The independent influence of peak oxygen uptake (V̇o2 peak) on changes in thermoregulatory responses during exercise in a neutral climate has not been previously isolated because of complex interactions between V̇o2 peak, metabolic heat production (Hprod), body mass, and body surface area (BSA). It was hypothesized that V̇o2 peak does not independently alter changes in core temperature and sweating during exercise. Fourteen males, 7 high (HI) V̇o2 peak: 60.1 ± 4.5 ml·kg−1·min−1; 7 low (LO) V̇o2 peak: 40.3 ± 2.9 ml·kg−1·min−1 matched for body mass (HI: 78.2 ± 6.1 kg; LO: 78.7 ± 7.1 kg) and BSA (HI: 1.97 ± 0.08 m2; LO: 1.94 ± 0.08 m2), cycled for 60-min at 1) a fixed heat production (FHP trial) and 2) a relative exercise intensity of 60% V̇o2 peak (REL trial) at 24.8 ± 0.6°C, 26 ± 10% RH. In the FHP trial, Hprod was similar between the HI (542 ± 38 W, 7.0 ± 0.6 W/kg or 275 ± 25 W/m2) and LO (535 ± 39 W, 6.9 ± 0.9 W/kg or 277 ± 29 W/m2) groups, while changes in rectal (Tre: HI: 0.87 ± 0.15°C, LO: 0.87 ± 0.18°C, P = 1.00) and aural canal (Tau: HI: 0.70 ± 0.12°C, LO: 0.74 ± 0.21°C, P = 0.65) temperature, whole-body sweat loss (WBSL) (HI: 434 ± 80 ml, LO: 440 ± 41 ml; P = 0.86), and steady-state local sweating (LSRback) (P = 0.40) were all similar despite relative exercise intensity being different (HI: 39.7 ± 4.2%, LO: 57.6 ± 8.0% V̇o2 peak; P = 0.001). At 60% V̇o2 peak, Hprod was greater in the HI (834 ± 77 W, 10.7 ± 1.3 W/kg or 423 ± 44 W/m2) compared with LO (600 ± 90 W, 7.7 ± 1.4 W/kg or 310 ± 50 W/m2) group (all P < 0.001), as were changes in Tre (HI: 1.43 ± 0.28°C, LO: 0.89 ± 0.19°C; P = 0.001) and Tau (HI: 1.11 ± 0.21°C, LO: 0.66 ± 0.14°C; P < 0.001), and WBSL between 0 and 15, 15 and 30, 30 and 45, and 45 and 60 min (all P < 0.01), and LSRback (P = 0.02). The absolute esophageal temperature (Tes) onset for sudomotor activity was ∼0.3°C lower (P < 0.05) in the HI group, but the change in Tes from preexercise values before sweating onset was similar between groups. Sudomotor thermosensitivity during exercise were similar in both FHP (P = 0.22) and REL (P = 0.77) trials. In conclusion, changes in core temperature and sweating during exercise in a neutral climate are determined by Hprod, mass, and BSA, not V̇o2 peak.
- heat stress
- V̇o2 max
- thermal modeling
exercise physiologists evaluating changes in core temperature and thermoregulatory sweating in populations, such as the obese (11, 13), children (15, 42), burn patients (35), different ethnic groups (50), and those with health disorders (10, 47) or spinal cord injuries (22) use an independent group experimental design to compare responses to a reference. Common practice is to match groups for peak oxygen uptake (V̇o2 peak) and/or administer exercise using relative workloads (% V̇o2 peak) (11, 22, 35).
These procedures are grounded in the widely held notion that differences in V̇o2 peak profoundly influence thermoregulatory responses during exercise in physiologically compensable environments (5, 16, 18, 26, 32). In 1966, Saltin and Hermansen (43) reported exercise at the same % of V̇o2 peak at 22°C yields similar core temperatures, irrespective of aerobic capacity. These findings were apparently confirmed in 2000 by Fritzche and Coyle (16) and in 2004 by Gant et al. (18) in a similar climate (22–24°C). Greenhaff (23) also reported smaller core temperature changes at the same absolute workload (and metabolic heat production) in fitter individuals exercising at 23°C. In addition, greater whole body and local sweat rates are regularly reported with increasing fitness at a fixed % V̇o2 peak in neutral to warm climates (30, 31, 43, 49). The prevailing logic is that despite the greater heat production of fitter people at a given % V̇o2 peak, their heat loss mechanisms are proportionally greater, resulting in similar heat storage and body temperatures, as their less fit counterparts under conditions where heat loss is not limited by climate (26).
