Arch Dermatol Res (1992) 284:135-140
9 Springer-Verlag 1992
Thermal and sweating responses in normal and atopic subjects
under internal and moderate external heat stress
B. Bothorel 1 A. Heller 2, E. Grosshans 2, and V. Candas 1
1 CNRS-INRS, Laboratoire de Physiologie et de Psychologie Environnementales, 21, rue Becquerel, F-67087 Strasbourg Cedex, France
z Clinique Dermatologique, Facult~ de M~decine, 1, place de l'H6pitat, F-67091 Strasbourg Cedex, France
Received September 12, 1991
Summary. To compare the thermoregulatory responses of
normal subjects and patients with atopic dermatitis, the
effect of exercise under moderate heat stress was examined
in two groups of subjects. Each group of eight subjects
(controls or clinical atopics) underwent a 90-min exper-
iment after being equipped with probes for measurement
of core and skin temperatures, heart rate and overall and
local sweating rates. Sweat surface tension was determined
from sweat collection made at the end of the session. The
experimental procedure was as follows: 30 min rest at
thermoneutrality, 30 min cycling at 90 W at 36 ~ follo-
wed by 30 min recovery sitting at rest at 36 ~ on the
cycloergometer. None of the registered variables differed
significantly between the normal and atopic subjects except
for variations in mean skin temperature, core-to-skin
temperature difference and sweat surface tension. Since
local sweating on non-affected skin under a controlled
thermal clamp was not altered by atopy, it can be concluded
that the thermoregulatory modifications under heat stress
is of vasomotor origin, the benefit of which, in terms of
heat dissipation capacity, remains uncertain. There may
be alterations in the constituents of sweat, but not in its
excretion rate. Local, rather than central, factors are
probably involved in this qualitative change, which remains
to be investigated.
Key words: Core temperature - Skin temperature -
Local sweating - Sweat tensioactivity
Atopic dermatitis encompasses skin diseases associated
with atopy, which are characterized by allergenic re-
sponses. Diagnosis of atopic dermatitis is based on three
out of the following major features: pruritus, local skin
lichenification, chronically relapsing dermatitis, personal
or family history of atopy [8].
Atopic subjects often claim to have difficulty in
tolerating heat, the difficulty being greater in the case of
humid heat exposure. It is generally thought that atopic
Correspondence to: B. Bothorel
patients suffer from reduced heat dissipation mechanisms,
especially because of possible decreased sweating capaci-
ty. However, atopy has previously been associated with
reduced [4, 5, 19], normal [11, 13] or enhanced sweating
[18, 22] or with variable sweat responses depending upon
the season [21]. To our knowledge, most of the studies
quantified sweat output following injection of various
pharmacological substances. Sweat gland sensitivity may
be increased under acetylcholine stimulation due to
fl-adrenergic receptor blockade. Rovensky and Saxl [18]
compared the sweat output of both normal and lichenified
skin of atopic subjects and found that sweat gland
sensitivity was increased only on the lesional skin. If the
so-called 'normal' skin is unaffected, the mechanism
involved has to be local in origin and should not have
an overall effect on the body heat content regulation.
Kiistala [11], using methacholine or epinephrine injec-
tions, concluded that the unaffected skin of atopic subjects
and that of controls were not statistically different.
Nevertheless, even in the absence of any modification
in sweating capacity, alteration of heat dissipation may
be affected by sweat composition. One preliminary study
[7], carried out on the qualitative properties of sweat,
concluded that sweat surface tension was greater in
patients with dermatological problems, probably because
of a decreased concentration of mucopolysaccharides. It
has been hypothesized that increased surface tension
could reduce the capacity for wetting the skin. Skin
wetness is an important factor for the mechanism of
cutaneous heat dissipation under humid conditions [2].
If wetness adjustments are inadequate, the evaporation
efficiency of sweat may be altered in atopic subjects, who
can then suffer from excessive heat retention in warm
humid climates.
The aim of the present study was to compare the
thermophysiological responses of atopic and normal
subjects exposed to heat strain, by combining heat
exposure with metabolic heat production induced by
physical exercise. It is well known that, due to a local
temperature effect [17], any local temperature increase
will result in an increase in sweat gland output. Thus, to
avoid erroneous conclusions, the evaluation of local
sweating in reaction to a rise in the central command,
136

