High ambient and radiant temperature, absolute humidity, and factors such as urbanization and heat storage in crowded stadiums may cause “heat stress” in athletes (Bongers et al., 2020). Under these conditions, an athlete’s ability to dissipate the exercise-induced heat production is limited, leading to significantly elevated core temperature and increased risk of performance impairments (Bongers et al., 2020) and heat-related illness (Wendt et al., 2007), including disease (heat edema, heat rash, heat cramps, heat syncope), heat exhaustion, and the most severe form, heat stroke (Howe and Boden, 2007). Typical symptoms of heat exhaustion are dizziness, malaise, nausea, vomiting, or excessive fatigue, and without treatment, the potentially life-threatening heat stroke may develop as core temperature elevates >40°C (Howe and Boden, 2007).

Exertional heat illness has been reported at a rate of 0.47 per 10,000 athlete-exposures among US collegiate athletes (Yeargin et al., 2019), with heat cramps, heat exhaustion, and dehydration being the most prominent types and 8.2% of all cases requiring emergency transport. However, the prevalence of heat-associated diseases might rise since athletes will have to train and compete under more challenging thermal conditions. Besides the fact that global warming will probably lead to an increase in the frequency and length of heat waves, including heat waves occurring in previously temperate environments (McGeehin and Mirabelli, 2001), also major sports events are often organized in extremely hot and/or humid conditions [e.g., Olympic Games in Tokio 2020, Fédération Internationale de Football Association (FIFA) World Cup in Qatar 2022].

To reduce the risk of performance impairments and exercise-induced heat illness when exercising in hot-humid conditions cannot be avoided, athletes need to apply cooling strategies before (pre-cooling) or during (mid-cooling) exercise to lower core temperature. Cooling applications, in general, can improve exercise performance in hot environments due to reductions in thermal strain and an increased heat storage capacity (Bongers et al., 2017). They can be classified into external (i.e., cold-water immersion, ice packs, ice vests) and internal (ingestion of ice, cold-water, and menthol) applications. Several systematic reviews and meta-analyses demonstrate a positive effect of cooling on physical performance (Jones et al., 2012; Bongers et al., 2015; Ruddock et al., 2017; Choo et al., 2018; Douzi et al., 2019; Jeffries and Waldron, 2019; Zhang, 2019; Bongers et al., 2020; Keringer et al., 2020; Rodríguez et al., 2020), with external applications such as cold water immersion and ice vests being the most effective strategies for pre- and mid-cooling, respectively (Bongers et al., 2020).

However, not every cooling method providing performance benefits in the laboratory setting is feasible for real-life competition. For example, due to sport-specific regulations, practical considerations, local environmental conditions, high performance costs (even though regulations and conditions would allow for external cooling) it might not be possible to employ external cooling such as cold-water immersion or ice vests (Bongers et al., 2020). In contrast, internal cooling applications are more applicable during exercise in a field-based setting and are usually well-tolerated and cheap (Bongers et al., 2017).

There exist two different types of internal cooling: physical and perceptional cooling. The application of physical cooling with a medium of high heat capacity (e.g., ice or cold-water ingestion) might cause a decrease in core temperature and a consequent delay in the onset of thermally induced fatigue (Wegmann et al., 2012). In a recent meta-analysis by Zhang 2019, the ingestion of ice-slurries was associated with moderate performance improvements in hot environments (Zhang, 2019). However, internal heat losses caused by the ingestion of cold fluids might decrease the evaporative potential of the skin (Morris et al., 2016; Jay and Morris, 2018). Therefore, some authors recommend to ingest cold-water/ice-slurry only during exercise in hot, humid, and calm conditions, but not in warm, dry, and windy environments (Jay and Morris, 2018). On the other hand, high-intensity exercise may cause excessive elevations in heat production and sweat rate, and small reductions in sweat rate would only minimally reduce evaporative heat loss, suggesting a net beneficial effect (Bongers et al., 2020). Furthermore, it could be argued that a reduction in sweat rate following the ingestion of cold-water or ice-slurry could prevent dehydration-dependent performance decrements (Murray, 2007).

