Oxygen
consumption
J, M. HAGBERG, J. P. MULLIN, Departments of Physical Education University of Wisconsin, Madison,
during
constant-load
exercise
AND F. J. NAGLE and Physiology, Wisconsin 53706
HAGBERGJ. M.,J. P, MULLIN, AND F. J, NAGLE. Oxygen consumption during constant-load exercise. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 4513): 381-384, 1978. Previous investigators have reported that oxygen consumption (vo,) continues to rise after the initial Z- to 3-min transient period of exercise when work exceeds approximately 60% of was to examine the vo* n-lax*The purpose of this investigation possible causes of this slow rise in $70~. Eighteen subjects exercised for 20 min at 65% and at 80% of VO, MaX on the bicycle ergometer. vo,, ventilation (VE), and respiratory exchange ratio were monitored by a continuous computerbased system. Blood lactate concentration and rectal temperatures were measured at 2- to 3-min intervals during the exercise. VO, increased significantly from the 5th to 20th min of exercise in 81% of the tests at both levels of work intensity. The magnitude of the rise was not different for the two work loads. No evidence was found to support the lactacid explanation proposed for this rise. Increased temperature could account for 30% of the rise; the estimated cost of increased VE could account for 30 and 81% of the rise at the two work loads. The sum of these factors could account for 60 and 111% of the rise in vo, at the 65 and 80% of vo, max work loads.
ranged from 39.7 to 60.7 ml. kg+ min-l with a mean of 51.9 -+ 1.4 ml wkg+ min? All subjects were scheduled to complete two ZO-min exercise bouts on a bicycle ergometer at work loads chosen to elicit 65 and 80% of VO, maxfor each subject. A number of the exercise tests had to be eliminated from the analysis because of technical difficulties. This left a total of 25 complete tests for the final analysis. The missing data included the Vo2 data from two runs, the lactate data from three runs, and the temperature data from three runs. In addition, three subjects were unable to complete the 20-min test at 80% of VO, max* Blood samples were drawn through a Teflon catheter placed in an antecubital vein at rest and after 1, 3, 5, 8, 11, 14, 17, and 20 min of exercise. Blood samples were immediately pipetted into chilled trichloroacetic acid (TCA) solution and later analyzed for lactate concentration by the method of Barker and Summerson (2). A Yellow Springs telethermometer was used to monitor core temperature at the time each blood sample was taken. oxygen uptake; oxygen uptake kinetics; ventilatory control; voZ was monitored continuously on a breath-bycost of ventilation; rectal temperature; blood lactate breath basis throughout the exercise. The computerbased system was similar in design to that previously described (5) and utilized a PDP-12 digital laboratory computer, Beckman OM-11 and LB-2 rapid response gas A NUMBER OF INVESTIGATORS (8, 18, 23) have reported analyzers, and an expiratory pneumotachometer atthat oxygen uptake (VQ) continues to increase after the initial 3 min of exercise at work rates requiring more tached to a Statham differential pressure transducer. The pneumotachometer was positioned approximately 6 than 60% of vo, max. The slow rise of vo, after the ft from the breathing valve (5). Gas samples were initial minutes of exercise has been attributed by Henry continuously drawn to the analyzers at a rate of 200 ml/ (13) and Volk ov et al, (22) to the removal of blood min through narrow-gauge tubing connected directly to lactate. However, it is well known that other, potenthe mouthpiece; expired flow was corrected for this tially calorigenic, factors, are also changing during continous sampling rate. i70, was calculated as delong-term exercise. The purpose of this study was to scribed by Beaver et al. (5) and included a correction for investigate the possible role of substrate utilization, expired water vapor dilution of the respiratory gas rectal temperature, and ventilation in causing the slow fractions (4). rise of vo, during constant-load exercise in addition to The vo2 data points for the entire run were fitted to indirectly investigating the possibility of an explanation the two-component exponential model described by based on the metabolism of lactate. Henry (13) and Volkov et al. (22) using an iterative nonlinear regression technique. The regression data METHODS concerning the second component, which is slower and Eighteen male subjects, aged 20-33 yr, volunteered to quantitatively smaller than the first component, were take part in the experiment. All signed informed conutilized to describe the time course of the Voz beyond sent statements after the design of the study and its the initial minutes of exercise. Also, the raw VO, data potential hazards were outlined. They first completed a were averaged in the 5th and 20th min of exercise and continuous incremental bicycle ergometer test to assess compared using a t test to determine if the second, slow their voZ max; the initial load ranged from 360 to 720 component of the rise in ire, was statistically evident. kpmimin and was increased by 180 kpmimin until Expired volume (VE) and respiratory exchange ratio subjective exhaustion occurred (11). vo, max values (RER) were also averaged in the 5th and 20th min to OOZl-8987/78/0000-0000$01.25
Copyright
0 1978 the American
Physiological
l
l
Society
381
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
382
HAGBERG,
determine the change in these variables during the exercise. The oxygen cost of the VE values in the 5th and 20th min were calculated from the data of Cournand et al. (9), Data concerning the oxygen cost of ventilation are highly variable; however, the data of Cournand et al. (9) were used since they represent the median of 10 studies reviewed by Shephard (20). The cost of the change in RER was calculated from the table of Zuntz presented by Swift and French (21); their table reveals that a 0.01 decrease in RER would necessitate a 0.25% increase in 00, to maintain the same energy production. The approximate change in vo, that would result from a known change in rectal temperature was calculated from the data of Grimby (12) who found a 5.5% increase in vo2 during exercise for a 1.3”C increase in rectal temperature.
