Journal Logo

BASIC SCIENCES: Integrative Physiology & Exercise

Control of Oxygen Uptake during Exercise


Author Information
Medicine & Science in Sports & Exercise: March 2008 - Volume 40 - Issue 3 - p 462-474
doi: 10.1249/MSS.0b013e31815ef29b
  • Free


As distinct from the steady state of oxygen uptake (V˙O2) that may be achieved during exercise, the dynamic transition of V˙O2 (V˙O2 kinetics), initiated at exercise onset, provides a unique window into understanding metabolic control. One major and contentious issue that has intrigued physiologists and exercise scientists for decades concerns whether, in healthy individuals, the speed of V˙O2 kinetics at the onset of exercise is limited by muscle O2 delivery or, rather, some rate limitation of the oxidative machinery itself (16,22,23,40,45,84). Resolution of this issue is essential to understanding metabolic control in health, and also for defining the mechanistic bases for the impaired (slowed) V˙O2 kinetics found in patient populations. Specifically, prevalent chronic diseases such as heart failure and diabetes may incur slowed V˙O2 kinetics, which, at a given V˙O2, mandates generation of an increased O2 deficit and exacerbates perturbations of the intramuscular milieu: that is, ↓[phosphocreatine, PCr], ↑[ADP]free, ↑[Pi] (70). These changes are associated with a reduced exercise capacity, and it is possible that a greater understanding of the causes of slowed V˙O2 kinetics in such patients will facilitate development of therapeutic strategies to reverse this problem and improve exercise tolerance and patients' quality of life.

There is a substantial body of scientific literature that has examined the site of V˙O2 kinetics limitation during moderate-intensity exercise, and the reader is referred to the excellent reviews by Whipp and Mahler (85), Hughson and colleagues (45), Hughson (40), Grassi (22,23), and Delp (16). In brief, proponents of the O2-delivery limitation theory point out that there is biochemical evidence for O2 adjusting the phosphorylation and redox potentials needed to drive oxidative metabolism (38,91,92). Furthermore, they cite studies demonstrating that the rate of rise to a given V˙O2 (i.e., V˙O2 kinetics) is slowed when muscle O2 delivery is impaired by 1) inspired hypoxia (i.e., reduced arterial O2 content (18)), 2) supine posture (41), 3) β-adrenergic receptor blockade (39), 4) transition from prior exercise (42), and 5) diseases that impair cardiovascular dynamics and pulmonary function (70,71). In contrast, proponents of the "oxidative enzyme inertia" hypothesis base their position on the following observations: 1) most studies demonstrate that muscle O2-delivery kinetics (assessed from muscle arterial inflows or cardiovascular dynamics) are appreciably faster than V˙O2 kinetics (16), and it is difficult to envisage a faster process limiting a slower one. 2) It is possible to alter V˙O2 and cardiovascular kinetics independently (95). Specifically, prior one-legged exercise speeds cardiac output and V˙O2 kinetics during a second bout with the same leg. In contrast, when the other leg performs the second exercise bout, V˙O2 kinetics are not speeded, even though cardiac output kinetics are faster in comparison with control values. 3) For moderate-intensity exercise, neither increased arterial O2 content (62) nor muscle O2 availability (24,25) speed V˙O2 kinetics, and even in the heavy/severe-intensity domains their O2 dependence may be very modest or absent (22,86). 4) The temporal correspondence between PCr and V˙O2 kinetics and their monoexponential profile support a phosphate-linked control of mitochondrial energetics (70,77). 5) Reductions in nitric oxide inhibition of mitochondrial function by L-NAME speed V˙O2 kinetics, even in the face of a potentially reduced muscle O2 delivery (50,54). 6) In muscles with a preponderance of slow-twitch fibers, microvascular oxygenation does not fall immediately or precipitously at the onset of contractions; instead, it decreases exponentially to the steady state following a 10- to 20-s delay or even exhibits a slight increase (5,6,65).

Notwithstanding the positions expressed above, both camps agree that at some point altering O2 delivery will impact V˙O2 kinetics, and this is illustrated schematically in Figure 1. Notice that, as O2 delivery is reduced moving from right to left, at some "tipping point" an O2 limitation occurs, and V˙O2 kinetics (designated as the time constant,τ) become progressively slowed. The dissent arises principally from consideration of where the exercising individual lies with respect to the tipping point. The O2-delivery proponents would place the healthy exercising individual on the ascending limb of the response, where either an increase or decrease in O2 delivery would speed or slow V˙O2 kinetics, respectively. In contrast, if V˙O2 kinetics are limited by an intrinsic sluggishness of the muscle metabolic machinery, the exercising individual would lie at some point to the right of the tipping point, and thus V˙O2 kinetics would be insensitive to either increased O2 delivery or to modest reductions in O2 delivery, provided that such reductions did not take the individual across the tipping point.

Model demonstrating the effects of altering muscle O2 delivery on V˙O2 kinetics (i.e., increased V˙O2 time constant (τ) denotes slower kinetics). Moving from right to left, note that τ is unchanged as O2 delivery decreases (O2-delivery-independent V˙O2 kinetics) until some critical "tipping point" is reached, beyond which V˙O2 kinetics become progressively slowed (larger τ) with further reductions in O2 delivery. The question posed in this review is, Where on this continuum are healthy individuals as they perform upright cycle or running exercise? Redrawn from Poole and Jones (70).

In recent years, significant developments in technology and innovative experimental models and designs have facilitated advances in our understanding of the control of V˙O2 kinetics in health and disease. This review uses Figure 1 as a framework within which to reexplore the limitations to V˙O2 kinetics during exercise, taking into account some of these latest experimental findings. As presented in Indianapolis, IN, at the ACSM Integrative Physiology of Exercise meeting in September 2006, this paper is sequenced to demonstrate how novel experimental approaches and data help reconcile (apparently) opposed viewpoints. A. M. Jones will explore the opposing viewpoints that V˙O2 kinetics are or are not limited by O2 delivery. T. J. Barstow will present near-infrared spectroscopy data suggesting that capillary blood flow kinetics are considerably slower than those of muscle blood flow measured at more remote sites, and that humans therefore may exercise in the proximity of the tipping point. Finally, P. McDonough will demonstrate that the temporal relationship between O2 delivery and V˙O2 kinetics is fiber type dependent, such that slow- and fast-twitch fibers might lie on opposite sides of the tipping point.


The concept of an O2 transport limitation to V˙O2 kinetics has traditionally been championed by Dr. Richard L. Hughson of the University of Waterloo, Canada. This section therefore attempts to provide a fair synopsis of the evidence that might be harbored in favor of this concept, which implies that there is a specific "rate-limiting step" in the O2 cascade from the lungs to the mitochondria. There are certain conditions under which it might be possible to identify an O2-transport-related factor that results in a change in V˙O2 kinetics-for example, breathing hypoxic gas (18,61,66), following β-adrenergic receptor blockade (39), or (possibly) in the transition from light- to moderate-intensity exercise (8,42,64). As indicated in the introduction, it is now widely accepted by researchers in this field that there are conditions under which O2 transport can impact V˙O2 kinetics. Controversy remains, however, over the exact nature of these conditions. Arguably, this debate can be distorted by a focus on the "either/or" nature of O2-transport versus O2-use limitations. The terms "regulation" or "modulation" of V˙O2 kinetics by O2 transport have recently been used more widely, and these might present a more appropriate way to understand the interaction of factors that establish the increase in V˙O2 during a transition to a higher work rate (40,83).

One specific example of how a change in V˙O2 kinetics might be related to a change in O2 transport is provided by the experiment conducted by Hughson and Morrissey (42) some 25 yr ago. These authors studied different exercise transitions to work rates that were strictly below the gas-exchange threshold (GET). The kinetics of V˙O2 were slower in the 40-80% GET transition than during the rest (or unloaded pedaling (43)) to 80% GET transition. The kinetics of heart rate showed a very similar response profile during these studies. Thus, Hughson and his associates proposed that the slower kinetics of V˙O2 in the 40-80% GET transition were the consequence of slower O2-transport kinetics to the exercising muscles. Subsequently, advances in Doppler ultrasound technology enabled muscle blood flow kinetics to be examined during exercise transients, with some studies indicating a relationship between V˙O2 kinetics and O2-transport kinetics, such as when supine exercise was compared with upright exercise (63). A recent study by McPhee et al. (64) has also indicated that muscle O2 transport is slower in the upper compared with the lower region of the moderate-intensity exercise domain.

