average 70% higher in subjects lacking cold-activated BAT.

Based on visual inspection by the expert reviewer, the fixed threshold method using SUVbw>1.5, generated regions that extended beyond the cold-activated BAT and included other tissue. We included fixed threshold method analysis to allow a direct comparison with the variable threshold method. Using the variable threshold method, the optimal SUVbw threshold was always greater than the fixed value of 1.5 (mean=1.98, min=1.8, max=2.0), as predicted by BARCIST criteria in the population with obesity. However, the resulting minimum SUVlean values were highly variable (mean=1.14, min=0.86, max=1.34). The fixed threshold method overestimated the volume of cold-activated BAT by 31% (mean=300cc) when compared to the user-defined variable threshold (mean=228cc). Conversely, the fixed threshold method underestimated the mean SUVbw of cold-activated BAT by 20% (mean=3.6) when compared to the variable threshold method (mean=4.3). However, the correlation between the two methods for volume and mean SUVbw was excellent (R2=0.94 and 0.96, respectively) (Figure 4). The total cold-activated BAT FDG uptake was overestimated with the fixed threshold method (mean=1048) when compared to the variable threshold method (mean=856). The coefficient of variation of cold-activated BAT SUVbw was also greater using the fixed threshold method compared to the variable threshold method (174% vs. 157%). In the two subjects who reported feeling cold we did not find evidence of altered biodistribution of FDG due to physiologic uptake in muscle.

Figure 4:

Figure 4:

Open in a new tab

Cold-activated BAT volume and FDG uptake (mean SUVbw) measured with variable versus fixed threshold methods. Note: Points represent regions (cervical, thoracic) with BAT, not participants with BAT.

We found no significant relationships between BAT activity or BAT volume and leptin, A/L ratio, glucose, insulin, HOMA-IR, LDL-C, BP, waist circumference or age. However, there was a strong positive correlation between BAT activity (mean SUVbw) and HDL-C (r=0.5958, P=0.0091; r=0.6383, P = 0.0044, for the fixed threshold and variable threshold methods, respectively, n=18; Figure 5), and a moderate inverse correlation between BAT volume and triglycerides (r=−0.4728, P=0.0475; r=−0.4703, P=0.0489, for the fixed threshold and variable threshold methods, respectively, n=18; Figure 5). BAT volume by the variable threshold method had a weak positive correlation with adiponectin (r=0.5125, P=0.0354, n=17; Figure 5).

Figure 5:

Figure 5:

Open in a new tab

Associations between BAT volumes and BAT activity (mean SUVbw) with triglycerides, HDL-cholesterol and adiponectin. Results from fixed and variable threshold calculation methods are depicted side-by-side for comparison. Regression lines are shown for the significant associations only.

Our optimal variable threshold for defining the minimum SUVbw had a median of 2.0 with a narrow range (1.8–2.0). This was consistent with the BARCIST example of a person with obesity with LBM/BM=0.6, resulting in an SUVbw of 2.0. However, the LBM/BM ratio was highly variable in our study population with obesity with a range of 0.43–0.67, and the BARCIST proposed subject specific-LBM corrected SUV for body mass (SUVbw) thresholds were also highly variable (1.79–2.79) with the majority (21/28, 75%) having an overestimation of the SUVbw threshold.

Finally, in subjects with detectable cold-activated BAT, males had significantly lower leptin concentrations (35±30ng/mL vs 98±65ng/mL, P=0.031) and higher A/L ratios (439±377 vs 138±132, P=0.033) than females. Overall, we have treated the findings as hypothesis-generating, and did not formally adjust for multiple significance testing.

Discussion

Our goal in conducting this study was to enhance understanding of the mechanism underlying metabolically healthy obesity. We tested the hypothesis that people with obesity who are metabolically healthy have metabolically active BAT that is known to secrete adiponectin, which protects them from metabolic dysfunction.

This report presents results from a cross-sectional study of a moderate-sized cohort of young adults with obesity and demonstrable cold-activated BAT. Our study differs substantially from previous studies in that we enrolled only young adults (women and men) with obesity.18,24,27 We documented that both men and women with obesity and with cold-activated BAT have a lower degree of metabolic dysfunction compared to those without cold-activated BAT. At least two components of metabolic dysfunction (HDL cholesterol and triglycerides), and one adipokine (adiponectin) showed strong or modest correlations with cold-activated BAT. Our findings are similar to those of Chondronikola et al.,27 who demonstrated that both improved glucose utilization and insulin sensitivity are associated with cold-activated BAT in middle-aged, overweight healthy men. Furthermore, our findings regarding the prevalence of metabolically active BAT in young adults with obesity are similar to those which Yoneshiro et al.28 observed in adults of all ages and with normal weight. Our study shows that, in the presence of obesity ranging from mild to severe, the previously described gender difference of women having more cold-activated BAT than men, as well as the inverse relationship between cold-activated BAT and age are present in even a relatively narrow age range (18–31 years old).3,4 Interestingly, males with cold-activated BAT had lower leptin and higher A/L ratios than females, but the reason for the gender difference cannot be gleaned from our study.

