thermoregulation exercisingmen in the morning rise internal ...performed bicycle exercises at 30%...

UTTERWORTHE IN E MA NNP

Br. J. Sp. Med., Vol. 29, No. 2, pp. 113-120, 1995Elsevier Science Ltd

Printed in Great Britain0306-3674/95 $10.00 + 00

Thermoregulation of exercising men in the morningrise and evening fall phases of internal temperature

Masafumi Torii*t PhD, Hideaki Nakayamat MD PhD and Takashi SasakitMD PhD*Bioregulation and Physical Fitness Laboratory, Faculty of Engineering, Kyushu Institute of Technology,Kitakyushu 804, Japan; tDepartment of Hygiene, Faculty of Medicine, Tottori University, Yonago 683, Japan andtEnvironmental Physiology Laboratory, Ginkyo Junior College, Kumamoto 860, Japan

The purpose of this study was to compare the thermoreg-ulatory responses during exercise in the morning rise(0900 h) and evening fall phases (2000 h) in circadianvariation of body temperature. Five healthy volunteersperformed bicycle exercises at 30% and 60% of maximalaerobic power (VO2m.) at 260C with a relative humidity of50%. Whole-body sweat rate (SR), rectal (Tre), mean skin(TSk) and mean body (Tb) temperature, pulmonary ventila-tion (VE), oxygen uptake (Vo2), and carbon dioxide output(VCO2) and heart rate (HR) were measured during theexperimental period. SR during exercise at 30% VO2maxwas significantly higher at 2000 h than at 0900 h. However,the circadian variation of SR during exercise was notobserved at 60% VO2maX. At the two experimental times,there were also no significant differences in VO2, VCO2, VEand Tsk in both workloads. In HR, Tb and Tre circadianeffects were demonstrated as well as in workload levels.As Tb was plotted against SR during exercise, positivecorrelations were observed. The data showed that therewas a parallel shift in the SR to Tb relationship duringexercise in the morning and evening. This rightward shiftindicated that there was an increased Tb threshold for theonset of sweating in the evening. Resting Tb at 2000 h wassignificantly higher when compared with Tb at 0900 h. Thepresent results suggest that the circadian influence on thethermoregulatory response to exercise may be evident onlyat low workloads.

Keywords: temperature regulation, sweat rate, workintensity, circadian rhythm

The circadian rhythm in human body temperatureappears to be regulated, rather than just theconsequence of passive imbalances in the rate of heatproduction and heat loss'-4. Several of the followingthermal physiological parameters to exercise and heatstress were shown to exhibit circadian rhythmicity:sweating5-8, skin blood flow5'8 9, body temperature5-8and heart rate10. Aschoff and Heise11 estimated thatin a resting man the circadian variations of heat losswere responsible for about 75% of the range ofoscillation in internal temperatures, while the varia-tion in heat production contributed only 25%.

Hirdebrandt12 reported that the circadian control ofhuman body temperature was the result of thefollowing thermoregulatory adjustments: warming-up in the morning and cooling-down in the after-noon. However, identification of the factors thatcause the circadian fluctuation has been morecontroversial.

It was previously reported by Niwa et al.6 that therewere a circadian rhythm of thermoregulatory re-sponse. Under the condition of an environmentaltemperature at 130C with a relative humidity of 60%,the sweating response during bicycle exercise atworkload of 450kpmmin-1 (75W) was at its max-imum at 1800 h and its minimum at 0600 h, but heatproduction during exercise was identical at variousperiods during the day. However, no report wasgiven on thermoregulatory response at differentwork intensities. The aim of the present study,therefore, was to investigate temperature regulatoryresponses at two different work intensities in themorning rise and evening fall phases in the humanbody temperature, and to discuss the circadiancontrol mechanism contributed by the work factor.

