The series of drops in PO2 which comprise the oxygen cascade from the air to mammalian tissue can provide useful information about O2 transport during exercise in both health and disease, but the complete cascade has been challenging to study in vivo. This paper reviews a series of in vivo human knee-extensor studies which focus on the determinants of maximal O2 consumption (˙VO2max) in exercising muscle and concludes with a characterization of the complete O2 cascade in maximally exercising human muscle. Specifically, three issues have been addressed: 1) determinants of O2 extraction under conditions of very high muscle blood flow; 2) the role of O2 diffusivity in determining the maximum O2 flux rate(˙VO2max); and 3) myoglobin associated PO2 as a indicator of O2 transport and cellular respiration rate. In summary, these investigations demonstrate that in humans O2 extraction can be uncompromised despite high mass specific blood flows, perhaps in part because of an increased capillary density in exercise trained subjects. Exercise in hypoxia reduces ˙VO2max, but as calculated diffusability of O2 from blood to muscle is constant this suggests that a fixed O2 diffusivity plays a key role in limiting maximal O2 uptake. Supporting evidence of a substantial PO2 gradient from blood to myoglobin also suggests a resistance to the diffusion of O2 between red cell and sarcolemma, which may be present even at submaximal exercise. Finally, the proportionate relationship between myoglobin associated PO2 and ˙VO2max in conditions of normoxia and hypoxia additionally supports the hypothesis that maximal respiratory rate of muscle cells is limited by O2 supply.
University of California, San Diego, Department of Medicine, La Jolla, CA 92093-0623
Submitted for publication March 1997.
Accepted for publication April 1997.
The purpose of this paper is to briefly summarize data that we have collected during in vivo human muscle studies designed to clarify the determinants of skeletal muscle O2 consumption during maximal exercise. This series of investigations used the human dynamic knee-extensor model which recruits only a small muscle mass, and therefore, even at peak exercise, cardiac output is not maximal (1,3). This unique exercise paradigm allows the in vivo study of a single human muscle group under varying conditions (1), now free from the potential O2 supply-limitation dictated by a finite cardiac output. Specifically, data are presented which examine: 1) the potential limitations to O2 extraction during the very high mass specific blood flows recorded during this form of exercise(3,20,21); 2) the calculation of O2 diffusivity under conditions of hypoxia and normoxia and its role in determining human muscle ˙VO2max (19); and finally 3) myoglobin associated PO2 as an indicator of O2 transport and cellular respiration studied by combining proton magnetic resonance spectroscopy (MRS), to measure myoglobin saturation(29), and the isolated human quadriceps muscle model(1).
Subjects. This series of studies involved healthy nonsmoking male competitive athletes regularly bicycling 200-400 miles·wk-1. Informed consent was obtained according to both the University of California San Diego and the University of Pennsylvania Institutional Review Board for Human Subjects Research Committees.
Exercise model. To perform these studies several knee-extensor ergometers were constructed to replicate the device reported in previous research (1,3,21,24). One ergometer was constructed from nonmetallic materials to allow its use in both the human physiology lab in San Diego and the MRS facility in Philadelphia(Fig. 1). With these ergometers, dynamic contractions of the quadriceps muscles (60 rev·min1) caused the lower part of the leg to extend from approximately 90 to 170° flexion, and the momentum of the flywheel assisted in the passive return of the relaxed leg to the start position.
