Share this article on:

Thermoregulatory Responses to Constant versus Variable-Intensity Exercise in the Heat


Medicine & Science in Sports & Exercise: November 2008 - Volume 40 - Issue 11 - p 1945-1952
doi: 10.1249/MSS.0b013e31817f9843
BASIC SCIENCES: Original Investigations

Purpose: To compare the thermoregulatory responses between constant (CON) and variable-intensity exercise (VAR) in a dry-hot environment (36°C, 29% relative humidity, and 2.5 m·s−1 airflow).

Methods: In a random order, seven endurance-trained, heat-acclimated subjects cycled either at 60% V˙O2max (CON) or alternating 1.5 min at 90% V˙O2max with 4.5 min at 50% V˙O2max (VAR). Total work output (915 ± 100 kJ) and exercise duration (90 min) were identical in both trials.

Results: Net metabolic heat production was not different between trials (394 ± 12 vs 408± 11 W·m−2 for VAR vs CON). However, heat storage (60 ± 3 vs 48 ± 4 W·m−2), the increase in rectal temperature (1.6 ± 0.1 vs 1.3 ± 0.1°C), and final heart rate (HR; 147 ± 5 vs 141 ± 4 beats·min−1) were all higher for VAR than for CON (P < 0.05). During VAR, averaged forearm skin blood flow (SKBF) was lower, whereas whole-body sweat rate (1.23 ± 0.1 vs 1.11 ± 0.1 L·h−1) and dehydration (2.8 ± 0.1% vs 2.5 ± 0.2%) were higher than during CON (P < 0.05). Final blood lactate during VAR was higher than during CON (3.5± 0.4 vs 2.1 ± 0.3 mmol·L−1; P < 0.05).

Conclusion: Ninety minutes of variable-intensity exercise in a hot environment increases heat storage and fluid deficit in comparison to the same amount of work performed in a constant-load mode. VAR increases not only thermal (i.e., heat storage) but also cardiovascular (i.e., heart rate) and metabolic (i.e., blood lactate) stresses, which makes it less advisable than CON when the goal is to minimize physiological stress.

University of Castilla-La Mancha, Exercise Physiology Laboratory at Toledo, SPAIN

Address for correspondence: Ricardo Mora-Rodriguez, Ph.D., University of Castilla-La Mancha, Avda. Carlos III, s/n. 45071, Toledo, Spain; E-mail:

Submitted for publication February 2008.

Accepted for publication May 2008.

The thermoregulatory responses to prolonged (>60 min) constant-load submaximal exercise (<75% V˙O2max) in the heat have been extensively described (12,13). However, most sport activities are characterized by frequent bouts of high-intensity, short-duration exercise alternated with recovery periods. During the recovery periods, exercise is either continued at a lower intensity (i.e., variable-intensity exercise) or interrupted (i.e., intermittent exercise). Even in sports typically considered to be constant (i.e., cycling or running), the intensity varies in response to terrain condition or to the changes in pace imposed by the opponents (30). Despite being an exercise mode widely used, the information about the thermoregulatory responses to variable-intensity exercise in a hot environment is incomplete.

For a given environmental heat, the core temperature achieved during exercise largely depends on the relative exercise workload (i.e., percent V˙O2max (26)). On the basis of this study, it could be thought that constant-load (CON) and variable-intensity (VAR) exercise in the heat (when matched by workload and duration) may result in similar elevations in core temperature. This is the case when comparing long bouts (15-30 min) of variable-intensity to constant-intensity walking in a hot environment (2,17). However, when using shorter and more intense bouts of VAR exercise, it has been found either a tendency (8,15) or a significantly faster rise (10) in final core temperature, when compared to CON exercise. Because metabolic heat production is mainly determined by workload (albeit influenced by efficiency), the higher core temperature during VAR exercise in comparison to CON exercise suggests reduced heat dissipation.