associated with a rise in b o d y heat c o n t e n t , requires
t h e r m a l c l a m p i n g of the local area at a c o n s t a n t level.
Therefore, b o d y t e m p e r a t u r e s , whole b o d y sweat p r o d u c -
tion a n d local sweating, u n d e r c o n s t a n t local skin tem-
perature, were m e a s u r e d to assess possible sweat o u t p u t
alterations a n d / o r m o d i f i c a t i o n of t e m p e r a t u r e r e g u l a t i o n
m e c h a n i s m s . The sweat surface t e n s i o n was also m e a s u -
red.
Methods
Subjects
Eight healthy young male subjects and eight atopic patients, not
acclimatized to heat and not trained to exercise, volunteered for
these experiments and gave their written consent.
The mean age, height, body mass and skin surface area of the
patients were 24 (SD 7) years, 1.80 (SD 0.05) m, 77.8 (SD 11.7) kg
and 1.96 (SD 0.14)m z, respectively. The corresponding values for
healthy subjects were 23 (SD 3) years, 1.81 (SD0.05)m, 72.0
(SD 5.5) kg and 1.91 (SD 0.09) m 2, respectively.
The criteria for the diagnosis of atopic dermatitis were derived
from the guidelines given by Hanifin and Rajka [8]. Diagnosis of
atopic dermatitis requires the presence of at least three basic features
(pruritus, lichenification, chronically relapsing course, atopic hi-
story) and three minor features (xerosis, ichthyosis, facial ery-
thema ...). The clinical profiles of the patients on the day of the
experimental session are detailed in Table 1.
Experimental protocol
All the experiments were carried out during the same period of the
year (June) to avoid seasonal variations [21, 22] and at the same
time of day to avoid circadian variations. The subjects, wearing
briefs and sports shoes, were exposed to the environmentalcondition
in a climatic chamber in which air, wall and dew-point temperatures
and air velocity were accurately regulated. The subject sat on the
cycloergometer suspended on a scale (sensitivity _+1 g). During
cycling, rotation speed (60 rotations per min) and work load
intensity (90 W) were controlled.
The experimental protocol was as follows:
- Subjects, equipped with all sensors, first sat quietly on the bicycle
for at least 30 min at thermoneutrality: air and wall temperatures
at 28 ~ dew-point temperature at 10 ~ and air at a low
velocity of 0.3 m s- 1.
- Subjects then exercised for 30 min at 90 W external work load
at a speed of 60 rotations per min. At the start of the exercice,
the air and wall temperatures were increased stepwise to 36 ~
For all subjects the same work load was imposed (rather than
some constant percentage of the maximal oxygen consumption)
in order to reduce the variability in the sweat rate per unit of
skin surface. Given a 0.23 net mechanical work efficiency and
a 100 W resting metabolism, the total heat produced in the
body during exercise reached approximately 400 W.
- Subjects then recovered for 30min in the maintained warm
environment.
Recorded variables
Rectal temperature (probe at 11 cm beyond the anal sphincter),
ten local skin temperatures for mean skin temperature determination
[9], heart rate (from electrocardiogram counts), whole body sweat
rate (from body mass variation not corrected for pulmonary
evaporative rate) and local sweat output (with the dew-point
measuring technique as explained below) were recorded each minute
throughout the experiment. Both chest and thigh sweat rates were
obtained continuously using 12-cmz capsules highly ventilated with
air taken from the climatic chamber at a constant flow rate

