See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/11379083

Oxygen uptake kinetics during treadmill

running across exercise intensity domains

Article in European Journal of Applied Physiology · February 2002

DOI: 10.1007/s00421-001-0556-2 · Source: PubMed

CITATIONS READS

108 695

4 authors, including:

Jamie S M Pringle Andrew M Jones

34 PUBLICATIONS 965 CITATIONS University of Exeter
354 PUBLICATIONS 13,464 CITATIONS
SEE PROFILE

SEE PROFILE

Jonathan H Doust
University of Brighton
70 PUBLICATIONS 3,652 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Longitudinal development of oxygen uptake kinetics and muscle deoxygenation in youth endurance
athletes View project

Impact of dietary nitrate on skeletal muscle vascular and metabolic control during exercise View
project

All content following this page was uploaded by Andrew M Jones on 22 May 2014.

The user has requested enhancement of the downloaded file.

Eur J Appl Physiol (2002) 86: 347–354
DOI 10.1007/s00421-001-0556-2
O R I GI N A L A R T IC L E
Helen Carter Æ Jamie S. M. Pringle Æ Andrew M. Jones
Jonathan H. Doust
Oxygen uptake kinetics during treadmill running
across exercise intensity domains
Accepted: 19 October 2001 / Published online: 11 January 2002

