Liam P. Kilduff, and
Show fewer authors
04 Apr 2025https://doi.org/10.1152/ajpregu.00186.2024
INTRODUCTION
Adult humans rely on eccrine sweat production to facilitate evaporative cooling and maintain thermal balance, particularly in hot and/or humid environments [high wet-bulb globe temperature (WBGT) (1, 2)]. In hot conditions, evaporation typically represents the primary heat transfer avenue (2), which offsets heat production at rest and during exercise or occupational work (3, 4). If heat is not sufficiently dissipated from the body to the environment, positive heat storage ensues, leading to rises in core temperature (Tcore; uncompensable heat stress) (5). If left uncorrected, heat strain can cause heat exhaustion, heat stroke, and even death in extreme scenarios (6, 7). Thus, understanding factors that control thermal sweating is of great importance for the health and well-being of many people.
Thermoregulatory capacity is largely determined by three primary modifiable factors: metabolic heat production, vasodilation (i.e., dry heat loss) and sweating [i.e., evaporative heat loss; (1, 8, 9)]. Consequently, the ability to activate thermoregulatory defenses (i.e., thermoeffectors) and to tolerate exposure to hot environmental conditions can be improved, with sweating being the primary manipulable pathway (2, 10, 11). For example, endurance training and heat acclimation regimes are capable of lowering resting Tcore, and the oxygen cost of exercise at a given intensity, but are notable in their capacity to accelerate sweating onset and increase sweat rate, plasma volume and skin blood flow (SkBF) (10–14). In various ways, these physiological adaptations augment avenues of heat transfer, control heat production and, ultimately, aid in maintaining thermal equilibrium during heat exposure. Given the importance of thermal sweating in achieving this, further understanding of the capacity for adaptation in sweating variables in response to various interventions is required.
More recently, the notion that dietary supplementation may offer thermoregulatory benefits or, alternatively, heighten the risk of heat illness when ingested in hot conditions has been considered (15–17). There are a number of motivations for individuals to consider dietary supplementation, such as ensuring adequate intake of certain nutrients, improving health, or supporting specific physiological functions (18, 19). Approximately 50% of US adults (20) and between 15 and 41% of UK adults (21) report dietary supplement use, with only a quarter of users taking supplements that have been recommended by a healthcare professional (22). Although such dietary supplements are not commonly consumed for the purpose of influencing thermoregulation, they may inadvertently affect it (16). As the popularity of dietary supplements continues to rise, in a world that is likely to experience more frequent, prolonged, and intense heatwaves (23), research is needed to better understand the potential thermoregulatory effects upon human health and performance. For example, the amino acid taurine, often ingested for its antioxidative and antihypertensive effects (24, 25), has more recently been reported to increase sweating rate/loss (a key modifiable heat dissipation pathway) by ∼13% (26) and 27% (27) as well as reducing Tcore compared with placebo in the heat (26, 28). Furthermore, another dietary supplement, creatine, is not typically considered to help in offsetting hyperthermia, but it is commonly taken to improve high-intensity exercise performance (29). However, a review of its thermoregulatory effects highlighted that supplementation may be beneficial during exercise in high ambient temperatures due to its effects on fluid balance (17). In addition, a recent meta-analysis established that preexercise hyperhydration with glycerol and/or creatine supplementation decreased the rate of rise in Tcore after constant work exercise in both thermoneutral and hot conditions, compared with placebo (15).
Other commonly used supplements, such as dietary nitrate (NO−3NO3−), which has a key role in blood pressure regulation and endurance exercise enhancement (30, 31), does not appear to maintain all of these effects when humans are exposed to the heat (16). This is surprising, as there is a plausible mechanistic basis for thermoregulatory enhancement following ingestion of dietary nitrate and l-arginine, as both are known to improve NO (nitric oxide) bioavailability (32, 33). Specifically, NO bioavailability could have direct and indirect effects on eccrine sweat gland and microvascular function (34, 35). Indeed, other supplements, such as anti-oxidants (i.e., polyphenols), may support thermoregulation through protection of NO against oxidative destruction, thereby improving its bioavailability (36) and enhancing or preserving peripheral vasodilation. However, given that body fluid loss, and secondary hypovolemia, is accelerated in the heat (37), the reported reductions in blood pressure following supplementation with NO donors (38) could increase the risk of acute hypotension, particularly in the postexercising state (39).
Branched-chain amino acids (BCAAs) have been extensively researched for their potential ergogenic role among athletes (40), yet have several health-related applications (41) and can be supplemented to account for age-related decline in lean muscle mass (42). Although less commonly supplemented for such reasons among the general population (43), BCAAs and other amino acids have a wide variety of biological roles. For example, tyrosine is used to enhance cognitive function (44, 45) and BCAAs have been reported to alleviate skeletal muscle damage and soreness following exhaustive and resistance exercise (46, 47). Given that both tyrosine and BCAAs may compete for the same blood-brain-barrier transporters, coupled with their wider roles in neurotransmitter biosynthesis pathways (48–50), sufficient balance of both supplements may be important during heat exposure. In a previous meta-analysis, the use of orally administered tyrosine or BCAAs (used separately), were capable of enhancing endurance exercise performance in the heat, but there was no effect on submaximal or maximal Tcore responses (16), thereby questioning their thermoregulatory role. Although many of the above-mentioned supplements are used more modestly across the population (43, 51), caffeine features in the daily intake of ∼80–85% of people globally (52, 53) and is a prominent dietary supplement among athletes (54). However, caffeine has been reported to increase Tcore when ingested before or during exercise in the heat (16), but its effects in the resting state have not been evaluated meta-analytically. Given the high prevalence of caffeine consumption, mixed with the understanding of its cardiometabolic side-effects (55, 56), this perhaps places one of the greatest risks to the general population when consumed in the heat. Collectively, it is apparent that supplementing the diet with some substances, could have implications for thermoregulatory capacity, and further research is required to understand the consistency and magnitude of effects reported across the empirical literature.
Based on the evidence, to date, a systematic evaluation of the effect of all dietary supplements on the primary modifiable thermoregulatory process of sweating, and subsequent Tcore responses, is warranted. This is necessary to provide clarity on the magnitude and consistency of the effect of supplements on thermal balance during rest and exercise. This impact has not previously been fully considered, and there remains limited official guidance on dietary supplement intake for those exposed to thermally stressful conditions, such as athletes (57–59), and military personnel (60) or, indeed, the general public (61). Given the range of effects that different supplements appear to have on Tcore, at least in the exercising state in the heat (16), coupled with the clear lack of specific guidance on this topic, a comprehensive evaluation of the collective evidence is an important step in developing an evidence-based understanding of the benefits or risks associated with using dietary supplements in hot conditions.
The aims of the current meta-analysis were to investigate the effects of all known orally administered dietary supplements on Tcore and sweating responses in the heat. The effect of rehydration solutions, such as electrolytes, on thermal sweating have been thoroughly evaluated (37, 62, 63) and this was not replicated here; however, a number of factors were considered as moderators of Tcore and sweating responses, such as hydration status among participants in studies evaluating dietary supplementation in the heat (64, 65). Likewise, training and acclimatization/acclimation status (11), protocol (rest vs. exercise) and exercise intensity were considered to potentially impact thermoregulatory sweating and Tcore (66–68). Environmental conditions, such as WBGT (and/or heat stress index and vapor pressure) will also influence the ability to evaporatively cool (69, 70). Therefore, to evaluate the effects of dietary supplements on thermoregulation in the heat, these factors were considered as potential moderating variables and formed part of a secondary meta-regression analysis.
METHODS
Search Strategy
All of the available literature was searched and obtained according to the PRISMA guidelines, with a predetermined search strategy (71). Medical subject heading (MeSH) terms were active during the searches. There was no limit on the status, date or language of the publication. The final Boolean searches were performed in PubMed, SPORTDiscus (EBSCO) and Scopus on April 9th, 2024. The search terms used were “(dietary supplements OR dietary supplementation OR nutritional supplements OR nutritional supplementation OR supplements OR supplementation OR ergogenic OR ergogenic aids OR nutraceuticals OR amino acids OR anti-oxidants OR vitamins OR minerals OR stimulants OR herbs OR herbal) AND (heat OR temperature OR sweat OR sweating OR sweat response OR sweating response OR sudomotor OR body temperature regulation OR thermoregulation OR thermoregulatory OR heat loss OR cooling OR evaporative OR evaporation OR thermal stress OR heat stress OR hyperthermia OR hyperthermic).” As there is no a-priori list of dietary supplements that effect thermal balance, no supplements were searched individually by name. Two authors (J.P. and M.W.) verified the search terms and the accuracy of the returned results. “Other sources” were also identified, such as through social media (Twitter or “X”), the reference lists of included papers and additional database (Google Scholar) searching using various combinations of the above search terms.