Table 1. Clinical caracteristics of the eight atopic patients

Patient Associated atopic Family Skin lesions Topography Pruritus Heat

no. manifestations history intolerance

P1 Asthma Yes Erythema, squama (+) Back Yes

Rhinitis Excoriations (+ + +) Trunk
Lichenification (+ +) Nape of the neck, folds + + +
Xerosis (+) General
P2 Asthma No Lichenification (+) Popliteal folds + No
Rhinitis
P3 Asthma Yes Erythema, squama (+ +) Face, nape of the neck Yes
Rhinitis Lichenification (+ +) Popliteal folds + +
Xerosis (+ +) General
P4 Asthma No Erythema, squama (+ + +) Limbs, trunk + + + No
Excoriations (+ + +)
Lichenification (+ +) Folds, hands
Xerosis (+ + +) General
P5 Food intolerance Yes Lichenification (+ +) Nape of the neck, folds + No
Xerosis (+)
P6 Asthma Yes Erythema, squama (+ + +) Face, hands + + Yes
Excoriations ( + +) Hands
Lichenification (+ + )
Xerosis (+)
P7 Yes Erythema, squama (+) Back, face Yes
Excoriations (+)
Lichenifications (+) Limbs + +
Xerosis (+ +)
P8 Asthma Yes Erythema, squama (+ + +) Face Yes
Excoriations (+ + ) Back, Limbs
Lichenification (+ + +) Limbs, nape of the neck ++ +
Xerosis (+ +)
 137

(5lmin-1). All sweat excreted by the underlying glands was 37.3

immediately evaporated by the airstream, thus avoiding any water
accumulation on the skin under the capsule. Reheating the air in i OttoBBmBBBWmml BBIBImBBWIBBB~BBBBmWWBml
37.8
the outflow before the measurement of the dew-point temperature
(Tap, from Schlumberger hygrometers) prevented water vapour
condensation anywhere in the system. Continuous monitoring of 37.4
0
dew-point differences between the upstream air (climatic chamber
P
air) and the downstream air allowed on-line calculations of the local I-

sweating rate. Before entering the capsule, the air temperature was 37.2
regulated by a servosystem to keep the average local skin tem-
perature (T~0, obtained from four thermistors (KTY l 1-2A), at a 37.0
constant target level of 36.0 • 0.2 ~ Under such experimental
conditions, the effect of the local skin temperature upon local
sweating is constant and changes in the sweating rates can be related 36.8 r P = ~ I I I I

to factors other than local thermal ones. 0 10 20 30 40 50 60 70 80

Four sweat capsules were used: two on the chest and two on Time (min)
the external flat part of the thigh. The values given by each area
were averaged to obtain the mean sweat rate of each area (chest Fig. l. Average rectal temperatures (• dotted lines') as a
or thigh). The accuracy of all temperature readings was 0.05 ~ function of time. Thin line (+SEM), normal subjects; thick line
and sensitivity to variations was 0.0l ~ (-SEM), atopic subjects
A plastic sheet of approximately 500 cm2 was stuck onto the
skin of the back (previously washed with distilled water) to collect
sweat for the determination of surface tension, which was determined 360
at 20 ~ using a tensiometer (Kriiss). Results are given in dynes per
centimetre.
35.0

Statistical analysis
34.0
For statistical analyses, independent Student's t-test were used and
p < 0.05 was accepted as significant for comparing the physiological
responses of the eight normal subjects with those of the eight 33.0
patients. Results are given as means of eight subjects, plus or minus
standard errors (figures) or deviations (text and tables) of the means.
i i i i i i , i
320
10 20 30 40 50 60 70 80
Results
Time (min)

F o r each variable, the figures shown in this paper depict
the last 20 min at thermoneutrality followed by the 30 min
of exercise in the heat (20 < t < 50 min) and the 30 rain
of recovering at rest in the same environment (50 < t <
80 min).
Core temperature
Figure 1 shows the rectal temperature (T~e) pattern in
both groups of subjects. Before exercise, no difference at
all could be found between the two groups with a mean
T~e of 37.04 ~ Although the core temperature showed
a slight difference of 0.12 ~ at the end of exercise (normal,
37.55 _+ 0.21 ~ atopic, 37.43 _+ 0.14 ~ between the
two groups, this was not significant (p = 0.18). When
statistical analysis was performed on the change in T~
with time (dT~/dt), again no significant difference could
be discerned (p = 0.26). Increases in T~ were 0.52 _+
0.17 ~ and 0.39 + 0.23 ~ in normal and atopic subjects,
respectively.
Fig. 2. Average mean skin temperature (• SEM, dotted lines) as a
function of time. Thin line (-SEM), normal subjects; thick line
(+ SEM), atopic subjects
the groups before the heat exposure, thus the mean ~k
was 32.91 ~ When the subjects exercised in the heat,
~k increased less in normal subjects than in the patients
(Fig. 2). At the end of the work period, ~k levels were
somewhat different, although not significantly (p = 0.21):
T~k = 34.7 _+ 0.5 ~ in control subjects and ~k = 35.1 _+
0.4 ~ in patients. The changes in mean skin temperature
from minute 20 to minute 50 were, however, significantly
larger (p = 0.038) in the atopic subjects. ~k was + 1.7
_--1-0.5 ~ in control subjects and + 2 . 2 +_ 0.4 ~ in the
patients.
N o systematic differences with a t o p y were observed
in the studied local temperatures of the extremities such
as forehead, hands and feet.
Core to skin temperature difference