Exercise studies with blood flow and blood gas measurements. For each exercise bout, exercise work rate was increased from 25 to 50 to 75 and then to 90 and 100% of normoxic WRMAX, with data obtained at each level. Two catheters (radial artery and left femoral vein) and a thermocouple(left femoral vein) were emplaced using sterile technique as previously reported (18,21). During exercise, iced saline was infused through the femoral venous catheter and the calculation of blood flow was performed on thermal balance principles as detailed by Andersen and Saltin(3). Samples of arterial and venous blood were taken to measure PO2, PCO2, pH, O2 saturation, blood lactate, and hemoglobin concentration ([Hb]) (described in detail inreference 19). Blood O2 concentration was calculated as 1.39 × [Hb] × measured O2 saturation + 0.003× measured PO2. Arterial-venous O2 concentration([O2]) difference was calculated from the difference in radial artery and femoral venous [O2]. This difference was then divided by arterial concentration to give O2 extraction. Leg ˙VO2 was calculated as the product of arterial-venous [O2] difference and blood flow.
For MRS measurements of deoxy-myoglobin the exercise protocol was reproduced in a 2.0 Tesla Oxford imaging magnet with custom-built spectrometer(Fig. 1). A 7-cm diameter surface coil double-tuned to proton (85.45 Mhz) and phosphorus (34.59 MHz) was placed over the rectus femoris portion of the quadriceps group, approximately 20-25 cm proximal to the knee (26). Spectra were collected from the muscle region beneath the surface coil. For these studies, this “sensitive region” was 100 cc, which isolated signal detection predominantly from the region of the rectus femoris. Initial resting measurements (2 min) were recorded and then ischemic measurements (10 min) were made through the inflation of a thigh cuff to 270 mm Hg (proximal to the surface coil), with the subject positioned in the bore of the magnet, as for exercise(Fig. 1). Greater detail of the theory and methodology behind oxygen-sensitive myoglobin signals have been published(4,20). The fraction of deoxy-Mb(Fdeoxy-Mb) during exercise was determined by normalizing signal areas to the average signal obtained during the 9th and 10th minutes of the suprasytolic cuff ischemia. At rest, intramuscular oxygen depletes within 6-8 min of occlusion (5). Therefore, the signals obtained during the last 2 min of the cuff represent complete deoxygenation of myoglobin and may be used to estimate total Mb content within the muscle. Conversion to PO2 values were then calculated from the O2-binding curve for myoglobin: Equation  where 1-f is the fraction of myoglobin that is oxygenated, f is the fraction of myoglobin that is not oxygenated, and P50 is the oxygen pressure where 50% of the myoglobin binding sites are bound with oxygen. The temperature-dependent myoglobin P50 of 3.2 mm Hg was used(23), based on an approximate muscle temperature of 39°C (25).
Mean capillary PO2 and muscle O2 conductance(DO2). Using the measured intracellular PO2 values, mean capillary PO2 and DO2 were calculated by the numerical integration of measured arterial and venous blood gases as described previously (28).
Quadriceps femoris muscle mass. For consistent comparisons with the previous findings, thigh length, circumference, and skinfold measurements were used to determine thigh volume, as suggested by Andersen and Saltin and others (3,14,21). Thus, allowing an estimate of variables in mass specific terms.
Statistical analyses. Least-squares regression, repeated measures ANOVA (Tukey post hoc), and t-test analyses were computed using a commercially available software package (SPSS PC+). Variables were considered significantly different when P-value was 0.05 or less. Data are presented as the mean ± SE throughout the manuscript.