Using our simple and easily reproducible cooling methods, BAT activation occurred in almost half of our study participants (50% of females, 35% of males). FDG-PET/CT scans did not reveal visually observable FDG uptake in skeletal muscle; thus, no evidence of (imperceptible) shivering was present. Other studies have reported the use of specialized cooling suits to promote BAT activation, telemetric capsules to monitor internal body temperatures, and electrodes placed to detect muscle twitching in order to determine the most appropriate temperature for BAT activation.18,29 The percentage of individuals with cold-activated BAT has varied from 23% to 100%.18,30 Our results, which are similar, indicate that technologically advanced cooling methods used by others may not be necessary to produce and demonstrate BAT cold-activation.

Several prior studies have demonstrated a higher prevalence of BAT among females than among males, as well as an inverse association between BAT and BMI.3,18,29 Our findings have documented the presence of cold-activated BAT in almost half of a large cohort of men and women with obesity. In comparison, one study of 15 (13 female; mean age 39.2y) subjects with severe obesity (BMI>40) found low cold-activated BAT among only 3 (23%) of the females.18 Another study, which evaluated 10 metabolically healthy men with overweight/obesity (mean BMI 32.9, mean age 36.0), found cold-activated BAT in 60% of subjects.31 These differences in cold-activated BAT prevalence may be due to the younger age of our cohort. Recently, Grundy et al.32 documented the absence of a relationship between waist circumference and metabolic syndrome. Thus, it may not be surprising that we observed no relationship between waist circumference and cold-activated BAT (volume or mean SUVbw), as well as between cohorts with mild to moderate obesity (BMI 29.2-< 40) and extreme obesity (BMI≥40). These findings suggest that total abdominal adiposity may not affect BAT activity in young adults with obesity.

We did not detect any significant changes in temperature before and after cooling and after 18FDG-PET/CT imaging in those with and without cold-activated BAT. This suggests that short-term activation of BAT does not significantly contribute to changes in body temperature after cooling.

Our findings demonstrate that young adults with obesity and cold-activated BAT have less metabolic dysfunction than young adults with obesity without cold-activated BAT. From a clinical perspective, less metabolic dysfunction results in a lower Framingham Risk Score for coronary heart disease.33 Some have suggested that BAT has a physiologic role in systemic energy metabolism, specifically in diet-induced thermogenesis and whole-body fat utilization.34,35 Furthermore, previous studies have assessed the potential for BAT, as well as for WAT induced to behave like BAT, to serve as therapeutic targets for metabolically unhealthy obesity, diabetes and dyslipidemia.34,36–38 We agree with previous and current attempts to induce BAT activity in people with obesity and/or type 2 diabetes, in hopes of improving the overall, including and especially the metabolic, health of these and perhaps other patient populations.6,39

Several limitations of this study should be recognized. We did not report percent body fat because currently available skinfold and bioelectrical impedance analysis methods are notoriously inaccurate in individuals with this degree of adiposity.40,41 Many of our participants had such extensive subcutaneous fat that the calipers were not large enough to provide reproducible skinfold measurements. It is possible that our cold exposure protocol may have been too mild to assess BAT activity in adults with obesity; consequently, we may have underestimated the prevalence of BAT activity in our subjects. Additionally, we did not conduct formal room temperature measurements at frequent intervals, although we did monitor the temperature to ensure it remained below 16⁰C for the duration of the cooling protocol.

