fisiologi 1

CALL FOR PAPERS Physiology and Pharmacology of Temperature Regulation

A three-compartment thermometry model for the improvedestimation of changes in body heat content

Ollie Jay,1 Louise M. Gariepy,1 Francis D. Reardon,1 Paul Webb,2

Michel B. Ducharme,1,3 Tim Ramsay,4 and Glen P. Kenny1

1Laboratory of Human Bioenergetics and Environmental Physiology, School of Human Kinetics, Faculty of Health Sciences,University of Ottawa, Ottawa, Ontario, Canada; 2Yellow Springs, Ohio; 3Defence R&D Canada, Toronto, Ontario, Canada;and 4Ottawa Health Research Institute, The Ottawa Hospital, General Campus, Ottawa, Ontario, Canada

Submitted 18 May 2006; accepted in final form 21 August 2006

Jay, O, Gariepy LM, Reardon FD, Webb P, Ducharme MB,Ramsay T, Kenny GP. A three-compartment thermometry model forthe improved estimation of changes in body heat content. Am JPhysiol Regul Integr Comp Physiol 292: R167–R175, 2007. Firstpublished August 24, 2006; doi:10.1152/ajpregu.00338.2006.—Theaim of this study was to use whole body calorimetry to directlymeasure the change in body heat content (�Hb) during steady-stateexercise and compare these values with those estimated using ther-mometry. The thermometry models tested were the traditional two-compartment model of “core” and “shell” temperatures, and a three-compartment model of “core,” “muscle,” and “shell” temperatures;with individual compartments within each model weighted for theirrelative influence upon �Hb by coefficients subject to a nonnegativeand a sum-to-one constraint. Fifty-two participants performed 90 minof moderate-intensity exercise (40% of VO2 peak) on a cycle ergometerin the Snellen air calorimeter, at regulated air temperatures of 24°C or30°C and a relative humidity of either 30% or 60%. The “core”compartment was represented by temperatures measured in the esoph-agus (Tes), rectum (Tre), and aural canal (Tau), while the “muscle”compartment was represented by regional muscle temperature mea-sured in the vastus lateralis (Tvl), triceps brachii (Ttb), and uppertrapezius (Tut). The “shell” compartment was represented by theweighted mean of 12 skin temperatures (T� sk). The whole bodycalorimetry data were used to derive optimally fitting two- andthree-compartment thermometry models. The traditional two-com-partment model was found to be statistically biased, systematicallyunderestimating �Hb by 15.5% (SD 31.3) at 24°C and by 35.5% (SD21.9) at 30°C. The three-compartment model showed no such bias,yielding a more precise estimate of �Hb as evidenced by a meanestimation error of 1.1% (SD 29.5) at 24°C and 5.4% (SD 30.0) at30°C with an adjusted R2 of 0.48 and 0.51, respectively. It isconcluded that a major source of error in the estimation of �Hb usingthe traditional two-compartment thermometry model is the lack of anexpression independently representing the heat storage in muscleduring exercise.

body heat storage; calorimetry; muscle temperature; thermoregulation

THE DERIVATION OF THE CHANGE in body heat content (�Hb) is offundamental importance to the physiologist assessing the ex-posure of the human body to environmental conditions thatresult in thermal imbalance. In theory, the measurement ofbody heat exchange using simultaneous measures of direct and

indirect calorimetry is the only method whereby �Hb can bedirectly determined. Thus the difference between metabolicheat production using the stoichiometric relationship of theproducts and reactants of oxidative metabolism (indirect calo-rimetry) and the total heat lost from the body can be used toestimate �Hb. By definition, �Hb is the product of the changeof the mean temperature of the tissues of the body (�T� b), thetotal body mass (bm), and the average specific heat of thetissues of the body (CP). Hence, for a given body mass of aknown CP, �T� b is an acceptable surrogate measure of �Hb.Because of the limited accessibility of direct calorimeters,thermometry is most often used to estimate �T� b and therebyderive �Hb. The most common thermometry approach is thetwo-compartment model (4) that estimates �T� b by separatingthe body into a “core” compartment temperature measuredusing the change in rectal temperature (�Tre) and a “shell”compartment temperature measured using the change in meanskin temperature (�T� sk). The relative contribution of eachcompartment to �T� b is determined by a sum-to-one ratio ofweighting coefficients that is dependent upon the externalenvironment. Recommended weighting coefficients are vari-able, however, and for a hot environment, the suggested “shell”coefficient ranges from as low as 0.05 (40, 41) to as high as 0.2or 0.3 (5, 11, 20, 26).

Previous research has shown that the two-compartmentalthermometry model greatly underestimates �T� b and therefore�Hb (15, 18, 38, 43). Despite this, the two-compartment modelis still extensively used for the estimation of �T� b and �Hb,both as an analytical tool for assessing individual heat loadstatus for a wide range of subpopulations (1, 7, 36, 49) andextensively as a physiological criterion for maximal heat ex-posure (2, 16, 17, 19, 22).

A major source of error in the prediction of �Hb may beaccounted for by independently considering the thermal influ-ences of muscle tissue using a three-compartment thermometrymodel of “core,” “muscle,” and “shell.” Such a model was firstdescribed by Nadel et al. (28) using partitional calorimetry andintermittent intramuscular temperature measurements taken inthe quadriceps and later by Webb (45) using suit calorimeterybut with no concurrent measurements of muscle temperature.Total body mass is composed of �40% muscle mass compared

Address for reprint requests and other correspondence: O. Jay, Univ. ofOttawa, School of Human Kinetics, 125 Univ., Montpetit Hall, Rm. 367,Ottawa, Ontario, Canada K1N 6N5 (e-mail: [email protected]).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Am J Physiol Regul Integr Comp Physiol 292: R167–R175, 2007.First published August 24, 2006; doi:10.1152/ajpregu.00338.2006.

0363-6119/07 $8.00 Copyright © 2007 the American Physiological Societyhttp://www.ajpregu.org R167

with �5% skin mass (44), and the specific heat capacity ofmuscle tissue is very similar to that of skin (13). Despite this,temperature of the skin is independently considered in thetwo-compartment thermometry model (i.e., “shell”), whereasmuscle temperature is not. Furthermore, the traditional two-compartment thermometric approach arbitrarily includes mus-cle mass as part of the “core.” However, active muscle tissuecan be the primary source of thermogenesis during exercise,and inactive muscle may be a major heat sink during exercise.Therefore, overall muscle tissue temperature is typically sub-ject to a greater change relative to measures of core and skintemperature.

The aim of the present study was to compare the change inbody heat content, as estimated using a two-compartmentthermometry model, and a three-compartment model, incorpo-rating the thermal influences of muscle mass, with those valuesdirectly measured using whole body direct calorimetry. It washypothesized that the difference between the estimates forchange in body heat content by a three-compartment thermom-etry model relative to direct calorimetry, will be less than bythe two-compartment approach.

