Thermoregulation of competitive artistic gymnastic athletes and non-athlete girls exercising in the heat

It’s unclear whether the combination of intense, chronic training and heat exposure during prepubescence improves thermoregulatory responses to exercise in artistic gymnastics athletes. The objective of this study was to compare thermoregulatory and perceptual responses between artistic gymnastics athletes and non-athlete girls while exercising both in heat and thermoneutral conditions. Seven athletes (8.7 ± 1.3 yrs) and 7 nonathletes (9.4 ± 1.5 yrs) cycled for 30 min at load (W) of ~55% VO2peak, on two separate occasions in a randomized order: heat (35 ̊C, 40% relative humidity) and thermoneutral conditions (24 ̊C, 50% relative humidity). Rectal temperature, heart rate, rate of perceived exertion, thermal sensation, thermal comfort and irritability were measured throughout the exercise. Initial rectal temperature was similar between athletes and non-athletes in both heat (37.2 ± 0.4 vs. 37.4 ± 0.2 ̊C, respectively) and thermoneutral conditions (37.3 ± 0.2 vs. 37.3 ± 0.3 ̊C). Final rectal temperature was similar between groups (38.0 ± 0.2 vs. 38.2 ± 0.2 ̊C in heat and 37.8 ± 0.2 vs. 37.9 ± 0.2 ̊C in thermoneutral conditions). Initial heart rate was lower in athletes in the heat (76 ± 7 vs. 91 ± 11 bpm, P = 0.01); however, throughout cycling, it became similar between groups. Athletes reported similar perceptual responses compared to non-athletes, with the exception of higher thermal comfort in the 10th minute of exercise in thermoneutral conditions (P = 0.003). It was concluded that athletes were similar to non-athletes with respect to thermoregulatory and perceptual responses during 30 min of cycling at similar relative intensities.


INTRODUCTION
Girls participating in artistic gymnastics (AG) usually start focusing on competitive training at very early ages (5-7 years old) and prior to puberty 1 .Such training involves intense efforts and a combination of aerobics, flexibility and strength exercises each session that may last ~ 3 to 4 h, 5 to 6 days per week 1 .In addition, AG training may be held in a warm environment, which would add an additional challenge to the training regimen.Such chronic exposure to heat may result in thermoregulatory adaptations, which may prevent excessive and potentially dangerous rise in body temperature during exercise due to efficient heat dissipatory mechanisms, such as improved efficiency of evaporative heat loss as a result of alterations in the sweating response, greater plasma volume and stability of cardiovascular function.However, it is unclear whether such intense and chronic training, as well as the heat adaptation during prepubescence, improves thermoregulatory responses to exercise as observed in adults 2,3 .There is no information about thermoregulatory responses during exercise in the heat in young AG athletes.Most of the information involving thermoregulatory responses in athletes and non-athletes are from studies performed in the adult population.
Exercise training in adults improves thermoregulatory effectiveness, which may improve exercise tolerance and perceptual responses [2][3][4][5] .Exercise training improves thermal tolerance, which may partially result from improved evaporative cooling 4,5 .Also, aerobic training may increase the slope of body core temperature-sweat rate 6 and reduce sweating threshold 7 .The thermoregulatory responses in athletes are thought to be comparable to those produced by heat adaptation.Heat adaptation may be affected by years of training, as a response to repeated stress application.In prepubescence, the effect of physical adaptations on thermoregulatory responses is still inconsistent 8 , which may be due to insufficient years of training to achieve such adaptations.The limited studies that have investigated thermoregulatory responses of exercising children have focused on hydration aspects [8][9][10] and groups of non-athletes 11 .
The impact of exercise in the heat on thermoregulatory and perceptual responses between AG athletes and non-athletes is still unclear.The purpose of this study was to compare the thermoregulatory and perceptual responses between two groups of prepubescent girls (AG athletes and non-athletes) cycling at similar relative intensities in both heat (HC) and thermoneutral (TC) conditions.

Participants
Fourteen prepubescent girls (7 AG athletes and 7 non-athletes ─ but physically active) participated in this study.Girls in the group of athletes trained AG for at least 5 days per week, > 3 h per day, during the last year and also participated in competitions.Non-athlete girls were physically active (level ≥ 3), according to the Physical Activity Questionnaire for Older Children (PAQ-C) 12 .Girls did not present any chronic disease and were not taking any medication at the time of participation.The study approval was obtained from the Federal University of Rio Grande do Sul Research Ethics Board (Protocol # 19624).All procedures respected the Declaration of Helsinki (1975) principles and all participants and respective guardians were fully informed about potential risks and freely signed the informed consent and assent forms.