The term "regulation" rather than "limitation" of V˙O2 kinetics by O2 transport is now preferred by Hughson and colleagues, because this term incorporates the obvious interaction with metabolic conditions in the determination of V˙O2, both in the steady state and across an exercise transient (40). The rate of ATP hydrolysis is precisely matched by ATP synthesis in the steady state of exercise. As exercise intensity increases, the sum of the high-energy phosphates (ATP + PCr), which might be termed the "energetic state" of the muscle (3,14), is progressively reduced. Across the transition to exercise (i.e., in the nonsteady state), there is a progressive reduction in the energetic state until the steady-state level is finally attained. The energetic state during exercise at a constant work rate can be modified by changes in O2 supply, such as with hypoxia or hyperoxia (37,61). In vivo observations that high-energy phosphates and substrate concentrations can change at the same steady-state ATP demand might be considered to be consistent with the complex regulation of metabolism by O2, as described by Wilson and Rumsey for in vitro preparations (92). It might be further argued that this must also happen in exercise transitions when the intracellular PO2 is changing from resting values above 30 mm Hg to those less than 5 mm Hg (75,76). Hughson (40) has recently proposed a multidimensional model to explain these complex interactions (Fig. 2). In this figure, the x-y plane represents the metabolic control where a constant rate of ATP flux can be achieved at lower mitochondrial PO2 by reductions in the concentrations of PCr and ATP (92). With a reduction in PO2 (as reflected by moving from point 1 to point 2 on the graph), the altered energetic state maintains the necessary balance between PO2 and the phosphorylation and redox potentials to provide sufficient drive to keep ATP production at the required level (45). The third dimension in Figure 2 (in which there is constant ATP production across the full surface) allows for the effects of interventions such as prior exercise (62), drugs to inhibit nitric oxide synthase (50), or exercise training (31), to be explored. With exercise training, for example, a muscle's metabolic state might move from point 2 to 3 on the figure. Although hypothetical, this model recognizes the complex nature of metabolic control at a constant ATP flux rate.

A hypothetical scheme is presented to show how a constant rate of ATP production across the entire surface can be accomplished by the interactions of intracellular PO2 (PintracellularO2) on the x-axis, energetic state as indicated by the concentration of ATP + PCr on the y-axis, and muscle enzyme or substrate concentration changes on the z-axis. Moving from point 1 on the graph to point 2, which would occur in hypoxia, requires reduction of the energetic state as described by Wilson and colleagues (91). Movement from point 2 to point 3 could represent exercise in the trained state, where breakdown of ATP + PCr would be less, and intracellular PO2 could be slightly higher. Figure adapted from Hughson (40).

Hughson's model can be expanded to understand V˙O2 kinetics by adding a fourth dimension (time) that considers the increase in oxidative production of ATP along with changes in intracellular PO2, the activation of enzymes, and the availability of substrate. Across the rest-to-exercise transition, metabolism is controlled by a complex series of interactions; however, one of the regulating factors is PO2. As described by Ereciñska and Wilson (19), "although a fall in oxygen tension does cause a decrease in the rate of ATP synthesis, it does not affect the rate of ATP utilization. However, such a situation leads to an immediate decline in the [ATP]/[ADP][Pi] which induces, through the near equilibrium relations in the first two sites of oxidative phosphorylation, an increase in reduction of cytochrome c and activation of cytochrome c oxidase. The consequent enhancement of respiration proceeds to the point at which the rate of ATP synthesis again matches the rate of ATP utilization." Therefore, in contrast to the concept of an "O2-delivery-independent zone" in Figure 1, Hughson and his associates would argue that O2always regulates the kinetics of V˙O2. The model proposed in Figure 2 also highlights why researchers might have difficulty in detecting the influence of O2 transport on V˙O2 kinetics (45). Under conditions of typical intracellular PO2 the relatively flat surface on the three dimensional plot (e.g., to the right of point 1 on Fig. 2) will have small, but real, effects on metabolic control compared with the steeper slope (e.g., from points 1 to 2). On this basis, Hughson and coworkers have proposed that all exercise transitions are in the O2-delivery-dependent zone depicted in Figure 1.


In this section, evidence against O2 as a limitation to V˙O2 kinetics (with appropriate caveats) and evidence for an intracellular control of V˙O2 kinetics will be reviewed. The important influence of muscle fiber type and motor unit recruitment profiles in determining the characteristics of the V˙O2 response to exercise will also be highlighted.

There is no question that muscle O2 availability has the potential to influence metabolic control: as can be seen from equation 1, if mitochondrial O2 supply is truly insufficient, the rate of oxidative metabolism will be restricted, and this will be manifest as slower V˙O2 kinetics across a metabolic transient.

While cellular metabolic adjustments can occur to maintain ATP flux rate in the face of falling PO2 (91,92), muscle O2 supply can only be considered "limiting" if V˙O2 kinetics are demonstrably impaired. There are a number of situations in which muscle O2 availability might be considered to be among the factors responsible for relatively slow V˙O2 kinetics: for example, during exercise where the muscle perfusion pressure is reduced (supine and prone leg exercise; arm exercise performed above the level of the heart (44,46)), muscle O2 supply is deliberately restricted (ischemia, hypoxia, β-blockade (18,39)), and in older age and a variety of disease conditions (chronic heart failure, peripheral vascular disease, diabetes (70,71)). These situations would clearly lie in the O2 delivery-dependent zone shown in Figure 1. However, the contention in this section of this review is that muscle O2 delivery does not limit V˙O2 kinetics during most normal forms of exercise (i.e., those in which the heart is positioned above the bulk of the working muscle mass such as in running, cycling, and rowing) in most normal healthy, physically active people below the age of approximately 50 yr, and even when the exercise intensity is high (> 80% of the V˙O2max).

The first line of evidence against O2 as a limitation to V˙O2 kinetics under the conditions described above is that bulk muscle blood flow kinetics (and thus the kinetics of muscle O2 delivery) are almost always faster than muscle V˙O2 kinetics during low- as well as high-intensity exercise (4,30,63). It is difficult to envisage a situation in which the kinetics of a relatively slow physiological process can be limited by a faster one, unless, of course, there are substantial O2-distribution heterogeneities across the recruited muscle(s). Also, although experimental interventions that might be expected to reduce muscle O2 supply during upright exercise in healthy subjects have the potential to slow V˙O2 kinetics, this is not consistently the case: following blood withdrawal (13), hemodilution (7), and with the application of lower-body positive pressure (90), V˙O2 kinetics are not significantly altered. The fact that compensatory adjustments can apparently be made to maintain adequate muscle O2 supply in these circumstances indicates that the exercise must be located some distance to the right of the tipping point in Figure 1. In any case, "proof" that muscle O2 delivery is limiting in these conditions requires that V˙O2 kinetics are speeded when O2 supply is increased-evidence that is almost completely absent. Classic studies by Grassi and colleagues in the isolated in situ canine gastrocnemius preparation have shown that setting muscle blood flow at the required steady-state level across a metabolic transient does not alter muscle V˙O2 kinetics at 60% V˙O2max (24) and that it only barely does so at 100% V˙O2max (22,23). Moreover, enhancing the potential for muscle O2 diffusion through a combination of pump perfusion, hyperoxia, and administration of a drug to right-shift the HbO2 dissociation curve, had no further effect on V˙O2 kinetics (25). Similarly, in humans, improving the potential for increased muscle O2 availability through administration of recombinant human erythropoietin (89), or having subjects inspire a hyperoxic gas mixture (62,86), does not speed phase II pulmonary V˙O2 kinetics, even during high-intensity exercise. If muscle were really lacking O2 in these conditions, it is difficult to appreciate how it would not avail itself of the increased O2 afforded by these interventions.