METHODS

Participants

After approval of the experimental protocol from the University ofOttawa Research Ethics Committee, 52 healthy, nonsmoking normo-tensive participants volunteered (26 males, 26 females) for the study.Of the participants, 22 (10 males, 12 females) were exposed to an airtemperature of 30°C and relative humidity (RH) of 30%; 6 (3 males,3 females) to 30°C, 60% RH; 14 (9 males, 5 females) to 24°C, 30%RH; and 10 (4 males, 6 females) to 24°C, 60% RH. The characteristicsof the participants are given in Table 1.

The body composition of each participant was measured using dualenergy X-ray absorptiometry (DEXA) by which the body mass ispartitioned into fat tissue mass (mf), lean tissue mass (ml), and bonemass (mb). Lean tissue mass (ml) is further subdivided into musclemass (51.0% of ml), skin mass (11.0%), white matter, gray matter,eye, nerve, lens, and cartilage mass (12.9%), blood mass (25.0%), andcerebral spinal fluid mass (0.1%) (12, 37). Using these components(13), we determined the mean average specific heat of the body (CP)(Table 2).

Instrumentation

Esophageal temperature (Tes) was measured by placing a pediatricthermocouple probe of �2 mm in diameter (Mon-a-therm Nasopha-ryngeal Temperature Probe; Mallinckrodt Medical, St. Louis, MO)through the participant’s nostril while they were asked to sip waterthrough a straw. The location of the probe tip in the esophagus wasestimated to be at the T8/T9 level, in proximity to the left ventricleand aorta. This position is based upon the equation of Mekjavic andRempel (27). Rectal temperature (Tre) was measured using a pediatric

thermocouple probe (Mon-a-therm General Purpose TemperatureProbe) inserted to a minimum of 12 cm past the sphincter. Aural canaltemperature (Tau) was measured using a aural canal thermocoupleprobe (Mon-a-therm Tympanic) placed in the aural canal until restingagainst the tympanic membrane (determined by the participant report-ing an audible scratching sound), following which it was withdrawnslightly. The aural canal probe was held in position and isolated fromthe external environment with cotton and ear protectors. Skin temper-ature was measured at 12 points over the body surface using 0.3-mmdiameter T-type (copper/constantan) thermocouples integrated intoheat flow sensors (Concept Engineering, Old Saybrook, CT). Ther-mocouples were attached using surgical tape (Blenderm, 3M, St. Paul,MN). Mean skin temperature (T� sk) was calculated using the 12 skintemperatures weighted to the regional proportions, as determinedby Hardy and DuBois (14): head 7%, hand 4%, upper back 9.5%,chest 9.5%, lower back 9.5%, abdomen 9.5%, bicep 9%, forearm7%, quadriceps 9.5%, hamstring 9.5%, front calf 8.5%, and backcalf 7.5%.

Regional muscle temperature was measured using a flexible intra-muscular temperature probe (Physitemp Instruments, Clifton, NJ,model IT-17:18, type T, time constant of 0.1 s) inserted into the vastuslateralis (Tvl), triceps brachii (Ttb), and upper trapezius (Tut). Using anaseptic technique, we anesthetized the skin, subcutaneous tissue, andmuscle to a maximum depth of 40 mm by infiltrating �3 ml oflidocaine with 2% epinephrine. An 18-gauge, 45-mm nonradiopaqueFEP polymer catheter (Medex Canada, Toronto, ON, Canada) wasthen inserted at an angle and parallel to the long axis of the muscleinto the anesthetized tract to the required depth (�3 cm). The catheterstylet was then withdrawn, and the temperature probe was insertedinto the catheter shaft. The probe assembly, including the cathetershaft, was secured to the skin with sterile, waterproof dressing (23,25). The implant site for the vastus lateralis was approximatelymidway between, and lateral to, a line joining the anterior superioriliac spine and the superior aspect of the centre of the patella (23–25).The triceps brachii muscle temperature probe was inserted approxi-mately midway between, and lateral to, a line joining the greater

Table 1. Mean descriptive characteristics for male and female participants

n Age, yr Weight, kg Height, cm BSA, m2 BMI, kg/m2 VO2peak, ml�kg�1�min�1

Male 26 25.8 (9.0) 78.8 (12.9) 179.0 (6.0) 1.97 (0.17) 24.5 (3.3) 49.5 (7.3)51–18 106.8–60.1 193.0–167.6 2.37–1.69 31.8–19.6 63.1–30.8

Female 26 23.7 (4.2) 63.1 (10.0) 167.3 (6.2) 1.71 (0.14) 22.7 (3.3) 40.5 (8.1)38–18 97.2–41.5 180.0–154.9 2.04–1.35 35.7–17.3 56.7–28.0

Values given are means, standard deviation (in parentheses), and range (maximum–minimum). Body surface area (BSA) is estimated using the equation ofDuBois and DuBois (8). BMIs, body mass indices; VO2peak, volume of peak oxygen consumption.

Table 2. Mean DEXA results for male andfemale participants

nLean Mass,

kgFat Mass,

kgBone Mass,

kgCp,

kJ�kg�1�°C�1

Male 26 61.89 (6.56) 14.96 (8.30) 3.27 (0.46) 3.48 (0.05)74.01–50.48 31.72–3.24 4.02–2.38 3.56–3.39

Female 26 44.66 (6.14) 15.88 (6.26) 2.59 (0.34) 3.43 (0.05)58.81–33.40 38.53–4.73 3.18–1.91 3.53–3.34

Values given are means, standard deviation (in parentheses) and range(maximum–minimum). Specific heat of the human body (Cp) calculated frompartitioning total body mass into lean, fat, and bone; and lean mass subdividedinto muscle mass (51.0%); skin mass (11.0%) combined white matter, graymatter, eye, nerve, lens, and cartilage mass (12.9%); blood mass (25.0%); andcerebral spinal fluid mass (0.1%); and assigning a specific heat for eachcomponent (12, 13, 37). DEXA, dual energy X-ray absorptiometry.

R168 BODY HEAT CONTENT AND DIRECT CALORIMETRY

AJP-Regul Integr Comp Physiol • VOL 292 • JANUARY 2007 • www.ajpregu.org

tubercle of the humerus and the superior aspect of the olecranon of theulna (23, 25). The upper trapezius muscle temperature probe wasinserted 3 cm superior to the center point between the acromionprocess and superior angle of the scapula.

All temperature data were collected using a HP Agilent dataacquisition module (model 3497A) at 15-s intervals. These data weresimultaneously displayed and recorded in spreadsheet format on apersonal computer (IBM ThinkCentre M50) with LabVIEW software(Version 7.0, National Instruments, Austin, TX).

Direct calorimetry. The modified Snellen whole body air calorim-eter was employed for the purpose of measuring whole body changesin evaporative and dry heat loss, yielding an accuracy � 2.3 W for themeasurement for total body heat loss. A full technical description ofthe fundamental principles of the original Snellen calorimeter hasbeen published (39), and a further technical report describing allmodifications and performance characteristics is also available (34).

