| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Children's Health and Exercise Research Centre, School of Sport and Health Sciences, University of Exeter, Exeter EX1 2LU, United Kingdom
Submitted 25 July 2003 ; accepted in final form 17 March 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2) for 9 min on an electronically braked cycle ergometer. Each subject's breath-by-breath responses were interpolated to 1-s intervals, time aligned, and averaged. The data after phase 1 were fit with 1) a double exponential model and 2) a single exponential model within a fitting window that was previously identified to exclude the slow component. There were no significant differences in the parameters of the primary component between each model. Subsequent analysis was carried out using model 2. The O2 slow component was computed as the difference between the amplitude of the primary component and the end-exercise O2 and was expressed as the percent contribution to the total change in O2. Over the 2-yr period, the primary time constant (boys 16.8 ± 5.3 and 21.7 ± 5.3 s, girls 21.1 ± 8.1 and 26.4 ± 8.4 s, first and second occasion, respectively) and the relative amplitude of the slow component (boys 9.4 ± 4.6 and 13.8 ± 5.3%, girls 10.3 ± 2.4 and 15.5 ± 2.8%, first and second occasion, respectively) significantly increased with no sex differences. The data demonstrate that children do display a slow-component response to exercise and are consistent with an age-dependent change in the muscles' potential for O2 utilization. slow component; age; modeling; confidence intervals
O2) kinetic responses to both moderate-intensity exercise [below the anaerobic threshold (AT)] and heavy-intensity exercise (above AT and below critical power) are well documented in adults, but data from children are relatively sparse (6, 34). After the onset of exercise, three phases in O2 dynamics have been identified. The phase 1 or cardiodynamic phase is followed by a primary and rapid exponential process (phase 2), which in the moderate domain achieves a steady state (phase 3) within 
3 min (6, 35). In the heavy-intensity domain, with adult subjects, the third phase is characterized by a delayed rise in O2 (slow component) that causes O2 to increase toward an elevated steady state (9). The mechanisms by which this slow component causes an elevation in the O2 cost of exercise (gain) above that predicted by the subthreshold O2-work rate relationship remain to be confirmed, although it is widely accepted that the additional O2 cost originates from within the exercising muscle (24).
Data from child subjects are conflicting regarding the existence of a slow component of O2 and the effect of age on children's kinetic response to heavy-intensity exercise. Two studies have suggested that the slow component in children is significantly smaller than in adults and that it is associated with lower end-exercise blood lactates. Armon et al. (1) and Williams et al. (36) reported a faster kinetic response to the exercise transition and negligible slow components in children compared with adults during heavy-intensity cycle ergometry and treadmill running, respectively. Both these authors concluded that a single exponential model fit the response profiles of children more appropriately than a model with an additional linear or secondary exponential term. Contrary to this, Obert et al. (21) identified positive slopes in O2 during the third phase of the response to 90% maximal aerobic power. Fawkner and Armstrong (6) exercised prepubertal children at 40% of the difference between AT and peak O2 (40%
) and observed that the response to heavy-intensity exercise in children clearly displayed a rapid primary exponential component, followed by an emerging slow component that projected toward a steady state.
Despite the varying exercise intensities imposed between these studies, there is ambiguity as to whether children do display a slow component and whether in fact it increases in magnitude with age. These contradictions between studies may be due to a number of reasons that need to be addressed. First, cross-sectional studies with small sample numbers rely heavily on selection of subjects of different ages, whereas the most suitable way to address the influence of age on the response variables is to use well-defined subjects and longitudinal analysis. Second, the signal-to-noise ratio in children's response profiles is inherently small and questions the validity of applying mathematical formulas unless a suitable number of exercise transitions have been averaged. Confidence intervals in response parameters need to be taken into consideration but have not previously been reported for children exercising in the heavy-intensity domain. Third, the modeling procedures used, which have varied widely between studies, are known to heavily influence the interpretation of the response data. Finally, it is critical to impose carefully selected relative exercise intensities such that subjects are exercising within a known intensity domain.