Study Selection
Any duplicates were removed, and titles and abstracts were screened for inclusion by two investigators (J.P. and M.W.), in accordance with agreed inclusion criteria. The single paper retrieved which had been published in a language other than English was translated digitally using two separate translation software programs; Google Translate and DeepL Translator (DeepL GmbH, Cologne, Germany). The reference lists of the initial papers were reviewed independently by two authors (J.P. and M.W.). The remaining articles were then assessed separately (and without influence) by J.P. and M.W. against the inclusion and exclusion criteria. There was 100% agreement in study selection between the two reviewers. Papers were required to have been published in a peer-reviewed journal as original research articles with a crossover, randomized control trial, an intervention, or an independent groups design. They must also have included a control or placebo group, and participants were required to be healthy adults (≥18 yr). To be included in this analysis, the studies must have: 1) administered a dietary supplement (by the definition provided in the subsequent paragraph); 2) been conducted in an ambient dry-bulb temperature of ≥30°C or WBGT ≥ 20°C or small ranges up to those temperatures in either a laboratory or field setting. A WBGT of ≥20°C was considered to provide sufficient heat stress, even when dry-bulb temperature was <30°C (72). Of the remaining papers, 71 were removed for the reasons outlined in Fig. 1, which were primarily that they included supplements that were: a drug; not orally administered; a macronutrient or a rehydration solution (e.g., electrolytes or a supplement with a mechanism of action considered to be directly related to hydration). Other reasons were the absence of measures of Tcore and the sweating response, or environmental issues.
Figure 1.The process of study selection.
Download figureDownload PowerPoint
A dietary supplement was defined by adapting the International Olympic Committee (IOC) position statement (58) and the European Food Safety Authority statement (19) as: a nonfood, nonpharmacological, food component, nutrient, or nonfood compound that is purposefully orally ingested in addition to the habitual diet, for its nutritional or physiological effects. This may be to maintain sufficient intake of certain nutrients, correct deficiencies, or support physiological function, including thermoregulatory responses to the heat. The supplement is not being consumed for its calorific value, its effects on hydration (the mechanism of action is not through rehydration), and is not an energy drink. Ingestion of the supplement is also recognized to be legal as per the Misuse of Drugs Act 1971 (73) and is not on the World Anti-Doping Association’s prohibited substances list (74).
Data Extraction and Quality Assessment
Data were manually extracted independently by two authors (J.S.P. and M.W.) and entered into a custom-designed Microsoft Excel spreadsheet. Any discrepancies were verified by a third independent reviewer. Extracted data included: 1) characteristics of the sample (sex, age, health, training, and heat acclimation/acclimatization status); 2) study design; 3) supplement, dose, and timing of intake; 4) fluid intake before and during exercise (i.e., hydration status); 5) environmental conditions (temperature and humidity); 6) trial type (i.e., exercise type or rest and length); 7) peak Tcore (rectal, gastrointestinal, esophageal or tympanic); and 8) bias. Risk of bias was assessed independently by two authors (J.S.P. and M.W.) according to the Cochrane Collaboration guidelines (75). Where details of the study were unclear, the authors of the relevant papers were contacted for specific information or to clarify the method that was used. There was 100% agreement between the investigators concerning the outcome of this quality assurance procedure, hence, it was not considered necessary to include a third independent reviewer. There were three outcome measures for this meta-analysis: 1) Tcore reported at the end of the trial, the end of the exercising portion of the trial, or at the point of the highest thermal strain, hereafter referred to as “peak Tcore”; 2) whole body sweat rate (WBSR) across the trial or exercising portion of the trial; and 3) local sweat rate (LSR) reported at the end of the trial or at the point of the highest thermal strain.
Statistical Analysis
Data analysis was performed by one author (J.S.P.). Data were extracted from the qualifying papers in the form of a means, standard deviation (SD) and sample size (n) for the meta-analysis. Publicly available software (WebPlotDigitizer, Version 4.3) was used to extrapolate any unreported values from the figures to means and SD data. Where data were expressed as mean and standard error (SE or SE) or CI, they were converted to means and SD. Authors of the original research articles were contacted for any missing data; however, if mean data were not accessible, these articles were excluded. If SE or CI data were missing, they were imputed using the sample pooled SD from similar included studies in accordance with Cochrane guidelines (75, 76). There were 17 instances (7 in the Tcore meta-analysis and 10 in the WBSR meta-analysis) where no dispersion data (SD, SE, or CI) were provided. For selected study designs (i.e., intervention studies with prepost supplementation), the postintervention values were extracted as the outcome measures for the “supplementation condition” and the preintervention values as the “placebo or control condition” (75). For crossover trials (within-subject) or independent designs, the outcome measures for the supplementation condition were considered against the placebo or control condition. Standardized mean difference (SMD) was used to compare the results between studies using different protocols and measures. Peak Tcore outcome data were reported as peak Tcore (°C) or rate of rise (°C·h−1) of Tcore. Mean, maximum, peak, and mean body temperature were also included if peak Tcore data were not provided. Whole body sweating response outcome data were reported as WBSR (mL·min−1) and body mass change (%). Outcome data representing WBSL (i.e., body mass or sweat loss and body mass change), reported in absolute L, mL, kg, or g were converted to WBSR (mL·min−1) using trial length data, and WBSR reported in L·h−1 or mL·h−1 were directly converted to mL·min−1. LSR outcome data reported in nL·min−1 were converted to mg·cm·min−1 and reported as such.
Three meta-analyses were conducted, one for each outcome measure. These were performed in RStudio [R Core Team; (77)] and included 135, 106, and 11 comparison groups for the peak Tcore, WBSR, and LSR meta-analyses, respectively. Not all studies reported Tcore or sweating response data; hence, they were excluded from the respective analyses. All data were analyzed with a random-effects model, with heterogeneity assessed using the I2 statistic. Outliers were detected using a function in RStudio and influence on analysis was investigated. Publication bias was accounted for by funnel plots and conducting Egger’s test (78). Any adjustments to the effect sizes based on this procedure are reported in the results. Hedges’ g and 95% CIs were used to express SMD between dietary supplementation and placebo groups across studies. Subanalysis of the different dietary supplements included were conducted for all three meta-analyses. Meta-regressions were also conducted to determine the effect of candidate moderators on peak Tcore, WBSR, and LSR outcomes, as reported in each study: training status (highly trained vs. recreationally active); heat acclimation status (heat acclimated vs. nonheat acclimated); hydration status (euhydrated vs. hypohydrated); fluid ingestion during exercise (fluid ingestion vs. no fluid ingestion); duration of trial; WBGT; trial type (exercise vs. rest); supplement dose (where sufficient no. of studies); and duration of supplementation (where applicable). The thresholds for the magnitude of effects were < 0.2, 0.2, 0.5, and 0.8 for “trivial,” “small,” “medium,” and “large” effects, respectively (79). Alpha (α) was set at P ≤ 0.05 for all analyses.
RESULTS
Study Selection
The initial searches retrieved 39,383 articles, which were reduced to 37,907 after removal of duplicates. After further screening and removal of reviews, animal studies, and other irrelevant papers, 177 articles remained. Searches of social media (Twitter or “X”), additional databases, and reference lists within the 177 papers provided 32 further papers. Of the 209 articles, 52 were removed based on their incomplete compliance with the inclusion criteria and further 33 were removed due to having: no full-text available, duplicate data with another paper, or no extractable data. This left 124 papers, of which 117, 93, and 10 papers were included in the peak Tcore, WBSR, and LSR analyses, respectively (Fig. 1). Sixteen papers had more than one comparison group, therefore, one or more additional data sets were added to the analysis for each study. As these additional comparison groups shared participants, the sample size was reduced to mitigate any unit-of-analysis error, as per the Cochrane guidelines (75). Four papers also included multiple comparison groups; however, as these did not share participants, they were included without sample size adjustment. One paper was included without the addition of the duplicate peak Tcore data.
