| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
O2 max in humans
1Department of Physiology, St. George's Hospital Medical School, Cranmer Terrace, London, United Kingdom; 2Department of Medicine, Division of Physiology, University of California, San Diego, California 92093-0623; 3Department of Exercise Science and Physiology, Hiroshima Women's University, Hiroshima 724-8558, Japan; and 4Division of Respiratory and Critical Care Physiology and Medicine, Harbor-University of California at Los Angeles Medical Center, Torrance, California 90502
Submitted 11 December 2002 ; accepted in final form 25 March 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
L) and CP thresholds, W', and maximum oxygen
uptake (
O2 max).
Subsequently, subjects cycled to the limit of tolerance (for 
360 s) on
three occasions, each followed by a work rate reduction to 1) 110%
CP, 2) 90% CP, and 3) 80% L for a 20-min
target. W' averaged 20.9 ± 2.35 kJ or 246 ± 30 J/kg. After
initial fatigue, 110% CP was tolerated for only 30 ± 12 s. Each subject
completed 20 min at 80% L, but only two sustained 20 min at
90% CP; the remaining four subjects fatigued at 577 ± 306 s, with
oxygen consumption at 89 ± 8%
O2 max. The results
support the suggestion that replenishing W' after fatigue necessitates a
sub-CP work rate. The variation in subjects' responses during 90% CP was
unexpected but consistent with mechanisms such as reduced CP consequent to
prior high-intensity exercise, variation in lactate handling, and/or regional
depletion of energy substrates, e.g., muscle glycogen. maximal oxygen uptake; fatigue; gas exchange dynamics
135% of the work rate (WR) associated with
maximal oxygen consumption (O2
max) (10) or for
durations of 
60 min (5),
for durations between these limits, the hyperbolic relationship appears to
provide a good characterization of the physiological response. The curvature
constant of the hyperbola (W') is mathematically equivalent to a
constant amount of work, i.e., the product of WR (above CP) and time
(6,
19,
22), and has been suggested to
be notionally equivalent to an energy store consisting of oxygen, high energy
phosphates, and a source related to anaerobic glycolysis
(17,
19). However, it may also
relate to the buildup of fatigue-inducing metabolites or conditions to a
critical level, such as low intramuscular pH and/or high intramuscular
Pi concentration ([Pi]).
The CP threshold has also been shown to be a close correlate of the highest metabolic rate that is associated with pulmonary oxygen uptake
(O2), and acid-base
status and blood lactate concentration ([La]) being maintained at a constant
level (10,
22). As such, CP has been
suggested to characterize the transition between, what Whipp
(30) has termed, the heavy-
and very heavy-intensity exercise domains. Above CP, in the very
heavy-intensity domain,
O2 and [La] increase
inexorably, terminating at fatigue with
O2 attaining its maximum
(e.g., Refs. 21,
22) and depletion of (or
attainment of the critical metabolite accumulation determining) W'.
Therefore, Fukuba and Whipp (6)
have hypothesized that continued exercise after exhaustion (i.e., with
W' depleted) would only be possible if the power output were reduced to
a level below CP, where predominantly aerobic energy transfer can occur, i.e.,
permitting W' to be partially restored or replenished. Conversely, it is
suggested that if at the point of fatigue (where fatigue is used in the sense
of a level of fatigue that limits exercise tolerance) the power output were
reduced to a level greater than CP, then continued exercise would not be
possible because repletion of W' would not occur in a domain where
obligatory increases in
O2, [La], and
[Pi] were evident. We were interested in testing this hypothesis by
determining whether, after induction of limiting fatigue with an initial
intense exercise bout, it was possible to continue to exercise at a reduced WR
if that rate was either above or below CP.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
O2 max and
the lactate threshold (L) to be estimated independently by two
experimenters using standard noninvasive pulmonary gas exchange criteria
(3,
32).
|
Exercise tests were then performed to exhaustion at three different high-intensity constant-load WRs, chosen to span a range of fatigue times of
4–10 min. The subjects were required to maintain a pedal cadence
between 60 and 80 rpm for all constant-load tests. The times to exhaustion
were measured to the nearest second with a stopwatch and used to estimate each
subject's CP by using a three-point linear regression and extrapolation of the
time-1 vs. WR relationship. The slope of this relationship provides
the W', and its WR intercept corresponds to CP (e.g., Refs.