Materials and methodsSubjectsThe subjects selected were five healthy male collegestudents, mean(s.e.m.) age 21.6(0.7) years, 171.8(2.7)cm in height, 68.8(2.5) kg in weight, 1.83(0.04) m2 ofbody surface area (= heighto 178 x weight0_427 x 71.84of Takahira's equations, and with maximal aerobicpower (VO2max) of 3.07(0.16)1min-1 (mean(s.e.m.)).During the first visit to the laboratory, each subjectwas oriented to the basic equipment and proceduresused in the experiment. Practice time was providedfor riding the bicycle ergometer and performing themouth piece for oxygen uptake (Vo2) measurement.Before the main experiments, each participant'sVO2max was determined by an incremental work rateprotocol on a Monark bicycle ergometer using theDouglas bag technique14. The pedalling rate was keptconstant at 50 rpm and timed with a metronome.After 2 min of pedalling with a constant load720-780kpm min- (120-130 Wi), the work intensitywas increased by 150kpm min- (30 W) every minuteup to exhaustion. Before the main experiment, we

113

Address for correspondence: Dr M. Torii, Bioregulation andPhysical Fitness Laboratory, Faculty of Engineering, KyushuInstitute of Technology, Sensui 1 -1, Tobata, Kitakyushu 804,Japan

on April 19, 2021 by guest. P

rotected by copyright.http://bjsm

.bmj.com

/B

r J Sports M

ed: first published as 10.1136/bjsm.29.2.113 on 1 June 1995. D

ownloaded from

http://bjsm.bmj.com/

Circadian variation of thermoregulation during exercise: M. Toni et al.

also measured circadian variation of oral temperaturein our subjects (Figure 1), and confirmed morning riseand evening fall in oral temperature.

Experimental protocolThe subjects arrived at the laboratory at least 1 hbefore the experiments, and sat quietly until thedesignated time. They undressed except for their

Tor (°C)37.5

37.0

36.5

36.0

35.56 8 10 12 14 16 18 20 22 24

TIME OF DA Y (hours)Figure 1. Circadian variations of mean oral temperature(Tor) of resting subjects (n = 5). Shaded areas representthe time period of the experiment. Resting oral tempera-ture shows a diurnal increase and a nocturnal decrease,followed by an early morning increase. In the figure aquadratic line indicates its equation, y = 35.3 + 0.229x-8.3160x2, r = 0.933. 0 = 1st measurement, * = 2ndmeasurement. The periods of the present experiments arerepresented by two shaded areas

/ End

underwear. The experiments were done at twodifferent times of the day, and each experiment wason a different day. The experiments were at 0900hand 2000 h, with all studies starting within 30 min ofthe target time. The four experiments were carriedout from October to early November, with the orderrandomized. The work intensities were both 30% and60% of VO2max in each subject. After sitting on a chairfor 30min at the condition of thermoneutrality, theyconducted a bicycle exercise for 40 min in a mean(s.d.)ambient temperature (Ta) of 260C(0.70C) with amean(s.d.) relative humidity (rh) of 50%(7.3%). Anexperimental work test was performed on 20 separatedays, two at each of following times: 0900 h and2000 h. No muscular work was performed for 24 hbefore any test. In the morning on an experimentalday, the subjects conducted the cycle exercisewithout taking breakfast, and in the case of theexperiments in the evening on another day kept nokaloric intake six to eighth after taking lunch (usualfoods) at 1300-1400 h.

Measurements

Sweating rate (SR) was measured continuously fromwhole body weight loss, using a bed scale (Model33B, sensitivity, ± 1.0 g, J.A. Potter Co., Southington,Conneticut, USA) technique, as described pre-viouslyl5 16. The best curve shown in Figure 2 wasdrawn through the weight record for each experi-ment and rate of whole-body weight loss was thencalibrated prior to each experiment. Respiratoryevaporative weight loss was estimated by Mitchell etal.' .

The temperatures of four skin surface locations andrectum (Tre) at a depth of 15cm from the anus wererecorded every 5 min by a copper-constantan thermo-couple recording system (AM-300, Ohkura ElectricCo. Ltd, Japan) with an accuracy of ±0.05°Cthroughout the experimental period. Mean skintemperature (TSk) and mean body temperature (Tb)

_ _ ~~~Starta

Ta 25 °Crh 55%Subj. EK

600 kpm/min

50 40 30 20 10I Time (min)0

9.30 am 9.00 am

Figure 2. A recording example of the whole body weight loss measured by a bed scale. After 15 min of exercise (at 100W =.600 kpm min-1) the weight loss increased markedly. 100 g scales were indicated in the centre of a chart. The curve of weightlosses was modified at the point a, because a thermocouple sensor dropped