RESULTS AND DISCUSSION
High muscle blood flows and O2 extraction. Measurements of muscle blood flow in isolated muscle preparations in various animal species have revealed a spectrum of exercise blood flows ranging from 45-120 mL·min·100 g-1 (7,10,27). Original in vivo studies of human muscle blood flow placed our species at the low end of this spectrum with values ranging from 46-52 mL·min·100 g-1 (9,13,15). New approaches and the study of conscious animals have indicated that blood flows within the range of 120-260 mL·min·100 g-1 are not uncommon, while Armstrong and Laughlin (17) demonstrated that exercise blood flows of 500 mL·min·100 g-1 in the rat are indeed possible. In parallel with the acceptance of greater exercise blood flows in other species, the introduction of the dynamic human knee-extensor model with the thermodilution technique for measuring blood flow elevated reported human flows to within a new range of 250-375 mL·min·100 g-1(3,20,21,24). With these high muscle blood flows came the potential issue of limited O2 extraction because of reduced red blood cell transit time. This was implied in early single leg knee-extensor studies (Fig. 2,(3,24). However, in contrast to these initial studies that reported a maximal O2 extraction of 70% during knee-extensor exercise (1,3), we have since demonstrated much greater extraction (averages ranging from 78-87% at WRmax), accompanied by even more rapid flows, greater leg work rates, and increased leg ˙VO2 (19-21). In fact, as a function of leg ˙VO2, O2 extraction rose in a hyperbolic fashion similar in nature to previous conventional cycle ergometer measurements (Fig. 3). We hypothesize that these increased work rates and other suitably elevated physiological parameters were the result of a more rapid protocol that reduced the influence of fatigue during testing and/or the characteristics of the physically trained subjects studied(21). Specifically, if previous capillary density measurements in trained and untrained people are compared(2,6), it is likely that the subjects in this study had a capillary density of at least 400 cap·mm-1 or greater, placing their data more to the right side of Figure 4, which illustrates that adequate transit time may be available for O2 extraction, even at these high flows, if a sufficient capillarity exists(8,12).
Diffusion as a determinant of O2 movement from blood to cell. Conventional two-leg exercise in humans and the in situ canine gastrocnemius muscle preparation have provided extensive evidence that˙VO2 can reach a maximum (i.e., failure to further increase with an increase in WR) under many prescribed conditions(11,22). Furthermore, these studies illustrate that˙VO2max can be predicted for any situation by considering the quantitative interaction of the convective (eq. A, Fig. 5) and diffusive (eq. B, Fig. 5) elements in the transport of O2 from the environment to the mitochondrion (Fig. 5) (22,28). The single leg knee-extensor paradigm is unique in that it produces a situation in man where cardiac output is not maximal because of the small exercising muscle mass. Hence, muscle blood flow is not limited by cardiac output under these conditions, as it may be in conventional cycle exercise, and thus permits an increase in O2 delivery by increasing leg blood flow if necessary. With this model Rowell et al. (Fig. 6) (24) found that breathing hypoxic gas did not compromise peak ˙VO2. Leg blood flow increased to offset a lower arterial O2 content, and thus peak˙VO2 was protected. One interpretation of this finding is that if˙VO2max was achieved, an increase in O2 extraction and therefore an increase in diffusional conductance (DO2) had occurred. Yet another interpretation is that in this scenario ˙VO2max is unrelated to O2 supply. To address these questions we re-examined the response of the quadriceps to single knee-extensor exercise in both hypoxia and normoxia and employed a rapid protocol to produce both WRmax and˙VO2max in both conditions. Our study revealed a decrease in˙VO2max in hypoxia and tissue DO2 was not significantly different between normoxia and hypoxia (Fig. 6). The proportional relationships between femoral venous PO2 (and mean capillary PO2) and ˙VO2max accompanying a reduction in FIO2 are consistent with the tissue diffusion limitation of˙VO2max in normal man at maximal exercise(28). In Figure 6, the left panel contrasts our maximal data with our submaximal data at a work rate similar to that producing the peak ˙VO2 reported by Rowell et al.(24). This panel illustrates the small reduction in venous PO2 from submaximal to maximal O2 in our subjects and the similarity between our submaximal data and the peak values of Rowell et al.(24). These data strongly suggest that previous knee-extensor studies may not represent ˙VO2max for this muscle, and therefore our analysis of the determinants of maximal O2 transport is not applicable to data collected in previous studies. The estimation of DO2 depends on the assumption that, at ˙VO2max, the residual O2 in the muscle venous blood is the result of diffusion-limited O2 efflux and not perfusional shunts or blood flow-O2 heterogeneity (16). Thus from these data and within the bounds of these assumptions, it can be concluded that the human quadriceps muscle conforms to the DO2 limitation to ˙VO2max theory during maximal knee-extensor exercise. This implies that when O2 supply to a highly perfused skeletal muscle performing maximum exercise is reduced, factors which determine diffusional conductance of O2 from blood to muscle play a key role in determining ˙VO2max.