Ekblom et al. (10) observed that intermittent cycling exercise reduced sweat production in comparison to continuous exercise of the same average work rate and duration. The authors hypothesized that the diminished sweat production may have been responsible for the higher rectal temperature during intermittent exercise. However, Drust et al. (8) did not find a difference in sweat production to explain the trend for a higher rectal temperature during intermittent in comparison to continuous treadmill running. Beyond these apparently contradictory findings stands the fact that neither of these studies were held in a hot environment, and it is likely that the heat dissipative mechanisms were not fully stressed. To our knowledge, no study has compared the sweat rate response of VAR versus CON exercise specifically in a hot-dry environment where sweat rate is the main contributor of heat dissipation.

Skin blood flow (i.e., SKBF) increases when exercise intensity is raised from mild to moderate (i.e., 50% to 70% V˙O2max (11,28)). However, when intensity is raised to 90% V˙O2max, its response plateaus (11) or even declines (28). These studies were held in a thermoneutral environment, and thus, the observed skin vasoconstriction (when plotted against core temperature) may not be present in a hot environment that induces skin vasodilation. On the contrary, the high-intensity bouts of VAR exercise could set a limit for the increases in SKBF. Blood flow is redistributed away from the skin during bouts of high-intensity exercise in conjunction with the secretion of adrenomedullary hormones (i.e., catecholamines (25)). We have previously observed that the infusion of adrenaline to simulate the plasma concentrations achieved during high-intensity exercise (i.e., 90% V˙O2max) reduces SKBF during exercise in the heat (21). However, to our knowledge, no data ontheSKBF response to VAR exercise in theheat in comparison to CON exercise are available.

The first purpose of the present investigation was to compare the amount of heat stored during prolonged cycling in a hot-dry environment when performing the same amount of work using either CON (60% V˙O2max) or VAR (90-50% V˙O2max) exercise. A second purpose was to measure the response of heat dissipation indexes (i.e., sweat and SKBF) that would affect heat storage. We also evaluated the effects of the exercise mode on the physiological strain index (PSI) that gauges in conjunction the cardiovascular and thermal stresses. On the basis of the available literature, we hypothesized that during prolonged exercise in a hot environment, heat storage will be larger when the same amount of work is performed alternating exercise intensities (VAR) than when a constant-intensity mode is used (CON).

Back to Top | Article Outline



Seven endurance-trained cyclists volunteered to participate in this investigation that was approved by the local hospital's research ethics committee. Subjects' means ± SD for age, height, and weight were 27 ± 0.6 yr, 1.8 ± 0.1m, and 72 ± 7 kg, respectively, and their maximal oxygen consumption was 4.3 ± 0.1 L O2·min−1. All subjects completed a health history screening questionnaire, and a preliminary physical examination (including ECG) ensured that each participant was in good health. They gave their written informed consent to participate and retained the right to withdraw from the study at any time.

Back to Top | Article Outline

Preliminary test.

Oxygen consumption was measured every 15 s during an incremental cycling test to volitional fatigue. Maximal oxygen consumption (i.e., V˙O2max) was defined as the highest plateau (two successive maximal readings within 0.15 L·min−1) reached. Acclimatization (seven consecutive days of cycling for 90 min at 63% V˙O2max) to a hot-dry environment (37°C and 30% relative humidity) was conducted to avoid the bias of having subjects in progressive stages of acclimatization during the experimental trials.

Back to Top | Article Outline

Experimental protocol.

Each subject performed in two random-order experimental trials separated by 48 h. Trials consisted on 90 min of cycle ergometer (Monark® 818; Monark Exercise, Varberg, Sweden) exercise in a hot-dry environment (36 ± 0.3°C, 29 ± 1% relative humidity, and 2.5 m·s−1 airflow). Subjects pedaled either at a constant load that elicited 60% of their V˙O2max (169 ± 7 W; CON) or alternating 1.5 min at 90% V˙O2max (254 ± 11 W) with 4.5 min at 50% V˙O2max (141 ± 6 W; VAR) for 15 consecutive bouts (Fig. 1). Total work output was the same for both trials (i.e., 915 ± 100 kJ). Trials were carried out at the same time of the day to circumvent the effects of circadian variation in body core temperature. Subjects were asked to record their diet for 48 h before the first experimental trial and to repeat this same diet before the subsequent trial. They were also requested to refrain from physical activity 24 h before each trial.



