Exertional heat stress typically occurs with the combination of physical activity and high ambient air temperature and relative humidity. This can result in core temperatures of 39.5°C or higher (27) leading to an increased risk of heat related injury. If not recognized and treated immediately, heat injuries can potentially result in loss of consciousness, severe tissue damage, organ failure, and even death (12). The severity of the illness is directly related to the duration and magnitude of core temperature elevation (24). Therefore, rapid cooling of hyperthermic individuals to a near-normal resting core temperature should be the main goal of any treatment strategy (3,6,8,20,24).
Current guidelines for the treatment of hyperthermic individuals state that any treatment modality should achieve a rectal cooling rate of at least 0.10°C·min−1 (6,25), which is most effectively achieved by immersion in cold water ranging from 2 to 10°C (6). However, these guidelines do not take into account variations in physical characteristics between individuals. Particularly in regards to males and females, differences in body size, shape, and composition can influence the rate of whole-body cooling during water immersion (2,14,17,18). From a biophysical point of view, if two objects of similar mass and shape are heated to the same temperature and then immersed in water of a given temperature, then both objects would cool at a similar rate. However, it is well known that males and females typically differ in body size and composition. Males typically have a greater body mass (mostly because of a greater lean body mass) and a lower percentage of body fat, whereas females typically have a greater body surface area-to-mass ratio. These physical determinants have all been shown to influence the rate of core cooling when normothermic individuals are immersed in water (1). However, it remains unclear how these physical characteristics may influence core cooling rates in individuals rendered hyperthermic.
Exercise-induced hyperthermia is associated with a thermal status (i.e., elevated core and/or muscle temperatures) and a tissue blood distribution (elevated skin blood flow), which may reduce the relative influence of tissue insulation and physical characteristics on core cooling rates during the water immersion period. For example, in normothermic individuals, cold-induced vasoconstriction combined with high levels of tissue insulation (i.e., subcutaneous fat) is inversely related to the rate of core temperature decline for a given cold stress (4,23). However, Lemire et al. (16) reported that a ∼10% difference in whole-body adiposity between hyperthermic low- and high-fat males did not significantly influence rectal cooling rates. This was explained by high levels of postexercise skin blood flow owing to hyperthermia causing heat to bypass the subcutaneous fat layers during cold water immersion thus reducing the insulative effect of greater body adiposity.
To the best of our knowledge, studies examining differences between sexes in core cooling rates during cold water immersion have not been evaluated in previously hyperthermic individuals. The purpose of this study was to evaluate possible differences between sexes in rectal cooling rates during cold water immersion (2°C) after exercise-induced hyperthermia. To eliminate some possible confounding influences because of the differences in physical characteristics between sexes, participants were matched for body surface area-to-mass ratio. Because males and females were matched for body surface area-to-mass ratio, and modest variations in body adiposity do not influence rectal cooling rates after exercise-induced hyperthermia, the only physical difference between males and females that could modify core cooling rate would be lean body mass. Considering that lean body mass has been shown to be important in contributing to whole-body insulation during cold water immersion of normothermic individuals, we hypothesized that rectal cooling rates in previously hyperthermic females would be greater because of a smaller lean body mass.
After approval of the experimental protocol from the University of Ottawa Research Ethics Committee, 19 healthy and physically active participants (10 males and 9 females) volunteered and gave written consent to participate in this study. Participants were matched for body surface area-to-mass ratio, which was calculated from the measurements of height and weight (7). As such, each participant was matched with a similar participant of the opposite sex. Five to 7 days before the experimental trial, maximum oxygen consumption was measured during a progressive treadmill running protocol, and the data were used to select the submaximal workload for the experimental exercise phase of the study. Body density was also measured using the hydrostatic weighing technique, and body fat percentage was then calculated using the equation of Siri (22). Females were tested in the follicular phase of their menstrual cycle. The physical characteristics of the participants are presented in Table 1.
Rectal temperature was measured using a pediatric thermocouple probe (Mon-a-therm General Purpose Temperature Probe; Mallinckrodt Medical, St. Louis, MO) inserted to a minimum of 12 cm past the anal sphincter. Skin temperature was monitored at 12 sites by Type T thermocouples integrated into heat flow sensors (Concept Engineering, Old Saybrook, CT). These are commercially available heat flow disks that can read heat flux and skin temperature. Because the head and the chest were not entirely immersed in water, the area-weighted mean skin temperature and mean nonevaporative heat loss were calculated by assigning the following regional percentages: upper back 12%, lower back 12.5%, abdomen 12.5%, bicep 9.5%, forearm 9.5%, hand 2%, quadricep 12%, hamstring 12%, front calf 9%, and back calf 9% (11,15). All temperature data were collected and digitized (IBM ThinkCentre M50, New York, NY) with LabVIEW software (Version 7.0; National Instruments, Austin, TX) at 5-s intervals, displayed graphically on a computer screen, and recorded in spreadsheet format on a hard disk (IBM ThinkCentre M50) with LabVIEW software (Version 7.0; National Instruments).