One intervention that has received considerable attention with regard to its effect on the kinetics of V˙O2 is that of prior high-intensity ("priming") exercise. In situations where muscle O2 supply might be expected to be limiting in the control condition (e.g., during arm exercise or leg exercise performed in the supine position, and in senescent or sedentary/unfit subjects), there is some evidence that the performance of prior exercise, which will increase muscle vasodilatation and right-shift the HbO2 dissociation curve, is associated with a speeding of the phase II V˙O2 kinetics (21,47,58,80). During upright cycle exercise in young healthy subjects, however, the overwhelming majority of studies (21 out of 22 at the time of writing) have reported that the performance of prior exercise does not speed the phase II V˙O2 kinetics during subsequent high-intensity exercise (12). Rather, the principal effect of prior exercise seems to be to reduce the magnitude of the so-called V˙O2 slow component, which is characteristic of exercise performed above the lactate threshold (LT or GET), and thus to bring the overall V˙O2 response back toward a monoexponential profile (11,12) (Fig. 3). Although these data do not rule out a possible role for (regional) muscle O2 insufficiency in the development of the V˙O2 slow component, they do suggest that the fundamental response of V˙O2 across a metabolic transient is principally regulated by factors more proximal to the contracting muscle. Similarly, the performance of prior exercise has not been found to alter the phase II V˙O2 kinetics during subsequent moderate-intensity exercise (12), although, interestingly, there are two recent exceptions to this finding (33,34). It should be noted, however, that any observation of a speeding of the phase II V˙O2 kinetics following prior exercise can never be attributed solely to enhanced muscle O2 supply, because numerous muscle metabolic and other factors will also be altered by the intervention (12,33).

Influence of prior high-intensity (priming) cycle exercise on pulmonary V˙O2 kinetics during subsequent high-intensity cycle exercise in a representative subject from the study of Burnley et al. (11). Closed symbols represent responses during the first exercise bout, and open symbols represent responses during the second exercise bout. The top panel demonstrates that the kinetics of the V˙O2 response were faster overall in the second (primed) bout. However, the time constant describing the adaptation of V˙O2 during phase II of the response was similar in the first and second bouts (bout 1 τ: 22 s; bout 2 τ: 25 s). The facilitated V˙O2 response in the second bout therefore resulted from an increased amplitude of the primary or fundamental component and a reduced amplitude of the slow component. The close correspondence ofthe V˙O2 kinetics in phase II is underlined in the middle panel, in which the V˙O2 responses in the two bouts are normalized to the amplitude of the primary response. The bottom panel shows the mean integrated electromyogram response of four lower-limb muscles during the first and second exercise bouts. The profiles in the first and second bouts mirror those for V˙O2 suggesting that (changes in) motor unit recruitment might be mechanistically associated with the (altered) V˙O2 responses.

Mitochondrial respiration is intimately linked to the rate of muscle ATP hydrolysis, and one or more of the reactants of this process (e.g., [ADP], [Pi], phosphorylation potential, and/or [PCr] and [Cr]) is/are thought to activate oxidative phosphorylation through feedback control. Rossiter et al. (77-79) have provided evidence consistent with this theory by demonstrating close agreement between muscle [PCr] kinetics (as estimated using 31P-MRS techniques) and pulmonary V˙O2 kinetics during both moderate- and heavy-intensity exercise. Moreover, Kindig et al. (52) report that acute inhibition of creatine kinase (CK) in isolated Xenopus myocytes led to significantly faster intracellular PO2 kinetics (equivalent to faster V˙O2 kinetics in this model). These data indicate that the CK reaction buffers changes in [ADP] across a metabolic transient, thus attenuating one of the principal signals responsible for an acceleration of oxidative phosphorylation.

The kinetics of V˙O2 therefore seem to be principally under feedback control through the CK reaction (it is worth noting here that this will also be true even when other limitations to the V˙O2 response, such as O2 availability, are "superimposed" (36)). However, it is possible that other factors also contribute to the inertia of muscle oxidative metabolism that is evident in the transition from a lower to a higher work rate. Theoretically, any rate-limiting or flux-generating enzyme catalyzing a nonequilibrium reaction might limit the rate of oxidative metabolism. Pharmacological activation of pyruvate dehydrogenase with dichloroacetate (DCA) reduces substrate-level phosphorylation during subsequent exercise, suggesting an enhancement of the contribution of oxidative phosphorylation to energy turnover (32). Although to date this intervention has not been shown to speed muscle or pulmonary V˙O2 kinetics, there are suggestions that muscle efficiency might be improved with DCA, thus reducing the amplitudes of the fundamental and/or slow components of V˙O2 and reducing the magnitude of the O2 deficit (26,48,79). Another possible limitation to the dynamics of V˙O2 is the potentially pernicious influence of nitric oxide (NO) on mitochondrial function. In addition to its well-known role in the regulation of muscle vasodilatation, NO has the potential to inhibit several mitochondrial enzymes and to compete with O2 for the binding site at cytochrome c oxidase (9). Recent studies in horses (54) and in humans (50,87) have shown that inhibition of NO synthesis with l-NAME results in a significant speeding of phase II V˙O2 kinetics. The effect, at least in humans, seems to be greatest at higher work rates, suggesting that it might be especially pronounced in type II muscle fibers (87); if so, this might also explain why a speeding of V˙O2 kinetics was not observed following L-NAME administration in isolated canine muscle known to have a high percentage of type I fibers (23). As mentioned earlier, NO also plays an important role in muscle vasodilatation, and thus the inhibition of NO synthesis might impair muscle blood flow. The speeding of phase II V˙O2 kinetics with L-NAME therefore potentially provides simultaneous evidence for an NO-linked muscle metabolic limitation to V˙O2 kinetics and against an important role for O2 supply, at least under the conditions of these studies.

There is often a tendency for phase II V˙O2 kinetics to become slower at higher work rates, especially those that are above the LT (70). Although this has been attributed to an (increasing) muscle O2-delivery limitation by some authors (40), another explanation is that the slower overall V˙O2 kinetics reflect the increasing contribution of muscle fibers that are higher in the recruitment hierarchy (i.e., type II fibers) to force production. There is evidence to suggest that these higher-order fibers might have slower V˙O2 kinetics (and also lower efficiency) relative to earlier-recruited fibers (49,59,65). Alterations in motor unit recruitment might underpin the reduced V˙O2 slow component (and faster overall V˙O2 kinetics) observed for the same work rate following interventions such as endurance training, hyperoxic gas breathing, and priming exercise (49) (Fig. 3), as well as the greater V˙O2 slow component observed for the same work rate following glycogen depletion of the type I muscle fiber population (59). When exercise is initiated from a higher compared with a lower baseline metabolic rate, markedly slower phase II V˙O2 kinetics and a higher V˙O2 response gain are typically reported (8,43,88). Again, these observations are consistent with the metabolic responses that would be expected when a population of higher-order fibers is recruited to meet the augmented muscle force-production requirements, and they are not necessarily indicative of any impairment in muscle O2 delivery. It should be acknowledged, however, that these two effects might not be mutually exclusive (6,49,65).


Insight into the factors that determine the rate of rise in muscle oxygen uptake (V˙O2m) following exercise onset ultimately requires assessment of this process where it is actually occurring, that is, at the capillary/myocyte interface. However, with the exception of isolated muscle/capillary network preparations (5), which allow for measurements of intracapillary oxygen tension profiles and time course of red blood cell flux, observations have been limited to those made distant from the capillary bed (e.g., blood flow and oxygen extraction measured across an exercising muscle or limb (4,24,25)). The first two sections of this review elegantly summarized our understanding of the controlling features of V˙O2m based primarily on findings from distant sites (conduit artery, lungs, etc.). However, we sought a methodology that would allow us to "observe" capillary gas exchange noninvasively in human skeletal muscle during exercise, in a somewhat analogous manner to the direct observations of microvascular PO2 (PmvO2) in isolated muscles (5).