Also, there was a lack of racial/ethnic diversity among our largely non-Hispanic white participants. We recruited most of our participants from the University of Utah campus and the surrounding Salt Lake City metropolitan area, where the general population is 65% non-Hispanic white (https://www.census.gov/quickfacts/fact/table/saltlakecitycityutah,saltlakecountyutah,UT/PST040217#viewtop). Furthermore, we collected fasting blood samples only once before participants underwent the cooling protocol. While the results based on fasting indices of metabolic health are interesting, more extensive studies are needed to determine whether differences in BAT cold-activation underlie differences in metabolically healthy or unhealthy obesity, and if these differences are modulated by adiponectin. Our cross-sectional study was not designed to investigate if metabolically active BAT directly secretes adiponectin, which has been previously demonstrated in humans.12

An additional limitation is that we did not perform tissue biopsies to confirm the presence of BAT; however, our observed anatomical areas of FDG-avid BAT are identical to those described by several other investigators.3,42,43 Financial constraints and increased exposure to (and risk from) radiation prevented us from doing multiple time-point PET/CT scans in this young cohort in order to evaluate differences in FDG-avid versus non-FDG-avid BAT.10

While this study was performed prior to the development of BARCIST criteria,25 it confirms the BARCIST recommendation that a different SUVbw threshold is more optimal for people with obesity. However, the correction for FFM to estimate SUVlean results in increased variability and a generalized overestimation of subject-specific SUVbw. Despite the objective nature of using either a fixed SUVbw threshold or a patient-specific threshold based on LBM, our results demonstrate that these approaches result in discordances in the automated segmentation of BAT compared to the determination based on guidance from an expert radiologist. This is to be expected, given the known variability in PET scanner SUV quantification at the voxel level as well as the biological variability in glucose metabolism across individuals. However, since SUVlean measurements are more comparable across males and females, we suggest that SUV normalization for FFM may be more appropriate than normalizing for body mass when evaluating potential differences due to gender with FDG-PET.

Conclusion

We have demonstrated a high prevalence of metabolically active BAT among young adults with obesity, and documented associations between the presence and activity of cold-activated BAT with a more favorable cardiometabolic profile. We suggest that increasing BAT thermogenesis be a target for treating metabolically unhealthy obesity.

What is already known about this subject?

Brown adipose tissue (BAT) is present in lean and overweight adults. Brown adipose tissue is associated with lower body weight, lower serum glucose and better lipid profile.

What does this study add?

Cold-activated BAT is observed in young adults (females and males) with obesity and its presence is associated with a healthier cardiometabolic profile. In addition, our larger cohort of young adults (female and male) with obesity corroborates previous research in smaller cohorts (primarily males), and we have shown that technologically advanced methods are unnecessary in order to successfully study cold-activation. Furthermore, our findings suggest that there might not be an effect of total abdominal adiposity on BAT activity in young adults with obesity, and that short-term BAT activation is not a significant contributor to body temperature changes following cooling.

How the results of this study might change the direction of research or the focus of clinical practice:

Further development of the BARCIST criteria for improved accuracy in the evaluation of BAT in people with obesity. We suggest additional research on the use of fat free mass corrections to estimate SUVlean and the proposed thresholds for segmenting BAT.

Acknowledgements

The authors are grateful for the support of Jaci Skidmore, Regan Butterfield, Fumiko Alger, and Elizabeth Phillips for assistance with data collection; Paul Young and Susan Schulman for editing and assistance with manuscript preparation. Support for infrastructure and personnel was also provided by the Center for Quantitative Cancer Imaging at the Huntsman Cancer Institute. The authors thank the study participants. Individual deidentified participant data will be available beginning 3 months and ending 5 years following article publication. Data include demographics, anthropometry, study protocol, statistical analysis plan, and informed consent form. Researchers who provide a methodologically sound proposal may request access to the data in order to achieve aims in the approved proposal. All proposals should be directed to nicole.mihalopoulos@hsc.utah.edu.

Funding:

This work was supported by the National Heart, Lung and Blood Institute (grant number HL#092069) and by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR001067. Additional infrastructure and personnel support were provided by the Center for Quantitative Cancer Imaging at the Huntsman Cancer Institute at the University of Utah. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References