The purpose of this study was therefore to test the hypotheses that a slow component exists in prepubertal children and that O2 kinetic responses to heavy-intensity exercise change over a 2-yr period in this group.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Resting measures. Subjects visited the laboratory on at least four occasions over a 2-wk period, and testing took place at approximately the same time of day for each subject. Stature was measured with a Seca 220 stadiometer (Vogel & Halke, Hamburg, Germany) and body mass was determined by use of Seca electronic scales (Vogel & Halke). Sexual maturity was assessed on the first test occasion only by using the indexes of pubic hair described by Tanner (30). All observations were made by the same nurse. Exercise was carried out on the same electronically braked cycle ergometer (Lode Excalibur Sport, Groningen, the Netherlands) with the seat height, handlebar height, and crank length adapted to each child and subsequently maintained throughout that testing period. The Lode ergometer was calibrated according to the manufacturer's recommendations and had a baseline pedaling resistance equivalent to 10 W at 70 rpm.
Measurement of peak O2 and AT.
Peak O2 and AT were determined by a ramp test to voluntary exhaustion. During exercise, gas-exchange variables were measured and displayed online by use of an EX670 mass spectrometer and analysis suite (Morgan Medical, Rainham, UK) that was calibrated according to the manufacturer's instructions. Expired volume was measured by a turbine flowmeter (Interface Associates) with a dead space volume of 90 ml. Volume calibration was achieved by using a handheld calibration syringe (Hans Rudolph, Kansas City, MO) over a range of flow speeds. The sum of the gas-transport and analyzer-response delay terms was determined, and appropriate adjustments were made in the software. All calibration procedures were repeated before each experimental test. Breath-by-breath responses were subsequently interpolated to 1-s intervals.
After a 3-min warm-up of unloaded pedaling, the resistance increased continuously at either 10 or 15 W/min to attain a test 8–10 min in duration. Subjects pedaled at a cadence of 70 ± 5 rpm, and the children were actively encouraged to continue until voluntary exhaustion. Maximal effort was considered to have been given if, in addition to subjective indications of intense effort (e.g., excessive hyperpnea, facial flushing, sweating, discomfort), respiratory exchange ratio reached a value >1.00. All subjects satisfied these criteria. Peak O2 was taken as the highest recorded 10-s stationary average value during the maximal exercise test.
AT was determined noninvasively by the V-slope method as the point at which CO2 production (CO2) began to rise at a more rapid rate than O2 (3, 7). The gas-exchange and ventilatory responses were analyzed with purpose-designed software developed by use of LabVIEW (National Instruments, Newbury, UK). The first 60 s of data after the onset of the exercise and data after the respiratory compensation point were deleted. The respiratory compensation point was recorded as the time at which ventilation (E) began to rise more rapidly than CO2 and was determined subjectively from a plot of E dependent on CO2, with data smoothed by use of a 9-s moving average. By using the data from the selected time period only, V-slope was determined from a plot of CO2 dependent on O2. After smoothing the data using a 9-s moving average, we determined the V-slope by systematically dissecting the CO2/O2 data and plotting linear regression lines using all data from either side of that point. The V-slope was recorded as the point at which the ratio of the largest standard error of the two lines and the distance from the intersection of the two lines to a single regression line drawn through the data set was minimized. The data and resulting regression lines were displayed graphically, and the experimenter visually confirmed the computer-generated selection of the inflection point.
Constant-work rate exercise tests.
On subsequent visits, subjects completed a step-change exercise test that consisted of 6 min of unloaded pedaling, followed instantaneously by a work rate that, from extrapolation from the ramp response, corresponded to 40% of the difference between the O2 at V-slope and peak O2 (40%) for 9 min. A pedal cadence of 70 rpm was maintained throughout. Fingertip blood samples were taken immediately after the end of exercise and assayed for blood lactate concentration by use of a whole-blood automated and self-calibrating analyzer (YSI 2300 Stat Plus, Yellow Springs Instruments, Yellow Springs, OH).
A single transition was completed on each visit, and at least three and in most cases four transitions were completed in total. This was the number of transitions required to obtain 95% confidence intervals in the primary time constant of approximately ±5 s. The 1-s interpolated responses for each individual to each rest-to-exercise transition were time aligned to the start of exercise and averaged together to form a single data set for analysis.
Kinetic analysis.