Springer-Verlag 2002
Abstract The purpose of the present study was to exam-
ine comprehensively the kinetics of oxygen uptake (V_ O2 )
during treadmill running across the moderate, heavy and
severe exercise intensity domains. Nine subjects [mean
(SD age, 27 (7) years; mass, 69.8 (9.0) kg; maximum V_ O2 ,
V_ O2max , 4,137 (697) mlÆmin–1] performed a series of
‘‘square-wave’’ rest-to-exercise transitions of 6 min
duration at running speeds equivalent to 80% and 100%
of the V_ O2 at lactate threshold (LT; moderate exercise);
and at 20%, 40%, 60%, 80% and 100% of the difference
between the V_ O2 at LT and V_ O2max (D, heavy and severe
exercise). Critical velocity (CV) was also determined using
four maximal treadmill runs designed to result in
exhaustion in 2–15 min. The V_ O2 response was modelled
using non-linear regression techniques. As expected, the
amplitude of the V_ O2 primary component increased with
exercise intensity [from 1,868 (136) mlÆmin–1 at 80% LT to
3,296 (218) mlÆmin–1 at 100% D, P<0.05]. However, there
was a non-significant trend for the ‘‘gain’’ of the primary
component to decrease as exercise intensity increased [181
(7) mlÆkg–1Ækm–1 at 80% LT to 160 (6) mlÆkg–1Ækm–1 at
100% D]. The time constant of the primary component
was not different between supra-LT running speeds (mean
value range = 17.9–19.1 s), but was significantly shorter
during the 80% LT trial [12.7 (1.4) s, P<0.05]. The V_ O2
slow component increased with exercise intensity from
H. Carter (&)
School of Sport, Exercise and Leisure,
University of Surrey Roehampton, West Hill,
London, SW15 3SN, UK
E-mail: helen.carter@roehampton.ac.uk
Tel.: +44-20-83923564
Fax: +44-20-83923749
J.S.M. Pringle Æ A.M. Jones
Exercise Physiology Group,
Manchester Metropolitan University,
Hassall Road, Alsager, ST7 2HL, UK
J.H. Doust
Dept. of Sport Science, University of Wales,
Aberystwyth, Old College, King Street,
Aberystwyth, Ceredigion, SY23 2AX, UK
139 (39) mlÆmin–1 at 20% D to 487 (57) mlÆmin–1 at 80% D
(P<0.05), but decreased to 317 (84) mlÆmin–1 during the
100% D trial (P<0.05). During both the 80% D and 100%
D trials, the V_ O2 at the end of exercise reached V_ O2max
[4,152 (242) mlÆmin–1 and 4,154 (114) mlÆmin–1, respec-
tively]. Our results suggest that the ‘‘gain’’ of the primary
component is not constant as exercise intensity increases
across the moderate, heavy and severe domains of tread-
mill running. These intensity-dependent changes in the
amplitudes and kinetics of the V_ O2 response profiles may
be associated with the changing patterns of muscle fibre
recruitment that occur as exercise intensity increases.
Keywords Running exercise Æ Oxygen consumption
slow component Æ Oxygen consumption primary
component
Introduction
The pulmonary oxygen uptake () response in the tran-
sition from rest to constant-load exercise has been well
described for cycling exercise. During moderate-inten-
sity exercise (i.e. below the lactate threshold, LT), fol-
lowing the cardiodynamic phase (phase 1), rises in an
approximately monoexponential fashion (phase 2) to
attain a new steady state (phase 3) within 2–3 min.
Previous research in cycling exercise has shown the
amplitude of the primary component of [Ac+p¢; i.e. the
combined amplitudes of phase 1 (cardiodynamic phase,
Ac) and phase 2 (primary phase, Ap)] to be linearly re-
lated to exercise intensity. The primary ‘‘gain’’ (Ac+p¢/
power output) therefore remains essentially constant
over a broad range of submaximal power outputs, with
values of @10 mlÆmin–1ÆW–1 commonly reported (Bar-
stow and Mole 1991; Paterson and Whipp 1991). During
heavy exercise above the LT, the kinetics are more
complex and have been suggested to be dependent upon
the proximity of the exercise intensity to the critical
power (CP). CP is the asymptote of the hyperbolic
power/time relationship, and therefore represents the
348
upper limit of sustained, tolerable work (Poole et al.
1988a). Between the LT and the CP, Ac+p¢ is supple-
mented by an additional slow component (As¢) that
causes to rise for several minutes until a delayed steady
state is achieved. This slow component causes to rise
above the value expected from extrapolation of the sub-
LT/power output relationship. In the severe exercise
domain, that is above CP, a steady state in may not be
attained and maximum () may be reached if exercise is
continued (Poole et al. 1988a).
Controversy exists over whether the time constant for
the primary component (sp) is slower (Paterson and
Whipp 1991) or is unchanged (Barstow and Mole 1991)
for exercise above, as compared to below the LT. The
accurate estimation of sp is of great importance in
understanding the control of the on-kinetic response.
A slower sp in heavy exercise has been interpreted to
indicate a reduced oxygen availability (MacDonald et al.
1997; Paterson and Whipp 1991). In contrast, an
invariant sp supports the notion that oxygen delivery is
not limiting during heavy exercise and that muscle
kinetics are controlled by intracellular processes (Grassi
2000; Grassi et al. 1998).
To date, research in this area has focussed almost
exclusively on cycle ergometry. Recent research from
our laboratory has shown a tendency for sp to be slower
in heavy- than in moderate-intensity treadmill running,
although the difference did not reach statistical signifi-
cance (Carter et al. 2000a, b). Previous studies reporting