In summary, the calorimeter incorporates a semirecumbent con-stant load cycle ergometer and is housed within a climatic chamberslightly pressurized (�8.25 mmHg) to nullify potential air leakagethrough the calorimeter walls. Differential air temperature and humid-ity are measured over the calorimeter by sampling the influent andeffluent air. The water content is measured using precision dew pointthermometry (model 373H; RH Systems, Albuquerque, NM), whilethe air temperature is measured using RTD high-precision thermistors(�0.002°C, Black Stack model 1560, Hart Electronics, AmericanFork, UT). Air mass flow through the calorimeter is estimated bydifferential thermometry over a known heat source (2 � 750 Wheating elements) placed in the effluent air stream. Differential tem-perature over the heater is measured using a third aforementionedhigh-precision thermistor placed downstream from the heater. Airmass flow rate (kg air/min) is continuously measured during each trial.Data from the calorimeter were collected continuously at 8-s intervalsthroughout the trials. The real-time data were displayed and recordedon a personal computer (Dell OPTIPLEX GX270) with LabVIEWsoftware (Version 7.0, National Instruments).

Evaporative heat loss per minute was calculated using the follow-ing equation

Evaporative Heat Loss

� Mass flow � (Humidityout � Humidityin) �2.427(1)

where (Humidityout � Humidityin) is the difference in absolutehumidity across the calorimeter (g water �kg air�1), and 2.427 is thelatent heat of vaporization of sweat (kJ �kg sweat�1) (50).

Dry heat loss per minute from radiation, conduction, and convec-tion was calculated using the following equation

Dry Heat Loss

� Mass flow � (Temperatureout � Temperaturein) �1.005(2)

where (Temperatureout � Temperaturein) is the difference in airtemperature across the calorimeter (°C), and 1.005 is the specific heatof air [kJ � (kg air �°C)�1].

Indirect calorimetry. Oxygen consumption (VO2) was measured bythe open circuit technique using expired gas samples drawn from a6-liter fluted mixing box. Expired gas was analyzed using calibratedelectrochemical gas analyzers (AMETEK model S-3A/1 and CD 3A,Applied Electrochemistry, Pittsburgh, PA). Expired air was recycledback into the calorimeter chamber to account for respiratory conduc-tive and evaporative heat loss. Before each session, gas mixtures of4% CO2, 17% O2, balance N2 were used to calibrate the gas analyzersand a 3-liter syringe was used to calibrate the turbine ventilometer.

Metabolic energy expenditure (M) was calculated from minute-average values for VO2 and respiratory exchange ratio using thefollowing equation

M � �VO2 ��RER � 0.7�

0.3ec� � ��1 � RER�

0.3ef�� (3)

where ec is the caloric equivalent per liter of oxygen for the oxidationof carbohydrates (21.13 kJ), and ef is the caloric equivalent per liter ofoxygen for the oxidation of fat (19.62 kJ).

Change in body heat content. Change in body heat content (�Hb)was measured using the temporal summation of metabolic heatproduction by indirect calorimetry and the net evaporative and dryheat exchange of the body with the environment by direct calorimetry.The cumulative change in heat storage over the exercise period wastherefore calculated using the following equation

�Hb � t0

t

�M � �R � C � K� � E � W�dt (4)

where M metabolic rate, (R � C � K) rate of dry heat loss(radiation, convection, and conduction), E rate evaporative heatloss, and W rate of external work being performed.

Experimental Protocol

All participants volunteered for two separate testing sessions. Onthe first day, an incremental cycle ergometer VO2 peak test was per-formed. On the second day, the calorimetry experimental protocol wasperformed. Testing days were separated by a minimum of 72 h. Allcalorimeter trials were performed at the same time of day, with eachparticipant entering the calorimeter at 8:45 AM. Participants wereasked to arrive at the laboratory in a fasted state, consuming no tea,coffee, or food that morning, and also avoiding any major thermalstimuli on their way to the laboratory. Participants were also asked tonot drink alcohol or exercise for 24 h before experimentation.

Following instrumentation, the participant entered the calorimeterregulated to an ambient air temperature of either 24°C or 30°C andeither 30% or 60% relative humidity. The participant, seated in thesemirecumbent position, rested for a 45-min habituation period untila steady-state baseline resting condition was achieved. Subsequently,the participant cycled at 40% of their predetermined VO2 peak for amaximum of 90 min. The exercise duration was such that a steady-state condition defined as a rectal temperature stable within 0.1°C, avariation of less than 3% for metabolic heat production (M � W) andconstant total heat loss (dry heat loss � evaporative heat loss), wouldbe achieved for at least the final 10 min of exercise (31, 46–48).

For all experimentation, clothing insulation was standardized at�0.2 to 0.3 clo [i.e., cotton underwear, shorts, socks, sports bra (forwomen) and athletic shoes].

Statistical Analyses

The data from all participants were pooled and analyzed accordingto ambient air temperature (24 and 30°C). Data were not separatedfurther according to relative humidity due to the confounding effect ofa reduced number of data points upon predictive power, and thetraditional two-compartment thermometry approach employingweighting coefficients based upon air temperature not relative humid-ity (4). This also ensures an optimal statistical validity by attaining awide variation in the calorimetric and thermometric measures betweenparticipants, under each air temperature condition.

Change in body heat content (�Hb) as measured using calorimetrywas solved for mean body temperature (�T� b) using the followingequation

R169BODY HEAT CONTENT AND DIRECT CALORIMETRY


�T� b � �Hb/�bm �Cp� (5)

where �Hb is the change in body heat content by calorimetry (kJ), bm

is total body mass (kg), and CP is specific heat of the human body asestimated using DEXA (in kJ �kg�1 �°C�1).

Two-compartment thermometry model of mean body temperature.The traditional two-compartment thermometry model (4) for meanbody temperature (�T� b) is

�T� b � �X ��Tre� � ��1 � X� � �T� sk� (6)

where �Tre is the change in rectal temperature and �T� sk is the changein mean skin temperature. The value for X is the proportion of thebody representing the body “core.” The value of X may not exceed 1or be less than 0.

Three-compartment thermometry model of mean body temperature.The three-compartment thermometry model (28, 45) for mean bodytemperature (�T� b) is

�T� b � �X1 ��Tcore� � �X2 ��T� sk� � �X3 ��Tmus� (7)

where �Tcore is the change in core temperature represented by eitherrectal (Tre), esophageal (Tes), aural (Tau) temperature or an un-weighted mean of the three measurements (Tc); �T� sk is the changein mean skin temperature; �Tmus is the change in muscle temperaturerepresented by either vastus lateralis (Tvl), upper trapezius (Tut),triceps brachii (Ttb), the unweighted mean of inactive muscle temper-ature (Tinact), or an unweighted mean of all three measurements (Tm).Values for coefficients X1, X2, and X3 may not exceed 1 or be lessthan 0, and the sum of all coefficients must equal 1.

