Physiology and Biochemistry
Thermoregulatory and Physiological Responses of
Wheelchair Athletes to Prolonged Arm Crank
and Wheelchair Exercise
M. J. Price1, I. G. Campbell 2
1
School of Chemical and Life Sciences, University of Greenwich, Wellington Street, Woolwich, London, England
2
School of Sport Studies and Sports Sciences, Edge Hill University College, Omskirk, Lancashire, England
Price, MJ, Campbell, IG. Thermoregulatory and Physiological
Responses of Wheelchair Athletes to Prolonged Arm Crank
and Wheelchair Exercise. Int J Sports Med 1999; 20: 457 – 463
Accepted after revision: May 25, 1999
■■■■ Seven wheelchair athletes participated in this study. On
separate occasions all athletes performed 60 min of arm crank
ergometry and wheelchair ergometry at 60 % of the ergometer
specific V̇O2peak in cool conditions (21.5 ± 1.3 8C; 54.2 ± 6.3 % re-
lative humidity, 21.2 ± 1.9 8C; 55.5 ± 11.9 % relative humidity,
respectively). The order of testing was randomised. Aural and
skin temperatures were continually measured throughout the
60 min test. Expired air was collected at 5, 15, 30, 45, and
60 min during the exercise period. Oxygen consumption was
similar for both trials (1.09 ± 0.21 and 1.16 ± 0.33 l × min–1, for
the ACE and WCE trials, respectively). Heat storage was calculat-
ed at these time-points. Aural temperature was elevated from
rest between 25 to 45 min of wheelchair ergometry
(0.5 ± 0.3 8C; P < 0.05) when compared to between 20 min of ex-
ercise and 5 min of recovery (0.6 ± 0.3 8C; P < 0.05) during the
arm crank ergometry trial. On the cessation of arm crank ergo-
metry, heat storage was elevated above values observed at
5 min of exercise (P < 0.05). On the cessation of wheelchair ergo-
metry, heat storage was not elevated above values at 5 minutes
of exercise. Upper arm skin temperature was cooler during
wheelchair ergometry when compared to arm crank ergometry
(P < 0.05). All other skin temperature responses were similar
during both exercise modes. The efficiency of arm crank ergo-
metry was greater than wheelchair ergometry throughout the
exercise period (18.5 ± 3.5 % and 8.9 ± 3.7 % at 60 minutes of ex-
ercise, respectively; P < 0.05). The results of this study suggest
that although ACE demonstrates greater efficiency than WCE
prolonged arm crank ergometry elicited greater thermal and
physiological strain when compared with prolonged wheelchair
ergometry. The lower thermal strain during WCE was suggested
to be related to the propulsion biomechanics which may result
in some degree of local cooling, and consequently heat dissipa-
tion, when compared to ACE. Due to the greater thermal strain
Int J Sports Med 1999; 20: 457 – 463
© Georg Thieme Verlag Stuttgart · New York
ISSN 0172-4622
457
during arm crank ergometry, it is recommended that for studies
examining the exercise responses of wheelchair users wheel-
chair ergometry should be employed.
Downloaded by: University of British Columbia. Copyrighted material.
■ Key words: Arm crank ergometry, wheelchair ergometry,
aural temperature, skin temperature, heat storage, efficiency
of exercise.
Introduction
The thermoregulatory responses of wheelchair athletes during
prolonged arm crank ergometry (ACE) have been examined
[6,16,18]. Whilst this research has provided important infor-
mation regarding thermoregulatory responses, wheelchair
athletes regularly perform wheelchair exercise. Despite this
little is known about the thermoregulatory responses of these
athletes during their habitual exercise mode. One problem of
using wheelchair ergometry (WCE) in such studies is that,
due to significant physiological and biomechanical differences
that have been reported to exist between able-bodied subjects
and wheelchair users during WCE [2], it is difficult to employ
an appropriate able-bodied control group. A comparison of the
thermoregulatory responses during prolonged ACE and WCE
would consequently enable such responses of wheelchair ath-
letes to be compared to those of an appropriate able-bodied
control group. This would allow differences between the two
modes of exercise to be incorporated into comparisons of ther-
moregulatory responses of wheelchair athletes during WCE
and able-bodied athletes during ACE.