MULLIN,
AND
NAGLE
TABLE 2. Blood lactate and temperature during constant-load work Relative Work Load (% b ma,>
n
Freexercise Lactate, mg1100 ml
Lactate at 8 min, mg1100 ml
65 80
13 12
10.0 ?I 0.9 15.1 + 1.9
26.9 * 3.3 50.8 2 4.6
ALA
I
mg,lOO ml
-0.3 -t- 1.7 15.3 IL 2.9
ATemp,
“C
0.5 5 0.1 0.8 -t 0.1
All values are means + SE. ATemp value is the difference in that variable between 5th and 20th min of exercise; ALA is the difference between 8th and 20th min.
a 7
‘400
I’
El
Portion
not
Portion
incrsarsd
occountsd cost
accounted
Portion Increased
accountad tmmperature
of
for for by ventilation for
by
Erceu
accounted
for
RESULTS
A statistically significant (P < 0.05) rise in vo, from the 5th to 20th min of exercise occurred in 76% of the tests at 65% of vo 2 max and 85% of the tests at 80% of 2 max (Table 1); however, the magnitude of the rise at these two work loads was not significantly different. The time course of the second component of exercise vo2 at the two work loads was not significantly different. The average half time of the slow rise in vo, was 3,7 min. The maximum blood lactate level during exercise at 65% of VO, max was 28.7 t 3.3 mg/lOO ml (n = 13, mean t SE). Blood lactate concentration remained essentially unchanged after the 8th min of exercise (Fig. 1). The mean lactate concentrations for exercise tests l
vo
TABLE 1. Work for constant-load Relative Work Load (% 90, ,&
65 80
load and Ih2 data exercise Absolute Work Load, kpm/min
IE
13 12
1,005 1,299
k 61 k 79
vo, at 5 min I * min-’
Avo2*
ml 0min-’
2.25 * 0.21 2.98 2 0.29
184 k 36 330 -c 63
Values are expressed as means + SE. * Avo, in vo, between 5- and ZO-min exercise value.
is the difference
80r &Q 13, -E
Stop Exercise
Rest
0
5 TIME
FIG, vO2
max.
1. Blood
lactate
during
10 (Minutes) 20 min
of exercise
15
20 at 65 and 80% of
0
80 RELATIVE WORKLOAD (% of i/o, max) 45
FIG. 2, Causes of rise in %, of exercise at 65 and 80% of vo,
observed max.
between
5th and 20th
min
requiring 80% of ‘cio2 max are presented in Fig. 1. The subjects had average maximum lactate values of 66.1 -+ 6.8 mg/lOO ml; an increase in lactate concentration of 15.3 -t- 2.9 mg/lOO ml occurred between min 8 and20 of exercise (Table 2). Sixteen runs (5 at 65% and 11 at 80% of vo2 max> exhibited maximum blood lactate levels in excess of 30 mg/lOO ml and were included in the analysis of the relationship of the change in lactate to the change in vo2. The correlation between these two variables was 0,48 (P < 0.05) in these 16 runs; however, in the 9 remaining tests where lactate was only slightly elevated above resting levels, the rise in vo, was still evident. The change in rectal temperature could account for 31% of the rise in vo2 at both work loads (Fig. 2). Ventilation increased between the 5th and 20th min of exercise at both work loads (Table 3). The estimated cost of this increased ventilatory work could account for 30% of the rise in VO, at 65% of VO, maxand 81% of the rise Of 80% Of V02maxm It is interesting to note that at the heavier work load, where a marked hyperventilation occurred, the increase in ventilatory work could account for a major portion of the increase in vo2. The increase in ventilation at both loads was brought about by an increase in breathing frequency (Table 3). Paired t tests revealed no significant differences between the tidal volumes at the 5- and 20-min point at each work
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
%bp DURING
CONSTANT-LOAD
TABLE 3. Ventilatory changes constant-load exercise Relative Work Load 9% mmlax
VT,
liters
EXERCISE
383
during
served. In fact, the half time of the response reveals that more than 97% of the slow rise in VO, had already occurred by the 20th min of exercise. The cause of the observed rise in 00, during constantload work is unknown. Henry (13) and Volkov et al. (22) both theorized that it was due to an oxidative removal of lactate from the blood. Our results, thou .ghindirect, offer two refutations to this “metabolism of lactate” explanation offered for the slow rise in Vo, observed during constant-load exercise. First, though there was a small, but significant, correlation between the change of 00, and blood lactate in tests which elicited blood lactate concentrations in excess of twice the resting level, the rise in voZ was also evident at work rates where virtually no change in blood lactate was observed from rest to exercise. Second, at 80% of Vo2 max, all of the rise in iToZ was accounted for by the increases in Thus, our data support temperature and ventilation. the contention th .at the slow rise of VoZ in exercise appears not to be solely a “lactacid” phenomenon (1, 3); in fact, little experimental evidence exists to assign any role to such a mechanism. One calorigenic factor involved in causing the increase in Vo, during constant-load exercise is the direct effect of temperature. The observed increases in rectal temperature could account for a 57 and 103 ml/min rise in VO, at the 65 and 80% of iTo maxwork loads. Both values are approximately 3% of the total vo, in 1 min-I, implying that the metabolic effect of the temperature rise is proportional to the energy utilization, and hence heat release, of the subjects. Ventilation increased from the 5th to 20th min of exercise at both work loads, Other authors (7, 8, 18) have noted increases in ventilation during constantload exercise similar to those in the present investigation. Our results concerning the cost of the increased ventilation should probably be interpreted qualitatively, rather than quantitatively, since the oxygen cost of increased ventilation has been shown to be highly variable (20), and in fact, may be due to cellular alkalosis rather than increased work of the respiratory muscles (6, 15). Our data reveal that one-third of the rise in VO, at 65% of VO, max can be attributed to increased ventilation. The ventilatory cost, however, can account for four-fifths of the i70, rise at a heavier work load where a more marked hyperventilation occurs. Our data support the fmdings of Katch et al. (16) who found that only a portion of the rise in ire, during constant-load exercise could be attributed to the cost of increased ventilation. The rise in ventilation during this period is interesting from another standpoint. Though VE is increasing, which will inherently decrease exercise efficiency because of the additional work of the respiratory muscles, an attempt to limit the decrease in efficiency is made by causing an in crease in breathing frequency rather than an increased tidal volume. It ha.s been shown that an increase in frequency is the more efficient method, in terms of oxygen cost and mechanical work, to increase ventilation (9). The cause of this increased ventilation that occurs after the initial minutes of constant-load exercise is