In contrast, perceptional cooling may affect physiological outcomes and performance indirectly by inducing a sensation of cooling (Keringer et al., 2020). The most comment agent is menthol, a cyclic monoterpene alcohol that possesses various biological properties such as antimicrobial, anticancer, anti-inflammatory activities, and well-known cooling characteristics (GPP et al., 2013). Internal menthol application leads to an activation of “transient receptor potential melastatin-8" (TRPM8)-channels causing a reduced thermal sensation and physiological effects similar to “physical cooling” (Zheng, 2013). A recently published study by Han et al., 2020 (Han et al., 2020) showed that intranasal menthol activated several brain regions related to nociceptive and trigeminal processing. However, it remains unclear whether and to what extent this activation has a performance impact. Three systematic reviews assessed the effects of internal and external menthol application on performance in the heat (Douzi et al., 2019; Jeffries and Waldron, 2019; Keringer et al., 2020). According to one review, internal menthol was superior to an external application (Jeffries and Waldron, 2019), whereas another review reported contrary results (Keringer et al., 2020), and one did not show any effect (Douzi et al., 2019).

So far, no systematic review has focused exclusively on the effect of different internal cooling methods (ice-/cold-water and menthol ingestion). In most of the above-mentioned systematic reviews, no separate analysis was performed, differentiating between internal and external or various internal cooling methods. Therefore, this review aims to systematically screen and evaluate the literature on the effects of internal cooling on various outcomes (n = 11), including performance, physiological, and perceptional parameters. Several previous reviews assessed the effect of cooling on aerobic performance without the differentiation between performance and capacity (Jones et al., 2012; Bongers et al., 2015; Douzi et al., 2019; Jeffries and Waldron, 2019; Zhang, 2019; Bongers et al., 2020). However, aerobic performance relates to completing a certain task as fast as possible (e.g., time trials), whereas endurance capacity refers to the exercise time to volitional fatigue at a constant workload or speed (e.g., time to exhaustion) (Saris et al., 2003). Endurance capacity is more often studied since the technique is relatively easy to control, and the constant workload allows comparison of metabolic and other measurements between intervention and control trials. Yet, for the assessment of true aerobic performance, time trials are the more valid and realistic approach since there are only a few events where athletes have to exercise as long as possible (Saris et al., 2003). Therefore, in this systematic review, the effect of internal cooling on performance was further differentiated by the exercise protocol used (e.g., time trials vs. time to exhaustion).

Data was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Page et al., 2021). The meta-analysis was registered in PROSPERO (no. CRD42022336623).

The purpose of the present review and meta-analysis was to systematically analyze and quantify the effects of internal cooling methods on performance and physiological as well as perceptional parameters while exercising in the heat. Our main findings demonstrate that internal cooling improves physical performance and reduces overall sweat rate, core temperature and thermal sensation at the end of the exercise. These effects, however, depend on the method (physical vs. perceptional cooling) as well as the time of administration (pre-vs. mid-vs. pre- and mid-cooling).

Our main finding of the present study was that internal cooling resulted in improvements in physical performance, which is in agreement with some (Bongers et al., 2015; Zhang, 2019; Rodríguez et al., 2020) but not all (Jones et al., 2012; Ruddock et al., 2017; Choo et al., 2018) systematic reviews on the topic. We overcame this apparent discrepancy in the literature by differentiating between time to exhaustion, typically used as a measure of aerobic capacity, and time trial performance, considered a more realistic and valid measure of performance (Saris et al., 2003). Along this line, we found that time to exhaustion was significantly prolonged by internal cooling, whereas the effect of internal cooling on time trial performance was only borderline significant. The positive impact on aerobic capacity is likely linked to our finding of improved perceptional outcomes, such as a lower rate of perceived exertion. For example, in several studies in which cold water or ice ingestion resulted in a significantly longer time to exhaustion (Lee et al., 2008b; Siegel et al., 2012), the authors reported lower rates of perceived exertion during exercise, but notably not at the end of the exercise bout. Considering that subjective exertion is closely linked to the time to exhaustion (Presland et al., 2005), lower rates of perceived exertion, as seen in our analysis, likely allowed the subjects to exercise for a longer duration until exhaustion was achieved. Thermal sensation, which is an essential mediator of behavioral thermoregulation that integrates with the rate of perceived exertion as the predominant controller of the self-selected work rate of exercise (Flouris and Schlader, 2015), was also significantly reduced in our analysis, suggesting that performance improvements are likely linked to lower sensations of heat and exertion.