The efficiency of WCE is consistently reported to be lower than
conventional exercise modes [2, 26 – 28]. This may result in
greater increases in body temperature and a greater thermal
strain for athletes undertaking WCE when compared to more
conventional exercise modes such as ACE. However, compari-
sons of ACE and WCE have previously been concerned with
maximal exercise responses [11,13, 21]. Although two studies
have examined the physiological responses during short term,
submaximal ACE and WCE exercise bouts [23, 24], no studies
have compared the thermoregulatory or physiological respon-
ses of wheelchair athletes during prolonged ACE and WCE.
Therefore, the aim of this study was to compare the thermore-
gulatory and physiological responses to prolonged arm crank
and wheelchair ergometry in a group of wheelchair athletes.
458 Int J Sports Med 1999; 20
Subjects and Methods
Subjects
Seven paraplegic wheelchair athletes (PA; T3/T4 – L1) volun-
teered to participate in this study which had received ethical
committee approval. More specifically, five athletes were
wheelchair racers and two were basketball players. All trained
and competed regularly at national and international level.
The mean (± SD) age, body mass, and sum of four skinfolds
were 29.3 ± 5.9 yrs, 64.3 ± 1.7 kg, and 52.5 ± 21.6 mm, respec-
tively. Athletes visited the laboratory on three separate occa-
sions and were fully familiar with testing procedures. All sub-
jects gave written informed consent.
Preliminary tests
On the first visit to the laboratory athletes undertook a sub-
maximal exercise test and an incremental exercise test for
peak oxygen uptake (V̇O2peak) for both ACE and WCE. The or-
der of testing was randomised with at least 3 hours rest be-
tween exercise modes. All submaximal exercise stages were
four minutes in duration separated by two minute recovery
periods to minimise local fatigue [17, 23]. Expired air was col-
lected during the final minute of each stage and analysed for
oxygen (V̇O2) and carbon dioxide content (V̇CO2). For the ACE
test, exercise stages were performed at work loads of 35, 70,
105, and 140 W on a Monark cycle ergometer (Model 814E, Var-
berg, Sweden) adapted for upper body exercise. On cessation
of the fourth work load athletes rested for at least 5 minutes
or until heart rate had returned to below 100 bts × min–1. Sub-
jects then performed a test for V̇O2 peak [19]. Expired air was
collected during the final minute of the V̇O2 peak test. A 20 µl
blood lactate sample was obtained from the subjects' earlobe
for the determination of blood lactate (BLa; YSI 2000 Sport,
Yellow Springs Instruments, Yellow Springs, USA) at volitional
exhaustion. The WCE trial involved athletes performing sub-
maximal exercise stages on a wheelchair ergometer (Bromking
Turbo Trainer, Loughborough, England) in their own specific
sports chair. Athletes exercising in track wheelchairs per-
formed exercise stages at work loads of 30, 40, 50, and 60 W.
Subjects then undertook an incremental test for V̇O2 peak. This
involved subjects propelling their wheelchairs at an increment
of 5 W × min–1 from an initial work load of 30 W until volitional
exhaustion. As the basketball players were unable to push
their sports chairs at these power outputs, a protocol employ-
ing push rate, developed by Wicks et al. [30], was undertaken.
This involved submaximal exercise stages of 30, 40, 50, and 60
pushes per minute with the test for V̇O2 peak involving in-
creases in push rate of 10 pushes per minute until volitional
exhaustion.
Prolonged exercise tests
On subsequent visits to the laboratory all athletes performed
60 minutes of ACE and WCE at a work load set to elicit 60 % of
the ergometer specific V̇O2 peak. This work load was deter-
mined from the results of the preliminary exercise tests. Both
the prolonged ACE and WCE trials were performed in a cool en-
vironment (21.5 ± 1.3 8C, 54.2 ± 6.3 % relative humidity, and
21.2 ± 1.9 8C and 55.5 ± 11.9 % relative humidity for the ACE
and WCE trials, respectively). The order of testing was rando-
mised.