Oxygen consumption was continuously recorded during the immersion period with the portable Jaeger-Oxycon mobile system (VIASYS Healthcare, Hoechberg, Germany). Rate of metabolic heat production was subsequently calculated from minute-average values of oxygen consumption and RER (19).
Each participant took part in one experimental trial. All trials were performed at the same time of day to avoid circadian variations in skin and rectal temperatures. Participants were asked to fast at least 3 h before experimentation and were given instructions on what the last meal should be (i.e., two slices of brown toasts with butter and orange juice). Water ingestion was permitted ad libitum during this time to ensure euhydration. Upon arrival at the laboratory, the participants clothed in shorts, and athletic shoes were fitted with the appropriate instruments.
After instrumentation, the participants remained resting in the upright seated posture for 15 min at an ambient air temperature of ∼25°C and a relative humidity of ∼35%. They then entered a temperature-controlled chamber regulated at an ambient temperature of 40.0°C and a relative humidity of ∼18%, where they ran on a treadmill at 65% of their predetermined maximum oxygen consumption until rectal temperature reached 39.5°C. Participants were then transferred (∼1.5 min) and immersed in a recumbent position in a circulated water bath (Jacuzzi, J-315; Advanced Spas, Cincinnati, OH) maintained at 2.0°C. Before entering the circulated water bath, participants were fitted with neoprene mitts and socks to reduce discomfort. Participants remained in the water until rectal temperature reached 37.5°C.
Physical characteristics between sexes were compared using independent sample t-tests. For each trial, baseline resting and end-exercise temperatures, exercise duration, cooling times, time to reach (in min) a rectal temperature of 39°C, 38.5°C, 38°C, and 37.5°C, and the overall rectal cooling rates (start to end of immersion) were calculated and analyzed using independent sample t-tests. Area-weighted mean skin temperature and nonevaporative heat loss as well as rate of metabolic heat production were also calculated and analyzed using a two-way mixed-model ANOVA during the immersion period using the repeatable factor of time (levels: 0, 1, 2, 3, 4, and 5 min until the first participant was out of the water) and the nonrepeatable factor of sex (levels: male and female). In addition, correlations between overall rectal cooling rates and body surface area as well as body surface area-to-lean body mass ratio were analyzed. The level of significance was set at 0.05, and the alpha level was adjusted during multiple comparisons to maintain the rate of Type I error at 5% during the Holm-Bonferroni post hoc analysis. The data are presented as mean ± SD, unless otherwise indicated. All analyses were performed using the statistical software package SPSS 15.0 for Windows (SPSS, Inc., Chicago, IL).
As participants were matched for body surface area-to-mass ratio, there were no differences between sexes (P = 0.337). However, there were significant differences in body fat percent (P < 0.001), lean body mass (P < 0.001), body surface area (P = 0.008), body surface area-to-lean body mass ratio (P < 0.001), and relative maximum oxygen consumption (P < 0.001) between sexes (Table 1).
There were no significant differences between groups in baseline rectal temperature (P = 0.482). Exercise time taken to reach the experimental withdrawal criterion (i.e., 39.5°C) was not different between groups (35.21 ± 7.48 and 34.17 ± 7.29 min for females and males, respectively, P = 0.763). Exercise resulted in a similar increase from baseline in rectal temperature of 2.24 ± 0.23°C and 2.15 ± 0.28°C (P = 0.461) for the female and the male groups, respectively. Furthermore, rectal temperature was similar between sexes at the end of exercise (P = 0.898). Finally, mean skin temperature was similar (P = 0.505) at the end of exercise between groups (Table 2).
Cold water immersion.
After the transfer from the thermal chamber to the water bath, rectal (P = 0.200) and mean skin (P = 0. 575) temperatures did not differ between groups before the start of immersion. Immersion times were significantly different between groups (10.89 ± 4.49 and 18.13 ± 4.47 min for females and males, respectively, P = 0.003). The times taken to reach a rectal temperature of 39.0°C (P = 0.178) and 38.5°C (P = 0.097) were not different between groups, but the times to reach a rectal temperature of 38.0°C (P = 0.035) and 37.5°C (P = 0.003) were significantly greater in males (Fig. 1). In addition, significant differences occurred in the overall rectal cooling rates (P = 0.001; Table 3).
