Abstract The study investigated how the pressure exert-ed on the skin by clothing worn while working in thedaytime affected the urinary excretion of adrenaline, nor-adrenaline and cortisol, heart rate, and also melatonin se-cretion at night. Nine young women (experiment I) andseven young women (experiment II) participated. Partic-ipants wore either a 100% cotton jacket (tight clothes,TC) or a 100% cotton T-shirt (loose clothes, LC). Loose-fitting, 100% cotton tank tops and panties were worn asunderwear in both the TC and the LC groups. The mainresults can be summarized as follows: (1) urinary excre-tion of adrenaline, noradrenaline and cortisol was facili-tated, and the amounts of urinary excretion were signifi-cantly higher when TC were worn. Heart rate was signif-icantly higher in the TC group; (2) nocturnal urinary me-latonin excretion was significantly greater in the TCgroup. These results are discussed in terms of an en-hancement of diurnal sympathetic nervous system activi-ty caused by pressure on the skin produced by tightclothing.

Keywords Intermittent skin pressure · Tight clothing ·Hormonal responses · Melatonin · Heart rate

Introduction

Tokura and his group have found a marked influence ofskin pressure (exerted by clothing) on the rate of saliva

secretion (Tokura 1989; Okura et al. 2000), core temper-ature (Lee et al. 2000, 2001), digestive function (Takasuet al. 2000) and melatonin and noradrenaline secretion(Lee et al. 2000). Participants were exhausted morequickly in performing a hand-grip exercise and their re-action time to acoustic stimuli was significantly delayedwhen their trunk had been compressed by a cuff (Zhangand Tokura 1995). These studies were performed usingsedentary participants, but people normally move andwalk during working hours in their everyday life. There-fore it is more relevant to study how physiological pa-rameters respond to pressure on the skin due to the cloth-ing worn during working hours.

Park and Tokura (1998) found that the way in whichthe surface of the body was covered during waking hoursinfluenced the nocturnal minimum core temperature,even if the participants slept at night wearing identicalclothes. Also, the nocturnal rise of salivary melatonin wassignificantly suppressed when the participant slept atnight under the influence of pressure upon the skin pro-duced by foundation garments (brassiere, girdle) (Lee etal. 2000; Tokura et al. 2001). It was therefore of interestto investigate whether pressure upon the skin, producedby clothing during the daytime while the participantswere working, could influence physiological parametersduring the following night. We have focused our attentionon urinary melatonin excretion during sleep.

Materials and methods

Participants

Nine healthy women, 20 years of age, participated in the study.Their stature, weight and bust measurements are summarized inTable 1. Seven out of nine participants (p1–p7) also took part inexperiment II.

Clothes

Participants wore either a 100% cotton jacket (blazer), referred toas tight clothes (TC), or a 100% cotton T-shirt, referred as loose

Y. Mori (✉)Department of Human Environment, Konan Women’s University,Kobe 658-0001, Japane-mail: moriyuki@konan-wu.ac.jpTel.: +81-78-4133004, Fax: +81-78-4133004

E. KiokaDepartment of Apparel and Space Design, Kyoto Women’s University, Kyoto 605-8501, Japan

H. TokuraDepartment of Working Physiology & Ergonomics, Nofer Institute of Occupational Medicine, 8 Teresy Str.,P.O. Box 199, 90-950 Lodz, Poland

Int J Biometeorol (2002) 47:1–5DOI 10.1007/s00484-002-0145-z

O R I G I N A L A RT I C L E

Yuki Mori · Etsuko Kioka · Hiromi Tokura

Effects of pressure on the skin exerted by clothing on responses of urinary catecholamines and cortisol, heart rate and nocturnal urinary melatonin in humans

Received: 8 March 2002 / Revised: 10 July 2002 / Accepted: 11 July 2002 / Published online: 12 September 2002© ISB 2002

clothes (LC) (Fig. 1). The thickness and weight were 0.85 mm and524 g for the TC, and 0.97 mm and 426 g for the LC. The clotheswere selected to fit the body size of the participants. The pressureexerted upon the skin by the two types of clothes during workingposture is shown in Table 2. Using equipment for the measure-ments of skin pressure by clothing (AMI 3037, AMI Co., Tokyo),this was determined by the application of small bag of air to theventral area of the forearm, the dorsal area of the upper arm andM. latissimus dorsi. The skin pressure applied by the TC was inthe range 0.54–2.50 kpa in the regions of the forearm, upper armand M. latissimus dorsi., while it was in the range of0.069–1.00 kpa at identical regions when LC was worn. The par-ticipants wore a loose-fitting, 100% cotton tank top and panties asunderwear in both the TC and the LC conditions.