As a result of the tendency to lower core temperatures

M e a n skin temperature
The mean skin temperature (f~k) is a reliable indicator
of the overall average shell temperature of the body. In
the present study, no difference in ~k was found between
and higher skin temperatures in atopic patients, the
( T r o - ~k) difference was found to be significantly
lower in these patients (p = 0.04). The difference was
2.8 _+ 0.4 ~ in normal and 2.4 _ 0.4 ~ in atopic sub-
jects at the end of the work period.
138

Whole-body sweat rate Local sweating

Whole-body sweating was the same (p = 0.42) in both
groups:during the 60-min heat exposure, control subjects
lost 465 _+ 68 g and atopic subjects lost 494-t-59 g
(difference not significant, p = 0.35). The sweat rates were
identical (p = 0.44) during work in the heat: the last
5 rain of exercise led to instantaneous sweating of 609 _+
93 g h - 1 in control subjects and of 645 _+ 79 g h - ~ in
patients. A slight, non-significant, difference (p = 0.14)
occurred at the end of the recovery period: 349 __
48 g h -~ in control against 425 +_ 120 g h -1 in atopic
subjects.
As shown in Figs. 3 and 4 for chest and thigh sweat rates,
respectively, no difference existed between subjects (for
clarity of illustration SEMs have been omitted). The
average thigh sweat rate 0.56 + 0 . 1 5 m g m i n -1 cm -2
was significantly (p < 0.001) smaller than the chest sweat
rate 0.97 ___ 0.14 m g m i n - t c m -z, but no difference exi-
sted relating to atopy, either during the exercise load or
the recovery period, as shown in Table 2.
Heart rate

Although no statistical difference appeared, average heart

rates were lower in atopic subjects compared with the
control group at rest (p = 0.13) and at the end of exercise
(p = 0.13). Average heart rates increased from 82 + 6
1.00
beats min -1, during the 5 min period preceding the
exercise, to 135 +_ 12 beats min -~ during the last 5 min
.80
of exercise in atopic subjects, and from 86 _+ 5 to
145 • 12 beats min-1 in normal subjects. The increase
~7 .60
was not significantly different between the two groups
(p = 0.13).
.40

.20 Sweat surface tension

0
0 10 20 30 40 50 60 70 80
The values of surface tension at 20~ were
38.3dyncm-1 (+2.0) for normal subjects and
Time (rain)
41.8 dyn cm-1 (+_ 3.1) for atopic subjects. The difference
Fig. 3. Average chest sweat rates as a function of time. Thin line, between subject groups was significant (p = 0.016).
normal subjects; thick l#Te, atopic subjects

Discussion
.80

In the literature, conclusions relating to thermophysiolo-

.60
gical responses of atopic subjects are controversial. A
number of studies have investigated functional variations
in temperature regulation by examining the vasomotricity
~-' .40 of body segments following local thermal stimuli. The
so-called paradoxical temperature response of the arm
in response to warming the contralateral limb has been
"~ .20 well described [3, 10, 12]. In contrast to the normal
vasodilation of control subjects, atopic patients show
either a rigid or a paradoxical response (i.e. vasoconstric-
0 tion) during exposure of the contralateral arm to warm
0 10 20 30 40 50 60 70 SO
water. Moreover, it has also been stated that the acral
Time (rain}
skin atopic subjects exhibits a lower skin temperature
Fig. 4. Average thigh sweat rates as a function of time. Thin line, than that of normal subjects. From our results it cannot
normal subjects; thick line, atopic subjects be concluded that the thermal state of the extremities