M. J. Price, I. G. Campbell
On arrival at the laboratory for the 60 minute test subjects res-
ted quietly for 15 minutes. At the end of this period resting
heart rate (HR) was recorded (Polar Sports Tester PE4000,
Kempele, Finland) and a resting expired air sample was obtain-
ed. A small 20 µl capillary blood sample was obtained from the
antecubital vein from which haemoglobin (Clandon HemoCue,
HemoCue Ltd. Sheffield, England) and haematocrit (Hawsksley
Reader, Hawsksley & Sons, Sussex, England) were subsequent-
ly analysed to determine plasma volume [7]. Body mass was
then recorded (Seca 710, seated scales, Hamburg, Germany)
after individuals had evacuated their bladders. Skinfold meas-
urements were taken from the biceps, triceps, subscapular and
suprailiac sites using Harpenden skinfold callipers (British In-
dicators Ltd, Luton, England), in accordance with the proce-
dures of Durnin and Wormersley [8]. Subjects wore light-
weight tracksuit trousers, socks, and training shoes during
both trials.
Thermistors were positioned for measures of aural and skin
temperatures. Aural temperature was measured by an aural
Downloaded by: University of British Columbia. Copyrighted material.
thermistor inserted into the subjects' auditory canal [1] and
securely plugged and taped in position. The external ear was
then insulated with cotton wool. Aural temperature was em-
ployed in the study as rectal temperature has been considered
to be inappropriate as a measure of core temperature in para-
plegics during exercise [12], and subject discomfort is often re-
ported from the use of oesophageal probes [22]. Skin thermis-
tors were placed at the forehead, forearm, upper arm, back,
chest, abdomen, thigh, and calf in order to establish the whole
body thermoregulatory response.
Thermistors were attached to the skin using narrow strips of
water permeable surgical tape (3M Transpore, Loughborough,
England) in a criss-cross pattern. The ends of the strip were an-
chored with surgical tape. This technique maintained an ap-
propriate skin-to-thermistor interface while minimising the
area of skin covered by the surgical tape [14]. Subjects then un-
dertook a standardised five minute warm up at a work rate of
25 – 35 W on the arm crank ergometer or at a power output
equivalent to the initial stage of the WCE V̇O2 peak test as ap-
propriate. Subjects then exercised at an intensity of 60 % V̇O2
peak for 60 minutes. On completion of the exercise test sub-
jects remained in the seated position and a second 5 ml venous
blood sample was obtained. Post-exercise haemoglobin con-
centration and haematocrit were subsequently analysed. Sub-
jects then rested quietly for 30 minutes and were re-weighed.
Aural and skin temperatures were recorded at rest, post warm-
up, and every 5 minutes during the exercise period and during
the first 30 minutes of recovery. Values were recorded by a
Grant Squirrel meter logger (Grant Instruments, SQ8-16U,
Cambridge, England) via Edale thermistors (Edale Instru-
ments, Cambridge, England). Heat storage was calculated from
the following formula employed by Havenith et al. [15] where:
Heat Storage = (0.8 ∆Tcore + 0.2 ∆Tskin) × Cb and Cb is the
specific heat capacity of the body tissue (3.49 J × g–1 × 8C–1).
Values were calculated from changes in aural and skin tem-
perature [20] from resting values at 5, 15, 30, 45, and 60 min-
utes of exercise. Small 20 µl capillary blood samples were ob-
tained from the earlobe at 5, 15, 30, 45, and 60 minutes during
the exercise period and 5 minutes post-exercise. One minute
expired air samples were collected via the Douglas bag tech-
nique during minutes 5, 15, 30, 45, and 60 of the exercise peri-
Thermoregulation and Efficiency of Wheelchair Athletes Int J Sports Med 1999; 20 459

od. The efficiency of exercise at these time points was also cal-
culated [9]. Ratings of perceived exertion were obtained dur-
ing each expired air collection (Borg Scale). Expired air sam-
ples were analysed for fractions of expired oxygen and carbon
dioxide (Servomex Analyser Series 1400, Crowborough, Eng-
land) and evacuated (Harvard Dry Gas Meter, Harvard Appara-
tus Ltd. Kent, England) to determine ventilation rate (VE). Val-
ues for oxygen consumption (V̇O2), carbon dioxide production
(V̇CO2), and respiratory exchange ratio (RER) were subsequent-
ly calculated. Both analysers were calibrated before each series
of measurements with nitrogen, a calibration gas, and room
air.