Protocol

Experiment I

The participants dressed in the experimental clothes (TC or LC) at10:30 hours after having voided their bladder completely. After30 min rest in the sitting position, they worked at a computer for

the next 90 min. They then produced a urine sample, the volumeof which was measured by a cylinder. Approximately 5 ml urinewas placed in a sample tube and frozen at –20 °C for later analysisfor adrenaline, noradrenaline and cortisol. Adrenaline and norad-renaline were analyzed by HPLC. Cortisol was analyzed by radio-immunoassay. The total excretion of these substances during theworking time (2 h) was calculated [µg/l × urine volume (ml/2 h)]and these levels were expressed as µg/2 h.The participants wererequested to type several papers during their work on the comput-er. After a 1-h lunch break, they worked on the computer for90 min from 1330 hours to 1500 hours. The food and water intakewas strictly controlled to be equal for the two kinds of clothing.Food influencing the hormonal levels, such as orange, coffee andbeer, was not provided. A further urine sample was then collectedand treated as before. The heart rate was measured evenly 5 minby a T.K.K. data logger (model 1866W, Takei Kiki Kogyo, Osaka)from 1030 hours to 1530 hours.

Two days later the same procedure was repeated, but with eachparticipant wearing the other type of clothes. The order of wearingthe TC or the LC was randomized.

These experiments were conducted from 4 February to 14March 2001. The temperature of the experimental room was keptat 25 ± 2 °C.

Experiment II

Seven (p1–p7) of the same nine participants also took part in ex-periment II, where the effects of the TC or the LC over the courseof 7–8 h (from 0700 or 0800 hours to 1600 hours) on the nocturnalexcretion of melatonin were studied. The participants rose at 0700hours or 0800 hours (they were at their homes) and wore the TCor the LC. They spent their time normally, usually attending lec-tures at the University. If they wore the TC in the morning, theychanged into the other type of clothing at 1600 hours and returnedhome. When they wore the LC clothing in the morning, they didnot change into the other type of clothing, continuing to wear theLC. If they moved their upper extremities when they were wearingthe TC, then a considerable pressure was exerted on their skin bythe clothing. The participants retired at 2300 hours. One day later,the participants underwent the same routine, but wearing the othertype of clothing. The order of wearing the TC or the LC was ran-domized.

Participants produced a urine sample at 2300 hours and retiredat their usual bedtime. As soon as they rose (at 0700 hours or 0800hours in the following morning) they produced a urine sample thatwas treated as in the previous experiment. Urinary melatonin wasanalyzed by radioimmunoassay method. The total amount excret-ed during sleeping (8 h) was calculated [pg/ml × urine volume(ml)] and its level was expressed as ng/8 h.

Statistics

Mean values of urinary adrenaline, urinary noradrenaline, urinarycortisol and urinary melatonin were compared between the TC andthe LC groups using Student’s paired t-test. The average heart ratewas compared between the TC and the LC groups using a two-way analysis of variance with repeated measure.

Results

Experiment I

Rates of urinary excretion of adrenaline, noradrenalineand cortisol were compared in the morning and after-noon for the two types of clothing (Fig. 2). Urinary ad-renaline and cortisol were significantly higher for thosewearing the TC in the morning and afternoon; urinary

2

Table 1 Physical characteristics of the participants

Participants Height (cm) Weight (kg) Bust (cm)

p1 160.6 52.0 84.2p2 150.9 39.0 76.8p3 151.4 57.0 91.0p4 155.4 48.6 79.7p5 167.3 61.6 85.3p6 166.4 53.8 81.2p7 157.0 50.7 86.2p8 165.7 55.0 94.0p9 160.0 47.0 81.4Mean 159.4 51.6 84.4SD 6.2 6.5 5.5

Fig. 1 Experimental clothing. TC tight clothes, LC loose clothes

Table 2 The skin pressure applied by the clothes while the wearerworked on the computer. TC tight clothes, LC loose clothes

Position Clothing type Pressure on skin (kPa)

Forearm TC 0.54–0.78LC 0.069–0.25

Upper arm TC 1.02–2.09LC 0.38–0.55

M. latissimus dorsi TC 1.07–2.50LC 0.25–1.00

3

noradrenaline was significantly higher in the TC groupin the morning only.