differences in sp between moderate and heavy exercise
have selected relatively few exercise intensities, typically
examining the response at 80–90% LT and halfway
between the LT and (Barstow et al. 1996; Engelen et al.
1996; Paterson and Whipp 1991). Furthermore, none of
these studies measured CP, the parameter that demar-
cates the transition between the heavy and severe exer-
cise intensity domains (Poole et al. 1988a). This may
have implications for the interpretation of sp, since
kinetics differ markedly in these two domains (Poole
et al. 1988a).
The purpose of this study was to describe compre-
hensively kinetics throughout the moderate, heavy and
severe exercise intensity domains during treadmill run-
ning. Based upon previous studies, we hypothesised that
sp would be shorter during exercise at a sub-LT intensity
than at a supra-LT intensity, that the gain (Gc+p) of the
primary component (calculated in this study as per kg
body mass per unit of distance run; i.e. mlÆkg–1Ækm–1)
would be constant across the range of exercise intensities,
and that the amplitude of the slow component would
increase as exercise intensity increased above the LT.
Methods
Subjects
Nine recreationally active subjects [six men, mean (SD): age, 27
(7) years; mass, 69.8 (9.0) kg; , 4,137 (697) mlÆmin–1] volunteered to
take part in this study, which was approved by the Institution
Ethics Committee. The subjects gave written informed consent to
participate, after the experimental procedures, the associated risks
and the benefits of participation were explained. The subjects were
all fully familiar with laboratory exercise testing procedures.
The subjects were instructed to arrive at the laboratory in a
rested and fully hydrated state, at least 3 h post-prandial, and to
avoid strenuous exercise in the 48 h preceding a test session. For
each subject, tests took place at the same time of day (±2 h) to
minimise the effects of diurnal biological variation on the results.
Experimental design
The subjects were required to visit the laboratory for three stages of
experimentation. The first stage involved the determination of LT
and . The second stage involved seven laboratory sessions, the
subjects performing 2–3 repetitions of ‘‘square-wave’’ transitions
from rest to one of seven exercise intensities: 80% LT, 100% LT,
20% D (20% of the difference in between that at LT and ), 40% D,
60% D, 80% D, and 100% D. No more than two transitions were
completed in 1 day, with at least 1 h of recovery in between. The
third stage involved determination of the subjects’ critical velocity
(CV, being analogous to CP in cycling exercise) from four treadmill
runs to exhaustion (on separate days). The square-wave transitions
and the exercise bouts to determine CV were performed in random
order. The study was completed within 3–4 weeks for all subjects.
Procedures
All tests were performed on a motorised treadmill (Woodway,
Cardiokinetics, Salford, UK) with the grade set at 1% (Jones and
Doust 1996b). During the exercise tests, pulmonary gas exchange
was determined breath-by-breath. Subjects breathed through
a low-dead-space (90 ml), low-resistance (0.65 mmH2OÆl–1Æs–1 at
8 lÆs–1) mouthpiece and turbine assembly. Gases were drawn con-
tinuously from the mouthpiece through a 2-m capillary line of
small bore (0.5 mm) at a rate of 60 mlÆmin–1, and analysed for
oxygen, carbon dioxide and nitrogen concentrations by a quadru-
pole mass spectrometer (CaSE QP9000, Gillingham, Kent, UK),
which was calibrated before each test using gases of known con-
centration. Expiratory volumes were determined using a turbine
volume transducer (Interface Associates, California, USA). The
sampling rate for both gas flow and gas concentration was 50 Hz.
The volume and concentration signals were integrated by com-
puter, following analog-to-digital conversion, with account taken
of the gas transit delay through the capillary. Respiratory gas
exchange variables (carbon dioxide output and pulmonary venti-
lation) were calculated and displayed for every breath. Heart rate
was recorded telemetrically throughout the exercise tests (Polar
Electro Oy, Kempele, Finland).
Subjects performed incremental exercise to volitional exhaus-
tion in order to determine LT. The initial running speed was
6.0–7.0 kmÆh–1 for the female subjects and 8.0–9.0 kmÆh–1 for the
male subjects. Subjects completed 6–8 submaximal stages of 4 min
duration, with running speed increased by 1.0 kmÆh–1 between
stages (Jones and Doust 1997). At the end of each stage, subjects
supported their weight with their hands and moved their feet to the
sides of the treadmill belt. Fingertip capillary blood samples
(@25 ll) were collected in capillary tubes and subsequently analysed
for lactate concentration ([La])using an automated analyser (YSI
2300, Yellow Springs, Ohio, USA). Subjects recommenced running
within 10–15 s. When blood [La] exceeded 4 mM, or heart rate
exceeded 90% of the known or age-predicted maximum heart rate,
the treadmill gradient was increased by 1% every minute until the
subject reached volitional exhaustion. In a previous study (Jones
and Doust 1996a), we demonstrated that the measured following a
25-min LT determination was not significantly different from that
measured with a conventional 10-min fast-ramp protocol.
Plots of blood [La] against running speed and were provided to
two independent reviewers who determined the LT as the first