Myoglobin PO2 as an indicator of human cellular respiration. The assumption that cellular PO2 is close to zero in maximally exercising muscle is essential for the hypothesis that O2 transport between blood and mitochondria has a finite conductance that determines maximum O2 consumption (28). By combining the human knee-extensor exercise model (1) and proton MRS technology to detect myoglobin saturation(29), it has been possible to study the complete series of oxygen gradients from ambient air to human muscle in vivo during both maximal and submaximal exercise (20)(Fig. 7). With vascular measurements, it was determined that at the reduced hypoxic leg ˙VO2max both venous PO2(17.4 (hypoxia, H) and 22.1 mm Hg (normoxia, N)) and calculated mean capillary PO2 (29.7, H and 37.5 mm Hg,N) were significantly lower than at ˙VO2max in normoxia. At each submaximal WR, below 90%, the calculated DO2 was elevated in hypoxia in comparison with normoxia (20), but there was no significant difference in DO2 between normoxia at WRmax(36.3, H and 35.3 mL O2·min-1·mm Hg-1, N). This was the case no matter whether mean capillary or femoral venous PO2 was used to illustrate muscle DO2 and so was consistent with our previous findings illustrated inFigure 6 (19). The proton MRS indicated that during unweighted knee-extensor exercise the deoxy-myoglobin signal rose to an average of 38% of the maximal deoxy-Mb signal (PMbO2= 5 Torr). In normoxia, as exercise progressed the deoxy-Mb signal increased rapidly (within 20 s) to ≅50% of the maximum signal (PMbO2 = 2.1 mm Hg) and maintained this value through WRMAX, (Fig. 8). (Hypoxic response is discussed below). In both conditions, the cessation of exercise, after reaching WRMAX, produced a rapid reduction in the deoxy-myoglobin signal indicating a large increase in PMbO2, again within 20 s.
These data illustrate that myoglobin becomes significantly desaturated at submaximal exercise levels (≅ 50% of ˙VO2max) and remains at this level of desaturation even as leg ˙VO2 is increased to maximum(Fig. 7). However, a secondary observation was that in hypoxia the degree of myoglobin desaturation was significantly greater than in normoxia and that this difference in O2 saturation was still evident at leg ˙VO2max (Fig. 8). It is not apparent from these data why at normoxic WRmax myoglobin associated PO2 did not fall to the level reached in hypoxia and why under both conditions desaturation was far less at WRmax than under conditions of cuff occlusion. In an effort to reconcile these questions, Figure 9 illustrates that the present data support the hypothesis that leg˙VO2max is dependent on myoglobin associated PO2(Fig. 9B) and may represent an in vivo correlate(in myocytes) of the effect of O2 tension on cellular respiration rate as previously described in vitro by Wilson et al. (in kidney cells)(30) (Fig. 9A). The proportional relationship illustrated between leg ˙VO2 and measured myoglobin associated PO2 and that this relationship, if continued, would pass through the origin (Fig. 9B) are similar to the findings of Wilson et al. (30) who found a hyperbolic relationship between O2 tension and cellular respiratory rate in kidney cells(Fig. 9A). It is speculated that these findings may represent initial data in the development of a hyperbolic relationship betweenin vivo muscle ˙VO2 and intracellular PO2, supporting the concept that maximal respiratory rate (˙VO2max) is limited by O2 supply (Fig. 9B). We hypothesize that our data may describe a similar relationship to those of Wilson et al.