f’
STPD
20 min
ALE, llmin
ARER
5 min
27.5 + 1.8 29.9 k 1.7
31.6 2 1.8 38.9 -t 2.5
11.4 k 1.9 33.8 + 7.9
0.00 t 0.01 - 0.02 2 0.03
n 5 min
65
13
80
12
20
2.11 IL 0.17 2.33 2 0.20
Values are expressed between 5- and 20-min
min
2.20 2 0.16 2.66 IL 0.18
as means k SE, ALE is the difference exercise value,
in
VE
TABLE 4. Calculated PACT, values at 5 and 20 min of exercise Relative
5-min ZO-min Values equations
PkO, P&,
are means of Otis (19).
f
SE,
Work
Load
65% vo2 max
80% vo, max
41.9 + 1.0 39.6 + 1.3
38.7 * 1.5 32.3 k 1.9
expressed
in
Tory
calculated
from
load, though the tendency for a slight increase was evident. However, breathing frequency increased by 14% from the 5th to 20th min of exercise at 65% of 00, maxand by 30% at 80% of VOW max. The average change in RER from the 5th to the 20th min of exercise was 0 and -0.02 at the 65 and 80% of VOW max work loads. However, it must be pointed out that we measured RER at the lungs which is influenced by hyper- or hypoventilation and thus may not be indicative of respiratory quotient (RQ) at the muscle in either of these situations, The change in substrate utilization at 65% of vo 2maxwould have been evident from the change in RER because there was very little change in PkO, between the 5th and 20th min of exercise (Table 4). However, at 80% of VO, maxa larger change in substrate utilization would have been evident, if a progresqive hyperventilation had not occurred between the 5th and 20th min of exercise (Table 4). The observed changes in RER could account for only a 0.5% rise in VO, at the 80% of Vo2 maxwork load. DISCUSSION
The slow rise of Vo2 during constant-load work was evident at both work loads, 65 and 80% of iTo maxi investigated in this study. The magnitude of the rise in vo2 from the 5th to the 20th min of exercise is similar, at the heavier work load, to those reported by Nagle et al. (18) for treadmill running at 82-89% of vo2 maxaThis slow rise in vo2 from the 5th to 20th min of exercise must be caused by changes occurring after the 5th min of exercise and must not be caused directly by the work load itself because the VoZ value in the 5th min of exercise is essentially that which would be predicted for that absolute work load on a bicycle ergometer (Table 1). The mathematical models of vo, in exercise (13, 22) predict that the observed rise in Voe after the initial minutes of exercise is an exponential function of time. Thus the changes in Vo, during successive 3,7-min periods will decrease until virtually no change is ob-
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
384 unknown. The ventilation is increased beyond that required to maintain arterial Pcoz homeostasis; as shown in Table 4, the result is an arterial hypocapnia. One ventilatory control factor often implicated in mediating this hypocapnic ventilation is an increased temperature. Dempsey and Reddan (10) have demonstrated that abolishing the rise in core temperature during exercise also eliminated the hypocapnic hyperventilation. The correlation between the increase in rectal temperature and the rise in ventilation during the last 15 min of exercise was 0.59 (P < 0.001) in the present investigation, which would lend support to the hypothesis that temperature may play a predominant role in altering ventilation during long-term exercise. The change in substrate utilization indicated by the RER, as measured in the expired air, contributed very little to the elevation of vo, during the exercise. A portion of the decrease in tissue RQ, which would indicate a switch to fatty acids as the carbon source for the TCA cycle, was masked by a hyperventilation at the lung. Thus, if the RQ could have been measured at the tissue level, a much larger increase in Vo, could have been attributed to the shift to an energetically less efficient oxidation of free fatty acids. The total calculated effect of the three calorigenic factors, ventilation, temperature, and respiratory quotient, could account for 60% of the rise in vo2 during exercise requiring 65% of vo, maxand for essentially all (actually 111%) of the rise in VO, at 80% of VO, max. Thus, it seemsthat at a lighter work load, 65% of Vo, max in this study, an as yet unknown factor must account
HAGBERG,
MULLIN,
AND
NAGLE
for much of the increase in ire,. Various other factors (circulating catecholamines and mechanical efficiency) may also contribute to the increase in Vo2 during the exercise. The effect on Vo, during long-term work of these two, possibly calorigenic, factors can not be ascertained, either qualitatively or quantitatively, at this time. The net result of the data, concerning the rise in Vo,, VE, and temperature and the slight decline in RER at the work loads investigated, implies that what people have called the “steady state” is really a period of nonsteady-state exercise. Even though the changes in these variables are relatively small when compared to those that occur in the initial minutes of exercise, their absolute magnitude is such that they can hardly be considered invariable with time. Our data indicate that the cause of the slow increase in vo, is primarily a temperature effect, both directly, as evidenced by a QIO effect, and indirectly, by causing a slight hyperventilation which increases the vo2 of the respiratory muscles. The authors thank Michael Clark and James Carlson of the University of Wisconsin Laboratory Computer Facility for their assistance during the development of the computer-based voe system. Appreciation is also expressed to Dr. M. L. Birnbaum for the use of his laboratory facilities. A preliminary report of this investigation was presented at Chicago, Illinois at the meeting of the Federation of American Societies for Experimental Biology in April, 1977. Received
5 August
1977; accepted
in final
form
19 April
1978.