Statistical analysis
The peak physiological responses to ACE and WCE for the PA
athletes were compared using paired t-tests. All physiological
and thermoregulatory data from the prolonged ACE and WCE
protocols were compared using two way Analysis of Variance
with repeated measures. Significance was accepted at the

Downloaded by: University of British Columbia. Copyrighted material.

P < 0.05 level. Where significance was obtained, Tukey post-
hoc analysis was undertaken.

Results

Responses to maximal arm crank and wheelchair ergometry

The peak physiological responses to incremental ACE and WCE
for the paraplegic (PA) athletes are shown in Table 1 (Mean ±
SD). No differences were observed for V̇O2 peak, blood lactate
(BLa peak), and heart rate (HRpeak) between incremental ACE
and WCE. Peak power output (POpeak) was greater during ACE
when compared to WCE (P < 0.05). Those athletes who
achieved the highest V̇O2peak values during ACE also achieved
the highest value during WCE (r = 0.93; P < 0.01).

Table 1 Peak physiological response to incremental arm crank ergo-

Fig. 1 Oxygen consumption (above) and efficiency (below) for the
metry (ACE) and wheelchair ergometry (WCE) for the paraplegic
paraplegic wheelchair athletes at rest and during prolonged arm
wheelchair athletes (Mean ± SD)
crank (ACE) and wheelchair ergometry (WCE) (R = Rest).
ACE WCE

V̇O2peak (l × min–1) 1.90 ± 0.40 1.96 ± 0.40 during both trials with values tending to increase from
–1
V̇O2peak (ml × kg × min ) –1
29.7 ± 8.2 31.5 ± 8.8 139 ± 10 and 137 ± 23 bts × min–1 at 5 minutes of exercise and
HRpeak (bts × min–1) 185 ± 7 183 ± 14 157 ± 21 and 159 ± 27 bts × min–1 at 60 minutes of exercise for
the ACE and WCE trials, respectively. The efficiency of exercise
VEpeak (l × min–1) 74.3 ± 16.0 73.3 ± 16.0
was greater during ACE when compared to WCE throughout
POpeak (Watts) 125* ± 24 55 ± 31 the exercise period (P < 0.05). At 5 minutes of exercise the effi-
BLapeak (mmol × l–1) 7.04 ± 1.37 5.58 ± 1.74 ciency of ACE and WCE were 18.5 ± 3.5 % and 8.9 ± 3.7 %, respec-
RERpeak 1.13 ± 0.07 1.08 ± 0.13 tively (P < 0.05). At the end of the ACE trial efficiency tended to
decrease from initial values (15.1 ± 1.2 %), whereas during WCE
* Significantly different from WCE at the P < 0.05 level
values remained at similar levels throughout exercise
(8.1 ± 3.3). The lowest efficiency values were observed for the
athletes exercising in basketball wheelchairs (3.3 % and 5.3 %
Physiological responses during prolonged arm-crank and at 5 minutes of exercise). No differences were observed for
wheelchair ergometry the VE, HR, or RPE responses during prolonged ACE or WCE.
Ratings of perceived exertion tended to be one point greater
Oxygen consumption during prolonged ACE and WCE and the during the final 30 minutes of the ACE trial when compared
efficiency of both modes of exercise are shown in Fig. 1. Oxygen to the WCE trial.
consumption was similar throughout exercise for both trials
(1.09 ± 0.21 and 1.16 ± 0.33 at 5 minutes of exercise and The blood lactate (BLa) and respiratory exchange ratio (RER)
1.28 ± 0.36 and 1.30 ± 0.35 at 60 minutes of exercise for the responses during the ACE and WCE trials are shown in Fig. 2.
ACE and WCE trials, respectively). Heart rates were also similar Blood lactate was greater during the ACE trial at 15 minutes
460 Int J Sports Med 1999; 20
Fig. 2 Blood lactate (above) and respiratory exchange ratio (below)
for the paraplegic wheelchair athletes at rest, during prolonged arm
crank (ACE) and wheelchair (WCE) exercise and recovery (R = Rest;
Rec = Recovery).
