Journal Logo

APPLIED SCIENCES: Physical Fitness and Performance

Clothing fabric does not affect thermoregulation during exercise in moderate heat


Author Information
Medicine & Science in Sports & Exercise: December 2001 - Volume 33 - Issue 12 - p 2124-2130
  • Free


In humans, the primary means of cooling the body during exercise is through the evaporation of sweat from the skin surface (25). Clothing generally represents a layer of insulation and, as such, imposes a barrier to heat transfer (3). Clothing can also hinder sweat evaporation from the skin (25). Increases in skin and core temperature and a reduction in cooling efficiency are observed when clothing interferes with the evaporation of sweat from the skin (11,17). The rate of water evaporation from the wet skin surface is dependent on the air velocity and skin-clothing-ambient air vapor pressure gradients (2). Thus, in addition to its insulative properties, the effect of clothing on evaporation is also of critical concern when thermoregulatory parameters are discussed. With this in mind, many fabrics have been introduced to the athletic apparel market with manufacturers’ claims of improved evaporative characteristics. It has been suggested that these improved evaporative characteristics are a result of improved vapor permeation, which leads to increased sweat evaporation, and potentially lower skin and core temperatures, as well as reduced evaporative water loss. Increases in skin wettedness can affect thermal comfort (17).

The purpose of the present study was to investigate whether the proposed thermoregulatory advantages of a fabric with improved evaporative characteristics could be measured. Specifically, we questioned whether these differences would be detectable when the amount of skin covered by the clothing and the environmental conditions used were similar to those commonly incurred during running, cycling, and triathlon competitions. It was hypothesized that important physiological variables classically accepted to be affected by exercise in the heat (i.e., skin temperature, core temperature, heart rate, and sweat loss) would show less perturbation with garments made from a fabric with improved evaporative characteristics as compared with ensembles of more traditional fabrics. In this report, we demonstrate that neither modest differences in the amount of clothing worn nor the fabric characteristics of the clothing alter physiological, thermoregulatory, or comfort sensation responses during exercise in a moderately warm environment.



Eight men volunteered and gave informed written consent after written and verbal explanations of the content and intent of the study. All subjects were healthy and nonsmoking, with no history of cardiopulmonary disease. Approval was previously granted by the Indiana University Committee for the Protection of Human Subjects. All experiments took place within the Human Performance Laboratory. The physical characteristics of the subjects are listed in Table 1.

Table 1
Table 1:
Subject characteristics (N = 8).


Testing occurred in Bloomington, Indiana, during the month of January. All subjects were well trained during this time and, as a result, were assumed to be heat acclimated (18). Each subject performed one running treadmill test for maximal aerobic capacity (V̇O2max) and three randomly assigned running submaximal treadmill tests for determination of the thermoregulatory responses to each clothing ensemble. On the morning of each test, the subjects were instructed to ingest 2 L of water 2 h before the start of their testing to ensure euhydration. Subjects were tested every other day, in the morning between 0800 and 1100 h.

The initial mass of each piece of the clothing ensemble to be tested was determined and the garment stored in a sealed plastic freezer bag to minimize evaporation from or water vapor absorption by the clothing before each trial. All ensembles were stored in the environmental chamber at 30°C before testing. Before the beginning of each test, the subjects voided and their body mass (WBODY) was determined. Next, a vinyl-covered rectal thermistor (Model 401, Yellow Springs Instruments, Yellow Springs, OH) (TRE) was self-inserted to a distance approximately 12 cm beyond the anal sphincter and was secured in place with gauze wrap. It has recently been demonstrated that at the end of steady-state exercise, TRE is a valid estimate of core temperature (9). Eight surface thermistors (Model 409B, Yellow Springs Instruments) were anatomically placed according to sites as defined by Stolwijk and Hardy as follows: head, chest, back, upper arm, lower arm, hand, thigh, and calf (23). The ambient dry bulb temperature and black bulb temperature (Model 405, Yellow Springs Instruments) were continuously recorded to ensure environmental conditions. All thermistors were interfaced to a switching box (Model 4002, Yellow Springs Instruments) and thermometer (Model 4009A, Yellow Springs Instruments). Before testing, all thermistors were calibrated against an NIST traceable thermometer (Model 21002, Syborn/Taylor, Fletcher, NC). On completion of the thermistor placement, subjects dressed and initial readings were taken at 22°C. After the initial reading, subjects were escorted into the environmental chamber and the test begun.