Derivation of optimal two- and three-compartment models formean body temperature. The two- and three-compartment thermom-etry models were individually fit to the mean body temperature dataobtained using calorimetry with the optimization technique of qua-dratic programming. In summary, the quadratic programming prob-lem is to derive coefficient values that minimize a quadratic functionwhile simultaneously satisfying the set of linear constraints (32).These constraints were that individual coefficients within each modelmay not exceed 1, be less than 0, and the sum of all coefficients withineach model must equal 1. Quadratic programming was performedusing the statistical programming language “R” (the open-sourcesoftware R can be downloaded at http://www.r-project.org/).

Goodness-of-fit. To demonstrate and compare the predictive powerof the optimal two- and three-compartment thermometry models for�T� b, the goodness-of-fit was measured for each by simply adaptingthe R2 statistic from linear regression. For n observations and kparameters in a given model, the quadratic programming problemincorporates j equality constraints (in the present case, j1). Let theith response be denoted by yi (for each thermometry model, yi �T� bi),the ith fitted value be denoted by yi, and let the mean response bedenoted by y� . Then the variance of the response about the mean isestimated by SSM [¥i 18n (yi � y�)2]/(n � 1) and the residualvariance, with respect to the quadratic programming model, is esti-mated by SSE [¥ (yi � yi)2]/(n � k � j).

Defined as the proportion of the variance in the response explainedby the model, the R2 statistic is given by the expression [1 �(SSE/SSM)]. As with linear regression, the R2 statistic in a quadraticprogramming model has a maximum value of 1. However, as SSEmay be greater than SSM, R2 may be less than 0. It is possible for SSEto be greater than SSM as the model does not contain a constantintercept. In the event of this, the model is considered biased, that is,a systematic under- or overestimation of the response. For a biasedmodel, the average observed response will actually perform better asa predictor than the model itself. In other words, the variance aboutthe mean (SSM) will be less than the variance about the fitted values(SSE) and R2 will be negative.

As is the case with linear regression, if there are many parametersin the model, it is possible for the R2 statistic to be biased by

overfitting. The adjusted R2 statistic, which takes into account thepossibility of overfitting, is given by the following expression: 1 �{[(n � 1)SSE]/[(n � k � j)SSM]}. When k is large relative to n, theadjusted and unadjusted R2 statistics will be somewhat different, withthe adjusted R2 statistic being lower. With this in mind, the adjustedR2 statistic is reported in the present study.

RESULTS

Thermometry Data

Mean values for mean skin temperature and all of themeasurements of core and regional muscle temperature at eachair temperature condition are given for baseline preexerciserest and across the final 10 min of exercise (Table 3). Thesedata show that during the final 10 min of exercise, rectaltemperature (Tre) was 0.24°C (SD 0.23) and 0.25°C (0.15)higher than esophageal temperature (Tes) and 0.62°C (0.23)and 0.45°C (0.27) higher than aural canal temperature (Tau) at24°C and 30°C, respectively. Active muscle temperature, vas-tus lateralis (Tvl), was higher than the two inactive muscletemperature sites of the triceps brachii (Ttb) and the uppertrapezius (Tut) by 1.23°C (0.98) and 0.53°C (0.99), respec-tively, at 24°C, and by 0.63°C (0.74) and 0.62 (0.87), respec-tively at 30°C. Mean core temperature (Tc), Tes, and Tre washigher than Tvl by 0.29°C (0.53), 0.33°C (0.59), and 0.57°C(0.56), respectively, at 24°C, and by 0.08°C (0.37), 0.07°C(0.42), and 0.32°C (0.39), respectively, at 30°C. Whereas Tau

was 0.03°C (0.23) and 0.14°C (0.36) lower than Tvl at 24° and30°C, respectively.

Comparison of Thermometry with Calorimetry

An example of the minute-by-minute calorimetry data withconcurrent thermometry data is given in Fig. 1. The meandifferences between calorimetry and the two thermometricmodels for the change in mean body temperature (�T� b) andchange in body heat content (�Hb) at 24 and 30°C are detailedin Table 4.

The optimal coefficients for the estimation of �T� b using thetraditional two-compartment thermometry model of “core”represented by Tre and “shell” represented by T� sk for the direct,whole body calorimetry data measured in the present studywere

24 C; �T� b � �0.52�Tre� � �0.48�T� sk�

Adjusted R-squared: 0.07

30 C; �T� b � �0.90�Tre� � �0.10�T� sk�

Adjusted R-squared: � 0.37

The resultant estimation of �Hb using the two-compartmentmodel for �T� b shows a very poor predictive capability at 24°Cand a systematic shortfall compared with calorimetry (as indi-cated by the negative adjusted R2 value) at 30°C. Change inbody heat content (�Hb) is below the line of identity betweenthermometry and calorimetry in 43 of the 52 total participants,and in 26 of the 28 participants at 30°C (Fig. 2A).

Results for the quadratic programming analyses of the three-compartment model of core, muscle, and skin for the estima-tion of �T� b are detailed for 24°C (Table 5) and 30°C (Table 6).At 24°C, it is evident that in all of the models T� sk has aconsistent influence with a weighting coefficient between



�0.20 and 0.30. Models employing Tre as a representation ofthe “core” yield the higher adjusted R-squared statistics, with“muscle” effectively represented by Tvl, Ttb, and Tinact. Incontrast, the three-compartment models at 30°C are minimallyinfluenced by T� sk; however, Tre again provides the best repre-sentation of the “core” relative to Tes, Tau, and Tc. By incor-porating Tre as “core”, Tvl, Ttb, and Tinact again provideeffective representations of the “muscle” compartment. At both24 and 30°C, all models using Tut as “muscle” yielded thelowest adjusted R-squared statistics. The optimal three-com-partment models for the data in the present study were

24 C; �T� b � �0.63�Tre� � �0.24�T� sk� � �0.13�Tvl�

Adjusted R-squared: 0.48

30 C; �T� b � �0.51�Tre� � �0.03�T� sk� � �0.46�Tinact�

Adjusted R-squared: 0.51Tab

le3.

Mea

nva

lues

for

skin

tem

pera

ture

and

all

ofth

em

easu

rem

ents

ofco

rean

dre

gion

alm

uscl

ete

mpe

ratu

re

Cor

eT

empe

ratu

res

Reg

iona

lM

uscl

eT

empe

ratu

res

Skin

T�sk

Tes

Tre

Tau

Tc

Tvl

Ttb

Tut

Tin

act

Tm

Bas

elin

e(2

4°C

)36

.64

(0.3

0)36

.82

(0.2

0)36

.48

(0.3

0)36

.65

(0.2

3)34

.17

(0.7

5)32

.77

(0.8

3)35

.11

(0.6

7)34

.01

(0.6

5)34

.08

(0.6

2)31

.29

(0.7

4)St

eady

-sta

te(2

4°C

)37

.27

(0.2

5)37

.51

(0.2

5)36

.90

(0.3

2)37

.23

(0.2

2)36

.94

(0.6

4)35

.71

(1.1

2)36

.38

(0.7

6)36

.06

(0.8

2)36

.33

(0.6

5)32

.36

(1.0

2)D

elta

(�)

(24°

C)

0.63

(0.2

3)0.