Near-infrared spectroscopy (NIRS) noninvasively measures oxygenated and deoxygenated (HHb) forms of hemoglobin + myoglobin (Hb + Mb) in the microcirculation and myocytes of muscle tissue. As such, HHb has been used to estimate fractional O2 extraction within the microcirculation (15,20,29). Using technology that provides either an approximation (15,29) or an absolute concentration (20), the response of HHb demonstrates a similar pattern following exercise onset: there is a period of 5-10 s of no change, followed by a second phase of rapid increase (average time constants ranged from 6-10 s) to a third, relatively constant plateau (Fig. 4 middle panel). The initial period of relatively constant HHb, similar to that observed in microvascular PO2 in isolated muscle (5), coupled with the observation that V˙O2m begins to rise immediately with exercise onset with no apparent time delay (53,70), implies that during the first few seconds after exercise onset, oxygen delivery (QO2) is precisely matched to the rising V˙O2m, such that fractional O2 extraction remains constant (dashed arrows in Fig. 4). Whether this matching is fortuitous or the result of the integration of muscle pump and/or rapid vasodilatory mechanisms (16,83,84) remains to be elucidated. The rapid increase in HHb during the second phase (Fig. 4, solid arrows) reflects a transient imbalance between V˙O2m and QO2, which seems to be restored during the third, plateau, phase (Fig. 4, dotted arrows). The kinetics of adjustment of HHb during this second phase, whether expressed as a time constant or as an MRT (time constant + time delay) are consistently found to be faster than those of phase II pulmonary V˙O2 (15,20,29,33,34), suggesting that HHb per se is not a good surrogate for V˙O2m kinetics. The likely reason for this lack of agreement between HHb and V˙O2 kinetics, and the implications for the plateau phase, are discussed below.

Data showing both the derivation of capillary blood flow ( cap) and the underlying phenomena for the response of deoxy (Hb + Mb) (HHb) for the transition from light to moderate exercise. The upper panel is estimated muscle V˙O2 (V˙O2m), derived from the phase II kinetics of pulmonary V˙O2. The middle panel shows the measured response of HHb, and the bottom panel is the resulting calculated cap (as V˙O2m/HHb). Following exercise onset, there is a period of 5-10 s, during which time HHb remains relatively constant (dashed arrows). Given that V˙O2m is rising exponentially with little or no delay, this implies that cap rises in proportion to V˙O2m. The rapid increase in HHb during the second phase of the response (solid arrows) suggests that the increase in cap slows down, such that O2 extraction has to increase to facilitate the continued rise in V˙O2m. However, in another few seconds, HHb plateaus. Given that this occurs sooner than the steady state for V˙O2m is attained, this implies that cap continues to increase in a fashion again tightly coupled to the rising V˙O2m. The suggestion is offered that at least during this region, the increase in blood flow may be integrated with the rise in metabolic rate through an oxygen-sensing mechanism, such as the release of NO from hemoglobin as PO2 falls along the capillary.

As noted in the previous sections, the kinetics of adjustment of conduit artery blood flow are either faster than (24,57,63,70) or similar to (27,30) those of V˙O2m. Although these data have been used to speak against O2 flow limitation to V˙O2(m) kinetics (i.e., points to the right of the tipping point in Fig. 1), they do not necessarily provide evidence for the lack of an O2 delivery limitation at the level of the microcirculation. Further insights into the underlying integrated responses implied by the HHb dynamics have been attained by the development of the ability to estimate the kinetics of capillary blood flow (cap) (20). By rearranging the Fick equation, cap can be estimated noninvasively as cap = V˙O2m/(CaO2 − CvO2), assuming that the phase II kinetics of pulmonary V˙O2 are a good surrogate for V˙O2m (30,77) and that the deoxy (Hb + Mb) signal (HHb) from NIRS is proportional to (CaO2 − CvO2) (15,20,29,35). Based on these assumptions, the following results have been observed: 1) for both cycling (20) and knee extension exercise (35), the estimated cap kinetics are biphasic, reflecting an initial rapid rise to plateau during the first 10-15 s, followed by a slower exponential rise to a steady-state level for moderate exercise (Fig. 4, lower panel). This biphasic nature of the cap kinetics is similar to that observed in capillaries of isolated rat spinotrapezius muscle (55) and in conduit arteries (74,81), which reinforces the appropriateness of the NIRS approach to estimate cap. 2) The kinetics of conduit (femoral) artery blood flow are significantly faster than those of cap (35). This finding has significant implications for both the control, and initial distribution, of limb/muscle blood flow following exercise onset. First, they suggest that measurement of blood flow kinetics in the conduit artery upstream of the contracting muscle(s) may not reflect the kinetics of red blood cell flux in the microcirculation where gas exchange is occurring. Second, the apparent disparity between conduit artery kinetics (faster) and capillary kinetics (relatively slower) suggests that control of the distribution of flow within the contracting limb/muscle is an important factor in determining the kinetics of capillary RBC flux. 3) The overall kinetics of cap were either similar to (cycling (20)) or slower than (knee extension (35)) those of V˙O2m.

These latter two findings suggest the following scenario: immediately on initiation of muscle contractions, conduit artery and capillary blood flow increase in concert with the rising V˙O2m. At the level of the capillaries, this increased flow matches V˙O2m, such that HHb changes little. Within a few seconds, this initial blood flow response begins to plateau, necessitating an increase in O2 extraction to sustain the rising V˙O2m. Within a few more seconds, the blood flow response is again coupled to the V˙O2m, and HHb becomes relatively constant. The overall kinetics (as mean response time MRT) of cap are similar to those of V˙O2m (20,35) and imply that there may not be as much of a reserve of O2 delivery as has been intuited by the generally faster kinetics observed in conduit arteries (i.e., even in healthy subjects performing moderate exercise, the relationship is closer to the tipping point in Figure 1 than previously thought). Further, this implies that disease processes such as diabetes or peripheral arterial disease might not need to dramatically slow O2 delivery in order to have a pernicious effect on V˙O2m kinetics. This coupling of blood flow to the rising V˙O2m is impressive, and it suggests a heretofore unappreciated dynamic integration between V˙O2m and blood flow. While the precise mechanism(s) remain to be elucidated, one likely candidate is some form of oxygen sensing within the microcirculation (10), which could be effected through the release of nitric oxide (NO) from S-nitrosohemoglobin as oxygen partial pressure falls in the capillaries (1).

It should be noted here that the increase in HHb (as fractional O2 extraction, or the fall in microvascular PO2) with exercise, either transiently or in the steady state, is not an obligatory sign of an inadequate blood flow response (15,51). Rather, the increase in HHb with increasing metabolic rate is predicted by the positive intercept of the blood flow/V˙O2 relationship, as observed across the entire body (as cardiac output and pulmonary V˙O2), exercising limbs (e.g., femoral artery blood flow and leg V˙O2 (69)), and even individual muscles.


Up to this point in the review, the methodologies used have not facilitated differentiating, with certainty, the dynamic responses among different fiber types. Thus, when the linear relationship between blood flow () and metabolic rate (V˙O2) is described for the exercising legs as = (S × V˙O2) + I, where S represents the slope (i.e., 5.3; data from ref. (56)) and I the intercept (i.e., 2.8 L·min−1; data from Knight et al. (56)), the tacit presumption is made that all fibers and microvascular units operate in close proximity to these mean values (69). Moreover, inter-fiber-type differences in the dynamics of the increased immediately following exercise onset, particularly as this relates to that of V˙O2, have not been explored.

It is well established, however, that human leg muscles constitute a mosaic of slow-twitch (type I) and fast-twitch (type II) fibers, which are discriminated by very different vascular () and metabolic (V˙O2) control processes. Understanding the net effect of these control processes is crucial, in part, because it is the instantaneous balance between O2 delivery (QO2) and V˙O2 that determines the microvascular O2 pressure (PmvO2), which, in turn, drives blood-myocyte O2 flux and (via its effect on intracellular PO2) modulates the intracellular energy state (3,38,65,91,92).

Investigations in animals at rest and during exercise indicate that there is a pronounced heterogeneity of among different muscles and fiber types (2,68,72). This spatial stratification of is likely attributable to those mechanisms that control arteriolar vasodilation across the different fiber types. Specifically, endothelial nitric oxide synthase mRNA expression, as well as sensitivity and maximal responsiveness to endothelium-dependent vasodilation, is far greater in arterioles from slow- compared with fast-twitch muscles (17,93,94). Moreover, first-order arterioles from fast-twitch muscles elicit a higher sensitivity-that is, a greater vasoconstriction to noradrenaline (16). From these observations, it would be predicted that contraction-induced hyperemia would be exacerbated in muscles comprising slow-twitch, highly oxidative fibers. Whether this impacts the PmvO2 depends, of course, on the relationship between any augmentation of the response with that of V˙O2. For example, if dynamics are speeded equivalently with those of V˙O2, the PmvO2 profile would remain unchanged. The same could be said for the steady-state response if the contracting absolute and V˙O2 are not different. Based on the above considerations, Behnke et al. (6) and McDonough et al. (65) judiciously selected rat hindlimb muscles for their fiber-type content to test the hypothesis that, following the onset of contractions, the exercising response would be faster and more robust (i.e.,quantitatively greater), compared with that of V˙O2, in slow- compared with fast-twitch muscles. The prediction, therefore, was that following the onset of contractions, PmvO2 would fall faster and farther (i.e., to lower steady-state contracting values) in the fast-twitch gastrocnemius (65) and peroneal (6) muscles than the slow-twitch soleus.

