Acta Anaesthesiol Scand 2002; 46: 43–50
Printed in Denmark. All rights reserved
Construction and evaluation of a manikin for perioperative
heat exchange
A. BRÄUER1, M. J. M. ENGLISH2, H. SANDER1, A. TIMMERMANN1, U. BRAUN1 and W. WEYLAND3
1
Department of Anesthesiology, Emergency and Intensive Care Medicine, University of Göttingen, Göttingen, Germany and 2Department of Anaesthesia,
Montreal General Hospital and McGill University, Montreal, Canada and 3Department of Anaesthesia and Intensive Care Medicine, Evangelisches
Bethesda-Krankenhaus, Essen, Germany
Background: During surgery hypothermia can be avoided only
if the heat exchange between the body surface and the environ-
ment can be controlled. To allow a systematic analysis of this
heat exchange, we constructed and evaluated a copper manikin
of the human body.
Methods: The manikin consists of six tubes (head, trunk, two
arms and two legs) painted matt-black to simulate the emiss-
ivity of the human skin. Hot-water mattresses are bonded to the
inner surface of the copper tubes to set the surface temperature.
Calibrated heat flux transducers were placed on the following
points to determine the heat exchange coefficient for radiation
and convection (hRC) of the manikin: Forehead, chest, abdomen,
upper arm, forearm, dorsal hand, anterior thigh, anterior leg
and foot. Room temperature was set to 22æC. Surface tempera-
ture of the manikin was set between 22æC and 38æC. The hRC
was determined by linear regression analysis as the slope of the
temperature gradient between the manikin and the room versus
P ERIOPERATIVE hypothermia is still a very common
problem during anaesthesia and is associated
with various medical risks. Intraoperatively the dis-
turbed function of thrombocytes causes a prolonga-
tion of the bleeding time, and this can increase intra-
operative blood loss (1, 2). In the postoperative phase
hypothermic patients with coronary heart disease
have a significant higher risk of arrythmias, myocar-
dial ischaemia and unstable angina pectoris (3, 4).
Furthermore, perioperative hypothermia can lead to a
higher incidence of surgical-wound infection and to a
longer hospitalisation of patients (5).
The physical principles of the mechanisms of en-
vironmental heat exchange and their susceptibility to
manipulation have been understood for decades (6,
7). The application of this knowledge has made it
possible for humans to remain normothermic even
under extreme thermal conditions, e.g. in the ocean
depths and outer space. In contrast, surgical patients
become hypothermic under relatively mild environ-
mental conditions.
Copyright C Acta Anaesthesiol Scand 2002
ACTA ANAESTHESIOLOGICA SCANDINAVICA
ISSN 0001-5172
the measured heat flux. Subsequently we studied five minimally
clothed volunteers in a climate chamber. Initial chamber tem-
perature was set to 29æC and was lowered slowly to 12æC. The
hRC was determined as described above for each volunteer.
Results: The hRC of the manikin was 11.0 W mª2 æCª1 and hRC
of the volunteers was 10.8 W mª2 æCª1.
Conclusion: The excellent correlation of hRC between the volun-
teers and the manikin will allow the manikin to be used for
standardised studies of perioperative heat exchange.
Received 24 January, accepted for publication 25 May 2001
Key words: Hypothermia; perioperative; heat exchange; meas-
urement techniques; manikin.
c Acta Anaesthesiologica Scandinavica 46 (2002)
During long surgical operations perioperative hy-
pothermia can only be avoided if the patient’s heat
loss is offset by an equal heat gain, either from meta-
bolic heat production or from an external heat source.
It is important to understand the principles of heat
exchange between the body surface and the environ-