69(0

.23)

0.42

(0.1

8)0.

58(0

.22)

2.78

(0.7

9)2.

94(0

.88)

1.27

(0.4

2)2.

04(0

.64)

2.25

(0.6

2)1.

08(0

.61)

Bas

elin

e(3

0°C

)36

.81

(0.2

3)36

.98

(0.2

3)36

.74

(0.3

5)36

.84

(0.2

3)34

.97

(0.5

9)34

.58

(0.8

3)35

.78

(0.8

4)35

.18

(0.7

6)35

.11

(0.5

8)33

.20

(0.7

7)St

eady

-sta

te(3

0°C

)37

.43

(0.3

9)37

.68

(0.3

7)37

.22

(0.4

1)37

.44

(0.3

6)37

.36

(0.3

9)36

.74

(0.8

3)36

.74

(0.8

7)36

.74

(0.7

5)36

.95

(0.5

6)33

.75

(0.8

8)D

elta

(�)

(30°

C)

0.62

(0.3

1)0.

70(0

.31)

0.48

(0.3

3)0.

60(0

.30)

2.40

(0.4

9)2.

16(0

.65)

0.95

(0.3

7)1.

56(0

.46)

1.84

(0.3

8)0.

55(0

.46)

All

valu

esar

em

eans

,st

anda

rdde

viat

ion

(in

pare

nthe

ses)

for

base

line

pree

xerc

ise

rest

,fin

al10

-min

ofex

erci

se(s

tead

yst

ate)

and

chan

ges

betw

een

stea

dyst

ate

and

base

line

(Del

ta�

).D

ata

are

pres

ente

dfo

rn

24

at24

°Can

dn

28

at30

°C.C

ore

tem

pera

ture

mea

sure

din

the

esop

hagu

s(T

es),

rect

um(T

re),

aura

lcan

al(T

au),

and

the

unw

eigh

ted

mea

nof

allt

hree

mea

sure

men

ts(T

c).R

egio

nalm

uscl

ete

mpe

ratu

rem

easu

red

inth

eva

stus

late

ralis

(Tvl),

tric

eps

brac

hii

(Ttb

),up

per

trap

eziu

s(T

ut),

anun

wei

ghte

dm

ean

ofin

activ

e(T

lban

dT

ut)

mus

cle

tem

pera

ture

(Tin

act),

and

anun

wei

ghte

dm

ean

all

mus

cle

tem

pera

ture

mea

sure

men

ts(T

m).

Skin

tem

pera

ture

(T�sk

)m

easu

red

asa

wei

ghte

dm

ean

of12

site

s(1

4).

Fig. 1. Example of the minute-by-minute whole body direct calorimetry datafor total heat gain and heat loss (top) with concurrent thermometry data(bottom) during exercise at 40% VO2 peak.



The resultant estimation of �Hb using the optimal three-compartment thermometry model for �T� b shows an unbiasedrelationship with calorimetry at both 24 and 30°C (Fig. 2B).

DISCUSSION

The main findings from this study show that the traditionaltwo-compartment thermometry model of “core” and “shell”underestimates changes in body heat content (�Hb) duringmoderate-intensity, steady-state exercise by between 15 and35%. Upon investigating change in mean body temperature(�T� b) within the two-compartment thermometry model, therewas a systematic underestimation of �T� b relative to calorim-etry. At 30°C, the adjusted R-squared statistic for the two-compartment model was negative (�0.37), indicating a bias,that is, simply using the group mean value for �T� b of 1.11°Cmeasured using calorimetry provides a better estimation thanthe two-compartment model. The three-compartment ther-mometry model of core, muscle, and skin at both 24 and 30°Cwas unbiased and was found to consistently yield a moreprecise estimate of �T� b and therefore �Hb than the two-compartment model.

The notion of a three-compartment thermometry model forthe improved estimation of �Hb has been supported for sometime. However, the present study, with the exception of Snellen(38), is the first to use whole body direct air calorimetry toassess such a concept. Nadel et al. (28) developed the follow-ing theoretical three-compartment model for the calculation ofabsolute mean body temperature (T� b) during positive andnegative work on a semirecumbent bicycle ergometer, byestimating the mass of working muscles from the data ofStolwijk and Hardy (42)

T� b � 0.67Tes � 0.23Tm � 0.10T� sk

where Tes is esophageal temperature; Tm is quadriceps temper-ature, and T� sk is mean skin temperature.

Webb (45) proposed an alternative three-compartment ther-mometry model for the estimation of change in mean bodytemperature (�T� b) during level and uphill walking with a suitcalorimeter. While muscle temperature was not measured, theestimated weighting coefficients for muscle suggested substan-tial heat storage during exercise

�T� b � 0.5�Tre � 0.4�Tm � 0.1�T� sk

where �Tre is the change in rectal temperature, �Tm is the

change in mean muscle temperature; and �T� sk is change inmean skin temperature.

The three-compartment models derived in the present studysuggest that the role of skin temperature in the estimation of�T� b becomes progressively less with increasing ambient airtemperature as is the case with the traditional two-compartmentapproach (4). As such, the model derived at 30°C is similar tothat proposed by both Webb (45) and Nadel et al. (28).Furthermore, the “core” weighting coefficient derived byWebb (45) is almost identical to the 30°C model in our study,while the weighting coefficient derived by Nadel et al. (28) iscloser to that of the 24°C model; however, Tre provided thebest representation of the “core” compartment at both 24 and30°C. The “muscle” compartment weighting coefficient of0.46 at 30°C is very similar to that proposed previously byWebb (45), but the coefficient of 0.13 at 24°C is lower thanpreviously suggested with the body “shell” (i.e., �T� sk) havinga more prominent effect.

Whole body, direct air calorimetry was used by Snellen (38)to develop an improved estimation of �Hb and therefore �T� b.This was a unique study that incorporated multiple tissuemeasurements, including the estimation of subcutaneous tem-perature, for individuals exposed to ambient conditions from12.3 to 35.0°C. However, there were considerable limitationsof the methodology, in that only seven young male subjectswere tested and subcutaneous temperatures were not directlymeasured, but estimated, using zero-flux heat devices, whichhave been since demonstrated not to give a reliable estimate ofmuscle temperature (3). Furthermore, the multiple linear re-gression method used for deriving the improved estimatingequation for �T� b appears statistically unstable due to thenumber of variables introduced for such a small sample size,potential collinearity between variables, and the use of anonintercept design, giving disproportionate R-squared statis-tics. In the case of the present study, a large subject group of52 was used, ranging in age and physical characteristics; activeand inactive muscle temperatures were directly measured usingintramuscular probes; and the analytical techniques for deriv-ing an improved estimation of �T� b were meticulously consid-ered, so that collinearity between variables was avoided andfallacious R-squared statistics were not attained.