Heart rate was compared between the TC and the LCgroups (Fig. 3). It was significantly higher in those wear-ing TC than in the LC group throughout the experimen-tal period. The first heart rate measurement before thestart of computer working was higher in the TC than inthe LC group. Each participant had been wearing the ex-perimental clothing for about 30 min before the start ofthe experiment, therefore the initial higher heart rate val-ue may reflect the influence of the TC.

Experiment II

Figure 4 shows the amount of melatonin excreted in theurine in 8 h during the night, following the wearing ofthe two types of clothing during the daytime. The noc-turnal excretion of melatonin was significantly higherwhen the TC had been worn. This means that the pres-sure exerted by clothing during the daytime influencedthe melatonin level during sleep.

Fig. 2 A comparison of the urinary excretion of adrenaline, nor-adrenaline and cortisol between TC and LC. Left columns, morn-ing values, right columns afternoon values. Black columns TC,white columns LC. The values are means with SD. **P < 0.01, *P < 0.05 (n = 9)

Fig. 3 A comparison of aver-age heart rates between TCand LC throughout the experi-mental period. • TC, ●● LC.**P < 0.01 (n = 9)

Fig. 4 A comparison of the urinary excretion of melatonin duringsleep, following the wearing of TC or LC during the daytime.Black column TC, white column LC. *P < 0.05 (n = 7)

Discussion

The most interesting findings were that when the cloth-ing exerted pressure on the skin while the participantsworked during the daytime i.e. in the TC group (1) theamounts of urinary adrenaline, noradrenaline and corti-sol excreted were significantly higher, as was the heartrate, and (2) the amount of urinary melatonin excreted inthe following night was significantly higher when theparticipants had worn the TC during the daytime.

Lee et al. (2000) found that the urinary excretion ofnoradrenaline was lower when subjects wore foundationgarments while sitting and resting than when these gar-ments were not worn. Sugimoto (1991) found that thepressure exerted by a girdle and body suit increased theurinary excretion of noradrenaline. Our present resultsare consistent with those of Sugimoto (1991), but differfrom those of Lee et al. (2000). The reason for such adiscrepancy might be that the effect depends uponwhether the pressure exerted on the skin is constant (Leeet al. 2000) or intermittent (Sugimoto 1991; presentcase). In our current experiments, the pressure was inter-mittent because of body movements. The amount ofpressure exerted was nearly the same in the present caseas in the conditions used by Lee et al. (2000).

The finding that the rates of urinary excretion of ad-renaline and noradrenaline were significantly higherwhen TC was worn suggests that the sympathetic nervoussystem was more excited in this case, and that this result-ed in the higher heart rate. That is, a constant pressure ex-erted upon the skin seems to inhibit the sympathetic ner-vous system, whereas an intermittent pressure (as in thepresent experiment) seems to stimulate it. The physiolog-ical mechanisms involved remain to be studied.

According to Nagayama et al. (1995), Watanuki andMurata (1994) and Watanuki (1994), the heart rate wasreduced under the when the skin was subjected to pres-sure by a girdle and pantyhose in resting women. Theirresults were inconsistent with our results. Similarly, itseems that working on a computer with the accompany-ing intermittent skin pressure generated by the TC couldincrease the heart rate, while resting conditions accom-panied by the nearly constant skin pressure of a girdleand pantyhose stock could reduce heart rate, suggestingthat intermittent and constant skin pressures exerted byclothing have different effects on heart rate.

The appropriate pressure to be exerted on the skin byclothing, in order to produce the most appropriateamount of sympathetic activation, remains to be studied;extended periods of excessive stimulation might lead tofatigue. The increase of urinary excretion of melatonin atnight following the wearing of TC during the daytimesuggests a positive effect of such clothing in terms ofhealth maintenance. This is because it is well known thatmelatonin is involved in the induction of sleep at night(Dawson and Encel 1993; Brzezinski, 1997) and in anenhancement of immune activity (Park and Tokura,1999; Maestroni and Conti 1993; Pierpaoli and Maestroni1987).