REFERENCES 1. ALPERT, N. R., AND N. S. ROOT. Relationship between excess respiratory metabolism and utilization of intravenously infused sodium racemic lactate and sodium L (-> lactate. Am. J. Physiol. 177: 455-462, 1954. 2. BARKER, S. B., AND W. H. SUMMERSON. The calorimetric determination of lactic acid in biological materials. J. Biol. Chem. 138: 535-537, 1941. 3. BARNARD, R. J., M. L. Foss, AND C. M. TIPTON. Oxygen debt: involvement of the Cori cycle. Intern. 2. Angew. Physiol. 28: 105-119,197o. 4. BEAVER, W. L. Water vapor corrections in oxygen consumption calculations, J. AppZ. Physiol. 35: 928-931, 1973. 5. BEAVER, W. L., B. J. WHIPP, AND K. WASSERMAN. On-line computer analysis and breath-by-breath graphical display of exercise function tests. J. Appl. Physiol. 34: 128-132, 1973. 6. CAIN, S. M. Increased oxygen uptake with passive hyperventilation of dogs. J. AppZ. PhysioZ. 28: 4-7, 1970. 7. CLAREMONT, A. D., F. J. NAGLE, W. D. REDDAN, AND G. A. BROOKS. Comparison of metabolic, temperature, heart rate, and ventilatory responses to exercise at extreme ambient temperatures. Med. Sci. Sports 7: 150-154, 1975, 8. COSTILL, D. L. Metabolic responses during distance running. J. AppZ. Physiol. 28: 251-255, 1970. 9. COURNAND, A., D. W. RICHARDS, R. A, BADER, M. E. BADER, AND A. P. FISHMAN. The oxygen cost of breathing. Trans. Assoc. Am. Physicians 67: 162-173, 1954. 10. DEMPSEY, J. A., AND W. G, REDDAN. Hyperventilation during prolonged work: cause and effect. Med. Sci. Sports 8: 62, 1976. If. DILL, D. B. Assessment of work performance. J. Sports Med. Phys. Fitness 6: 3-8, 1966. 12. GRIMBY, G. Exercise in man during pyrogen-induced fever. Acta Physiol. Stand. SuppZ. 67: l-114, 1962. 13. HENRY, F. M. Aerobic oxygen consumption and alactic debt in
muscular work. J. AppZ. Physiol. 3: 427-438, 1951. 14. HENRY, J. G., AND C. R. BAINTON. Human core temperature increase as a stimulus to breathing during moderate exercise. Respiration Physiol. 21: 183-191, 1974. 15. KARETZKY, M. S., AND S. M. CAIN. Effect of carbon dioxide on oxygen uptake during hyperventilation in normal man. J. Appl. PhysioZ. 28: 8-12, 1970, 16. KATCH, F. I., R. N. GXRANDOLA, AND F. M. HENRY. The influence of the estimated cost of ventilation on oxygen deficit and recovery oxygen intake for moderately heavy bicycle ergometer exercise. Med. Sci. Sports 4: 71-76, 1972, 17. KAYNE, H. L., AND N. P. ALPERT, Oxygen consumption, following exercise in anesthetized dogs. Am. J. PhysioZ. 206: 51-56, 1964. 18. NAGLE, F. J., D. ROBINHOLD, E. HOWLEY, J. DANIELS, G. BAPTISTA, AND K. STOEDEFALKE. Lactic acid accumulation during running at submaximal aerobic demands. Med. Sci. Sports 2: 182-186, 1970. 19. OTIS, A. B. Quantitative relationships in steady-state gas exchange. In: Handbook of Physiology. Respiration. Washington, D.C.: Am. Physiol. Sot., 1964, sect. 3, vol. I, chapt. 27, p. 681698. 20. SHEPHARD, R. J. The oxygen cost of breathing during vigorous exercise. Quart. J. ExptZ. Physiol. 51: 336-350, 1966. 21. SWIFT, R. H., AND C. E. FRENCH. Energy Metabolism and Nutrition. Washington, D.C.: Scarecrow, 1954. 22. VOLKOV, N. I., V. N. CHEREMISINOV, AND E. N. RAZUMOVSKII. Oxygen exchange in man during muscular activity. In: The Oxygen Regime of the Organism and Its Regulation, edited by N. V. Lauer and A. 2. Kolchinskaya. NASA Technical Translation, 1969. 23. WHIPP, B. J., AND K. WASSERMAN. Oxygen uptake kinetics for various intensities of constant-load work. J. AppZ. Physiol. 33: 351-356. 1972.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
consumption
J, M. HAGBERG, J. P. MULLIN, Departments of Physical Education University of Wisconsin, Madison,
during
constant-load
exercise
AND F. J. NAGLE and Physiology, Wisconsin 53706
HAGBERGJ. M.,J. P, MULLIN, AND F. J, NAGLE. Oxygen consumption during constant-load exercise. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 4513): 381-384, 1978. Previous investigators have reported that oxygen consumption (vo,) continues to rise after the initial Z- to 3-min transient period of exercise when work exceeds approximately 60% of was to examine the vo* n-lax*The purpose of this investigation possible causes of this slow rise in $70~. Eighteen subjects exercised for 20 min at 65% and at 80% of VO, MaX on the bicycle ergometer. vo,, ventilation (VE), and respiratory exchange ratio were monitored by a continuous computerbased system. Blood lactate concentration and rectal temperatures were measured at 2- to 3-min intervals during the exercise. VO, increased significantly from the 5th to 20th min of exercise in 81% of the tests at both levels of work intensity. The magnitude of the rise was not different for the two work loads. No evidence was found to support the lactacid explanation proposed for this rise. Increased temperature could account for 30% of the rise; the estimated cost of increased VE could account for 30 and 81% of the rise at the two work loads. The sum of these factors could account for 60 and 111% of the rise in vo, at the 65 and 80% of vo, max work loads.