Mean body temperature is defined as the average tempera-ture of the tissues of the body (4). In the two-compartmentthermometry model, change in mean skin temperature is inde-

Table 4. Comparison of whole body, direct calorimetry, and traditional thermometry

Calorimetry

Thermometry

Two-compartmentThree-

compartment

Change in body heat content in kJ, �Hb 24°C 260.7 (94.1) 220.0 (108.1) 259.6 (109.1)30°C 258.7 (95.7) 161.7 (61.1) 261.6 (82.6)

Change in mean body temperature in °C, �T� b 24°C 1.06 (0.36) 0.87 (0.35) 1.04 (0.34)30°C 1.11 (0.47) 0.68 (0.30) 1.09 (0.35)

%Difference between thermometry and calorimetry for �Hb 24°C — �15.5% (31.3) �1.1% (29.5)30°C — �35.5% (21.9) �5.4% (30.0)

Values for calorimetry obtained using direct whole body calorimetry. Values for thermometry obtained using �Hb �T� b � bm � Cp, where bm is total bodymass (kg), Cp is specific heat of the human body (kJ �kg�1 �°C�1), and �T� b is change in mean body temperature. Optimal two-compartment models: �T� b (0.52 � �Tre) � (0.48 � �T� sk) for 24°C, and �T� b (0.90 � �Tre) � (0.10 � �T� sk) for 30°C; Optimal three-compartment models: �T� b (0.63 � �Tre) � (0.24 � �T� sk) �(0.13 � �Tvl) for 24°C and �T� b (0.51 � �Tre) � (0.03 � �T� sk) � (0.46 � �Tinact) for 30°C, where �Tre is change in rectal temperature, �T� sk is change in mean skintemperature, �Tvl is change in vastus lateralis temperature and �Tinact is change the unweighted mean of the two inactive muscle sites.



pendently included in the estimation of the change mean bodytemperature. However, in a typical person, �40% of total bodymass is muscle mass, where as only �5% is skin mass (44).The specific heat of muscle (3.639 kJ �kg�1 �°C�1) is verysimilar to that of skin (3.662 kJ �kg�1 �°C�1), and in the presentstudy, change in mean muscle temperature was 2.08 and 3.35times greater than the change in mean skin temperature at 24°Cand 30°C, respectively. It therefore seems logical to considerchanges in muscle temperature when estimating changes inmean body temperature. It has previously been viewed in thetwo-compartment thermometry model that the “muscle” com-

partment is enclosed in the “core” compartment once musclebecomes well perfused with blood during exercise. However,the data from the present study show that the change in bothactive and inactive muscle temperature is poorly representedby any measure of core temperature. After 90-min of exercise,change in Tvl, Ttb, and Tut were 4.0, 4.3, and 1.8 times greaterthat of Tre at 24°C; and 3.4, 3.1, and 1.4 times greater than Tre

at 30°C.

Fig. 2. A comparison between change in body heat content measured by wholebody and direct calorimetry and estimated by thermometry using the optimaltwo-compartment model (A) and the optimal 3-compartment model (B).Triangle (‚) denotes 24°C; Circle (E) denotes 30°C. Solid line indicates theline of identity (y x).

Table 5. Results for quadratic fitting of three-compartmentmodel of core (�Tcore), skin (�T� sk)and muscle (�Tmus) for 24°C

Measures Optimal CoefficientsAdjusted

R-Squared�Tcore �Tmus X1 (�Tcore) X2 (�T� sk) X3 (�Tmus)

Tc Tm 0.58 0.19 0.23 0.48Tes Tvl 0.61 0.23 0.16 0.37

Ttb 0.62 0.24 0.14 0.33Tut 0.31 0.30 0.40 0.22Tinact 0.54 0.25 0.22 0.32

Tre Tvl 0.63 0.24 0.13 0.48Ttb 0.64 0.24 0.12 0.47Tut 0.37 0.29 0.34 0.32Tinact 0.57 0.24 0.19 0.45

Tau Tvl 0.58 0.20 0.22 0.44Ttb 0.59 0.21 0.20 0.38Tut 0.24 0.30 0.46 0.20Tinact 0.49 0.22 0.29 0.36

Data obtained for the model of �T� b (X1 � �Tcore) � (X2 � �T� sk) �(X3 � �Tmus). The term �Tcore is represented by esophageal (Tes), rectal (Tre),aural canal temperature (Tau), or the unweighted mean of all three coretemperature measures (Tc). The term �T� sk is represented by mean skintemperature. The term �Tmus is represented by vastus lateralis (Tvl), tricepsbrachii (Ttb), upper trapezius (Tut) temperature, the unweighted mean of allthree measurements (Tm) or the unweighted mean of the two inactive musclesites (Tinact). Model constraints are the sum of X1, X2 and X3 must equal 1, andall coefficients must be between 0 and 1.

Table 6. Results for quadratic fitting of three-compartmentmodel of core (�Tcore), skin (�T� sk),and muscle (�Tmus) for 30°C

Measures Optimal CoefficientsAdjusted

R-Squared�Tcore �Tmus X1 (�Tcore) X2 (�T� sk) X3 (�Tmus)

Tc Tm 0.49 0.10 0.41 0.45Tes Tvl 0.51 0.23 0.27 0.25

Ttb 0.48 0.20 0.31 0.31Tut 0.00 0.00 1.00 0.17Tinact 0.36 0.11 0.52 0.40

Tre Tvl 0.65 0.12 0.23 0.42Ttb 0.63 0.10 0.27 0.48Tut 0.13 0.00 0.87 0.19Tinact 0.51 0.03 0.46 0.51

Tau Tvl 0.51 0.19 0.30 0.32Ttb 0.49 0.17 0.35 0.39Tut 0.00 0.00 1.00 0.17Tinact 0.37 0.08 0.56 0.45

Data obtained for the model of �T� b (X1 � �Tcore) � (X2 � �T� sk) �(X3 � �Tmus). The term �Tcore is represented by esophageal (Tes), rectal (Tre),aural canal temperature (Tau), or the unweighted mean of all three coretemperature measures (Tc). The term �T� sk is represented by mean skintemperature. The term �Tmus is represented by vastus lateralis (Tvl), tricepsbrachii (Ttb), upper trapezius (Tut) temperature, the unweighted mean of allthree measurements (Tm) or the unweighted mean of the two inactive musclesites (Tinact). Model constraints are the sum of X1, X2, and X3 must equal 1 andall coefficients must be between 0 and 1.