sudden and sustained increase in blood [La] above resting con-
centrations. The breath-by-breath gas-exchange data collected
during the incremental tests were averaged over consecutive 30-s
periods. The was defined as the highest attained in the test.
Extrapolation of the relationship between and running speed for
exercise at sub-LT intensities was used to estimate the running
speed at and to calculate the running speeds requiring 80% and
100% of the at LT (moderate-intensity exercise); and 20%, 40%,
60% and 80% of the difference in between LT and (heavy-
intensity exercise); and 100% D (i.e. 100% ).
Subsequently, subjects performed a series of square-wave tran-
sitions of 6 min duration at the seven exercise intensities on separate
days. The exercise protocol began with 2 min of standing rest with
feet astride the moving treadmill belt and hands holding the hand-
rails. Subjects supported their body mass with their hands on the
handrails until their leg speed matched treadmill belt speed, after
which they let go of the handrails and began running. The transition
from rest to exercise was complete within 5 s. Fingertip capillary
blood samples were taken immediately before and after the 6-min
exercise period. The difference between the end-exercise [La] and the
resting [La] is expressed as a delta value (D [La]). For the moderate
exercise trials (80% and 100% LT), the subjects performed a total of
three transitions, while for the heavy and severe exercise trials (20%
D–100% D), the subjects performed two transitions.
For the determination of CV, subjects completed a treadmill
run to exhaustion at four different exercise intensities: @90%,
100%, 110% and 120%. The running speeds chosen were designed
to elicit exhaustion in 2–15 min. Subjects were instructed to run for
as long as possible and were given strong verbal encouragement to
continue at all times. No indication was given as to the time
elapsed. The time to exhaustion was recorded to the nearest second
and the CV was calculated using the linear model of velocity versus
time–1 (Whipp et al. 1982).
Data analysis
For each exercise transition, the breath-by-breath data were in-
terpolated to give second-by-second values. The data from the
transitions for each intensity were then time-aligned to the start of
exercise and averaged in order to enhance the underlying response
characteristics. Non-linear regression techniques were used to fit
the data after the onset of exercise with an exponential function.
An iterative process ensured the sum of squared error was mini-
mised. The mathematical model consisted of two (moderate exer-
cise) or three (heavy exercise) exponential terms, each representing
one phase of the response (Barstow and Mole 1991; Barstow et al.
1996). Based on previous literature (Barstow et al. 1996), the model
was constrained to aid in identification of the key parameters. The
first exponential term started with the onset of exercise (time, t=0),
whereas the other terms began after independent time delays;
V_ O2 ðtÞ ¼ V_ O2 ðbÞ þ Ac  ð1  et=sc ÞPhase 1ðcardiodynamic componentÞ
þ Acþp  ð1  eðtTDpÞ=sp ÞPhase 2ðprimary componentÞ
þ As  ð1  eðtTDsÞ=ss ÞPhase 3ðslow componentÞ
ð1Þ
where (b) is the average value over the 2 min of resting baseline; Ac, Ap,
As are the asymptotic amplitudes for the exponential terms; sc, sp and ss
are the time constants; and TDp and TDs are the time delays. The phase 1
term was terminated at the start of phase 2 (i.e. at TDp) and was assigned
the value for that time (Ac¢).
The at the end of phase 1 (Ac¢) and the amplitude of phase 2
(Ap) were summed to calculate the amplitude at the end of the
primary component (Ac+p¢). The slow component at the end of
exercise (As¢) was calculated and is used in preference to the
asymptotic value (Barstow et al. 1996).
Previous research has suggested that the response at exercise
intensities equating to and above reverts back to a two-phase
response (Hebestreit et al. 1998). This notion was tested using the
procedures of Motulsky and Ransnas (1987) by fitting the response
349
during the 100% D trial with both a two-phase and a three-phase
model. However, the residual sum of squares was significantly
greater for the two-phase model, and despite having fewer
parameters than the three-phase model, an F-test supported the use
of the more complex fit (F=–24.8, P<0.001).
Statistical analysis
One-way repeated measures analysis of variance, and where
appropriate, paired t-tests, were used to test for differences between
the exercise intensities. The level of statistical significance was set at
P<0.05. The results are presented as mean (SEM).
Results
The mean at LT was 3,036 (199) lÆmin–1, equating to
73.2 (1.6)% . When the at CV was predicted using the /
running speed relationship, it was found to occur at 84.6
(1.7)% (t8=–4.8, P=0.001). When expressed relative to
the at LT and , CV occurred at 41 (7.6)% D.
The response of a typical subject is shown in Fig. 1
and the mean data for the subject group is given in
Table 1. As expected, Ac+p¢ increased with exercise in-
tensity [from 1,868 (136) mlÆmin–1 at 80% LT to 3,296
(218) mlÆmin–1 at 100% D, F6, 62=9.16, P=0.048).
However, in the suprathreshold square waves the pre-
dicted from the sub-LT /running speed relationship was
not achieved at the end of the primary component. As
Fig. 2 illustrates, there was a non-significant trend for
the gain of the primary component (Ac+p¢/D running
speed, expressed relative to body mass and termed
Gc+p,) to decrease as exercise intensity increased [181
(7) mlÆkg–1Ækm–1 at 80% LT to 160 (6) mlÆkg–1Ækm–1 at
100% D].
Two major patterns emerged in the temporal
parameters of the response. The time constant for the
primary component was not different between supra-LT
running speeds, but was significantly shorter during the
80% LT trial (F6, 62=11.5, P=0.035, see Table 1). There
was also a tendency for the onset of the slow component
(TDs) to occur earlier as exercise intensity increased,
reducing from @130 s at 20% D to @80 s at 100% D.
For heavy and severe exercise, the slow component
(As’) increased with exercise intensity from 139
(39) mlÆmin–1 at 20% D to 486 (57) mlÆmin–1 at 80% D
(F6, 62=53.6, P=0.004), but decreased to 317 (84) mlÆ-
min–1 during the 100% D trial. The additional slow
component caused the gain calculated from end exercise
to increase with exercise intensity (Fig. 2). During both
the 80% D and 100% D trials, the gross at the end of
exercise reached [4,152 (242) mlÆmin–1 and 4,154
–1
(114) mlÆmin , respectively]. In the 60% D trial, only
two subjects reached at end-exercise.
Discussion
This study is the first to describe comprehensively
the kinetic response to treadmill running across the
350