However, changes in core temperature among individuals of different fitness vary by as much as 1.6°C (26) at a given % V̇o2 peak, and by 1.0°C (33) at a similar metabolic rate. The training groups of Fritzche and Coyle (16) differed substantially in body mass and body surface area (BSA), while Gant et al. (18) compared two relatively fit groups of dissimilar mass during weight-bearing exercise. Furthermore, the negative correlation of rectal temperature (Tre) with V̇o2 peak at an absolute workload reported by Havenith et al. (26) also coincided with a negative correlation of Tre with body mass. Because V̇o2 peak and body mass were themselves strongly correlated, the effects of mass and fitness on Tre could not be fully dissociated (26). Finally, the greater sweat rates previously reported in fitter individuals at a % V̇o2 peak in physiologically compensable climates (30, 31, 43, 49) can be explained by the fact that a greater rate of evaporation would be required to balance their higher metabolic heat production (17).
It is clear that previous studies have been unable to identify the independent influence of V̇o2 peak on changes in core temperature and sweating because of complex interactions between V̇o2 peak, body mass, BSA, and metabolic heat production within their data, with these factors were only partially dissociated using a multiple regression approach (27, 28). The only exception (48) used a repeated-measures design to compare thermoregulatory responses during exercise at a fixed heat production before and after 8 wk of aerobic training, which elicited a ∼10% increase in V̇o2 peak but no changes in body mass or BSA. Similar changes in core temperature, local sweat rate, and whole body evaporation during exercise were reported both pretraining and posttraining. However, the training-induced increase in V̇o2 peak was less than 5 ml·kg−1·min−1, and a greater difference in fitness may be necessary to observe differences in thermoregulatory responses (48). A repeated-measures training study is unlikely to yield sufficiently large changes in V̇o2 peak without a greater training stimulus and concurrent reductions in body mass. The optimal solution is, therefore, to select two independent groups matched for body mass, BSA, and sex, and similar in age, acclimation status, and ethnicity, but differing greatly in V̇o2 peak (i.e., by ∼20 ml·kg−1·min−1), with both groups exercising at the same absolute heat production in a physiologically compensable (neutral) climate. If V̇o2 peak does not truly influence changes in thermoregulatory responses during exercise, the robustness of this finding can be tested further by comparing the same participants at a fixed % of their V̇o2 peak. Under this scenario, results should contradict previous studies that have stated that relative exercise intensity determines core temperature, irrespective of fitness level (1, 16, 29, 43), with much greater core temperature changes in the high V̇o2 peak group by virtue of their higher metabolic heat production, despite greater sweat rates.
The purpose of the present study was to determine whether large differences in V̇o2 peak independently alter changes in core temperature and sweating during exercise in a neutral climate. It was hypothesized that after matching high and low V̇o2 peak participants for similar body mass, BSA, age, acclimation status, sex, and ethnicity, changes in core temperature and sweating during exercise would be 1) the same, irrespective of V̇o2 peak, at a fixed metabolic heat production (540 W), despite distinctly different relative workloads between groups and 2) significantly greater in the high V̇o2 peak group at the same relative workload (60% V̇o2 peak). The findings would provide very strong evidence that contrary to conventional thinking, large differences in V̇o2 peak between participants do not independently alter changes in thermoregulatory responses during exercise in a neutral climate.
