Evidence that diffusion limitation determines oxygen uptake kinetics during exercise in humans

Evidence That Diffusion Limitation Determines Oxygen Uptake Kinetics during Exercise in Humans Akira Koike, Karlman Wasserman, David K. McKenzie, Stefania Zanconato, and Daniel Weller-Ravell Division ofRespiratory and Critical Care Physiology and Medicine, Harbor-UCLA Medical Center, Torrance, California 90509 Abstract To determine the role of arterial 02 content on the mechanism of muscle 02 utilization, we studied the effect of 2, 11, and 20% carboxyhemoglobin (COHb) on 02 uptake (V02), and CO2 output (VCO2) kinetics in response to 6 min of constant moderate- and heavy-intensity cycle exercise in 10 subjects. Increased COHb did not affect resting heart rate, V02 or VC02. Also, the COHb did not affect the asymptotic V02 in response to exercise. However, V02 and VCO2 kinetics were affected differently. The time constant (TC) of V02 significantly increased with increased COHb for both moderate and heavy work intensities. V02 TC was positively correlated with blood lactate. In contrast, VCO2 TC was negatively correlated with increased COHb for the moderate but unchanged for the heavy work intensity. The gas exchange ratio reflected a smaller increase in CO2 stores and faster VCO2 kinetics relative to V02 with increased COHb. These changes can be explained by compensatory cardiac output (heart rate) increase in response to reduced arterial 02 content. The selective slowing of V02 kinetics, with decreased blood 02 content and increased cardiac output, suggests that 02 is diffusion limited at the levels of exercise studied. (J. Clin. Invest. 1990. 86:1698-1706.) Key words: anaerobic threshold * carbon monoxide * carboxyhemoglobin heart rate kinetics * lactate - Introduction 02 uptake (V02)' by the lungs reflects V02 by the muscles. The V02 differs from the 02 requirement only when the 02 delivery to the muscles is inadequate to maintain an 02 partial pressure (Po2) difference between muscle capillaries and muscle mitochondria to meet the immediate bioenergetic 02 requirement. Whereas a primary determinant of Vo2 kinetics in response to exercise must be muscle bioenergetics (1), the adeThis study was presented in part at the 74th Annual Meeting ofFederation of American Societies for Experimental Biology (FASEB), Washington, DC, April 1990. Address reprint requests to Dr. Wasserman, Division of Respiratory and Critical Care Physiology and Medicine, A-15 Annex, Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, CA 90509. Receivedfor publication 19 January 1990 and in revisedform 25 June 1990. 1. Abbreviations used in this paper: COHb, carboxyhemoglobin; VCO2, CO2 output; V02, 02 uptake. J. Clin. Invest. © The American Society for Clinical Investigation, Inc. 0021-9738/90/1 1/1698/09 $2.00 Volume 86, November 1990, 1698-1706 1698 Koike et al. quacy of the 02 supply to the muscles must also determine the kinetics. Reducing the arterial 02 content, without changing arterial Po2 and without an anticipatory increase in blood flow, would result in a more rapid rate of fall in capillary Po2 than normal. This would affect the pattern of V02 by the muscles, only if 02 transport from muscle capillaries to mitochondria was diffusion limited. Inhaling low concentrations of carbon monoxide (2-5) can affect 02 content of the arterial blood without affecting the diffusion equilibrium between 02 in the alveolar gas and the pulmonary capillary blood. At the tissue level, capillary Po2 would fall more rapidly than normal because of the reduced blood 02 content and the leftward shift of the oxyhemoglobin dissociation curve caused by the increased carboxyhemoglobin (COHb) (6). The partial pressure of carbon monoxide (Pco) required to increase COHb to 20%o (- 0.15 mm Hg) does not significantly affect oxymyoglobin (binding of carbon monoxide to myoglobin is only 10% of that to hemoglobin) and mitochondrial cytochromes (7). Any increase in blood flow caused by the reduced arterial 02 content would be compensatory and incomplete, resulting in a reduced mean and end capillary P02 (4), and therefore reduced 02 availability to the exercising muscles. Thus if 02 uptake by the muscle mitochondria were limited by diffusion, rather than muscle bioenergetics alone, V02 kinetics in response to a given work rate will be slowed when capillary P02 was decreased. In order to determine the role of diffusion in 02 utilization during exercise, we measured V02, CO2 output (VCO2) and heart rate kinetics and end-exercise lactate while COHb was systematically controlled at 1 1% (level of heavy cigarette smoker) and 20%, during constant work rate exercise. Methods Subjects. 