The duration of phase 1 was estimated from the averaged response profile as the time at which there was a marked inflection point in the O2, CO2, and E response and change in end-tidal O2, end-tidal CO2, and respiratory exchange ratio from baseline values (35). All data before the end of phase 1 were removed from the data set.
Model 1.
A double exponential model (Eq. 1) was applied to the averaged response file, and parameters and their 95% confidence intervals were estimated by least squares nonlinear regression analysis (GraphPad Prism, GraphPad Software, San Diego, CA)
 | (1) |
O2(t) is the increase in O2 at time t above the prior control level, which was calculated as the mean O2 from the last minute of baseline pedaling; A1 and A2, 
1 and 2, and 
1 and 2 are the amplitudes, time constants, and independent time delays of each exponential, respectively.
Model 2.
Model 2 was designed to remove the possible influence of arbitrarily parametizing the slow component on the dependent parameters of the primary component. By using a purpose-designed software program developed with LabVIEW (National Instruments, Newbury, UK), a single exponential with a delay term was fit to data after the end of phase 1 (Eq. 2). The fitting window was iteratively widened by 1-s intervals, starting from a 60-s fitting window and finishing with a fitting window that encompassed the entire data set (27).
 | (2) |
O2(t) is the increase in O2 at time t above the prior control level, which was calculated as the mean O2 from the last minute of baseline pedaling; A1, 1, and 1 are the amplitude, time constant, and time delay.
The estimated time constant (1) for each fitting window was plotted against time. This was to allow for the beginning of the slow component to be determined through visual inspection and therefore to identify the optimal fitting window with which to estimate the parameters of the primary component. The onset of the slow component was defined as the point at which a plateau in the estimated 1 was followed by a progressive increase in the estimated 1 (Fig. 1).
|
1), delay term (1), and amplitude (A1) and their 95% confidence intervals were subsequently estimated using the predetermined fitting window (GraphPad Prism, GraphPad Software, San Diego, CA).
In the case of both models 1 and 2, the confidence interval for A1 was considered as a percentage of A1. The amplitude of the slow component (A2) was calculated as the difference in the mean of the O2 amplitude over the last 30 s of exercise (O2tot) and A1. The slow component was expressed in relative terms as the percentage contribution of A2 to O2tot. The functional primary and secondary gain (G1 and G2, respectively) were expressed as the O2 relative to the change in work rate (G1, A1/W; G2, O2tot/W).
Statistical analyses.
Peak O2 was expressed relative to body mass as a ratio (ml·kg–1·min–1) and as a power function ratio (ml·kg–b·min–1). The b exponent represents the gradient of the loge peak O2 (l/min) – loge body mass (kg) relationship and was derived by log-linear analysis of covariance. Repeated-measures ANOVA was used to investigate differences in estimated response parameters between the two models and to test for differences in the slow component after 6 and 9 min of exercise. Repeated-measures ANOVA was performed to identify significant changes in response measures between the two test occasions (independent variable) and any sex interaction (between-subjects variable). Sex differences in anthropometric and exercise responses were assessed with ANOVAs on each test occasion. Correlation coefficients were used to investigate relationships between response variables. Statistical significance was set at the P < 0.05 level.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 identified a common b exponent for the whole group (0.71, SE 0.09). There was a significant increase in all physical characteristics and responses to the ramp test with age apart from peak O2 relative to body mass (ml·kg–1·min–1), peak blood lactate, and V-slope (as a percentage of peak O2). Significant sex differences were found on both test occasions in peak O2 expressed relative to body mass (ml·kg–1·min–1) and as a power function ratio (ml·kg–0.71·min–1), and in peak O2 in absolute terms (l/min) on the first occasion only. The boys were significantly (P < 0.01) younger than the girls on the first test occasion. The only interaction between test occasion and sex was found for stature.
|
1, A2 and 2 were 3.5 ± 1.7%, 4.3 ± 1.5 s, 44 ± 51%, 188 ± 181 s and 4.2 ± 2.9%, 6.4 ± 3.3 s, 40 ± 21%, 141 ± 129 s for test occasions 1 and 2, respectively. Table 3 presents the model parameters using model 2. The mean ± SD confidence intervals for A1 and 1 were 2.7 ± 1.0%, 4.2 ± 1.4 s and 2.9 ± 1.0%, 5.5 ± 2.3 s for test occasions 1 and 2, respectively. There was no significant difference in A1 or 1 between models 1 and 2 and no significant difference between the determination of the onset of the slow component between each analysis method.
