Unravelling the true influences of fitness and sex on sweating during exercise

Even at moderate environmental temperatures (∼25°C), physical activity imposes a thermoregulatory stress, by virtue of the internal heat generated as a byproduct of cellular metabolism. This elevated rate of metabolic heat production during exercise requires a parallel increase of heat dissipation from the skin, primarily from an increase in skin blood flow and the evaporation of sweat, to avoid critical levels of physiological heat strain. For the past several decades, researchers have endeavoured to develop a better understanding of the various factors that contribute to the large interindividual variation commonly observed in the human heat stress response. Historically, aerobic fitness and sex have been considered to be two influential determinants of the changes in core temperature and sweat rate during exercise. In 1966, a seminal study by Saltin & Hermansen concluded that the core temperature during exercise was closely related to the relative exercise intensity (i.e. the percentage of maximal oxygen uptake, ); the prevailing wisdom was that individuals with a greater aerobic capacity were able to shed a proportionally greater amount of heat to balance their higher rate of metabolic heat production at a given , resulting in the same core temperature. This notion has apparently been supported by the observation of greater sweat rates in fitter individuals exercising at a fixed on numerous occasions. It follows that many studies over the past few decades comparing thermoregulatory sweating between independent groups have consequently administered exercise using a fixed , while often also matching groups for aerobic capacity, in an attempt to isolate the influence of the between-group factor under investigation from the apparently confounding effects of aerobic capacity. Using this experimental approach, female subjects have regularly demonstrated significantly lower sweat rates relative to fitness-matched male subjects exercising at the same , leading to the common belief that sex independently alters thermoregulatory sweating. However, when viewed from a heat balance perspective, an alternative picture is painted. It is widely acknowledged that the effective goal of the human themoregulatory system is to attain a steady-state core temperature; and it stands to reason that such a scenario can be achieved only when the prevailing rate of metabolic heat production is balanced by an equally elevated rate of net heat dissipation to the surrounding environment. The increase in heat production during exercise is mostly determined by the change in absolute rate of oxygen consumption (; in litres of oxygen per minute), whereas the increase in heat dissipation occurs mainly due to an elevated rate of sweat evaporation from the skin. The whole-body sweat rate during exercise is therefore governed by the rate of evaporation required to balance the rate of heat production, which itself is determined by absolute . However, an aerobically fit person exercising at a fixed will consume more oxygen and thus generate more heat, logically leading to the question: do fit individuals still demonstrate greater sweat rates when exercising at the same heat production as their less fit counterparts? Following this rationale, Cramer et al. (2012) reported that, contrary to conventional wisdom, there is no difference in whole-body sweat rate between an aerobically fit group (∼60 ml O2 kg-1 min-1) and a relatively unfit group (∼40 ml O2 kg-1 min-1) during exercise at a fixed heat production of ∼275 W m-2 despite large differences in . Furthermore, the only fitness-related differences in local sweat rate were observed on the forehead, where sweating was in fact lower, not higher, in individuals with a greater aerobic capacity. The proposed mechanism for this response was not a partial heat acclimatization effect associated with aerobic training, but rather a greater sympatho-adrenal activity associated with a greater cardiovascular, perceptual and ventilatory strain in the less fit individuals at a fixed heat production, possibly eliciting a greater sweat production in glaborous skin areas only because no parallel differences in local sweating were observed on the non-glaborous skin of the forearm (Cramer et al. 2012). Together with another study published a year earlier (Jay et al. 2011), these findings led to the conclusion that future studies wishing to compare whole-body sweat rate and the local sweat rate of non-glaborous local skin areas during exercise between independent groups should administer an exercise intensity that elicits the same metabolic heat production irrespective of . It was also concluded that some previous studies that employed a approach might have incorrectly attributed differences in whole-body and/or local sweating to the between-group factor that was thought to be isolated when divergent rates of metabolic heat production between groups may instead have been the culprit. Such a scenario could have occurred even between groups matched for aerobic capacity but different in body mass, because dissimilar rates of heat production would still have arisen at a fixed . For example, the classic study of Horstman & Christensen (1982) reported lower sweat rates in females during exercise at 40 in a hot environment in comparison to males with a similar aerobic capacity (47 versus 51 ml O2 kg-1 min-1). However, the lower body mass of the female participants (67 versus 77 kg) would have meant that they consumed less oxygen [0.40(47 × 67) = 1.26 l O2 min-1] than the males participating in the study [0.40(51 × 77) = 1.57 l O2 min-1]. The lower resultant rate of heat production in females would therefore have necessitated a smaller amount of evaporation from the skin and subsequently a lower sweat production to satisfy biophysical heat balance requirements. Later in 2012, Gagnon & Kenny (2012) reassessed the influence of sex on thermoregulatory sweating during exercise by comparing male and female subjects working at the same absolute levels and therefore the same rates of metabolic heat production. They demonstrated, for the first time, that whole-body and local sweating rates on the chest, forearm and back were not different as a function of sex at fixed rates of heat production ≤250 W m-2. However, they also found that at a fixed heat production of 300 W m-2, sweating was lower in females due to a lower maximal evaporative capacity secondary to a smaller maximal output of the individual sweat glands (Gagnon & Kenny, 2012). Their data also illustrate nicely the importance of assessing between-group differences in sweating using a heat balance perspective as opposed to the conventional approach. As a result of circumstantial differences in both aerobic capacity and body size between the male and female participants in their study, 250 W m-2 of heat production in the male group yielded a similar relative intensity (∼55 ) to 200 W m-2 in the female group but different sweat rates due to dissimilar evaporative heat balance requirements between sexes. It follows that had exercise intensity been set to elicit the same , sex-related differences in sweating would probably have been reported incorrectly at all exercise intensities when sex per se exerts an influence only at higher rates of heat production. Collectively, these studies provide an improved methodology for physiologists wanting to isolate the independent influence of physiological factors (e.g. injury, disease, factors disrupting skin blood flow) on the sudomotor response to exercise using a between-subject design. Firstly, matching independent groups for aerobic capacity does not seem necessary, and exercise should be administered to elicit the same metabolic heat production, irrespective of . By focusing on metabolic heat production, net mechanical efficiency can also be accounted for in the exercise stimulus. However, when using this approach the local sweat rate responses of glaborous skin regions such as the forehead should be treated with caution. Secondly, a mixture of male and female subjects may be used within each group only if employing exercise intensities that elicit rates of heat production of 250 W m-2 or lower. Readers are invited to give their opinion on this article. To submit a comment, go to: http://ep.physoc.org/letters/submit/expphysiol;99/10/1265 None declared. None declared.