Table 2. Average local sweating Chest Thighs

(mg rain-1 cm-2 + SD) observed on chest
and thighs at the end of both the exercise Control Atopic Control Atopic
and recovery periods and the level of
significance (p) obtained from the t-tests Exercise 0.95 (0.25) 0.98 (0.26) 0.54 (0.14) 0.58 (0.19)
for comparison of means p 0.79 0.65
Recovery 0.47 (0.19) 0.55 (0.16) 0.25 (0.08) 0.26 (0.10)
p 0.35 0.85

was significantly lowered. However, it can be hypo-
thesized that the paradoxical vasomotor responses, if any,
take place locally, and that this does not necessarily reflect
a general disturbance in thermoregulatory responses. In
fact, close inspection of previously published results
demonstrates that the qualitative vasomotor changes are
somewhat difficult to quantify [i0] and sometimes not
reproducible [12].
The results of the present study show a slight tendency
for a lower heart rate in patients with atopy, and suggest
a reduced core to skin temperature difference which, for
the same requirement for heat dissipation, implies a larger
vasodilation in the skin of the atopic subjects. No other
explanation can be put forward since all external factors,
as well as exercise intensity, were controlled and were
the same for both groups. Given the theory of Szentivanyi
[20], a decrease in/~-receptor activity could be an explana-
tion for the lower heart rate level. However, this theory
also implies an associated increase in the c~-receptor
sensitivity, which is not in accordance with a larger
capacity for dissipating the internal heat towards the
periphery. Further investigations are clearly needed to
clarify this point.
The modification of the eccrine sweat output in atopic
subjects is even more controversial. Two studies have
investigated the sudomotor axon reflex on the normal
skin of atopic patients [4, 5]. Although the results of
Greene et al. [5] appeared rather heterogeneous, it was
concluded from these studies that atopy is associated
with reduced sweat gland function. Certain studies have
investigated the sudomotor responses of the lichenified
skin of atopic subjects and have concluded that sweating
is increased at the level of abnormal skin [18, 22].
However, only a few authors have reported changes in
the sweat output of the normal skin of patients with atopy.
In atopic eczematous subjects, no difference is appa-
rent in autonomic function [16], in sweat gland density
[1], or in the sweat response of intact skin [11]. The latter
point is confirmed in the present study which strongly
suggests that the intact skin of atopic patients overlying
active or non-active muscles sweated at the same rate as
that of normal subjects.
The surface tension of sweat collected under thermal
heat stress from atopic subjects, in the present study, was
found to be significantly higher than that from normal
subjects. It is known that there is a disturbance of essential
fatty acid metabolism in atopics [15], which could be
responsible for a decrease in sebaceous secretion [6] and
alterations in the stratum corneum [23], but it has not
been shown that an increased sweat surface tension is
related to this metabolic deficiency. The possible de-
creased concentration of tensioactive constituent (s) [6] in
eccrine thermal sweating with atopy requires detailed
chemical analysis of the sweat secreted to gain better
knowledge of the nature of the surfactants involved
in these tensioactivity changes.
As a consequence of the significant difference that we
found between values of the sweat surface tension of
normal subjects and patients, it may be concluded that
sweat gland function, in terms of composition of sweat
secretion, may be altered by atopic dermatitis [6, 14], but
139
there is no effect of atopy on the central mechanisms of
temperature regulation. The higher level of mean skin
temperature found in atopic subjects could result from
reduced evaporative efficiency and/or from a larger core
to skin heat transfer. However, our study, showing no
quantitative change in evaporation and no significantly
different heart rates between the groups, does not resolve
this point. The thermoregulatory modifications, if any,
may be related more to changes in peripheral vasomotor
adjustments rather than in central modifications of swea-
ting mechanisms. Neural thermal input or output altera-