As clearly depicted in Figure 5 for the soleus and mixed and white portions of the gastrocnemius, these predictions were borne out. In the soleus muscle, the response was sufficiently vigorous that PmvO2 remained close to resting values for approximately 15-30 s before falling exponentially with a time constant of 25-30 s to steady state at 20 mm Hg. In contrast, PmvO2 in the fast-twitch mixed and white gastrocnemius plummeted immediately (time delay, 5-9 s; time constant, 6-11 s) following the onset of contractions to steady state at approximately 10 mm Hg, which was only half that found in the soleus. In addition, in the nonsteady state, gastrocnemius PmvO2 actually fell below that seen in the subsequent steady state. These responses are symptomatic of a relatively parsimonious QO2-to-V˙O2 in fast-twitch compared with slow-twitch muscles.

Microvascular O2 partial pressure (PmvO2) response for rat soleus (sol, slow twitch), as well as mixed (MG, fast twitch) and white gastrocnemius (WG, fast twitch) following the onset of 1-Hz contractions. Thin lines, real data; thick lines, model fits. Note the longer time delay, slower subsequent fall, and higher steady-state PmvO2 for sol than for MG or WG. Also note the biphasic response of PmvO2 in MG and WG that temporarily reaches PmvO2 values below the subsequent steady state. This figure supports the contention that O2 delivery may be limiting V˙O2 kinetics in fast-twitch (MG, WG) but not slow-twitch (sol) muscles. Reproduced from McDonough et al. (65), with permission.

An additional interesting feature of the slow- versus fast-twitch fiber dichotomy was evidenced when these same muscles were compared during the steady state of low- and high-intensity contractions. Moving from the lower to higher intensity, the soleus increased its V˙O2 predominantly by increasing , whereas the gastrocnemius, in the face of a relatively small increase of at the higher intensity, increased fractional O2 extraction through an elevation of its diffusional O2 conductance (DO2). Thus, while transitioning from contractions of low to higher intensity, when the soleus increased DO2 by only 30%, the gastrocnemius increased it by 60% (white) to 120% (mixed), and this accounted for a further fall in steady-state PmvO2 in the mixed gastrocnemius (65).

The first time that PmvO2 was measured in muscle following the onset of contractions, Behnke and colleagues (5) hypothesized that, if O2 delivery (QO2) were limiting, PmvO2 would be expected to fall precipitously at exercise onset. Furthermore, PmvO2 should fall transiently below steady-state levels reflecting the lag in the hyperemic response prior to achieving steady-state values commensurate with the exercise intensity. These features of the response were not found in the spinotrapezius, which is a specialized postural muscle of mixed fiber type (5). However, they are clearly evident in the mixed and white gastrocnemius shown in Figure 5, but not in the slow-twitch soleus. These observations suggest that, if the exercise bout is sufficiently intense that it recruits a mosaic of slow-and fast-twitch fibers, the latter may be O2-delivery limited (ascending limb of Fig. 1). Moreover, for those fast-twitch fibers or muscle portions in which PmvO2 has reached very low levels, blood-myocyte O2 diffusion may be compromised, and thus V˙O2 kinetics may be slowed. This increasing recruitment of fast-twitch fibers likely explains the reduced gain of the primary component of V˙O2 kinetics as well as the overall slowing (i.e., presence of a slow component that prolongs the MRT) observed at higher exercise intensities (70). Moreover, it may also help explain how different interventions such as hyperoxia (62) and priming exercise (11,12) (as discussed above in Evidence for an intracellular control of V˙O2 kinetics) may modulate V˙O2 kinetics if they either reduce the proportional recruitment of fast-twitch fibers or, alternatively, improve O2 delivery to those fibers at the onset of exercise.

Accepting that lower PmvO2 is indicative of a reduced intracellular PO2 (65), one fascinating consideration that arises from the differential PmvO2 profiles observed between fiber types is that one locus of control for the energetic response of the myocyte is moved further upstream from the myocytes themselves-at least for fast-twitch fibers. Historically, the metabolic potentials of the different fiber types have not only helped categorize them into oxidative and glycolytic subpopulations, but their metabolic potentials also have been invoked to explain their reliance on oxidative versus substrate-level phosphorylation when activated. Given the recent findings presented herein, the possibility must be acknowledged that arteriolar vasodilation as it affects the QO2-to-V˙O2 balance dictates the PmvO2 and, hence, directly influences the intracellular PO2 (65). In turn, that intracellular PO2 dictates the energy charge and modulates the degree of metabolic perturbation (Δ[ADP]free, Δ[PCr], Δ[Pi]) necessary to generate the required ATP (3,38,91,92).

In addition to the implications for understanding exercise and muscle energetics in health, these findings may explain some of the perturbations characteristic of major diseases. For example, type II diabetes slows V˙O2 kinetics following the onset of exercise and compromises exercise tolerance (67,71). There are at least two plausible mechanistic bases for these effects based on the differential regulation of PmvO2 demonstrated herein: 1) the diabetes-induced lowering of slow-twitch fiber oxidative capacity will force recruitment of a greater population of fast-twitch fibers with their lower PmvO2 and the energetic consequences of such (i.e., ↑Δ[ADP]free, ↑Δ[PCr], ↑Δ[Pi]), and 2) impaired arteriolar vasodilation consequent to decreased bioavailability of nitric oxide and elevated plasma endothelin-1 concentrations, for example, will reduce PmvO2 in slow-twitch muscles, thereby modulating their cellular energetics toward those of their fast-twitch counterparts. Both these mechanisms would be expected to slow V˙O2 kinetics, increase the O2 deficit incurred, and exacerbate substrate-level phosphorylation.

From the above, it may not be sufficient to characterize a given bout of "exercise" as simply lying to the left or right of the tipping point (i.e., QO2 limited, or not; Fig. 1). Rather, unless the exercise is of a type and intensity that solely recruits slow-twitch fibers (e.g., in the moderate-intensity domain), there may be discrete populations of fibers operating concomitantly on the right (slow-twitch, non-O2-delivery limited) and left (fast-twitch, QO2 limited) of the tipping point. Exercise conditions (e.g., inspired hypoxia/hyperoxia) or diseases (e.g., diabetes, heart failure) that either change the QO2 (and PmvO2) of a muscle fiber(s) or the fiber type recruitment profile will modify the balance of that exercise with respect to the tipping point.


Although O2 has the potential to "regulate" metabolic control, it is clear from the above that agreement is still lacking regarding the role of, and conditions under which, muscle O2 delivery "limits" V˙O2 kinetics. Hughson (40) proposes that V˙O2 kinetics is always dependent on O2 delivery-that is, that PO2 interacts with the cellular redox and phosphorylation potentials to ensure an appropriate ATP flux rate-but that the resolution of available experimental techniques is not always sufficient to enable alterations in V˙O2 response profiles to be detected. The work of Hughson (40) (Fig. 2) therefore rejects the notion, implicit in Figure 1, that there are both O2-delivery-dependent and O2-delivery-independent zones. The counterpoint position championed by A. M. Jones in the Evidence for an intracellular control of V˙O2 section espouses the view that muscle O2 delivery cannot be considered to be definitively "limiting" unless V˙O2 kinetics can be measurably speeded following removal of the putative restriction. Specifically, there is compelling evidence that O2 delivery does not limit the speed of V˙O2 kinetics during rhythmic leg exercise such as cycling or running in young healthy individuals: experimental interventions designed to either reduce or enhance muscle O2 delivery have no discernible effect on the time constant of the phase II V˙O2 response, even during high-intensity exercise (70-100% V˙O2max (7,11,13,86,89)), suggesting that exercise of this type lies some distance to the right of the tipping point portrayed in Figure 1. With respect to this issue, heavy priming exercise is an intervention that may facilitate both intracellular metabolic processes and enhanced muscle O2 delivery. And, whereas the overwhelming majority of evidence demonstrates that priming exercise does not speed the primary component of the V˙O2 kinetics (12), two recent studies have shown a positive result, particularly in subjects with lower fitness levels (33,34). This priming exercise-induced speeding of the V˙O2 kinetics occurred concomitantly with an elevated pyruvate dehydrogenase activation (33), suggesting a role for reduced "metabolic inertia" in this response. In other populations (the elderly, diseased, or unfit), and in exercise modes in which muscle perfusion pressure is compromised, V˙O2 kinetics might indeed be limited, in part, by an O2-supply limitation, because, in these conditions, interventions that would be expected to alter muscle O2 delivery very often also alter the V˙O2 kinetics (70).