The findings of the present study suggest that the source oferror observed with the two-compartment thermometry modelfor the estimation of �T� b and therefore �Hb is an underesti-mation of the tissue temperature transients of the body duringexercise. The three-compartment model by no means reflectsthe level of complexity of the thermal interactions betweenvarious tissues of the body; however, the inclusion of a “mus-cle” compartment does give a degree of representation to theconsiderable influences of muscle heat load upon body heatcontent. Indeed, regional muscle temperature at any point intime is the result of regional differences in metabolic rate,conductive heat loss to adjacent tissue, and deep and peripheralconvective blood flow (9). Furthermore, the convective transferof muscle heat load to cooler tissues in the body has beendemonstrated to significantly prolong the elevation of coretemperature and presumably body heat content after exercise,with hyperemic previously active musculature considered tohave the most profound influence (23, 25).

When using a “muscle” compartment for the estimation of�T� b, the present findings suggest a minimal predictive role ofthe “shell” compartment at warmer ambient temperatures. Inaddition to the fact that total body mass is composed of arelatively small proportion of skin mass, the contribution ofmean skin temperature to the estimation of �T� b is furtherconfounded by several factors. Skin temperature is stronglyinfluenced by skin blood flow, which itself can be significantlymodified independently of whole body thermal state. Forexample, varying levels of exercise intensity result in differentskin-to-muscle perfusion ratios, with increasing blood flowshunted away from skin to working muscle groups with greaterlevels of exercise (23). Furthermore, factors such as hydrationstatus (29), training status (30), level of acclimatization (10),and the administration of topically applied medications, suchas corticosteroids and nicotinates (21), have been demonstratedto alter skin blood flow during exercise. The use of mean skintemperature as an estimate of the thermal status of the skin isitself also subject to potential error. Measurement methodsrange from as few as four sites (33) to as many as 12 sites (14),and as mean skin temperature is in reality an interface temper-ature between the body surface with the external environment,factors such as clothing and environmental conditions will alsohave an influence.

Similarly, as with the two-compartment thermometry model,the data from the present study indicated that Tre was the“core” measurement that best associated with �T� b within thethree-compartment thermometry model. Models using Tes asan indicator of “core” temperature provided the worst associ-ation with �T� b. This is thought to be a consequence of Tes

generally representing the central arterial blood temperature(6). Although a response lag is inevitable with Tre, it issuggested that during steady-state exercise Tre provides a betterrepresentation of equilibrated tissue temperatures of the deepvisceral/splanchnic region. The use of Tau also provided anacceptable means by which changes in core temperature couldbe represented in the three-compartment thermometry model;however, Tau was consistently lower than both Tre and Tes,possibly due to insufficient insulation of the probe in the auralcanal from the air within the calorimeter and are thereforethought to be less reliable.

The present study does have limitations in terms of the rangeof ambient conditions tested and the employment of only one

workload. Indeed core and active muscle temperature has beendemonstrated to be dependent upon relative workload (35);therefore, thermometry models for �Hb may also differ acrossthe range of workloads that steady-state exercise is possible.Furthermore, the employment of the three-compartment ther-mometry model does require the direct measurement of intra-muscular temperature; however, in many cases, such an inva-sive technique may not be possible. The development of anaccurate noninvasive method of estimating muscle temperatureusing novel zero-heat-flux methods is ongoing (3, 51).

Despite the inclusion of a “muscle” compartment yielding animproved estimation of �T� b relative to the traditional two-compartment thermometry model, the optimal models onlyexplained 48% and 51% of the variation found in �T� b at 24°Cand 30°C, respectively. Increasing the number of tissue tem-perature measurements intermediate to the “core” and “shell”may further improve the estimation of �T� b somewhat; how-ever, the determination of the primary sources of individualvariability in body heat content is of paramount importance forits more precise estimation. Further research must therefore beconducted to elucidate the relative effects of factors suchas adiposity, age, gender, physical fitness, and acclimationupon �Hb.

In conclusion, whole body direct air calorimetry shows thata two-compartment thermometry model of “core” and “shell”for the derivation of �T� b underestimates �Hb by between 15and 35% under the conditions tested in this study. A three-compartment thermometry model independently removed thestatistical bias seen with the two-compartment model, includ-ing the thermal influences of “muscle,” and consistentlyyielded a more precise estimate of �T� b and therefore �Hb.

GRANTS

This research was supported by the U.S. Army Medical Research andMaterial Command’s Office of the Congressionally Directed Medical ResearchPrograms and Natural Sciences and Engineering Research Council (to Dr. G.Kenny).

REFERENCES

1. Arngrimsson SA, Petitt DS, Stueck MG, Jorgensen DK, and CuretonKJ. Cooling vest worn during active warm-up improves 5-km run perfor-mance in the heat. J Appl Physiol 96: 1867–1874, 2004.

2. Blockley WV, Mc CJ, Lyman J, and Taylor CL. Human tolerance forhigh temperature aircraft environments. J Aviat Med 25: 515–522, 1954.

3. Brajkovic D and Ducharme MB. Confounding factors in the use of thezero-heat-flow method for non-invasive muscle temperature measurement.Eur J Appl Physiol 94: 386–391, 2005.

4. Burton AC. The average temperature of the tissues of the body. J Nutr 9:261–280, 1935.

5. Colin J, Timbal J, and Boutelier C. Experimental determination of theequation permitting the calculation of the mean body temperature inneutral and hot environment. J Physiol 63: 229–231, 1971.

6. Cooper KE and Kenyon JR. A comparison of temperatures measured inthe rectum, oesophagus, and on the surface of the aorta during hypother-mia in man. Br J Surg 44: 616–619, 1957.

7. Cotter JD, Sleivert GG, Roberts WS, and Febbraio MA. Effect ofpre-cooling, with and without thigh cooling, on strain and enduranceexercise performance in the heat. Comp Biochem Physiol A 128: 667–677,2001.

8. DuBois D and DuBois EF. A formula to estimate surface area if heightand weight are known. Arch Intern Med 17: 863, 1916.

9. Ducharme MB and Tikuisis P. Role of blood as heat source or sink inhuman limbs during local cooling and heating. J Appl Physiol 76: 2084–2094, 1994.

10. Fox RH, Goldsmith R, Kidd DJ, and Lewis HE. Blood flow and otherthermoregulatory changes with acclimatization to heat. J Physiol 166:548–562, 1963.



11. Gagge AP, Fobelets AP, and Berglunda LG. A standard predictiveindex of human response to the thermal environment. ASHRAE Trans 92:709–731, 1986.

12. Gandhi OP, Lazzi G, and Furse CS. Electromagnetic absorption in thehuman head and neck for mobile telephones at 835 and 1900 MHz. IEEETrans Microwav Theor Techn 44: 1884–1897, 1996.

13. Geddes LA and Baker LE. The specific resistance of biological ma-terial—a compendium of data for the biomedical engineer and physiolo-gist. Med Biol Eng 5: 271–293, 1967.

14. Hardy JD and DuBois EF. The technique of measuring radiation andconvection. J Nutr 461–475, 1938.

15. Hardy JD and Stolwijk JA. Partitional calorimetric studies of manduring exposures to thermal transients. J Appl Physiol 21: 1799–1806,1966.