A sample size was determined using a power calculation (β = 0.9, α = 0.05). The effect size of 15%, and the standard deviations of 8% for this calculation were estimated from the changes in core temperature for fit and unfit populations exercising at a relative workload in the existing literature (29, 49). A minimum sample size of six for each of the two V̇o2 peak groups (12 in total) was needed to express our results at a 95% level of confidence. After approval of the experimental protocol from the University of Ottawa Research Ethics Committee and following receipt of written informed consent, 14 [7 low V̇o2 peak: ∼40 mlO2·kg−1·min−1 (LO group); and 7 high VO2peak: ∼60 mlO2·kg−1·min−1 (HI group)] healthy, nonsmoking, and normotensive Caucasian male participants volunteered for the study consisting of 1 preliminary trial and two experimental trials [1), a fixed heat production of 540 W (FHP trial) and 2), 60% of V̇o2 peak (REL trial)]. During the preliminary trial, total body mass, height, and maximal oxygen consumption were measured. Body surface area (BSA) was calculated from the measurements of height and weight, according to DuBois and DuBois (12). Peak oxygen consumption (V̇o2 peak) was measured using a semirecumbent cycle ergometer protocol consisting of a 2-min warmup at 40 W followed by a 20 W increase every minute until physical exhaustion, in accordance with recommendations by the Canadian Society of Exercise Physiology (8). Participants in the HI and LO groups were also specifically selected for similar total body mass, BSA, and age (Table 1). Body fat percentage of each participant was measured using Dual energy X-ray absorptiometry.
Rectal temperature (Tre) 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. Aural canal temperature (Tau) was measured using a tympanic thermocouple probe (Mon-a-therm Tympanic; Mallinckrodt Medical) placed in the aural canal until resting near the tympanic membrane. The tympanic probe was held in position and isolated from the external environment with cotton, surgical tape (Transpore; 3M, London, Ontario, Canada), and an ear defender (model 1000; Mastercraft, Bobcaygeon, Ontario, Canada). Esophageal temperature (Tes) was measured by placing a pediatric thermocouple probe (Mon-a-therm Nasopharyngeal Temperature Probe; Mallinckrodt Medical) through the participant's nostril into the esophagus. The location of the probe tip in the esophagus was estimated to be at the level of the eighth and ninth thoracic vertebrae reflecting the region of the left ventricle and aorta (36). Measurements for Tes were only obtained in 5 HI and 5 LO participants and were only performed for the initial 30-min of exercise so that Tes onset thresholds and subsequent thermosensitivites could be calculated. Skin temperature (Tsk) was measured at 4 points over the right side of the body using T-type (copper/constantan) thermocouples. The probes were attached using surgical tape. Mean skin temperature was weighted using the following regional proportions: shoulder 30%, chest 30%, quadriceps 20%, and back calf 20% (40). All temperature data were collected using a National Instruments data acquisition module (model NI cDAQ-9172) at a sampling rate of 15 s. Data were simultaneously displayed and recorded in spreadsheet format on a personal computer (Dell Inspirion 545) with LabVIEW 2009 software (National Instruments, Austin, TX).
Whole-body sweat rate (WBSR) was estimated by measuring the body mass of the participant to the nearest 2 g using a platform scale (Combics 2, Sartorius, Mississauga, Ontario, Canada). Measurements were taken at rest directly before exercise and then immediately after completing the 15th, 30th, 45th, and 60th min of exercise, with participants stopping exercise for each measurement, which took ∼20 s. Participants were not towelled down prior to body mass measurements. Average values were calculated for each 15-min period and subsequently divided by BSA to give whole body sweat rate values in grams per square meter.
Local sweat rate of the upper back (LSRback).
was continuously measured using a ventilated capsule. Anhydrous compressed air was passed through the capsule over the skin surface at a rate of 1.80 l/min. Flow rate was measured using an Omega FMA-A2307 flow rate monitor (Omega Engineering, Stamford, CT). Water content of the effluent air was measured using a 473 precision dew point mirror (RH Systems, Albuquerque, NM). Local sweat rate was calculated using the flow rate, and the difference in water content between effluent and influent air was obtained. This value was normalized for the skin surface area under the capsule (3.8 cm2) and was expressed in milligrams per minute per centimeter square.
Metabolic energy expenditure was determined using a Vmax Encore Metabolic Cart (CareFusion, San Diego, CA). Subjects were equipped with a mouth piece and nose clip and were instructed to breath normally through the mouth piece. Metabolic data were recorded for the first 15 min and last 15 min of exercise. Metabolic energy expenditure (M) was obtained from minute-average values for oxygen consumption (V̇o2) in liters per minute, and the respiratory exchange ratio (RER) using Eq. 1 (37): (1) where: ec is the caloric equivalent per liter of oxygen for the oxidation of carbohydrates (21.13 kJ), and ef is the caloric equivalent per liter of oxygen for the oxidation of fat (19.62 kJ). To estimate heat production in W/m2, M was divided by BSA.