9,
22). All tests were preceded
by (3–4 min) and followed by (6 min) a 20-W control period. All
constant-load tests were assigned in a randomized sequence.
Finally, each subject completed three subsequent tests consisting of an initial exhaustive WR (WRexh; targeted from the power-duration
relationship to exhaust the subject in 6 min), followed by, at the point
of fatigue (when subjects were no longer able to maintain a pedal cadence of
at least 60 rpm), an abrupt reduction in the WR to each of three intensities
for a 20-min target period: 1) 110% CP (very heavy), 2) 90%
CP (heavy), and 3) 80% L (moderate). Each subject
performed the experiments requiring the recovery intensities of very heavy,
heavy, and moderate in a randomized sequence and separated by at least 48 h.
If the subject was unable to maintain the recovery exercise for the required
target of 20 min, the fatigue time was recorded and the WR was reduced to 20 W
for the remainder of the 20 min.
Equipment. The subjects breathed through a mouthpiece connected to a low-dead space (90 ml), low-resistance (<1.5 cmH2O at 3 l/s)
turbine volume transducer (Interface, Irvine, CA) for the measurement of
inspiratory and expiratory volume. Respired gas was continuously sampled (at 1
ml/s) from the mouthpiece and analyzed by mass spectrometry (QP9000, Morgan
Medical, Gillingham, UK) for the concentrations of oxygen, carbon dioxide, and
nitrogen. Before each experiment, calibration was made from two
precision-analyzed gas mixtures chosen to span the range of respired gas
concentrations; this calibration was verified at the end of each experiment.
The time delay between the gas concentrations and the volume signals was
measured by passing a bolus of known gas mixture through the system by using a
low dead-space solenoid valve
(2). The electrical signal
outputs were digitally converted every 20 ms and sampled by computer for the
calculation of O2, carbon
dioxide output (CO2), and
minute ventilation (E) by using the
algorithms of Beaver et al. (1,
2).
Analyses. Editing of data was performed from O2 in the time domain to
exclude occasional aberrant breaths caused by swallowing, coughing, sighing,
etc., which were considered not to reflect the underlying response, i.e.,
values greater than three standard deviations from the local mean were omitted
(15). All breath-by-breath
O2 responses were
interpolated on a second-by-second basis and time averaged by using discrete
10-s time bins to produce a standard weighted response.
Times to fatigue were compared by using ANOVA and post hoc Neuman-Keuls tests. Differences between the six square-wave exhaustive exercise bouts were considered significant if P < 0.05. The dispersion about the mean is expressed as ± standard deviation.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
O2 increased linearly throughout the incremental ramp test (after a delay due to the
O2 time constant), up to
O2 max (averaging 4.5
± 0.3 l/min). The slope of the
O2 response as a function
of WR (or gain) averaged 12.3 ± 0.6 ml · min-1
· W-1, falling within the expected normal limits for cycle ergometer exercise (28).
L averaged 2.7 ± 0.3 l/min, which corresponded to 60.0
± 4.6% of O2 max
(Table 1).
Constant-load tests to estimate CP. In the subsequent exhaustive constant-load tests, O2
max was not significantly different from that obtained during the
incremental ramp test (4.5 ± 0.4 l/min). As expected, the time to
fatigue was inversely correlated (P < 0.05) to power output in all
subjects. An example from subject 1 is shown in
Fig. 1A. With the use
of the time to fatigue at each of these WRs, a linear regression and
extrapolation of the power vs. time-1 relationship was used to
calculate CP (the abscissa intercept) and W' (the slope of the linear
fit) for each subject (an example from subject 1 is shown in
Fig. 1B). The 95%
confidence limits of the fit (dotted lines in
Fig. 1B) were also
calculated, and the fit was deemed acceptable if the confidence of CP
estimation was within ±5%, which was the case in all but one instance.
In this case, a fourth constant-load test was performed at another WR to
improve the accuracy of CP estimation. The CP threshold occurred at 246
± 24 W, which was equivalent to 41.8 ± 9.0% of the change in WR
range [defined as the difference between the WRs at L and
O2 max taken from the
ramp incremental test, c.f., Özyener et al.