Subjects were instructed to ingest 500 mL of water 2 h before arriving at the laboratory to increase the likelihood that they would begin the test in a euhydrated state. On arrival, a urine sample was collected, and its specific gravity (i.e., U sg) was measured by refractometry (Uricon-Ne, Atago, Japan) to confirm euhydration (U sg < 1.020 g·mL−1). Then, nude body mass was recorded, a rectal temperature (T REC) thermistor was inserted, and a heart rate (HR) transmitter was secured around the chest. The subjects then put on cycling shorts and cleated shoes, entered the climatic chamber, and sat on the cycle ergometer for 15 min. During this time, their lower back was cleaned and dried before positioning two large sweat patches to measure local sweat rate and composition. Thereafter, four skin temperature (T SK) thermistors were attached to the skin, and a laser probe for measuring SKBF was secured in the dorsum of the left forearm. After 15 min of seated rest, a blood sample was drawn using finger prick, and subjects began the 90-min exercise period.

At rest and during exercise, T REC, T SK, and SKBF were recorded every minute. Heart rate and the rates of O2 consumption (V˙O2) and CO2 production (V˙CO2) were measured every 5 s during a 6-min period early on (18-24 min) and at the end of exercise (84-90 min). These periods included a complete set of high- and low-intensity bouts of VAR (Fig. 1). During exercise, blood samples (125 μL) were taken at 12 and 90 min. Immediately after exercise, sweat patches were collected, and the remaining instruments were rapidly removed from the subjects. The subjects then removed their clothing and toweled dry before their postexercise nude body weight was recorded.

Back to Top | Article Outline

Metabolic heat production, heart rate, and gross efficiency.

V˙O2 and V˙CO2 were analyzed using an automated breath-by-breath, open-circuit spirometer (Quark b2; COSMED, Rome, Italy) calibrated before each single test (0.01% accuracy; Carburos Metalicos, Madrid, Spain). Heart rate (HR) was measured by short-range telemetry (Polar Advantage, Kempele, Finland). V˙O2 and V˙CO2 measurements were used for calculating net metabolic heat production [M˙NET = ((3.869V˙O2) + (1.195V˙CO2))(4.186 · 60−1)]-workload in watts (4). To calculate M˙NET, expired gases were measured during two 6-min periods of exercise (18-24 min and 84-90 min), which included a set of high- and low-intensity exercise during VAR. Data were collected at 5-s intervals, and 72 data points were averaged per period. Gross efficiency (GE) was calculated as the quotient between the work rate and the rate of energy expenditure: GE (%) = (work rate [W] / energy expended [W]).

Back to Top | Article Outline

Thermoregulatory responses and PSI.

Skin temperature was measured on four sites, namely, lateral calf, thigh, chest, and upper arm, using superficial skin thermistors (Model 409; Yellow Springs Instruments Inc., Yellow Springs, OH). Mean skin temperature was calculated using the weighting formula from Ramanathan (24): T SK = 0.3(T CHEST + T ARM) + 0.2(T THIGH + T LEG). Rectal temperature (T REC) wasmeasured with a flexible thermistor (Model 401; Yellow Springs Instruments Inc.) positioned 15 cm past the anal sphincter. All thermistors were calibrated before the study using a water bath (Vertex; Velp, Italy) and a reference high-resolution (0.1°C) mercury in-glass thermometer traceable to the German Bureau of Standards (Select, Proton, Spain). Forearm SKBF was measured using a laser Doppler flowmeter (Moor Lab, Moor Instruments, Devon, United Kingdom) positioned at the dorsum of the left forearm. At rest and during exercise, the cyclists' forearms rested on triathlete handlebars to avoid spurious SKBF values due to movement artifacts. SKBF was normalized using each subject's preexercise resting value for that trial. All these probes were connected to a multichannel A/D board (Power Lab 8SP, ADI, UK) and associated software that displayed and stored data throughout the trials. Mean body temperature (T body) was calculated from rectal and skin temperature using weighting factors for exercise in a hot environment (T body = 0.79(T REC) + 0.21(T SK)) (6). Heat storage (H˙S) was calculated from the increase in T body during exercise, pre-exercise body weight (BW), and surface area (AD = 0.202BW0.425height0.725 (9)) according to the equation proposed by Adams et al. (1) (H˙S = 0.965BWΔT body · AD)−1. PSI was calculated from rectal temperature and heart rate following the formulae proposed by Moran et al. (23).