114 Br J Sp Med 1995; 29(2)



.bmj.com

/B

r J Sports M


ownloaded from


Circadian variation of thermoregulation during exercise: M. Torii et al.

were calculated from Ramanathan's18 and Stolwijk-Hardy's19 equations, respectively.Heart rate (HR) was recorded by electrocardiogra-

phy with a telemeter system (Model 270 and 1418,Sanei Sottki, Japan). Vo2 was determined by theDouglas bag technique at rest and during exercise, 5-,10-, 20-, 30-, 40-min after the beginning of exerciseand during recovery. The gas samples were immedi-ately analysed for oxygen (F-3, Beckman Fullerman,California, USA) and carbon dioxide (MCD-L, HoribaSeisakusho, Japan). Expired gas volume was mea-sured with a gas meter (WT-10, Shinagawa Seisa-kusho Co, Japan).

Statistical analysisValues represent the mean ±s.e.m., and statisticallysignificant differences of mean values were assessedby a paired t test and one-way ANOVA. We alsoevaluated time of day and work intensity to thermalresponse during exercise by means of two-wayANOVA. The significance of the regression line wasalso evaluated by covariance analysis. A probabilitylevel of 0.05 or less was accepted as a significantdifference.

ResultsNo individual differences were found among thesubjects in all the data related to sweating responsesat rest and during exercise (P> 0.05, one-wayANOVA).

Figure 3 shows the time courses of Vo2 duringexercise at two different work intensities at 0900hand 2000 h. While all subjects sat quietly on a bicycleergometer, their mean(s.e.m.) Vo2 was not signifi-cantly different between 0900 h (309(19.2) ml min-',n = 10) and 2000 h (306(12.2) ml min-', n = 10). At the

V02 (I-mini1)

2.0 30%VC

mean ± s1.5 -

1.0 -

0.5

0.0

end of the two test periods at 30% VO2max and 60%VO2max, V02 measured during exercise averaged1.000(0.06) and 1.611p.06)1min-1 and 1.041 (0.05)and 1.635 (0.03) 1 min- (P> 0.05), respectively. At allphases of the two test periods, Vo2 during exercise at60% VO2max was significantly higher than 30%VO2max (two-way ANOVA). There were also nosignificant differences in carbon dioxide output andpulmonary ventilation in both workloads and at theconducted experimental time.The time courses of HR during exercise at two

different work intensities at 0900 h and 2000 h arerepresented in Figure 4. HR, at 25-40min after theonset of exercise at 60% VO2max, was significantlyhigher at 2000 h than 0900 h. However, there was nota significant difference in the HR between 0900 h and2000 h during exercise at 30% VO2max. At all phases ofthe two test periods, HR during exercise at 60%VO2max was significantly higher than 30% VO2max(two-way ANOVA).The time courses of SR during exercise at two

different work intensities at 0900h and 2000h areshown in Figure 5. SR during exercise at 30%VO2max(left), was significantly higher at 2000h than 0900h.At 60% VO2max exercise (right), SR was not signifi-cantly different, except for 20min after the onset ofexercise at 0900h in comparison with 2000h. At allphases of the two test periods, SR during exercise at60% VO2maxc was significantly higher than 30%VO2mgx (two-way ANOVA). As shown in Table 1, at30% VO2max exercise TSR and SRmax were significant-ly higher at 2000 h than at 0900 h. In contrast, otherparameters showed no significant differences be-tween 0900h and 2000h in exercise at 60% VO2max.Furthermore, at 60% VO2max exercise, the time ofreaching SRmax was faster at 2000h than at 0900h(P< 0.01), but there was not significant difference ofthe time of reaching SRmax between 0900h and 2000 h

V02 (Imminl)2.0

1.5

1.0

0.5

0.0

Exercise ExerciseRest 5 10 20 30 40 50 Rest 5 10 20 30 40 50

Time (min) Time (min)Figure 3. Time courses of Vo2 in exercising men at two different workloads, about 30% [left] and 60% [right] of VO2max in thelater morning [O] and evening [E]. Horizontal bars indicate an exercise period. Data represent mean(s.e.m.) for fivesubjects

Br J Sp Med 1995; 29(2) 115



.bmj.com

/B

r J Sports M


ownloaded from


Circadian variation of thermoregulation during exercise: M. Torn et al.

in exercise at 30% V02max. At rest insensibleperspiration also was significantly higher at 2000 hand 0900h (Table 1).