ranged from 39.7 to 60.7 ml. kg+ min-l with a mean of 51.9 -+ 1.4 ml wkg+ min? All subjects were scheduled to complete two ZO-min exercise bouts on a bicycle ergometer at work loads chosen to elicit 65 and 80% of VO, maxfor each subject. A number of the exercise tests had to be eliminated from the analysis because of technical difficulties. This left a total of 25 complete tests for the final analysis. The missing data included the Vo2 data from two runs, the lactate data from three runs, and the temperature data from three runs. In addition, three subjects were unable to complete the 20-min test at 80% of VO, max* Blood samples were drawn through a Teflon catheter placed in an antecubital vein at rest and after 1, 3, 5, 8, 11, 14, 17, and 20 min of exercise. Blood samples were immediately pipetted into chilled trichloroacetic acid (TCA) solution and later analyzed for lactate concentration by the method of Barker and Summerson (2). A Yellow Springs telethermometer was used to monitor core temperature at the time each blood sample was taken. oxygen uptake; oxygen uptake kinetics; ventilatory control; voZ was monitored continuously on a breath-bycost of ventilation; rectal temperature; blood lactate breath basis throughout the exercise. The computerbased system was similar in design to that previously described (5) and utilized a PDP-12 digital laboratory computer, Beckman OM-11 and LB-2 rapid response gas A NUMBER OF INVESTIGATORS (8, 18, 23) have reported analyzers, and an expiratory pneumotachometer atthat oxygen uptake (VQ) continues to increase after the initial 3 min of exercise at work rates requiring more tached to a Statham differential pressure transducer. The pneumotachometer was positioned approximately 6 than 60% of vo, max. The slow rise of vo, after the ft from the breathing valve (5). Gas samples were initial minutes of exercise has been attributed by Henry continuously drawn to the analyzers at a rate of 200 ml/ (13) and Volk ov et al, (22) to the removal of blood min through narrow-gauge tubing connected directly to lactate. However, it is well known that other, potenthe mouthpiece; expired flow was corrected for this tially calorigenic, factors, are also changing during continous sampling rate. i70, was calculated as delong-term exercise. The purpose of this study was to scribed by Beaver et al. (5) and included a correction for investigate the possible role of substrate utilization, expired water vapor dilution of the respiratory gas rectal temperature, and ventilation in causing the slow fractions (4). rise of vo, during constant-load exercise in addition to The vo2 data points for the entire run were fitted to indirectly investigating the possibility of an explanation the two-component exponential model described by based on the metabolism of lactate. Henry (13) and Volkov et al. (22) using an iterative nonlinear regression technique. The regression data METHODS concerning the second component, which is slower and Eighteen male subjects, aged 20-33 yr, volunteered to quantitatively smaller than the first component, were take part in the experiment. All signed informed conutilized to describe the time course of the Voz beyond sent statements after the design of the study and its the initial minutes of exercise. Also, the raw VO, data potential hazards were outlined. They first completed a were averaged in the 5th and 20th min of exercise and continuous incremental bicycle ergometer test to assess compared using a t test to determine if the second, slow their voZ max; the initial load ranged from 360 to 720 component of the rise in ire, was statistically evident. kpmimin and was increased by 180 kpmimin until Expired volume (VE) and respiratory exchange ratio subjective exhaustion occurred (11). vo, max values (RER) were also averaged in the 5th and 20th min to OOZl-8987/78/0000-0000$01.25
Copyright
0 1978 the American
Physiological
l
l
Society
381
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
382
HAGBERG,
determine the change in these variables during the exercise. The oxygen cost of the VE values in the 5th and 20th min were calculated from the data of Cournand et al. (9), Data concerning the oxygen cost of ventilation are highly variable; however, the data of Cournand et al. (9) were used since they represent the median of 10 studies reviewed by Shephard (20). The cost of the change in RER was calculated from the table of Zuntz presented by Swift and French (21); their table reveals that a 0.01 decrease in RER would necessitate a 0.25% increase in 00, to maintain the same energy production. The approximate change in vo, that would result from a known change in rectal temperature was calculated from the data of Grimby (12) who found a 5.5% increase in vo2 during exercise for a 1.3”C increase in rectal temperature.