Study Characteristics
The characteristics of the 124 included studies are summarized in Table 1. The studies included a total of 1,553 participants, comprising both males and females (males 90%; both males and females 9%; unreported 1%) of varying training (highly trained 43%; recreationally active 40%; unreported 18%) and heat acclimation statuses (heat acclimated 13%; nonheated acclimated 38%; unreported 49%). One hundred and six studies were crossover designs, 12 studies were independent group designs, and 6 studies were pre-to-post interventions. Thirty-nine different types of dietary supplements or supplement combinations were included in varying doses (Table 1). These were a combination of acute doses (single day; n = 81; 65%) and chronic administration (≥2 days; n = 43; 35%). The trial types included were exercise (90%) and rest (10%). The measures of Tcore were rectal (62%), tympanic (10%), esophageal (9%), gastrointestinal (14%), oral (1%), and unreported (4%). The measures of body mass or sweat loss or sweat rate, representing WBSR were reported in L (7%), mL (7%), kg (17%) or g (3%), g·m−2·h−1 (1%), % change (11%), L·h−1 (16%), mL·h−1 (1%), mL·min−1 (8%), or were unreported (29%). Ambient dry-bulb temperature (mean 33.8°C; range 25°C–46.6°C), WBGT (mean 27.5°C; range 18.5°C–35.1°C), and RH% (mean 47%; range 12%–80%) are reported herein. There were no adverse health-related events noted in any of the studies.
Table 1. Summary of studies included in the meta-analyses (n = 124)
Meta-Analysis
The results of the peak Tcore meta-analysis (n = 135) are reported in Fig. 2. Overall, the pooled analysis of all supplements revealed that there was a trivial nonsignificant positive effect on peak Tcore compared to placebo (Hedges’ g = 0.004, 95% CI: −0.091 to 0.100, P = 0.930). The I2 statistic demonstrated 20.6% heterogeneity. The results of the WBSR (n = 106) and LSR (n = 11) meta-analyses are reported in Figs. 3 and 4, respectively. Overall, WBSR (Hedges’ g = 0.041, 95% CI: −0.095 to 0.176, P = 0.559) and LSR (Hedges’ g = 0.021, 95% CI: −0.224 to 0.266, P = 0.869) had a “trivial” nonsignificant increase with dietary supplementation compared to placebo, with 1.7% and 0% heterogeneity (I2), respectively.
Figure 2.Effect of dietary supplementation on peak core temperature. CI, confidence interval; SMD, standardized mean difference.
Download figureDownload PowerPoint
Figure 3.Effect of dietary supplementation on whole body sweat rate. CI, confidence interval; SMD, standardized mean difference.
Download figureDownload PowerPoint
Figure 4.Effect of dietary supplementation on local sweat rate. CI, confidence interval; SMD, standardized mean difference.
Download figureDownload PowerPoint
Subgroup Analysis
Peak core temperature.
Subgroup analysis demonstrated a significant overall pooled effect of the different dietary supplement categories on peak Tcore (P = 0.015). However, the following supplements all demonstrated nonsignificant trivial effects on peak Tcore: creatine (Hedges’ g = −0.12, 95% CI: −0.453 to 0.203, P = 0.456), nitrate (Hedges’ g = 0.07, 95% CI: −0.237 to 0.381, P = 0.648), l-glutamine (Hedges’ g = 0.07, 95% CI: −0.246 to 0.384, P = 0.667), bovine colostrum (Hedges’ g = 0.13, 95% CI: −0.327 to 0.588, P = 0.575), probiotics (Hedges’ g = 0.00, 95% CI: −0.653 to 0.653, P = 1.000), blackcurrant extract (Hedges’ g = −0.11, 95% CI: −0.461 to 0.453, P = 0.986), tyrosine (Hedges’ g = −0.04, 95% CI: −0.427 to 0.345, P = 0.835), BCAAs (Hedges’ g = −0.004, 95% CI: −0.327 to 0.588, P = 0.575), betaine (Hedges’ g = −0.14, 95% CI: −0.749 to 0.464, P = 0.646), vitamin C (Hedges’ g = −0.07, 95% CI: −0.891 to 0.743, P = 0.859), Eurycoma longifolia Jack (Hedges’ g = 0.00, 95% CI: −0.800 to 0.800, P = 1.000), polyphenols (Hedges’ g = 0.00, 95% CI: −0.800 to 0.800, P = 1.000), folic acid (Hedges’ g = 0.00, 95% CI: −0.924 to 0.924, P = 1.000), amino acids (Hedges’ g = 0.13, 95% CI: −0.492 to 0.753, P = 0.681), combined caffeine and taurine (Hedges’ g = 0.07, 95% CI: −1.063 to 1.200, P = 0.906), and combined creatine, glycerol, glucose, and alpha lipoic acid (Hedges’ g = 0.00, 95% CI: −0.924 to 0.924, P = 1.000).
There were a number of caffeine-based supplements that increased Tcore, with isolated caffeine (Hedges’ g = 0.44, 95% CI: 0.275–0.603, P < 0.001) having a small significant positive effect and combined caffeine and ginseng demonstrating a large significant positive effect (Hedges’ g = 1.19, 95% CI: 0.163–2.208, P = 0.023). l-arginine (Hedges’ g = 0.22, 95% CI: −0.765 to 1.203, P = 0.663), sodium bicarbonate (Hedges’ g = 0.26, 95% CI: −0.309 to 0.829, P = 0.370), β-glucan (Hedges’ g = 0.28, 95% CI: −0.217 to 0.784, P = 0.268), ginseng (Hedges’ g = 0.38, 95% CI: −0.554 to 1.316, P = 0.424), and combined alpha-ketoglutaric acid (α-KG) and 5-hydroxymethylfurfural (5-HMF; Hedges’ g = 0.35, 95% CI: −0.709 to 1.408, P = 0.518) had “small” nonsignificant positive effects. Glycerol (Hedges’ g = −0.28, 95% CI: −0.700 to 0.130, P = 0.179), sodium citrate (Hedges’ g = −0.46, 95% CI: −1.261 to 0.341, P = 0.261), GABA (Hedges’ g = −0.46, 95% CI: −1.401 to 0.480, P = 0.337), vitamin E (Hedges’ g = −0.23, 95% CI: −0.889 to 0.423, P = 0.487), curcumin (Hedges’ g = −0.28, 95% CI: −1.268 to 0.704, P = 0.575), quercetin (Hedges’ g = −0.24, 95% CI: −0.973 to 0.496, P = 0.524), menthol (Hedges’ g = −0.46, 95% CI: −1.299 to 0.380, P = 0.283), Thermo Speed Extreme (Hedges’ g = −0.23, 95% CI: −0.785 to 0.327, P = 0.420), effective microorganism X (Hedges’ g = −0.30, 95% CI: −1.441 to 0.841, P = 0.606), whey protein (Hedges’ g = −0.36, 95% CI: −1.201 to 0.486, P = 0.407), and combined creatine and glycerol (Hedges’ g = −0.47, 95% CI: −1.460 to 0.530, P = 0.359) had “small” nonsignificant negative effects.
There were some medium-to-large effects of supplementation on peak Tcore, such as oligonol (Hedges’ g = −0.50, 95% CI: −0.907 to −0.101, P = 0.014) and taurine (Hedges’ g = −0.66, 95% CI: −1.296 to 0.022, P = 0.043), which had a medium significant negative effect, and combined creatine, glycerol and glucose (Hedges’ g = −0.66, 95% CI: −2.187 to 0.873, P = 0.400) had “medium” nonsignificant effects. Catechin (Hedges’ g = −0.80, 95% CI: −1.825 to 0.235, P = 0.130) had “large” nonsignificant negative effects on peak Tcore.
Whole body sweat rate.