10 normal nonsmoking subjects without cardiac or pulmonary disease ranging in age from 18 to 45 yr were studied (Table I). The nature and purpose of the study and the risks involved were explained. Each subject voluntarily consented to participate in the study. The protocol and procedures for this study were reviewed and approved by the Institution's Human Subjects Committee. Carbon monoxide loading. Carbon monoxide loading was accomplished using a procedure modified from Vogel and Gleser (4). This technique involved establishing a titration curve by monitoring venous COHb levels after breathing successive levels of 5-10 liters of 1% carbon monoxide in air using a co-oximeter (Instrument Laboratory, Inc., Lexington, MA) to measure COHb. This titration was used to estimate the volume of 1% carbon monoxide that the subject needed to breathe in order to achieve 11% or 20% COHb levels on the study days. No subject experienced symptoms from breathing the gas, and resting heart rate and minute ventilation were not changed. Exercise protocol. Exercise tests were employed using an upright, electromagnetically braked cycle ergometer (Gould Godhart BV, Bilthaven, The Netherlands). Each subject performed two levels of exercise during three sessions on different days in randomized order as follows: one session without added carbon monoxide (control), one Table L Physical Characteristics and Work Rates ofModerate and Heavy Intensity Studiedfor Each Subject Work Subject Age Sex yr 1 2 3 4 5 6 7 8 9 10 45 31 40 32 29 39 31 18 28 35 Mean±SD 32.8±7.1 M M M M F M M M M M Height Weight cm kg 180 168 168 175 168 168 186 165 175 173 79 59 70 67 60 71 86 62 69 74 120 55 50 145 100 60 130 80 80 90 200 110 110 260 200 130 260 190 200 205 172.6±6.2 69.7±8.1 91.0±30.9 186.5±51.6 session with - 11% COHb, one session with 20% COHb. During each session, moderate and heavy work intensity exercise of6 min each starting from rest were performed. Three repetitions of both moderate and heavy work intensity exercise were completed during each session whenever possible. The subjects rested at least 30 min between repetitions. The moderate and heavy work intensity exercise levels were determined by finding the work rates during air breathing corresponding to the Vo2 at 80% of the anaerobic threshold (moderate exercise intensity), and the work rate at a V02 which was 40% of the difference between the anaerobic threshold and maximal V02 (heavy exercise intensity). The actual work rates performed at the moderate and heavy work intensities for each subject are shown in Table I. The anaerobic threshold and maximal Vo2 were previously determined for each subject from an incremental cycle ergometer exercise test while breathing room air without added carbon monoxide. The same protocol was used for all three levels of COHb. To obviate the energetic effect of inertia during acceleration of the flywheel at the start of exercise, an electric motor was used to drive the flywheel at 60 rpm during the rest periods. The motor was switched off when the subject began pedaling. The start of exercise was signaled by the change of a light from red to green within the subject's view. To avoid startle responses, no verbal command was given. To maintain the COHb at the 11% or 20% level, subjects breathed 0.023% carbon monoxide in air during exercise. Measurements. Heart rate was continuously monitored by a cardiotachometer, and superficial forearm vein blood was obtained before and 2 min after the exercise for measurement of lactate (enzymatic method [8]). COHb level was also measured before and at the end of exercise. Subjects breathed through a mouthpiece attached to a turbine volume transducer (Alpha Technologies, Hayward, CA) for measurements of ventilatory volumes during the test. Dead space of the system was 170 ml. Respired gases were sampled continuously from a site at the mouthpiece for analysis of 02, C02, and N2 by mass spectrometry (Perkin-Elmer Corp., Norwalk, CT). Breath-by-breath calculations of alveolar VO2 and VCO2 were performed as previously described (9). Data analysis. To enhance the signal to noise ratio, three repetitions of each test were performed and superimposed for both moderate and heavy work intensity exercise in six subjects. Two repetitions were performed and superimposed for only moderate work intensity exercise in two subjects. For the remaining two subjects with less noisy breathing, a single transition of each test was performed and analyzed. The repetitions were more important for low work rate exercise in subjects with noisy breathing. Moderate Heavy W The breath-by-breath data from each of the three repetitions for each 6-min test were interpolated to give values second by second. These values were time-aligned to a mark at the start of exercise, and superimposed to average random noise and enhance the underlying response patterns for each test with each COHb level for each subject (10). These averaged 60 responses (two work rates, three COHb levels, and 10 subjects) were then used for data analysis. 