Fig. 1 The breath-by-breath

oxygen uptake (V_ O2 ) response
across exercise intensity do-
mains in one typical subject.
Data from the exercise per-
formed at 80% of the lactate
threshold (LT; closed circles),
100% LT (open triangles), 20%
of the difference between the
V_ O2 at LT and V_ O2max (D;
closed squares), 40% D (open
diamonds), 60% D (closed tri-
angles), 80% D (open circles)
and 100% D (closed diamonds)
are shown. The V_ O2 at LT
(solid line), the predicted V_ O2 at
critical velocity (CV; dashed
line), and V_ O2max (dotted line)
are also shown

Table 1 Parameters of the oxygen uptake (V_ O2 ) response as a
function of exercise intensity. Please refer to Eq. 1. Values are
presented as the mean (SEM). (LT Lactate threshold, BL baseline,
TDp time delay of the primary component, sp time constant for the
primary component, TDs time delay for the slow component, ss
time delay for the slow component, Ap’ amplitude of primary the
Variable 80% LT 100% LT 20% D
component, Ac¢ amplitude at end of the cardiodynamic phase,
Ac+p¢ sum of Ac¢ in Eq. 1 and Ap¢, As¢ value of slow component at
the end of exercise, Rel. As¢ (As¢/Ac+p¢+As¢) relative contribution
of the slow component to a net increase in V_ O2 at end exercise;
EEV_ O2 increase in V_ O2 above baseline at the end of exercise,
TotalV_ O2 EEV_ O2 plus baseline V_ O2
40% D 60% D 80% D 100% D