ment to control the heat losses of the patient. In an
animal experiment it has been demonstrated that even
during large abdominal operations the major source
of heat loss was the radiative and convective heat loss
from the skin. Although evaporation accounted for
the largest heat loss from the abdominal cavity, it was
a minor source due to the smaller heat-losing area (8).
As a consequence the radiative and convective heat
exchange of the body surface is of great clinical im-
portance.
Radiative and convective heat exchange can be de-
scribed as follows:
Q̇/A Ω hRC ¡ DT (Eqn. 1)
where:
43
A. Bräuer et al.
Q̇/A Ω heat flux per area [W mª2]
hRC Ω combined heat exchange coefficient for
radiation and convection [W mª2 æCª1]
DT Ω temperature gradient between the skin and
the environment [æC]
To apply this physical principle to surgical patients
the heat exchange must be measured under the speci-
fic conditions in the operating room (O.R.). A manikin
allows single factors to be changed in a systematic
way without any risk to patients. Therefore, we have
Fig. 1. Schematic diagram of the manikin. Overview and cross-section.
The manikin consists of six copper tubes painted matt-black. The ex-
posed surface is 1.98 m2. Hot-water mattresses are bonded to the inner
surface of the copper tubes to set the surface temperature and to trans-
port heat inside the manikin. These mattresses are connected to a Hico-
Variotherm 530 Hyper-Hypothermia System (Hirtz, Germany).
44
constructed and evaluated a copper manikin of the
human body. The present study was designed to de-
termine whether the combined heat exchange coef-
ficient for radiation and convection (hRC) of this
manikin correlates well with the hRC of volunteers.
Methods
The manikin (Fig. 1)
The manikin consists of six tubes. Two tubes serve as
arms (circumference 330 mm, thickness 2 mm, length
705 mm), two as legs (circumference 485 mm, thick-
ness 2 mm, length 750 mm), one as the head (circum-
ference 500 mm, thickness 3 mm, length 330 mm) and
one as the trunk (circumference 840 mm, thickness 3
mm, length 740 mm). The tube representing the trunk
was pressed into an ellipsoid form to simulate the
shape of the human body. The total surface area of all
tubes is 1.98 m2. In order to set surface temperature
and achieve steady-state conditions, hot-water mat-
tresses (Maxi-Therma, Cincinnati Sub-Zero Products
Inc., Cincinnati, Ohio, USA) are bonded to the inner
surface of the copper tubes. The water is warmed or
cooled by a hypo-hyperthermia system (Hico-Vario-
therm 530, Hirtz & Co. Hospitalwerk, Cologne, Ger-
many).
The high thermal conductivity of copper (389 W
mª1 æCª1) facilitates homogeneous distribution of
heat on the manikin surface. The manikin was painted
matt-black to simulate the radiant emissivity of the
human skin, 0.98 (7).
Determination of the emissivity of the manikin
surface
The emissivity of the manikin surface was determined
by measuring its surface temperature with a calibrated
thermistor incorporated in a heat flux transducer (Heat
Flow Sensor Model FR-025-TH44033-F16, Concept En-
gineering, Old Saybrook, CT, USA) and simul-
taneously, and in the same area, with an infrared ther-
mometer (Non-contact Infrared Thermometer THI-300,
Tasco Japan Co. ltd., Osaka, Japan). The emissivity con-
trol of the infrared thermometer was then adjusted un-
til its temperature reading equalled the value of the
thermistor. These measurements were performed on
the head, on the trunk, on an arm and on a leg of the
manikin. The average of the emissivity values was
taken as the emissivity of the manikin surface.
Measurement of heat flux
To measure the heat flux from the surface of the
manikin to the environment, we used heat flux trans-
ducers (Heat Flow Sensor Model FR-025-TH44033-F16,