Subgroup analysis demonstrated a nonsignificant overall pooled effect of the different supplement categories on WBSR (P = 0.434). Caffeine (Hedges’ g = 0.04, 95% CI: −0.129 to 0.203, P = 0.660), creatine (Hedges’ g = 0.15, 95% CI: −0.208 to 0.517, P = 0.403), sodium citrate (Hedges’ g = −0.05, 95% CI: −0.616 to 0.525, P = 0.875), l-glutamine (Hedges’ g = −0.07, 95% CI: −0.473 to 0.337, P = 0.742), bovine colostrum (Hedges’ g = 0.10, 95% CI: −0.457 to 0.659, P = 0.723), probiotics (Hedges’ g = −0.01, 95% CI: −0.666 to 0.642, P = 0.972), tyrosine (Hedges’ g = −0.001, 95% CI: −0.426 to 0.423, P = 0.995), BCAAs (Hedges’ g = −0.08, 95% CI: −0.703 to 0.540, P = 0.797), betaine (Hedges’ g = −0.13, 95% CI: −0.967 to 0.706, P = 0.760), l-arginine (Hedges’ g = 0.00, 95% CI: −0.980 to 0.980, P = 1.000), vitamin C (Hedges’ g = 0.02, 95% CI: −1.141 to 1.191, P = 0.967), polyphenols (Hedges’ g = 0.00, 95% CI: −0.800 to 0.800, P = 1.000), curcumin (Hedges’ g = −0.14, 95% CI: −1.126 to 0.838, P = 0.774), quercetin (Hedges’ g = 0.08, 95% CI: −0.992 to 1.156, P = 0.881), menthol (Hedges’ g = 0.09, 95% CI: −0.473 to 0.660, P = 0.746), β-glucan (Hedges’ g = −0.14, 95% CI: −0.635 to 0.362, P = 0.591), amino acids (Hedges’ g = 0.08, 95% CI: −0.544 to 0.965, P = 0.877), and combined α-KG and 5-HMF (Hedges’ g = −0.08, 95% CI: −1.131 to 0.965, P = 0.877) all had “trivial” nonsignificant effects.
For WBSR, nitrate (Hedges’ g = 0.32, 95% CI: −0.083 to 0.716, P = 0.120), blackcurrant extract (Hedges’ g = 0.28, 95% CI: −0.227 to 0.791, P = 0.277), Eurycoma longifolia Jack (Hedges’ g = 0.38, 95% CI: −0.429 to 1.188, P = 0.358), sodium bicarbonate (Hedges’ g = 0.20, 95% CI: −0.368 to 0.768, P = 0.490), combined creatine and glycerol (Hedges’ g = 0.48, 95% CI: −0.332 to 1.296, P = 0.246), combined creatine, glycerol and glucose (Hedges’ g = 0.39, 95% CI: −0.195 to 0.975, P = 0.192), and combined creatine, glycerol, glucose, and α lipoic acid (Hedges’ g = 0.26, 95% CI: −0.671 to 1.187, P = 0.586) had “small” nonsignificant positive effects. Glycerol (Hedges’ g = −0.27, 95% CI: −1.163 to 0.632, P = 0.562) and oligonol (Hedges’ g = −0.25, 95% CI: −0.886 to 0.392, P = 0.448) had a “small” nonsignificant negative effect.
There were a number of medium-to-large effects on WBSR, including GABA (Hedges’ g = −0.78, 95% CI: −1.514 to −0.053, P = 0.036), which had a “medium” significant negative effect and folic acid (Hedges’ g = −0.57, 95% CI: −1.523 to 0.373, P = 0.235), which had a “medium” nonsignificant negative effect. Taurine (Hedges’ g = 0.79, 95% CI: 0.225–1.363, P = 0.006) had a “medium” significant positive effect and whey protein (Hedges’ g = −1.31, 95% CI: −2.248 to −0.371, P = 0.006) had a “large” nonsignificant negative effect.
Local sweat rate.
Subgroup analysis demonstrated a nonsignificant effect of the different supplement categories on LSR (P = 0.886). Caffeine (Hedges’ g = −0.005, 95% CI: −0.529 to 0.519, P = 0.985), GABA (Hedges’ g = 0.08, 95% CI: −0.905 to 1.056, P = 0.880), nitrate (Hedges’ g = 0.17, 95% CI: −0.471 to 0.804, P = 0.608) and sodium bicarbonate (Hedges’ g = −0.15, 95% CI: −0.950 to 0.653, P = 0.717) had “trivial” nonsignificant effects. Catechin (Hedges’ g = 0.20, 95% CI: −0.787 to 1.180, P = 0.695), and taurine (Hedges’ g = 0.29, 95% CI: −0.261 to 0.835, P = 0.305) had “medium” nonsignificant positive effects. Oligonol (Hedges’ g = −0.21, 95% CI: −0.849 to 0.427, P = 0.517) and folic acid (Hedges’ g = −0.47, 95% CI: −1.414 to 0.467, P = 0.323) also had “medium” nonsignificant effects but decreased the local sweating response.
Meta-Regression
Across the three meta-analyses, there were no significant moderating effects. The effect of several moderating variables on WBSR and LSR could not be assessed due to either an insufficient number of studies included in the analysis (supplement dose) or a lack of variation within the moderating variables in the included studies (e.g., training, heat acclimation and hydration status).
Risk of Bias
The studies included had a generally “low” or “unclear” risk of bias, with only 14 studies stating randomization procedures (26–28, 93, 94, 138, 143, 145, 146, 152, 157, 158, 163, 189) and three studies with prepost intervention designs not randomizing or blinding (137, 188, 199). Allocation concealment was “high” in three studies (137, 188, 199) (Fig. 5). A number of outliers were detected in the peak Tcore meta-analysis (82, 86, 98, 100, 102, 124, 130, 150, 186, 197, 199), owing to the large effects that were elicited by some supplements on peak Tcore responses, but Egger’s test showed that there was no publication bias (P = 0.427). Several outliers were detected in the WBSR meta-analysis (119, 125, 128, 195); however, Egger’s test indicated no publication bias (P = 0.358), and influence analysis demonstrated no outcome changes when these were removed. No outliers or publication bias (P = 0.638) were detected in the LSR meta-analysis (Fig. 6).
Figure 5.Risk of bias.
Download figureDownload PowerPoint
Figure 6.Publication bias for peak core temperature (A), whole body sweat rate (B), and local sweat rate (C).
Download figureDownload PowerPoint
DISCUSSION
The main findings of the current meta-analyses were that, overall, pooled analysis of all dietary supplements demonstrated no effect on peak Tcore in the heat (Fig. 2). However, there were differences between supplements, with caffeine, taurine, and oligonol significantly affecting peak Tcore responses, but to varying degrees and directions. Caffeine supplementation appeared to induce a thermogenic effect, while other supplements, such as taurine and oligonol lowered Tcore responses compared with placebo. This is consistent with our previous meta-analytical findings on the thermal effects of caffeine and taurine during exercise in the heat (16), but this work has now expanded the evidence to a wider pool of supplements, across both resting and exercising conditions. Furthermore, the additional analyses of sweating responses revealed that, collectively, dietary supplements may increase WBSR (Fig. 3) but have limited effects on LSR, which is likely to be due to the smaller number of studies included in the analysis (Fig. 4). Despite this, there was variation across supplements regarding their effects on sweating, with taurine demonstrating the greatest increases in WBSR and LSR, and others such as oligonol and folic acid, reducing these responses. In the heat, sweating is the primary heat loss avenue, and as such, is responsible for limiting thermal gain (i.e., Tcore increases). The findings herein, which indicate that dietary supplements may influence these aspects of thermoregulation, have implications for individuals exposed to hot environmental conditions. Furthermore, as demonstrated by the meta-regression analysis (Table 2), there were also no moderating effects of training and heat acclimation status, hydration status, fluid ingestion during the trial, and WBGT on the supplements’ capacity to alter Tcore and sweating responses. Thus, the effects reported are unlikely to be altered by other modifiable factors.
Table 2. Meta-regression of potential moderating variables of the peak core temperature, whole-body sweat rate and local sweat rate meta-analyses
Nitrate, l-Arginine, and Folic Acid
In regard to WBSR and LSR, the current analysis revealed a “small” and “trivial,” nonsignificant positive effect of nitrate, which is a supplement well-established to improve NO bioavailability (32, 201). It is theorized that NO may contribute to eccrine sweat gland function, as local inhibition of NO synthase – NO’s precursor – with NG-nitro-l-arginine methyl ester (l-NAME), has been shown to attenuate sweat rate during moderate exercise in the heat (35, 202), though not all studies provide support for this (203). Furthermore, NO also appears to have a role in regulating cutaneous vasodilation (204–206). Interestingly, numerous studies, in isolation, reported no significant increases in sweating (136–138, 140–142), which equated to a “small” nonsignificant effect based on the collective evidence of the current meta-analysis. Therefore, the degree to which nitrate supplementation augments the sweating response is likely to be insufficient to elicit substantial thermoregulatory benefits. This is supported by the findings of the peak Tcore analysis herein, where increases in sweating did not translate to reductions in Tcore. These findings, in combination, therefore, question whether nitrate supplementation has the capacity to aid thermal balance in hot environmental conditions. In line with this, supplementation with other precursors to NO, specifically l-arginine and folic acid, demonstrated no significant thermoregulatory improvements, as peak Tcore was not lower and sweat rate was not increased following their supplementation. Together, these results suggest that supplementation strategies aiming to increase NO bioavailability fail to enhance the sweating responses, and thus, the opportunity to evaporatively cool when used in a hot environment. Interestingly, many of these studies also demonstrated no greater cutaneous blood flow or dry heat transfer capacity, thereby questioning any thermoregulatory advantage of nitrate or NO donors in the heat (138, 173, 188). Consequently, supplementation with ∼4.2–16.8 mmol dietary nitrate, 10 g l-arginine, or 5 mg folic acid cannot currently be recommended during heat exposure to improve thermal balance, but do not appear to have any deleterious effects.