(21)]
(Table 1). The W' value
averaged 20.9 ± 2.35 kJ or 246 ± 30 J/kg for the six
subjects.
|
Intensity-dependent recovery tests. WRexh, predicted from the power-duration relationship, to fatigue the subject in 6 min
averaged 305 ± 22 W (Table
2). Subjects completed an average of 348 ± 40 s at
WRexh before fatigue (there being no significant difference between
each of the 3 bouts). The subsequent WRs for the three experimental conditions
(110 and 90% CP, and 80% L) are given in
Table 2. When the WR was
reduced to 110% CP (very heavy-intensity domain; a reduction in WR averaging
34 ± 7 W), exercise was sustained, on average, for only 30 ± 12
s beyond the point of initial fatigue (i.e., after exhaustion at
WRexh; Table 2).
Even though WR was reduced by an average of 11 ± 2%, subjects were
still unable to sustain a significant duration of exercise. This was
calculated to be equivalent to only 3.5 ± 1.7% of W' (on the
basis of the WR above CP multiplied by the recovery duration expressed as a
percentage of W'). Examples of this response are shown in
Fig. 2, A and
B.
|
|
In contrast, when WR was dropped to 80% L (the moderate-intensity domain; a reduction in WR averaging 170 ± 22 W), all
subjects completed the 20-min target period, as expected
(Table 2;
Fig. 2A). This
corresponded to a 56 ± 4% reduction in WR.
Interestingly, however, only two of six subjects completed the target 20 min at the 90% CP WR (the heavy-intensity exercise domain; a reduction in WR
averaging 83 ± 5 W) with the remaining four subjects fatiguing before
the 20-min target. The exercise duration at 90% CP was highly variable and
averaged 785 ± 400 s (range: 188–1,200 s;
Table 2). The drop in WR from
WRexh averaged 27 ± 2%. Two examples of the
O2 response are shown in
Fig. 2, A and
B. Figure
2A provides an example of a subject who could not sustain
the 20-min target exercise period, and Fig.
2B provides an example of a subject who was able to
sustain the 20 min of exercise after initial fatigue at WRexh.
Figure 3 illustrates the times
to fatigue for each subject (after the initial fatiguing WRexh
bout) at the three different exercise intensities.
|
The four subjects who were unable to sustain exercise for the 20-min target period all fatigued at a value of
O2 less than the maximum
(e.g., subject 1 in Fig.
2A). In all cases, standard respiratory measurements
revealed that metabolic demands were falling at the time of fatigue. The
responses of O2,
CO2, and
E were falling in all cases. In
these four subjects, O2
at fatigue averaged 4.1 ± 0.4 l/min, or 89 ± 8% of their O2 max,
CO2 averaged 3.9 ±
0.3 l/min, or 73 ± 5% of
CO2 max (of the
WRexh constant load), and
E averaged 113 ± 9 l/min, or
77 ± 10% of O2 max
(of the WRexh constant load). Figure 4, A and
B, shows examples of
CO2 and
E in subjects 1 and
2, respectively (subject 1 fatigued before the 20-min
target; subject 2 achieved the 20-min target). Heart rate, however,
was observed to have a secondary rise during the 90% CP recovery exercise
bout, causing the four subjects who fatigued in recovery to attain a heart
rate of 100 ± 3% of their maximum heart rate (HRmax),
compared with the two subjects who achieved the 20-min target who attained 96
± 2% of HRmax (2 individual examples are shown in
Fig. 4, A and
B).
|
ANOVA and post hoc (Neuman-Keuls) tests revealed that, although there were no differences between exercise durations at WRexh on the three
tests, there was a significant difference between the exercise duration during
the recovery from a fatiguing bout (WRexh) within each of the
moderate, heavy, and very-heavy exercise intensity domains (P <
0.05). Figure 5 shows a summary
of the group mean responses. The mean power duration curve and the CP and
L thresholds are shown, and the arrows depict the mean
durations in recovery after depletion of W' during WRexh.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
O2 was not able to
fall, and it is inferred, therefore, that [La] would also not reduce (c.f.,
Ref. 22), and hence pH would
be likely to continue to fall, high-energy phosphate stores, e.g.,
phosphocreatine concentration ([PCr]), would continue to decrease
(24), and [Pi] [a
potent inductor of fatigue
(29)] would continue to
rise. In the very heavy-intensity domain the functional volume of work that is described by W' is progressively depleted until exhaustion, and, as the present results suggest, that recovery or repletion of W' is not possible at any level above CP. It is, therefore, suggested that for exercise to continue at a supra-CP WR a proportion of W' must remain. If W' were depleted during an exercise bout then a reduction in the WR to a level below CP is required for W' be replenished. In a practical sense [similar to Fukuba and Whipp (6)], this suggests that an athlete who performs with such an ill-considered pace strategy that W' is depleted before the end of the race would have to not only reduce their WR to continue but reduce it to a sub-CP level.