(2.44 ± 0.63 mmol × l–1) and 30 minutes of exercise (2.12 ±
1.05 mmol × l–1) when compared to the same time points dur-
ing the WCE trial (1.46 ± 0.63 mmol × l–1 and 1.32 ±
0.40 mmol × l–1, respectively; P < 0.05). Respiratory exchange
ratio was greater during the ACE trial at 5 minutes of exercise
(1.02 ± 0.06) when compared to the WCE trial (0.93 ± 0.05,
P < 0.05). Both BLa and RER decreased throughout the remain-
der of the exercise period during both trials.
Thermoregulatory responses during exercise and recovery
The aural temperature responses for the paraplegic athletes at
rest, during prolonged exercise and recovery for the ACE and
WCE trials are shown in Fig. 3. Similar increases in aural tem-
perature were noted for the PA athletes during the ACE and
WCE trials (0.6 ± 0.3 8C and 0.5 ± 0.3 8C, respectively). Values
were greater than during rest from 20 minutes of exercise until
5 minutes of recovery during the ACE trial (P < 0.05) but only
between 25 minutes to 45 minutes of exercise during the
WCE trial (P < 0.05). During recovery from exercise aural tem-
perature decreased at similar rates during both the ACE and
WCE trials.
M. J. Price, I. G. Campbell
Fig. 3 Aural temperature for paraplegic wheelchair athletes at rest,
during prolonged exercise and recovery for the arm crank ergometry
(ACE) and wheelchair ergometry (WCE) trials (R = Rest).
Downloaded by: University of British Columbia. Copyrighted material.
Fig. 4 Upper arm skin temperature for the wheelchair athletes at
rest, during prolonged exercise and recovery for the arm crank ergo-
metry (ACE) and wheelchair ergometry (WCE) trials (R = Rest).
During the WCE trial upper arm skin temperature was cooler
throughout the exercise period when compared to the ACE
trial (P < 0.05). On the cessation of exercise upper arm skin
temperature returned to similar values as those observed dur-
ing the ACE trial (Fig. 4). The majority of skin temperatures
demonstrated similar responses during both the ACE and
WCE trials.
Heat storage for the PA athletes post warm-up, during pro-
longed exercise and recovery for the ACE and WCE trials is
shown in Fig. 5. No differences were observed for heat storage
during the ACE trial or the WCE trial. At the end of the exercise
period heat storage was 2.34 ± 1.37 J × g–1 and 0.79 ± 1.51 J × g–1
during the ACE and WCE trials, respectively. A group-time in-
teraction (P < 0.05) revealed that during ACE heat storage was
elevated from warm-up values from 15 minutes of exercise un-
til 15 minutes of recovery (P < 0.05), whereas during WCE heat
storage was not significantly elevated during the exercise peri-
od at any time point.
Thermoregulation and Efficiency of Wheelchair Athletes
Fig. 5 Heat storage for the wheelchair athletes post-warm-up, dur-
ing prolonged exercise and recovery for the arm crank ergometry
(ACE) and wheelchair ergometry (WCE) trials (R = Rest; Rec = Recov-
ery).
Fluid balance
Similar weight losses, fluid intake, and changes in plasma vol-
ume during both the ACE and WCE trials were observed
(0.67 ± 0.32 kg and 0.70 ± 0.36 kg; 426 ± 378 ml and 388 ±
305 ml; – 3.3 ± 3.4 % and – 4.0 ± 4.1 % for the ACE and WCE
trials, respectively).
Discussion
Peak physiological responses during arm crank
and wheelchair exercise
The V̇O2peak values achieved by the PA athletes during incre-
mental ACE and WCE were similar and consistent with pre-
vious studies employing athletes with a range of spinal cord
lesions [5, 21]. Some studies have demonstrated greater V̇O2-
peak values during WCE when compared to ACE [4,11]. This
may be due to the training status of the athletes employed in
these studies. Both Gass and Camp [11] and Campbell et al.
[4] employed elite athletes, whereas the present study em-
ployed a more heterogeneous group of trained athletes. Great-
er maximal exercise responses have been reported previously
for homogeneous groups of elite athletes when compared to
heterogeneous groups with specific exercise modes being
more sensitive to differences in training status than non
specific exercise modes [3, 25]. The athletes studied by Gass
and Camp [11] and Campbell et al. [4] may therefore have ex-
hibited greater V̇O2peak values due to greater training status
and the specificity of the exercise mode.
