Next Article in Journal
Associations between ZNF676, CTC1 Gene Polymorphisms and Relative Leukocyte Telomere Length with Myopia and Its Degree
Previous Article in Journal
Advances in the Treatment of Kidney and Upper Urinary Tract Cancers
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Biomedicines
Volume 12
Issue 3
10.3390/biomedicines12030537
biomedicines-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Metabolic Effects of Brown Adipose Tissue Activity Due to Cold Exposure in Humans: A Systematic Review and Meta-Analysis of RCTs and Non-RCTs

by
Shirin Tabei
1,2,3,
Rodrigo Chamorro
1,4,
Sebastian M. Meyhöfer
1,2,3 and
Britta Wilms
1,2,3,*
1
Institute of Endocrinology and Diabetes, University of Lübeck, 23562 Lübeck, Germany
2
Center of Brain, Behavior, and Metabolism (CBBM), University of Lübeck, 23562 Lübeck, Germany
3
German Center for Diabetes Research (DZD), Partner Düsseldorf, 85764 München-Neuherberg, Germany
4
Department of Nutrition, Faculty of Medicine, University of Chile, Santiago 8380453, Chile
*
Author to whom correspondence should be addressed.
Biomedicines 2024, 12(3), 537; https://doi.org/10.3390/biomedicines12030537
Submission received: 12 December 2023 / Revised: 7 February 2024 / Accepted: 18 February 2024 / Published: 28 February 2024
(This article belongs to the Section Endocrinology and Metabolism Research)
Browse Figures
Review Reports Versions Notes

Abstract

:
Brown adipose tissue (BAT), specialized in thermoregulation in mammals, has been linked to improved glucose and lipid homeostasis when activated by cold exposure (CE). This systematic review and meta-analysis assessed the metabolic effects of CE-induced BAT activation in healthy humans, examining changes in glucose and lipid metabolism compared to thermoneutrality (TN). A literature search was conducted, identifying relevant human studies, including randomized controlled trials (RCTs) and non-RCTs, based on predefined inclusion criteria. Seven studies (a total of 85 participants) fully met the criteria. Data on plasma glucose, insulin, triglycerides (TGs), and free fatty acids (FFAs) were extracted for meta-analysis. When comparing TN and CE under fasting conditions, there were no significant changes in glucose, insulin, or TG concentrations (all p > 0.36). In contrast, CE significantly increased FFA concentrations (p = 0.002; n = 38). Bias was absent for all parameters, but heterogeneity was observed for insulin (I2 = 74.8%). CE primarily affects FFA concentration, likely reflecting cold-induced BAT activity. This suggests that circulating FFAs, serving as the primary fuel for thermogenesis, could indicate BAT activation. However, understanding the effects of BAT activation on overall metabolism requires a broader approach beyond fasting glucose and lipid concentration measurements.

1. Introduction

Obesity has reached epidemic proportions [1]. Obesity is associated with metabolic diseases such as type 2 diabetes (T2D), dyslipidemia, or metabolic syndrome. In terms of thermodynamics, if energy intake exceeds energy expenditure, a positive energy balance results in subsequent body weight and fat mass gain and, if persistent, in the development of obesity. However, a low success rate of weight reduction by decreasing caloric intake has been reported [2]. Long-term maintenance of weight loss is difficult and may ultimately lead to weight gain after an initial phase of weight loss [3,4].
Several animal and human studies have addressed strategies to increase energy expenditure to force a negative energy balance with subsequent weight loss, for example, by increasing physical activity [5,6].
In recent years, and after the rediscovery of brown adipose tissue (BAT) in adult humans [7,8,9], activation and recruitment of this fat tissue moved to the center of scientific interest as a potential anti-obesity and anti-diabetes target.
BAT is a highly specialized type of fat tissue mainly found in newborns of mammals and human infants, whose primary functions involve non-shivering thermogenesis (NST) [10] and the production of heat to maintain normal body temperature [11]. Breastfeeding plays an important role in the development of thermogenic fat in human newborns [12,13,14]. After birth, maternal cues including free fatty acids (FFAs) and alkylglycerols are delivered through milk [12,15] and microbial products of the gut microbiota to infants [16], which triggers thermogenic (beige/brown) fat development. The mechanism behind this development is due to beta-adrenergic signaling [15].
In adults, BAT is most often located in the supraclavicular region and, to a lesser extent, in the abdomen, around the spine, and around the heart and kidney [17]. Both experimental and clinical studies have shown that activated BAT can potentially enhance glucose metabolism [18] and lipid metabolism [19], increase metabolic rate [20], and promote the loss of fat mass in humans [21]. Interestingly, preserving body core temperature in adults depends on the subcutaneous white adipose tissue and its insulating function rather than thermogenic fat [22].
Researchers have investigated different mediators to activate BAT, including diet, substrates (e.g., capsaicin), medications such as beta-adrenergic (β-AR) agonists (e.g., mirabegron), and cold exposure (CE) [8,23,24,25]. CE has been reported as the most effective activator of BAT [26]. Mild CE stimulates the activity of the sympathetic nervous system, leading to the release of norepinephrine (NE). NE then binds to β-AR receptors located on the membrane of brown adipocytes [27]. Furthermore, intracellular signaling cascades lead to triglyceride (TG) degradation into FFA, which interacts with uncoupling protein 1 (UCP1). UCP1 is a hallmark of BAT [28] and is highly abundant on the inner mitochondrial membrane. If activated, UCP1 uncouples the electron transport chain independently of the consumption of adenosine triphosphate, leading to a free flow of protons across the inner mitochondrial membrane. This process, in turn, leads to NST and the production of heat [11]. By comparing Positron Emission Tomography and Computed Tomography (PET/CT) images before and after 2 h of CE at 19 °C, [18F]-Fluorodeoxyglucose ([18F]-FDG) uptake in BAT is increased, specifically in the supraclavicular region, indicating the potential of mild CE to activate BAT [8,9].
Studies in rodents have provided evidence of improved insulin sensitivity and browning in white adipose tissue in response to CE in obese mice [29,30,31]. Bartlett and colleagues showed a TG clearance after 24 h of CE in hypertriglyceridemic mice, specifically mediated by BAT [32]. In humans, Blondin et al. reported a 45% increased BAT oxidative metabolism and BAT volume in healthy young lean and overweight men by daily CE at 10 °C, who experienced low shivering for 2 h over 4 weeks [33]. We have also shown that 2 h of moderate CE at 18 °C increased BAT activity and improved insulin sensitivity in healthy lean participants [34].
In sum, substantial evidence suggests that CE leads to increased BAT activity and subsequent beneficial metabolic adaptations in humans.
This systematic review and meta-analysis comprehensively gather and evaluate the impact of CE without experience of shivering on healthy humans compared to TN. Specifically, it focuses on glucose and lipid metabolism only at the fasting level to ascertain whether mild CE influences the aforementioned parameters at a basal level.

2. Methods

2.1. Search Strategy

Computerized searches were performed in PubMed (MEDLINE) and Web of Science, respectively, with the initial search conducted on 7 March 2022, and the latest search performed on 29 March 2023. The following search terms were used:
“(((“Brown Adipose Tissue”“ OR ““BAT”“ OR ““Brown Fat”“) AND “(“Cold Exposure”“ OR ““Cold Expos*”“ OR ““Cold Stimulus”“ OR ““Cold Induced”“ OR ““Cold Acclimation”“ OR ““Cold Effect”“) AND “(“Metabolic Homeostasis”“ OR ““Metabolic Effect”“ OR ““Metabolic Response”“ OR ““Metabolic Consequence”“ OR ““Metabolic Regulation”“ OR ““Metabolic Impact”“ OR ““Metabolism”“) AND “(“Humans”“ OR ““Subjects”“ OR ““Participants”“ OR ““Individuals”“)) NOT “(“Animals”“)).