02 pulse was calculated every second by dividing V02 by heart rate. To determine the kinetics of V02, the time constant was determined for all the data from the start ofexercise, assuming exponential kinetics, using least square nonlinear regression through the data. The asymptotic VO2 was determined by fitting the data to the sum of two exponentials, as previously described by Linnarsson (11). The first exponential term described the kinetics to 180 s and the second exponential term described the kinetics for the data between 180 and 360 s. The overall dynamics were then expressed as a single exponential time constant through all the data from the resting Vo2 to the asymptotic Vo2, when V02 did not reach steady state by 6 min, or actual steadystate V02 if steady state was established before 6 min. The same analysis was used for VCO2, heart rate, and 02 pulse. While Vo2 kinetics may be more complex than a single exponential increase for heavy exercise (12-14), this approach is a useful model, as previously reported (15). The increase in V02 at 6 min as compared to 3 min of constant work rate tests [AVO2 (6-3)], a measurement which was described by Roston et al. (16) to be highly correlated with the increase in blood lactate, was also calculated. COHb levels for a given test were determined from the average of the values before and after the test. Statistical methods. Differences in the parameters without added carbon monoxide (control), with - 11 % COHb and 20% COHb were determined by analysis of variance for repeated measures. When the F test was significant, individual comparisons were made by NewmanKeuls' multiple-range test. Variations about the mean are expressed as ± 1 SD and differences were considered significant at the P < 0.05 level. Results The physical characteristics of the subjects are shown in Table I. Venous lactate concentration at rest before exercise averaged 1.0±0.4 mM/liter for control studies, 1.2±0.5 mM/liter for 11% COHb studies, and 1.2±0.7 mM/liter for 20% COHb studies. The differences were not significant. Diffusion Limitation and 02 Uptake Kinetics 1699 Table II. COHb Level of Each Exercise Test Heavy intensity tests Moderate intensity tests 19% COHb Subject Control 11% COHb 20% COHb Control 11% COHb 1 2 3 13.5 11.3 11.4 11.6 12.4 12.1 9.6 10.2 9.2 11.3 19.6 19.4 19.4 21.2 20.5 19.0 20.1 23.3 21.6 19.4 2.0 2.5 2.5 2.0 13.7 12.6 11.8 11.8 11.2 20.2 19.3 6 7 8 9 10 2.0 2.5 2.5 2.0 1.1 3.0 0.8 0.9 0.8 0.6 13.1 9.7 10.7 18.9 18.6 20.4 0.8 0.6 10.0 11.1 19.4 17.8 Mean±SD 1.6±0.8 11.2±1.2 20.3±1.3 1.6±0.9 11.5±1.2 19.2±0.7 4 5 The mean COHb level of each test during exercise is given in Table II. The actual COHb level without added carbon monoxide (control tests), and with added carbon monoxide during exercise were 1.6±0.8%, 11.2+1.2%, and 20.3±1.3%, for the moderate work intensity tests, and 1.6±0.9%, 11.5±1.2%, and 19.2±0.7% for heavy work intensity tests, respectively. These increased COHb levels were kept almost constant during exercise by breathing 0.023% carbon monoxide. Superficial forearm venous lactate concentration of the control (air breathing) study obtained 2 min after the moder- 1.1 3.0 0.8 0.7 18.5 19.4 19.7 ate work intensity exercise was 1. 1±0.6 mM/liter and increased to a small but significant amount at the 20% COHb level (Table III, Fig. 1). For the heavy work intensity exercise, lactate concentration was 5.0±2.0 mM/liter for the control study and increased to 6.7±1.8 and 9.4± 1.1 mM/liter, for the 11.5% and 19.2% COHb levels, respectively. These increases were significant at each level of COHb (Table III, Fig. 1). Fig. 2 shows the responses of V02, VCO2, and gas exchange ratio in one representative subject (subject 4 in Table I) during both moderate and heavy work intensity tests without added carbon monoxide and with - 20% COHb. The V02 of Table III. V02, VCO2, Heart Rate, and OrPulse Responses to Moderate and Heavy Work Rate Tests and Lactate Concentration Obtained 2 min after Exercise Heavy intensity tests Moderate intensity tests Time constant of V02 (S) ±SD V02 at 6 min (liter/min) ±SD Vo2 asymptote (liter/min) ±SD AVO2 (6-3) (liter/min) ±SD Time constant of VCO2 (S) ±SD VCO2 asymptote (liter/min) ±SD HR asymptote (beats/min) ±SD 02-pulse asymptote (mI/beat) ±SD Lactate (mM/liter) +SD Control 1 1% COHb 20% COHb P value Control 1 1% COHb 19% COHb 28.8 30.5 5.5 1.70 0.38 1.69 0.38 0.036 0.045 57.5 12.6 1.51 0.36 118.9 12.8 14.0 3.1 1.5 0.8 33.6 5.5 1.71 0.39 1.74 0.40 0.041 0.062 53.9 12.8 1.57 0.36 131.7 11.4 13.1 2.6 1.9 0.6 <0.0001 51.4 11.7 2.88 0.67 2.95 0.67 0.145 0.051 70.0 17.5 2.81 0.65 157.4 6.8 18.9 3.9 59.5 12.8 2.90 0.67 3.00 0.68 0.209 0.073 73.1 15.9 2.98 0.64 166.6 5.1 18.0 4.1 5.0 6.7 1.8 67.5 16.5 2.87 0.60 2.97 0.61 0.265 0.089 75.4 22.4 3.07 0.59 173.6 10.6 16.9 3.4 9.4 4.1 1.69 0.37 1.71 0.36 0.030 0.047 68.3 10.2 1.54 0.35 115.0 8.3 14.8 2.8 1.1 0.6 NS NS NS 0.007 NS <0.0001 <0.000 1 <0.02 2.0 P value 0.0001 NS NS 0.003 NS <0.02 0.0002 0.0002 <0.0001 1.1 AVO2 (6-3) = the increase in 02 uptake at 6 min as compared to 3 min of constant work rate tests. P value was determined by analysis of variance for repeated measures. HR, heart rate. 1700 Koike et al. MODERATE WORK HEAVY WORK 