External work (W) was regulated and measured directly using semirecumbent cycle ergometer (Lode Corival, AEI Technologies, Naperville, IL).
Participants were told to refrain from ingesting any caffeine or alcohol, or exercising 24 h prior to experimentation. They were asked to arrive at the Thermal Ergonomics Laboratory in Ottawa, Ontario, Canada, after eating a small meal. All trials were conducted in the morning and began between 9 and 10 AM throughout the winter and spring months (October to May) to ensure no difference in acclimatization status between HI and LO groups. It should also be noted that even outside of these months no summer heat acclimatization is observed in the present age group (22 ± 2 years) residing in this geographical region (3). To assure proper hydration, participants were instructed to drink plenty of fluids the night before testing. An additional 500 ml of water was also consumed immediately prior to exercise; however, no fluids were permitted during the ensuing 60-min exercise bout. Although hydration status was not monitored, this additional 500 ml of water consumed prior to the exercise bout was sufficient to assume euhydration throughout exercise due to the neutral climate throughout testing. Following instrumentation and subject preparation, 30 min of rest was recorded to obtain baseline values. Afterward, all subjects cycled for 60 min at either 1) a target metabolic heat production (M-W) of 540 W (FHP trial) or 2) an oxygen uptake of 60% of each individual's V̇o2 peak (REL trial). Pedaling cadence throughout both trials was fixed at 80 revolutions per minute. Ambient air temperature and relative humidity were 24.8 ± 0.6°C and 0.8 ± 0.3 kPa (26 ± 10% RH), respectively. A mechanical fan was placed in front of the participants at a distance of 2.0 m producing an air velocity of 1.3 m/s. These environmental conditions were selected to ensure a physiologically compensable environment at the highest rates of metabolic heat production expected in the study (i.e., HI group at 60% V̇o2 peak). Environmental conditions are also comparable to previous studies investigating the influence of V̇o2 peak on thermoregulatory responses during exercise (16, 18, 23, 43). All trials were completed seminude. Clothing consisted of only light and brief cotton running shorts, athletic shoes, and light cotton socks.
All data are expressed as a mean with standard deviation and were analyzed within exercise trials (i.e., FHP and REL trial). A two-way mixed ANOVA with the repeated factor of “time” (five levels: rest, 15, 30, 45, and 60 min of exercise) and the nonrepeated factor of “V̇o2 peak group” [2 levels: high V̇o2 peak (HI) and low V̇o2 peak (LO)] was used to analyze the dependent variables of Tre, Tau, Tsk, LSRback, and WBSR. When significance was found, individual differences were assessed using independent Student's t-tests. In addition, mean M-W and W throughout each trial, as well as Tes onset sweating thresholds and thermosensitivity of sweating with subsequent changes in Tes after onset were compared between HI and LO groups using a one-way ANOVA. The probability of making a Type I error in all tests was maintained at 5% using a Holm-Bonferroni correction. All analyses were performed using the statistical software package SPSS 18.0 for Windows (SPSS, Chicago, IL, USA).
Participants were successfully selected to ensure no differences between HI and LO groups for body mass (P = 0.884) and body surface area (P = 0.462). However, the LO group had a significantly greater body fat percentage compared with the HI group (P = 0.024). Mean participant characteristics separated according to V̇o2 peak group are presented in Table 1.
Heat Production and External Workload
Average values for heat production and external work are presented in Table 2. Absolute metabolic heat production was successfully maintained at the same levels for both groups of participants in the FHP trial (P = 0.754). The external workload required to attain these fixed levels of heat production did not differ between groups (P = 0.170), but the resultant relative exercise intensity was significantly greater in the LO group compared with the HI group (P < 0.001). Absolute metabolic heat production in the REL trial, was significantly higher (P < 0.001) in the HI group. The external workload in the REL trial was also significantly greater (P < 0.001) in the HI group relative to the LO group but relative exercise intensity was the same (P = 0.579).