2.2. Inclusion and Exclusion Criteria

The protocols of studies investigating the effects of CE on BAT activity vary in design (i.e., comparing CE vs. TN), the temperature level of the CE (i.e., from 4 °C to 21 °C), and regarding cooling procedures (i.e., using cooling feet and hands or water-perfused suits), and the duration of CE. Some studies have used cooling protocols that allowed low shivering [33,35,36,37], whereas others avoided shivering (see Section 3.1). Therefore, to extract data for this meta-analysis, we aimed to select studies that are as homogenous and comparable as possible (Table 1 and Table 2). We included original studies, RCTs, and non-RCTs conducted in healthy adults (≥18 years of age), both male and female, and of any geographic area. Studies were included if the temperature range of intervention was between 12–19 °C and between 20–25 °C for CE and TN conditions, respectively. The range of temperature refers to the general average of previous applied cooling studies in the literature.
Additionally, studies with personalized cooling protocols, in which the CE temperature was set above the shivering threshold of each subject individually, were also included. Lastly, concerning metabolic outcome parameters, research was restricted to studies reporting basal glucose, insulin, TG, and FFA concentrations. Studies with either lower (<12 °C) or higher (>26 °C) temperatures were considered “not mild CE” or “warm but not TN” and were, therefore, excluded. Studies reporting a shivering observation during the CE protocol used and studies with shorter (i.e., nighttime fasting lower than 8 h) or longer fasting periods (i.e., 24 h or more) were also excluded.

2.3. Data Selection

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [45], three investigators (ST, RC, and BW) independently screened all titles and abstracts of papers to assess them for potential relevance. If no agreement was achieved, the respective titles were reviewed by a fourth researcher (SMM). The systematic review and meta-analysis included studies conducted on healthy adults who underwent both TN and CE and fulfilled all inclusion and exclusion criteria (see Section 2.2).

2.4. Meta-Analytic Approach

The meta-analysis procedure was performed using the RStudio (version 2023.06.02, Boston, MA, USA) library package “metafor” (version 2.0-0). Means and standard deviations (SDs) of fasting glucose, insulin, TG, and FFA concentrations during both CE and TN were extracted (Table 2).
If needed, a 95% confidence interval (CI) was converted to an SD using the following equation:
SD   = N × upper   limit   −   lower   limit / 4.128
A random-effects model was used to measure the impact of TN vs. CE on glucose, insulin, TG, and FFA concentrations. The effect size was provided by the standardized mean difference (SMD), calculated as follows:
SMD = new treatment improvement − comparator (placebo)
improvement/pooled standard deviation
The studies’ 95% CI and corresponding Z values (i.e., ±1.96) were estimated. The findings of the meta-analysis were graphically displayed by forest plots. When the 95% CI from a single study (or the pooled estimate) crosses the line of no effect, the difference in outcome between CE and TN is not statistically significant. Heterogeneity was assessed with Cochran’s Q test and I2 statistics. Cochran’s Q test is calculated as the weighted sum of squared differences between individual study effects and the pooled effect across studies. The I2 statistic describes the variability in effect estimates due to heterogeneity rather than sampling error. The following formula was used to quantify the degree of inconsistency or heterogeneity among the individual studies included in the analysis (df: degrees of freedom; Q: chi-squared statistics) [46]:
I2 = (Q − df/Q) × 100
Conventions were followed regarding the interpretation of I2: 0–40% may represent no heterogeneity; 30–60% may represent moderate heterogeneity; 50–90% may represent substantial heterogeneity, and 75–100% may represent considerable heterogeneity [47]. Publication bias was assessed by funnel plots, where the standard error was plotted on the vertical axis with a reversed scale, placing larger studies towards the top, and effect estimates were given on the horizontal axis [48]. In addition, Egger tests, in which the standardized effect sizes were regressed on their standard error, were applied. Within this test, the regression intercept is expected to be zero in the absence of publication bias. p values lower than 0.05 were considered statistically significant.

3. Results

3.1. Study Selection and Characteristics

Literature research was performed in March 2023 in Web of Science and PubMed (MEDLINE) databases and identified 844 potentially relevant articles (Web of Science n = 358; PubMed n = 486). In addition, 30 articles were included after reviewing the references of identified articles. After removing duplicates, articles were assessed for more detailed evaluation. Following the selection process, 173 articles were assessed for eligibility. After reading full texts, 151 articles were removed, and ten studies fulfilled the protocol criteria (TN and CE conditions). Seven studies provided data on at least one of the selected metabolic parameters and were included in the meta-analysis. The PRISMA flow chart is shown in Figure 1.

3.2. Systematic Review and Meta-Analysis

Based on the predefined criteria, seven studies were included in this meta-analysis [38,39,40,41,42,43,44]. Details regarding these studies and the extracted raw data are presented in Table 1 and Table 2.
Plasma glucose and TG values were converted into mmol/L, insulin values were converted into pmol/L, and plasma FFA values were converted into µmol/L.

3.3. Exclusion of Parameters after Conversion

Wijers and colleagues presented the insulin concentration in nM [39], and converting it into pmol/L (by multiplying by 1000) resulted in unphysiologically high values. Thus, these values were not considered for the meta-analysis. Also, the conversion of insulin values in the study of Vosselman et al. revealed unphysiological concentrations, and they were not included in the meta-analysis [40].

3.4. Characteristics of Participants

A pooled sample size of n = 85 healthy adults was included in the analysis. The sample size of respective studies ranged from 7 to 18 participants. The mean age of participants across the studies was 31.8 ± 11.0 years, with a mean BMI of 24.9 ± 3.4 kg/m2. The mean temperature under the TN condition was 24.0 ± 1.9 °C, and under the CE condition 17.1 ± 1.1 °C, respectively (Table 1). All studies reported a non-shivering observation using hourly questionnaires, personalized cooling, self-reporting, or electromyography (EMG). The experimental visits were performed after an overnight fast.

3.5. Meta-Analytic Results

The average data of glucose, insulin, TG, and FFA concentrations obtained in enrolled studies are provided in Table 2.

3.5.1. Effects of CE on Plasma Glucose

Four studies are included (k = 4) with a pooled sample size of n = 38. For all studies, the 95% CI intersects the line of no effect size. The pooled effect size of glucose was 0.19 (SMD), and results indicated no effect of CE in comparison to TN (95% CI = [−0.27, 0.65], p = 0.41). The test of heterogeneity was not significant (Q = 2.8, p = 0.43, I2 = 2.3%) (Figure 2A). Visual inspection of the funnel plot and the result of the Egger intercept (z = 0.59, p = 0.56) suggest the absence of publication bias (Figure 2B).

3.5.2. Effects of CE on Plasma Insulin

Four studies are included (k = 4) assessing the effects of CE as compared to TN on fasting plasma insulin concentrations, with a pooled sample size of n = 50. The pooled effect size of insulin concentration was 0.39 (SMD), and results indicated no significant changes in insulin concentration post-CE compared to TN (95% CI = [−0.44, 0.12], p = 0.36). Q and I2 tests indicated the presence of statistical heterogeneity (Q = 11.53, p = 0.009; I2 = 74.8%) (Figure 3A). No publication bias existed by visual inspection of the funnel plot and interpretation of the Egger intercept (z = 0.39, p = 0.69) (Figure 3B).

3.5.3. Effects of CE on Plasma Triglyceride

Four studies are included (k = 4) with a pooled sample size of n = 45. The pooled effect size was −0.17 (SMD), and the results indicated no differences between TN and CE (95% CI = [−0.58, 0.25], p = 0.43). Heterogeneity was not evident according to Q and I2 statistics (Q = 3, p = 0.68, I2 = 0.0%) (Figure 4A). Visual inspection of the funnel plot and the Egger intercept (z = −1.09, p = 0.28) negates a publication bias (Figure 4B).