Back to Top | Article Outline

Sweat analyses.

Whole-body sweat rate (whole-body m˙sw) was determined by subtracting postexercise from preexercise nude body weight using a ±0.05-kg sensitive scale (WildCat; Metler, Toledo, OH) correcting for respiratory water and carbon losses (18). Percent dehydration was calculated by dividing the body weight lost by the initial nude body weight. For the measurement of regional sweat rate, two sweat patches composed of sterilized gauze (5 × 5 cm; Indas, Madrid, Spain) covered with powder-free latex (Aposan, Spain) were attached to the skin of the lower back using an adhesive wound dress (10 × 12 cm; Tegaderm; 3M, Neuss, Germany). One patch was collected after 60 min and the other after 90 min of exercise. Regional sweat rate (lower back m˙sw) was determined gravimetrically (±0.001 g; Model XB 220-A; Precisa, Dietikon, Switzerland) by dividing the gauze's mass gain by the skin area patched. Thereafter, sweat was separated by centrifugation (MPW-350R; Med Instruments, Warsaw, Poland) and analyzed for osmolality (Osmolsweat) by freezing point (Micro-osmometer 300; Advanced Instruments) and sodium and potassium concentration ([Na+]sweat; [K+]sweat) using flame photometry (Clinical FP7; Jenway Ltd., Essex, United Kingdom).

Back to Top | Article Outline

Blood analysis.

Arterialized capillary blood was collected (125 μL) from prewarmed fingertips into a heparinized microtube (CriGel Li-He; Radiometer, Copenhagen, Denmark) at rest and after 12 and 90 min of exercise. Blood was immediately analyzed for hemoglobin concentration (i.e., [Hb]; ABL-520; Radiometer, Madrid, Spain), and hematocrit was measured in triplicate by microcentrifugation and corrected for trapped plasma and venous sampling. Relative changes in plasma volume were calculated with the equations outlined by Dill and Costill (7). Blood lactate concentration was measured in the remaining blood (40 μL) using an electroenzymatic analyzer (YSI 1500 Sport; Yellow Springs Instruments Inc.).

Back to Top | Article Outline

Statistical analysis.

Differences between trials in preexercise hydration status (i.e., urine specific gravity, body weight and hemoglobin concentration) and in variables measured at a single time point (i.e., sweat rate and percent dehydration) were analyzed using Student's paired t-test. Data collected repeatedly over time were analyzed using two-way (time × trial) repeated measures ANOVA. After a significant F ratio (Greenhouse-Geisser adjustment for sphericity), pairwise differences were identified using Tukey's (HSD) post hoc procedure. A paired t-test was also used to determine differences between trials in the average 90 min of exercise responses. Figures are illustrated as means ± SEM for clarity of presentation, and all other data are presented as means ± SD. For all statistical analyses, the significance level was set at P < 0.05.

Back to Top | Article Outline


Fluid balance.

Our indexes of hydration status weresimilar before each trial (i.e., U sg = 1.013 ± 0.005 vs1.016 ± 0.005 g·mL−1, BW = 71.7 ± 5.6 vs 71.9± 5.4 kg, and [Hb] = 14.5 ± 1.2 vs 14.3 ± 1.2 g·dL−1 for VAR vs CON). During the VAR trial, subjects hadhigher whole-body and lower-back sweat rates (Table 1; P < 0.05) than during CON. This resulted in higher percentdehydration for VAR than CON (2.8 ± 0.2% vs 2.5 ± 0.4%, respectively; P< 0.05). There were no differences in sweat electrolyte composition between CON and VAR (Table 1).



Back to Top | Article Outline

Metabolic heat production and gross efficiency.

During VAR, V˙O2, M˙NET, and gross efficiency showed a pulsed response. However, when averaged during the 6-min period that included a high- and low-intensity bout, the responses were similar to the steady-state responses of the CON trial. Average V˙O2 was similar between VAR and CON during the 18- to 24-min period (2.6 ± 0.2 and 2.7 ± 0.2 L·min−1, respectively) and increased similarly (∼4%) in both trials at the end of exercise (i.e., 84- to 90-min period). Average M˙NET was not significantly different between VARand CON at the initial measurement or at the end ofexercise (Fig. 2A). Average gross efficiency during these 6-min periods was similar between VAR and CON (i.e., 18-19%; Table 2).