|
|
1 and G2, and there was a significant reduction in G1 (Figs. 2 and 3). The contribution of the O2 slow component to the total change in O2 after 6 min was boys, 8.4 ± 7.4, 12.3 ± 5.2; and girls, 7.8 ± 3.3, 13.2 ± 4.0%, for test occasions 1 and 2, respectively. The relative magnitude of the slow component was significantly greater after 9 min of exercise than after 6 min on both test occasions. There was no sex interaction or sex difference in any of the variables. Correlation analysis revealed that there was no relationship between the slow component or the gains and peak O2 or end-exercise blood lactate.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 is apparent in prepubertal children, contributing 10% of the final O2 after 9 min of exercise, and that the response to heavy-intensity exercise undergoes changes over the following 2 yr. This therefore conflicts with previous studies that have suggested that the slow component is negligible in children but supports the view that the magnitude of the slow component is age dependent (1, 36).
Methodological considerations.
This study has taken into consideration the 95% confidence intervals with which the nonlinear regression procedure was able to estimate the model parameters. The parameters derived by using the double exponential model (model 1) were not significantly different from those derived by the alternative approach (model 2). However, the wide confidence intervals identified for 2 and A2 identify the concern with applying model 1, especially with response patterns that have an inherently low signal-to-noise ratio (i.e., with children). As well, the properties of the nonlinear regression process are such that there is dependence between estimated parameters, and thus it is not advisable to force a model to define data if the physiological properties may not match the model design (such as may be the case when applying an exponential to the slow component). These concerns are illustrated by examining the 95% confidence intervals of the primary component for models 1 and 2. Confidence intervals are estimated by using the error of the entire data set, and the inclusion of "noisy" data during the slow component in model 1 inevitably resulted in wider 95% confidence intervals of the parameters of the primary response (1 and A1). This was not the case in model 2, in which the modeling process focused on data with known exponential properties.
Why we have clearly identified a slow component in children whereas previous authors (1, 36) have not is not readily apparent but might be due to methodological issues. With a group of boys and girls (10.0 ± 2.2 yr), Armon et al. (1) concluded that a single exponential was sufficient to describe children's O2 responses both above and below AT. This was despite finding a positive linear term in 73% of the children's responses above AT and positive slopes of the regression of O2 between the 3rd and 6th minute of exercise in 50% of the children exercising at 50% (mean 0.27 ± 0.73 ml·kg–1·min–2). To investigate whether the conflicting results could be explained by the length of the exercise periods, we analyzed the slope of the linear regression between the 3rd and 6th minute with the present data. For both the boys and girls collectively, on test occasion 1 when the mean age was similar to the children in the study of Armon et al., the slope was substantially greater than that previously reported (0.54 ± 0.38 ml·kg–1·min–2). The slope increased to 0.70 ± 0.31 ml·kg–1·min–2 on test occasion 2.
Williams et al. (36) reported that, after 6 min of treadmill running, the slow component response in eight 11- to 12-yr-old boys contributed 0.9 ± 1.2% to the total change in O2 and that a single exponential model could suitably describe the responses. Our analysis revealed a substantially greater contribution in the present data after 6 min than was reported by Williams et al. and that the relative contribution of the slow component to the total change in O2 was significantly greater after 9 min of exercise. Although it is apparent that terminating the exercise test after only 6 min excludes some development of the slow component, conflicting conclusions drawn from the present data and from previous studies cannot be explained fully by differences in the period of exercise. Other methodological variance between the studies may be influential, including the ergometer used and the imposed relative exercise intensities.
Phase 1.
The shorter phase 1 reported presently on the first test occasion supports the work by Hebestreit et al. (14), who reported that a shorter phase 1 (15.3 ± 8.5 s) was found in 9- to 12-yr-old boys compared with men (22.5 ± 6.3 s) exercising at 50% peak O2. Nevertheless, caution must be taken in interpreting the duration of phase 1 visually because of the limited amount of available data and the breath-by-breath noise that invariably masks any clear changes in ventilatory variables. Further investigation into the properties of phase 1 with children may prove valuable but will require more rigorous technologies and procedures in its determination than have been used previously.