Procedures
This study consisted of a preliminary session and two experimental sessions (separated by 4-7 days).Experimental sessions were performed between January and April, where temperatures ranged from 28 to 42˚C and relative humidity from 40 to 95% (Southern Brazil).Data were obtained from the National Meteorological Institute.Participants were similarly acclimatized to heat.

Preliminary session
Stage 1 of biological maturation (prepubertal) was confirmed through observation of breast and the pubic hair development 13 .Height (stadiometer, Urano PS 180) and body mass (G-TECH scale, model BALGLA3C, GTech Technology Ltd, Zhuhai, Guangdong, China) were assessed wearing minimal clothing (shorts and tops with child barefoot).Body surface area (BSA) was determined according to the equation of Dubois & Dubois 14 .Fat mass and total body mass were measured via dual-energy X-ray absorptiometry (DXA) (GE-LUNAR Prodigy, GE medical systems, Madison, WI).
The progressive exercise test to determine VO 2peak was conducted on a cycle ergometer (Ergo Fit 167, five-watt resolution; ERGO-FIT, Pirmasens, Germany) using the McMaster protocol 15 .The test began at 25 W and increased in 25 W increments every 2 min while maintaining cadence between 60 and 80 rpm.All girls were verbally encouraged to give their best performance.An open-circuit indirect calorimeter was used (Medgraphics O 2 and CO 2 Analyzer CPX/D [breath-by-breath]; Medical Graphics Corporation, St Paul, Minnesota), and peak was considered to be the highest VO 2 value.The test ended when two of the five following criteria were reached: 1) participant's request to interrupt the test, 2) inability to maintain cycling cadence above 60 rpm, 3) HR (Polar S610; Polar Electro Oy, Kempele, Finland) > 200 bpm, 4) rate of perceptual exertion (RPE) > 19, and 5) respiratory exchange rate (RER) > 1.0.VO 2peak was also corrected for the total muscle mass to avoid a confounder effect of fat mass and total body mass.

Experimental design
Participants were asked to avoid strenuous physical activity 24 h prior to experimental sessions and to maintain their eating habits during interval between sessions.The experimental protocol was identical for both sessions except for environmental conditions (HC or TC) for which the order was randomized by lottery during the preliminary session.All experimental trials were scheduled in the morning to eliminate systematic differences between groups due to circadian variation, and participants ate a standardized breakfast that consisted of two portions of white bread (40 g), one portion of jelly (15 g), 200 mL of fruit juice and 200 mL of chocolate milk.
On arrival, each participant provided a urine sample, which was immediately analyzed for urine specific gravity (USG, cut-off value < 1.020) (Atago refractometer, URC-Ne, Japan; ATAGO CO Ltd, Tokyo, Japan) and color 16 , using an 8-point scale that ranges from very pale yellow (1) to brownish green (8), to ensure that initial hydration status was similar between groups.Body mass was measured with participants dressed in shorts, top and barefoot.Each participant was equipped with a HR monitor and T re was measured using a flexible thermometer (RET-1, 0.1˚C resolution; Physitemp Instruments, Inc, Clifton, New Jersey) with a disposable cover, which was inserted 10 cm beyond the anal sphincter.
Participants received standardized instructions on how to answer the following three perceptual scales: thermal sensation 17 , thermal comfort 17 and irritability.Thermal sensation was categorized on a 9-point scale, ranging from (1) "very cold" to ( 9) "very hot".Thermal comfort was assesses on a 6-point scale ranging from (1) "very comfortable" to ( 6) "very uncomfortable", and irritability was assesses on a 5-point scale ranging from (1) "nothing" to (5) "very strong".
Participants exercised while exposed to environmental conditions inside a climate chamber (Modular Construction Walk-In Temperature and Humidity Test Chamber; Russells Technical Products, Holland, Michigan).Environmental conditions were similar between AG athletes and non-athlete girls for HC (35.2 ± 0.9˚C and 38.4 ± 3.9 % relative humidity) and TC (24.0 ± 1.1˚C and 50 ± 8.4 % relative humidity).
Upon entering the climate chamber, participants rested for five minutes, while initial T re , HR, RPE, thermal sensation, thermal comfort, and irritability measurements were recorded.The exercise protocol consisted of pedaling on a cycle ergometer (Ergo Fit; Toledo, Spain) for 30 min with load (W) corresponding to 55% of the VO 2peak, and cadence between 60 and 80 rpm.At the 15 th min of cycling of submaximal sessions (HC and TC), VO 2 was measured for three minutes to check the target intensity.T re , HR, and RPE were recorded every five min, and thermal sensation, thermal comfort, and irritability were measured at 0, 10, 20, and 30 min.Exercise was stopped early if T re achieved values > 39˚C or, two events of the following 4 criteria simultaneously occurred: HR 200 bpm, RPE > 19, heat exhaustion (nausea, disorientation, headache, and dizziness), and inability to maintain cycling frequency between 60 and 80 rpm 18 .
During cycling, water (~15 ˚C) was available for consumption ad libitum, and the following instruction was given: "Water will remain available within your reach; you can drink whenever you wish".The bot-tle was weighed (Ohaus Compact Scale, CS2000; Ohaus Compact scale, Parsippany, New Jersey) before and after cycling to calculate water volume intake.Following the completion of exercise, girls urinated and dried their bodies, and their body mass was measured to calculate sweat volume (∆ body mass before and after cycling, corrected by the volume of water intake and urine volume).