Back to Top | Article Outline

Thermoregulatory responses.

Resting rectal temperature (T REC) was similar in both trials (37.5 ± 0.4 and 37.6 ± 0.3°C for VAR and CON, respectively). During the last 30min of VAR, T REC was 0.2°C higher than during CON, although the differences did not reach significance. However, the 90-min average and the increase in T REC were larger during VAR than during the CON trial (P < 0.05; Fig. 3A and Table 2, respectively). Mean skin temperature (T SK) was not different between trials (Fig. 3B). In both trials,most of the increase in forearm SKBF took place within the first 10 min of exercise. However, during VAR, the increase was not as high as during the CON trial, leading to lower average values for the 90 min of exercise (P < 0.05; Fig. 3C). The alternation of workloads during VAR did not have a visible effect on SKBF or T SK, which fluctuated similarly in both trials. During the VAR trial, the heat stored was statistically higher for VAR than for CON (60 ± 7 vs 48 ± 9 W·m−2, respectively, P < 0.05; Table 2).



Back to Top | Article Outline

Heart rate, PSI, and blood lactate.

There was no difference in preexercise heart rate between trials (74 ± 12 beats·min−1). During the 18- to 24-min period of exercise, heart rate (HR) was elevated during the VAR trial in comparison to the CON trial (136 ± 5 vs 128 ± 7 beats·min−1, respectively, P < 0.05; Fig. 2B). HR drifted similarly in both trials (∼9%), and thus, the differences among them were maintained during the 84- to 90-min period of exercise (147 ± 9 vs 141 ± 9 beats·min−1, respectively). After 12 min of exercise, plasma volume declined from resting values further during VAR than during CON (P < 0.05; Table 2). However, this difference in plasma volume between trials was not maintained until the end of exercise. PSI was similar between trials after 24 min of exercise, but it was larger during VAR than during CON after 90min of exercise (7.3 ± 0.9 vs 6.4 ± 0.9 arbitrary units, P < 0.05; Table 2). Two bouts of VAR exercise (i.e., 12min) increased blood lactate above resting levels, whereas after 12 min of CON, lactate concentrations remained at resting levels. The differences between trials in blood lactate were maintained throughout the exercise (P < 0.05; Table 2).

Back to Top | Article Outline


The main findings of the present study were that during prolonged cycling in a hot-dry environment, rectal temperature increases more when the same amount of work is performed alternating exercise intensities (VAR) than when a constant-intensity mode is used (CON). During VAR, body temperature and heat storage increased higher than CON (i.e., 26%, P < 0.05; Table 2). The PSI and blood lactate concentration were higher at the end of VAR in comparison to CON (Table 2). In summary, we have observed that during exercise in the heat, VAR increases the thermoregulatory (i.e., rectal temperature), cardiovascular (i.e., heart rate) and metabolic (i.e., blood lactate) physiological stresses in comparison to the same amount of work completed in the CON mode.

Comparing pulsed work (VAR) to steady work (CON) at the same energy expenditure in the heat, Lind (17) and Belding et al. (2) found no difference in core temperature. However, in these early studies, the bouts of VAR were long (15-30 min) and the intensity was low (i.e., walking at 5.6 km·h−1). The authors already speculated that with shorter and higher-intensity pulses of VAR, a steady-state thermal response may not have been achieved. We observed that when 15 bouts of high-intensity cycling in the heat were performed (i.e., VAR trial), heat storage was higher than during CON. Furthermore, the mismatch between heat production and dissipation seemed to be accumulative because rectal temperature (T REC) during VAR progressively separated from CON (Fig. 3A), being around 0.2°C higher during the 45- to 90-min period. Although the difference in T REC after90 min of exercise was not significant, possibly longer exercise duration would have produced significant differences.