BL V_ O2 (mlÆmin–1) 465 (30) 474 (27) 496 (35) 513 (45) 507 (31) 528 (33) 541 (43)
Ac¢ (mlÆmin–1) 925 (80) 1,018 (145) 995 (107) 1091 (152) 908 (98) 940 (74) 1,048 (107)
sc (s) 6.2 (1.8) 5.0 (1.9) 5.8 (2.1) 5.5 (2.1) 2.8 (1.1) 3.7 (1.1) 4.0 (1.4)
TDp (s) 27.1 (1.3) 21 (0.6) 19.2 (1.3) 18.8 (1.0) 15.5 (0.8) 15.6 (0.8) 13.7 (0.8)
Ac+p¢ (mlÆmin–1) 1,869 (136) 2,515 (204) 2,688 (165) 2,856 (170) 3,028 (181) 3,137 (183) 3,296 (218)
sp (s) 12.7 (1.4) 20.9 (3.5) 18.8 (0.6) 19.1 (0.8) 18.3 (0.8) 17.9 (1.4) 17.9 (1.8)
TDs (s) 131.5 (9.6) 107.2 (7.5) 106.0 (10.2) 89.6 (7.1) 81.3 (5.8)
As¢ (mlÆmin–1) 139 (39) 264 (33) 399 (51) 487 (57) 317 (84)
ss (s) 221.7 (26) 289.4 (24) 247.1 (23) 255.3 (53) 224 (53)
Rel. As¢ (%EEV_ O2 ) 4.5 (1.2) 8.3 (1.0) 13.3 (1.1) 13.3 (1.1) 8.3 (2.0)
EEV_ O2 (mlÆmin–1) 1,869 (136) 2,515 (183) 2,827 (190) 3,120 (193) 3,472 (208) 3,624 (217) 3,613 (256)
Total V_ O2 (mlÆmin–1) 2,334 (146) 2,989 (204) 3,323 (210) 3,633 (225) 3,935 (220) 4,152 (242) 4,154 (114)

moderate, heavy and severe exercise intensity domains.
The LT was found to occur at @73% of the , similar to
previous values reported in trained endurance runners
(Carter et al. 2000a, b). The CV was found to occur at a
running speed substantially higher than the LT, mea-
sured at @85% . These markers thus allowed us to
identify the boundary between the heavy and severe
domains to be @40% D.
The existence of the slow component during tread-
mill running has been challenged by some investigators
(Billat et al. 1998a, b). However, we have shown rep-
eatedly that the slow component is evident in treadmill
running, although its magnitude tends to be somewhat
smaller than in cycling exercise (Carter et al. 2000a, b;
Jones et al. 1999). In the present study, the amplitude of
the slow component reached @600 mlÆmin–1 in some
subjects. As can be clearly seen in Fig. 2, the presence of
the slow component resulted in significant increases in
the end-exercise gain as exercise intensity increased.
Contrary to what has been shown in previous work in
cycling (Poole et al. 1988b), the slow component did not
cause to reach when working just above the CV.
During the 60% D trial, in which all subjects except one
were above their CV, increased to only @96% and
appeared to plateau.
The physiological mechanisms for the slow com-
ponent are not well understood. Glycogen depletion
studies in cycling (Vollestad and Blom 1985) and
running (Costill et al. 1973) have demonstrated that
type IIa and IIb motor units are recruited at the
exercise intensities at which the slow component
develops. Furthermore, although not universally
observed (Scheuermann et al. 2001), electromyographic
studies indicate increased neuromuscular activation
consistent with a progressive recruitment of type II
fibres over the time course in which the slow compo-