3.5.4. Effects of CE on Plasma Free Fatty Acids

Three studies (k = 3) with a pooled sample size of n = 38 reported data on plasma FFA concentration. The pooled effect size was −0.75 (SMD), and the results indicated a significant increase in FFA concentration post-CE compared to TN (95% CI = [−1.21, −0.28], p = 0.002). Further analysis revealed no heterogeneity (Q = 2, p = 0.67, I2 = 0.0%) (Figure 5A) as well as no publication bias by visual inspection of the funnel plot and by the Egger intercept (z = −0.39, p = 0.70) (Figure 5B).

4. Discussion

The current systematic review and meta-analysis aimed to analyze the data of glucose and lipid parameters and analyze the effect of mild cold only at the fasting level. Prior research has established that mild CE improves glucose and lipid metabolism in humans and rodents. Generally, a high volume of active BAT is associated with lower glycated hemoglobin (HbA1c), and improved diabetes status [31]. In this scenario, an increased expression of glucose transporter type 4 in brown adipocytes post-CE was reported in a mouse model, which was associated with increased glucose clearance [49,50]. Comparable effects of improved insulin sensitivity in humans have been confirmed.
Interestingly, one of the main findings of the present meta-analysis showed that basal glucose concentration was not changed by mild CE as compared to TN. However, subsequent animal and human studies reported improvements in glucose homeostasis parameters, such as improved insulin sensitivity and increased glucose infusion by performing the Botnia Clamp technique after short- and long-term CE in both healthy individuals and patients with T2D [51]. However, this did not lead to any significant changes in fasting plasma glucose. The Botnia Clamp method enables the detection of a more comprehensive picture of glucose metabolism [34,52]. In a large study including 260 healthy and lean participants, after 6–12 h of fasting, they underwent 2 h of mild CE at 19 °C, which did not change fasting plasma glucose and insulin levels compared to TN (p = 0.07 and p = 0.47, respectively). However, in the same study, HbA1c showed a negative association with BAT volume [53]. Similarly, no glucose and insulin level differences were observed between CE and TN after 4 h of fasting [54]. In mice, 14 days of CE improved glucose tolerance and insulin sensitivity but not basal glucose levels [55]. Additionally, CE combined with a high-fat diet for 6 weeks improved insulin sensitivity and decreased plasma glucose levels during an oral glucose tolerance test (OGTT) in mice, whereas the fasting blood glucose level again remained unchanged [56]. Lebbe et al. reported an increased plasma glucose level in cold-housed mice compared to the mice housed at room temperature; however, the BAT glucose uptake increased after cold exposure [57]. Taken together, these findings emphasize that fasting glucose levels may not capture the nuanced changes in glucose homeostasis induced by CE, thereby necessitating the use of gold-standard methods to assess its true impact on metabolism. Additionally, different results between human and rodent studies can be explained by the physiological differences and applied cooling protocols.
Multiple different variables can impact BAT activity. For instance, BMI negatively correlates with BAT volume and activity after CE [58]. However, it could be shown that CE, compared to TN, did not improve fasting glucose levels in healthy young subjects with either normal weight, overweight, or obesity [59]. This systematic review incorporated studies with a varied BMI range, encompassing participants categorized as lean (n = 4) and overweight (n = 3) and with an overall mean BMI of 24 kg/m2. According to existing literature, this mean BMI suggests the inclusion of participants exhibiting active BAT volume.
In contrast, a study, after dividing the cohort based on the presence of BAT into positive and negative subgroups, could show a decreased fasting glucose plasma level after CE compared to TN in only the BAT+ subgroup [60]. The research strategy of this work was restricted to only healthy individuals who were exposed to almost the same cooling temperature irrespective of their anthropometric or environmental parameters, which may affect the changes in fasting glucose levels. Furthermore, assessment of only fasting glucose concentration might not provide a reliable indicator of changes in glucose metabolism, i.e., due to cold [61,62].
At the molecular level, activation of the sympathetic nervous system and epinephrine secretion due to CE can inhibit pancreatic insulin secretion. Studies in rats have shown decreased insulin secretion post-CE [63,64,65]. In contrast, inhibition of the sympathetic and activation of the parasympathetic nervous systems in rats stimulate β-cell proliferation [66]. In cold-induced BAT activity, enhancing the insulin receptor signaling pathway appears to be a crucial mechanism through which BAT optimizes its glucose uptake efficiency after 4 h at 4 °C [63]. In contrast to animal studies, this meta-analysis found no changes in fasting insulin levels after CE. This is supported by the study from Iwen and colleagues that did not report any changes in peripheral fasting insulin secretion or first-phase insulin response post-CE compared to TN. However, whole-body insulin sensitivity was improved [34]. Generally, insulin concentration in serum is a sensitive parameter, which can be easily affected by others, such as meals before the measurement. Remarkably, the liver extracts 50% of postprandially secreted insulin in healthy individuals before entering the systemic circulation.
Additionally, serum insulin is an unstable molecule with a short half-life. Therefore, in the clinic, C-peptide is often a more reliable alternative to insulin for assessing β-cell activity and diagnosing diabetes status since C-peptide is generally found in the bloodstream in approximately equal proportions to insulin [67]. As previously mentioned, to ensure a more homogeneous cohort concerning BAT prevalence and activity, it is possible to divide the cohort into BAT+ and BAT− subgroups based on cold-induced glucose uptake in the supraclavicular region assessed by PET/CT. Crandall et al. showed a decrease in insulin levels, specifically in the BAT+ group, but not in the BAT− group after 2 h of CE. Another study that compared BAT+ and BAT− individuals following 2 h of CE observed a significant reduction in insulin levels in the BAT+ subgroup [68]. These data point to the necessity to take a closer look at individual BAT activity to elucidate the effects of CE on metabolism in humans.
Regarding the impact of CE on FFAs, findings supported the observation of elevated FFA levels following CE compared to TN. This increase in FFAs was consistent across all studies included in this meta-analysis. Brown adipocytes utilize fatty acids released from white adipose tissue for adaptive thermogenesis [32]. Furthermore, an association exists between BAT volume and elevated lipolysis in mice due to activation with β-ARs agonist [69], which stimulates lipolysis and increases FFA levels. By comparing individuals with detectable vs. undetectable BAT, it has been shown that the BAT+ group exhibits higher rates of FFA oxidation during CE than the BAT− group [70].
In hyperlipidemic Apoa5-/- mice, it was demonstrated that BAT, only after 4 and later 24 h of CE at 4 °C compared to mice housed at room temperature, effectively extracts a substantial amount of TG from circulation due to the high oxidative capacity of activated BAT [32], which underpins the role of TG as fuel for NST due to CE. However, human studies did not find a reduction in circulatory TG post-CE. Indirect evidence from human studies reported that intracellular TG is the primary fuel of BAT thermogenesis [71]. In rats, following both acute and chronic CE at the temperature of 10 °C for 6 h and 21 days, respectively, the inhibition of TG using nicotinic acid (NiAc) results in reduced iBAT oxidative metabolism, manifested by decreased glucose and lipid uptakes [57]. Inhibition of TG lipolysis by NiAc application suppressed cold-induced thermogenesis in healthy humans [72]. Experimental studies in humans have reported either no changes or an increase in TG levels during CE [35,36,37,72,73]. In contrast, Iwen and colleagues showed a significantly decreased TG level after 2 h of acute mild CE compared to TN [34]. In a study comparing BAT+ and BAT− groups after an acute CE before PET/CET imaging, no differences in TG concentration were observed between the two subgroups.
When evaluating BAT activity, PET/CT counts as a gold standard. However, its limitations are due to high radiation exposure and its invasive nature. For infants and children, infrared thermography offers an alternative, non-invasive, and indirect method, providing initial insights into the presence of thermogenic tissue in the supraclavicular region. Nevertheless, its application in this demographic is restricted by the requirement for individuals to remain calm and motionless during the assessment.
From a conceptual point of view, the application of an oral fat tolerance test (OFTT) has been known to assess intestinal fat metabolism in rodents and humans by consumption of high-fat meals [74,75], which could provide a more comprehensive insight into the hypothesized effects of CE on lipid metabolism.