Concept Engineering, Old Saybrook, CT, USA). The
transducers have a diameter of 2.5 cm and a thickness
of 2.25 mm. The thermal conductivity is specified by
the manufacturer as 0.316 W mª1 æCª1. Therefore, the
thermal resistance is 7.12 ¡ 10ª3 æC m2 Wª1.
Calibration of the heat flux sensors
The heat flux sensors were calibrated at the Thermal
Insulating Material Testing Laboratory at the School
of Building, Concordia University, Montreal, Canada,
with a Dynatech R-Matic heat-flow meter (Dynatech,
Cambridge, MA, USA). It conforms to ASTM C-518,
the standard test method for measuring steady-state
thermal transmission by means of a heat flow meter.
The overall average accuracy of calibration is ex-
pected to be ∫3%.
Measurement of the manikin surface temperature
and room temperature
Surface temperature of the manikin was measured
with thermistors (YSI Inc. ‘‘Thermolinear’’ 44018) in-
corporated into the heat flux sensors. Room tempera-
ture was measured in the middle of the room using a
thermocouple (MAT Myocardial sensor 18 mm, Mal-
linckrodt Medical, Hennef/Sieg, Germany).
Calibration of the temperature probes
The thermistors were calibrated in a water bath at
temperatures of 22æC and 40æC. The reference ther-
mometer was a Hewlett Packard Model 2801 (Palo
Alto, CA, USA) with an accuracy of ∫0.01æC. The
thermistors were corrected to the value of the refer-
ence thermometer. The thermocouples were also
tested in the same way and thermocouples which
showed an error of more than ∫0.1æC were excluded.
Measurement of environmental conditions
Air humidity and air velocity were measured using a
thermoanemometer (VELOCICALC PLUS TSIA
Model 8388-M-D, TSI Incorporated, St. Paul, MN,
USA). This device measures relative air humidity be-
tween 0% and 95% with a maximal error of ∫3% and
air velocities between 0.15 m/s and 50 m/s with a
maximal error of ∫3%.
Since mean radiant temperature does not differ sig-
nificantly from room temperature under normal con-
ditions, we did not measure mean radiant tempera-
ture (9, 10).
Determination of the heat exchanging properties
for radiation and convection of the manikin
Nine heat flux sensors were placed on the following
points: Forehead, chest, abdomen, upper arm, fore-
A manikin for perioperative heat exchange
arm, dorsal hand, anterior thigh, anterior leg and foot
of the manikin. These points were chosen because
they represent measuring points of a 14-point formula
to determine the mean skin temperature (11). Five
measuring points on the back of the manikin were
excluded because they were not exposed to the sur-
rounding air; this part of the manikin loses heat by
conduction instead of radiation and convection.
The manikin evaluation took place in a room with
a constant temperature of 22æC. Relative humidity
was 40–50% and air velocity was ⬍0.2 m/s. The tem-
perature gradient between the surface of the manikin
and the room was changed by varying the manikin
surface temperature in small steps between 22æC and
38æC. After each step we waited till steady-state con-
ditions were achieved (approximately 20 min) and
data were recorded continuously for a period of 3
min.
The average values of these periods from each
single heat flux transducer and each single surface
temperature measurement were used to determine the
combined heat exchange coefficient for radiation and
convection (hRC). The slope of the regression line of
the temperature gradient between the surface of the
manikin and the room, and the heat flux, represents
hRC. Heat flux from the manikin to the environment
was called ‘‘heat loss’’ and was assigned a negative
value.
The regression line was forced through zero be-
cause at a temperature gradient of 0æC there can be
no heat flux (9).
Determination of the heat exchanging properties
for radiation and convection of the volunteers
After informed consent and approval by the local in-
stitutional ethics committee, five healthy men were
studied in a climate chamber. These volunteers had
been fasting for at least 8 h. Measurements were car-
ried out between 8:00 and 11:00 a.m. During the ex-
periment the volunteers were only minimally clothed.
Heat flux sensors were placed on the same points
as on the manikin. Measurements of heat flux, skin
temperature, room temperature, air velocity and air
humidity were performed as described in the manikin
evaluation.
At the beginning of each experiment the air tem-
perature of the climate chamber was set at a ther-
moneutral temperature of 29æC. Relative humidity
was 40–50% and air velocity was 0.24 ∫0.05 m/s.
After 60 min the temperature of the climate chamber
was lowered slowly to 12æC within 2 h. From the end
of the control period and during the cooling period
heat flux and temperature data were recorded every
45
A. Bräuer et al.