In addition to changes in exertion perception, physiological outcomes likely explain performance improvements. Our meta-analysis demonstrated significant reductions in core temperature and sweat rate and a borderline significant reduction in heart rate following internal cooling. Ingesting (ice-) cold beverages will lower core temperature as considerable amounts of internal heat will be absorbed, and a consequent delay in the onset of thermally induced fatigue might occur (Wegmann et al., 2012). In addition, brain temperature might be reduced (Onitsuka et al., 2018), increasing the probability of thermal sensation and performance improvements during later stages of exercise. Due to the activation of gastrointestinal thermoreceptors (Villanova et al., 1997), positive effects on the inhibitory feedback on core temperature and subsequent performance improvements might occur. However, several studies reported no differences or even reductions (Byrne et al., 2011) in core temperature, although exercise intensity was higher or exercise duration prolonged (Lee et al., 2008b; Burdon et al., 2010; Schulze et al., 2015; Takeshima et al., 2017) after internal cooling. This finding might be explained by the fact that athletes were able to perform at higher exercise intensity when applying internal cooling. Therefore, internal heat production might be greater, affecting physiological and perceptional outcomes. Since we chose to limit our analysis of physiological and perceptional outcomes to measurements taken at the end of the exercise, our results might be diluted by differences in exercise intensities or duration. We can conclude that studies showing no differences in end-exercise physiological outcomes with higher exercise intensity or duration support the positive effects of the internal cooling intervention on physiological parameters. In contrast, findings of increased core temperature at exhaustion following internal physical cooling are likely due to higher intensity or longer exercise duration in the trial (Siegel et al., 2012; Nakamura et al., 2020). When interpreting the impact of internal cooling on physiological or perceptional outcomes, it is therefore always crucial to take into account the time course and their relationship with exercise intensity and performance.

In contrast to previous systematic reviews (Ruddock et al., 2017; Choo et al., 2018), we observed a significant reduction in whole-body sweat rate following internal cooling. According to Morris et al. (Morris et al., 2016), human abdominal thermoreceptors detect intra-abdominal temperature changes, and due to their sufficient integration within the central nervous system, they can further elicit thermoeffector responses at the skin surface. A reduction of whole-body sweat rate and further evaporative heat loss from the skin might result in a lower, rather than greater, net heat loss and subsequently a greater heat storage during exercise (Morris et al., 2016). Therefore, the authors recommend to ingest beverages of any temperature, but not ice-cold drinks, during competition in hot and dry environments, where evaporative heat loss plays a greater role in total heat dissipation (Morris et al., 2016). On the other side, a lower core temperature likely reduces the sweat rate necessary for cooling (Montain et al., 1995). A lower sweat rate might further reduce the risk of dehydration. Since performance impairments might occur with sweat loss rates >2–4% of body mass (Thomas et al., 2016), the performance improvements of internal cooling might also be partially explained by lower sweat loss.

In the present study, we found only a borderline significant reduction in heart rate following internal cooling. These results concur with previous systematic reviews, which reported no internal cooling effect on heart rate (Bongers et al., 2015; Ruddock et al., 2017; Choo et al., 2018). As for other physiological and perceptional outcomes, this discrepancy might be explained by our inclusion of data collected at the end of the exercise only. Further, our results do not imply there were no positive effects of internal cooling on skin blood flow or stroke volume, as heart rate in this context is only an index of these variables (Ruddock et al., 2017). Since, in the present study, we did not include skin blood flow, an essential factor for thermoregulation, as an outcome, our analysis does not provide conclusive evidence about the underlying physiological mechanisms related to reductions in sweat and heart rate.

Physical cooling seems more effective than perceptional cooling in improving physiological parameters and physical performance when comparing cooling strategies. In agreement with another meta-analysis (Keringer et al., 2020), we found no effects of perceptional cooling on physiological outcomes, as menthol is a non-thermal cooling stimulus that acts on thermoreceptors, inducing sensations of coolness without physical reductions in body temperature (Watson et al., 1978). We further found no effects of menthol cooling on performance, which is in line with two previous (Douzi et al., 2019; Keringer et al., 2020) but in contrast with one meta-analysis (Jeffries and Waldron, 2019). The positive performance effects of perceptional cooling reported in some studies may probably be due to previously reported changes in perceptional outcomes (Jeffries and Waldron, 2019; Keringer et al., 2020), which we were also unable to demonstrate in our analysis. These perceptional effects are likely caused by an activation of cold sensors, leading to reduced thermal sensation and physiological reactions similar to physical cooling (Zheng, 2013).