Primary component.
The higher O2 cost of work during the primary component displayed by the younger children is supportive of previous studies that have identified a higher O2 cost of exercise above AT in children compared with adults. Zanconato et al. (38) reported that, after 1 min of cycling at 50%, the mean O2 cost in 10 children (7–11 yr) was 10.9 ± 2.1 ml O2·min–1·W–1 compared with 7.4 ± 1.2 ml O2·min–1·W–1 in 13 adults (26–42 yr). Hebestreit et al. (14) observed that after 2 min cycling at 100% peak O2 the O2 cost was 10.4 ± 1.4 and 8.3 ± 1.0 ml O2·min–1·W–1 in children and adults, respectively. Exercising at 50%, the children in the study of Armon et al. (1) demonstrated a mean O2 cost after 6 min of cycling of 11.47 ± 1.71 ml O2·min–1·W–1 compared with 9.90 ± 0.71 ml O2·min–1·W–1 in the adults. These authors noted also that after 6 min of cycling at 75% the O2 cost of exercise after 3 min was significantly higher in the children than the adults but that there was no significant difference by the end of the exercise period (12.56 ± 1.33 vs. 11.4 ± 1.30 ml O2·min–1·W–1 in children and adults, respectively). Similarly, in the present study, we found a greater primary gain on the first test occasion but no difference between the two occasions in the gain at the end of the exercise period. This is also in accord with the results of Williams et al. (36), who compared boys with men, although in this latter study the O2 was reported in relation to body mass rather than power output, which confounds further interpretation.
The greater O2 cost of exercise in younger subjects during the dynamic response has provided other authors with the notion either that the aerobic capacity for exercise is enhanced in children or that they have a lesser ability to generate energy anaerobically (1, 36, 38). However, if the control theories regarding the dynamics of O2 at the onset of exercise are correct, the primary response is considered to depend predominantly on the mitochondrial potential to generate the required ATP for exercise (32). We observed a faster primary time constant (1) response when the children were younger, which is consistent with studies that have reported a faster kinetic adjustment to exercise above AT in children compared with adults (1, 19, 20, 26, 28, 36). Our laboratory has previously reported a faster 1 in children exercising at moderate-intensity exercise compared with adults (8). Together, the data suggest that a developmental influence on the O2 utilization potential, possibly a function of mitochondrial enzyme activation or intracellular concentrations of putative metabolic controllers (12, 31, 33), presides over a reduced glycolytic potential. Although there is some evidence that adolescents may have an enzyme profile supportive of a greater rate of pyruvate oxidation than adults (13), there are to the authors' knowledge no data monitoring changes in enzyme profiles from prepubescence through to adolescence or adulthood. Of significance though is that these responses are characteristic of subjects who differ in the fiber-type profile of the muscle. A greater O2 cost (2, 25) and faster primary component time constant (25) have been reported in adult subjects with a high ratio of type I to type II muscle fibers. There is presently limited evidence to suggest that the fiber-type profile of the muscle changes during growth and maturation. Some studies have indicated that the proportion of type I fibers undergoes no change from childhood through to adulthood (4, 22), whereas Lexell et al. (17) have identified a reduction in the percentage of type I fibers in subjects ranging from age 5 (69%) to age 15 (58%) and age 35 (45%).
There is also evidence that O2 delivery (muscle blood flow per unit tissue) may decrease from the ages of 12 to 14 yr (15). The extent to which O2 delivery limits the phase 2 response to heavy-intensity exercise is clouded by contradictory literature (5, 10, 11, 16, 18, 29). Therefore, whether these results are indicative of a change in the fiber-type profile of the muscle after the prepubertal period and/or the properties of those fibers and delivery of O2 remains to be elicited. It should be noted, however, that although this is the first study to report the confidence intervals for 1 with children during heavy-intensity exercise, we do not suggest that these are suitably tight to draw any robust conclusions.
O2 slow component.