Drust et al. (8) tested soccer players during 46 short bouts of running alternating velocities (between 6 and 21 km·h−1) in comparison to the same amount of work performed continuously (12 km·h−1). Similar to our data, their rectal temperature (T REC) tended to be higher during VAR than during CON (∼0.3°C). Only one publication in this topic has been able to show a statistically significant increase in T REC when comparing VAR to CON (i.e., 0.35°C (10)). In this study, Ekblom et al. (10) tested three subjects during 60 bouts of 30-s cycling exercise interspersed with 30 s of rest. To start moving the ergometer's flywheel in each bout of VAR, additional work was needed in comparison to CON. To match the oxygen consumed due to the extra work, they reduced 7% of the workload during VAR. Consequently, metabolic heat production (M˙NET = energy expenditure (V˙O2) − work performed (workload)) was 5% higher. It is unclear how much of the higher T REC during VAR could be attributed to the higher M˙NET.

To avoid differences in M˙NET, we did not stop the cycle ergometer flywheel, but the workload was lowered instead. In average, M˙NET was not different between trials (Fig. 2A). Although during VAR, the workload was below the maximal aerobic capacity (i.e., 90% V˙O2max), the V˙O2 response to a sudden increase in workload was not immediate, and anaerobic metabolism contributed to energy provision. Our indirect calorimetry analysis only accounts for aerobic energy expenditure and thus underestimates the heat produced by the anaerobic metabolism. Heat liberation per mole of ATP produced through anaerobic metabolism is less than when aerobic metabolism is predominant (14). In addition, it has been observed that aerobic metabolism contribution to energy increases with the number of high-intensity exercise bouts (3,16). Both of these factors reduced our error for not being able to measure anaerobic heat production in our VAR trial. In addition, gross efficiency reached 45% during the high bout of VAR exercise (90% V˙O2max), evidencing that oxygen consumption lagged behind the ATP demands. During recovery (50% V˙O2max), gross efficiency dropped to 12%, evidencing an excess of oxygen consumption for that workload. Of note, when gross efficiency was averaged during the high and low bout of VAR, there was no difference with CON (∼18%; Table 2). This suggests that the oxygen deficit (an estimate of the anaerobic energy contribution) incurred during the 1.5 min at 90% V˙O2max of VAR was fully compensated during the 4.5 min of 50% V˙O2max and that our M˙NET calculation based only in aerobic metabolism may be correct.

If M˙NET was similar between trials, the larger heat storage during VAR must have been caused by a reduction in heat dissipation. We observed that during VAR, SKBF tended tobe reduced from the first stages of exercise (i.e., 11%; Fig. 3C). However, due to the high variability in the SKBF measurement, the differences did not reach statistical significance. It has been reported that high-intensity exercise (i.e., 90% V˙O2max) attenuates the SKBF response to exercise (11,27) and reduces skin conductance (29) during constant-load mode of exercise. However, we fail to detect a rapid transient reduction in forearm SKBF during the high-intensity bouts of exercise in VAR. In apparent contradiction with the reductions in SKBF, the skin temperature (T SK) was elevated during the first stages of VAR (Fig. 3B). Ekblom et al. (10) also reported an initially higher T SK that progressively dropped during intermittent work. T SK responds in parallel with SKBF when airflow is minimal, but the presence of airflow potentiates evaporation that affects T SK (20).

Using uncompensable heat stress (i.e., no airflow), Kranning and Gonzalez (15) observed during VAR a reduction in T SK during walking and jogging and a sudden increase during seated recovery. Apparently, upright exercise induced skin vasoconstriction limiting heat dissipation, constriction that was released during the seated recovery. Unfortunately, in that study, it is not possible to separate the effects of varying exercise intensity on T SK from the effects of varying posture. In the present study, with subjects remaining in a similar cycling position during the whole trial, we did not observe a biphasic response of T SK or SKBF to the changes in exercise intensity during VAR. The increases in workload during VAR surely raised blood pressure but not SKBF (if something tended to decrease), which strongly suggests that VAR reduced skin conductance. In turn, the reduction in SKBF might be responsible for limiting heat dissipation and increasing heat storage during VAR.