Physiological responses during prolonged arm crank and
wheelchair exercise
The greater blood lactate concentration during ACE when com-
pared to WCE may be related to the method of force applica-
tion to the ergometer in order to generate the required power
output. Arm crank ergometry involves a continuous applica-
Int J Sports Med 1999; 20 461
tion of force to the flywheel of the ergometer and no resting
phase, whereas propulsion during WCE involves the synchro-
nous application of force to the flywheel with a resting phase
of typically 70 % of the push cycle [17, 23, 29]. Consequently,
the active muscles may be in a contractile state and applying
force for longer durations during ACE when compared to
WCE. The greater BLa was reflected in the higher RER values
and the greater RPE values during the ACE trial.
The efficiency of ACE for all athletes was within the range com-
monly reported for both upper body [3] and lower body exer-
cise [10]. The lower efficiency during WCE when compared to
ACE is consistent with previous studies [2, 27, 28]. When com-
pared to the continual force application during ACE, WCE in-
volves the application of force to the handrims of the wheel-
chair during the forward arm swing of the propulsion phase
only, with the backward swing resulting in wasted energy
[23]. These inefficient propulsion biomechanics result in lower
efficiency values for WCE when compared to ACE of a similar
metabolic rate [23, 26, 28]. Therefore, during exercise at similar
Downloaded by: University of British Columbia. Copyrighted material.
metabolic rates, such as in the present study, differences in
propulsion technique and power output contribute to differen-
ces in the exercise efficiency between the two modes of exer-
cise.
Thermoregulatory responses during prolonged arm crank
and wheelchair exercise
Aural temperature was elevated above resting values through-
out exercise and the initial stages of recovery during ACE. Dur-
ing WCE aural temperature was elevated above resting values
between 25 and 45 minutes of exercise only. This may indicate
that less heat was produced or that less heat was retained dur-
ing WCE when compared to ACE. As the metabolic rates, and
therefore heat production, during both trials were equal and
less work accomplished during WCE when compared to ACE,
more waste heat would be liberated during the WCE trial.
However, as heat storage and aural temperature were not
elevated for the same time period during WCE when compared
to ACE, more heat must have been dissipated during the WCE
trial. As heat storage is a composite of both aural and skin tem-
perature changes, and both trials demonstrated similar aural
temperatures, the lower heat storage values observed during
WCE suggest cooler skin temperatures. Cooler upper body skin
temperatures, in particular those of the upper arm, may have
resulted from the generation of convective cooling currents
from the wheelchair wheels and a greater potential for periph-
eral heat loss. This in turn may have resulted in the lower heat
storage values during WCE.
Similar arm skin temperature responses during WCE to those
in the present study were observed by Gass et al. [12] for para-
plegic athletes exercising in track wheelchairs. Due to the
crouched position of the athletes in such wheelchairs, with
the arm in close proximity to the moving wheel, direct convec-
tive cooling of the arm may occur. However, two subjects in
the present study demonstrated similar results when exercis-
ing in basketball wheelchairs with a much more upright body
position and the arms a greater distance away from the
wheels. The decrease in upper arm skin temperature during
WCE may therefore be due to movement of the arm relative to
the body itself more so than local convective air currents from
the moving wheel of the wheelchair. The effectiveness of this
462 Int J Sports Med 1999; 20
local cooling on thermal strain for wheelchair athletes may be-
come more apparent during studies of wheelchair exercise in
hot conditions.
The research question of the present study suggested that
WCE, being less efficient than ACE, may result in greater in-
creases in body temperature. The data though suggest that
the opposite response occurs. For the same metabolic rate
ACE demonstrated elevations in heat storage during exercise
whereas WCE did not. The efficiency of ACE though tended to
decrease during the exercise period with the consequence of
more waste heat being produced and resulting in greater in-
creases in heat storage as observed during the ACE trial. As
ACE does not demonstrate the advantageous local cooling, as
shown by the decreasing upper arm skin temperatures during
WCE, this extra heat would not be dissipated. Therefore, at the
same metabolic rate the inefficient biomechanics of wheel-
chair propulsion which result in lower efficiency than ACE
[26] appear to provide for compensatory local cooling and
greater potential for heat dissipation, thus resulting in lower
heat storage and thermal strain.