In the FHP trial, the changes in both rectal temperature (Tre) and aural canal temperature (Tau) from preexercise resting values were almost identical throughout exercise (Fig. 1). After 60 min of exercise, the change in Tre was 0.87 ± 0.15°C in the HI group and 0.87 ± 0.18°C in the LO group (P = 1.00). Likewise, the change in Tau was 0.70 ± 0.12°C in the HI group and 0.74 ± 0.21°C in the LO group (P = 0.65). Absolute Tre and Tau were significantly lower (P < 0.05) at rest in the HI group compared with the LO group (Fig. 1). Preexercise Tes was also significantly lower in the HI group (36.9 ± 0.2°C vs. 37.2 ± 0.2°C). After 60 min of exercise in the FHP trial, absolute core temperatures remained lower (P < 0.05) in the HI group relative to the LO group (Fig. 1). End-exercise Tes values were not compared between V̇o2 peak groups since measurements were only performed for the initial 30 min of exercise.
In the REL trial, the change in both Tre and Tau from preexercise resting values was significantly greater (P < 0.001) in the HI group relative to the LO group for the last 45 min of exercise (Fig. 2). At the end of 60 min of exercise, the change in Tre was 1.43 ± 0.28°C in the HI group and 0.89 ± 0.19°C in the LO group (P < 0.001), and the change in Tau was 1.10 ± 0.21°C in the HI group and 0.66 ± 0.14°C in the LO group (P < 0.001). Preexercise resting Tre and Tau were again significantly lower (P < 0.05) in the HI group compared with the LO group (Fig. 2), but after 60 min of exercise, both absolute Tre and Tau were significantly higher in the HI group (Fig. 2). As in the FHP trial, preexercise Tes was significantly lower in the HI group (37.0 ± 0.2°C vs. 37.2 ± 0.2°C). Again, end-exercise Tes values could not be compared between V̇o2 peak groups.
Mean Skin Temperature
In the FHP trial, end-exercise skin temperature (Tsk) was 31.2 ± 1.3°C and 31.9 ± 1.4°C in the HI and LO group, respectively (P = 0.374). These values were used to calculate convective heat loss (HI: 43 ± 10 W/m2, LO: 48 ± 10 W/m2; P = 0.368) and radiative heat loss (HI: 26 ± 6 W/m2, LO: 29 ± 6 W/m2; P = 0.363) toward the end of exercise, as well as the maximum rate of evaporation possible in the ambient environment (Emax) (HI: 418 ± 47 W/m2, LO: 428 ± 34 W/m2; P = 0.365).
In the REL trial, Tsk at the end of exercise was 32.6 ± 1.0°C in the HI group and 32.5 ± 1.3°C in the LO group (P = 0.916). Values for convective (HI: 51 ± 3 W/m2, LO: 56 ± 10 W/m2; P = 0.218) and radiative (HI: 31 ± 2 W/m2, LO: 33 ± 6 W/m2; P = 0.291) heat loss as well as Emax (HI: 469 ± 50 W/m2, LO: 451 ± 29 W/m2; P = 0.416) were not different between groups.
Whole Body Sweat Rate
In the FHP trial, whole body sweat loss across the entire 60-min exercise bout of 434 ± 80 ml in the HI group was almost identical to the whole body sweat loss of 440 ± 41 ml in the LO group (P = 0.863). Likewise, the WBSR was almost identical between HI and LO groups at each time point throughout exercise (P = 0.914) (Fig. 3).
In the REL trial, whole body sweat loss after 60 min of exercise was 807 ± 155 ml in the HI group compared with 486 ± 59 ml in the LO group (P < 0.001). Similarly, WBSR was significantly greater in the HI participants relative to their LO counterparts at each time point throughout exercise (P < 0.001) with the difference between HI and LO groups becoming progressively greater (fitness × time: P < 0.001) as exercise continued (Fig. 3).
Local Sweat Rate of the Upper Back
The local sweat rate of the upper back (LSRback) was similar between HI and LO groups throughout 60 min of exercise in the FHP trial (P = 0.267) (Fig. 4). However, in the REL trial, LSRback was significantly greater in the HI group after 15 min and for the remainder of exercise (P < 0.05) (Fig. 4).