Statistical Analyses
To verify data normality and homogeneity of variance, the Shapiro-Wilk test and the Levene test were used, respectively.For independent samples, the t-test was used for intergroup comparisons.Condition-specific 2-way ANOVAs were used to compare groups over time.When group × time interaction was found, post-hoc analyses were performed using Bonferroni test.All data are expressed as mean and standard deviation (SD).Statistical significance was set at ≤ 0.05, and all analyses were performed using the statistical SPSS software version 18.0 (SPSS Inc, Chicago, Illinois).

RESULTS
Table 1 shows that most of the physical characteristics of AG group were similar to that of non-athlete group, with the exception that they showed lower body fat (kg) (P = 0.04) and % body fat (P = 0.01).BSA (m 2 and m 2. kg -1 ) and aerobic fitness (mL .min -1 , mL .kg -1.min -1 , and mL .kg -1MuscleMass.min -1 ) were similar between groups.Urine color and USG were similar between groups prior to exercise both in HC (2.6 ± 1.3 and 1.018 ± 0.007 in athletes vs. 3.4 ± 1.8 and 1.014 ± 0.007 in non-athletes) and TC (2.4 ± 1.3 and 1.018 ± 0.007 in athletes vs. 3.2 ± 0.8 and 1.016 ± 0.008 in non-athletes).At the end of the exercise session, urine color and USG remained similar between groups at ambient conditions, and continued to indicate mild hypohydration.