Although these viewpoints might seem polarized, they are not necessarily irreconcilable. T. J. Barstow's novel application of near-infrared spectroscopy technology yields the surprising conclusion that whereas the dynamics of muscle or limb blood flow may be much faster than V˙O2 kinetics (15,16,57,84), those of capillary blood flow may not be. The logical interpretation of this finding is that even at moderate exercise intensities, muscles may operate precariously close to the tipping point. Hence, the pernicious effects of disease on V˙O2 kinetics may arise from only a minor impediment to the normal hyperemic control processes, or, possibly, from a pathologically induced shift in fiber type recruitment toward more fast-twitch fibers. In this latter regard, P. McDonough's demonstration of different O2 supply-to-V˙O2 relationships in muscles predominantly comprising fast-twitch or slow-twitch fibers leads to the conclusion that the operating position on Figure 1 is governed by the relative recruitment of these different fiber types. Moreover, the finding of extremely low O2 levels in microvessels supplying fast-twitch fibers suggests that their metabolic behavior may be dictated by upstream hyperemic control processes to a heretofore unappreciated degree.

Both A. M. Jones and P. McDonough suggest that the slower V˙O2 kinetics that are sometimes observed at higher exercise intensities might relate to the recruitment of higher-order (fast-twitch) muscle fibers. These fibers have O2 supply-to-V˙O2 relationships which make it more likely that at least some fibers will be operating to the left of the tipping point, where increased or decreased muscle O2 delivery would serve to speed or slow the overall V˙O2 kinetics, respectively. Moreover, the often-profound slowing of V˙O2 kinetics observed in various disease conditions, and the effects observed with interventions such as endurance training, priming exercise, and hyperoxia, might be related, at least in part, to effects on muscle fiber type and/or fiber recruitment profiles and the associated alteration to the O2 supply-to-V˙O2 relationships. Interestingly, this hypothesis is in keeping with Figure 1, and yet it does not conflict with the predictions of the model proposed by Hughson (40) (Fig. 2) or with the data presented by T. J. Barstow. Taking endurance training as an example, the increased mitochondrial density in all fiber types might be expected to delay and/or reduce the recruitment of type II muscle fibers (A.M.J. and P.M.), to maintain PO2 at a higher level and thus reduce the fall in cell energy state (40) (Fig. 2), and, perhaps, to enhance the distribution of blood flow to cells with higher metabolic activity, such that capillary blood flow kinetics are less likely to limit muscle V˙O2 kinetics (T.J.B.). Therefore, although the role of O2 as a limitation to V˙O2 kinetics remains controversial and certainly complex, the application of new technologies and experimental techniques is already beginning to reconcile apparently disparate positions and is helping to unravel the secrets of metabolic control in health and disease.

The authors gratefully acknowledge the role of the National Institutes of Health (HL-50306), the American Heart Association, Heartland Affiliate, and the American College of Sports Medicine, without whose generous support this work would not have been possible. Special thanks are accorded Professors Scott K. Powers, George A. Brooks, and Timothy I. Musch for their encouraging and facilitating the presentation of this symposium at the ACSM Integrative Physiology of Exercise meeting held in Indianapolis, IN, in September 2006. Professor Richard L. Hughson presented in the Indianapolis symposium and was an integral part of the preliminary drafts of this paper. Unfortunately, he chose to withdraw from authorship before publication.