Sweating Onset Thresholds and Thermosensitivities
The absolute esophageal temperature (Tes) threshold for the onset of sweating and the subsequent thermosensitivity of the sweating response with changes in Tes for both the HI and LO groups in the FHP and REL trials are depicted in Fig. 5. The mean absolute Tes for the onset of sweating was significantly greater in the LO group relative to the HI group in both the FHP trial (37.6 ± 0.1°C vs. 37.2 ± 0.1°C, P = 0.032) and the REL trial (37.5 ± 0.1°C vs. 37.2 ± 0.1°C, P = 0.031). However, the magnitude of this difference was similar to the discrepancy between preexercise Tes values in the HI and LO group (i.e., 0.2 to 0.3°C). The thermosensitivity of sweat rate with changes in Tes beyond the onset threshold was not different between the HI and LO groups in the FHP trial (P = 0.220) or REL trial (P = 0.770).
The present study clearly demonstrates that large differences in V̇o2 peak do not influence changes in core temperature or sweating during exercise in a neutral climate, independently of metabolic heat production, body mass, and BSA. Also, when mass and BSA are similar, exercise at a relative intensity (60% V̇o2 peak) results in a much greater change in core temperature and sweating in a high V̇o2 peak group, not due to an influence of V̇o2 peak per se, but due to differences in heat production.
Stapleton et al. (48) reported similar changes in core temperature and whole body evaporation during exercise at a fixed heat production before and after a ∼10% increase in V̇o2 peak from aerobic training. However, they hypothesized that a greater difference in fitness may be required to alter thermoregulatory responses (48). The present study now shows that even a ∼50% difference in V̇o2 peak between groups matched for body mass and BSA, does not alter changes in core temperature or sweating during exercise at the same heat production.
To date, the prevailing wisdom has apparently been that V̇o2 peak alters the regulation of core temperature during exercise (5, 16, 18, 26, 32), and therefore, the confounding effect of fitness is typically eliminated between independent groups by matching them for V̇o2 peak and/or administering a relative exercise intensity (% V̇o2 peak). In the FHP trial, the HI and LO groups exhibited very similar changes in core temperature (Fig. 1), with little within-group variability, showing that if heat production and mass are the same between groups, matching for V̇o2 peak is not necessary to compare core temperature changes. In addition, relative exercise intensity was distinctly different in the FHP trial (HI: 39.7 ± 4.2% vs. LO: 57.6 ± 8.0% V̇o2 peak) showing that the change in core temperature is independent of relative exercise intensity. In fact, exercise at 60% V̇o2 peak (REL trial) yielded far greater changes in core temperature in the HI group (Fig. 2).
The matching of body mass between independent groups is no more straightforward than matching for fitness; however, any differences in body mass between participants could be accounted for by administering exercise that elicits the same heat production per unit mass (i.e., W/kg). In the FHP trial, heat production per unit mass was the same between HI (7.0 ± 0.6 W/kg) and LO (6.9 ± 0.9 W/kg) groups (Table 2), whereas in the REL trial, it was much greater in the HI (10.7 ± 1.3 W/kg) relative to the LO (7.7 ± 1.4 W/kg) group. The greater changes in core temperature in the HI group in the REL trial but not in the FHP trial suggest that heat production per unit mass primarily determines the change in core temperature during exercise, irrespective of V̇o2 peak and, therefore, relative exercise intensity. Previously reported comparisons of core temperature responses between independent groups exercising at the same %V̇o2 peak but apparently producing different rates of heat production per unit mass may, therefore, merit re-examination. For example (35), the greater changes in core temperature reported in nonburned children (V̇o2 peak: 36.1 ± 9.8 ml·kg−1·min−1) relative to severely burned children (V̇o2 peak: 24.1 ± 6.5 ml·kg−1·min−1) exercising at 75% of V̇o2 peak are likely explained by their greater heat production per unit mass (nonburned: ∼9.2 W/kg; burned: ∼6.1 W/kg), possibly masking any underlying differences in thermoregulatory function between groups.
As in many other previous studies (30, 31, 48), absolute core temperatures at rest were ∼0.2°C lower in the HI group (Figs. 1 and 2). Because changes in core temperature were identical in the FHP trial, the HI group had ∼0.2°C lower end-exercise core temperatures at a fixed heat production. In the REL trial, changes in core temperature were so much greater in the HI group that their end-exercise values were much higher than the LO group (Fig. 2). This runs counter to the concept that absolute core temperature is determined by relative exercise intensity, irrespective of aerobic capacity (1, 9, 16, 18, 43) and further supports heat production, body mass, and BSA, and not V̇o2 peak, as the primary determinants of end-exercise core temperature.