Each trial consisted of 20 min of quiet, seated rest in the environmental chamber, 30 min of treadmill running at 70% of V̇O2max, 15 min of treadmill walking at 40% of V̇O2max, and 15 min of seated rest. The environmental conditions consisted of a temperature of 30 ± 1°C and a relative humidity of 35 ± 5%. During pre- and postexercise rest, a simulated wind of 3 km·h−1 was applied. During the 30 min of running, a simulated wind of ∼11 km·h−1 was applied, whereas a simulated wind of ∼6 km·h−1 was applied during walking. These wind speeds were chosen to match with the subjects’ running and walking speeds, respectively.

Immediately after the completion of each test, the subject disrobed and the garments were placed in the same plastic freezer bags and sealed. The subject then removed all thermistors. Postexercise values for WBODY and the mass of the garments were then immediately determined. Body mass loss (WTLOSS) was calculated as the difference in pre- and postexercise WBODY, which was determined using an electronic balance (Sauter, Germany) accurate to ± 10 g. The mass of the clothing was determined using a triple arm balance (American Scientific Products, McGaw, IL) accurate to ± 0.01 g.

Minute ventilation (V̇E), oxygen uptake (V̇O2), and carbon dioxide production (V̇CO2) were monitored continuously from the beginning of the test until 10 min into the treadmill running bout and from the completion of treadmill running until the end of the test via open circuit spirometry as described previously (10). Heart rate (fc) was monitored continuously (Vantage XL, Polar Vantage, Finland). Temperature readings were recorded every 5 min. Subject perceptions of thermal comfort and sweating sensations were obtained every 5 min (24). The scales for each sensation are listed in Table 2.

Table 2
Table 2:
Subjective sensations.


The clothing consisted of three ensembles: seminude (S-N), a synthetic material (SYN), and cotton (COT). The S-N ensemble consisted of a Lycra racing swimsuit, standard polyester anklet socks, and running shoes. The SYN ensemble was made from a synthetic fabric claimed by the manufacturer to promote sweat evaporation. Both the SYN and COT ensembles consisted of a crew neck, short sleeve T-shirt, form-fitting cycling shorts that extended to midthigh, anklet socks, and running shoes. The same running shoes were used for all three ensembles. Total ensemble intrinsic insulation, less shoes, was estimated as 0.09 clo, 0.27 clo, and 0.28 clo for S-N, SYN, and COT, respectively (15). Each ensemble was prewashed and worn once. Each subject had their own set of clothing for each of the three clothing ensembles.


Metabolic energy production (MEP) was calculated from the measurements of V̇O2, using the equation of Gagge and Nishi (8) :


where, RER is the respiratory exchange ratio, V̇O2 is oxygen consumption in L·min−1, and k is the energy equivalent of oxygen (5.873 W·h−1·L−1·min−1).

Mass of sweat evaporated (SWE) from the subjects during the exercise bout was determined from WTLOSS during the exercise bout, using the equations of Nielsen and Endrusick (16) :


where RESLOSS is the water loss attributable to respiration:


and METLOSS is the metabolic mass loss:


where Pa is the ambient water vapor pressure in pascals and AD is the DuBois surface area (6). The mass of nonevaporated sweat (SWNE) was calculated as the difference in clothing mass measured before and after exercise. It is assumed that all nonevaporated sweat was absorbed by the clothing, since no dripping was observed. Sweat efficiency was calculated as the amount of sweat evaporated divided by total sweat produced.