Figure 6 illustrates the time courses of Tre (top), Tsk(middle) and Tb (bottom) during exercise at two

HR (beats-miff1)

30%VO2ma)150 -

mean ± s.e.m140-

0901130- 4 2001

120 -

110

90100 <

80 -

70

different work intensities at 0900 h and 2000 h.Mean(s.d.) Tre had an initial value of 37.49(0.03)0C (n= 5), and did not change during the 40 min of work at2000 h (Figure 6, top and left). At 0900 h, Tre increasedslightly reaching values of 37.17(0.12)0C (n = 5) at

HR (beats-mif'1)

Rest 0 5 10 15 20 25 30 35 40 45 50

Time (min)

Rest 0 5 10 15 20 25 30 35 40 45 50

Time (min)Figure 4. Time courses of HR in exercising men at two different workloads, about 30% [left] and 60% [right] of Vo2m"X in thelater morning [0] and evening [0]. Horizontal bars indicate an exercise period. Data represent mean(s.e.m.) for fivesubjects. Significantly different (paired t test), O.900h vs. 2000h, *P<0.05, **P<0.01, ***P<0.001

SR (g.mh2.h) SR (g.ni24V)

I I T-- I I I

test 0 5 10 15 20 25 30I

35 40 45 50

300 -

250 -

200 -

150 -

100 -

50

URestO 5 10 15 20

I I I

25 30 35 40 45 50

Time (min) Time (min)Figure 5. Time courses of SR in exercising men at two different workloads, about 30% [left] and 60% [right] of VO2ma, in thelater morning [0] and evening [0]. Horizontal bars indicate an exercise period. Data represnt mean(s.e.m.) for fivesubjects. Significantly different (paired t test) 0900 h vs. 2000 h, *P< 0.05, **P< 0.01

116 Br J Sp Med 1995; 29(2)

300 -'

250 -

200 -

150 -

100 -

50 -

0R

30% V02max

mean ± s.e.m

[ 0902I* 2000*

*

**

**

** **

*60%V02max

****

J4

:$



.bmj.com

/B

r J Sports M


ownloaded from



Table 1. Circadian variation of thermoregulatory responses in exercising men at two different work intensities

30% VO2max 60% VO2max

0900 h 2000 h 0900 h 2000 h

TSR (g m-2 h-') 83.0(8.0) 117.3(2.8)* 194.4(8.7)t 216.2(11.0)tSRma, (g m-2 h-1) 138.4(10.0) 167.4(6.1)* 283.4(17.1)t 287.4(5.2)tTR SRmax (min) 37(1.2) 33(3.0) 38(1.2) 29(2.4)**AN (0C) 0.27(0.07)* 0.11(0.04) 0.70(0.10)t 0.61(0.02)tATk (0C) 1.20(0.11) 1.42(0.36) 1.21(0.16) 0.85(0.02)ATb (QC) 0.45(0.05) 0.37(0.09) 0.79(0.07)t 0.61 (0.07)t

Data represent mean(s.e.m.) for five subjects; TSR, total sweat rate for 40 min of exercise; SRma,, maximum rate of sweating duringexercise; TR SRmax, time of reaching SRmax after the onset of exercise; ATre, ATAk and ATb, changes from pre-exercise value in Tre, TSk andTb, respectively; Significant difference (paired t test), 0900 h vs 2000 h, *P< 0.05, **P<< 0.01; 30% Vo2max vs 60% VO2max, tp< 0.01

Tre ('

37.8 -

37.4 -

37.0

36.6 -

DC)

30ma02max OMmean ±s.e.m.I * 200

***~~~~~~~~~~~~~~~~~

Tre (°C)

37.8-

37.4 -

37.0-

36.6--

Tsk

Tb (°C) Tb (°C)37.4-J"