12 - 121 10- 10 - S 0 8 E E 6 a) Subject 6- 10 0-0 *-@ A-6 0 4 u *-- 24- 0-0 *-,-, V-V 2- 21C C n 1 1% 20% 1 2 4 3 5 6 7 8 0-0 9 *-* 10 I 11% C 19% Figure 1. Venous lactate concentration of each test obtained 2 min after exercise. The symbols for each subject are indicated on the figure, and the subject numbers correspond to those in Tables I and II. *P < 0.05, **P < 0.01 by Newman-Keuls' multiple-range test. C, control studies; 11%, studies with 11.2% COHb for moderate work intensity and 11.5% COHb for heavy work intensity exercise; 19%, studies with 19.2% COHb; 20%, studies with 20.3% COHb. For moderate-intensity exercise, the two levels of COHb averaged 11.2% and 20.3%. For heavy work intensity exercise, the two levels of COHb averaged 11.5% and 19.2%. the moderate work intensity test of the control study reached a steady state within -1 min and remained constant until the end of exercise. Moderate work Mean V02 at 6 min of the control, 11% and 20% COHb test were 1.69±0.37, 1.70±0.38, and 1.71±0.39 liter/min for the moderate work intensity tests and 2.88+0.67, 2.90±0.67, and 2.87±0.60 liter/min for the heavy work intensity tests, respectively (Table III). There was no difference in V02 at 6 min as related to COHb level. For both moderate and heavy work intensity tests, the V02 asymptote, calculated by a single exponential curve fit to the V02 response, did not change with increased COHb levels (Table III). However, the time constant for Vo2 kinetics were significantly increased with increased COHb levels (Fig. 3). The time constants of the heavy work intensity tests were higher than those of the moderate work intensity tests, and increased to a greater degree with increased COHb levels (Fig. 3). Fig. 4 shows the relationship between the time constant of Vo2 and the lactate concentration. The time constant of Vo2 increased with the level of lactate, showing a strong statistical correlation (r = 0.869, P < 0.0001). Fig. 5 shows AVO2 (6-3) of each test. Although there was no significant difference in AVO2 (6-3) for moderate work intensity tests, AVO2 (6-3) ofheavy work intensity tests significantly increased with increased COHb. Despite the fact that the time constant is dominated by data during the first 3 min of testing, whereas AVO2 (6-3) is determined by data between 3 and 6 min of exercise, AVO2 (6-3) correlated well with lactate concentration (r = 0.784, P < 0.0001) (Fig. 6). The VCO2 asymptote did not change for moderate work Heavy work 4. .-1 0 ._q -4 ._ 0 I-I CQ 0 I .E 0 1- CII 0 3 1-1 r) 2 CO, CU 0 2 1I - 01.2 Control (2.0% COHb) 21.2% COHb - ti - Control (2.0% COHb) 20.2% COHb 1.21 1.0 - 0.8 0 1.0 I' k. !. -% 0.8 - - I/ -60 0 60 120 180 240 300 360 Time (s) Figure 2. Effects of increased COHb on Vo2, VCO2, and gas exchange ratio (R) during moderate (145 W) and heavy (260 W) work intensity exercise without added carbon monoxide and with 20% COHb in one subject (subject 4 in Table I). The values to the left of time "0" are measured at rest. The shaded area denotes cardiodynamic phase (15 s after the onset of exercise). -60 0 60 120 180 240 300 360 Time (s) Difusion Limitation and 02 Uptake Kinetics 1701 oN 80 0 -.1 0 60- 80- i 40 - 0 C) v) 40- _ 40- C- I* I lio C n 20 -J l- 0- 20'% C 1170 Figure 3. The time constant of V02 in response to moderate and heavy work intensity exercise for 10 subjects. Time constant of V02 for both moderate and heavy work intensity tests was significantly increased with increased COHb levels. *P < 0.05, **P < 0.01 by Newman-Keuls' multiple-range test. C, control studies; 11%, studies with 11.2% COHb for moderate work intensity and 11.5% COHb for heavy work intensity exercise; 19%, studies with 19.2% COHb; 20%, studies with 20.3% COHb. intensity tests, but was significantly higher for the heavy work intensity 19% COHb studies (3.07±0.59 liter/min) than the control studies (2.81±0.65 liter/min) (Table III). The time constant of VCO2 of the moderate work intensity tests was 68.3±10.2 s for the control study, 57.5±12.6 s for the 11% COHb study, and 53.9±12.8 s for the 20% COHb study, showing a significant decrease with increased COHb levels (P < 0.05 between the control and 11% COHb study, and P < 0.01 between the control and 20% COHb study) in contrast to the V02 changes (Table III). There was no significant difference in the VCO2 time constant for heavy work intensity tests. Fig. 7 relates the ratio of the VCO2 and VO2 time constants to blood lactate. The ratio, while averaging 2.4 for the air- A 0 ----00 01 -0.1 C 11% 19% 17% 20% Figure 5. The increase in 02 uptake at 6 min as compared to 3 min of constant work rate tests [AVO2 (6-3)] for each test. AVO2 (6-3) of heavy work intensity tests significantly increased with increased COHb. *P < 0.05, **P < 0.01 by Newman-Keuls' multiple-range test. C, control studies; 11%, studies with 11.2% COHb for moderate work intensity and 11.5% COHb for heavy work intensity exercise; 19%, studies with 19.2% COHb; 20%, studies with 20.3% COHb. C