Mean skin temperature (TSK) was calculated as an average of eight skin sites according to Gagge and Nishi (8) :


Body temperature (TBODY) was calculated according to Hardy and DuBois (12) :


A computation of “covered” mean skin temperature (TCSK) and “shirt” mean skin temperature (TSHIRT) was made using a modification of the covered skin equation of Nielsen and Endrusick (16) :



Statistical analysis.

Repeated-measures analysis of variance (ANOVA) was used to determine if any significant effect occurred as a result of the different clothing ensembles. Significant F values were further analyzed using a Bonferroni multiple comparison test. Analyses were run for the data at time 0, 20, 30, 50, 65, and 80 min, respectively. Significance was established at P ≤ 0.05 for all statistics.


Temperature measurements.

There was no difference in TRE or TBODY between clothing ensembles during preexercise rest (Fig. 1 A and B). During preexercise rest, TSK (Fig. 1 C), TCSK (data not shown), and TSHIRT (data not shown) were all lower for S-N. There were no differences in any temperature variable between ensembles during running, walking, and postexercise rest. In response to changes in exercise conditions, there were similar rates of change in TBODY, TRE, and TSK between all ensembles (Fig. 1).

Temporal changes in mean body temperature (A), rectal temperature (B), and mean skin temperature (C) among different clothing ensembles. Values represent means. *Significantly different than COT or SYN (P ≤ 0.05).

Metabolic measurements.

There was no difference in V̇O2 between ensembles at any time point (Fig. 2 A). During both walking and postexercise rest, both RER (Fig. 2 B) and V̇CO2 (data not shown) were significantly lower for SYN compared with COT. Figure 3 C demonstrates no difference in fc during rest, running, or walking with respect to clothing ensemble.

Oxygen consumption (V̇O2) (A) and respiratory exchange ratio (RER) (B) among different clothing ensembles as measured during preexercise seated rest (PRE), treadmill running (RUN), treadmill walking (WALK), and postexercise seated rest (POST). Temporal changes in heart rate (fc) among different clothing ensembles (C). Values are means ± SE. *SYN significantly different than COT (P ≤ 0.05).
Scaled sensation reactions of thermal comfort (A) and sweating (B) among different clothing ensembles as measured during preexercise seated rest (PRE), treadmill running (RUN), end of treadmill running (END), treadmill walking (WALK), and postexercise seated rest (POST). Values are means ± SE (P ≥ 0.05).

Perceptions of clothing ensembles.

To determine if differences in clothing fabrics altered perception, subjects were asked to rate their thermal comfort and sweating sensations. There were no differences in thermal comfort (Fig. 3 A) or sweating (Fig. 3 B) sensations between ensembles. At rest, subjects were thermal neutral and dry. During running, there was a progressive increase in thermal comfort and sweating sensations from 10 min (RUN) to the end of the running (END). These sensations decreased during walking and postexercise rest.

Before to after exercise.

There were no differences in MEP, WTLOSS, METLOSS, or RESLOSS between ensembles (data not shown). There was no difference in SWE between ensembles (Fig. 4 A). There was a difference in the retention of sweat by the clothing ensembles, excluding shoes, which were the same between trials, with COT retaining approximately three times the amount of sweat as either S-N or SYN (Fig. 4 B). Sweat efficiency was lower in COT than either S-N or SYN (Fig. 4 C).

Mass of sweat evaporated from (SWE) (A) and retention of sweat (excluding shoes) by (B) the different clothing ensembles. Values are means ± SE. *Significantly different than S-N and SYN (P ≤ 0.05).


The main finding of the present study is that in the environmental conditions tested, a synthetic fabric designed to promote evaporation does not provide any thermoregulatory, physiological, or comfort advantage when compared with a traditional cotton ensemble during or after exercise. The S-N condition was included in this experiment to represent an ensemble with a minimal insulative load and a maximal evaporative capacity. During and after exercise, S-N provided no thermoregulatory or metabolic advantage when compared to either SYN or COT. These results suggest that in moderately warm temperature and dry conditions, clothing as used in this study does not improve or impair temperature regulation.