Limitations

The main limitation of the present systematic review and meta-analysis was the inclusion of both RCT and non-RCT studies. This decision was made based on the limited number of human studies that met all the predetermined inclusion and exclusion criteria.
Studying the metabolic effects of BAT activity in humans is complex due to their susceptibility to different variables such as dietary intake and physical activity days before measurements, the metabolic health of participants, and BAT activation and volume.
Furthermore, although gender is an important factor in any metabolic analysis such as BAT activity [76], the studies incorporated in this systematic review predominantly featured male participants, which counts as one of the limitations of this systematic review.
Moreover, age constitutes another parameter influencing metabolic outcomes, including a negative correlation with BAT activity [76]. Within this systematic review, five out of seven studies involved participants aged between 20–30 years, while only two studies included individuals with a mean age of 47 years old. Due to the limited sample size, we refrained from dividing the included studies based on age for our analysis.
Moreover, the cooling protocols employed across various studies stand out as the pivotal determinant influencing outcomes concerning the impact of CE on BAT activity [77]. Therefore, in this systematic review, we defined rigorous criteria based on cooling temperature for the study selection, which led to a very small number of included studies. However, it is important to mention that the duration of the applied protocols varied between included studies. In future studies, a standardized cooling protocol, addressing both temperature and duration, will be very helpful for comparing the metabolic impact and BAT activity in human studies.
Additionally, dividing cohorts into BAT+ and BAT− subgroups would be a promising approach to better investigate the effects of CE on metabolism with more nuance.
Nonetheless, concentrating on these two subgroups would have further reduced the already limited pool size. Hence, it was not incorporated into this meta-analysis.

5. Conclusions

This meta-analysis evaluated the CE-induced changes in metabolic parameters under fasting conditions in humans. Taken together, we were able to establish that increased fasting FFAs are likely the predominant substrate fueling BAT oxidative metabolism in humans and could serve as an indicator of BAT activation at the basal level due to CE. FFA concentration was elevated after mild CE in all included studies, which indicates that the body undergoes increased lipolysis and releases more FFAs as an energy source for BAT to maintain the body temperature due to NST. By analyzing the additional parameters, TG, glucose, and insulin, no significant changes at the fasting level have been observed after mild CE compared to TN. Regarding glucose homeostasis, relying only on fasting glucose and insulin levels may not provide sufficient insight into the effect of CE on BAT activity, although the included studies have already provided evidence of activated BAT as increased energy expenditure post-cold exposure. Therefore, for a more comprehensive understanding of the beneficial effects of CE on glucose homeostasis, gold-standard methods such as the Botnia Clamp will be very important, which is the gold standard to assess beta cell function and insulin sensitivity. Considering lipid homeostasis, an OFTT has been known to determine intestinal fat metabolism in rodents and humans by consumption of high-fat meals [75,78], and monitoring blood TG levels could count as a complementary method to the monitoring of fasting FFA and TG.
Generally, in clinical practice, the assessment of BAT volume in the early stage of obesity is important. It serves as an early indicator, signaling the potential risk of metabolic syndrome associated with obesity in the future. Additionally, dividing cohorts into BAT+ and BAT− subgroups would be a promising approach to an in-depth assessment of the metabolic effects of CE in humans.

Author Contributions

Writing—original draft preparation, methodology, S.T.; writing—review and editing and data curation, R.C. and B.W.; conceptualization, funding acquisition, and supervision, S.M.M. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the grant of Deutsche Forschungsgemeinschaft (DFG) to the Graduiertenkolleg (GRK) 1957 “Adipocyte-Brain Crosstalk”. R.C. was supported by the Vicerrectoría de Investigación y Desarrollo (VID), Universidad de Chile, Grant/Award Number: 0278/2022. R.C. thanks the Chilean National Research and Development Agency (Fondecyt Initiation Project #11230075).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Dataset available on request from the authors.