If a greater thermal strain, based on heat storage values, exists
during ACE when compared to WCE, a greater risk of exercise
induced hyperthermia may also exist. Studies examining the
thermoregulatory responses of spinal cord injured subjects
during ACE may overestimate the thermal strain that these
athletes would be under during WCE, an effect which may be
accentuated by the presence of an external heat load. Based on
this, in order to accurately determine the thermoregulatory re-
sponses of spinal cord injured subjects specifically during pro-
longed exercise in a hot environment, wheelchair ergometry
should be employed. Furthermore, when the results of studies
involving spinal cord injured and able-bodied subjects during
ACE are compared to those obtained for wheelchair users dur-
ing WCE, differences in heat production due to the exercise
mode should be considered.
Due to the differences in physiological and thermoregulatory
responses during prolonged ACE and WCE observed in this
study, it would seem appropriate to employ WCE in order to
accurately assess the exercise responses of wheelchair ath-
letes. Furthermore, differences in responses due to the exercise
mode should be considered when comparing the physiological
and thermoregulatory responses of able-bodied athletes dur-
ing ACE and wheelchair athletes during WCE.
References
1. Benzinger TH, Taylor GW. Cranial measurements of internal
temperature in man. In: Hardy JR (ed). Temperature: Its Meas-
urements and Control in Science and Industry. 3, Biology and
Medicine. New York: Reinhold, 1973: 111 – 120
2. Brown DD, Knowlton RG, Hamill J, Schneider TL, Hetzler RK.
Physiological and biomechanical differences between wheel-
chair-dependent and able bodied subjects during wheelchair
ergometry. Eur J Appl Physiol 1990; 60: 179 – 182
3. Bunc V, Heller J. Ventilatory threshold and work efficiency on a
bicycle and paddling ergometer in top canoeists. J Sports Med
Phys Fitn 1991; 31: 376 – 379
4. Campbell IG, Lakomy HKA, Williams C. Physiological character-
istics of wheelchair athletes determined from arm crank ergo-
metry and wheelchair treadmill exercise. In: Atkinson G, Reilly
T (eds). Proceedings of the Third International Conference on
M. J. Price, I. G. Campbell
Sport, Leisure and Ergonomics. London: Burton Manor, E. & F.
N. Spon, 1995: 25 – 29
5. Coutts KD, Rhodes EC, McKenzie DC. Maximal exercise respon-
ses of tetraplegics and paraplegics. J Appl Physiol 1983; 55:
479 – 482
6. Dawson B, Bridle F, Lockwood RJ. Thermoregulation of paraple-
gic and able bodied men during prolonged exercise in hot and
cool climates. Paraplegia 1994; 32: 860 –870
7. Dill DB, Costill DL. Calculation of percentage changes in vol-
umes of blood, plasma and red cells in dehydration. J Appl Phys-
iol 1974; 37: 247 – 248
8. Durnin JGVA, Wormesley J. Body fat assessment from total body
density and its estimation from skinfold thickness: Measure-
ments of 481 men and women aged 16 – 71 years. Br J Nutr
1974; 32: 77 – 97
9. Fox EL, Mathews DK. The physiological basis of physical educa-
tion and athletics. Japan: Saunders College Publishing, Holt-
Saunders, 1981; 3rd ed.
10. Gaesser GA, Brooks GA. Muscular efficiency during steady-rate
exercise: effects of speed and work rate. J Appl Physiol 1975; 38:
1132 – 1139
11. Gass GC, Camp EM. The maximum physiological responses dur-
Downloaded by: University of British Columbia. Copyrighted material.
ing incremental wheelchair and arm cranking exercise in male
paraplegics. Med Sci Sports Exerc 1984; 16: 355 – 359
12. Gass GC, Camp EM, Nadel ER, Gwinn TH, Engel P. Rectal and rec-
tal vs. oesophageal temperature in paraplegic men during pro-
longed exercise. J Appl Physiol 1988; 64: 2265 – 2271
13. Glaser RM, Sawka MN, Brune MF, Wilde SW. Physiological re-
sponses to maximal effort wheelchair and arm crank ergome-
try. J Appl Physiol 1980; 48: 1060 – 1064
14. Goss FL, Herbert WG, Kelso TB. A comparison of mean skin tem-
peratures during prolonged cycle exercise. Res Quart Exerc
Sport 1989; 60: 292 – 296
15. Havenith G, Luttikholt VGM, Vrijkotte TGM. The relative influ-
ence of body characteristics on humid heat stress response.