Temperature regulation.

During preexercise rest, TSK was higher for SYN and COT than S-N (Fig. 1 C). This higher skin temperature can be attributed to the increase in clothed surface area for these ensembles, since both the SYN and COT consisted of a T-shirt and midthigh shorts, which covered a greater skin surface area. The issue of covered skin surface area does address an important issue concerning the effects the small amount of clothing would have on TSK and TBODY. In all ensembles, a substantial amount of the skin surface area was not covered with the respective materials. As a result, the effect that each fabric would have on TSK and TBODY would be reduced. In an attempt to address this issue, we modified the TSK equation of Nielsen and Endrusick (16) to develop the calculations for TCSK and TSHIRT, as a means of analyzing the effect the fabric had on the skin surface area it actually covered. However, there was no difference in TCSK, TSHIRT, or TTHIGH between SYN and COT (data not shown). To observe a difference between these fabrics at these environmental conditions, future studies may need to consist of full-length garments made from each clothing fabric. Jeong and Tokura have demonstrated that at normal room temperature and humidity (20°C and 65% relative humidity), TRE is kept significantly higher during rest and recovery from exercise while wearing full-length garments compared with clothing covering trunk, upper arms, and thighs (13). However, it should be noted that the ensembles tested here are representative of clothing ensembles used in running, cycling, and triathlon training and competitions in moderate and hot conditions.

After preexercise rest, there were no differences in any of the measured temperature readings during running, walking, or postexercise rest (Fig. 1). Body temperature is a highly regulated physiological variable; thus, a greater disturbance may be needed to see differences in these temperature variables. This disturbance could be in one of several different forms. First, a greater physiological stress could be imposed by increasing the exercise time or intensity. Core temperature does increase linearly with increases in exercise intensity (22); thus, a significant increase in work intensity could produce a sufficient heat load during which an evaporative fabric would prove beneficial in the maintenance of body temperature. However, as exercise intensity increases, the time to fatigue decreases (20). Therefore, an increase in exercise intensity must not be so severe as to limit exercise duration to an extent that the thermoregulatory system is not stressed.

Second, the severity of the environmental conditions could be increased, specifically, by increasing temperature. In exercise-trained men, the environmental temperature significantly influences the thermoregulatory response to exercise (18,19). At the same workload, an increase in environment temperature from 40°C to 50°C increases rectal temperature and exercise heart rate. In addition, exercise in 50°C, which results in near to complete exhaustion in trained men, results in linear increases in both rectal temperature and heart rate throughout the exercise bout, physiological responses that are not observed during similar exercise in 40°C. Increases in humidity would not prove fruitful. The proposed thermoregulatory advantage of the synthetic material tested in this report is enhanced evaporation (Fig. 4 B). Exercise in a high-humidity environment would negate this advantage because of the reduction in the water vapor pressure gradient between the wetted skin and the environment. In many athletic events, the wearing of clothing in warm or hot environments creates conditions of uncompensable heat stress where the evaporative heat loss required (Ereq) to maintain a thermal steady state exceeds the evaporative heat loss capacity (Emax) of the environment (5). Whereas Ereq is determined predominantly by metabolic rate, Emax can be altered by changes in ambient vapor pressure, air movement, temperature, and clothing (1,4). Thus, differences in temperature regulation between fabrics may require harsher environmental condition (decreased Emax), which would lower the tolerance for changes in heat production as a result of increased metabolic work (increased Ereq).

Metabolic regulation.

There was no difference in V̇O2 between clothing ensembles during rest, exercise, or walking (Fig. 2). Given that the treadmill speed was held constant for any subject between exercise bouts, this would be anticipated. We did not observe any difference in heart rate at any time point. During exercise, clothing fabric can influence exercise heart rate (14). Kwon et al. suggest that these differences may reflect differences in tympanic temperature (14); however, we did not observe any difference in temperature during the experimental bouts (Fig. 1). Thus, the thermoregulatory stress associated with each exercise bout was similar.