Acknowledgments

We would like to express our gratitude to all groups, that provided us with the raw data for the meta-analysis.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. WHO. WHO European Regional Obesity: Report 2022; World Health Organization, Regional Office for Europe: Copenhagen, Denmark, 2022. [Google Scholar]
  2. Contreras, R.E.; Schriever, S.C.; Pfluger, P.T. Physiological and Epigenetic Features of Yoyo Dieting and Weight Control. Front. Genet. 2019, 10, 1015. [Google Scholar] [CrossRef] [PubMed]
  3. Sumithran, P.; Delbridge, E.; Kriketos, A. Long-Term Persistence of Hormonal Adaptations to Weight Loss. N. Engl. J. Med. 2011, 365, 1597–1604. [Google Scholar] [CrossRef] [PubMed]
  4. van Baak, M.A.; Mariman, E.C.M. Dietary Strategies for Weight Loss Maintenance. Nutrients 2019, 11, 1916. [Google Scholar] [CrossRef] [PubMed]
  5. Jaroslawska, J.; Gospodarska, E.; Korytko, A. Increasing energy expenditure through exercise and low ambient temperature offers oxidative protection to the hypothalamus after high-fat feeding to mice. J. Neuroendocrinol. 2022, 34, e13095. [Google Scholar] [CrossRef] [PubMed]
  6. Kaikkonen, K.M. Long-term effects on weight loss and maintenance by intensive start with diet and exercise. Scand. J. Med. Sci. Sports 2023, 33, 246–256. [Google Scholar] [CrossRef]
  7. Cypess, A.M.; Lehman, S.; Williams, G.; Tal, I.; Rodman, D.; Goldfine, A.B.; Kuo, F.C.; Palmer, E.L.; Tseng, Y.H.; Doria, A.; et al. Identification and Importance of Brown Adipose Tissue in Adult Humans. N. Engl. J. Med. 2009, 360, 1509–1517. [Google Scholar] [CrossRef]
  8. Van Marken Lichtenbelt, W.D.; Vanhommerig, J.W.; Smulders, N.M.; Drossaerts, J.M.A.F.L.; Kemerink, G.J.; Bouvy, N.D.; Schrauwen, P.; Teule, G.J.J. Cold-Activated Brown Adipose Tissue in Healthy Men. N. Engl. J. Med. 2009, 360, 1500–1508. [Google Scholar] [CrossRef] [PubMed]
  9. Virtanen, K.A.; Lidell, M.E.; Orava, J.; Heglind, M.; Westergren, R.; Niemi, T.; Taittonen, M.; Laine, J.; Savisto, N.J.; Enerbäck, S.; et al. Functional Brown Adipose Tissue in Healthy Adults. N. Engl. J. Med. 2009, 360, 1518–1525. [Google Scholar] [CrossRef]
  10. Shinde, A.B.; Song, A.; Wang, Q.A. Brown Adipose Tissue Heterogeneity, Energy Metabolism, and Beyond. Front. Endocrinol. 2021, 12, 651763. [Google Scholar] [CrossRef]
  11. Gaspar, R.C.; Pauli, J.R.; Shulman, G.I.; Muñoz, V.R. An update on brown adipose tissue biology: A discussion of recent findings. Am. J. Physiol.-Endocrinol. Metab. 2021, 320, E488–E495. [Google Scholar] [CrossRef]
  12. Yu, H.; Dilbaz, S.; Coßmann, J.; Hoang, A.C.; Diedrich, V.; Herwig, A.; Harauma, A.; Hoshi, Y.; Moriguchi, T.; Landgraf, K.; et al. Breast milk alkylglycerols sustain beige adipocytes through adipose tissue macrophages. J. Clin. Investig. 2019, 129, 2485–2499. [Google Scholar] [CrossRef] [PubMed]
  13. Hoang, A.C.; Sasi-Szabó, L.; Pál, T.; Szabó, T.; Diedrich, V.; Herwig, A.; Landgraf, K.; Körner, A.; Röszer, T. Mitochondrial RNA stimulates beige adipocyte development in young mice. Nat. Metab. 2022, 4, 1684–1696. [Google Scholar] [CrossRef] [PubMed]
  14. Gyurina, K.; Yarmak, M.; Sasi-Szabó, L.; Molnár, S.; Méhes, G.; Röszer, T. Loss of Uncoupling Protein 1 Expression in the Subcutaneous Adipose Tissue Predicts Childhood Obesity. Int. J. Mol. Sci. 2023, 24, 16706. [Google Scholar] [CrossRef] [PubMed]
  15. Pena-Leon, V.; Folgueira, C.; Barja-Fernández, S.; Pérez-Lois, R.; Lima, N.D.S.; Martin, M.; Heras, V.; Martinez-Martinez, S.; Valero, P.; Iglesias, C.; et al. Prolonged breastfeeding protects from obesity by hypothalamic action of hepatic FGF21. Nat. Metab. 2022, 4, 901–917. [Google Scholar] [CrossRef]
  16. Tsukada, A.; Takata, K.; Aikawa, J.; Iwase, D.; Mukai, M.; Uekusa, Y.; Metoki, Y.; Inoue, G.; Miyagi, M.; Takaso, M.; et al. White adipose tissue undergoes browning during preweaning period in association with microbiota formation in mice. iScience 2023, 26, 107239. [Google Scholar] [CrossRef]
  17. Sacks, H.; Symonds, M.E. Anatomical Locations of Human Brown Adipose Tissue. Diabetes 2013, 62, 1783–1790. [Google Scholar] [CrossRef] [PubMed]
  18. Cheng, L.; Wang, J.; Dai, H.; Duan, Y.; An, Y.; Shi, L.; Lv, Y.; Li, H.; Wang, C.; Ma, Q.; et al. Brown and beige adipose tissue: A novel therapeutic strategy for obesity and type 2 diabetes mellitus. Adipocyte 2021, 10, 48–65. [Google Scholar] [CrossRef]
  19. Xiang, A.S.; Meikle, P.J.; Carey, A.L.; Kingwell, B.A. Brown adipose tissue and lipid metabolism: New strategies for identification of activators and biomarkers with clinical potential. Pharmacol. Ther. 2018, 192, 141–149. [Google Scholar] [CrossRef]
  20. Carpentier, A.C.; Blondin, D.P.; Virtanen, K.A.; Richard, D.; Haman, F.; Turcotte, É.E. Brown Adipose Tissue Energy Metabolism in Humans. Front. Endocrinol. 2018, 9, 447. [Google Scholar] [CrossRef]
  21. Yoneshiro, T.; Aita, S.; Matsushita, M.; Kayahara, T.; Kameya, T.; Kawai, Y.; Iwanaga, T.; Saito, M. Recruited brown adipose tissue as an antiobesity agent in humans. J. Clin. Investig. 2013, 123, 3404–3408. [Google Scholar] [CrossRef]
  22. Hoffmann, M.E.; Rodriguez, S.M.; Zeiss, D.M.; Wachsberg, K.N.; Kushner, R.F.; Landsberg, L.; Linsenmeier, R.A. 24-h core temperature in obese and lean men and women. Obesity 2012, 20, 1585–1590. [Google Scholar] [CrossRef]
  23. O’Mara, A.E.; Johnson, J.W.; Linderman, J.D.; Brychta, R.J.; McGehee, S.; Fletcher, L.A.; Fink, Y.A.; Kapuria, D.; Cassimatis, T.M.; Kelsey, N.; et al. Chronic mirabegron treatment increases human brown fat, HDL cholesterol, and insulin sensitivity. J. Clin. Investig. 2020, 130, 2209–2219. [Google Scholar] [CrossRef] [PubMed]
  24. Vosselman, M.J.; Brans, B.; van der Lans, A.A.; Wierts, R.; A van Baak, M.; Mottaghy, F.M.; Schrauwen, P.; Lichtenbelt, W.D.v.M. Brown adipose tissue activity after a high-calorie meal in humans. Am. J. Clin. Nutr. 2013, 98, 57–64. [Google Scholar] [CrossRef] [PubMed]
  25. Yoneshiro, T.; Matsushita, M.; Hibi, M.; Tone, H.; Takeshita, M.; Yasunaga, K.; Katsuragi, Y.; Kameya, T.; Sugie, H.; Saito, M. Tea catechin and caffeine activate brown adipose tissue and increase cold-induced thermogenic capacity in humans. Am. J. Clin. Nutr. 2017, 105, 873–881. [Google Scholar] [CrossRef] [PubMed]
  26. Cypess, A.M.; Chen, Y.-C.; Sze, C.; Wang, K.; English, J.; Chan, O.; Holman, A.R.; Tal, I.; Palmer, M.R.; Kolodny, G.M.; et al. Cold but not sympathomimetics activates human brown adipose tissue in vivo. Proc. Natl. Acad. Sci. USA 2012, 109, 10001–10005. [Google Scholar] [CrossRef] [PubMed]