Decreases in mean skin temperature at the beginning of immersion became gradually less as a function of time (P < 0.001) and were significantly different between males and females from the first minute to the end of the mean immersion period for females (P = 0.001; Fig. 2A). Similarly, decreases in the rate of nonevaporative heat loss at the beginning of immersion became gradually less as a function of time (P < 0.001) but were not significantly different between males and females for the duration of the immersion period (P = 0.180; Fig. 2B). Furthermore, decreases in the rate of metabolic heat production became less as a function of time (P < 0.001), but no difference was observed between sexes up to the last common point of immersion time for females and males (P = 0.307). It is worth noting that the rate of metabolic heat production was gradually elevated in males after the 10th min of immersion (Fig. 3).
Finally, when pooling data from both sexes, the overall rectal cooling rates correlated significantly with body surface area-to-lean body mass ratio (r = 0.70, P = 0.001; Fig. 4A). In contrast, the overall rectal cooling rates were not significantly correlated with the absolute body surface area (r = −0.29, P = 0.235; Fig. 4B).
The main finding of this study was that the rectal temperature of previously hyperthermic females cooled ∼1.7 times faster than males when immersed in a 2°C circulated water bath despite males and females being matched for body surface area-to-mass ratio. This was evidenced by greater rectal cooling rates in females and by a shorter time for females to reach a rectal temperature of 38°C and 37.5°C. Females had a lower mean skin temperature throughout the immersion period, although both groups had a similar temporal pattern for the rate of nonevaporative heat loss. Similar responses were seen in the rate of metabolic heat production between both groups until the mean end immersion time of females was reached, after which it increased gradually for males.
During cold water immersion, the primary avenue for heat loss is via nonevaporative mechanisms (i.e., conduction and/or convection). Heat transfer must occur through layers of muscle, adipose tissue and skin to the water. The rate of nonevaporative heat loss is dependent on 1) the thermal conductivity of the mass of tissue (i.e., muscle, adipose tissue, and skin) and 2) the surface area across which heat is exchanged. In this manner, it is the amount of heat loss relative to the amount of heat produced which is the determinate of the body surface area-to-mass ratio. Typically, females have a greater body surface area-to-mass ratio and, therefore, cool at a quicker rate than males during normothermia (17). Sloan and Keatinge (23) have shown that the cooling of rectal temperature of males and females immersed in 20°C water was primarily the result of differences in body surface area-to-mass ratio between sexes. In contrast, our data suggest that body surface area-to-mass ratio does not primarily determine rectal cooling rates of hyperthermic individuals because females cooled faster than males despite having similar body surface area-to-mass ratios. Furthermore, females demonstrated a greater overall rectal cooling rate despite having a ∼10% greater percentage of body fat compared with males. Although a greater fat mass could possibly explain the lower skin temperature for females during the immersion period (9), our results suggest that greater levels of adipose tissue insulation, in particular, subcutaneous fat, have little effect in reducing the rate of heat transfer in previously hyperthermic females. This is in accordance with the findings of Lemire et al. (16) who concluded that differences in body adiposity (within a range of 8%-29%) had no significant effect on the rate of overall core cooling rate in previously hyperthermic males.
Despite the fact that 1) our participants were matched for body surface area-to-mass ratio, 2) greater levels of body adiposity do not influence rectal cooling rates after hyperthermia, and 3) the rates of nonevaporative heat loss and metabolic heat production were not different between males and females, females did have greater overall rectal cooling rates. In the present study, the nonevaporative heat loss data are corrected for differences in body surface area (i.e., W·m−2 of body surface area). Therefore, females actually had lower absolute rates of nonevaporative heat loss (in W) because they had a smaller body surface area (Table 1). The fact that females cooled at a faster rate compared with males despite losing less heat (on an absolute scale) may seem contradictory. However, the fact that both groups differed greatly in body size and composition resulted in a faster decrease in rectal temperature for females. Therefore, differences in rectal cooling rates may possibly be explained by differences in lean body mass. In fact, it is generally accepted that, when normothermic individuals are immersed in water, core cooling rate is influenced by lean body mass. Furthermore, differences in response to cold stress between sexes have been suggested to be caused by the males' greater lean body mass (2). This said, differences between sexes are thought to be greatly reduced if lean body mass is taken into account (2).