  • 1.Alberti KG, Zimmet P, Shaw J. Metabolic syndrome--a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med. 2006;23(5):469–480. [DOI] [PubMed] [Google Scholar]
  • 2.Poon VT, Kuk JL, Ardern CI. Trajectories of metabolic syndrome development in young adults. PLoS One. 2014;9(11):e111647. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Cypess AM, Lehman S, Williams G, et al. Identification and importance of brown adipose tissue in adult humans. N Engl J Med. 2009;360(15):1509–1517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Ouellet V, Routhier-Labadie A, Bellemare W, et al. Outdoor temperature, age, sex, body mass index, and diabetic status determine the prevalence, mass, and glucose-uptake activity of 18F-FDG-detected BAT in humans. J Clin Endocrinol Metab. 2011;96(1):192–199. [DOI] [PubMed] [Google Scholar]
  • 5.Saito M, Okamatsu-Ogura Y, Matsushita M, et al. High incidence of metabolically active brown adipose tissue in healthy adult humans: effects of cold exposure and adiposity. Diabetes. 2009;58(7):1526–1531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Iwen KA, Backhaus J, Cassens M, et al. Cold-Induced Brown Adipose Tissue Activity Alters Plasma Fatty Acids and Improves Glucose Metabolism in Men. J Clin Endocrinol Metab. 2017;102(11):4226–4234. [DOI] [PubMed] [Google Scholar]
  • 7.Koskensalo K, Raiko J, Saari T, et al. Human Brown Adipose Tissue Temperature and Fat Fraction Are Related to Its Metabolic Activity. J Clin Endocrinol Metab. 2017;102(4):1200–1207. [DOI] [PubMed] [Google Scholar]
  • 8.Blondin DP, Labbé SM, Noll C, Kunach M, Phoenix S, Guérin B, Turcotte ÉE, Haman F, Richard D, Carpentier AC. Selective Impairment of Glucose but Not Fatty Acid or Oxidative Metabolism in Brown Adipose Tissue of Subjects With Type 2 Diabetes. Diabetes. 2015;64(7):2388–97. [DOI] [PubMed] [Google Scholar]
  • 9.Sampath SC, Sampath SC, Bredella MA, Cypess AM, Torriani M. Imaging of Brown Adipose Tissue: State of the Art. Radiology. 2016;280(1):4–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Upadhye T, Gandhi A, Basu S. Evaluation of (18)F-FDG Uptake Pattern in Brown Adipose Tissue Over Extended Time Period as Assessed by Multiple Time Point (18)F-FDG-PET. Nucl Med Mol Imaging. 2013;47(2):89–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Galic S, Oakhill JS, Steinberg GR. Adipose tissue as an endocrine organ. Mol Cell Endocrinol. 2010;316(2):129–139. [DOI] [PubMed] [Google Scholar]
  • 12.Iacobellis G, Di Gioia C, Petramala L, et al. Brown fat expresses adiponectin in humans. Int J Endocrinol. 2013;2013:126751. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Falahi E, Khalkhali Rad AH, Roosta S. What is the best biomarker for metabolic syndrome diagnosis? Diabetes Metab Syndr. 2015;9(4):366–372. [DOI] [PubMed] [Google Scholar]
  • 14.Selthofer-Relatic K, Radic R, Stupin A, et al. Leptin/adiponectin ratio in overweight patients - gender differences. Diab Vasc Dis Res. 2018;15(3):260–262. [DOI] [PubMed] [Google Scholar]
  • 15.Berbee JF, Boon MR, Khedoe PP, et al. Brown fat activation reduces hypercholesterolaemia and protects from atherosclerosis development. Nat Commun. 2015;6:6356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Wang TY, Liu C, Wang A, Sun Q. Intermittent cold exposure improves glucose homeostasis associated with brown and white adipose tissues in mice. Life Sci. 2015;139:153–159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hoeke G, Kooijman S, Boon MR, Rensen PC, Berbee JF. Role of Brown Fat in Lipoprotein Metabolism and Atherosclerosis. Circ Res. 2016;118(1):173–182. [DOI] [PubMed] [Google Scholar]
  • 18.Vijgen GH, Bouvy ND, Teule GJ, Brans B, Schrauwen P, van Marken Lichtenbelt WD. Brown adipose tissue in morbidly obese subjects. PLoS One. 2011;6(2):e17247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Stuebe AM, Rich-Edwards JW. The reset hypothesis: lactation and maternal metabolism. Am J Perinatol. 2009;26(1):81–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Warnick GR, Benderson J, Albers JJ. Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin Chem. 1982;28(6):1379–1388. [PubMed] [Google Scholar]
  • 21.Nanjee MN, Miller NE. Evaluation of long-term frozen storage of plasma for measurement of high-density lipoprotein and its subfractions by precipitation. Clin Chem. 1990;36(5):783–788. [PubMed] [Google Scholar]