breathing moderate work intensity tests, decreased strikingly and approached 1 as lactate increased. This ratio significantly decreased with increased COHb for both moderate and heavy work intensity exercise, as well as between moderate and heavy exercise at each COHb level. This decrease was primarily due to slowed V02 kinetics with faster or unchanged VCO2 kinetics. Fig. 8 shows heart rate responses during both moderate and heavy work intensity exercise in one subject (the same subject as shown in Fig. 2). There was no difference in heart rate at rest but the rate of rise and peak heart rate response was greater the higher the COHb level for both moderate and heavy work intensity. The asymptotic values were significantly increased with increased COHb for the group as a whole (Table III). Fig. 9 shows the kinetics of 02 pulse during moderate and heavy work intensity exercise tests in one subject (the same subject as shown in Fig. 2). Peak values are seen by 1 min for A C/) 0.5 - Go 80 0 0.4- 60 0.3 Control 7 * 1 1% COHb F. 40 0 g 0.2 - -0.1 1970 c CD t* -J I 0.3 v 0. 1- 0 \ v 0.1 N, * Ed 0.4 0 L- . * n 0.4- 0.2- 60- - .2 0.5 0.3- \. CD o V) 0.5 - 100 100U) 0 HEAVY WORK MODERATE WORK HEAVY WORK MODERATE WORK - - - AL 20% COHbI L 20 y=24.7+4.8x r=0.869 P(O.OOO1 -1 0 4 6 8 10 12 Lactate (mM/1) Figure 4. Relationship between the time constant of V02 and superficial forearm vein lactate concentration obtained 2 min after exercise. Data are for both moderate and heavy work intensity exercise at each COHb level; symbols for 20% COHb studies in this figure include 20% COHb studies of moderate work intensity exercise and 19% COHb studies of heavy work intensity exercise. C 0.2 - - r- Control ~-L/ I* 1 1% COHbI IL 20% COHb L o 0.1 0 y=O.008+0.026x r=0.784 p(O.OOO1 2 1702 Koike et al. -0.1 4 6 8 10 12 Lactate (mM/1) Figure 6. Relationship between the increase in 02 uptake at 6 min as compared to 3 min of constant work rate tests [WvV02 (6-3)] and superficial forearm vein lactate concentration obtained 2 min after exercise. Symbols are same as those noted in Fig. 4. 0 2 TC of o Cnr 0 Control * 11% COHb A 20% COHb 02 C)2 Discussion VCO,/TC of V0, Moderate 2.40+0.36 Heavy 1.37±0.18 Constant work rate exercise does not require a subject's maximum effort if the work rate is not unreasonably high and if exercise duration is not too long. Using a constant work rate test, the patterns of VO2, VCO2, heart rate, and 02-pulse increase in response to an exercise stimulus can be quantified and be used as descriptors of cardiopulmonary adaptations to exercise (10, 18-21). The binding of hemoglobin by carbon monoxide reversibly decreases the 02 carrying capacity, producing a useful model ofthe effect of a modest acute reduction in 02 transport. Thus this model was applied in this study to determine'the sensitivity of VO2 kinetics to detect small 1.91±0.36 1.24±0.20 62±0.29 1.13±0.20 1 0 0 (9 2- 00 L________ 'I- (9 A 0 f4U 1- 0 A000 * 0 A A A 2 4 6 8 10 12 Lactate (mM/1) Heavy work Figure 7. Relationship between the ratio of the time constant (TC) of VCo2 and that of Vo2 and forearm vein lactate concentration obtained 2 min after exercise. Each symbol includes the data of both moderate and heavy work intensity exercise of each COHb level, as described in Fig. 4. The inset relates TC of VCO2/TC of V02 to the intensity of work rate and the COHb level. the moderate work intensity. For the heavy work intensity with increased COHb, 02 pulse increased further after the 1 stmin increase. In all instances 02 pulse abruptly increased at the start of exercise and then oscillated for 15 s before increasing to its asymptote. The asymptotic values were significantly decreased with increased COHb for both work rate tests (Table III). Since 02 pulse is equal to stroke volume X arterial-venous 02 difference, and stroke volume is known not to change (4), the decrease likely reflects the'expected reduction in arterialvenous 02 difference. For both moderate and heavy work intensity tests, the gas exchange ratio decreased during the first minute, after a 15-s delay (Fig. 2). The near-constant gas exchange ratio at resting levels during the first 15 's of exercise, before the gas exchange ratio decreases, is postulated to be due to the proportional increase in VCO2 and V02 caused by increase in pulmonary blood flow at the start of exercise with the arterial-venous difference being that of the before-exercise resting state (17). The subsequent decrease in the gas exchange ratio is postulated to be due to arrival of blood at the lungs from exercising muscle. Because of CO2 solubility in tissues, 'part of the early metabolic CO2 is retained as tissue Pco2 increases consequent to the smaller increase in cardiac output relative to metabolic rate, i.e., increase in arterial-venous 02 difference is greater than increase in venous-arterial CO2 difference during this non-steady-state period. With increased COHb, the cardiac output response is increased, even for moderate intensity work, as evidenced by the increased heart rate response (Fig. 8 and Table III). This would result in a reduced arterial-venous 02 (lower 02 pulse, Fig. 9) and CO2 concentration difference and therefore less CO2 retention in'the tissues. Thus CO2 output kinetics would be expected to be faster (approach that of 02) even for moderate work intensity exercise (Table III), and the decrease in the gas exchange ratio during the first minute would be expected to be attenuated as found in this study (Fig. 2). 