tions, hypothalamus dysfunction, as well as neuroglandu-
lar deficiency are unlikely to be the origin of the small
but noticeable changes in temperature regulation associa-
ted with atopic dermatitis.
The present study demonstrates clearly that there are
no central modifications of the temperature regulatory
mechanisms associated with atopy. The difficulty in
withstanding heat exposure is probably due to acute
symptoms, such as pruritus or asthma, which can take
place in atopic patients, rather than to inadequate
vegetative thermoregulatory responses.
Acknowledgements. The authors thank M. Gartner, F. Hinkel and
G. Ober for technical assistance, A. Hoeft for computing assistance
and J. Saini and J. Nooney for correction of the manuscript.
References
1. Bachmann-Andersen L, Thestrup-Petersen K (1987) Sweat pore
density on the fingertips of atopic patients. Br J Dermatol 117:
225-230
2. Candas V (1987)Adaptation to extreme environments.Thermo-
physiological changes in man during humid heat acclimation.
In: Dejours P (ed) Comparative physiology of environmental
adaptation, vol 2. Eighth ESCP Conference, Strasbourg, 1986.
Kager, Basel, pp 76-93
3. Champion RH (1963) Abnormal vascular reactions in atopic
eczema. Br J Dermatol 75:12-15
4. Fujita M, Aoki T (1975) Sweat suppression in patients with
atopic dermatitis and infantile xerotic eczema. Jpn J Dermatol
85:291-294
5. Greene RM, Winkelman RK, Opfer-Gehrkring TL, Low PA
(1989) Sweating patterns in atopic dermatitis patients. Arch
Dermatol Res 281:373-376
6. Grosshans E, Woehl M (1982) Troubles vasomoteurs, sudoraux
et s6bac6s de l'ecz6ma constitutionnel. Ann Dermatol Venereol
(Paris) i09:151-162
7. Grosshans E, Juillard J, Libert JP, Candas V, Vogt JJ (1980)
Physiopathology of sweating response in individuals with
eccrine glands of the rare reticulated clear cell type. J Invest
Dermatol 74:455-456
8. Hanifin JM, Rajka G (1980) Diagnostic features of atopic
dermatitis. Acta Derm Venereol Suppl 92:44-47
9. Hardy JD, DuBois EF (1938) Basal metabolism, radiation,
convection and evaporation at temperatures of 22-35 ~ J.
Nutr 15:477-496
10. Heyer G, Sch6nberger A, Hornstein OP (1988) St6rungen der
Thermoregulation auf externe Kfilte- und W/irmereize bei
Patienten mit atopischem Ekzem. Hautarzt 39:18-22
11. Kiistala R (1982) Stimulated local sweating response in atopic
dermatitis (abstract). XVI Congressus Internationalis Der-
matologiae, Tokyo, May 1982. p 320
12. Kocsard E, Ofner F, Broe GA (1973) Paradox temperature
response in atopic dermatitis. Dermatologica 146:8-14
140
13. Lobitz WC, Campbell CJ (1953) Physiologic studies in atopic
dermatitis (disseminated neurodermatitis). I. The local cuta-
neous response to intradermally injected acetylcholine and
epinephrine. Arch Dermatol Syphilol 67:575-589
14. MaliJWH (1971) Skin lesions of the atopic syndrome. The
sweat gland as a mucous gland. Dermatologists Nijmegen,
March 1970. Dermatologica 142:103-104
15. Manku MS, Horrobin DF, Morse NL, Wright S, Burton JL
(1984) Essential fatty acids in the plasma phospholipids of
patients with atopic eczema. Br J Dermatol 110:643 648
16. Murphy GM, Smith SE, Smith SA, Greaves MW (1984) Auto-
nomic function in cholinergic urticaria and atopic eczema. Br
J Dermatol 110:581-586
17. Ogawa T, Asayama M (1986) Quantitative analysis of the local
effect of skin temperature on sweating. Jpn J Physiol 36:
417-422
18. Rovensky J, Saxl O (1964) Differences in the dynamics of sweat
secretion in atopic children. J Invest Dermatol 43:171-176
19. Sulzberger MB, Herrmann F, Zak FG (1947) Studies of swea-
ting. I. Preliminary report with particular emphasis on a sweat
retention syndrome. J Invest Dermatol 9:221-242
20. Szentivanyi A (1968) The beta-adrenergic theory of the atopic
abnormality in bronchial asthma. J Allergy 42:203-232
21. Thune P, Kocsis M (1975) Effects of adrenergic stimulating and
blocking agents on the eccrine sweat secretion in atopic
dermatitis and psoriasis. Arch Dermatol Res 253:97-103
22. WarndorffJA (1970) The response of the sweat gland to
acetylcholine in atopic subjects. Br J Dermatol 83:306-311
23. Werner Y, Lindberg M (1985) Transepidermal water loss in dry
and clinically normal skin in patients with atopic dermatitis.
Acta Derm Venereol (Stockh) 65:102-105