The determinants of W' are presently poorly understood but have been suggested by Knuttgen (14),
Monod and Sherrer (18), and
Moritani et al. (19), and in
review by Hill (9) to be
related to the anaerobic work capacity or maximum oxygen deficit
(9,
16). These reflect energy
stores, such as PCr (and ATP), stored oxygen, and stores of muscle glycogen
related to anaerobic glycolysis
(17). Because of its
relationship with the oxygen deficit, however, W' may also reflect an
inverse function of this, such as the attainment of maximum rates of aerobic
function (attainment of O2
max), or even a buildup of fatigue-inducing metabolites such as
H+ and Pi. What is interesting about the CP concept is
that it predicts that fatigue will ensue as soon as W' is depleted, an
event that occurs only above CP; these data add that W' may also not be
replenished in this intensity domain despite reducing the WR, i.e., the flux
of W' is unidirectional above CP.
W' in the present study averaged 20.90 ± 2.35 kJ or 246 ± 30 J/kg, consistent with values reported by Hill and Smith
(9,
11) in humans. The amount of
work actually performed by the subjects in the very heavy-intensity domain
after initial fatigue corresponds to only a small proportion of this value
(3.5 ± 1.7%). However, the 30-s tolerance duration at the supra-CP
intensity may actually reflect a tolerance time of, or close to, zero were the
subject (and W') actually exhausted by the fatiguing ride. That is, the
duration of the recovery period was determined from the point at which the
subject could no longer maintain a pedal cadence of >60 rpm, at which point
the WR was dropped and the subject was instructed to again increase the pedal
cadence; this recovery then continued until the subject could again no longer
maintain 60 rpm. It was, however, necessary to allow the subjects some time at
the new, reduced WR to determine whether they could actually sustain the
required exercise (during which time they received verbal encouragement from
the experimenters and the pedal cadence was in some cases continually falling
throughout). This time amounted to 10 s. Thus we had to allow the
subjects sufficient W' reserve to be able to bring the cycling cadence
back up to the target level. In a few subjects, it appears we did not judge
this to a sufficiently fine tolerance. This seems more likely than W',
having been completely depleted at end exercise and being immediately repleted
by 3–4% as the WR reduced. These findings are summarized in
Fig. 5.
It was not surprising that all the subjects were able to continue exercise in the moderate-intensity domain (below the L). This
necessitates a greatly reduced rate of ATP provision and, therefore,
O2 requirement. This
domain does not require a significant proportion of energy derived from
anaerobic mechanisms and, therefore, allows for the clearance of
fatigue-inducing substances.
What was more surprising, however, was that four of six subjects were unable to maintain exercise in the heavy-intensity domain for the required
20-min target; the overall average was 13 min. The hypothesis, based on
the models of Fukuba and Whipp
(6), suggested that subjects
would be able to sustain exercise in this intensity domain after initial
fatigue. Although we cannot rule out the possibility that the wide variability
observed between the recovery duration tolerated at this intensity may, at
least in part, be due to day-to-day variability in performance, we believe
this to be unlikely on the basis of both the magnitude of the differences and
the different response profiles in those who attained the target recovery time
and those who did not. Although each recovery bout was not repeated, the
initial WRexh was undertaken on three occasions that yielded an
average coefficient of variation for constant load performance of 10%. This
value is far lower than the 50% coefficient of variation for the duration
of the 90% CP recovery bout. The mechanism of this fatigue is unclear. The
four subjects fatigued at
aO2 averaging 89 ±
8% of their maximum but a heart rate of 100 ± 3% of their
HRmax, compared with the two subjects who achieved the target at 86
± 4% O2 max and 96
± 2% of HRmax (2 individual examples are shown in Figs.
2 and
4). It is possible, therefore,
that the heart rate in the four subjects who fatigued reflects a condition,
presumably (although not absolutely certainly) in the contracting muscles,
that is not tolerable. This is despite a reduction in metabolic demand and
oxygen pulse inferred from the
O2 and heart rate
responses (e.g., Fig.