Interestingly, our measurements of sweat rate (whole body and lower back) coincided to show larger values during VAR than during CON (Table 1). One possible explanation for the larger sweat rate despite higher heat storage during VAR is that the extra sweat produced was not evaporated (i.e., inefficient sweating). We have recently found that sweat evaporation is greatly influenced by airflow velocity (20). Our airflow velocity was moderate (2.5 m·s−1) to develop a high level of heat storage, but, in turn, it sets a limit for the maximal sweat that could be evaporated. Any sweat produced above that limit was probably not evaporated. High-intensity exercise increases sweat rate (5,19) by increasing the adrenergic nerve signal to the sweat gland. Probably during VAR, the sweat gland's response to the adrenergic stimulation during the 90% V˙O2max was not fully decreased during the 50% V˙O2max low bout of exercise, resulting in a higher total sweat output.

The higher sweat rate during VAR did not influence sweat electrolyte composition, which was similar to CON. According to a recent study of Buono et al. (5), an increased sweat gland output would increase sweat electrolyte concentration due to the reduced tubule reabsorption time. The reason why the elevated sweat rate with VAR did not produce a higher electrolyte secretion is currently unclear. The VAR mode of exercise used in the present experiment makes our data not comparable to the steady-state exercise mode. We found a tendency for reductions in lower-back sweat rate from 60 to 90 min of exercise (Table 1). This reduction in sweat rate was paralleled by a reduction in sweat osmolality and electrolytes' concentration in agreement with the reabsorption mechanism proposed by Buono et al. (5).

Finally, after 12 min of the VAR trial, plasma volume was reduced to below the CON trial, likely due to the high blood pressure built during the 1.5 min at 90% V˙O2max. Some of that plasma concentration may be behind the elevated lactate level after 12 min of exercise during VAR exercise. At the end of the exercise, the percent reduction in plasma volume was similar between trials (Table 2); however, lactate was still elevated in the VAR trial. This suggests either that glycolysis was further activated during VAR or that lactate clearance was reduced. We found that the VAR trial elevated HR by 6 to 8 beats·min−1 higher than that during CON. When HR and T REC were combined into a unique index (i.e., PSI), VAR revealed as a more stressful trial despite amounting the same total work. PSI responds to both, increases in exercise intensity and dehydration (22). The observed differences in PSI between VAR and CON are of similar magnitude as the ones found by Moran et al. (22) when exercise intensity is raised by 20% or when dehydration is increased by 1%.

In summary, VAR exercise performed in a hot-dry environment produces higher heat storage than CON exercise despite a similar heat production. The reduction in heat dissipation could be mediated by a reduction in SKBF. VAR in the heat increases not only the thermal stress (i.e., heat storage) but also the cardiovascular (i.e., heart rate) and metabolic stress (i.e., blood lactate) indexes, which, in conjunction, makes VAR less advisable than the CON exercise mode when stress is to be avoided.