  27. Townsend, K.; Tseng, Y.H. Brown adipose tissue: Recent insights into development, metabolic function and therapeutic potential. Adipocyte 2012, 1, 13–24. [Google Scholar] [CrossRef] [PubMed]
  28. Ricquier, D. UCP1, the mitochondrial uncoupling protein of brown adipocyte: A personal contribution and a historical perspective. Biochimie 2017, 134, 3–8. [Google Scholar] [CrossRef] [PubMed]
  29. Hepler, C.; Shao, M.; Xia, J.Y.; Ghaben, A.L.; Pearson, M.J.; Vishvanath, L.; Sharma, A.X.; Morley, T.S.; Holland, W.L.; Gupta, R.K. Directing visceral white adipocyte precursors to a thermogenic adipocyte fate improves insulin sensitivity in obese mice. Elife 2017, 6, e27669. [Google Scholar] [CrossRef] [PubMed]
  30. McKie, G.L.; Shamshoum, H.; Hunt, K.L.; Thorpe, H.H.A.; Dibe, H.A.; Khokhar, J.Y.; Doucette, C.A.; Wright, D.C. Intermittent cold exposure improves glucose homeostasis despite exacerbating diet-induced obesity in mice housed at thermoneutrality. J. Physiol. 2022, 600, 829–845. [Google Scholar] [CrossRef]
  31. Zhu, P.; Zhang, Z.; Huang, X.; Shi, Y.; Khandekar, N.; Yang, H.; Liang, S.; Song, Z.; Lin, S. Cold exposure promotes obesity and impairs glucose homeostasis in mice subjected to a high-fat diet. Mol. Med. Rep. 2018, 18, 3923–3931. [Google Scholar] [CrossRef]
  32. Bartelt, A.; Bruns, O.T.; Reimer, R.; Hohenberg, H.; Ittrich, H.; Peldschus, K.; Kaul, M.G.; I Tromsdorf, U.; Weller, H.; Waurisch, C.; et al. Brown adipose tissue activity controls triglyceride clearance. Nat Med. 2011, 17, 200–205. [Google Scholar] [CrossRef] [PubMed]
  33. Blondin, D.P.; Labbé, S.M.; Tingelstad, H.C.; Noll, C.; Kunach, M.; Phoenix, S.; Guérin, B.; Turcotte, É.E.; Carpentier, A.C.; Richard, D.; et al. Increased Brown Adipose Tissue Oxidative Capacity in Cold-Acclimated Humans. J. Clin. Endocrinol. Metab. 2014, 99, E438–E446. [Google Scholar] [CrossRef] [PubMed]
  34. Iwen, K.A.; Backhaus, J.; Cassens, M.; Waltl, M.; Hedesan, O.C.; Merkel, M.; Heeren, J.; Sina, C.; Rademacher, L.; Windjäger, A.; et al. Cold-Induced Brown Adipose Tissue Activity Alters Plasma Fatty Acids and Improves Glucose Metabolism in Men. J. Clin. Endocrinol. Metab. 2017, 102, 4226–4234. [Google Scholar] [CrossRef] [PubMed]
  35. Blondin, D.P.; Labbé, S.M.; Noll, C.; Kunach, M.; Phoenix, S.; Guérin, B.; Turcotte, É.E.; Haman, F.; Richard, D.; Carpentier, A.C. Selective Impairment of Glucose but Not Fatty Acid or Oxidative Metabolism in Brown Adipose Tissue of Subjects with Type 2 Diabetes. Diabetes 2015, 64, 2388–2397. [Google Scholar] [CrossRef] [PubMed]
  36. Blondin, D.P.; Nielsen, S.; Kuipers, E.N.; Severinsen, M.C.; Jensen, V.H.; Miard, S.; Jespersen, N.Z.; Kooijman, S.; Boon, M.R.; Fortin, M.; et al. Human Brown Adipocyte Thermogenesis Is Driven by β2-AR Stimulation. Cell Metab. 2020, 32, 287–300.e7. [Google Scholar] [CrossRef] [PubMed]
  37. Ouellet, V.; Labbé, S.M.; Blondin, D.P.; Phoenix, S.; Guérin, B.; Haman, F.; Turcotte, E.E.; Richard, D.; Carpentier, A.C. Brown adipose tissue oxidative metabolism contributes to energy expenditure during acute cold exposure in humans. J. Clin. Investig. 2012, 122, 545–552. [Google Scholar] [CrossRef]
  38. Wijers, S.L.J.; Saris, W.H.M.; van Marken Lichtenbelt, W.D. Individual Thermogenic Responses to Mild Cold and Overfeeding Are Closely Related. J. Clin. Endocrinol. Metab. 2007, 92, 4299–4305. [Google Scholar] [CrossRef]
  39. Wijers, S.L.J.; Schrauwen, P.; van Baak, M.A.; Saris, W.H.M.; van Marken Lichtenbelt, W.D. β-Adrenergic Receptor Blockade Does Not Inhibit Cold-Induced Thermogenesis in Humans: Possible Involvement of Brown Adipose Tissue. J. Clin. Endocrinol. Metab. 2011, 96, E598–E605. [Google Scholar] [CrossRef]
  40. Vosselman, M.J.; van der Lans, A.A.; Brans, B.; Wierts, R.; van Baak, M.A.; Schrauwen, P.; van Marken Lichtenbelt, W. Systemic β-Adrenergic Stimulation of Thermogenesis Is Not Accompanied by Brown Adipose Tissue Activity in Humans. Diabetes 2012, 61, 3106–3113. [Google Scholar] [CrossRef]
  41. Vosselman, M.J.; Hoeks, J.; Brans, B.; Pallubinsky, H.; Nascimento, E.B.M.; Van Der Lans, A.A.J.J.; Broeders, E.P.M.; Mottaghy, F.M.; Schrauwen, P.; Van Marken Lichtenbelt, W.D. Low brown adipose tissue activity in endurance-trained compared with lean sedentary men. Int. J. Obes. 2015, 39, 1696–1702. [Google Scholar] [CrossRef]
  42. Din, M.U.; Raiko, J.; Saari, T.; Kudomi, N.; Tolvanen, T.; Oikonen, V.; Teuho, J.; Sipilä, H.T.; Savisto, N.; Parkkola, R.; et al. Human brown adipose tissue [15O]O2 PET imaging in the presence and absence of cold stimulus. Eur. J. Nucl. Med. Imaging 2016, 43, 1878–1886. [Google Scholar] [CrossRef]
  43. Chondronikola, M.; Volpi, E.; Børsheim, E.; Porter, C.; Saraf, M.K.; Annamalai, P.; Yfanti, C.; Chao, T.; Wong, D.; Shinoda, K.; et al. Brown Adipose Tissue Activation Is Linked to Distinct Systemic Effects on Lipid Metabolism in Humans. Cell Metab. 2016, 23, 1200–1206. [Google Scholar] [CrossRef]
  44. Chondronikola, M.; Porter, C.; Malagaris, I.; Nella, A.A.; Sidossis, L.S. Brown Adipose Tissue is Associated with Systemic Concentrations of Peptides Secreted from the Gastrointestinal System and Involved in Appetite Regulation. Eur. J. Endocrinol. 2017, 177, 33–40. [Google Scholar] [CrossRef]
  45. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021, 372, n71. [Google Scholar] [CrossRef]
  46. Higgins, J.P.T. Measuring inconsistency in meta-analyses. BMJ 2003, 327, 557–560. [Google Scholar] [CrossRef]
  47. Higgins, J. Cochrane Handbook for Systematic Reviews of Interventions. Cochrane Training. 2022. Available online: https://training.cochrane.org/handbook/current (accessed on 3 March 2023).
  48. Sterne, J.A.C.; Sutton, A.J.; Ioannidis, J.P.A.; Terrin, N.; Jones, D.R.; Lau, J.; Carpenter, J.; Rücker, G.; Harbord, R.M.; Schmid, C.H.; et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 2011, 343, d4002. [Google Scholar] [CrossRef]
  49. Lee, P.; Bova, R.; Schofield, L.; Bryant, W.; Dieckmann, W.; Slattery, A.; Govendir, M.A.; Emmett, L.; Greenfield, J.R. Brown Adipose Tissue Exhibits a Glucose-Responsive Thermogenic Biorhythm in Humans. Cell Metab. 2016, 23, 602–609. [Google Scholar] [CrossRef] [PubMed]
  50. Shimizu, Y.; Nikami, H.; Tsukazaki, K.; Machado, U.F.; Yano, H.; Seino, Y.; Saito, M. Increased expression of glucose transporter GLUT-4 in brown adipose tissue of fasted rats after cold exposure. Am. J. Physiol.-Endocrinol. Metab. 1993, 264, E890–E895. [Google Scholar] [CrossRef] [PubMed]
  51. Hanssen, M.J.; Hoeks, J.; Brans, B.; van der Lans, A.A.; Schaart, G.; van den Driessche, J.J.; Jörgensen, J.A.; Boekschoten, M.V.; Hesselink, M.K.; Havekes, B.; et al. Short-term cold acclimation improves insulin sensitivity in patients with type 2 diabetes mellitus. Nat. Med. 2015, 21, 863–865. [Google Scholar] [CrossRef] [PubMed]