Fig. 2 The relationship
between ‘‘gain’’ calculated from
both Ac+p¢ (closed triangles)
and end exercise V_ O2 (open
circles), and exercise intensity
(mean ±SEM). The dashed line
represents the predicted primary
gain calculated from the sub-
LT V_ O2 /running speed rela-
tionship. The mean predicted
V_ O2 at CV of the subject group
is also represented (solid circle)
nent also develops (Mateika and Duffin 1994; Pringle
et al. 2000; Shinohara and Moritani 1992). As a con-
sequence, the development of the slow component has
been linked to the energetics of contraction of type II
fibres (Barstow et al. 1996; Whipp 1994). Barstow et al.
(1996) demonstrated that the contribution made by the
slow component to the total response during 8 min of
heavy constant-load cycling was negatively correlated
with the proportion of type I fibres. Crow and
Kushmerick (1982) have demonstrated that the con-
traction of type II muscle fibres is less efficient than
that of type I fibres (i.e. the phosphate:oxygen ratio is
@18% lower in the isolated mitochondria of type II
fibres). The inefficiency may be due, in part, to a
greater reliance on the a-glycerophosphate shuttle over
the malate-aspartate shuttle compared to type I fibres
(Crow and Kushmerick 1982; Willis and Jackman
1994). The lower efficiency of energy conversion by
cross-bridges in type II fibres means that the utilisation
of ATP is also less efficient, resulting in an increased
ATP cost per unit force generated (Saugen and Vol-
lestad 1995). It must be noted that much of the data
suggesting a lower efficiency of type II fibres was
obtained on rodent muscle, and the picture in humans
performing whole-body exercise is less clear.
An important feature of the data from the present
study is the tendency for the gain of the primary com-
ponent to decrease as exercise intensity increased. This is
in contrast to previous studies, comparing a limited
number of cycle power outputs, which have suggested
that the primary gain remains constant during exercise
above and below the LT (Barstow and Mole 1991;
Paterson and Whipp 1991). It is possible that the
reduction in the gain of the primary component is
peculiar to running exercise. It is unclear at this stage
what impact, if any, determinants of mechanical effi-
ciency such as contraction regimen, muscular duty cycle,
351
changes in stride length and frequency, centre of gravity
displacement, the stretch-shortening cycle, and involve-
ment of the upper body musculature have on the kinetic
response in running. Nevertheless, it is possible that this
phenomenon is also present in cycling, but that it has
not been recognised either because too few exercise
intensities have been studied (Barstow and Mole 1991;
Paterson and Whipp 1991), or because too few subjects
have been tested (Barstow et al. 1993; Ozyener et al.
2001). We observed recently a reduction in primary gain
from 12.3 mlÆmin–1ÆW–1 at 80% LT to 11.2 mlÆmin–1ÆW–1
at 25% D, 10.4 mlÆmin–1ÆW–1 at 50% D and finally
10.2 mlÆmin–1ÆW–1 at 75% D in seven subjects (Carter
et al. 2000a). This trend is also evident in the data of
Barstow et al. (1993), who showed that the primary gain
term was reduced in severe exercise compared to mod-
erate and heavy exercise, and in that of Ozeyener et al.
(2001), in which the primary gain decreased from
11.5 mlÆmin–1ÆW–1 at 90% LT to 10.0 mlÆmin–1ÆW–1 at
110%.
The phenomenon of a reduced primary gain may only
be detected if a mathematical model is used to examine
the kinetics. If, for the present data, the primary gain is
calculated using the at 3 min, as was done by Paterson
and Whipp (1991), it appears very similar across inten-
sities (177–180 mlÆkg–1Ækm–1). This is due to the inclu-
sion of a portion of the slow component, which started
to develop at 90–120 s after the onset of exercise
(Table 1). The primary gain during the 100% D trial (i.e.
at ) had decreased to @160 mlÆkg–1Ækm–1, with the slow
component bringing the end exercise nearer to the
‘‘target’’ (@178 mlÆkg–1Ækm–1, see Fig. 2). The end-
exercise gain cannot continue to increase at intensities
above 60% D, as the amplitude of the slow component
is ultimately limited by the attainment of.
We suggest that the increasing recruitment of type II
fibres with increasing exercise intensity (Henneman et al.
352
1974; Olson et al. 1968) may partially explain the
reduction in the primary gain and increased relative
contribution of the slow component as running speed
increases above the LT. At low exercise intensities (e.g.
at 80% LT as in this study), only type I fibres are likely
to be recruited (Gollnick et al. 1974; Vollestad and Blom
1985). As described by the size principle of Henneman
and colleagues (1974; Olson et al. 1968), the recruitment
of type IIa and IIb fibres will ensue as exercise intensity
increases. Exercise at intensities exceeding 90% may
require the progressive recruitment of motor units such
that the three major fibre types will eventually be
engaged (Gollnick et al. 1974). It has been observed
(Fig. 3) that individuals with a predominance of type II
fibres have a lower gain of the primary component and a
greater slow component for a given exercise intensity
(Barstow et al. 1996, 2000). Therefore, in our study the
reduction of the primary gain (Fig. 2) and increased
slow component amplitude may be explained by the
progressive recruitment of type II fibres at higher run-
ning speeds, and their increasing contribution to the
overall response.
Previous work in cycling exercise has suggested that
the slow component causes to project above the
expected , as predicted from the sub-LT /power output
relationship (Whipp 1994; Whipp and Mahler 1980).
However, as shown in Fig. 2, during treadmill running,
the slow component first makes up for the shortfall of
the primary component and then exceeds the ‘‘pre-
dicted’’ . This raises implications for interpreting the
true ‘‘oxygen cost’’ of running exercise within the heavy
and severe intensity domains. The question is whether
the true initial oxygen cost of exercise is represented by:
(1) the gain of the primary component, which then
becomes modified, or (2) the delayed steady-state or
end-exercise (Barstow et al. 1996; Whipp 1996).
Barstow et al. (1996) argued that since the Ac+p’/power
Fig. 3 Schematic of the gain of
the V_ O2 response to heavy-
intensity cycling exercise for
individuals with a predomi-
nance of type I and type II
muscle fibres (re-drawn from
Barstow et al. 1996)
output relationship during cycling exercise is linear
(Barstow and Mole 1991), the primary gain is the initial
target . This interpretation may not be applicable to
treadmill running in light of the present data, which
indicate that the primary gain term is not constant.
The time constant for the primary component in
heavy exercise for the present study was @18 s, similar to
values reported previously in trained individuals when
the primary component has been modelled as an inde-
pendent term (Barstow et al. 1996; Carter et al. 2000a,
b). Across the heavy and severe exercise intensity
domains, our data suggest that the time constant for the
primary component is invariant. However, in support of
Paterson and Whipp (1991), but in contrast to Barstow
and Mole (1991), sp appears to be shorter (i.e. the
kinetics may be faster) for exercise below compared to
above the LT. Previously, the equivocal findings had
been explained by the differences in the modelling
techniques used. Paterson and Whipp (1991) used a
monoexponential description of the kinetic response to
3 min of exercise, even in heavy exercise. The inclusion
of a portion of the slow component term will result in
an artificially slower phase 2 time constant (Barstow
1994). However, in the present study we found a slowing
of sp, despite using the same modelling procedures
developed by Barstow and colleagues. The nature of the
present study design (i.e. more exercise intensities) and
the mode of exercise used may explain the different
results. An invariant time constant for muscle has been
observed in isolated gastrocnemius muscle across work
rates (Piiper et al. 1968), but it is reasonable to suggest
that the situation in the whole-body response to exercise
is different. For exercise intensities above the LT, the
slower kinetics may be compensated for energetically by
increased anaerobic metabolism (Di Prampero et al.
1970). This is evidenced by the increased blood [La]
associated with exercise in this intensity domain (Pat-