DISCUSSION
The present study showed that, compared to non-athletes, when female AG prepubescent athletes cycled both in HC and TC, they showed similar: 1) initial T re and T re and HR responses while exercising; 2) sweat volume; and 3) RPE, thermal sensation, thermal comfort and irritability.The present study suggested that AG training of at least one year did not affect thermoregulatory and perceptual responses in prepubescent AG athletes compared to their non-athlete peers, with similar acclimatization levels and VO 2peak .
T re increase was similar among prepubescent girls in both environmental conditions.Some studies 19,20 have shown that trained adult men and women have lower increases in core temperature and skin temperature compared to untrained individuals while exercising (60-120 min) in the heat (30 -45˚C).Perhaps, the similar T re responses between groups in the present study is due to the relatively shorter duration (30 min) and lower intensity (55% VO 2peak ) of the exercise protocol compared to the usual high intensity and intermittent nature of AG training sessions.However, an important point in our study is that, in addition to similarities in their previous heat exposure, non-athletics were physically active and had aerobic fitness similar to AG athletes, as indicated by their VO 2peak .It is therefore possible that differences in the magnitude of thermoregulatory responses are accentuated when young athletes are compared with sedentary individuals rather than physically active individuals.In male adults, a study 21 showed that the magnitude of VO 2peak may not be the main factor to explain thermoregulatory responses when exercise elicits similar metabolic heat production per body mass unit.In the present study, absolute VO 2peak , RER and workload were similar between groups, which also resulted in similar metabolic heat production (in W and W . kg -1 ).It is possible that fitness improvement induced by AG training in prepubescent girls is not reflected by a single graded exercise test.Therefore, this supposed similar fitness capacity between groups could be a limitation of the present study.Further studies are needed to compare athletic and non-athletic children with different physical and fitness capacity exercising in the heat.
To compare our results with those obtained with children, it is important to consider how the protocol challenged the thermoregulatory system (i.e.method selected to induce heat exchange).For example in boys, lower T re increment was observed in trained vs. untrained prepubescents; however, they were resting in a seated posture in the heat (30˚C and 70% relative humidity) with their legs immersed in hot water (43˚C) 22 .This single heat exposure protocol should be observed, as this study 22 used passive and not active (exercise) protocol to compare groups.The choice of this protocol changes the relative contribution of heat gain and loss, and also isolates the physical conditioning effect on thermoregulatory responses.When exercising, heat gain occurs primarily from metabolic heat production while heat is lost via evaporative cooling, conduction and radiation.Therefore, when passive heat is used to compare thermoregulatory responses, radiation, conduction and convection are the main routes of heat gain, and losses are mostly dependent on evaporative cooling.
Physical training, even when regularly practiced in cool environments, may indirectly improve heat tolerance to exercise as endurance adult athletes with higher Vo 2max have some thermoregulatory advantages compared to non-athletes 4 .Physical training appears to have a great impact on heat adaptation in adults 4 , although the influence of individual VO 2max, as an isolated factor affecting thermoregulatory responses, is controversial 21 .Body composition is another feature that usually differs between athletes and non-athletes.The impact of this difference in thermoregulatory responses during exercise is also controversial.A study 21 suggested that body size (mass, composition and BSA) has no great impact on core temperature increase during exercise in the heat when exercise is prescribed to elicit similar metabolic heat production per body mass unit.Other studies suggested that subjects with greater % body fat have higher capacity to store heat and greater increase in core temperature 23,24 .On the other hand, obese and non-obese pubescent boys showed no difference in thermoregulatory responses during exercise in the heat 11 .In prepubescent girls, the lean group presented greater increase in body core temperature compared to the obese group when cycling in the heat; but not at thermoneutral conditions 25 .The extent at which adiposity affects thermoregulatory responses during exercise in the heat is still unclear.In our study, despite the similar body mass between groups, body fat was ~34.5% lower in AG athletes (13.3 vs. 22% in AG athletes and non-athletes, respectively), which is a characteristic in this sport modality 26 .In our study, differences in % body fat did not affect thermoregulation effectiveness of prepubescent girls; however both groups were lean.
Similar sweating responses were observed in AG athletes and nonathletes during both HC and TC.Water loss through sweating is largely dependent on exercise pattern such as intensity and duration.In adults, as aerobic fitness increases, thermoregulatory function and heat tolerance improve which, among other adaptations, may be also due to the earlier onset of sweating and greater sweating rates 27 .Therefore, it is possible that the training lifetime of female AG prepubescent athletes is still insufficient to affect sweating responses.In early-pubertal boys (aged 8-10 years), physiological adaptations due to heat acclimatization may be achieved either by exercise in heat (43˚C) or by further training (~65% of VO 2max ) even if the climate is neutral (24˚C) 28 .Children's lower sweat rate and the greater population density of their active glands suggest that the sweat production per gland is lower than in adults, mainly in prepubescent years 15 .Children may be less dependent on evaporation and present greater reliance on heat dissipation via convection, conduction and radiation.In adults, the effects of sports training on sweating responses were observed in runners compared to sedentary males during 30 min of cycling in the heat (35˚C) at similar relative intensity (60% VO 2max ).Runners showed higher sweat rate than sedentary males 29 .A study 30 showed the effect of age and training on thermoregulatory responses in three different groups: young cyclists (aged 19-35 years) at moderate or high training levels and older cyclists (51-63 years).Participants cycled for 60 min at 70% of their VO 2max in HC (35˚C and 40% relative humidity) and TC (20˚C and 40% relative humidity) conditions.Lower sweating rates and sweat loss by evaporation were observed in both young cyclists at moderate physical training levels and older cyclists compared with the group of young and highly trained individuals.Different from trained adults that have increased sweating rate and lower sweat threshold, the effect of training on thermoregulatory responses in prepubescent girls seems to be lower 15 .
There is limited information on perceptual responses of prepubescent girls while exercising in the heat.In the present study, RPE, thermal sensation, thermal comfort and irritability were similar between groups of AG athletes and non-athletes.Only after 10 min of exercise in TC, AG athletes showed greater thermal comfort compared to non-athletes.A study 11 showed that during exercise in the heat, the thermal sensation increased in both lean and obese boys; however, in general, it was higher in the obese group.

CONCLUSION
It was concluded that female athletes were similar to non-athletes with respect to thermoregulatory and perceptual responses, perceptual exertion, thermal sensation, thermal comfort and irritability to a single bout of aerobic exercise at similar relative intensity in either HC or TC.Different from trained adults that have increased sweating rate and lower sweat threshold, the effect of training on thermoregulatory responses in prepubescent AG athletes compared to non-athlete but physically active girls seems to be lower.It is possible that differences would be observed when comparing athletes and sedentary prepubescent girls.The cross-sectional design of the present study is a limitation, as well as the small sample size.A practical application is to consider the impact of hyperthermia related to chronic and prolonged exercise training and thermoregulatory adaptation in young athletes.

Figure 1 .
Figure 1.T re during cycling in HC (A) and TC (B) conditions.HR during cycling in HC (C) and TC (D).Results are presented as mean ± SD.

Table 1 .
Physical and Physiological Characteristics of Participants by Group.
Values are expressed as the mean ± SD. * P < 0.05.