Eur J Appl Physiol 1995; 70: 270 – 279
16. Hopman MTE, Oeseburg B, Binkhorst RA. Cardiovascular re-
sponses in persons with paraplegia to prolonged arm exercise
and thermal stress. Med Sci Sports Exerc 1993; 25: 577 – 583
17. Kerk JK, Clifford PS, Snyder AC, Prieto TE, O'Hagan KP, Schot TPK,
Mykleburst JB, Mykleburst BM. Effect of an abdominal binder
during wheelchair exercise. Med Sci Sports Exerc 1995; 27:
913 – 919
18. Price MJ, Campbell IG. Thermoregulatory responses of paraple-
gic and able-bodied athletes at rest during exercise and into re-
covery. Eur J Appl Physiol 1997; 76: 552 – 560
19. Price MJ, Campbell IG. Determination of peak oxygen uptake
during upper body exercise. Ergonomics 1997; 40: 491 – 499
20. Ramanathan NL. A new weighting system for mean surface
temperature of the human body. J Appl Physiol 1964; 19: 531 –
533
21. Rasche W, Janssen TWJ, VanOers CAJM, Hollander AP, Van der
Woude LHV. Responses of subjects with spinal cord injuries to
maximal wheelchair exercise: comparison of discontinuous
and continuous protocols. Eur J Appl Physiol 1992; 66: 328 –
331
22. Sato KT, Kane NL, Soos G, Gisolfi CV, Kondo N, Sato K. Re-exam-
ination of tympanic membrane temperature as a core tempera-
ture. J Appl Physiol 1996; 80: 1233 – 1239
23. Sawka MN, Glaser RM, Wilde SW, Lurthe TC. Metabolic and cir-
culatory responses to wheelchair and arm crank exercise. J Appl
Physiol 1980; 49: 784 – 788
24. Sedlock DA, Knowlton RG, Fitzgerald PI. Circulatory and meta-
bolic responses of women to arm crank and wheelchair ergo-
metry. Arch Phys Med Rehab 1990; 71: 97 – 100
Thermoregulation and Efficiency of Wheelchair Athletes Int J Sports Med 1999; 20 463

25. Stamford BA, Chuddihee RW, Moffatt RJ, Rowland R. Task specif- Corresponding Author:
ic changes in maximal oxygen uptake resulting from arm versus
leg training. Ergonomics 1978; 21: 1 – 9 Michael Price, Ph.D.
26. Tropp H, Samulsson K, Jorfeldt L. Power output for wheelchair School of Chemical and Life Sciences
driving on a treadmill compared with arm crank ergometry. Br University of Greenwich
J Sports Med 1997; 31: 41 – 44 Wellington Street
27. Van der Woude LHV, DeGroot G, Hollander AP, Schenau GJV, Ro- Woolwich
sendal RH. Wheelchair ergonomics and physiological testing of London SE18 6PF
prototypes. Ergonomics 1986; 29: 1561 – 1573 U.K.
28. Veeger HEJ, Van der Woue LHV, Rosendal RH. Effect of handrim
velocity on mechanical efficiency in wheelchair propulsion. Tel.+ 44 (181) 3318643
Med Sci Sports Exerc 1992; 24: 100 – 107 Fax.+ 44 (181) 3318305
29. Wang YT, Deutsch H, Morse M, Hendrick B, Milliken T. Three di- E-mail:m.j.price@gre.ac.uk
mensional kinematics of wheelchair propulsion across racing
speeds. Adapt Phys Activ Quart 1995; 12: 78 – 87
30. Wicks JR, Lymburner K, Dinsdale SM, Jones NL. The use of multi-
purpose exercise testing with wheelchair ergometry and arm
cranking in subjects with spinal cord lesions. Paraplegia 1977;
15: 252 – 261

Downloaded by: University of British Columbia. Copyrighted material.