Because body adiposity does not influence rectal cooling rate in previously hyperthermic individuals, lean body mass remains the main determinant of heat transfer to the water (28). Because the rate of heat transfer is greatly dependent on the surface across which it is exchanged, then it may be better to express the body surface area-to-mass ratio by removing body adiposity and accounting for differences in lean body mass, thus giving the body surface area-to-lean body mass ratio. We therefore examined a female and a male from each group that had very similar body surface area-to-lean body mass ratios (322 cm2·kg−1 of lean body mass for the male vs 323 cm2·kg−1 of lean body mass for the female). Both participants had similar rectal cooling rates (0.11 vs 0.09°C·min−1 for the male and female, respectively) during the immersion period, which resulted in a comparable total time of immersion (19.5 min for the male vs 21.5 min for the female). To better examine the relation between rectal cooling rates and body surface area-to-lean body mass ratio, we correlated both variables independently of sex (Fig. 4A) and observed a good correlation (r = 0.70). The importance of lean body mass is also demonstrated by the fact that a weaker correlation (r = −0.29) was observed when correlating only the absolute body surface area with rectal cooling rates (Fig. 4B). Tikuisis et al. (26) previously showed similar results between rectal cooling rates and body surface area (m2)-to-volume (m3) ratio, reporting a correlation of r = 0.42. Taking into consideration that our data does not reflect a full range of possible body surface area-to-lean body mass ratios, we suggest that this ratio may possibly explain core cooling variations in hyperthermic individuals, but future studies are required to demonstrate that this relationship holds true with individuals (for either males or females) across a wide spectrum of body surface area-to-lean body mass ratios.
Although physical differences between males and females seem to explain most of the variation in the observed rectal cooling rates, some physiological responses may have also influenced our results. During exercise, a large proportion of blood, which constitutes approximately two thirds of visceral mass at rest, is redistributed to the skin and working muscles (21). During postexercise recovery, a significant portion of this blood remains pooled in previously active musculature (10). Kenny and Jay (13) showed that females have a greater and more prolonged elevation in core temperature in active and inactive muscle temperatures after dynamic exercise. This was explained by a greater decrease in mean arterial blood pressure caused by an increase in blood pooling in the previously active limbs in females (13). This apparent hyperemia may cause a greater thermal conductivity of the limbs, which could provide a greater rate of nonevaporative heat loss in females. Although possible, we did not observe any differences in the rate of nonevaporative heat loss between males and females.
McArdle et al. (17) observed similar rates of metabolic heat production in both males and females during water immersion over the first 1°C decrease in rectal temperature. However, they showed that males have a greater metabolic response than females after the first 1°C decrease in rectal temperature. Similarly, we show no differences in metabolic response between males and females during the first degree of cooling (i.e., until rectal temperature reached 38.5°C). Further, females exited the cold water at approximately the 11th min mark corresponding to an exit rectal temperature of 37.5°C and a 2°C decrease from end-exercise values. However, no significant increase in metabolic heat production was observed. In contrast, males showed an increase in metabolic heat production during this same period albeit rectal temperature was greater (Fig. 4). Metabolic heat production remained elevated until they exited the water bath at approximately the 18th min of immersion. Hence, apparently hyperthermic males experience an increase in metabolic heat production at a greater rectal temperature compared with females during cold water immersion. This may be explained by the fact that males have a greater amount of lean body mass, which is known to correlate with shivering intensity (17). Furthermore, it has been reported that males have a heightened sensitivity to changes in skin temperature and a resultant increase in the core temperature at which onset of shivering occurs (9). It is plausible therefore that a higher onset threshold combined with a greater shivering intensity in males may have contributed to the slower rectal cooling rate.
Proulx et al. (20) reported an average rectal cooling rate of 0.35°C·min−1 in 2°C water and 0.19°C·min−1 in 8°C water. Our results show similar average rectal cooling rates of 0.22 and 0.12°C·min−1 for females and males, respectively. The average of our results (0.17°C·min−1) is comparable with previous studies (5). The rectal cooling rates for the males and females in this study agree with the current guideline stating that an effective treatment of exertional heat stress should provide a rectal cooling rate of at least 0.10°C·min−1 (6,25). This suggests that no changes to the proposed standards are required for males and females. However, it should be considered that immersion times will differ between sexes.
Despite being matched for body surface area-to-mass ratio and having a greater body adiposity, previously hyperthermic females cooled ∼1.7 times faster compared with males while immersed in 2°C water. Because hyperthermic individuals have a considerable reduction in adipose tissue insulation (because of elevated skin blood flow) and muscle tissue insulation (because of an elevated muscle perfusion), our findings suggest that differences between sexes in lean body mass were mainly responsible for the observed differences in rectal cooling rates between hyperthermic males and females.