The RER was lower in SYN than COT during walking and postexercise rest (Fig. 2 B). One possibility is that the differences in RER observed could be related to differences in substrate availability. Diet before each exercise bout was not controlled for in this study. It is possible that the difference in RER during postrunning walking and rest could result from differences in intramuscular temperature, which would have had to be warmer in COT than SYN, in spite of similar skin temperatures and the somewhat limited coverage of the clothing ensembles. Glycolysis is enhanced during exercise in the heat (41°C) compared with exercise in the cold (9°C) (7). However, this does not seem reasonable.

Thermal comfort.

Given the obvious suggestion that exercise tolerance can be influenced by perceived exertion and that increases in skin wettedness can affect thermal comfort (17,21), we measured thermal comfort and sweating sensations (Fig. 3). In spite of differences in the amount of skin surface covered by clothing and differences in fabric material, we did not observe any difference in thermal comfort or sweating sensations between exercise ensembles at any time point.

Sweat loss and sweat retention.

There were no significant differences in WTLOSS or SWE between clothing ensembles, suggesting that in the conditions tested neither clothing fabric nor skin coverage affects sweat loss. This is in contrast to Kwon et al., who demonstrated a difference in sweating rate between fabrics of different physical properties (14). It might be suggested that our inability to detect differences in either WTLOSS or SWE may be a function of the precision (± 10 g) of the electronic balance used for the measurements. However, the inherent variability of the sweating response demonstrated in our subjects between conditions prevents a difference in SWE of approximately 100 g from becoming significant (Fig. 4 A), a difference that is much greater than the precision of our balance.

The COT ensemble retained a significantly greater amount of sweat than either S-N or SYN, even when measured 30 min after the completion of running (Fig. 4 B). It is therefore evident that the synthetic material that constituted the SYN ensemble does promote greater evaporation, as claimed by the manufacturer. In spite of this, the COT ensemble did not impair temperature regulation in a warm environment. However, this may not be the case in a cold environment, where postexercise evaporation may create a significant cooling effect, especially during intermittent exercise. Thus, the use of fabrics that promote evaporation may have a potentially greater benefit during exercise in a cold environment.

In summary, a synthetic fabric that improves sweat evaporation does not provide a thermoregulatory, physiological, or comfort sensation advantage when compared to cotton in moderately warm environmental conditions. However, neither the synthetic nor cotton fabrics represent a thermoregulatory, physiological, or comfort sensation disadvantage when compared to a seminude ensemble. Thus, in moderately warm environmental conditions, neither the addition of a modest amount of clothing nor the fabric characteristics of the clothing alters temperature, physiological, or comfort sensation responses.

This study was funded in part by a grant from the National Institute for Fitness and Sport (NIFS) and the Indiana University School of Health Physical, Education, and Recreation (HPER).

Address for correspondence: Timothy P. Gavin, Ph.D., Human Performance Laboratory, 371 Ward Sports Medicine Bldg., East Carolina University, Greenville, NC 27858; E-mail:


1. Adams, W. C., G. W. Mack, G. W. Langhans, and E. R. Nadel. Effects of varied air velocity on sweating and evaporative rates during exercise. J. Appl. Physiol. 73: 2668–2674, 1992.
2. Berglund, L. G., and R. R. Gonzalez. Evaporation of sweat from sedentary man in humid environments. J. Appl. Physiol. 42: 767–772, 1977.
3. Burton, A. C., and O. G. Edholm. Man in a Cold Environment. London: E. A. Arnold Ltd., 1955, pp. 53–55.
4. Candas, V., and A. Hoeft. Clothing assessment and effects on the thermoregulatory responses of man working in humid heat. Ergonomics 38: 115–127, 1995.
5. Cheung, S. S., and T. M. McLellan. Heat acclimation, aerobic fitness, and hydration effects on tolerance during uncompensable heat stress. J. Appl. Physiol. 84: 1731–1739, 1998.
6. Dubois, D., and E. F. Dubois. A formula to estimate the approximate surface area if height and weight be known. Arch. Intern. Med. 17: 863–871, 1916.
7. Fink, W. J., D. L. Costill, and P. J. van Handel. Leg muscle metabolism during exercise in the heat and cold. Eur. J. Appl. Physiol. 34: 183–190, 1975.
8. Gagge, A. P., and Y. Nishi. Heat exchange between human skin surface and thermal environment. In: Handbook of Physiology, D. H. K. Lee (Ed.). Baltimore: Waverly Press, 1969, pp. 69–92.
9. Gass, E. M., and G. C. Gass. Rectal and esophageal temperatures during upper- and lower-body exercise. Eur. J. Appl. Physiol. 78: 38–42, 1998.
10. Gavin, T. P., P. A. Derchak, and J. M. Stager. Ventilation’s role in the decline in VO2max and SaO2 in acute hypoxic exercise. Med. Sci. Sports Exerc. 30: 195–199, 1998.
11. Givoni, B., and H. S. Belding. The cooling efficiency of sweat evaporation. In: Biometeorology, S. W. Tromp (Ed.). New York: Macmillan, 1962, pp. 304–314.
12. Hardy, J. D., and E. F. Dubois. Basal metabolism, radiation, convection, and vaporization at temperatures of 22 to 35°C. J. Nutr. 15: 477–497, 1938.
13. Jeong, W. S., and H. Tokura. Effects of wearing two different types of clothing on body temperatures during and after exercise. Int. J. Biometeorol. 33: 77–81, 1989.
14. Kwon, A., M. Kato, H. Kawamura, Y. Yanai, and H. Tokura. Physiological significance of hydrophilic and hydrophobic textile materials during intermittent exercise in humans under the influence of warm ambient temperature with and without wind. Eur. J. Appl. Physiol. 78: 487–493, 1998.
15. McCullough, E. A., and B. W. Jones. Institute for Environmental Technical Report 84–01: A Comprehensive Data Base for Estimating Clothing Insulation. Manhattan, KS: Kansas State University Press, 1984.
16. Nielsen, R., and T. Endrusick. Thermoregulatory responses to intermittent exercise are influenced by knit structure of underwear. Eur. J. Appl. Physiol. 60: 15–25, 1990.
17. Pascoe, D. D., T. A. Bellingar, and B. S. McCluskey. Clothing and exercise: II. Influence of clothing during exercise/work in environmental extremes. Sports Med. 18: 94–108, 1994.
18. Piwonka, R. W., and S. Robinson. Acclimatization of highly trained men to work in severe heat. J. Appl. Physiol. 22: 9–12, 1967.
19. Piwonka, R. W., S. Robinson, V. L. Gay, and R. S. Manalis. Preacclimatization of men to heat by training. J. Appl. Physiol. 20: 379–384, 1965.
20. Poole, D. C., and R. S. Richardson. Determinants of oxygen uptake: implications for exercise testing. Sports Med. 24: 308–320, 1997.
21. Robertson, R. J., and B. J. Noble. Perception of physical exertion: methods, mediators, and applications. In: Exercise and Sport Sciences Reviews, J. O. Holloszy (Ed.). Baltimore: Williams & Wilkins, 1997, pp. 407–452.
22. Saltin, B., and L. Hermansen. Esophageal, rectal, and muscle temperature during exercise. J. Appl. Physiol. 21: 1757–1762, 1966.
23. Stolwijk, J. A. J., and J. D. Hardy. Partitional calorimetric studies of responses of man to thermal transients. J. Appl. Physiol. 21: 967–977, 1966.
24. Vokac, Z., V. Køpke, and P. Keul. Physiological responses and thermal, humidity, and comfort sensations in wear trials with cotton and polypropylene vests. Textile Res. J. 46: 30–38, 1976.
25. W atkins , S. M. Clothing: The Portable Environment. Ames, IA: Iowa State University Publishers, 1984, pp. 1–24.


© 2001 Lippincott Williams & Wilkins, Inc.