  52. Tripathy, D.; Wessman, Y.; Gullström, M.; Tuomi, T.; Groop, L. Importance of Obtaining Independent Measures of Insulin Secretion and Insulin Sensitivity During the Same Test. Diabetes Care 2003, 26, 1395–1401. [Google Scholar] [CrossRef] [PubMed]
  53. Matsushita, M.; Yoneshiro, T.; Aita, S.; Kameya, T.; Sugie, H.; Saito, M. Impact of brown adipose tissue on body fatness and glucose metabolism in healthy humans. Int. J. Obes. 2014, 38, 812–817. [Google Scholar] [CrossRef]
  54. van Der Lans, A.A.J.J.; Hoeks, J.; Brans, B.; Vijgen, G.H.E.J.; Visser, M.G.W.; Vosselman, M.J.; Hansen, J.; Jörgensen, J.A.; Wu, J.; Mottaghy, F.M.; et al. Cold acclimation recruits human brown fat and increases nonshivering thermogenesis. J. Clin. Investig. 2013, 123, 3395–3403. [Google Scholar] [CrossRef]
  55. Wang, T.Y.; 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. [Google Scholar] [CrossRef]
  56. Stanford, K.I.; Middelbeek, R.J.; Townsend, K.L.; An, D.; Nygaard, E.B.; Hitchcox, K.M.; Markan, K.R.; Nakano, K.; Hirshman, M.F.; Tseng, Y.-H.; et al. Brown adipose tissue regulates glucose homeostasis and insulin sensitivity. J. Clin. Investig. 2013, 123, 215–223. [Google Scholar] [CrossRef]
  57. Labbé, S.M.; Caron, A.; Bakan, I.; Laplante, M.; Carpentier, A.C.; Lecomte, R.; Richard, D. In vivo measurement of energy substrate contribution to cold-induced brown adipose tissue thermogenesis. FASEB J. 2015, 29, 2046–2058. [Google Scholar] [CrossRef] [PubMed]
  58. Wang, Q.; Zhang, M.; Xu, M.; Gu, W.; Xi, Y.; Qi, L.; Li, B.; Wang, W. Brown Adipose Tissue Activation Is Inversely Related to Central Obesity and Metabolic Parameters in Adult Human. Sun Q, editor. PLoS ONE 2015, 10, e0123795. [Google Scholar]
  59. Raiko, J.R.H.; Tuulari, J.J.; Saari, T.; Parkkola, R.; Savisto, N.; Nuutila, P.; Virtanen, K. Associations Between Brain Gray Matter Volumes and Adipose Tissue Metabolism in Healthy Adults. Obesity 2021, 29, 543–549. [Google Scholar] [CrossRef]
  60. Zhang, Q.; Ye, H.; Miao, Q.; Zhang, Z.; Wang, Y.; Zhu, X.; Zhang, S.; Zuo, C.; Zhang, Z.; Huang, Z.; et al. Differences in the metabolic status of healthy adults with and without active brown adipose tissue. Wien. Klin. Wochenschr. 2013, 125, 687–695. [Google Scholar] [CrossRef]
  61. Leighton, E.; Sainsbury, C.A.; Jones, G.C. A Practical Review of C-Peptide Testing in Diabetes. Diabetes Ther. 2017, 8, 475–487. [Google Scholar] [CrossRef]
  62. Maddaloni, E.; Bolli, G.B.; Frier, B.M.; Little, R.R.; Leslie, R.D.; Pozzilli, P.; Buzzetti, R. C-peptide determination in the diagnosis of type of diabetes and its management: A clinical perspective. Diabetes Obes. Metab. 2022, 24, 1912–1926. [Google Scholar] [CrossRef]
  63. Wang, X.; Wahl, R. Responses of the Insulin Signaling Pathways in the Brown Adipose Tissue of Rats following Cold Exposure. Randeva HS, editor. PLoS ONE 2014, 9, e99772. [Google Scholar]
  64. Vallerand, A.L.; Perusse, F.; Bukowiecki, L.J. Stimulatory effects of cold exposure and cold acclimation on glucose uptake in rat peripheral tissues. Am. J. Physiol.-Regul. Integr. Comp. Physiol. 1990, 259, R1043–R1049. [Google Scholar] [CrossRef]
  65. Gasparetti, A.L.; de Souza, C.T.; Pereira-da-Silva, M.; Oliveira, R.L.; Saad, M.J.; Carneiro, E.M.; Velloso, L.A. Cold Exposure Induces Tissue-Specific Modulation of the Insulin-Signalling Pathway in Rattus Norvegicus. J. Physiol. 2003, 552, 149–162. [Google Scholar] [CrossRef]
  66. Moullé, V.S.; Tremblay, C.; Castell, A.-L.; Vivot, K.; Ethier, M.; Fergusson, G.; Alquier, T.; Ghislain, J.; Poitout, V. The autonomic nervous system regulates pancreatic β-cell proliferation in adult male rats. Am. J. Physiol. Metab. 2019, 317, E234–E243. [Google Scholar] [CrossRef]
  67. Petersen, M.C.; Shulman, G.I. Mechanisms of Insulin Action and Insulin Resistance. Physiol. Rev. 2018, 98, 2133–2223. [Google Scholar] [CrossRef]
  68. Saari, T.J.; Raiko, J.; U-Din, M.; Niemi, T.; Taittonen, M.; Laine, J.; Savisto, N.; Haaparanta-Solin, M.; Nuutila, P.; Virtanen, K.A. Basal and cold-induced fatty acid uptake of human brown adipose tissue is impaired in obesity. Sci. Rep. 2020, 10, 14373. [Google Scholar] [CrossRef]
  69. Berbée, J.F.; Boon, M.R.; Khedoe, P.P.; Bartelt, A.; Schlein, C.; Worthmann, A.; Kooijman, S.; Hoeke, G.; Mol, I.M.; John, C.; et al. Brown fat activation reduces hypercholesterolaemia and protects from atherosclerosis development. Nat Commun. 2015, 6, 6356. [Google Scholar] [CrossRef]
  70. Chondronikola, M.; Volpi, E.; Børsheim, E.; Porter, C.; Annamalai, P.; Enerbäck, S.; Lidell, M.E.; Saraf, M.K.; Labbe, S.M.; Hurren, N.M.; et al. Brown adipose tissue improves whole-body glucose homeostasis and insulin sensitivity in humans. Diabetes 2014, 63, 4089–4099. [Google Scholar] [CrossRef]
  71. Khedoe, P.P.S.J.; Hoeke, G.; Kooijman, S.; Dijk, W.; Buijs, J.T.; Kersten, S.; Havekes, L.M.; Hiemstra, P.S.; Berbée, J.F.P.; Boon, M.R.; et al. Brown adipose tissue takes up plasma triglycerides mostly after lipolysis. J. Lipid Res. 2015, 56, 51–59. [Google Scholar] [CrossRef]
  72. Blondin, D.P.; Frisch, F.; Phoenix, S.; Guérin, B.; Turcotte, É.E.; Haman, F.; Richard, D.; Carpentier, A.C. Inhibition of Intracellular Triglyceride Lipolysis Suppresses Cold-Induced Brown Adipose Tissue Metabolism and Increases Shivering in Humans. Cell Metab. 2017, 25, 438–447. [Google Scholar] [CrossRef]
  73. Hoeke, G.; Nahon, K.J.; Bakker, L.E.; Norkauer, S.S.; Dinnes, D.L.; Kockx, M.; Lichtenstein, L.; Drettwan, D.; Reifel-Miller, A.; Coskun, T.; et al. Short-term cooling increases serum triglycerides and small high-density lipoprotein levels in humans. J. Clin. Lipidol. 2017, 11, 920–928.e2. [Google Scholar] [CrossRef]
  74. Borén, J.; Matikainen, N.; Adiels, M.; Taskinen, M.-R. Postprandial hypertriglyceridemia as a coronary risk factor. Clin. Chim. Acta 2014, 431, 131–142. [Google Scholar] [CrossRef]
  75. Hou, X.; Song, A.; Guan, Y.; Tian, P.; Ren, L.; Tang, Y.; Wang, C.; Gao, L.; Song, G.; Xing, X. Identification of the Chinese Population That Can Benefit Most From Postprandial Lipid Testing: Validation of the Use of Oral Fat Tolerance Testing in Clinical Practice. Front. Endocrinol. 2022, 13, 831435. [Google Scholar] [CrossRef]
  76. Pfannenberg, C.; Werner, M.K.; Ripkens, S.; Stef, I.; Deckert, A.; Schmadl, M.; Reimold, M.; Häring, H.-U.; Claussen, C.D.; Stefan, N. Impact of Age on the Relationships of Brown Adipose Tissue with Sex and Adiposity in Humans. Diabetes 2010, 59, 1789–1793. [Google Scholar] [CrossRef]