LRestO 5 10 15 20 25 30 35 40 45 50 RestO 5 10 15 20 25 30 35 40 45 50

Time (min) Time (min)Figure 6. Time courses of Tre, Tb and T~k in exercising men at two different workloads, about 30% [left] and 60% [right] ofVO2max in the later morning [01 and evening [@]. Horizontal bars indicate an exercise period. Data represent mean(s.e.m.)for five subjects. Significantly different (paired t test), 0900 h vs. 2000 h, *P<0.05, **P<0.01, ***P<0.001

Br J Sp Med 1995; 29(2) 117

30%V02max***~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

137.0 -

36.6 -

36.2 -

35.8 i

T--



.bmj.com

/B

r J Sports M


ownloaded from



40 min. There was no significant difference betweenTre at 0900 h and at 2000 h, except for 40 min after theonset of exercise at 30% VO2max exercise. There was asignificant difference in the rate of increase in Trebetween 0900 h and 2000 h during exercise at30%VO2mac. In both periods, Tb levels peaked duringexercise at 60% V02max None of these changes weresignificant, suggesting no significant changes in Tbduring the exercise in both periods. However, at 30%VO2ma,, exercise Tb was significantly higher at 0900 hthan at 2000 h. No significant difference in Tsk wasfound between 2000 h and 0900 h in both workloads(Figure 6, bottom). After a transient fall, Tk slightlyincreased during exercise at the two differentworkloads of the two test periods. For sitting on abicycle ergometer (at rest) Tre and Tb also weresignificantly higher at 2000h and 0900h. There wasno significant difference in Trk between 0900h and2000 h in the resting condition on a bicycle ergometer(Table 1D.As Tb was plotted against SR during exercise,

positive correlations were observed (Figure 7). In all

300

N

E

I-

3I-LuU)

250 -

200

150'

100

50

36.0 36.2 36.4 36.6 36.8 37.0 37.2 37.4

300 1N

E

I-r

03n

250 -

200

150

100 -

50

036.0

. .

36.2 36.4 36.6 36.8 37.0 37.2 37.4

MEAN BODY TEMPERATURE (°C)

Figure 7. Relationship between SR and Tb in exercisingmen at two different workloads 30% [A] and 60% [B] ofVo2max in the later morning [0] and evening [0]. SR and Tbwere mean values during exercise for five subjects

subject data concerning SR vs TreTsk and Tb relations(or the relation of SR to Tre, Tsk or Tb), the bestcorrelation coefficient (r = 0.8201) was observed inthe SR vs Tb relation. The regression equations andcorrelation coefficients are presented in Figure 7.Analysis of covariance revealed significant differ-ences of the slope and intercept between theregression lines (SR vs Tb relationship) at 0900 h and2000 h to exercise at 30% Vo2rmax (P < 0.05, respective-ly). At 60% VO2max exercise the intercept shifted to ahigher temperature (P < 0.05).

DiscussionThe circadian rhythm of different physiological andbiochemical variations has been well established. Anumber of investigators studied the daily courses ofSR5' 6 8, peripheral blood flow5'8' 9, VO2ma, 20 and HR10in muscular working humans. A nocturnal loweringhuman core temperature has been correlated with alowering to the core temperature threshold forcutaneous vasodilation and sweating during exercise5and rest to a heat exposure21. Timbal et al.7investigated at 0200 h, 1000 h and 1800 h during a heatexposure of 90 min, sweat rate, body temperatureand heat storage of the body of human subjects.According to their calculating and measuring data,the relationship between the change of heat storageof the body (AS) in the morning and that during thenight is expressed in the following equation: ASnight= 0.69Smorning -6.24. In the present study, SR duringexercise at the lower work intensity was significantlyhigher in the evening than in the morning. In thiscase, we were in agreement with the previousreports6'8. However, the circadian variation of SRduring exercise was not observed at the heavy workintensity. In the previous studies, there was nocomparison made of the effects of work intensity onthe circadian control to temnfperature regulation. In thepresent study Tre and Tb at rest (values beforeexercise) were significantly higher at 0900 h than2000h. From our data, it suggests that heat dissipat-ing activity in the central control system is activatedmore in the evening than in the later morning.At the heavy workload, the HR of exercising men