MULLIN,
AND
NAGLE
TABLE 2. Blood lactate and temperature during constant-load work Relative Work Load (% b ma,>
n
Freexercise Lactate, mg1100 ml
Lactate at 8 min, mg1100 ml
65 80
13 12
10.0 ?I 0.9 15.1 + 1.9
26.9 * 3.3 50.8 2 4.6
ALA
I
mg,lOO ml
-0.3 -t- 1.7 15.3 IL 2.9
ATemp,
“C
0.5 5 0.1 0.8 -t 0.1
All values are means + SE. ATemp value is the difference in that variable between 5th and 20th min of exercise; ALA is the difference between 8th and 20th min.
a 7
‘400
I’
El
Portion
not
Portion
incrsarsd
occountsd cost
accounted
Portion Increased
accountad tmmperature
of
for for by ventilation for
by
Erceu
accounted
for
RESULTS
A statistically significant (P < 0.05) rise in vo, from the 5th to 20th min of exercise occurred in 76% of the tests at 65% of vo 2 max and 85% of the tests at 80% of 2 max (Table 1); however, the magnitude of the rise at these two work loads was not significantly different. The time course of the second component of exercise vo2 at the two work loads was not significantly different. The average half time of the slow rise in vo, was 3,7 min. The maximum blood lactate level during exercise at 65% of VO, max was 28.7 t 3.3 mg/lOO ml (n = 13, mean t SE). Blood lactate concentration remained essentially unchanged after the 8th min of exercise (Fig. 1). The mean lactate concentrations for exercise tests l
vo
TABLE 1. Work for constant-load Relative Work Load (% 90, ,&
65 80
load and Ih2 data exercise Absolute Work Load, kpm/min
IE
13 12
1,005 1,299
k 61 k 79
vo, at 5 min I * min-’
Avo2*
ml 0min-’
2.25 * 0.21 2.98 2 0.29
184 k 36 330 -c 63
Values are expressed as means + SE. * Avo, in vo, between 5- and ZO-min exercise value.
is the difference
80r &Q 13, -E
Stop Exercise
Rest
0
5 TIME
FIG, vO2
max.
1. Blood
lactate
during
10 (Minutes) 20 min
of exercise
15
20 at 65 and 80% of
0
80 RELATIVE WORKLOAD (% of i/o, max) 45
FIG. 2, Causes of rise in %, of exercise at 65 and 80% of vo,
observed max.
between
5th and 20th
min
requiring 80% of ‘cio2 max are presented in Fig. 1. The subjects had average maximum lactate values of 66.1 -+ 6.8 mg/lOO ml; an increase in lactate concentration of 15.3 -t- 2.9 mg/lOO ml occurred between min 8 and20 of exercise (Table 2). Sixteen runs (5 at 65% and 11 at 80% of vo2 max> exhibited maximum blood lactate levels in excess of 30 mg/lOO ml and were included in the analysis of the relationship of the change in lactate to the change in vo2. The correlation between these two variables was 0,48 (P < 0.05) in these 16 runs; however, in the 9 remaining tests where lactate was only slightly elevated above resting levels, the rise in vo, was still evident. The change in rectal temperature could account for 31% of the rise in vo2 at both work loads (Fig. 2). Ventilation increased between the 5th and 20th min of exercise at both work loads (Table 3). The estimated cost of this increased ventilatory work could account for 30% of the rise in VO, at 65% of VO, maxand 81% of the rise Of 80% Of V02maxm It is interesting to note that at the heavier work load, where a marked hyperventilation occurred, the increase in ventilatory work could account for a major portion of the increase in vo2. The increase in ventilation at both loads was brought about by an increase in breathing frequency (Table 3). Paired t tests revealed no significant differences between the tidal volumes at the 5- and 20-min point at each work
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
%bp DURING
CONSTANT-LOAD
TABLE 3. Ventilatory changes constant-load exercise Relative Work Load 9% mmlax
VT,
liters
EXERCISE
383
during
served. In fact, the half time of the response reveals that more than 97% of the slow rise in VO, had already occurred by the 20th min of exercise. The cause of the observed rise in 00, during constantload work is unknown. Henry (13) and Volkov et al. (22) both theorized that it was due to an oxidative removal of lactate from the blood. Our results, thou .ghindirect, offer two refutations to this “metabolism of lactate” explanation offered for the slow rise in Vo, observed during constant-load exercise. First, though there was a small, but significant, correlation between the change of 00, and blood lactate in tests which elicited blood lactate concentrations in excess of twice the resting level, the rise in voZ was also evident at work rates where virtually no change in blood lactate was observed from rest to exercise. Second, at 80% of Vo2 max, all of the rise in iToZ was accounted for by the increases in Thus, our data support temperature and ventilation. the contention th .at the slow rise of VoZ in exercise appears not to be solely a “lactacid” phenomenon (1, 3); in fact, little experimental evidence exists to assign any role to such a mechanism. One calorigenic factor involved in causing the increase in Vo, during constant-load exercise is the direct effect of temperature. The observed increases in rectal temperature could account for a 57 and 103 ml/min rise in VO, at the 65 and 80% of iTo maxwork loads. Both values are approximately 3% of the total vo, in 1 min-I, implying that the metabolic effect of the temperature rise is proportional to the energy utilization, and hence heat release, of the subjects. Ventilation increased from the 5th to 20th min of exercise at both work loads, Other authors (7, 8, 18) have noted increases in ventilation during constantload exercise similar to those in the present investigation. Our results concerning the cost of the increased ventilation should probably be interpreted qualitatively, rather than quantitatively, since the oxygen cost of increased ventilation has been shown to be highly variable (20), and in fact, may be due to cellular alkalosis rather than increased work of the respiratory muscles (6, 15). Our data reveal that one-third of the rise in VO, at 65% of VO, max can be attributed to increased ventilation. The ventilatory cost, however, can account for four-fifths of the i70, rise at a heavier work load where a more marked hyperventilation occurs. Our data support the fmdings of Katch et al. (16) who found that only a portion of the rise in ire, during constant-load exercise could be attributed to the cost of increased ventilation. The rise in ventilation during this period is interesting from another standpoint. Though VE is increasing, which will inherently decrease exercise efficiency because of the additional work of the respiratory muscles, an attempt to limit the decrease in efficiency is made by causing an in crease in breathing frequency rather than an increased tidal volume. It ha.s been shown that an increase in frequency is the more efficient method, in terms of oxygen cost and mechanical work, to increase ventilation (9). The cause of this increased ventilation that occurs after the initial minutes of constant-load exercise is