A greater potential for O2 utilization or O2 delivery during the primary phase in the subjects on the first test occasion is consistent with their subsequent response characteristics and the magnitude of the slow component of O2. Although there have been a number of possible explanations for the slow component [see Gaesser and Poole (9) for a discussion], work by Poole et al. (23) has indicated that 86% of the additional O2 originates from the exercising muscle, and therefore it is probable that the changes observed reside here. It is likely that this greater O2 cost of exercise manifested as the slow component is likely to be due to an increase in the phosphate cost of generating muscular force rather than the O2 cost of phosphate production (27). This may result from the recruitment of low oxidatively efficient fibers (type II) or the progressive recruitment of a greater number of high oxidatively efficient fibers. The smaller slow component that was demonstrated in the younger years resulted in an elevated final O2 gain that was equal on both test occasions, and this correlates closely with the pattern of comparison between high and low percentages of type I fibers with adults (2, 25). Hence changes in fiber-type recruitment may have contributed to the increases in the amplitude of the slow component with age.
It has been suggested that the primary amplitude represents the projected O2 for the exercise task, and the slow component reflects a metabolic process that is additive (29). However, contrary to this, the O2 cost by the end of the exercise period on both test occasions was equal, as has been found previously with adult experimental groups (2, 16, 25) and between children and adults (1, 36). This may suggest that the phosphate turnover required to maintain heavy-intensity exercise was in fact independent of age and that in the older children a lesser proportion of the required O2 for the given exercise intensity was achieved in the primary phase. In this sense, the slow component acts as a "catch-up" mechanism during which equilibrium in O2 delivery and fiber recruitment and efficiency is achieved.
Such a system may suggest that the higher blood lactates achieved in the older children are indicative of generating a lesser proportion of the required O2 aerobically during the primary phase and presumably during this catch-up phase. However, because there was no correlation between end-exercise blood lactates and the gain of either the primary phase or the slow component, the relationship between these response variables and blood lactate remains elusive, as has been indicated previously with both children and adults (21, 36, 37).
We found that, consistent with the literature, the peak O2 of children increased over the 2-yr interval, and the magnitude of the slow component also increased. There were significant sex differences in peak O2 when scaled to body mass on each occasion, but no sex difference in the slow component or any relationship between peak O2 and the slow component. These results support the work of Obert et al. (21), who reported that there was no difference in the magnitude of the slow component (after 90% maximal aerobic power until exhaustion) between 12 trained and 12 untrained children. Hence it seems that a relationship between the slow component and peak O2 is tenuous in children.
Interestingly, we found no significant sex differences in any of the response parameters to the heavy-intensity exercise. This was surprising, because earlier work from our laboratory identified longer primary time constants and larger slow components in girls compared with boys (6). Although the present data indicate sex differences in these directions, it is likely that the small sample sizes and large standard deviations of the response variables precluded statistical significance. Classification of sex differences in the response to heavy-intensity exercise will require further investigation with larger sample sizes and stringent methodological rigor.
Unfortunately, although we were able to assess pubic hair at the first test occasion, because of the prevailing sociological climate regarding such screening we were unable to repeat this procedure on the subsequent occasion. The interaction between test occasion and sex for stature suggests that some of the girls had initiated their growth spurt by test occasion 2, but this does not appear to have impacted on any of the measured responses to heavy-intensity exercise, which did not evidence any sex interaction on either occasion. It is possible that maturational changes have an independent influence over and above chronological age, but further investigation is restricted by the present difficulty in the rigorous assessment of changes in maturational status.
In conclusion, this is the first study that has attempted to investigate longitudinally the influence of age on the slow component and kinetic response to exercise in children. We have clearly identified that a slow component does exist in both male and female prepubertal children and that its magnitude increases over the subsequent 2-yr period independent of peak O2. The increase in the slow component is associated with a reduction in the relative amplitude of the primary phase and a slowing of the primary kinetics, and the entire response profile appears to project toward a predetermined O2 cost of the relative exercise intensity independent of age. These results are consistent with an age-dependent change in the muscles' potential for O2 utilization.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 on-kinetics in isolated insitu canine muscle. J Appl Physiol 85: 1404–1412, 1998.
O2 on-kinetics in canine muscle contracting at peak O2. J Appl Physiol 89: 1293–1301, 2000.
O2 kinetics in heavy submaximal exercise by hyperoxia and prior high-intensity exercise. J Appl Physiol 83: 1318–1325, 1997.
O2. Med Sci Sports Exerc 26: 1335–1340, 1994.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