in the evening was significantly higher than that inthe later morning. Vo2, however, increased inproportion to workloads, not to time of day (two-wayANOVA). No measurements were made of otherphysiological parameters in cardiorespiratory func-tion. The present study suggests that, for cardiospira-tory function and skin circulation during exercise, theregulatory system circulation mag be modified by thework factors. Cohen and Muehlt reported that therewas circadian variation of HR during submaximalexercise. In resting HR, they also noted 50 beatsmin-1 at 0400h and 70 beats min-' at 1800h duringthe time of day. We, however, did not observe HR ina resting condition, and did not compare directlywith their subject laying (supine) on a bed, becauseour subjects were seated on a bicycle ergometer.Further studies are required to measure otherphysiological parameters such as cardiac output,stroke volume and skin blood flow.A typical rightward shift in Tb threshold for SR was

118 Br J Sp Med 1995; 29(2)

0900, 600%/V02max * <.*e (B)- 8924.1 + 247.99x=0.992) - .. 0

0.'

*2000, 60%.0 maxy =- 9049.2 + 250.59x(r= 0.908)

0

Y=(r



.bmj.com

/B

r J Sports M


ownloaded from


Circadian variation of thermoregulation during exercise: M. Toni et al.

shown in the data (Figure 7). SR was graphed as afunction of Tb during two different exercises in themorning and evening. The data showed that therewas a parallel shift (rightward) in the SR to Tbrelationship during exercise in the later morning andevening. This rightward shift indicated that there wasan increased Tb threshold for the onset of sweating inthe evening. The increased threshold temperaturesfor sweating and vasodilation, observed by Stephen-son et al.8, were affected by the normal circadianrhythm in resting_internal temperature. In ourexperiment, resting Tb in the evening was significant-ly higher (P <0.001) in comparison with that in thelater morning. This result supported the concept thatthe regulation of sweating is related to a critical levelof Tb or heat storage of the body22. In anotherstudy', the regulation of sweating was analysed interms of the interaction between central andperipheral inputs to the thermoregulatory system.Inputs from the skin thermosensors stimulated by therate of change in skin temperature, have been shownto play a major role in the regulation of sweating inexercising subjects24.Although core temperature is the primary thermo-

regulatory drive in the control of skin sweating, skintemperature can also affect SR24. In the presentstudy, however, there were no significant differencesin skin temperature in the exercising men at twodifferent work intensities in the later morning and theevening. These data suggest that skin temperatureshifted the minimum levels to maximum levels in theearly morning, or the maximal levels to the minimumlevels in the afternoon, if we accept Hildebrandt'shypothesis12 on circadian control in human thermo-regulation; that is, circadian regulatory mechanismschange to warming-up after 0300 h and the regulatoryfunction change to cooling-down after 1500 h. Ex-amining the effect of a step change in environmentaltemperature upon the skin temperature and bloodflow of a seated human, Sasaki and Carlson25observed that the circadian change of core tempera-ture was paralleled by variations in both skintemperatures and peripheral blood flows. They havereported cycles in the heat dissipating responseswhich are not readily explained by simultaneouschanges in core temperature. On the other hand, therise in core temperature of an exercising man wasproportional to work intensities and largely indepen-dent of ambient temperatures between 5-30°C. Thisrise in core temperature was not due to a failure inbody temperature regulation, but was attributed tothe setting of the body thermostat at a higherlevel26' 27. At the same work intensity, our data differin the rate of increase of core temperature presentedby Tre due to exercise between the later morning andevening (Table 1). Moreover, evaporative coolingindexes as TSR, SRmax and the time to reach SRmxdid differ markedly by means of work intensity(Figure 5 and Table 1); that is, TSR and SRmax duringexercise at the heavy workload did not observe thecircadian variation.

Furthermore, our data indicate that the two majorindependent experimental variables in evaluating theeffect of exercise on temperature regulation are thecondition of the internal environment such as the

'biological clock'28 and the workload. Figures 5 and 6show how these two variables cause changes inregulatory sweating, core temperature and skintemperature. In the cooling-down period, heatdissipation activity was stimulated compared to thewarming-up period at the lower workload. However,effector responses dissipating heat at the heavyworkload are not distinguished in both phases.