1. Allen BW, Piantadosi CA. How do red blood cells cause hypoxic vasodilation? The SNO-hemoglobin paradigm. Am J Physiol Heart Circ Physiol. 2006;291:H1507-12.
2. Armstrong RB, Laughlin MH. Blood flows within and among rat muscles as a function of time during high speed treadmill exercise. J Physiol. 1983;344:189-208.
3. Arthur PG, Hogan MC, Bebout DE, Wagner PD, Hochachka PW. Modeling the effects of hypoxia on ATP turnover in exercising muscle. J Appl Physiol. 1992;73:737-42.
4. Bangsbo J, Krustrup P, Gonzalez-Alonso J, Boushel R, Saltin B. Muscle oxygen kinetics at onset of intense dynamic exercise in humans. Am J Physiol Regul Integr Comp Physiol. 2000;279:R899-906.
5. Behnke BJ, Kindig CA, Musch TI, Koga S, Poole DC. Dynamics of muscle microvascular oxygen pressure across the rest-exercise transition. Respir Physiol. 2001;126:53-63.
6. Behnke BJ, McDonough P, Padilla DJ, Musch TI, Poole DC. Oxygen exchange profile in muscles of contrasting fibre types. J Physiol (Lond.) 2003;549:597-605.
7. Berger NJ, Campbell IT, Wilkerson DP, Jones AM. Influence of acute plasma volume expansion on V˙O2 kinetics, V˙O2peak, and performance during high-intensity cycle exercise. J Appl Physiol. 2006;101:707-14.
8. Brittain CJ, Rossiter HB, Kowalchuk JM, Whipp BJ. Effect of prior metabolic rate on the kinetics of oxygen uptake during moderate-intensity exercise. Eur J Appl Physiol. 2001;86:125-34.
9. Brown GC. Nitric oxide and mitochondria. Front Biosci. 2007;12:1024-33.
10. Buehler PW, Alayash AI. Oxygen sensing in the circulation: "cross talk" between red blood cells and the vasculature. Antioxid Redox Signal. 2004;6:1000-10.
11. Burnley M, Doust JH, Ball D, Jones AM. Effects of prior heavy exercise on V˙O2 kinetics during heavy exercise are related to changes in muscle activity. J Appl Physiol. 2002;93:167-74.
12. Burnley M, Koppo K, Jones AM. 'Priming exercise' and V˙O2 kinetics. In: Jones AM, Poole DC, editors. Oxygen Uptake Kinetics in Sport, Exercise and Medicine. London (UK):Routledge; 2005. p. 230-60.
13. Burnley M, Roberts CL, Thatcher R, Doust JH, Jones AM. Influence of blood donation on O2 uptake on-kinetics, peak O2 uptake and time to exhaustion during severe-intensity cycle exercise in humans. Exp Physiol. 2006;91:499-509.
14. Connett RJ, Honig CR, Gayeski TEJ, Brooks GA. Defining hypoxia: a systems view of V˙O2, glycolysis, energetics, and intracellular PO2. J Appl Physiol. 1990;68:833-42.
15. DeLorey DS, Kowalchuk JM, Paterson DH. Relationship between pulmonary O2 uptake kinetics and muscle deoxygenation during moderate-intensity exercise. J Appl Physiol. 2003;95:113-20.
16. Delp MD. Control of skeletal muscle perfusion at the onset of dynamic exercise. Med Sci Sports Exerc. 1999;31(7):1011-8.
17. Delp MD, Colleran PN, Wilkerson MK, McCurdy MR, Muller-Delp J. Structural and functional remodeling of skeletal muscle microvasculature is induced by simulated microgravity. Am J Physiol. 2000;278:H1866-73.
18. Engelen M, Porszasz J, Riley M, Wasserman K, Maehara K, Barstow TJ. Effects of hypoxia on O2 uptake and heart rate kinetics during heavy exercise. J Appl Physiol. 1996;81:2500-8.
19. Erecñka M, Wilson DF. Regulation of cellular energy metabolism. J Membrane Biol. 1982;70:1-14.
20. Ferreira LF, Townsend DK, Lutjemeier BJ, Barstow TJ. Muscle capillary blood flow kinetics estimated from pulmonary O2 uptake and near-infrared spectroscopy. J Appl Physiol. 2005;98:1820-8.
21. Fukuba Y, Ohe Y, Miura A, et al. Dissociation between the time courses of femoral artery blood flow and pulmonary V˙O2 during repeated bouts of heavy knee extension exercise in humans. Exp Physiol. 2004;89:243-53.
22. Grassi B. Skeletal muscle V˙O2 on-kinetics: set by O2 delivery or by O2 utilization? New insights into an old issue. Med Sci Sports Exerc. 2000;32(1):108-16.
23. Grassi B. Limitation of skeletal muscle V˙O2 kinetics by inertia of cellular respiration. In: Jones AM, Poole DC, editors. Oxygen Uptake Kinetics in Sport, Exercise and Medicine. London (UK): Routledge; 2005. p. 212-29.
24. Grassi B, Gladden LB, Samaja M, Stary CM, Hogan MC. Faster adjustment of O2 delivery does not affect V˙O2 on-kinetics in isolated in situ canine muscle. J Appl Physiol. 1998;85:1394-403.
25. Grassi B, Gladden LB, Stary CM, Wagner PD, Hogan MC. Peripheral O2 diffusion does not affect V˙O2 on-kinetics in isolated in situ canine muscle. J Appl Physiol. 1998;85:1404-12.
26. Grassi B, Hogan MC, Greenhaff PL, et al. Oxygen uptake on-kinetics in dog gastrocnemius in situ following activation of pyruvate dehydrogenase by dichloroacetate. J Physiol. 2002;538:195-207.
27. Grassi B, Hogan MC, Kelley KM, et al. Role of convective O2 delivery in determining V˙O2 on-kinetics in canine muscle contracting at peak V˙O2. J Appl Physiol. 2000;89:1293-301.
28. Grassi B, Hogan MC, Kelley KM, Howlett RA, Gladden LB. Effects of nitric oxide synthase inhibition by L-NAME on oxygen uptake kinetics in isolated canine muscle in situ. J Physiol. 2005;568:1021-33.
29. Grassi B, Pogliaghi S, Rampichini S, et al. Muscle oxygenation and pulmonary gas exchange kinetics during cycling exercise on-transitions in humans. J Appl Physiol. 2003;95:149-58.
30. Grassi B, Poole DC, Richardson RS, Knight DR, Erickson BK, Wagner D. Muscle O2 uptake kinetics in humans: implications for metabolic control. J Appl Physiol. 1996;80:988-98.
31. Green HJ, Jones S, Ball-Burnett M, Farrance B, Ranney D. Adaptations in muscle metabolism to prolonged voluntary exercise and training. J Appl Physiol. 1995;78:138-45.
32. Greenhaff PL, Campbell-O'Sullivan SP, Constantin-Teodosiu D, Poucher SM, Roberts PA, Timmons JA. An acetyl group deficit limits mitochondrial ATP production at the onset of exercise. Biochem Soc Trans. 2002;30:275-80.
33. Gurd BJ, Peters SJ, Heigenhauser GJ, et al. Prior heavy exercise elevates pyruvate dehydrogenase activity and speeds O2 uptake kinetics during subsequent moderate-intensity exercise in healthy young adults. J Physiol. 2006;577:985-96.
34. Gurd BJ, Sheuermann BW, Paterson DH, Kowalchuk JM. Prior heavy-intensity exercise speeds V˙O2 kinetics during moderate-intensity exercise in young adults. J Appl Physiol. 2005;98:1371-8.
35. Harper AJ, Ferreira LF, Lutjemeier BJ, Townsend DK, Barstow TJ. Human femoral artery and estimated muscle capillary blood flow kinetics following the onset of exercise. Exp Physiol. 2006;91:661-71.
36. Haseler LJ, Kindig CA, Richardson RS, Hogan MC. The role of oxygen in determining phosphocreatine onset kinetics in exercising humans. J Physiol. 2004;558:985-92.
37. Haseler LJ, Richardson RS, Videen JS, Hogan MC. Phosphocreatine hydrolysis during submaximal exercise: the effect of FIO2. J Appl Physiol. 1998;85:1457-63.
38. Hogan MC, Arthur PG, Bebout DE, Hochachka PW, Wagner PD. Role of O2 in regulating tissue respiration in dog muscle working in situ. J Appl Physiol. 1992;73:728-36.
39. Hughson RL. Alterations in the oxygen deficit-oxygen debt relationships with beta adrenergic receptor blockade in man. J Physiol. 1984;349:375-87.
40. Hughson RL. Regulation of V˙O2 on-kinetics by O2 delivery. In: Jones AM, Poole DC, editors. Oxygen Uptake Kinetics in Sport, Exercise and Medicine. London (UK): Routledge; 2005. p. 185-211.
41. Hughson RL, Cochrane JE, Butler GC. Faster O2 uptake kinetics at onset of supine exercise with than without lower body negative pressure. J Appl Physiol. 1993;75:1962-7.
42. Hughson RL, Morrissey MA. Delayed kinetics of respiratory gas exchange in the transition from prior exercise. J Appl Physiol. 1982;52:921-9.
43. Hughson RL, Morrisey MA. Delayed kinetics of V˙O2 in the transition from prior exercise. Evidence for O2 transport limitation of V˙O2 kinetics. A review. Int J Sports Med. 1983;4:31-9.
44. Hughson RL, Shoemaker JK, Tschakovsky ME, Kowalchuk JM. Dependence of muscle V˙O2 on blood flow dynamics at onset of forearm exercise. J Appl Physiol. 1996;81:1619-26.
45. Hughson RL, Tschakovsky ME, Houston ME. Regulation of oxygen consumption at the onset of exercise. Exerc Sport Sci Rev. 2001;29(3):129-33.
46. Hughson RL, Xing HC, Borkhoff C, Butler GC. Kinetics of ventilation and gas exchange during supine and upright cycle exercise. Eur J Appl Physiol Occup Physiol. 1991;63:300-7.
47. Jones AM, Berger NJ, Wilkerson DP, Roberts CL. Effects of "priming" exercise on pulmonary O2 uptake and muscle deoxygenation kinetics during heavy-intensity cycle exercise in the supine and upright positions. J Appl Physiol. 2006;101:1432-41.
48. Jones AM, Koppo K, Wilkerson DP, Wilmshurst S, Campbell IT. Dichloroacetate does not speed phase-II pulmonary V˙O2 kinetics following the onset of heavy intensity cycle exercise. Pflugers Arch. 2004;447:867-74.