f’
STPD
20 min
ALE, llmin
ARER
5 min
27.5 + 1.8 29.9 k 1.7
31.6 2 1.8 38.9 -t 2.5
11.4 k 1.9 33.8 + 7.9
0.00 t 0.01 - 0.02 2 0.03
n 5 min
65
13
80
12
20
2.11 IL 0.17 2.33 2 0.20
Values are expressed between 5- and 20-min
min
2.20 2 0.16 2.66 IL 0.18
as means k SE, ALE is the difference exercise value,
in
VE
TABLE 4. Calculated PACT, values at 5 and 20 min of exercise Relative
5-min ZO-min Values equations
PkO, P&,
are means of Otis (19).
f
SE,
Work
Load
65% vo2 max
80% vo, max
41.9 + 1.0 39.6 + 1.3
38.7 * 1.5 32.3 k 1.9
expressed
in
Tory
calculated
from
load, though the tendency for a slight increase was evident. However, breathing frequency increased by 14% from the 5th to 20th min of exercise at 65% of 00, maxand by 30% at 80% of VOW max. The average change in RER from the 5th to the 20th min of exercise was 0 and -0.02 at the 65 and 80% of VOW max work loads. However, it must be pointed out that we measured RER at the lungs which is influenced by hyper- or hypoventilation and thus may not be indicative of respiratory quotient (RQ) at the muscle in either of these situations, The change in substrate utilization at 65% of vo 2maxwould have been evident from the change in RER because there was very little change in PkO, between the 5th and 20th min of exercise (Table 4). However, at 80% of VO, maxa larger change in substrate utilization would have been evident, if a progresqive hyperventilation had not occurred between the 5th and 20th min of exercise (Table 4). The observed changes in RER could account for only a 0.5% rise in VO, at the 80% of Vo2 maxwork load. DISCUSSION
The slow rise of Vo2 during constant-load work was evident at both work loads, 65 and 80% of iTo maxi investigated in this study. The magnitude of the rise in vo2 from the 5th to the 20th min of exercise is similar, at the heavier work load, to those reported by Nagle et al. (18) for treadmill running at 82-89% of vo2 maxaThis slow rise in vo2 from the 5th to 20th min of exercise must be caused by changes occurring after the 5th min of exercise and must not be caused directly by the work load itself because the VoZ value in the 5th min of exercise is essentially that which would be predicted for that absolute work load on a bicycle ergometer (Table 1). The mathematical models of vo, in exercise (13, 22) predict that the observed rise in Voe after the initial minutes of exercise is an exponential function of time. Thus the changes in Vo, during successive 3,7-min periods will decrease until virtually no change is ob-
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
384 unknown. The ventilation is increased beyond that required to maintain arterial Pcoz homeostasis; as shown in Table 4, the result is an arterial hypocapnia. One ventilatory control factor often implicated in mediating this hypocapnic ventilation is an increased temperature. Dempsey and Reddan (10) have demonstrated that abolishing the rise in core temperature during exercise also eliminated the hypocapnic hyperventilation. The correlation between the increase in rectal temperature and the rise in ventilation during the last 15 min of exercise was 0.59 (P < 0.001) in the present investigation, which would lend support to the hypothesis that temperature may play a predominant role in altering ventilation during long-term exercise. The change in substrate utilization indicated by the RER, as measured in the expired air, contributed very little to the elevation of vo, during the exercise. A portion of the decrease in tissue RQ, which would indicate a switch to fatty acids as the carbon source for the TCA cycle, was masked by a hyperventilation at the lung. Thus, if the RQ could have been measured at the tissue level, a much larger increase in Vo, could have been attributed to the shift to an energetically less efficient oxidation of free fatty acids. The total calculated effect of the three calorigenic factors, ventilation, temperature, and respiratory quotient, could account for 60% of the rise in vo2 during exercise requiring 65% of vo, maxand for essentially all (actually 111%) of the rise in VO, at 80% of VO, max. Thus, it seemsthat at a lighter work load, 65% of Vo, max in this study, an as yet unknown factor must account
HAGBERG,
MULLIN,
AND
NAGLE
for much of the increase in ire,. Various other factors (circulating catecholamines and mechanical efficiency) may also contribute to the increase in Vo2 during the exercise. The effect on Vo, during long-term work of these two, possibly calorigenic, factors can not be ascertained, either qualitatively or quantitatively, at this time. The net result of the data, concerning the rise in Vo,, VE, and temperature and the slight decline in RER at the work loads investigated, implies that what people have called the “steady state” is really a period of nonsteady-state exercise. Even though the changes in these variables are relatively small when compared to those that occur in the initial minutes of exercise, their absolute magnitude is such that they can hardly be considered invariable with time. Our data indicate that the cause of the slow increase in vo, is primarily a temperature effect, both directly, as evidenced by a QIO effect, and indirectly, by causing a slight hyperventilation which increases the vo2 of the respiratory muscles. The authors thank Michael Clark and James Carlson of the University of Wisconsin Laboratory Computer Facility for their assistance during the development of the computer-based voe system. Appreciation is also expressed to Dr. M. L. Birnbaum for the use of his laboratory facilities. A preliminary report of this investigation was presented at Chicago, Illinois at the meeting of the Federation of American Societies for Experimental Biology in April, 1977. Received
5 August
1977; accepted
in final
form
19 April
1978.