180 160 Control P1 140 (2.0X COHb) --I 120 < 100~ 80 60 -60 180 - 160 - 0 60 120 180 Time (s) 240 300 360 la 140120- - 100a) 80 60 4- -60 0 60 120 180 240 300 360 Time (s) Figure 8. The effect of increased COHb on heart rate response during moderate (145 W) and heavy (260 W) work intensity exercise in one subject (subject 4 in Table I). There was no difference in heart rate at rest but the rate of rise and peak heart rate response during exercise was greater the higher the COHb level for both moderate and heavy work intensity. Diffusion Limitation and 02 Uptake Kinetics 1703 Heavy work 30 25 t-P- .0 - 20- d %>- 15. N-- 100n 0 Control (2.0% COHb) ...20.2% COHb m -60 60 120 180 Time (s) 120 180 240 300 360 240 300 4i ,0 V oN -60 0 60 360 Time (s) Figure 9. The effect of increased COHb on 02 pulse as related to time during moderate (145 W) and heavy (260 W) work intensity exercise without added carbon monoxide and with 20% COHb in one subject (subject 4 in Table I). For the heavy work intensity with increased COHb, 02 pulse increased slightly after the rapid rise within the 1st min. - changes in arterial 02 content and transport without a change in Po2. Carbon monoxide has the effect of reducing 02 content and shifting the oxyhemoglobin dissociation curve to the left requiring a lower capillary P02 for 02 unloading from hemoglobin. It has been described by Root (22) that the concentration to which COHb can be increased, which is compatible with life, does not poison the cells of the body. While there are only limited studies on the level of Pco that starts to affect resting cell redox state or electron transport, it appears to be considerably higher than that required to raise COHb to 20% according to the studies of Wittenberg and Wittenberg (7) on cardiac myocytes and our lactate measurements. To maintain the level of COHb during exercise, our subjects breathed a concentration of 0.023% carbon monoxide in 1704 Koike et aL. air. During the moderate and heavy intensity exercise the COHb slightly increased for the 11% COHb (11.0±1.2 to 11.8±1.4%) and was unchanged or decreased for the 20% COHb level study (19.9±1.5 to 19.6±1.3%). Thus the blood and therefore tissue Pco must have been in the range of 0.1-0.2 mmHg, in agreement with the predicted Pco for 11 and 20% COHb from the COHb dissociation curve in air (6). This is well below the Pco that has a demonstrable effect on oxymyoglobin (affinity of carbon monoxide to myoglobin is only 10% of that to hemoglobin) and levels of Pco that affect the mitochondrial cytochromes (7). Further evidence that the low levels of Pco used in this study did not cause tissue toxicity is that V02 and ventilation (sensitive markers of acid-base balance and cell redox state) at rest were not affected. Also, resting lactate concentration was not affected by these levels of Pco. Furthermore, we found no difference in the 6-min Vo2 and Vo2 asymptote for both moderate and heavy work intensity tests as related to COHb level. It had been reported that the time constant of V02 is prolonged in patients with obstructive pulmonary disease (19) and patients with heart disease (10) compared with normal subjects. It had also been reported by Hughson and Smyth (20) and Petersen et al. (23) that beta-blockade slows the V02 increase during submaximal exercise in normal subjects. In contrast to the latter studies, which slowed Vo2 kinetics by attenuating the cardiac output increase, this study slowed Vo2 kinetics by effectively reducing capillary P02. Heart rate (cardiac output) was increased (Fig. 8), presumably as compensation for the impaired 02 delivery. In this study, the time constant for VO2 significantly increased with increasing levels of COHb (Fig. 3), even if the work intensity was moderate and lactate concentration was <2 mM/liter at the end of exercise. Since the heart rate increase suggests that cardiac output is actually increased under these circumstances (4), slowing of V02 kinetics must be attributed to the reduced blood 02 content and capillary-mitochondrial P02 difference. For higher-intensity work, the effect of reducing 02 content by increased COHb was more marked (Fig. 3) and V02 kinetics were slower, the higher the blood lactate (Fig. 4). These studies show that both lactate concentration and the time constant for V02 are increased in response to a relatively small reduction in 02 transport. V02 continues to increase slowly beyond 3 min at work rates associated with increased