Caffeine
The peak Tcore analysis demonstrated that caffeine and caffeine in combination with ginseng had “small” and “large” significant positive effects, respectively. Substantial rises in Tcore can cause heat strain and, ultimately, lead to heat exhaustion and heat stroke if not sufficiently controlled (6, 7). This is particularly the case during heat exposure, or exercise in hot conditions, when avenues of dry heat dissipation are reduced due to a smaller temperature gradient between the ambient air and skin’s surface (69, 207). There is evidence that caffeine supplementation increases ˙𝑉V˙O2 and, consequently, heat production at comparable exercise intensities compared to placebo (86, 208), potentially explaining this observed Tcore rise. Indeed, the V̇o2 response to exercise has been reported to be increased by 3% to 15% following caffeine supplementation in the heat (87, 98, 100, 101), although others have demonstrated negligible differences (81, 82, 103). These results could be attributed to the methodological differences between studies, as noted by John et al. (94), which calls for improved standardization of laboratory procedures in studies relating to thermoregulation and caffeine supplementation. In addition to the current results, regarding Tcore, WBSR and LSR were not increased with caffeine supplementation. Given the capacity of sweating to help dissipate excess metabolic heat through evaporative cooling, these “negligible” effects would place greater demand upon dry heat transfer. However, it appears that caffeine-mediated reductions in cutaneous blood flow, owing to vasoconstriction of the skin microvasculature, would preclude this possibility (93, 209). Thus, greater heat production, coupled with reduced SkBF and no increase in sweat production, explains why heat retention would ensue and the observed rises in Tcore reported herein. Indeed, the previous meta-analytical findings support this observation and have been corroborated further by a recent meta-analysis, which demonstrated a 0.1°C·h−1 greater Tcore rate of rise following caffeine supplementation (210). Interestingly, combined caffeine and taurine only had a “trivial” effect on peak Tcore, theoretically due to taurine’s thermolytic effect negating the additional heat production elicited by caffeine. However, without V̇o2 data and corroboration of these findings, this remains speculative. Overall, these results suggest that caffeine has an undesirable (i.e., heat-gaining) effect on thermal balance and question its use in hot environmental conditions. Therefore, acute caffeine intake of between 3 and 9 mg·kg−1 before exercise, or at rest during heat exposure, should potentially be avoided due to its thermogenic effect in such conditions. Although caffeine intake has not been directly linked to heat illness cases, substances that elevate Tcore could result in an increased risk of heat-related illnesses (211, 212).
Taurine
The current evaluation of sweating responses revealed that taurine had a “medium,” significant effect on peak Tcore and WBSR, as well as a “small,” nonsignificant positive effect on LSR. These studies suggest that ingestion of taurine before exercise in the heat augmented the sweating response by hastening sweat onset (26) and increasing sweat rate (26, 27). Theoretically, this would improve thermal balance, due to enhanced latent heat transfer and reduced heat storage, delaying rises in Tcore (1, 5). Although this indicates that taurine can exert a beneficial thermoregulatory effect, these findings require replication, including further insight into its mechanisms of action, which are poorly understood. More thorough elucidation of taurine’s effects on sweating and the consequential impact on heat transfer and heat tolerance is necessary, alongside investigation of these mechanisms. However, based on these studies, acute intake of 50 mg·kg−1 of taurine before exercise in the heat induces an earlier sweating onset, greater sweating rate, and lower Tcore responses and may offer a strategy to improve thermoregulatory capacity.
Tyrosine, BCAAs, and GABA
Tyrosine and BCAAs demonstrated no significant effect on peak Tcore or WBSR. Despite these amino acids previously being reported to provide some of the greatest performance effects of any supplements during exercise in the heat (16), they do not appear to confer thermoregulatory benefits. The ability to reduce central fatigue is often ascribed to these supplements to explain their erogenicity in such conditions (161, 213–215), but no apparent link to temperature regulation was found in this meta-analysis. However, tyrosine is an essential substrate for tyrosine hydroxylase, which is involved in axonal catecholamine synthesis [particularly norepinephrine; (216)]. Thus, sufficient tyrosine availability is required to maintain catecholamine levels and facilitate sympathetic vasoconstrictive effects on the subcutaneous vasculature (217). This has been reported to attenuate the rate of cold-induced decline in Tcore among those likely to have tyrosine deficiency (217), but there were no similar effects reported across studies conducted in the heat. Interestingly, another amino acid, GABA, has some potential to offer thermoregulatory benefits, with the peak Tcore analysis revealing a “small,” yet nonsignificant thermolytic effect. Although one of the two studies that supplemented GABA demonstrated null peak Tcore effects, there was a slower rate of rise in Tcore across the 30-min exercising period (170). GABA is a widely distributed neurotransmitter within the central nervous system, where it acts in the hypothalamus to regulate internal body temperature (218–220). Exogenous supplementation in humans is thought to increase GABA’s availability in the hypothalamus (221) and, thereby, influence temperature regulation (170). The hypothalamus contains cold-sensitive neurons, which have a role in controlling heat production upon detection of local and peripheral temperature changes (222, 223). In the animal model, suppression of these neurons appears to occur following GABA administration, leading to lowered Tcore responses in the heat (224). Indeed, the two original research articles included in the current meta-analysis (169, 170) observed reductions in metabolic heat production, which could in part explain the effects on Tcore in the GABA supplementation conditions. In addition, GABA attenuates activity of the sympathetic nervous system (225, 226), which would likely suppress epinephrine and norepinephrine release (227), as observed by Miyazawa et al. (169). Reductions in catecholamines have been associated with slower rises in Tcore during hyperthermic exercise (228), which is supported by the findings herein. Considering these effects on heat production, it is unsurprising that there was also a large, significant reduction in WBSR and a “trivial” effect on LSR, as it is a known driver of the thermal sweating response (5, 66, 68). Therefore, while GABA appears to reduce one avenue of heat dissipation (i.e., evaporative cooling), it has created a greater heat storage capacity, which would delay the onset of hyperthermic symptoms during heat stress and may be effective during short-duration exercise in the heat. Based on the two studies herein, the administration of 1 g of GABA directly before heat exposure (rest or exercise) appears to provide a beneficial effect on thermal balance through a reduction in heat gain. However, acute tyrosine and BCAA intake would not be a useful supplement in hot environmental conditions, due to the null impacts on thermoregulation.
Glycerol and Creatine
The peak Tcore analysis herein revealed “small,” nonsignificant negative effects for glycerol and combined creatine and glycerol supplementation. “Medium,” nonsignificant negative effects were observed for combined creatine, glycerol, and glucose, and “trivial,” nonsignificant effects for creatine and combined creatine, glycerol, glucose, and alpha lipoic acid. Although these results demonstrate that glycerol supplementation had a small-to-medium effect on peak Tcore, the variation across studies decreased the certainty of these findings, rendering them nonsignificant. When ingested, glycerol provides osmotic pressure in the plasma and intra- and extra-cellular water compartments—where concentrations are evenly distributed—and thereby increases their water content (15, 132, 229). Creatine acts similarly, as its transport into cells (primarily skeletal muscle) increases total body water through expansion of intra- and extracellular water compartments, with even fluid distribution (113, 230, 231). This increase in total body water and plasma volume expansion induces “hyperhydration” and affords excess fluid to compensate for sweat losses (116, 117), alongside providing greater availability to sweat glands to facilitate sweat production (232). This would likely improve thermoregulatory capacity, through increased evaporative cooling, but also because additional total body water enhances the specific heat-carrying capacity of the tissues and blood (108, 233, 234). Here, it can help transfer heat from the core to the periphery to be dissipated (235–237).