This research was supported by the Natural Sciences and Engineering Research Council of Canada (Grant # RGPIN-298159-2004, held by Dr. Glen P. Kenny). Dr. Glen P. Kenny was supported by a University of Ottawa Research Chair Award. The authors thank all the participants who volunteered for this study. The provision of financial support does not in any way infer or imply endorsement of the research findings by either agency. The results of the present study do not constitute endorsement by ACSM.
1. Anderson GS. Human morphology and temperature regulation. Int J Biometerol
2. Anderson GS, Ward R, Mekjavic IB. Gender differences in physiological reactions to thermal stress. Eur J Appl Physiol
3. Armstrong LE, Casa DJ, Millard-Stafford M, Moran DS, Pyne SW, Roberts WO. American College of Sports Medicine position stand. Exertional heat illness during training and competition. Med Sci Sports Exerc
4. Baker PT, Daniels F Jr. Relationship between skinfold thickness and body cooling for two hours at 15 degrees C. J Appl Physiol
5. Casa DJ, Armstrong LE, Ganio MS, Yeargin SW. Exertional heat stroke in competitive athletes. Curr Sports Med Rev
6. Casa DJ, McDermott BP, Lee EC, Yeargin SW, Armstrong LE, Maresh CM. Cold water immersion
: the gold standard for exertional heatstroke treatment. Exerc Sports Sci Rev
7. Dubois D, Dubois E. Clinical calorimetry. A formula to estimate the approximate surface area if height and weight be known. Arch Int Med
8. Gaffin SL, Gardner JW, Flinn SD. Cooling methods for heatstroke victims. Ann Intern Med
9. Graham TE. Thermal, metabolic, and cardiovascular changes in men and women during cold stress. Med Sci Sport Exerc
. 1988;20(5 suppl):S185-92.
10. Halliwill JR. Mechanisms and clinical implications of post-exercise hypotension in humans. Exerc Sports Sci Rev
11. Hardy JD, Dubois EF. The technique of measuring radiation and convection. J Nutr
12. Howe AS, Boden BP. Heat-related illness in athletes. Am J Sport Med
13. Kenny GP, Jay O. Sex difference in postexercise esophageal and tissue temperature response. Am J Physiol Regul Integr Comp Physiol
14. Kollias J, Barlett L, Bergsteinova V, Skinner J, Buskirk ER, Nicholas W. Metabolic and thermal responses of women during cooling in water. J Appl Physiol
15. Layton RP, Mints WH, Annis JF, Rack MJ, Webb P. Calorimetry with heat flux transducers: comparison with a suit calorimeter. J Appl Physiol
16. Lemire B, Gagnon D, Jay O, Dorman L, Ducharme MB, Kenny GP. Influence of adiposity on cooling efficiency in hyperthermic individuals. Eur J Appl Physiol
17. McArdle WD, Magel JR, Gergley TJ, Spina RJ, Toner MM. Thermal adjustment to cold-water exposure in resting men and women. J Appl Physiol
18. McArdle WD, Toner MM, Magel JR, Spina RJ, Pandolf KB. Thermal responses of men and women during cold-water immersion: influence of exercise intensity. Eur J Appl Physiol
19. Nishi Y. Measurement of thermal balance in man. In: Cena K, Clark J, eds. Bioengineering, Thermal Physiology and Comfort
. New York (NY): Elsevier; 1981. p. 29-39.
20. Proulx CI, Ducharme MB, Kenny GP. Effect of water temperature on cooling efficiency during hyperthermia in humans. J Appl Physiol
21. Rowell LB. Human cardiovascular adjustments to exercise and thermal stress. Physiol Rev
22. Siri WE. The gross composition of the body. Adv Biol Med Phys
23. Sloan RE, Keatinge WR. Cooling rates of young people swimming in cold water. J Appl Physiol
24. Smith JE. Cooling methods used in the treatment of exertional heat illness. Br J Sport Med
25. Tek D, Olshaker JS. Heat illness. Emerg Med Clin North Am
26. Tikuisis P, Jacobs I, Moroz D, Vallerand AL, Martineau L. Comparison of thermoregulatory responses between men and women immersed in cold water. J Appl Physiol
27. Yaqub B, Al Deeb S. Heat strokes: aetiopathogenesis, neurological characteristics, treatment and outcome. J Neurol Sci
28. Xu X, Castellani JW, Santee W, Kolka M. Thermal responses for men with different fat compositions during immersion in cold water at two depths: prediction versus observation. Eur J Appl Physiol
Keywords:© 2009 American College of Sports Medicine
COLD WATER IMMERSION; CORE TEMPERATURE; HEAT STRESS; RECOVERY