Why did the rate of excretion of urinary melatonin in-crease following the wearing of TC during the daytime?It has been established that the nocturnal elevation ofmelatonin can be suppressed by exposure to bright lightat night (Badia et al. 1990), and that this is due to thesuppression of noradrenaline secretion from postgangli-onic fibers innervating the pineal gland (Moore et al.1979). However, the elevation of salivary melatonin ex-cretion at night has been found to be suppressed if cloth-ing exerts pressure on the skin during sleep (Lee et al.2000). Our interpretation of the present finding – thatmelatonin increased after intermittent pressure was ex-erted on the skin by clothing worn during the daytime –is that it might be due to the activation of the noradrener-gic postganglionic fibers as an “after effect” of diurnalpressure upon the skin.

Thus, it is concluded that pressure exerted upon theskin in the range 0.54–2.50 kpa by clothing worn duringthe daytime increases the urinary excretion of adrenaline,noradrenaline and cortisol and the heart rate, and thatthere is also an increase in the urinary excretion of mela-tonin in the following night.

Acknowledgements The authors thank Dr. J. Waterhouse in Liv-erpool, U.K, for his kind editorial assistance.

References

Badia P, Culpepper J, Boecker M, et al (1990) The immediate effect of bright and dim light on body temperature. Sleep Res19:386

Brzezinski A (1997) Melatonin in humans. N Engl J Med 386:186–195

Dawson D, Encel N (1993) Melatonin and sleep in humans. J PinealRes 15:1–12

Lee YA, Hyun KJ, Tokura H (2000) The effects of skin pressureby clothing on circadian rhythms of core temperature and sali-vary melatonin. Chronobiol Int 17:783–793

Lee YA, Hyun KJ, Tokura H (2001) Circadian rhythms of corebody temperature and noradrenaline secretion under the influ-ence of skin pressure due to foundation garments worn duringwakefulness. Biol Rhythm Res 32:389–400

Maestroni GJM, Conti A (1993) Melatonin and the immunesystem. In: Touitou Y, Arendt J, Pevet P (eds) Melatonin andthe pineal gland – from basic science to clinical application.Elsevier Science, Amsterdam, pp 295–302

Moore DC, Paunier L, Sizonenko PC (1979) Effect of adrenergicstimulation and blockade on melatonin secretion in the human.Prog Brain Res 52:517–521

Nagayama Y, Nakamura T, Hayashida Y, Ohmura M, Inoue N(1995) Cardiovascular responses in wearing girdle – powerspectral analysis of heart rate variability (in Japanese with English summary). Jpn Res AssnText End-Uses 36:68–73

Okura K, Midorikawa T, Tokura H (2000) Effects of skin pressureapplied by cuffs on resting salivary secretion. J Physiol Anthropol 19:107–111

Park SJ, Tokura H (1998) Effects of different types of clothing oncircadian rhythms of core temperature and urinary catechola-mines. Jpn J Physiol 48:149–156

Park SJ, Tokura H (1999) Bright light exposure during the daytimeaffects circadian rhythms of urinary melatonin and salivaryimmunoglobulin A. Chronobiol Int 16:359–372

Pierpaoli W, Maestyoni GJM (1987) Melatonin: a principal neuro-immunoregulatory and anti-stress hormone: its anti-aging effects. Immunol Lett 16:355-362

4

of clothing and light. In: Honma K, Honma S (eds) Zeitgebers,entrainment and masking of the circadian system. HokkaidoUniversity Press, Sapporo, pp 147–154

Watanuki S (1994) Improvements on a design of girdle by usingcardiac output and pressure sensation. Ann Physiol Anthrop13:157–165

Watanuki S, Murata H (1994) Effects of wearing compressionstockings on cardiovascular responses. Ann Physiol Anthrop13:121–127

Zhang P, Tokura H (1995) Effects of skin pressure by cuffs on en-durance performance and voluntary reaction time. Jpn J ClothRes 38: 7–11

5

Sugimoto H (1991) Compression of body by clothing: increase inurinary norepinephrine excretion caused by foundation garments(in Japanese with English summary). Jpn J Hyg 46:709–714

Takasu N, Furuoka S, Inatsuji N, Rutkowska D, Tokura H (2000)The effect of skin pressure by clothing on whole gut transittime and amount of feces. J Physiol Anthropol 19:151–156

Tokura H (1989) Physiological significance of clothing and humanhealth. In: Tokura H (ed) Proceedings of international sympo-sium of clothing comfort studies in Mt. Fuji. The Japan ResearchAssociation for Textile End-Uses Publishing, Osaka, pp 203–222

Tokura H, Park SJ, Lee YA, Hyun KJ, Aizawa S (2001) Circadianrhythms of core temperature and malatonin under the influence