Creatine had a “trivial,” nonsignificant positive effect on WBSR, and combined creatine and glycerol, combined creatine and glycerol with the addition of glucose, and combined creatine and glycerol with the addition of glucose and alpha lipoic acid had “small,” nonsignificant positive effects. The role of glycerol combined with creatine was, therefore, also partially effective in promoting a sweating response but, as with the peak Tcore responses, the effects were inconsistent across studies, which increased the uncertainty of the “small” effects. Surprisingly, glycerol alone had a small, non-significant negative effect on WBSR. However, this was largely influenced by two glycerol supplementation conditions from the same study, where much larger fluid losses in the placebo group were reported (119). Collectively, it appears that these supplements may be capable of lowering Tcore and enhancing sweating responses compared to placebo, with the combination of creatine and glycerol potentially providing the greatest thermoregulatory benefit. However, with the large inconsistencies between studies and nonsignificant findings this is far from established, and further work is required to understand the heterogeneity of responses. For WBSR this could be dose-related, as for glycerol, Dini et al. (119) provided the highest glycerol dose (3 g·kg−1) and observed a large, negative effect on WBSR. Theoretically, glycerol ingestion of a large quantity may surpass concentrations that can be absorbed into the intra- and extracellular fluid, further elevating plasma concentrations and increasing osmotic pressure. This may prevent fluid from being drawn from the plasma to the sweat glands, thereby decreasing the sweat rate. Without further investigation into the effect of glycerol dose on the sweating response, more specifically, this remains speculative. Indeed, despite the outcome of this single study, the moderating effect of glycerol dose on WBSR was not significant. Overall, these findings demonstrate that lower doses of glycerol (1 to 1.4 g·kg−1), alone or in combination with 20 to 25 g·d−1 of creatine for between 3 and 9 days, appear to aid thermal balance during exercising heat stress, through hyperhydration. Additional investigation into whether this supplementation strategy would provide similar benefits during passive heating is also warranted. Higher doses of glycerol (e.g., 3 g·kg−1), however, may reduce this capacity due to lower sweat rates, though a greater understanding is necessary before providing definitive recommendations.
Sodium Citrate and Sodium Bicarbonate
As supplements often ingested before high-intensity exercise to improve blood buffering capacity, both sodium citrate and sodium bicarbonate have also been reported to increase plasma osmolality and plasma volume (238). This could feasibly help with thermoregulation in the same manner as creatine or glycerol loading; indeed, ingestion of sodium citrate had a “small,” yet nonsignificant negative effect on peak Tcore, which was similar to the previous supplements detailed above. However, this was not coupled with a greater WBSR. Conversely, sodium bicarbonate had a “small,” nonsignificant positive effect on Tcore. In these two studies, the placebo group ingested sodium chloride to match the sodium content of the two conditions, as they were investigating the buffering capacity of the supplement and not its effects on fluid balance (134, 135, 238). Therefore, it is likely that any potential osmotic effects that could theoretically have aided thermoregulation would be indistinguishable from the effects in the placebo condition. In support of this, there was a “small,” nonsignificant positive effect on WBSR and no effect on LSR. To establish whether sodium bicarbonate’s effects on fluid balance can aid thermoregulatory function in hot environmental conditions, studies would need to be conducted with a placebo group that does not contain any sodium. It appears that sodium citrate can potentially improve thermoregulatory capacity in the heat, though this was inconsistent across studies. Any thermolytic effect is likely due to its effects on plasma volume, as expansion was observed across all studies included in the analysis, but to a larger degree in the studies which demonstrated lower Tcore responses. As there was no greater WBSR associated with the lower Tcore response, a greater heat-carrying capacity of the blood may be responsible for these observed effects (233, 234). However, further research is required to corroborate these findings and establish whether sodium bicarbonate can elicit the same benefits. Therefore, based on the studies included in this analysis, there is evidence to suggest that acute sodium citrate ingestion of ∼100–600 mg·kg−1 can improve thermoregulatory capacity during exercise in the heat, but more research is needed to establish these effects at rest and with sodium bicarbonate supplementation.
Betaine
Betaine is an amino acid, which acts as both an osmolyte, to assist with cell volume regulation, and as a methyl group donor to convert homocysteine to methionine (239). It can be supplemented to reduce high plasma concentrations of homocysteine (240) or to improve endurance and resistance exercise performance (241). Due to its osmotic role, it has mechanistic potential to aid fluid balance and thermoregulation during exposure to heat stress (172). However, both the current peak Tcore and WBSR analyses demonstrated no effects. In addition to the measured variables in the current meta-analysis, one of the included studies observed plasma volume expansion across the study in response to betaine supplementation (171); however, the other did not (172). Together, these results question whether betaine is efficacious for fluid balance when ingested before exercise in the heat. There is also some evidence to suggest that betaine may attenuate thermal cellular stress in a similar manner to heat shock proteins (242, 243) and in animal models, it has repeatedly been demonstrated to reduce Tcore when chronically supplemented (244–246). Therefore, betaine may have the capacity to improve heat tolerability, and considering the limited and equivocal evidence in humans, this supplement requires further investigation.
Anti-Oxidants and Anti-Inflammatories
In the current meta-analysis, lower peak Tcore responses for several supplements with known antioxidative and anti-inflammatory properties were found. Oligonol and catechin had a “medium,” significant and a “large,” nonsignificant negative effect on peak Tcore, respectively. Furthermore, “small,” nonsignificant negative effects were observed for curcumin, vitamin E, and effective microorganism X (an antioxidant mixture), and “trivial,” nonsignificant negative effects for blackcurrant extract. The anti-inflammatory role of oligonol, catechins, curcumin, vitamin E, and effective microorganism X is most likely responsible for the lowered Tcore responses compared to placebo, where endogenous pyrogenic cytokines (247) may be attenuated. Indeed, oligonol supplementation lowered circulating levels of the pyrogenic cytokines, such as interleukins IL1-β and IL-6 (180), along with reductions in serum prostaglandin E2, a known intermediary in the development of fever (181, 248). The cytokine response can be acutely lowered with anti-inflammatory substances (249), theoretically leading to decreased thermal gain (250), explaining why the rate of rise in Tcore was reduced, despite no greater WBSR or LSR. However, not all of these studies observed reductions in pro-inflammatory cytokines (158, 183), despite attenuated increases in Tcore. All trials investigating oligonol and catechin—which had the greatest effects—induced heat strain via hot water immersion of the lower body at rest. This is a rapid means by which to facilitate heat gain, as water is highly conductive (251), yet it partially attenuates other avenues of heat dissipation, such as evaporation (1). It is possible that an immersive protocol induced greater thermal strain and production of pyrogenic cytokines, meaning that Tcore responses were more readily identified between conditions. However, the rise in Tcore within these trials was less than would be expected, only reaching an average of 37.52°C across all trials; although this was with tympanic measurement, which may explain the relatively low Tcore values. Another explanation for lower peak Tcore responses is increased heat dissipation; however, sweating was only greater following catechin and blackcurrant supplementation and these effects were nonsignificant. While this may partly explain the improved thermal balance, in this instance, it appears likely that reductions in endogenous pyrogenic cytokines have an important role to play in the efficacy of many of the anti-inflammatory supplements. Nevertheless, only oligonol has demonstrated significant impacts on aspects of thermoregulation and, therefore, further investigation of these supplements and mechanisms is required to provide definitive answers.
An additional role of antioxidants is to improve cellular oxidative capacity and, therefore, redox status (252). These effects could be directly extended to sudomotor function, based on the reported relationships between systemic markers of lipid peroxidation and sweat production (253). However, further research is needed to explore this possibility, owing to the failure of local anti-oxidant infusion to acutely alter the local sweating response during exercise-heat stress (254), which questions the likelihood that anti-oxidants play a major role in thermoregulatory sweating. Indeed, a greater sweating response was not observed for the majority of these supplements. A component of catechin, epicatechin, has been associated with greater cutaneous blood flow during heat exposure through improved NO signaling (255, 256). The results were nonsignificant, but if substantiated, the observed augmented sweating response may be due to the associated enhancement of skin blood demonstrated in response to catechin supplementation (191). In combination, these would improve evaporative and dry heat transfer, explaining the lower Tcore response. Similarly, there is evidence that anthocyanins, a key component of blackcurrants, promote the production of NO, through augmented NO synthase activity (257). Furthermore, Eurycoma longifolia Jack—another supplement with an antioxidative function—had a “medium,” positive nonsignificant effect on WBSR, yet no significant change in peak Tcore. Neither of these two latter studies measured or estimated SkBF, or characterized other indices of vascular function and, therefore, it can only be speculated that any potential greater WBSR observed—albeit nonsignificant—is in response to the aforementioned mechanisms.