REFERENCES 1. ALPERT, N. R., AND N. S. ROOT. Relationship between excess respiratory metabolism and utilization of intravenously infused sodium racemic lactate and sodium L (-> lactate. Am. J. Physiol. 177: 455-462, 1954. 2. BARKER, S. B., AND W. H. SUMMERSON. The calorimetric determination of lactic acid in biological materials. J. Biol. Chem. 138: 535-537, 1941. 3. BARNARD, R. J., M. L. Foss, AND C. M. TIPTON. Oxygen debt: involvement of the Cori cycle. Intern. 2. Angew. Physiol. 28: 105-119,197o. 4. BEAVER, W. L. Water vapor corrections in oxygen consumption calculations, J. AppZ. Physiol. 35: 928-931, 1973. 5. BEAVER, W. L., B. J. WHIPP, AND K. WASSERMAN. On-line computer analysis and breath-by-breath graphical display of exercise function tests. J. Appl. Physiol. 34: 128-132, 1973. 6. CAIN, S. M. Increased oxygen uptake with passive hyperventilation of dogs. J. AppZ. PhysioZ. 28: 4-7, 1970. 7. CLAREMONT, A. D., F. J. NAGLE, W. D. REDDAN, AND G. A. BROOKS. Comparison of metabolic, temperature, heart rate, and ventilatory responses to exercise at extreme ambient temperatures. Med. Sci. Sports 7: 150-154, 1975, 8. COSTILL, D. L. Metabolic responses during distance running. J. AppZ. Physiol. 28: 251-255, 1970. 9. COURNAND, A., D. W. RICHARDS, R. A, BADER, M. E. BADER, AND A. P. FISHMAN. The oxygen cost of breathing. Trans. Assoc. Am. Physicians 67: 162-173, 1954. 10. DEMPSEY, J. A., AND W. G, REDDAN. Hyperventilation during prolonged work: cause and effect. Med. Sci. Sports 8: 62, 1976. If. DILL, D. B. Assessment of work performance. J. Sports Med. Phys. Fitness 6: 3-8, 1966. 12. GRIMBY, G. Exercise in man during pyrogen-induced fever. Acta Physiol. Stand. SuppZ. 67: l-114, 1962. 13. HENRY, F. M. Aerobic oxygen consumption and alactic debt in
muscular work. J. AppZ. Physiol. 3: 427-438, 1951. 14. HENRY, J. G., AND C. R. BAINTON. Human core temperature increase as a stimulus to breathing during moderate exercise. Respiration Physiol. 21: 183-191, 1974. 15. KARETZKY, M. S., AND S. M. CAIN. Effect of carbon dioxide on oxygen uptake during hyperventilation in normal man. J. Appl. PhysioZ. 28: 8-12, 1970, 16. KATCH, F. I., R. N. GXRANDOLA, AND F. M. HENRY. The influence of the estimated cost of ventilation on oxygen deficit and recovery oxygen intake for moderately heavy bicycle ergometer exercise. Med. Sci. Sports 4: 71-76, 1972, 17. KAYNE, H. L., AND N. P. ALPERT, Oxygen consumption, following exercise in anesthetized dogs. Am. J. PhysioZ. 206: 51-56, 1964. 18. NAGLE, F. J., D. ROBINHOLD, E. HOWLEY, J. DANIELS, G. BAPTISTA, AND K. STOEDEFALKE. Lactic acid accumulation during running at submaximal aerobic demands. Med. Sci. Sports 2: 182-186, 1970. 19. OTIS, A. B. Quantitative relationships in steady-state gas exchange. In: Handbook of Physiology. Respiration. Washington, D.C.: Am. Physiol. Sot., 1964, sect. 3, vol. I, chapt. 27, p. 681698. 20. SHEPHARD, R. J. The oxygen cost of breathing during vigorous exercise. Quart. J. ExptZ. Physiol. 51: 336-350, 1966. 21. SWIFT, R. H., AND C. E. FRENCH. Energy Metabolism and Nutrition. Washington, D.C.: Scarecrow, 1954. 22. VOLKOV, N. I., V. N. CHEREMISINOV, AND E. N. RAZUMOVSKII. Oxygen exchange in man during muscular activity. In: The Oxygen Regime of the Organism and Its Regulation, edited by N. V. Lauer and A. 2. Kolchinskaya. NASA Technical Translation, 1969. 23. WHIPP, B. J., AND K. WASSERMAN. Oxygen uptake kinetics for various intensities of constant-load work. J. AppZ. Physiol. 33: 351-356. 1972.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (148.061.013.133) on August 13, 2018. Copyright © 1978 American Physiological Society. All rights reserved.