minute and sampled for analysis every 5 min. As de-

scribed above hRC was determined by a linear re-
gression analysis for each volunteer.

Determination of the influence of vasoconstriction

on heat exchanging properties for radiation and
convection
Peripheral vasoconstriction was measured as the tem-
perature gradient between the forearm and the
middle fingertip using two tested thermocouples
(Mon-a-therm Skin Temperature Probe, Mallinckrodt
Medical, Hennef/Sieg, Germany) (12).
The calculated hRC values of each single measure-
ment point were plotted against the forearm-fingertip
temperature gradient and a linear regression analysis
was performed to look for a possible relation between
these two values. A forearm-fingertip temperature
gradient of more than 0æC was defined as the begin-
ning of thermoregulatory vasoconstriction (12).

Results
Emissivity of the manikin Fig. 3. Combined heat exchange coefficient (hRC) of one volunteer. The
slope of the temperature gradient between room and skin of the volun-
Emissivity of the manikin surface: Head 0.93, trunk
teer versus the measured heat flux per area. Regression line and 95%
0.98, arm 0.93 and leg 1.0. Average manikin emissivity confidence intervals.
was calculated as 0.96.

Combined heat exchange coefficient for radiation

and convection of the manikin
A total of 270 data points was used for the determi-
nation of hRC of the manikin. The combined heat ex-
change coefficient for radiation and convection was
11.0 W mª2 æCª1 (SEM: 0.65). The coefficient of deter-
mination (r2) was 0.80 (Fig. 2).

Combined heat exchange coefficient for radiation

and convection of the volunteers
Age of the five volunteers ranged between 22 and 49
years, height ranged between 174 and 194 cm and
their body weight ranged between 68 and 95 kg.

Table 11
Results of the volunteers. Combined heat exhange coefficient for
radiation and convection (hRC), standard error of the mean (SEM),
coefficient of determination (r2) and the number of data points that
were used (n).

Volunteer hRC [W mª2 æCª1] SEM r2 n

1 10.7 0.80 0.73 180

Fig. 2. Combined heat exchange coefficient (hRC) of the manikin. The 2 11.0 0.81 0.76 180
slope of the temperature gradient between room and surface of the 3 11.3 0.75 0.77 225
4 10.6 0.70 0.61 224
manikin versus the measured heat flux per area. Regression line and
5 10.2 0.76 0.80 180
95% confidence intervals.