It is very well established (2, 21, 34, 45) that irrespective of core temperature, the rate of whole body and local sweating is determined by the rate of evaporation required for heat balance Ereq and the skin wettedness required for heat balance (i.e., Ereq/Emax). The present data clearly show that this relationship is not altered whatsoever by large differences in V̇o2 peak. Because HI and LO groups in the present study were also matched for BSA, exercise at a fixed heat production of 540 W (FHP trial) also yielded the same heat production per unit surface area (275 W/m2) but distinctly different relative exercise intensities. Skin temperatures were also similar between groups; therefore, the rate of evaporation required for heat balance (Ereq) (HI: 178 ± 20 W/m2 vs. LO: 172 ± 26 W/m2), the maximum evaporation possible (Emax) (HI: 418 ± 47 W/m2 vs. LO: 428 ± 34 W/m2) and subsequent Ereq/Emax values (HI: 0.43 ± 0.04 vs. LO: 0.40 ± 0.05) were almost identical, and no differences in whole body (Fig. 3) or local (Fig. 4) sweating occurred between HI and LO groups. Conversely, a relative exercise intensity of 60% V̇o2 peak (REL trial) elicited much greater levels of Ereq in the HI group (HI: 299 ± 39 W/m2 vs. LO: 189 ± 42 W/m2), and since Emax was not different, Ereq/Emax in the HI group (0.64 ± 0.09) was much greater than the LO group (0.42 ± 0.07), resulting in a much greater whole body and local sweat rates in the HI group (Figs. 3 and 4). It follows that the greater local sweat rates recently reported in trained males and females compared with untrained counterparts matched for mass (30) are, therefore, likely due to greater heat balance requirements rather than fitness per se since exercise was performed at a % of V̇o2 peak.
Collectively, the data of the present study support the notion that studies wishing to primarily compare sweating responses between independent groups should administer exercise intensities that yield similar Ereq and Ereq/Emax values (2, 34, 45) irrespective of %V̇o2 peak. To achieve equal Ereq and Ereq/Emax values between groups of different BSA, exercise must be administered to generate the same metabolic heat production per unit surface area (4, 27). It follows that the use of relative exercise intensities when comparing populations different in BSA-mass ratio, such as obese vs. lean (13, 46), or children of different ages (15, 42), could lead to potentially flawed conclusions. For example, the diminished sweat rates reported in prepubertal (age: 10.8 yr; mass: 34.6 kg; BSA: 1.17 m2; V̇o2 peak: 54 ml·kg−1·min−1) relative to late-pubertal boys (age: 16.2 y; mass: 63.4 kg; BSA: 1.74 m2; V̇o2 peak: 59 ml·kg−1·min−1) during exercise at 50% V̇o2 peak (14) can be explained by their lower heat production per unit surface area (prepubertal: ∼270 W/m−2; late-pubertal: ∼365 W/m−2) and, therefore, lower Ereq/Emax values.
Preexercise resting Tes was ∼0.3°C lower in the HI group. In parallel, the absolute Tes onset threshold for sweating was also ∼0.3°C lower (Fig. 5), but no alteration was observed in the slope of the relation between sweating and changes in Tes (thermosensitivity), indicating that only advantage of a high V̇o2 peak is a lower resting core temperature. Such a shift in the Tes threshold for the onset of sweating, a central adaptation associated with physical training (41), has been previously observed following an increase in V̇o2 peak of ∼6 ml·kg−1·min−1 with short-term aerobic training in women (31) and young men (38). These authors concluded that longer-term training would be required to alter the thermosensitivity of sweating; however, the present data show that thermosensitivities remains the same in men who differ greatly in V̇o2 peak (∼20 ml·kg−1·min−1) and, therefore, potentially long-term training volume (Fig. 5). It is also likely that the differences in V̇o2 peak between HI and LO groups may not be purely training-based but, in part, due to underlying genetic factors altering oxygen delivery and mitochondrial muscle oxidative capacity (24). However, while there may be differences in genetic factors associated with the different V̇o2 peak values between HI and LO groups, they clearly did not influence the outcome of the present study, since no differences in thermoregulatory responses during exercise were observed after eliminating differences in body mass and BSA.