It has been reported that in an exercising humanthe thresholds for sweating moved toward a lowercore temperature early in the morning (0400h)5. Onthe other hand, in the present study a tendency wasobserved in humans for the changes in coretemperature during thermal transient caused byexercise to be larger at the later morning phase that atthe evening phase. The rise in core temperature in aman during heat and cold exposure was reported tobe higher at early morning (0200 h) than latermorning (1000 h) and evening (0600 h)7. These find-ings may reflect the fact that the thermosensitivity ofan exercising man is reduced during the night. Asillustrated in Figure 7, the slopes of the temperaturecharacteristic lines might also be reduced as indicatedby temperature sensitivity during the light phase inaddition to the parallel shifts mentioned above. Thiscould cause a further downward shift of set-pointtemperature from day to evening.

Recently, the significance of exercise prescriptionfor a healthy lifestyle has been recognized29 but thereis little practical knowledge about the physiological-thermoregulating functions in humans, especially asaffected by internal conditions28. Many works30-32have reported that the risk factors for coronary heartdisease decrease with an increase in aerobic exerciseand sports (or muscular activity). Thus, an exercisingman is affected by the time of day, and prescribers ofexercise programmes must pay attention to factorssuch as exercise intensity. These findings areapplicable to practical-exercise activity; that is, ourdata may be important in planning the practice of andin evaluating the effectiveness of a physical exerciseprogramme.

In conclusion, the results of this study indicate thatthe circadian control of thermoregulatory response toexercise may be modulated by the workloads in thelater morning and evening. It is suggested that therise in human body temperature normally observedduring exercise at the light workload in the latermorning is due partly to the proportional nature ofthe circadian control mechanism and partly to thelower level in sweating which locally inhibits the heatdissipating responses due to evaporating sweat.These findings seem to have important implicationsfor further research; for example, disappearance orlowering of the circadian variation in sweatingresponse to exercise at the higher work intensity. Thecircadian variation of thermal regulation duringexercise may be modulated by this factor.

AcknowledgementsThe authors would like to express their gratitude to theircolleagues Dr M. Yamasaki (Department of Health and PhysicalEducation, Integrated Arts and Science, Hiroshima University) andMSc T. Miyabayashi (Laboratory of Work Physiology, Kumamoto

Br J Sp Med 1995; 29(2) 119



.bmj.com

/B

r J Sports M


ownloaded from



Junior College) and their contributions to the experimental workpresented here. We also thank Mr J. Russell (MA), AssistantProfessor in the Kyushu Institute of Technology, for checking theEnglish. This work was partly presented in the 1st InternationalConference of Human-Environment System, Tokyo, 1991.

References1 Aschoff J, Pohl H. Rhythmic variations in energy metabolism.

Federation Proc 1970; 29: 1541-52.2 Sasaki T. Body temperature and its circadianity. Jiritsushinkei

1984; 21: 177-83. (in Japanese with English abstract).3 Sasaki T. Circadian rhythm in body temperature. In: Itoh S.

Ogata K, Yoshimura H, eds. Advances in Climatic Physiology.Tokyo, Japan: Igaku Shoin, 1972; 319-33.

4 Smith RE. Circadian variations in human thermoregulatoryresponses. J Appl Physiol 1969; 26: 554-60.

5 Wenger CB, Roberts MF, Stolwijk JAJ, Nadel ER Nocturnallowering of thresholds for sweating and vasodilation. J ApplPhysiol 1976; 41: 15-19.

6 Niwa K, Nakayama T, Ohnuki Y, Midorikawa T. Circadianvariation of thermoregulatory responses during exercise inmale. Jpn J Phys Fitness Sports Med 1982; 3: 258-65 (in Japanesewith English abstract).

7 Timbal J, Colin J, Boutelier C. Circadian variations in thesweating mechanism. J Appl Physiol 1975: 39: 226-30.

8 Stephenson LA, Wenger CB, O'Donovan BH, Nadel ERCircadian rhythm in sweating and cutaneous blood flow. AmJ Physiol 1984: 246: R321-4.

9 Kaneko M, Zechman FW, Smith RE. Circadian variation inhuman peripheral blood flow levels and exercise responses. JAppl Physiol 1968; 25: 109-14.