4A). The falling oxygen pulse is consistent with stroke
volume and/or the arteriovenous oxygen content difference being lower than
that required for still-elevated metabolic demands. Further work is required
to elucidate the mechanism(s) of this novel feature. We assume that the four
subjects who could not recover sufficiently to meet the 20-min target would
have been successful had the WR been lower in the domain.
W' has previously been demonstrated to be dependent (at least in part) on the glycogen storage status of the subject
(17). Thus exercise designed
to deplete W' will also reduce muscle glycogen content in the
contracting units, but the degree to which this occurred in our subjects is
unknown. It is well established that during WRs that induce fatigue in 10
min, the average muscle glycogen content remains high at the point of
exhaustion (e.g., Ref. 25). At
WR ranges that demand between 60 and 85%
O2 max (on the basis of
the linear WR-O2
relationship from the ramp incremental test), muscle glycogen is almost completely depleted at fatigue (e.g., Ref.
4). Newsholme et al.
(20), for example, have
suggested that muscle glycogen limitation might even limit maximum performance
at durations of <30 min by using plausible estimates of muscle glycogen
storage and utilization rates
(13). Our subjects exercised
for a total of 19 min on average, of which 6 min was at a WR that
resulted in 100% of O2
max and an additional 13 min where
O2 declined, but only to
a value of 88% O2
max. It is, therefore, possible that fatigue ensued in some subjects
during the 90% CP bout, because of the differences in muscle glycogen
availability, but not in others. Furthermore, this degradation may be
exacerbated in different muscle fiber types
(26) and, therefore, may be
dependent on an individual's fiber-type profile and recruitment pattern during
the exercise. Another possible mechanism for the differences observed is on
the basis of lactate transport in the different individuals; this too,
however, may be exacerbated by a different fiber-type profile (e.g., Ref.
8). Because the fiber-type
characteristics of these subjects were not known, we hypothesized that
parameters of aerobic fitness, such as
O2 max or
L, which are related to fiber-type proportion, or, conversely,
anaerobic parameters, such as W', might provide a basis on which to
distinguish between subjects. We were, however, unable to find any correlation
between the tolerable duration of recovery at the 90% CP WR and these
parameters, but, of course, the subject number is small.
At the extremes, subject 5 exercised for a total of 21 min before fatiguing (i.e., 365 s at WRexh and 916 s at 90% CP),
whereas subject 3 managed only 7 min (294 s at WRexh
and 118 s at 90% CP), which would appear to be too short a duration to deplete muscle glycogen stores [by using the assumptions of Newsholme et al.
(20)]. Another explanation,
therefore, is that high-intensity exercise itself altered CP. CP has
previously been shown to be dependent on state of training
(7,
23) and on the inspired oxygen
fraction, being increased in hyperoxia
(31) and decreased in hypoxia
(19,
31). In contrast, glycogen
depletion affected W' only and not the value of CP
(17). It is possible that if
the fatiguing exercise itself (WRexh) affected the value of CP,
then the recovery exercise may have been at a level >90% CP. We believe
that this is unlikely, at least to within ±10% of CP because exercise
was not well tolerated at 110% CP but was better tolerated at 90% CP. It is
not possible, however, to establish CP immediately after fatigue by using
presently available techniques.
In conclusion, therefore, our findings demonstrate that functional recovery after fatiguing exercise is highly intensity dependent, and this has implications for the optimization of training strategies for athletic performance.
|
DISCLOSURES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
O2 max. Med
Sci Sports Exerc 34:
709-714, 2002.
This article has been cited by other articles:
|
|
 |
|
  C. Ferguson, B. J. Whipp, A. J. Cathcart, H. B. Rossiter, A. P. Turner, and S. A. Ward Effects of prior very-heavy intensity exercise on indices of aerobic function and high-intensity exercise tolerance J Appl Physiol, September 1, 2007; 103(3): 812 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. L. Calbet, C. Lundby, M. Sander, P. Robach, B. Saltin, and R. Boushel Effects of ATP-induced leg vasodilation on VO2 peak and leg O2 extraction during maximal exercise in humans Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2006; 291(2): R447 - R453. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P D Peeling, D J Bishop, G J Landers, and T Boone Effect of swimming intensity on subsequent cycling and overall triathlon performance * Commentary Br. J. Sports Med., December 1, 2005; 39(12): 960 - 964. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
), 90% CP (
), and 110% CP (