46

A total of 989 data points was used for the determi-
nation of hRC of the volunteers. The average combined
Fig. 4. Influence of vasoconstriction on the combined heat exchange
coefficient (hRC). Calculated hRC values of each single measurement
point versus the forearm-fingertip temperature gradient. Regression
lines of all five volunteers. The 95% confidence intervals were omitted
for clarity.
Fig. 5. Isocorrection factor diagram. The isocorrection factor lines are
plotted as a function of ratios of thermal resistances of tissues and
insulation to that of the heat flux transducer. The darkly shaded zones
are the zones of interest for heat exchange of humans and the manikin
under the conditions of the O.R.
A manikin for perioperative heat exchange
heat exchange coefficient for radiation and convection
was 10.8 W mª2 æCª1 (Table 1). Fig. 3 shows the data
of one volunteer.
Influence of vasoconstriction on heat exchanging
properties for radiation and convection
There was no influence of the forearm-fingertip tem-
perature gradient as a measure of vasoconstriction on
the heat exchange coefficient for radiation and con-
vection for any of the five volunteers (Fig. 4). The cor-
relation coefficients of all the linear regression analy-
ses were below 0.05.
Discussion
Anaesthetised patients often become hypothermic
during surgery. Hypothermia results from a combi-
nation of impaired thermoregulation and exposure to
the cold environment of the O.R. (13). The impairment
of thermoregulatory mechanisms leads to a redistri-
bution of heat from the warm core thermal compart-
ment to cooler peripheral tissues after induction of an-
aesthesia (14). Decreased heat production and in-
creased heat loss from the surgical field further
contribute to the disturbance in heat balance (14, 15).
To prevent hypothermia, the body’s heat balance must
be maintained, which means that heat loss of the body
must be balanced by an equal amount of heat gain by
the body either from the body’s own internal meta-
bolic heat production or from an external source of
heat. To understand this heat balance one must under-
stand the physical principles of heat exchange from
the body surface to the environment, because the skin
is the predominant source of heat loss during surgery
(8, 13). A manikin allows these principles to be
studied under the specific conditions of the O.R. with-
out any disturbance to clinical practice.
Thermal manikins are used extensively in environ-
mental physiology (7, 16), because they allow quick,
reliable and accurate measurements. Manikins are a
useful and valuable complement to experiments with
human volunteers. The first copper manikins were
made in the early 1940s and the development of new
manikins still continues (16). Main application areas
of thermal manikins are:
O Relevant simulation of human body heat exchange
O Integration of dry heat exchange in a realistic
manner
O Measurement of clothing thermal resistance
O Detailed assessment of thermal stress in different
environments
47
A. Bräuer et al.
O Development of prediction models
O Product development.
However, the specific aims and needs of the research
problem may require the manikin to have specific de-
sign and performance features. At the moment, most
manikins allow only measurement of heat loss (16).
However, in the O.R. heat gain occurs from clinical
warming systems and, therefore, a manikin for the
measurement of perioperative heat exchange must
measure both heat loss and heat gain. Our manikin
simulates heat loss and heat gain by radiation, con-
vection and conduction between its surface and the
environment. Evaporative heat losses from open body
cavities can be simulated when the heat flux trans-
ducers are covered with wet gauze (8).
In contrast, heat losses due to the physiologically
insensible evaporation from the skin cannot be simu-
lated, but these heat losses are not of great importance
and are estimated to be about 10% of the radiative
and convective heat loss. Although this may be an
invalid estimation, this estimation is used very often
in experiments with volunteers, because insensible
evaporation from the skin can not be measured by
simple techniques (17).
It has been claimed that manikins are not appropri-
ate for the simulation of perioperative heat exchange,
because they are not able to simulate thermoregulato-
ry vasoconstriction. However, thermoregulatory vaso-
constriction over the whole range of clinically ob-
served forearm-fingertip temperature gradients of
more than 15æC (12) does not influence the rate of heat
exchange from the skin, the value of hRC, as can be
seen in Fig. 4. Vasoconstriction will reduce the amount
of heat loss from the skin by reducing the skin surface
temperature and, therefore, the temperature gradient
between the skin and environment – but vasoconstric-
tion will not reduce the rate of heat loss. Therefore,
the calculation of heat exchange coefficients from this
manikin are valid and the fact that this manikin does
not simulate vasoconstriction does not reduce its abil-
ity to simulate heat exchange between the body sur-
face and the environment.
Use of heat flux transducers
The use of heat flux transducers in experiments study-
ing heat balance offers a unique advantage in com-
parison with temperature measurements, because
heat flux transducers directly measure heat loss or
heat gain, which temperature measurements do not.
Only if the heat exchange coefficient is known (which,
in itself, requires the use of heat flux transducers) and
the temperature gradient is measured, can heat loss
48
or heat gain be calculated from temperature measure-
ments alone.
Heat flux transducers contain two thermopiles sep-
arated by a matrix with a fixed thermal resistance (9,
18). When heat flows through a heat flux transducer
the matrix causes a temperature gradient to develop
between the two thermopiles. By the Seebeck effect,
each thermopile generates a voltage proportional to
its absolute temperature. The differential voltage be-
tween the two thermopiles is proportional to the tem-
perature gradient and therefore, since the thermal re-
sistance of the matrix is fixed, to the heat flow through
the heat flux transducer (9).
With heat flux transducers it is possible to measure
heat gain or heat loss by radiation, convection and
conduction. Ducharme et al. strongly recommend cali-
bration of the heat flux transducers because the cali-
bration constants given by the manufacturer are often
inappropriate (18). The calibration constants of our
heat flux transducers differed between 10.4% and 24%
from the value given by the manufacturer, with an
average error of 15.9%.
Comparison of the combined heat exchange
coefficient for radiation and convection (hRC)
between the manikin and the volunteers
The relationship between the measured temperature
gradient from the surface to the room and the ob-
served heat flux was linear. For every 1æC temperature
gradient the volunteers lost 10.8 W mª2 and the
manikin lost 11.0 W mª2.
The hRC value of the volunteers is slightly higher
than those of Clark and Edholm (7) and English and
co-workers (9, 20). These authors have described
values between 9.7 and 10.2 W mª2 æCª1 in rooms
with an air velocity below 0.2 m/s. Calculations based
on the formula of Kerslake (21) and English (9) show
that at an air velocity of 0.24 m/s the expected hRC is
between 10.8 W mª2 æCª1 and 10.7 W mª2 æCª1. Our
result of 10.8 W mª2 æCª1 suggests that our methods
of data acquisition and analysis are accurate and re-
liable.
The difference between human skin and a copper
surface regarding heat flux measurement error re-
quires discussion (18, 19).
Estimation of the heat flux measurement error
The very presence of a heat flux transducer on a sur-
face will influence the heat flux from that surface and,
therefore, create a source of error.
Frim and Ducharme developed a graphic method
to estimate the magnitude of this error in heat flux
measurement (19). This method considers that heat is