49. Jones A, Pringle JS, Carter H. Influence of muscle fibre type and motor unit recruitment on V˙O2 kinetics. In: Jones AM, Poole DC, editors. Oxygen Uptake Kinetics in Sport, Exercise and Medicine. London (UK): Routledge; 2005. p. 261-93.
50. Jones AM, Wilkerson DP, Koppo K, Wilmshurst S, Campbell IT. Inhibition of nitric oxide synthase by L-NAME speeds phase II pulmonary V˙O2 kinetics in the transition to moderate-intensity exercise in man. J Physiol. 2003;552:265-72.
51. Kemp G. Kinetics of muscle oxygen use, oxygen content, and blood flow during exercise. J Appl Physiol. 2005;99:2463-9.
52. Kindig CA, Howlett RA, Stary CM, Walsh B, Hogan MC. Effects of acute creatine kinase inhibition on metabolism and tension development in isolated single myocytes. J Appl Physiol. 2005;98:541-9.
53. Kindig CA, Kelley KM, Howlett RA, Stary CM, Hogan MC. Assessment of O2 uptake dynamics in isolated single skeletal myocytes. J Appl Physiol. 2003;94:353-7.
54. Kindig CA, McDonough P, Erickson HH, Poole DC. Nitric oxide synthase inhibition speeds oxygen uptake kinetics in horses during moderate domain running. Respir Physiol Neurobiol. 2002;132:169-78.
55. Kindig CA, Richardson TE, Poole DC. Skeletal muscle capillary hemodynamics from rest to contractions: implications for oxygen transfer. J Appl Physiol. 2002;92:2513-20.
56. Knight DR, Poole DC, Schaffartzik W, et al. Relationship between body and leg V˙O2 during maximal cycle ergometry. J Appl Physiol. 1992;73:1114-21.
57. Koga S, Poole DC, Shiojiri T, et al. Comparison of oxygen uptake kinetics during knee extension and cycle exercise. Am J Physiol Regul Integr Comp Physiol. 2005;288:R212-20.
58. Koppo K, Bouckaert J. Prior arm exercise speeds the V˙O2 kinetics during arm exercise above the heart level. Med Sci Sports Exerc. 2005;37(4):613-9.
59. Krustrup P, Soderlund K, Mohr M, Bangsbo J. Slow-twitch fiber glycogen depletion elevates moderate-exercise fast-twitch fiber activity and O2 uptake. Med Sci Sports Exerc. 2004;36(6):973-82.
60. Linnarsson D. Dynamics of pulmonary gas exchange and heart rate changes at start and end of exercise. Acta Physiol Scand Suppl. 1974;415:1-68.
61. Linnarsson D, Karlsson J, Fagraeus L, Saltin B. Muscle metabolites and oxygen deficit with exercise in hypoxia and hyperoxia. J Appl Physiol. 1974;36:399-402.
62. MacDonald M, Pedersen PK, Hughson RL. Acceleration of V˙O2 kinetics in heavy submaximal exercise by hyperoxia and prior high-intensity exercise. J Appl Physiol. 1997;83:1318-25.
63. MacDonald MJ, Shoemaker JK, Tschakovsky ME, Hughson RL. Alveolar oxygen uptake and femoral artery blood flow dynamics in upright and supine leg exercise in humans. J Appl Physiol. 1998;85:1622-8.
64. MacPhee SL, Shoemaker JK, Paterson DH, Kowalchuk JM. Kinetics of O2 uptake, leg blood flow, and muscle deoxygenation are slowed in the upper compared with lower region of the moderate-intensity exercise domain. J Appl Physiol. 2005;99:1822-34.
65. McDonough P, Behnke BJ, Padilla DJ, Musch TI, Poole DC. Control of microvascular oxygen pressures in muscles comprised of different fibre types. J Physiol. 2005;563:903-13.
66. Murphy PC, Cuervo LA, Hughson RL. Comparison of ramp and step exercise protocols during hypoxic exercise in man. Cardiovasc Res. 1989;23:825-32.
67. Padilla DJ, McDonough P, Behnke BJ, et al. Effects of type II diabetes on muscle microvascular oxygen pressures. Respir Physiol Neurobiol. 2007;156:187-95.
68. Piiper J, Pendergast DR, Marconi C, Meyer M, Heisler N, Ceretelli P. Blood flow distribution in dog gastrocnemius muscle at rest and during stimulation. J Appl Physiol. 1985;58:2068-74.
69. Poole DC. Influence of exercise training on skeletal muscle oxygen delivery and utilization. In: Crystal RG, West JB, Weibel ER, Barnes PJ, editors. The Lung: Scientific Foundations. New York (NY): Raven Press; 1997. p. 1957-67.
70. Poole DC, Jones AM. Towards an understanding of the mechanistic bases of V˙O2 kinetics. In: Jones AM, Poole DC, editors. Oxygen Uptake Kinetics in Sport, Exercise and Medicine. London (UK): Routledge; 2005. p. 294-328.
71. Poole DC, Kindig CA, Behnke BJ. V˙O2 kinetics in different disease states. In: Jones AM, Poole DC, editors. Oxygen Uptake Kinetics in Sport, Exercise and Medicine. London: Routledge; 2005. p. 353-72.
72. Poole DC, Sexton WL, Behnke BJ, Ferguson CS, Hageman KS, Musch TI. Respiratory muscle blood flows during physiological and chemical hyperpnea in the rat. J Appl Physiol. 2000;88:186-94.
73. Pringle JS, Doust JH, Carter H, et al. Oxygen uptake kinetics during moderate, heavy and severe intensity "submaximal" exercise in humans: the influence of muscle fibre type and capillarisation. Eur J Appl Physiol. 2003;89:289-300.
74. Radegran G, Saltin B. Muscle blood flow at onset of dynamic exercise in humans. Am J Physiol. 1998;274:H314-22.
75. Richardson RS, Duteil S, Wary C, Wray DW, Hoff J, Carlier PG. Human skeletal muscle intracellular oxygenation: the impact of ambient oxygen availability. J Physiol. 2006;571:415-24.
76. Richardson RS, Noyszewski EA, Kendrick KF, Leigh JS, Wagner PD. Myoglobin O2 desaturation during exercise-evidence of limited O2 transport. J Clin Invest. 1995;96:1916-26.
77. Rossiter HB, Ward A, Doyle VL, Howe FA, Griffiths JR, Whipp BJ. Inferences from pulmonary O2 uptake with respect to intramuscular [phosphocreatine] kinetics during moderate exercise in humans. J Physiol. 1999;518:921-32.
78. Rossiter HB, Ward SA, Kowalchuk JM, Howe FA, Griffiths JR, Whipp BJ. Dynamic asymmetry of phosphocreatine concentration and O2 uptake between the on- and off-transients of moderate- and high-intensity exercise in humans. J Physiol. 2002;541:991-1002.
79. Rossiter HB, Ward SA, Howe FA, et al. Effects of dichloroacetate on V˙O2 and intramuscular 31P metabolite kinetics during high-intensity exercise in humans. J Appl Physiol. 2003;95:1105-15.
80. Scheuermann BW, Bell C, Paterson DH, Barstow TJ, Kowalchuk JM. Oxygen uptake kinetics for moderate exercise are speeded in older humans by prior heavy exercise. J Appl Physiol. 2002;92:609-16.
81. Shoemaker JK, Hughson RL. Adaptation of blood flow during the rest to work transition in humans. Med Sci Sports Exerc. 1999;31(7):1019-26.
82. Tordi N, Perrey S, Harvey A, Hughson RL. Oxygen uptake kinetics during two bouts of heavy cycling separated by fatiguing sprint exercise in humans. J Appl Physiol. 2003;94:533-41.
83. Tschakovsky ME, Hughson RL. Interaction of factors determining oxygen uptake at the onset of exercise. J Appl Physiol. 1999;86:1101-13.
84. Tschakovsky ME, Sheriff DD. Immediate exercise hyperemia: contributions of the muscle pump vs. rapid vasodilation. J Appl Physiol. 2004;97:739-47.
85. Whipp BJ, Mahler M. Dynamics of pulmonary gas exchange during exercise. In: West JB, editor. Pulmonary Gas Exchange, Vol. II. New York (NY): Academic Press; 1980. p. 33-96.
86. Wilkerson DP, Berger NJ, Jones AM. Influence of hyperoxia on pulmonary O2 uptake kinetics following the onset of exercise in humans. Respir Physiol Neurobiol. 2006;153:92-106.
87. Wilkerson DP, Campbell IT, Jones AM. Influence of nitric oxide synthase inhibition on pulmonary O2 uptake kinetics during supra-maximal exercise in humans. J Physiol. 2004;561:623-35.
88. Wilkerson DP, Jones AM. Influence of initial metabolic rate on pulmonary O2 uptake on-kinetics during severe intensity exercise. Respir Physiol Neurobiol. 2006;152:204-19.
89. Wilkerson DP, Rittweger J, Berger NJ, Naish PF, Jones AM. Influence of recombinant human erythropoietin treatment on pulmonary O2 uptake kinetics during exercise in humans. J Physiol. 2005;568:639-52.
90. Williamson JW, Raven PB, Whipp BJ. Unaltered oxygen uptake kinetics at exercise onset with lower-body positive pressure in humans. Exp Physiol. 1996;81:695-705.
91. Wilson DF, Ericinska M, Drown C, Silver IA. Effect of oxygen tension on cellular energetics. Am J Physiol. 1977;233:C135-40.
92. Wilson DF, Rumsey WL. Factors modulating the oxygen dependence of mitochondrial oxidative phosphorylation. Adv Exp Biol Med. 1988;222:121-31.
93. Woodman CR, Shrage WG, Rush JWE, Ray CA, Price EM. Hindlimb unweighting decreases endothelium-dependent dilation and eNOS expression in soleus not gastrocnemius. J Appl Physiol. 2001;91:1091-8.
94. Wunsch SA, Muller-Delp J, Delp MD. Time course of vasodilatory responses in skeletal muscle arterioles at onset of exercise. Am J Physiol. 2000;279:H1715-23.
95. Yoshida T, Kamiya J, Hishimoto K. Are oxygen uptake kinetics at the onset of exercise speeded up by local metabolic status in active muscle? Eur J Appl Physiol Occup Physiol. 1995;70:482-6.


©2008The American College of Sports Medicine