  • 22.Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499–502. [PubMed] [Google Scholar]
  • 23.Graham MM, Wahl RL, Hoffman JM, et al. Summary of the UPICT Protocol for 18F-FDG PET/CT Imaging in Oncology Clinical Trials. J Nucl Med. 2015;56(6):955–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.van der Lans AA, Wierts R, Vosselman MJ, Schrauwen P, Brans B, van Marken Lichtenbelt WD. Cold-activated brown adipose tissue in human adults: methodological issues. Am J Physiol Regul Integr Comp Physiol. 2014;307(2):R103–113. [DOI] [PubMed] [Google Scholar]
  • 25.Chen KY, Cypess AM, Laughlin MR, et al. Brown Adipose Reporting Criteria in Imaging STudies (BARCIST 1.0): Recommendations for Standardized FDG-PET/CT Experiments in Humans. Cell Metab. 2016;24(2):210–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Janmahasatian S, Duffull SB, Ash S, Ward LC, Byrne NM, Green B. Quantification of lean bodyweight. Clin Pharmacokinet. 2005;44(10):1051–65. [DOI] [PubMed] [Google Scholar]
  • 27.Chondronikola M, Volpi E, Borsheim E, et al. Brown adipose tissue improves whole-body glucose homeostasis and insulin sensitivity in humans. Diabetes. 2014;63(12):4089–4099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Yoneshiro T, Aita S, Matsushita M, et al. Age-related decrease in cold-activated brown adipose tissue and accumulation of body fat in healthy humans. Obesity (Silver Spring). 2011;19(9):1755–1760. [DOI] [PubMed] [Google Scholar]
  • 29.Lee P, Smith S, Linderman J, et al. Temperature-acclimated brown adipose tissuemodulates insulin sensitivity in humans. Diabetes. 2014;63(11):3686–3698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Ouellet V, Labbe SM, Blondin DP, et al. Brown adipose tissue oxidative metabolism contributes to energy expenditure during acute cold exposure in humans. J Clin Invest. 2012;122(2):545–552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hanssen MJ, van der Lans AA, Brans B, et al. Short-term Cold Acclimation Recruits Brown Adipose Tissue in Obese Humans. Diabetes. 2016;65(5):1179–1189. [DOI] [PubMed] [Google Scholar]
  • 32.Grundy SM, Williams C, Vega GL. Upper body fat predicts metabolic syndrome similarly in men and women. Eur J Clin Invest. 2018;48(7):e12941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.D’Agostino RB Sr, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008;117:743–53. [DOI] [PubMed] [Google Scholar]
  • 34.Hankir MK. Loading and firing the brown adipocyte. Adipocyte. 2018;7(1):4–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hibi M, Oishi S, Matsushita M, et al. Brown adipose tissue is involved in diet-induced thermogenesis and whole-body fat utilization in healthy humans. Int J Obes (Lond). 2016;40(11):1655–1661. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hankir MK, Cowley MA, Fenske WK. A BAT-Centric Approach to the Treatment of Diabetes: Turn on the Brain. Cell Metab. 2016;24(1):31–40. [DOI] [PubMed] [Google Scholar]
  • 37.Li S, Li Y, Xiang L, Dong J, Liu M, Xiang G. Sildenafil induces browning of subcutaneous white adipose tissue in overweight adults. Metabolism. 2018;78:106–117. [DOI] [PubMed] [Google Scholar]
  • 38.Ridler C Adipose tissue: BAT affects lipid metabolism. Nat Rev Endocrinol. 2016;12(8):435. [DOI] [PubMed] [Google Scholar]
  • 39.Poher AL, Altirriba J, Veyrat-Durebex C, Rohner-Jeanrenaud F. Brown adipose tissue activity as a target for the treatment of obesity/insulin resistance. Front Physiol. 2015;6:4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Alvarez VP, Dixon JB, Strauss BJ, Laurie CP, Chaston TB, O’Brien PE. Single frequency bioelectrical impedance is a poor method for determining fat mass in moderately obese women. Obes Surg. 2007;17(2):211–221. [DOI] [PubMed] [Google Scholar]
  • 41.Das SK. Body composition measurement in severe obesity. Curr Opin Clin Nutr Metab Care. 2005;8(6):602–606. [DOI] [PubMed] [Google Scholar]
  • 42.Hui SCN, Ko JKL, Zhang T, et al. Quantification of brown and white adipose tissue based on Gaussian mixture model using water-fat and T2* MRI in adolescents. J Magn Reson Imaging. 2017;46(3):758–768. [DOI] [PubMed] [Google Scholar]
  • 43.Nedergaard J, Bengtsson T, Cannon B. Unexpected evidence for active brown adipose tissue in adult humans. Am J Physiol Endocrinol Metab. 2007;293(2):E444–452. [DOI] [PubMed] [Google Scholar]

RESOURCES