The time point of application might be an essential factor in evaluating the efficacy of internal cooling. We found that cooling before and during exercise significantly reduces core temperature, which can be explained by the continuous facilitation of heat storage capacity and extended exercise duration in the heat (Siegel and Laursen, 2012). Our finding that pre- or mid-physical cooling did not reduce core temperature might be explained by the aforementioned limitation to outcome data recorded only at the end of the exercise. However, it is also possible that the effects of pre-cooling might already disappear throughout the exercise. The time point of cooling also impacted effects on perceived exertion, which was reduced only for mid-exercise cooling. Further, physical cooling during exercise seems more effective in improving aerobic performance, whereas ingestion before exercise may be more beneficial to increase aerobic capacity. Taken together, the most benefits are likely to occur when cooling before and during exercise.

Although the benefits of internal cooling on physical performance have been demonstrated in several studies and were confirmed in the present analysis, the optimal dose and time point of ingestion remain unclear. Usually, a total dose of ∼500–700 mL of ice/cold-water (∼7.5 g · kg-1), divided into smaller amounts (∼1.25 g · kg-1 every 5 min until reaching the total dose), is recommended to offer greater cooling and better tolerance (Naito et al., 2017). In the present analysis, total doses ranged from 1.25–30 g kg^-1, and positive performance effects have been reported even in studies at the lower end of the spectrum (4.0–6.8 g kg^-1) (Ihsan et al., 2010; Burdon et al., 2013; Nakamura et al., 2020). Our meta-regression failed to confirm the previously reported dose-response effect of physical cooling on performance (Zhang, 2019), as the relationship between dose and improved time trial performance SMD was only borderline significant. Regardless, the study with the largest dose (∼21 g · kg-1) had the greatest positive effect on time trial performance (Burdon et al., 2013), also indicating that even with higher doses, the positive effect of cooling may outmatch possibly negative effects of weight gain due to increased fluid intake. However, athletes should always consider that overdrinking increases the risk for hyponatremia, causing several health and performance impairments (Thomas et al., 2016). Ingestion of cold drinks might also increase voluntary fluid consumption during exercise in the heat (Mündel et al., 2006). A greater volume of cold fluid might further act as a heat sink, thereby reducing heat stress’s effects and possibly increasing the time needed to reach an exercise-limiting core temperature. Furthermore, higher voluntary fluid ingestion might reduce the risk of dehydration and might have a beneficial impact on physical performance (Thomas et al., 2016). In conclusion, further studies are needed to determine the dose-response relationship with performance and what the minimum and maximum doses for performance improvements are.

Our results support previous findings (Zhang, 2019) that physical cooling improves performance independent of environmental conditions. These results indicate that internal cooling might already be effective in neutral-warm environments (20°C–30°C). Furthermore, our results show that physical cooling might improve aerobic performance independent of the exercise duration. Our results imply that even athletes exercising with high intensity and short duration might benefit from internal physical cooling.

Considering that most studies showed no beneficial effect of internal perceptional cooling on core and skin temperature while exercise intensity is increased, one might speculate that perceptional cooling could increase the risk of heat-related illness and non-desirable cardiovascular events (Gillis et al., 2010; Barwood et al., 2020). However, other authors claim that internal perceptional cooling with menthol is an effective and safe method to improve performance without adverse effects (Keringer et al., 2020). Menthol has toxic properties, and an acceptable daily intake value of 0–4 mg kg^-1 body mass was allocated (World Health Organization, 2019). According to the FAO/WHO Expert Committee on Food Additives the highest estimated dietary exposure of menthol is ∼51 mg d^-1, estimated by the maximized survey-derived intake (MSDI) method (World Health Organization, 2019). When menthol is used as a flavoring agent (as in mouth rinsing) at current intake levels, no safety concerns are raised (World Health Organization, 2019; Barwood et al., 2020). Therefore, the safety of internal menthol application, including dosage, likely depends on whether the substance enters the human body or not. Athletes wishing to use menthol cooling should be familiar with safe intake protocols, which should be well-practiced prior to competitive use (Barwood et al., 2020). However, the lack of beneficial effects on performance and the risk of adverse side effects (low, but possible) imply that menthol mouth rinsing should not be applied until further evidence is available.

Our research highlights that internal cooling has the potential to improve endurance performance and selected physiological and perceptional parameters. However, its effectiveness depends on the method used (i.e., physical vs. perceptional cooling) as well as the time of administration. Our results suggest that physical cooling is more effective for performance improvements than perceptional cooling, although the number of studies assessing the effects of perceptional cooling was low. Further studies are needed to formulate safe intake recommendations and evaluate possible side effects of internal cooling. In addition, more studies are required to assess the impact of internal cooling on exercise performance rather than exercise capacity. Future research should confirm the laboratory-based results in the field setting and involve a more inclusive study demographic with regard to sex and exercise type.