16. Henane R and Bittel J. Changes of thermal balance induced by passiveheating in resting man. J Appl Physiol 38: 294–299, 1975.

17. Holmer I. Body cooling with ice for warm-water diving operations.Undersea Biomed Res 16: 471–479, 1989.

18. Horstman DH and Horvath SM. Cardiovascular and temperature regu-latory changes during progressive dehydration and euhydration. J ApplPhysiol 33: 446–450, 1972.

19. Iampietro PF and Goldmen RF. Tolerance of man working in hot,humid environments. J Appl Physiol 20: 73–76, 1965.

20. Interantional Standards Organization. ISO 7933 - Hot Environments:Analytical Determination and Interpretation of Thermal Stress UsingCalculation of Required Sweat Rate. Geneva: International StandardsOrganization, 1989.

21. Jacobi U, Kaiser M, Koscielny J, Schutz R, Meinke M, Sterry W, andLademann J. Comparison of blood flow to the cutaneous temperature andredness after topical application of benzyl nicotinate [Online]. J BiomedOpt 11: 014025, 2006.

22. Kaufman WC. Human tolerance limits for some thermal environments ofaerospace. Aerospace Med 34: 889–896, 1963.

23. Kenny GP, Jay O, Zaleski W, Reardon ML, Sigal RJ, Journeay WS,and Reardon FD. Postexercise hypotension causes a prolonged pertur-bation in esophageal and active muscle temperature recovery. Am JPhysiol Regul Integr Comp Physiol 291: R580–R588, 2006.

24. Kenny GP, Reardon FD, Ducharme MB, Reardon ML, and ZaleskiW. Tissue temperature transients in resting contra-lateral leg muscle tissueduring isolated knee extension. Can J Appl Physiol 27: 535–550, 2002.

25. Kenny GP, Reardon FD, Zaleski W, Reardon ML, Haman F, andDucharme MB. Muscle temperature transients before, during, and afterexercise measured using an intramuscular multisensor probe. J ApplPhysiol 94: 2350–2357, 2003.

26. Malchaire J, Kampmann B, Havenith G, Mehnert P, and GebhardtHJ. Criteria for estimating acceptable exposure times in hot workingenvironments: a review. Int Arch Occup Environ Health 73: 215–220,2000.

27. Mekjavic IB and Rempel ME. Determination of esophageal probeinsertion length based on standing and sitting height. J Appl Physiol 69:376–379, 1990.

28. Nadel ER, Bergh U, and Saltin B. Body temperatures during negativework exercise. J Appl Physiol 33: 553–558, 1972.

29. Nadel ER, Fortney SM, and Wenger CB. Effect of hydration state ofcirculatory and thermal regulations. J Appl Physiol 49: 715–721, 1980.

30. Nadel ER, Pandolf KB, Roberts MF, and Stolwijk JA. Mechanisms ofthermal acclimation to exercise and heat. J Appl Physiol 37: 515–520,1974.

31. Nagle FJ, Webb P, and Wanta DM. Energy exchange in downhill anduphill walking: a calorimetric study. Med Sci Sports Exerc 22: 540–544,1990.

32. Nocedal J, and Wright SJ. Numerical Optimization. New York:Springer, 1999.

33. Ramanathan NL. A new weighting system for mean surface temperatureof the human body. J Appl Physiol 19: 531–533, 1964.

34. Reardon FD, Leppik LE, Wegmann R, Webb P, Ducharme MB, andKenny GP. The Snellen human calorimeter revisited, re-engineered andupgraded: Design and performance characteristics. Med Biol Eng Comput44: 721–728, 2006.

35. Saltin B and Hermansen L. Esophageal, rectal, and muscle temperatureduring exercise. J Appl Physiol 21: 1757–1762, 1966.

36. Saunders AG, Dugas JP, Tucker R, Lambert MI, and Noakes TD. Theeffects of different air velocities on heat storage and body temperature inhumans cycling in a hot, humid environment. Acta Physiol Scand 183:241–255, 2005.

37. Simunic D, Wach P, Renhart W, and Stollberger R. Spatial distributionof high-frequency electromagnetic energy in human head during MRI:numerical results and measurements. IEEE Trans Biomed Eng 43: 88–94,1996.

38. Snellen JW. An improved estimation of mean body temperature usingcombined direct calorimetry and thermometry. Eur J Appl Physiol 82:188–196, 2000.

39. Snellen JW, Chang KS, and Smith W. Technical description andperformance characteristics of a human whole-body calorimeter. Med BiolEng Comput 21: 9–20, 1983.

40. Stolwijk JA. A Mathematical Model of Physiological Temperature Reg-ulation in Man. New Haven, CT: J.B. Pierce Laboratory, 1971. (Report ofNASA contractor, CR 1855)

41. Stolwijk JA and Hardy JD. Control of body temperature. In: Handbookof Physiology. Reactions to Environmental Agents. Reaction to Environ-mental Agents. Bethesda, MD: Am. Physiol. Soc., 1977, sect 9, chapt. 4,p. 45–68.

42. Stolwijk JA and Hardy JD. Temperature regulation in man—a theoret-ical study. Pflugers Arch 291: 129–162, 1966.

43. Vallerand AL, Savourey G, Hanniquet AM, and Bittel JH. How shouldbody heat storage be determined in humans: by thermometry or calorim-etry? Eur J Appl Physiol Occup Physiol 65: 286–294, 1992.

44. Wang ZM, Pierson RN Jr, and Heymsfield SB. The five-level model: anew approach to organizing body-composition research. Am J Clin Nutr56: 19–28, 1992.

45. Webb P. Heat storage during exercise, especially in muscle. In: 8thInternational Conference on Environmental Ergonomics, edited by Hodg-don JA, San Diego, CA: National Health Research Center, 1998, p.121–124.

46. Webb P. The physiology of heat regulation. Am J Physiol Regul IntegrComp Physiol 268: R838–R850, 1995.

47. Webb P, Nagle FJ, and Wanta DM. Heat regulation during exercise withcontrolled cooling. Eur J Appl Physiol Occup Physiol 62: 193–197, 1991.

48. Webb P, Saris WH, Schoffelen PF, Van Ingen Schenau GJ, and TenHoor F. The work of walking: a calorimetric study. Med Sci Sports Exerc20: 331–337, 1988.

49. Webborn N, Price MJ, Castle PC, and Goosey-Tolfrey VL. Effects oftwo cooling strategies on thermoregulatory responses of tetraplegic ath-letes during repeated intermittent exercise in the heat. J Appl Physiol 98:2101–2107, 2005.

50. Wenger CB. Heat of evaporation of sweat: thermodynamic consider-ations. J Appl Physiol 32: 456–459, 1972.

51. Yamakage M, Iwasaki S, and Namiki A. Evaluation of a newly devel-oped monitor of deep body temperature. J Anesthesiol 16: 354–357, 2002.



Documents

fisiologi 1