The findings of the present study are only applicable to conditions that are physiologically compensable. Maximal sweating capacity (6, 20, 39) and subjective tolerance to the heat (7, 44) are no doubt improved by aerobic fitness, and, therefore, individuals with a high V̇o2 peak would certainly have a distinct advantage during exercise at a fixed heat production in a physiologically uncompensable (i.e., hot and humid) environment. While participant groups in the present study were matched for body mass and BSA, as well as other factors, such as age and ethnicity, body fat percentage was significantly different. Since fitness was inversely associated with fatness, any observed influence of a low V̇o2 peak upon thermoregulatory responses would have been difficult to dissociate from an influence of a high adiposity. The lower specific heat capacity of fat tissue (2.973 kJ·kg−1·°C−1) would have yielded a slightly lower average specific heat of the body in the LO group (3.53 kJ·kg−1·°C−1) relative to the HI group (3.45 kJ·kg−1·°C−1) (19). Therefore, a similar body heat storage would have theoretically elevated body temperature in the LO group by a greater magnitude. Nevertheless, the changes in both Tre and Tau were almost identical between V̇o2 peak groups at a fixed heat production, and changes in core temperature were significantly greater in the HI (and lean) group at 60% V̇o2 peak and, thus, opposite to the anticipated fat effect. Therefore, it appears that any thermoregulatory implications of the difference in adiposity between the HI and LO groups were insignificant, likely due to any insulative fat layers being bypassed by blood flow following the onset of exercise (28). One other implication of a higher adiposity in the LO group is a concomitant difference in body volume, and therefore, BSA, due to the lower density of fat (0.9 kg/l) relative to muscle (1.1 kg/l). The influence of differences in adiposity upon BSA was not accounted for in the present study since BSA was estimated using the equation of DuBois and DuBois (12) that employs only height and weight. A 10% greater body fat yields a BSA that is greater by between 3.5% (all additional fat on limbs) and 0.5% (all additional fat on torso) (25). Using the assumption that the additional fat in the LO group was evenly distributed between the limbs and torso, one can reasonably predict that BSA in the LO group was ∼2% greater than estimated using just height and weight; giving a mean adjusted BSA value (1.98 m2) that is even closer to that of the HI group (1.97 m2). Whole body evaporation was not directly measured in the present study, and although participants were not towelled down prior to each body mass measurement to ensure that mass changes better reflected the quantity of sweat that evaporated, some dripping of sweat may have occurred, particularly under conditions in which Ereq approached unity with Emax (i.e., HI group in the REL trial). Finally, future research is required to test the notion that heat production per unit mass determines the change core temperature across a range of exercise intensities. The interaction of this method with individual differences in BSA-mass ratio is also needed.
Perspectives and Significance
The present study suggests that changes in core temperature and thermoregulatory sweating should not be compared between independent groups using a relative exercise intensity (i.e., % of V̇o2 peak). Changes in core temperature are likely best assessed with exercise intensities administered to generate the same metabolic heat production per unit mass, whereas changes in thermoregulatory sweating are potentially best assessed with exercise intensities administered to generate the same metabolic heat production per unit surface area.
In conclusion, changes in core temperature and sweating during exercise in a neutral climate are apparently determined by metabolic heat production, body mass, and body surface area, not V̇o2 peak. Absolute core temperatures at the end of exercise at a fixed metabolic heat production were ∼0.2°C lower in individuals with a high V̇o2 peak, simply due to a ∼0.2°C lower core temperature prior to the onset of exercise.
This research was supported by a Natural Sciences and Engineering Research Council of Canada Discovery Grant (grant holder, Ollie Jay #386143-2010). Mr. Bain and Mr. Deren were supported by a University of Ottawa Master's Scholarship and Mr. Bain was also supported by an Ontario Graduate Scholarship.
No conflicts of interest, financial or otherwise, are declared by the authors.
The authors would like to thank the participants for volunteering for the study, as well as Nicole Lesperance for her assistance during data collection.
- Copyright © 2011 the American Physiological Society