Interestingly, beta-glucan and ginseng, other supplements with anti-inflammatory properties, had “small,” nonsignificant positive effects on peak Tcore, and beta-glucan also had no effect on WBSR. While other endogenous pyrogens were significantly reduced immediately postexercise in the beta-glucan condition compared with placebo, there was a transient elevation of macrophage inflammatory protein 1β, which may provide a potential explanation for these findings. However, without further investigation into beta-glucans’ thermoregulatory effects during heat exposure, any mechanistic explanations remain speculative. Ginseng is a herb with many bioactive ingredients, which has been demonstrated to increase body temperature in the animal model and may also partially explain this thermogenic effect (258).
Other antioxidants; vitamin C and polyphenols had no observable effects on peak Tcore or WBSR. Quercetin and combined α-KG and 5-HMF also had no effect on WBSR, but a “small,” nonsignificant negative and a “small,” nonsignificant positive effect on peak Tcore was observed, respectively. It has been theorized that quercetin, a well-characterized antioxidant, is capable of inhibiting (via ROS scavenging) the necessary molecular signaling events required to acquire the acclimated phenotype, by reducing the heat-shock factor or hypoxia-inducible factor response to heat exposure (184). Acutely, antioxidant intake would improve redox balance and potentially aid heat tolerance, but if supplemented chronically or alongside heat exposure may blunt adaptations (184, 259). A similar argument can be posed for supplements with anti-inflammatory properties (260). Indeed, a greater number of studies demonstrated beneficial peak Tcore and sweating responses (e.g., catechin, oligonol, quercetin), when supplemented acutely (1 day), but there is no clear consensus on dosing length and supplement efficacy within the current analysis. Nevertheless, based on the required time-course of these cellular adaptations, this mechanism could partially explain the lack of difference between antioxidants consumed over longer periods (≥7 days) and placebo supplements in the current meta-analysis. Across the antioxidant and anti-inflammatory supplements in the current meta-analysis, 10 of the 19 supplements were consumed repeatedly across 3 to 42 days, which means that any potential thermoregulatory effects may have been masked. In summary, the use of anti-oxidants results in variable responses in the heat, which could be partly explained by their multi-ingredient composition, or dosing period. It is important that the specific mechanisms by which these variable effects occur should be investigated, especially if chronic administration of anti-oxidants and anti-inflammatories may reduce adaptation to heat exposure and exacerbate heat illness. Indeed, many of these supplements are more likely to be ingested by people who require antioxidative or anti-inflammatory agents, such as older or clinical populations, who are also more vulnerable to heat stress (261, 262) and are also less likely to tolerate increases in Tcore (i.e., ginseng). Such individuals could benefit from reductions in Tcore and dietary supplements that may induce this (i.e., oligonol and catechin), assuming that there are no other apparent side effects.
The current analysis suggests that 100–200 mg oligonol ingested ∼30–60 min preheat exposure has a beneficial effect on thermal balance, by reducing heat gain. However, further investigation into oligonol’s efficacy during exercising heat stress is necessary to further elucidate its effects on thermoregulatory capacity. Catechin appears to have a similar effect, though corroboration of this finding is required, as only one study has been conducted thus far. There is also tentative evidence that prolonged intake of 800 mg curcumin, vitamin E, 70 mL effective microorganism X (3, 42, and 7 days, respectively), may reduce Tcore responses and 600 mg blackcurrant extract and 150 mg Eurycoma longifolia Jack may improve sweat rate during exercise in the heat. However, these results were nonsignificant and based on only one (curcumin, vitamin E, effective microorganism X, and Eurycoma longifolia Jack) or two studies (blackcurrant extract), so these results are not conclusive. Further research is required to establish these anti-inflammatory supplements’ efficacy during heat exposure and their effects on endogenous pyrogenic cytokines. Additional investigation into their effects when ingested acutely, at rest and during more ecologically valid conditions is warranted before more definitive thermoregulatory effects can be established. Similarly, the potential thermogenic effects of 200 mg ginseng, 250 mg beta-glucan and combined 4.8 g α-KG and 60 mg 5-HMF require additional examination, as their intake cannot currently be recommended based on the results herein. Further, ingestion of 250 to 1,500 mg of vitamin C and polyphenols does not appear to influence thermoregulatory responses (Tcore or sweat rate) during exercising heat stress and, therefore, while intake is unlikely to facilitate improved thermal balance, it is also unlikely to have detrimental performance or health consequences. However, establishing these effects following acute doses may reveal further impacts on thermoregulatory capacity.
l-glutamine, Bovine Colostrum, Probiotics, Whey Protein, and Amino Acids
l-glutamine, bovine colostrum, and probiotic supplementation all had no effects on peak Tcore and WBSR, suggesting that they confer no thermoregulatory benefit in the heat. These supplements, along with curcumin, are often ingested before exercise in hot environmental conditions, with the aim of maintaining gastrointestinal (GI) barrier integrity and reducing symptoms of GI dysfunction. Gastrointestinal injury and changes to epithelial permeability are relatively common during exercising heat stress (263), which consequently, leads to translocation of endotoxins and bacterial lipopolysaccharides into the central circulation, causing systemic endotoxemia (264). The subsequent release of proinflammatory cytokines can, in turn, cause cytokenemia and additional elevations in Tcore (264–266). However, evidence for these supplements’ efficacy is equivocal, along with their function as ergogenic aids in the heat. Favorable effects of supplementation with whey protein (195) and an amino acid beverage have been demonstrated on GI permeability during exercising heat stress, where there was a “small,” nonsignificant negative, and a “trivial” effect on peak Tcore, respectively. Whey protein supplementation also induced a large, significant negative effect on WBSR. In this study, the whey protein condition had a lower circulatory endotoxin concentration postexercise compared with placebo (195), which theoretically may explain any Tcore differences. The large inhibitory effect on sweating was unexpected, but given the lower Tcore—and likely heat production—the drive for sweating would be reduced (67). The only other supplement to display any potential improvements to thermoregulatory capacity is curcumin—as detailed previously—which is more likely to be due to its aforementioned anti-inflammatory role. Probiotics and bovine colostrum supplementation did not reduce circulating endotoxin or cytokine concentrations in the studies within which these were measured (147, 154–156) and only one study, which supplemented l-glutamine demonstrated reductions in endotoxins and TNF-α (151). These supplements may be less effective at preventing GI injury in the heat, due to greater redistribution of blood flow from the gut (GI ischemia) to the peripheral vasculature (267) and, consequently, have no influence on Tcore responses. Indeed, only a few studies identified improvements to GI barrier integrity (148, 150–152) and largely attributed this improvement to the upregulation of heat shock protein 70, which inhibits inflammation (144, 150, 151). Therefore, the long-term use of probiotics (7–28 days) and ∼20–140 g bovine colostrum (7–14 days) and acute use of 0.15–0.9 g·kg−1 l-glutamine to maintain GI barrier integrity in hot environmental conditions appears to provide no thermoregulatory advantage. Although supplements targeting the GI tract during heat stress are an area of ongoing interest, further research is required to establish other efficacious alternatives. Indeed, an acute dose of whey protein (15 g) may provide an effective option (195), but without replication of these results, this cannot be definitively stated.
Menthol and Thermo Speed Extreme
The oral supplementation of menthol nonsignificantly lowered peak Tcore. This reduction was unanticipated but the variability across studies explains the nonsignificant effect. Menthol is typically considered to be a nonthermal cooling agent (268), which evokes the perception of cooling via transduction of the TRPM8 receptors in the oral cavity (269–272) and possibly the viscera (273), without directly affecting thermal balance according to current literature (274–276). However, current findings were heavily influenced by a single study’s results, where Tcore was estimated by tympanic measurement (186), which can be less reliable if the correct procedures are not adhered to. Therefore, there is some doubt over these results. As discussed in Barwood et al.’s expert-led consensus article (268), there are some reports of menthol causing vaso-reactivity in the skin’s subcutaneous vasculature when applied externally, but no consensus was reached on any form of menthol administration and thermoregulatory effects. Therefore, replication of this single study may be required to confirm whether acute menthol ingestion can mechanistically affect temperature regulation. Additionally, there was no effect on WBSR, which supports the current consensus (268). Thermo Speed Extreme is another supplement that did not affect peak Tcore and given that this supplement contains ingredients such as caffeine (194), this finding is somewhat unexpected. It is possible that the tympanic measures used within this study were insufficiently sensitive to detect Tcore changes. However, significantly greater chest Tsk was observed at certain timepoints across the trial, which could enhance dry heat dissipation to the environment, particularly as the ambient air was much cooler (26.2°C) than average Tsk across participants (34°C). The ingredient piperine could be responsible for this likely enhancement of cutaneous vasodilation, as in vitro studies suggest it may have vaso-modulating effects (277). This could explain the tendency towards lower Tcore values, despite the thermogenic nature of the supplement. It should be stated that this supplement would, therefore, not be appropriate for use in ambient temperatures that exceed Tsk, where skin surface to ambient air temperature gradients, and dry heat transfer capacity are reduced.