blood lactate (16, 24). Roston et al. (16) measured the increase in V02 at 6 min as compared to 3 min [AV02 (6-3)], during constant work rate exercise of different intensities in normal men and showed a good correlation with the increase in blood lactate. In this study, we also showed a good correlation between AV02 (6-3) and lactate concentration (Fig. 6). At rest, gas exchange at the lungs is equal to gas exchange at the cells. Thus, the gas exchange ratio at rest reflects the metabolic respiratory quotient that is determined by the mixture of substrate used for energy (24, 25). After the onset of exercise, the immediate increase in pulmonary blood flow (resulting from increased heart rate and stroke volume) causes an abrupt increase in both V02 and VCO2 (25), during which the gas exchange ratio changes little for the first 15 s (Fig. 2) (17). The gas exchange ratio then decreases, i.e., VCO2 rise lags the increase in V02, because CO2 is more soluble in tissues and blood than 02. Thus the time constant of VC02 is longer than that of V02 for work rates below the anaerobic threshold (19, 24-27). However, at work rates accompanying lactic acidosis (Fig. 7), the rate of increase in VCo2 approaches and sometimes exceeds that of Vo2 as the latter slows and the former remains the same or becomes faster. Additional CO2 generated from HC0, as it buffers the increase in lactic acid, is undoubtedly responsible for the disparate changes in VCO2 and V02 kinetics observed in this study. Consistent with these findings, Springer et al. (28, 29) observed slowing of VO2 kinetics and speeding of VCo2 kinetics in response to hypoxia. Although heart rate at rest did not differ for different COHb levels, heart rate during exercise became significantly higher with increased COHb levels. This is presumably due to a compensatory stimulation of heart rate to maintain adequate 02 delivery to the working muscles when blood 02 content is reduced. Increasing the cardiac output relative to metabolic rate, without a reduction in 02 content, should increase capillary P02. The increase in cardiac output (heart rate) (Fig. 8), and the rightward shift in the oxyhemoglobin dissociation curve consequent to increased lactic acidosis, are apparently the mechanisms by which it was possible for V02 to increase to the same steady state (moderate intensity) or 6-min Vo2 (heavy intensity) in the control and increased COHb studies. 02 pulse, which did not differ at rest, significantly decreased during both moderate and heavy work intensity exercise with increased COHb levels (Fig. 9). As 02 pulse is mathematically equal to the product of stroke volume and the arterial-venous 02 difference, decreased 02 content of the arterial blood and decreased arterial-venous 02 difference must be responsible for the decreased 02 pulse during exercise observed with increased COHb. This is similar to the effect of anemia. Fick's law of diffusion states that the mass transfer (D) of a substance, such as 02, is directly proportional to the partial pressure difference between the high pressure point in the capillary (Pc) to the low pressure point in the mitochondria (Pm) and the surface area (A) (degree of capillary hyperemia), and inversely related to the diffusion distance (L) (capillary to mitochondria), that is, D02 = k(Pc-Pm) * A/L. The proportionality constant (k) is a function of the diffusibility and solubility of 02 in the tissue substance. Increasing the COHb should have no effect on arterial P02 (4, 30). However, as with anemia (31), the capillary and venous Po2 should be decreased at submaximal work rates (4) since the increase in blood flow is compensatory and does not completely adjust for the reduced 02 flow caused by the increased COHb. Increased COHb also results in a further lowering of capillary Po2 because it causes leftward shift in the oxyhemoglobin dissociation curve. Whereas the effect of the increased COHb is to reduce the capillary to mitochondrial P02 difference, any increase in capillary surface area resulting from increased blood flow, would facilitate diffusion. Also, capillary recruitment with increased blood flow should reduce diffusion distance and speed V02 kinetics. In this study, we experimentally reduced the capillary-mitochondrial 02 diffusion gradient without affecting the steadystate or asymptotic Vo2 and thus presumably mitochondrial function. We conclude that the slowing of V02 kinetics found in this study is most likely explained by diffusion limited 02 transport to the contracting muscle mitochondria at moderate as well as heavy work intensities. Acknowledaments This work was supported by U. S. Public Health Service grant HL-1 1907. Akira Koike was supported by funds from Otsuka Pharmaceutical Co. and Merrell Dow Pharmaceuticals K.K. References 1. Whipp, B. J., and M. Mahler. 1980. Dynamics of pulmonary gas exchange during exercise. In Pulmonary Gas Exchange. J. West, editor. Academic Press, Inc., New York. 33-96. 