(30) by reconciling the hyperbolic expression of O2 utilization in Figure 9A with the linear expression of O2 transport (equation B Fig. 5) as theoretically illustrated in Figure 9C. Thus, when˙VO2max is plotted against PMITO O2 (equation B,Fig. 5) it is a straight line of similar slope(DO2) in normoxia and hypoxia, but with a lower intercept in hypoxia caused by the lower PCAPO2 at ˙VO2max(Fig. 9C). The intersection of these lines with the intrinsic mitochondrial ˙VO2/PO2 hyperbolic relationship shows how the present myoglobin-associated PO2 data fit with O2 supply dependence of ˙VO2max in intact normal man. The conclusions are identical, but the data are essentially independent of those relating˙VO2max to mean capillary PO2 described earlier. These data provide evidence that there is a substantial O2 gradient from blood to tissue, suggesting a resistance to the diffusion of O2 between red cell and sarcolemma, and that this is present even during submaximal exercise. As WR is increased and O2 demand is elevated, presumably by increasing the area available for diffusion by simultaneously increasing both blood flow and capillary recruitment. Thus, the ability to measure tissue PO2 in hypoxia and normoxia reveals that at muscle ˙VO2max, tissue DO2 is constant and mean capillary PO2, femoral venous PO2, myoglobin PO2, and ˙VO2max are each reduced proportionately in hypoxia, supporting the concept that O2 supply plays a role in determining ˙VO2max.
The experiments reported in this article would not have been possible without the collaboration of Drs. Knight, D. C. Poole, B. Grassi, E. C. Johnson, S. S. Kurdak, M. C. Hogan, K. F. Kendrick, E. A Noyszewski, J. S. Leigh, B. K. Erickson, and P. D. Wagner.
Funding was provided by NIH HL17731, RRO2305, and Dr. Richardson was funded by a Parker B. Francis Fellowship in Pulmonary Research.
Address for correspondence: Russell S. Richardson, Ph.D., Department of Medicine 0623, 9500 Gilman Drive, University of California, La Jolla, CA 92093-0623. E-mail: firstname.lastname@example.org.
1. Andersen, P., R. P. Adams, G. Sjogaard, A. Thorbe, and B. Saltin. Dynamic knee extension as a model for study of isolated exercising muscle in humans. J. Appl. Physiol.
2. Andersen, P. and J. Henricksson. Capillary supply of the quadriceps femoris muscle in man: adaptive response to exercise. J. Physiol.
3. Andersen, P. and B. Saltin. Maximal perfusion of skeletal muscle in man. J. Physiol.
4. Bertinini, I. and C. Luchinat. NMR Paramagnetic Molecules in Biological Systems
. Menlo Park, CA: The Benjamin/Cummings Publishing Company, 1986.
5. Blei, M. L., K. E. Conley, and M. J. Kushmerick. Separate measures of ATP utilization and recovery in human skeletal muscle. J. Physiol.
6. Brodal, P., F. Ingjer, and L. Hermansen. Capillary supply of skeletal muscle fibers in untrained and endurance-trained men. Am. J. Physiol.
7. Folkow, B. and H. D. Halicka. A comparison between red and white muscle with respect to blood supply, capillary surface area and oxygen uptake during rest and exercise. Microvasc. Res.
8. Gayeski, T. E. I., W. J. Federspiel, and C. R. Honig. A graphical analysis of the influence of red cell transit time, carrier free layer thickness, and intracellular PO2 on blood tissue O2 transport.Adv. Exp. Med. Biol.
9. Grimby, G., E. Haggendal, and B. Saltin. Local xenon 133 clearance from the quadriceps muscle in man. J. Appl. Physiol.
22:305-310, 1970: 305-310, 1967.
10. Hogan, M. C., S. S. Kurdak, R. S. Richardson, and P. D. Wagner. Partial substitution of red cells with free hemoglobin solutioin does not improve maximal O2
uptake of working in situ
dog muscle. In: Advances in Experimental Medicine and Biology.