Moderating Effects
No candidate moderators, such as training and heat acclimation status, hydration status, and fluid ingestion during the trial, affected peak Tcore or sweating responses to dietary supplementation. For hydration status, this is likely due to the majority of papers stipulating the inclusion of hydrated participants. However, there was more variation in the training (highly trained; 43% vs. recreationally active; 39%) and heat acclimation (acclimated; 14% vs. nonacclimated; 36%) status of participants and whether the fluid was ingested during the trials (ingested; 46% vs. not ingested; 22%), yet no moderating effects were found. Nevertheless, it would be useful for future studies to consider investigating the efficacy of dietary supplements on thermoregulation among participants of different training and heat acclimation statuses, given the effect of these processes on sweating and Tcore responses (11). Some supplements, such as sodium citrate, nitrate, l-glutamine, and tyrosine, have only been used to assess thermoregulatory responses in nonacclimated participants, which limits the wider application of these to potential end-users. Whether this would augment or negate any effects seen from these supplements is of particular interest and important to establish for individuals in competitive sport, military, and occupational settings. In addition, all other meta-regressions (WBGT, trial type and length, and supplementation period) did not moderate the effect of dietary supplementation on peak Tcore or sweating. There is a large range of supplements included within the current meta-analyses, each with differing underpinning mechanisms and nuances in their efficacy. It is, therefore, unsurprising that there are no consistent moderating factors.
Although these trial moderators had no significant effects in the present analysis, they still require further investigation, particularly within the most efficacious supplements included here. For example, acute supplementation and the use of exercise were most common, with no variation within certain supplement categories. The effect of chronic supplementation of certain supplements, such as various anti-oxidants, glycerol, taurine, and other amino acids (e.g., l-glutamine), on Tcore and sweating responses in the heat is almost completely unknown. Taurine has been shown to elicit various physiological effects following chronic supplementation, which may be advantageous during heat exposure, such as enhanced vascular function (25) and an improved endurance trained phenotype (278–280). All studies investigating the effects of l-glutamine on GI barrier integrity in the heat have supplemented acutely and it is possible that a chronic dose may be more efficacious. Indeed, long-term administration (2 mo) has demonstrated beneficial effects on GI permeability in patients with Crohn’s disease (281). Longer-term glycerol intake has previously elicited hyperhydration for up to 49 h (282), but whether it can be maintained over a greater period of time is currently unknown. Further research into this, alongside potential side effects (e.g., hyponatremia) would help establish whether glycerol supplementation can provide long-term beneficial effects on thermal tolerance. In addition, as detailed in Anti-Oxidants and Anti-Inflammatories above, the chronic and acute effects of various antioxidants and anti-inflammatory supplements in the heat require investigation. Furthermore, establishing the efficacy of these supplements during differing trial types with differing physiological demands is necessary to be able to provide practical advice and application to athletes, workers, and the general population. The metaregression of WBGT demonstrated no effect, but ambient vapor pressure does have an established impact on avenues of heat dissipation (69). For example, supplements that augment thermal sweating (e.g., taurine) will likely be most effective in dry environments where any sweat produced can be evaporated, thereby providing a cooling effect. Depending on the mechanistic actions of certain supplements and beneficial thermoregulatory effects, this could have a large impact on their efficacy and ability to help individuals maintain thermal equilibrium. Investigation of these above-mentioned factors is important, particularly among the most efficacious of the supplements examined within these meta-analyses.
Limitations
Within these meta-analyses, several supplements were taken in combination, such as creatine and glycerol (198–200), caffeine and ginseng (197), caffeine and taurine (28), combined α-KG and 5-HMF (193), whey protein (195), amino acid beverage (196), effective microorgansim X (192), and Thermo Speed Extreme (194). However, as only a few studies employed a coingestion strategy, there is limited information on the thermoregulatory outcomes when using this approach across a wide range of different supplements. Herein, the combined effect of creatine and glycerol was beneficial for thermal balance, while the coingestion of caffeine and ginseng further exacerbated caffeine’s thermogenic effect. As such, the former could be recommended to improve fluid balance in the heat; however, the latter may pose a greater heat stress risk and potentially should be avoided in hot conditions. This is perhaps unsurprising given that caffeine alone causes an increase in Tcore. Considering these differing findings and the indication that co-ingestion potentially influences the thermoregulatory responses to these supplements, greater clarity across supplement types regarding these effects is certainly warranted. Indeed, athletes and military personnel often combine dietary supplements (283, 284), which may increase the risk of heat stress if a harmful combination is unwittingly ingested. Therefore, further research regarding the effect of dietary supplement co-ingestion on thermoregulatory responses during heat exposure is necessary and represents a key gap in the literature and a further lack of specific supplementation guidance for potential users.
Conclusions
In summary, for the first time, the effects of various dietary supplements on Tcore and sweating responses in the heat have been evaluated. The amino acids taurine and GABA, alongside whey protein, lowered peak Tcore, indicating an improvement in thermal balance. While GABA and whey protein negatively impacted WBSR, taurine increased the sweating response, demonstrating an enhancement to thermoregulatory capacity, albeit from a single study. However, other amino acids, such as tyrosine and BCAAs appeared to have no meaningful effect on thermoregulation. Various supplements with anti-oxidative and anti-inflammatory properties (e.g., oligonol, catechin, curcumin, vitamin E, and quercetin) provided beneficial effects on peak Tcore, which may in part be explained through improved redox balance and attenuation of endogenous pyrogenic cytokines. Nevertheless, not all of these supplements improved thermal balance, highlighting the need for additional research in this area. A number of supplements (e.g., glycerol, creatine, sodium citrate, and betaine), which appear to induce hyperhydration and/or expand plasma volume, nonsignificantly lowered Tcore responses. Mechanistically, this may be through increasing heat-carrying capacity and/or improving fluid availability to the sweat gland, as some supplements (e.g., combined glycerol and creatine) also demonstrated an effect on sudomotor function. However, Tcore and sweat rate findings were inconsistent across studies and supplement types, rendering the results nonsignificant overall. Many other supplements such as nitrate, l-arginine, folic acid (taken for their effects on NO bioavailability), l-glutamine, bovine colostrum, and probiotics (taken for their effects on GI barrier integrity) did not appear to provide any thermoregulatory benefit in the heat. Peak Tcore was greater following caffeine and combined caffeine and ginseng supplementation, without any increases in sweating responses. Consequently, caffeine when ingested during heat exposure appears to be thermogenic and, therefore, may have potential negative health implications. Several other supplements, such as ginseng, beta-glucan, and combined α-KG and 5-HMF also demonstrated “small” thermogenic effects, though these results were nonsignificant.
Although additional investigation is certainly required, some supplements have demonstrated the potential to improve thermoregulatory capacity in the heat. However, it appears that others have null or even deleterious effects on thermal balance when ingested in such conditions. These findings suggest that certain supplements, such as caffeine, should possibly be avoided in hot conditions, and others, such as taurine, may elicit a thermoregulatory benefit. This has potential implications for those ingesting dietary supplements for their health and/or performance effects during periods of heat exposure. Indeed, official guidance documents for the general population, athletes, and military personnel could also be updated to reflect the varying effects different dietary supplements appear to have on thermoregulation, detailing which to avoid and which may be advantageous in hot conditions. Additional investigation into many of these supplements is required to corroborate findings and provide greater understanding of their effects. Specifically, future research should focus on the thermolytic effects of various supplements such as taurine, GABA, oligonol, and catechin in varying conditions, alongside further mechanistic insight into these responses.