10 Cohen CJ, Muehl GE. Human circadian rhythms in restingand exercise pulse rate. Ergonomics 1977; 20: 475-9.

11 Aschoff J, Heise A. Thermal conductance in man: itsdependence on time of day and on ambient temperature. In:Itoh S, Ogata K, Yoshimura H eds. Advances in ClimaticPhysiology. Tokyo, Japan: Igaku Shoin, 1972; 334-48.

12 Hildebrandt G. Circadian variations of thermoregulatoryresponse in man. In: Scheving LE et al. eds. Chronobiology.Tokyo, Japan: Igaku Shoin, 1974; 234-40.

13 Takahira H. A study of metabolic rate in the Japanese; surfacearea in the Japanese and its estimation equation. EiyoKenkyusho Hokoku 1925; 1: 61-95 (in Japanese).

14 Miyamura M, Honda Y. Oxygen intake and cardiac outputduring maximal treadmill and bicycle exercise. J Appl Physiol1972; 32: 185-8.

15 Torii M, Yamasaki M, Sasaki T. Effects of heat stress ontemperature regulation in initial exercise. J Human Ergol 1986;15: 3-12.

16 Torii M, Yamasaki M, Sasaki T, Nakayama T. Fall in skintemperature of exercising man. Br J Sports Med 1992; 26:29-32.

17 Mitchell JW, Nadel ER, Stolwijk JAJ. Respiratory weightlosses during exercise. J Appl Physiol 1972; 32: 474-6.

18 Ramanathan NL. A new weighting system for mean surfacetemperature on the human body. J Appl Physiol 1967; 23:347-52.

19 Stolwijk JAJ, Hardy JD. Partitional calorimetric studies ofresponses of man to thermal transients. J Appl Physiol 1966; 21:967-77.

20 Wojtczak-Jaroszowa J, Banaszkiewicz A. Physical workcapacity during the day and night. Ergonomics 1974; 17: 193-8.

21 Hessenmer V, Bruck K. Influence of menstrual cycle onthermoregulatory, metabolic, and heart rate responses toexercise at night. J Appl Physiol 1985; 59: 1911-7.

22 Chappuis P. Pittet P. Jequier E. Heat storage regulation inexercise during thermal transient. J Appl Physiol 1976; 40:384-92.

23 Nadel ER, Bullard RW, Stolwijk JAJ. Importance of skintemperature in the regulation of sweating. J Appl Physiol 1971;31: 80-7.

24 Nadel ER, Mitchell JW, Saltin B, Stolwijk JAJ. Peripheralmodifications to the central drive of sweating. J Appl Physiol1971; 31: 828-33

25 Sasaki T, Carlson LD. Effect of a step change in temperatureon skin temperature and blood flow. Proc Soc Exper Biol Med1964; 117: 334-8.

26 Nielsen M. Die Regulations der Korpertemperatur beiMuskelarbeit. Scand Arch Physiol 1938; 79: 193-230.

27 Nielsen M. Heat production and body temperature duringrest and exercise. In: Hardy JD, Gagge AP, Stolwijk JAJ, eds.Physiological and Behavioral Temperature Regulation. Springfield,USA: CC Thomas, 1970; 205-14.

28 Chiba Y. Circadianity and its regulatory mechanisms. InSasaki T, Chiba Y eds. Chronobiology. Tokyo, Japan: AsakuraShoten, 1978; 11-40 (in Japanese).

29 Arikawa I. Active 80 health plan and sports medicine. 2ndToyko International Symposium on Sports and Exercise Medicine[TISSEM '89]. 1989; Abstract 32-3.

30 Wilson PK, Fardy PS, Froelicher VF. Cardiac Rehabilitation,Adult Fitness and Exercise Testing. Philadelphia, USA: Lea &Febiger; 1981.

31 Haskell WL. The influence of exercise on the concentrationsof triglyceride and cholesterol in human plasma. In: TerjungRL, ed. Exercise and Sport Sciences Reviews Volume 12, NewYork, USA: Macmillan Publishing, 1984; 205-44.

32 Ehsani LA. Cardiovascular adaptations to exercise training inthe elderly. Federation Proc 1987; 46: 1840-3.

120 Br J Sp Med 1995; 29(2)



.bmj.com

/B

r J Sports M


ownloaded from


thermoregulation exercisingmen in the morning rise internal ...performed bicycle exercises at 30%...

Documents