M. C. Hogan, Mathieu-Costello, O., Poole, D. C., and Wagner, P. D.(Eds). New York: Plenum Press, 1994.
11. Hogan, M. C., J. Roca, J. B. West, and P. D. Wagner. Dissociation of maximal O2
uptake from O2
delivery in canine gastrocnemius in situ. J. Appl. Physiol.
12. Honig, C. R., T. E. J. Gayeski, and K. Groebe. Myoglobin and oxygen gradients. In: The Lung: Scientific Founditions
. R. G. Crystal, and J. B. West (Eds). New York: Raven Press, 1991, pp. 1489-1496.
13. Humphreys, P. W. and A. R. Lind. The blood flow through active and inactive muscles of the forearm during sustained handgrip contractions. J. Physiol.
14. Jones, P. R. M. and J. Pearson. Anthropometric determination of leg fat and muscle plus bone volumes in young male and female adults. J. Physiol.
15. Jorfeldt, L. and J. Wahren. Leg blood flow during exercise in man. Clin. Sci.
16. Piiper, J. and P. Haab. Oxygen supply and uptake in tissue models with unequal distribution of blood flow and shunt.Respir. Physiol.
17. Pivarnik, J. M. and N. W. Sherman. Responses of aerobically fit men and women to uphill/downhill walking and slow jogging.Med. Sci. Sports. Exerc.
18. Poole, D. C., G. A. Gaesser, M. C. Hogan, D. R. Knight, and P. D. Wagner. Pulmonary and leg ˙VO2
during submaximal exercise: implications for muscular efficiency. J. Appl. Physiol.
19. Richardson, R. S., D. R. Knight, D. C. Poole, et al. Determinants of maximal exercise VO2
during single leg knee extensor exercise in man. Am. J. Physiol.
20. Richardson, R. S., E. A. Noyszeski, K. F. Kendrick, J. S. Leigh, and P. D. Wagner. Myoglobin O2
desaturation during exercise: evidence of of limited O2
transport. J. Clin. Invest.
21. Richardson, R. S., D. C. Poole, D. R. Knight, et al. High muscle blood flow in man: is maximal O2
extraction compromised?J. Appl. Physiol.
22. Roca, J., M. C. Hogan, D. Story, et al. Evidence for tissue diffusion limitation of ˙VO2max
in normal humans. J. Appl. Physiol.
23. Rossi-Fanelli, A., and E. Antonini. Studies of the oxygen and carbon monoxide equilibria of human myoglobin Arch. Biochem. Biophys.
24. Rowell, L. B., B. Saltin, B. Kiens, and N. J. Christensen. Is peak quadriceps blood flow in humans even higher during exercise with hypoxemia? Am. J. Physiol.
25. Saltin, B. and L. Hermansen. Esophageal, rectal and muscle temperature during exercise. J. Appl. Physiol.
26. Schnall, M. D., V. H. Subramanian, J. S. Leigh, and B. Chance. A new double-tuned probe for concurrent 1H and 31P NMR. J. Mag. Reson.
27. Terjung, R. L. and B. M. Engbretson. Blood flow to different rat skeletal muscle fiber type sections during isometric contractions in situ. Med. Sci. Sports. Exerc.
28. Wagner, P. D. Gas exchange and peripheral diffusion limitation. Med. Sci. Sports. Exerc.
29. Wang, Z., E. A. Noyszewski, and J. S. Leigh. In vivo
MRS measurement of deoxymyoglobin in human forearms. Magn. Reson. Med.
30. Wilson, D. F., M. Erecinska, C. Drown, and I. A. Silver. Effect of oxygen tension on cellular energetics. Am. J. Physiol.
Symposium: New Insights into the Control of Human Muscle Blood Flow and Metabolism Studied in Vivo
Keywords:©1998The American College of Sports Medicine
HYPOXIA; MYOGLOBIN SATURATION; QUADRICEPS; HUMAN; EXERCISE