being transferred from the body core through the
tissues and external insulation to the environment.
External insulation does not necessarily involve cloth-
ing; in the case of a nude subject it can be the layer of
relatively still air, the ‘‘boundary layer’’, lying over
the skin. It is assumed that the thermal resistance of
the heat flux transducer adds to the insulation over
the skin.
The true heat flux can then be calculated as follows:
Q̇True Ω (TCoreªTEnvironment)/(RTissuesπRInsulation)
where:
Q̇ Ω heat flux per area [W mª2]
T Ω temperature [æC]
R Ω thermal resistance [æC m2 Wª1] i.e. the
reciprocal of the heat exchange coefficient
The heat flux through the heat flux transducer can be
calculated as follow:
Q̇Measured Ω (TCoreªTEnvironment)/
(RTissuesπRInsulationπRHFT)
The heat flux correction factor is the ratio between
Q̇True and Q̇Measured:
Q̇True/Q̇Measured Ω 1π1/(RTissues/ (Eqn. 2)
RHFTπRInsulation/RHFT)
This correction factor is a function of ratios of the
thermal resistances of the tissues, the insulation and
the heat flux transducer. The thermal resistance of the
tissues of the human body can vary widely due to
vasoconstriction and vasodilation from 0.151æC m2
Wª1 to 0.005æC m2 Wª1 (19). The thermal resistance
of the copper manikin ranges between 5.1 ¡ 10ª6 æC
m2 Wª1 and 7.7 ¡ 10ª6 æC m2 Wª1, depending on the
thickness of the material. The thermal resistance of the
heat flux transducer is 7.12 ¡ 10ª3 æC m2 Wª1 and the
thermal resistance of any insulation added to the
body can vary enormously, depending on the type of
insulation and its thickness.
For practical reasons we took the upper limit for
added external insulation as 4 clo (0.62æC m2 Wª1).
No material used in the O.R. has such a high value.
The lower limit is the insulation of the air itself. The
insulating properties of air vary depending on air
velocity: the higher the air velocity, the lower the insu-
lation (6). In Germany, laminar flow O.R.s have a
maximal air velocity of 0.45 to 0.5 m/s with a corre-
sponding insulation of 0.55 clo (0.085æC m2 Wª1) (6).
A manikin for perioperative heat exchange
An isocorrection graph was constructed, as sug-
gested by Frim and Ducharme (19). RTissue/RHFT and
RInsulation/RHFT are used as the abscissa and ordinate,
respectively, of a log-log graph. Equation 2 was solved
for various values of Q̇True/Q̇Measured to produce a
family of isocorrection factor lines which represent
constant values of the correction factor as the ratios of
RTissue/RHFT and RInsulation/RHFT both vary.
After drawing the relevant resistance values for our
experiments in that isocorrection factor diagram the
possible errors for measurements from the metal sur-
face range between 1.3% for maximal insulation up to
nearly 10% for minimal insulation. Those errors are
acceptable, especially because the errors for measure-
ments from the human skin are the same for vasodil-
ated volunteers, as shown in Fig. 5.
These error estimations are based on one-dimen-
sional heat flow, and heat flows from other geometries
may need more complicated correction factors. How-
ever, experiments in volunteers have shown the prac-
ticality of this method of error calculation (19).
Conclusion
Since the value of the combined heat exchange coef-
ficient for radiation and convection is the same in a
manikin as in volunteers, we conclude that it is poss-
ible to use this manikin for standardised studies of
perioperative heat exchange, and that the results will
be comparable to those of human volunteers. The
manikin will also allow us to compare different pa-
tient warming devices in a standardised manner.
Acknowledgements
Thomas Schulze and Andrew Scott are thanked for program-
ming the data acquisition program and for technical help.