The help of Nassim Hamouti in reviewing the manuscript isgreatly appreciated. The results of the present study do notconstitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Adams WC, Mack CW, Langhans GW, Nadel ER. Effects of varied air velocity on sweating and evaporative rates during exercise. J Appl Physiol. 1992;73(6):2668-74.
2. Belding HS, Hertig BA, Kraning KK. Comparison of man's responses to pulsed and unpulsed environmental heat and exercise. J Appl Physiol. 1966;21(1):138-42.
3. Bogdanis GC, Nevill ME, Boobis LH, Lakomy HK. Contribution of phosphocreatine and aerobic metabolism to energy supply during repeated sprint exercise. J Appl Physiol. 1996;80(3):876-84.
4. Brouwer E. On simple formulae for calculating the heat expenditure and the quantities of carbohydrate and fat oxidized in metabolism of men and animals, from gaseous exchange (oxygen intake and carbonic acid output) and urine-N. Acta Physiol Pharmacol Neerl. 1957;6:795-802.
5. Buono MJ, Ball KD, Kolkhorst FW. Sodium ion concentration vs. sweat rate relationship in humans. J Appl Physiol. 2007;103(3):990-4.
6. Colin J, Tmbal J, Houdas Y, Boutelier C, Guieu JD. Computation of mean body temperature from rectal and skin temperatures. J Appl Physiol. 1971;31:484-9.
7. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol. 1974;37(2):247-8.
8. Drust B, Reilly T, Cable NT. Physiological responses to laboratory-based soccer-specific intermittent and continuous exercise. J Sports Sci. 2000;18(11):885-92.
9. Du Bois DBS, Du Bois EF. A formula to estimate the approximate surface area if height and weight be known. Arch Int Med. 1916;17:963-71.
10. Ekblom B, Greenleaf CJ, Greenleaf JE, Hermansen L. Temperature regulation during continuous and intermittent exercise in man. Acta Physiol Scand. 1971;81(1):1-10.
11. Fritzsche RG, Coyle EF. Cutaneous blood flow during exercise is higher in endurance-trained humans. J Appl Physiol. 2000;88(2):738-44.
12. Gonzalez-Alonso J, Mora-Rodriguez R, Below PR, Coyle EF. Dehydration reduces cardiac output and increases systemic and cutaneous vascular resistance during exercise. J Appl Physiol. 1995;79(5):1487-96.
13. Gonzalez-Alonso J, Mora-Rodriguez R, Coyle EF. Stroke volume during exercise: interaction of environment and hydration. Am J Physiol. 2000;278(2):H321-30.
14. Gonzalez-Alonso J, Quistorff B, Krustrup P, Bangsbo J, Saltin B. Heat production in human skeletal muscle at the onset of intense dynamic exercise. J Physiol. 2000;524(2):603-15.
15. Kranning KK, II, Gonzalez RR. Physiological consequences of intermittent exercise during compensable and uncompensable heat stress. J Appl Physiol. 1991;71(6):2138-45.
16. Krustrup P, González-Alonso J, Quistorff B, Bangsbo J. Muscle heat production and anaerobic energy turnover during repeated intense dynamic exercise in humans. J Physiol. 2001;1(536):947-56.
17. Lind AR. A physiological criterion for setting thermal environmentallimits for everyday work. J Appl Physiol. 1963;18:51-6.
18. Mitchell J, Nadel E, Stolwijk J. Respiratory weight losses during exercise. J Appl Physiol. 1972;32(4):474-6.
19. Montain SJ, Latzka WA, Sawka MN. Control of thermoregulatory sweating is altered by hydration level and exercise intensity. J Appl Physiol. 1995;79(5):1434-9.
20. Mora-Rodriguez R, Del Coso J, Aguado-Jimenez R, Estevez E. Separate and combined effects of airflow and rehydration during exercise in the heat. Med Sci Sports Exerc. 2007;39(10):1720-6.
21. Mora-Rodriguez R, González-Alonso JL, Below PR, Coyle EF. Plasma catecholamines and hyperglycaemia influence thermoregulation in man during prolonged exercise in the heat. J Physiol. 1996;491(2):529-40.
22. Moran DS, Montain SJ, Pandolf KB. Evaluation of different levels of hydration using a new physiological strain index. Am J Physiol. 1998;275(3):R854-60.
23. Moran DS, Shitzer A, Pandolf KB. A physiological strain index to evaluate heat stress. Am J Physiol. 1998;275(44):R129-34.
24. Ramanathan N. A new weighthing system for mean surface temperature of the human body. J Appl Physiol. 1964;19(3):531-3.
25. Rowell LB. General principles of vascular control. In: Human Circulation: Regulation during Physical Stress. New York (NY): Oxford University Press, Inc.; 1986. pp. 8-43.
26. Saltin B, Hermansen L. Esophageal, rectal, and muscle temperature during exercise. J Appl Physiol. 1966;21(6):1757-62.
27. Smolander J, Holmer I. Individual response to physical work in the heat in relation to sweating and skin blood flow. Int Arch Occup Environ Health. 1991;63(3):225-6.
28. Smolander J, Saalo J, Korhonen O. Effect of work load on cutaneous vascular response to exercise. J Appl Physiol. 1991;71(4):1614-9.
29. Taylor WF, Johnson JM, Kosiba WA. Roles of absolute and relative load in skin vasoconstrictor responses to exercise. J Appl Physiol. 1990;69(3):1131-6.
30. Vogt S, Heinrich L, Schumacher YO, et al. Power output during stage racing in professional road cycling. Med Sci Sports Exerc. 2006;38(1):147-51.


©2008The American College of Sports Medicine