erson and Whipp 1991). A factor that might contribute
to the slower time constant is the recruitment of type II
fibres in this domain. The kinetics of the mouse extensor
digitorum muscle, which is made up predominantly of
type II fibres, are considerably slower than those mea-
sured in muscle containing more type I fibres (Crow and
Kushmerick 1982). If more type II fibres are recruited at
higher intensities, it is reasonable to propose that their
slower kinetics will impact on the composite measured
at the mouth.
In conclusion, this study has described comprehen-
sively the kinetic responses across a wide range of
exercise intensities during treadmill running. In contrast
to cycling (Barstow and Mole 1991), the gain of the
primary component was not constant across the sub-
maximal exercise intensity range, and the primary time
constant was increased for exercise above the LT. The
assumption of linear first-order behaviour of the pri-
mary component may require fundamental reappraisal
for treadmill exercise.
References
Barstow TJ (1994) Characterization of VO2 kinetics during heavy
exercise. Med Sci Sports Exerc 26:1327–1334
Barstow TJ, Mole PA (1991) Linear and nonlinear characteristics
of oxygen uptake kinetics during heavy exercise. J Appl Physiol
71:2099–2106
Barstow TJ, Casaburi R, Wasserman K (1993) O2 uptake kinetics
and the O2 deficit as related to exercise intensity and blood
lactate. J Appl Physiol 75:755–762
Barstow TJ, Jones AM, Nguyen PH, Casaburi R (1996) Influence
of muscle fiber type and pedal frequency on oxygen uptake
kinetics of heavy exercise. J Appl Physiol 81:1642–1650
Barstow TJ, Jones AM, Nguyen PH, Casaburi R (2000) Influence
of muscle fibre type and fitness on the oxygen uptake/power
output slope during incremental exercise in humans. Exp
Physiol 85:109–116
Billat V, Binsse V, Petit B, Koralsztein J (1998a) High level runners
are able to maintain a VO2 steady state in all-out run over their
critical velocity. Arch Physiol Biochem 107:1–8
Billat VL, Richard R, Binsse VM, Koralsztein JP, Haouzi P
(1998b) The VO2 slow component for severe exercise depends
on type of exercise and is not correlated with time to fatigue.
J Appl Physiol 85:2118–2124
Carter H, Jones A, Barstow T, Burnley M, Williams C, Doust J
(2000a) Oxygen uptake kinetics in treadmill running and cycle
ergometry: a comparison. J Appl Physiol 89:899–907
Carter H, Jones A, Barstow T, Burnley M, Williams C, Doust J
(2000b) Effect of endurance training on oxygen uptake kinetics
during treadmill running. J Appl Physiol 89:1744–1752
Costill D, Gollnick P, Jansson E, Saltin B, Stein E (1973) Glycogen
depletion pattern in human muscle fibres during distance run-
ning. Acta Physiol Scand 89:374–383
Crow M, Kushmerick M (1982) Chemical energetics of slow- and
fast-twitch muscles of the mouse. J Gen Physiol 79:147–166
Di Prampero PE, Davies CT, Cerretelli P, Margaria R (1970) An
analysis of O2 debt contracted in submaximal exercise. J Appl
Physiol 29:547–551
Engelen M, Porszasz J, Riley M, Wasserman K, Maehara K,
Barstow TJ (1996) Effects of hypoxic hypoxia on O2 uptake and
heart rate kinetics during heavy exercise. J Appl Physiol
81:2500–2508
Gollnick PD, Piehl K, Saltin B (1974) Selective glycogen depletion
pattern in human muscle fibres after exercise of varying
353
intensity and at varying pedalling rates. J Physiol (Lond)
241:45–57
Grassi B (2000) Skeletal muscle VO2 on-kinetics: set by O2 delivery
or by O2 utilization? New insights into an old issue. Med Sci
Sports Exerc 32:108–116
Grassi B, Gladden LB, Samaja M, Stary CM, Hogan MC (1998)
Faster adjustment of O2 delivery does not affect VO2 on-
kinetics in isolated in situ canine muscle. J Appl Physiol
85:1394–1403
Hebestreit H, Kriemler S, Hughson RL, Bar-Or O (1998) Kinetics
of oxygen uptake at the onset of exercise in boys and men.
J Appl Physiol 85:1833–1841
Henneman E, Clamann HP, Gillies JD, Skinner RDJ (1974) Rank
order of motoneurons within a pool: law of combination.
Neurophysiology 37:1338–1349
Jones AM, Doust JH (1996a) A comparison of three treadmill
protocols for the determination of maximal aerobic power in
runners (abstract). J Sports Sci 14:89
Jones AM, Doust JH (1996b) A 1% treadmill grade most accu-
rately reflects the energetic cost of outdoor running. J Sports Sci
14:321–327
Jones AM, Doust JH (1997) Specific considerations for the
assessment of middle distance and long distance runners. In:
Bird S, Davison R (eds) The British Association of Sport and
Exercise Sciences physiological testing guidelines (3rd edn).
Bases, Leeds, pp 108–111
Jones AM, Carter H, Doust JH (1999) A disproportionate increase
in coincident with lactate threshold during treadmill exercise.
Med Sci Sports Exerc 31:1299–1306
MacDonald M, Pedersen PK, Hughson RL (1997) Accelera-
tion of VO2 kinetics in heavy submaximal exercise by
hyperoxia and prior high-intensity exercise. J Appl Physiol
83:1318–1325
Mateika JH, Duffin J (1994) Coincidental changes in ventilation
and electromyographic activity during consecutive incremental
exercise tests. Eur J Appl Physiol 68:54–61
Motulsky HJ, Ransnas LA (1987) Fitting curves to data using
nonlinear regression: a practical and nonmathematical review.
Faseb J 1:365–374
Olson CB, Carpenter DO, Henneman E (1968) Orderly recruitment
of muscle action potentials. Arch Neurol 19: 591–597
Ozyener F, Rossiter HB, Ward SA, Whipp BJ (2001) Influence of
exercise intensity on the on and off transient kinetics of pulmo-
nary oxygen uptake kinetics in humans. J Physiol 533:891–902
Paterson DH, Whipp BJ (1991) Asymmetries of oxygen uptake
transients at the on- and offset of heavy exercise in humans.
J Physiol (Lond) 443:575–586
Piiper J, Di Prampero PE, Cerretelli P (1968) Oxygen debt and
high-energy phosphates in gastrocnemius muscle of the dog.
Am J Physiol 215:523–531
Poole DC, Ward SA, Gardner GW, Whipp BJ (1988a) Metabolic
and respiratory profile of the upper limit for prolonged exercise
in man. Ergonomics 31:1265–1279
Poole DC, Ward SA, Whipp BJ (1988b) Control of blood-gas and
acid-base status during isometric exercise in humans. J Physiol
(Lond) 396:365–377
Pringle JSM, Carter H, Doust JH, Jones AM (2000) Oxygen
uptake kinetics and electromyographic activity during level and
graded treadmill running. J Physiol (Lond) 523P:243P
Saugen E, Vollestad NK (1995) Nonlinear relationship between
heat production and force during voluntary contractions in
humans. J Appl Physiol 79:2043–2049
Scheuermann BW, Hoelting BD, Noble ML, Barstow TJ (2001)
The slow component of O2 uptake is not accompanied by
changes in muscle EMG during repeated bouts of heavy exer-
cise in humans. J Physiol (Lond) 531:245–256
Shinohara M, Moritani T (1992) Increase in neuromuscular
activity and oxygen uptake during heavy exercise. Ann Physiol
Anthropol 11:257–262
Vollestad NK, Blom PCS (1985) Effect of varying exercise intensity
on glycogen depletion in human muscle fibers. Acta Physiol
Scand 125:395–405
354
Whipp BJ (1994) The slow component of O2 uptake kinetics during
heavy exercise [published erratum appears in Med Sci Sports
Exerc 1995 Feb;27(2):298]. Med Sci Sports Exerc 262:1319–
1326
Whipp BJ (1996) Domains of aerobic function and their limiting
parameters. In: Steinacker JM, Ward SA (eds) The physiology
and pathophysiology of exercise tolerance. Plenum, New York,
pp 83–89
View publication stats
Whipp BJ, Mahler M (1980) Dynamics of pulmonary gas exchange
during exercise. In: West JB (ed) Pulmonary gas exchange.
Academic, New York, pp 36–96
Whipp BJ, Huntsman DJ, Stoner N, Lamarra N, Wasserman K
(1982) A constant which determines the duration of tolerance of
high-intensity work. Fed Proc 41:1591
Willis WT, Jackman MR (1994) Mitochondrial function during
heavy exercise. Med Sci Sports Exerc 26:1347–1353