We are grateful to CMV Computer-Meßtechnik-Vertrieb,
Wolfsittard 10, D-41179 Mönchengladbach, Germany for lending
the Non-contact Infrared Thermometer THI-300 (Tasco Japan
Co. ltd., Osaka, Japan).
References
1. Valeri CR, Khabbatz K, Khuri SF, Marquardt C, Ragno G,
Feingold H et al. Effect of skin temperature on platelet func-
tion in patients undergoing extracorporal bypass. J Thorac
Cardiovasc Surg 1992: 104: 108–116.
2. Schmied H, Kurz A, Sessler DI, Kozek S, Reiter A. Mild hy-
pothermia increases blood loss and transfusion require-
ments during total hip arthroplasty. Lancet 1996: 347: 289–
292.
3. Frank SM, Beattie C, Christopherson R, Norris EJ, Perler BA,
Williams GM et al. Unintentional hypothermia is associated
with postoperative myocardial ischemia. Anesthesiology
1993: 78: 468–476.
49
A. Bräuer et al.
4. Frank SM, Fleischer LA, Breslow MJ, Higgins MS, Olsen KF,
Kelly S et al. Perioperative maintenance of normothermia
reduces the incidence of morbid cardiac events. A ran-
domized clinical trial. JAMA 1997: 277: 1127–1134.
5. Kurz A, Sessler DI, Lenhard R, and the Study of Wound
Infection and Temperature Group. Perioperative normother-
mia to reduce the incidence of surgical-wound infection and
shorten hospitalization. N Engl J Med 1996: 334: 1209–1215.
6. Burton AC, Edholm OG. Man in a cold environment. London:
Edward Arnold Publishers, 1955.
7. Clark RP, Edholm OG. Man and his thermal environment. Lon-
don: Edward Arnold Publishers, 1985.
8. English M, Papenberg R, Farias E, Scott WAC, Hinchey J.
Heat loss in an animal experimental model. J Trauma 1991:
31: 36–38.
9. English MJM, Farmer C, Scott WAC. Heat loss in exposed
volunteers. J Trauma 1990: 30: 422–425.
10. Kleiber M. The fire of life. New York: John Wiley and Sons
Inc., 1961.
11. Layton RP, Mints WH, Annies JF, Rack MJ, Webb P. Calori-
metry with heat flux transducers: comparison with a suit
calorimeter. J Appl Physiol 1983: 54: 1361–1367.
12. Rubinstein EH, Sessler DI. Skin-surface temperature gradi-
ents correlate with fingertip blood flow in humans. Anes-
thesiology 1990: 73: 541–545.
13. Sessler DI. Mild perioperative hypothermia. N Engl J Med
1997: 336: 1730–1737.
14. Matsukawa T, Sessler DI, Sessler AM, Schroeder M, Ozaki
M, Kurz A et al. Heat flow and distribution during induc-
tion of general anesthesia. Anesthesiology 1995: 82: 662–673.
15. Brismar B, Hedenstierna G, Lundh R, Tokis L. Oxygen up-
take, plasma catecholamines and cardiac output during neu-
50
rolept-nitrous oxide and halothane anaesthesias. Acta Anaes-
thesiol Scand 1982: 26: 541–549.
16. Holmér I. Thermal manikins in research and standards. In:
Nilsson HO, Holmér I, eds. Proceedings of the Third Interna-
tional Meeting on Thermal manikin Testing, 3IMM, at the Na-
tional Institute for Working Life, October 12–13, 1999, Stock-
holm: National Institute for Working Life, 2000: 1–7.
17. Kurz A, Sessler DI, Christensen R, Dechert M. Heat balance
and distribution during the core-temperature plateau in
anesthetized humans. Anesthesiology 1995: 83: 491–499.
18. Ducharme MB, Frim J, Tikuisis P. Errors in heat flux meas-
urement due to thermal resistance of heat flux disks. J Appl
Physiol 1990: 69: 776–784.
19. Frim J, Ducharme MB. Heat flux transducer measurement
error: a simplified view. J Appl Physiol 1993: 74: 2040–2044.
20. English M, Farmer C, Scott A. Environmental heat exchange
measured with heat flux transducers. Can J Anaesth 1989: 36:
S144–S145.
21. Kerslake D. McK. The stress of hot environments. Monographs
of the Physiological Society No 29. Cambridge: Cambridge
University Press, 1972.
Address:
Dr. Anselm Bräuer
Department of Anesthesiology, Emergency and
Intensive Care Medicine
University of Göttingen
Robert-Koch-Str. 40
D-37075 Göttingen
Germany
e-mail: abraeue/gwdg.de