2. Pirnay, F., J. Dujardin, R. Deroanne, and J. M. Petit. 1971. Muscular exercise during intoxication by carbon monoxide. J. Appl. Physiol. 31:573-575. 3. Ekblom, B., and R. Huot. 1972. Response to submaximal and maximal exercise at different levels of carboxyhemoglobin. Acta Physiol. Scand. 86:474-482. 4. Vogel, J. A., and M. A. Gleser. 1972. Effect of carbon monoxide on oxygen transport during exercise. J. Appl. Physiol. 32:234-239. 5. Raven, P. B., B. L. Drinkwater, R. 0. Ruhling, N. Bolduan, S. Taguchi, J. Gliner, and S. M. Horvath. 1974. Effect of carbon monoxide and peroxyacetyl nitrate on man's maximal aerobic capacity. J. Appl. Physiol. 36:288-293. 6. Roughton, F. J. W. 1964. Transport of oxygen and carbon monoxide. In Handb. Physiol. (Sect. 3 Respir.) 1:767-825. 7. Wittenberg, B. A., and J. B. Wittenberg. 1987. Myoglobin-mediated oxygen delivery to mitochondria of isolated cardiac myocytes. Proc. Nati. Acad. Sci. USA. 84:7503-7507. 8. Olsen, C. 1971. An enzymatic fluorimetric micromethod for the determination of acetoacetate, beta-hydroxybutyrate, pyruvate and lactate. Clin. Chim. Acta. 33:293-300. 9. Beaver, W. L., N. Lamarra, and K. Wasserman. 1981. Breathby-breath measurement of true alveolar gas exchange. J. Appl. Physiol. 51:1662-1675. 10. Sietsema, K. E., D. M. Cooper, J. K. Perloff, M. H. Rosove, J. S. Child, M. M. Canobbio, B. J. Whipp, and K. Wasserman. 1986. Dynamics of oxygen uptake during exercise in adults with cyanotic congenital heart disease. Circulation. 73:1137-1144. 11. Linnarsson, D. 1974. Dynamics of pulmonary gas exchange and heart rate changes at start and end of exercise. Acta Physiol. Scand. Suppl. 415:1-68. 12. Whipp, B. J., and K. Wasserman. 1972. Oxygen uptake kinetics for various intensities of constant-load work. J. AppL. Physiol. 33:351356. 13. Hagberg, J. M., J. P. Mullin, and F. J. Nagle. 1978. Oxygen consumption during constant-load exercise. J. Appl. Physiol. 45:381384. 14. Camus, G., G. Atchou, J. C. Bruckner, D. Giezendanner, and P. E. Di Prampero. 1988. Slow upward drift of Vo2 during constantload cycling in untrained subjects. Eur. J. Appl. Physiol. 58:197-202. 15. Sietsema, K. E., J. A. Daly, and K. Wasserman. 1989. Early dynamics of 02 uptake and heart rate as affected by exercise work rate. J. Appl. Physiol. 67:2535-2541. 16. Roston, W. L., B. J. Whipp, J. A. Davis, D. A. Cunningham, R. M. Effros, and K. Wasserman. 1987. Oxygen uptake kinetics and lactate concentration during exercise in humans. Am. Rev. Respir. Dis. 135:1080-1084. 17. Wasserman, K. 1988. New concepts in assessing cardiovascular function. Circulation. 78:1060-1071. 18. Karlsson, H., B. Lindborg, and D. Linnarsson. 1975. Time courses of pulmonary gas exchange and heart rate changes in supine exercise. Acta Physiol. Scand. 95:329-340. 19. Nery, L. E., K. Wasserman, J. D. Andrews, D. J. Huntsman, J. E. Hansen, and B. J. Whipp. 1982. Ventilatory and gas exchange kinetics during exercise in chronic airways obstruction. J. Appl. Physiol. 53:1594-1602. 20. Hughson, R. L., and G. A. Smyth. 1983. Slower adaptation of Diffusion Limitation and 02 Uptake Kinetics 1705 V02 to steady state of submaximal exercise with beta-blockade. Eur. J. AppL. Physiol. 52:107-110. 21. Casaburi, R., S. Spitzer, R. Haskell, and K. Wasserman. 1989. Effect of altering heart rate on oxygen uptake at exercise onset. Chest. 95:6-12. 22. Root, W. S. 1965. Carbon monoxide. Handb. Physiol. 1087-1098. 23. Petersen, E. S., B. J. Whipp, J. A. Davis, D. J. Huntsman, H. V. Brown, and K. Wasserman. 1983. Effects of beta-adrenergic blockade on ventilation and gas exchange during exercise in humans. J. Appl. Physiol. 54:1306-1313. 24. Wasserman, K., J. E. Hansen, D. Y. Sue, and B. J. Whipp. 1987. Principles of Exercise Testing and Interpretation. Lea & Febiger, Philadelphia. 6-20. 25. Wasserman, K. 1984. Coupling of external to internal respiration. Am. Rev. Respir. Dis. 129(Suppl.):S21-S24. 26. Cerretelli, P., R. Sikand, and L. E. Farhi. 1966. Readjustments 1706 Koike et aL. in cardiac output and gas exchange during onset ofexercise and recovery. J. Appl. Physiol. 2 1:1345-1350. 27. Diamond, L. B., R. Casaburi, K. Wasserman, and B. J. Whipp. 1977. Kinetics of gas exchange and ventilation in transitions from rest or prior exercise. J. Appl. Physiol. 43:704-708. 28. Springer, C., T. J. Barstow, K. Wasserman, and D. M. Cooper. 1990. Oxygen uptake and heart rate responses during hypoxic exercise in children and adults. Med. Sci. Sports Exercise. In press. 29. Springer, C., T. J. Barstow, and D. M. Cooper. 1989. Effect of hypoxia on ventilatory control during exercise in children and adults. Pediatr. Res. 25:285-290. 30. Ekblom, B., R. Huot, E. M. Stein, and A. T. Thorstensson. 1975. Effect of changes in arterial oxygen content on circulation and physical performance. J. Appl. Physiol. 39:71-75. 31. Woodson, R. D., R. E. Wills, and C. Lenfant. 1978. Effect of acute and established anemia on 02 transport at rest, submaximal and maximal work. J. Appl. Physiol. 44:36-43.