  77. Cypess, A.M.; Haft, C.R.; Laughlin, M.R.; Hu, H.H. Brown Fat in Humans: Consensus Points and Experimental Guidelines. Cell Metab. 2014, 20, 408–415. [Google Scholar] [CrossRef]
  78. Liu, L.; Hou, X.; Song, A.; Guan, Y.; Tian, P.; Wang, C.; Ren, L.; Tang, Y.; Gao, L.; Xing, X.; et al. Oral fat tolerance testing identifies abnormal pancreatic β-cell function and insulin resistance in individuals with normal glucose tolerance. J. Diabetes Investig. 2022, 13, 1805–1813. [Google Scholar] [CrossRef]
Biomedicines 12 00537 g001
Figure 1. PRISMA 2009 flow diagram of systematic review and meta-analysis.
Figure 1. PRISMA 2009 flow diagram of systematic review and meta-analysis.
Biomedicines 12 00537 g001
Biomedicines 12 00537 g002
Figure 2. (A) Forest plot represents the effect size and 95% CI for plasma glucose concentration post-CE compared to TN. (B) Funnel plot checks the existence of publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; CE: cold exposure; TN: thermoneutrality [39,40,41,42].
Figure 2. (A) Forest plot represents the effect size and 95% CI for plasma glucose concentration post-CE compared to TN. (B) Funnel plot checks the existence of publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; CE: cold exposure; TN: thermoneutrality [39,40,41,42].
Biomedicines 12 00537 g002
Biomedicines 12 00537 g003
Figure 3. (A) Forest plot represents the effect size and 95% CI for plasma insulin concentration post-CE compared to TN. (B) Funnel plot to check the publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; CE: cold exposure; TN: thermoneutrality [38,41,42,44].
Figure 3. (A) Forest plot represents the effect size and 95% CI for plasma insulin concentration post-CE compared to TN. (B) Funnel plot to check the publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; CE: cold exposure; TN: thermoneutrality [38,41,42,44].
Biomedicines 12 00537 g003
Biomedicines 12 00537 g004
Figure 4. (A) Forest plot represents the effect size and 95% CI in plasma TG concentration post-CE compared to TN. (B) Funnel plot to check the publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; TG: triglyceride; CE: cold exposure; TN: thermoneutrality [40,41,42,43].
Figure 4. (A) Forest plot represents the effect size and 95% CI in plasma TG concentration post-CE compared to TN. (B) Funnel plot to check the publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; TG: triglyceride; CE: cold exposure; TN: thermoneutrality [40,41,42,43].
Biomedicines 12 00537 g004
Biomedicines 12 00537 g005
Figure 5. (A) Forest plot represents the effect size and 95% CI in plasma FFA concentration post-CE compared to TN. (B) Funnel plot to check the publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; FFA: free fatty acid; CE: cold exposure; TN: thermoneutrality [40,41,43].
Figure 5. (A) Forest plot represents the effect size and 95% CI in plasma FFA concentration post-CE compared to TN. (B) Funnel plot to check the publication bias. RE: random effect; SMD: standardized mean difference; CI: confidence interval; FFA: free fatty acid; CE: cold exposure; TN: thermoneutrality [40,41,43].
Biomedicines 12 00537 g005
Table 1. Summary of the protocol included.
Table 1. Summary of the protocol included.
StudiesTNCE
ReferenceYearTemperature (°C)DurationTemperature (°C)DurationCooling MethodShivering Test
Wijers et al. [38]200722.036 h1636 hRespiration chamberHourly questionnaire
Wijers et al. [39]201122.036 h1648 hRespiration chamberHourly questionnaire
Vosselman et al. [40]201224.545 minPC120 minAir conditionerNon-shivering PC
Vosselman et al. [41]2015RT60 minPC120 minFlow-controlled circulating bathNon-shivering PC
Din et al. [42]201622.0120 minPC240 minAir conditionerObservation and self-report
Chondronikola et al. [43]201626.25 h18.25 hCooling blanketNon-shivering PC
Chondronikola et al. [44]201726.25 h18.25 hCooling blanketNon-shivering PC
Mean ± SD temperature -24.0 ± 1.9 °C-17.1 ± 1.1 °C---
Abbreviations: room temperature (RT); personalized cooling (PC); cold exposure (CE); thermoneutrality (TN).
Table 2. Characteristics of studies.
Table 2. Characteristics of studies.
StudiesParticipants’ CharacteristicsTNCE
ReferenceYearDesignnMale/
Female		Age (yrs.)BMI
(kg/m2)		Glucose
(mmol/L)		Insulin
(pmol/L)		TGs
(mmol/L)		FFAs
(µmol/L)		Glucose
(mmol/L)		Insulin
(pmol/L)		TGs
(mmol/L)		FFAs
(µmol/L)
Wijers
et al. [38]		2007Comparative
study		13male22.8 ± 1.722.9 ± 0.9NA101.4 ± 9.6NA-NA88.3 ± 6.2NA-
Wijers
et al. [39]		2011Non-RCT9male23.0 ± 0.822.6 ± 0.44.9± 0.1-NANA4.9 ± 0.04-NANA
Vosselman
et al. [40]		2012RCT10male22.5 ± 2.521.6 ± 1.64.9 ± 0.4-0.7 ± 0.2324 ± 844.7 ± 0.4-0.7 ± 0.2637 ± 398
Vosselman
et al. [41]		2015Comparative
study		12male23.0 ± 3.321.8 ± 1.95.0 ± 0.366.5 ± 35.70.9 ± 0.6567 ± 2175.1 ± 0.485.4 ± 44.10.9 ± 0.5671 ± 196
Din et al. [42]2016RCT75/236.0 ± 11.025.5 ± 3.35.2 ± 0.349.7 ± 56.70.8 ± 0.3NA5.2 ± 0.233.6 ± 19.61.1 ± 0.4NA
Chondronikola
et al. [43]		2016Crossover,
non-RCT		16male47.8 ± 16.030.3 ± 2.1NANA1.5 ± 0.9430 ± 220NANA1.6 ± 1.1630 ± 280
Chondronikola
et al. [44]		2017Crossover,
non-RCT		18male47.6 ± 17.829.7 ± 4.9NA36.6 ± 27.3NANANA32.9 ± 19.6NANA
Mean ± SD----31.8 ± 11.024.9 ± 3.4--------
DATA are mean ± SD. (-): unconverted values: after converting these data from the presented units in the papers to the chosen unit in this review, these values were not in the biological range. Therefore, they were excluded. For more details, see Section 3.3. Abbreviations: not assessed in respective studies (NA); randomized controlled study (RCT); body mass index (BMI); triglycerides (TGs); free fatty acids (FFAs); cold exposure (CE); thermoneutrality (TN).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Tabei, S.; Chamorro, R.; Meyhöfer, S.M.; Wilms, B. Metabolic Effects of Brown Adipose Tissue Activity Due to Cold Exposure in Humans: A Systematic Review and Meta-Analysis of RCTs and Non-RCTs. Biomedicines 2024, 12, 537. https://doi.org/10.3390/biomedicines12030537

AMA Style

Tabei S, Chamorro R, Meyhöfer SM, Wilms B. Metabolic Effects of Brown Adipose Tissue Activity Due to Cold Exposure in Humans: A Systematic Review and Meta-Analysis of RCTs and Non-RCTs. Biomedicines. 2024; 12(3):537. https://doi.org/10.3390/biomedicines12030537

Chicago/Turabian Style

Tabei, Shirin, Rodrigo Chamorro, Sebastian M. Meyhöfer, and Britta Wilms. 2024. "Metabolic Effects of Brown Adipose Tissue Activity Due to Cold Exposure in Humans: A Systematic Review and Meta-Analysis of RCTs and Non-RCTs" Biomedicines 12, no. 3: 537. https://doi.org/10.3390/biomedicines12030537

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop