4.2.1. Oxygen Extraction from the Blood

The increase in oxygen extraction from the blood lowers venous oxygen content such that the arterio-venous oxygen content difference ((A-V)O2) increases (Figure 10) [37, 144, 246–250]. The increased (A-V)O2 results not only from increased O2 extraction in the active muscles (heart, respiratory, abdominal and skeletal muscles) but also in peripheral organs whose blood supplies were reduced by baroreflex-mediated vasoconstriction to redistribute the cardiac output to ensure optimal oxygen utilization (Figure 10) [205, 207]. Thus, the changes in vascular resistance are opposite in active muscles vs peripheral organs and thereby limit the reduction in total peripheral vascular resistance that accompanies intense exercise.

Under resting conditions, oxygen extraction ranges between 20% and 40%. During heavy exercise, approximately 70–80% of the oxygen delivered to the active muscles may be extracted. This demonstrates that there is a reserve of oxygen in the blood that can be utilized immediately to meet the needs of the contracting muscles at the onset of exercise. Increased extraction of oxygen from the blood is driven by decreases in perivascular PO₂, which in turn are driven by reductions in cell PO₂ [205]. Other factors which contribute to enhanced oxygen extraction with increased muscle metabolism relates to changes in the oxygen dissociation curve that are brought about by the increased blood hydrogen ion and CO₂ levels that are released into the tissue spaces as a result of increased metabolism and subsequently diffuse into the circulation (Figure 10) [58]. This facilitates unloading of oxygen from hemoglobin in the contracting skeletal muscles.

4.2.2. Arterial and Arteriolar Vasodilation Increases Blood Flow Delivery and Microvascular Surface Area for Exchange

While increased oxygen extraction provides one important avenue for meeting tissue needs during exercise, dilating arterial and arteriolar segments of the vascular network is far more important (Figure 10). Vasodilation of the arterial tree results in increased blood flow, which carries more oxygen to the tissues per unit time. In addition, the enhanced blood flow increases microvessel hematocrit, which also supports increased oxygen delivery to the active muscles (Figure 10). Capillaries represent the terminal points for oxygen delivery to the tissues by the circulation, as well as removal of metabolites from the active muscles. Vasodilation of small arterioles also enhances functional capillary density (increased number of perfused capillaries) which shortens the diffusion distance for oxygen and other substrates (Figure 10) [116, 205, 431]. Perfused capillary density is proportional to oxygen consumption in red muscles, but correlates with lactate production in white muscle. In addition, small arteries also vasodilate in an axially coordinated fashion with the change in caliber in arterioles, the net effect of which is to produce a 2- to 5-fold larger increment in blood flow, and thus oxygen delivery to the active muscles.

4.6.1. Initial vs Later Vasoregulatory Mechanisms in Exercise Hyperemia

Blood flow increases within one second of the onset of muscular activity, a vasodilator response referred to as the fast vasodilator response (Figure 10) [72, 101, 138, 459, 578, 692, 709]. This is followed by a larger vasodilation that is sustained and dependent on the magnitude and duration of exercise. This second vasodilatory phase is thought to result from the production of vasodilator metabolites by the contracting muscle cells that can be modulated by other local vasoregulatory mechanisms (Figure 10) [375, 584, 622, 713]. Since it takes some time for active muscles to produce metabolites, which then have to diffuse to nearby arterioles, followed by binding to their receptors and activation of the second messenger signaling mechanisms that cause relaxation of vascular smooth muscle, the fast vasodilator mechanism must result from other processes [717]. Withdrawal of sympathetic tone does not explain this initial vasodilator phase because sympathectomy or treatment with ganglionic blockers does not alter the time to onset of this fast response [72, 138]. However, there is evidence that spillover of acetylcholine from motor endplates activates muscarinic receptors to initiate conducted vasodilator responses, at least in some muscle preparations [578, 692, 709]. In others, blockade of these receptors fails to alter the initial hyperemic response to exercise [24, 28, 70, 626]. Others have proposed that the potassium ions released during muscle action potentials, myogenic dilation, and the muscle pump may also be responsible for this early increase in muscle blood flow during exercise.

4.6.2. Mechanical Effects of Exercise on Blood Flow

Arterial inflow to active skeletal muscle decreases during contractions and increases when the muscle relaxes. In contrast, the venous outflow increases during rhythmic contractions but decreases during muscle relaxation. These mechanical effects of exercise are due to increased extravascular pressure during rhythmic contractions, which expels blood through the venous system and increases the arterio-venous pressure gradient for flow (Figures 6 and 10) [34, 92, 119, 180, 367, 525, 613, 628]. This muscle pump mechanism can contribute up to 60% of the hyperemic response, at least under some conditions [93, 311, 312, 470], and provides an attractive explanation for the observation that going from supine to upright position produces a twofold increase in exercise hyperemia (which is difficult to ascribe to other known mechanisms that contribute to the increased flow) [378]. The muscle pump also provides another explanation for the fast vasodilator mechanism because it may elicit a myogenic vasodilator response in skeletal muscle arterioles, owing to the decrease in arteriolar wall tension that occurs when extravascular pressure rises during the contractile phase. Some work suggest that vascular deformation may play a more substantial role in eliciting vasodilation in response to rhythmic contractions than that evoked by the myogenic mechanism, an effect related to circumferential compression and/or increased axial lengthening of arterioles [493].

4.6.3. Metabolic Mechanisms in Exercise Hyperemia

Although it is clear that the increased interstitial concentration of vasodilator metabolites derived from active muscle cells plays a major in producing exercise hyperemia, it does not appear that one single metabolite accounts for metabolic vasorelaxation of arteriolar vascular smooth muscle (Figure 10) [52, 263–265, 375, 488]. Depending on the muscle fiber types involved in contraction, different metabolites may predominate, owing to differences in metabolic activity [54, 322, 323, 373, 595]. In addition, the stimulus rate, intensity, and duration, as well as the interval between stimuli all influence the relative metabolite profile produced during muscular activity [536, 595]. Alterations in the type of contraction (isometric vs isotonic) and the mass of muscle involved in the contractile activity may also produce differential release of metabolites [46, 242, 243]. As muscle fatigues during ongoing exercise, vasodilation is attenuated, an effect due to diminished conducted responses [279].

While a large number of metabolites have been suggested to mediate exercise hyperemia, the most compelling evidence suggests that adenosine, potassium ions, and osmolarity changes may be quite important. In addition, decreases in arteriolar PO₂ (hypoxemia) that may occur in exercise owing to enhanced diffusion of the gas across the walls of these vessels may also contribute to active hyperemia. The notion that release of ATP from contracting myocytes and by erythrocytes in response to hypoxia and mechanical deformation of red cells as they pass through microvessels of contracting skeletal muscle is also gaining support.

4.6.4. Myogenic Mechanism and Exercise Hyperemia

During active muscle contraction, interstitial pressure increases, thereby reducing transmural pressure and wall tension. The reduction in transmural pressure decreases vascular wall tension, which is sensed by an unknown mechanism to elicit myogenic dilation (Figure 10). The dilatory response allows increased blood flow. Vascular wall tension is thought to be the sensed variable that initiates the myogenic response, as opposed to diameter or pressure changes, because myogenic constriction or dilation does not provide the appropriate feedback to maintain the vasomotor response. While myogenic contributions to exercise-induced increases in skeletal muscle blood flow is appealing in concept, pressure-induced responses may play a more important role in the autoregulation of blood flow and capillary pressure than in exercise hyperemia [412, 451, 452, 456, 458].

4.6.5. Skeletal Muscle Myokines

More recent work has established that in addition to the aforementioned vasodilator metabolites, exercising skeletal muscles also release cytokines (myokines) such as interleukin-6 (IL-6) and IL-8 [515–517]. Indeed, IL-6 concentrations increase up to 100-fold during exercise, which may induce vasodilation through the effects of this myokine on AMP-activated protein kinase (AMPK). Although this enzyme is best known as a master regulator of metabolism, AMPK is expressed by both vascular smooth muscle and endothelial cells, where its activity may produce vasodilation by virtue of its action to phosphorylate eNOS, thereby activating the enzyme to produce NO (Figure 10) [191, 584]. In skeletal muscle, neuronal NOS (nNOS) is highly expressed and NO derived from this source appears to be involved in the enhanced production of IL-6 [584]. Moreover, NO may also indirectly induce vasodilation secondary to its action to promote myosin light chain dephosphorylation.

The myokine IL-8 may be involved in angiogenic responses, perhaps serving as a mediator of exercise training-induced angiogenesis [395, 517, 584]. In this regard, pericytes have been shown to be essential for endothelial lumen formation during angiogenesis and IL-8 modifies pericyte function [468, 660]. Clearly, much additional work will be required to more definitively implicate myocyte-derived myokines and/or pericytes in the regulation of vascular responses in skeletal muscle. Skeletal muscle myokines have also been proposed as important mediators of the anti-inflammatory effects of exercise and serve as intermediaries in muscle-to-fat cross-talk, reducing body fat mass [516, 517].

4.6.6. Acetylcholine Spillover from Motor End-Plates and Adrenal Medullary Contributions to Exercise Hyperemia

Nerves and the adrenal medulla may also participate in the production of vasodilation during exercise. Muscle contraction is initiated by the release of acetylcholine from the motor end-plate, which spills over into periarteriolar regions to elicit vasodilation (Figure 10) [584, 709]. However, the conducted vasodilation elicited by acetylcholine exhibits features which distinguish its response from those induced by metabolites such as adenosine. Contraction-induced conducted responses produce remote dilations by K_ATP- and NO-dependent mechanisms that are not blocked by gap junctional uncouplers that inhibit acetylcholine-triggered responses [488, 489]. Moreover, the different potassium channel signatures (K_Ca for acetylcholine vs K_ATP for muscle contraction) involved in the two responses are also notable [245]. In contrast to the bidirectional conduction (upstream and downstream) induced by acetylcholine, muscular contraction induces only unidirectional, ascending transmission of the dilator signals [46]. In addition, there is evidence indicating that acetylcholine spillover may not be of sufficient magnitude to explain the response [584].

While adenosine and NO formed during muscular activity reduce sympathetic vasoconstrictor tone during exercise (functional sympatholysis), which contributes to exercise hyperemia, the simultaneous release of epinephrine from the adrenal medulla may contribute to the decreased vascular resistance in active muscles via activation of β2-adrenergic receptors (Figure 10) [584]. Norepinephrine is also released from the adrenal medulla, but because less of this vasoconstrictor catecholamine is released, the effect of epinephrine-induced vasodilation predominates. Adrenal medullary catecholamine release augments the effects induced by activation of sympathetic nerves supplying the heart to increase heart rate and contractility. Indeed, individuals with transected cardiac sympathetics (e.g., after heart transplant) are able to substantially increase cardiac output during exercise by this adrenal medullary mechanism.

4.6.7. Extracellular Matrix Components and Exercise Hyperemia

The extracellular matrix (ECM) surrounding tissue and vascular cells may also influence vascular caliber through matricryptic sites that are located within these molecules but are not normally exposed [111, 244]. Activation of receptors (integrins) that bind extracellular matrix components can elicit vasoconstriction or vasodilation, depending on which integrin ligand is evaluated. In addition to protein multimerization and proteolysis, matricryptic sites can be exposed on ECM proteins by mechanical forces, such as occurs during muscular contraction. Interestingly, recent work indicates that the ECM protein fibronection may contribute to active hyperemia in skeletal muscle by an NO-dependent mechanism that is independent of integrins, which normally act as ligands to direct other cell signaling events initiated by ECM proteins (Figure 10) [244]. According to this scenario [244], skeletal muscle contraction exerts a tensile force on the ECM surrounding arterioles that results in transient exposure of matricryptic sites on fibronection that allows ligation with its integrin receptor. In turn, this matricryptin/integrin ligation activates neuronal NOS to produce NO and elicit vasodilation, thereby offering a mechanism to mechanically couple skeletal muscle contraction to arteriolar dilation.

4.6.8. Flow- or Shear Stress-Induced Vasodilation

The increases in intraluminal blood flow velocity that accompany exercise cause vasodilation independent of changes in intraluminal pressure or transmural pressure (Figure 10). This occurs in both conduit and resistance arteries and is initiated by shear stress-induced signaling events in the endothelium [170, 189, 217, 330, 348, 349, 467, 635, 695]. While the sensor that activates this mechanotransduction signaling cascade during exercise is uncertain, it is clear that endothelial cells release a transferable factor that relaxes vascular smooth muscle because removal or destruction of the endothelium abolishes flow-dependent vasodilation [266, 348, 399, 567]. Work conducted in isolated arterioles suggests that NO is a likely candidate mediator, although hydrogen peroxide, epoxyeicosatrienoic acids, and prostacyclin may also play a role. Whether these data apply to flow-induced dilation in exercise is difficult to determine because pharmacologic inhibition may also target one or more of the constellation of local metabolic factors that contribute to active hyperemia in skeletal muscle.

4.6.9. Conducted Vasomotor Responses

Hyperpolarization of endothelial and vascular smooth muscle cells induces vasodilation in metabolite-sensitive arterioles which is conducted upstream to induce vasodilation (Figure 10). Endothelial cells provide the primary pathway for the spread of current along the arteriolar and arterial tree, with electrical signals being transmitted to underlying vascular smooth muscle at each point along the way [32, 145, 611, 612]. As the wave of hyperpolarization travels to adjacent arteriolar branches and feed arteries, vascular smooth muscle in these vessels relax, resulting in reduced vascular resistance and increased blood flow. Interestingly, as conducted hyperpolarization travels upstream from arterioles fed by the same parent vessel, there is partial summation of the vasodilator responses. This results in a larger vasodilation in the parent vessel than would occur if only one of the downstream vessels were activated by signals that induce conducted responses [32, 603, 686]. In addition, it is important to note that activation of just a few muscle fibers is sufficient to produce conducted responses that travel upstream through at least three branching orders of arterioles.

Initiating triggers for conducted responses include adenosine, which is released by active muscle cells in proportion to their increased metabolism, and acetylcholine, which may diffuse to adjacent arterioles near motor end plates [32]. When the hyperpolarization induced by adenosine, acetylcholine, and perhaps other molecules is conducted to upstream larger arterioles and feed arteries, it supplements shear stress-induced vasodilation to produce a 2- to 5-fold greater increase in blood flow than would occur if only small, metabolite-sensitive arterioles were to vasodilate [32, 111, 604, 610].

While conducted vasodilation has been clearly shown to occur in response to muscle contraction, there are differences between responses to acetylcholine vs exercise [113, 114]. Contraction-induced conducted responses produce remote dilations by K_ATP- and NO-dependent mechanisms that are not blocked by gap junctional uncouplers that inhibit acetylcholine-triggered responses [245]. The different potassium channel signatures (K_Ca for acetylcholine vs K_ATP for muscle contraction) involved in the two responses are also notable [584]. In contrast to the bidirectional conduction (upstream and downstream) induced by acetylcholine, muscular contraction induces only unidirectional, ascending transmission of the dilator signals [46]. It is unclear how this sort of directionality in transmission induced by muscle activity occurs.

4.6.10. Multiple Mediators Work in Concert to Induce Vasodilation in Active Skeletal Muscle

When taken together, the bulk of available evidence supports the notion that multiple vasodilator mechanisms work in concert to elicit exercise hyperemia (Figure 10). Indeed, coincident blockade of K_ATP channels, adenosine receptors, and nitric oxide synthase are necessary to attenuate the increase in blood flow to contracting skeletal muscle in some animal models [585] while concomitant blockade of cytochrome P450 metabolites, prostacyclin, and nitric oxide synthase is required to limit hyperemic responses to exercise in humans [59]. Importantly, the vasodilator effects of adenosine, NO, potassium ions, oxygen, and hyperosmolarity are pH-dependent [238, 277, 401, 659]. Thus, the acidic conditions produced by exercise (pH falls to 7.1 to 7.2) may potentiate the actions of these vasodilators, producing changes in arteriolar caliber than might otherwise occur.

4.6.11. Differential Control of Arteriolar Function along the Vascular Tree During Exercise

The arcading network of arterioles supplying skeletal muscle exhibits segmental responses to metabolites, pressure, flow, and neural influences, which may be very important for integrated vascular control during exercise [52, 60, 89, 108, 109, 137, 252, 266, 269, 316, 349, 350, 449, 584, 634, 635]. The smallest arterioles appear to be more responsive to certain metabolites, such as adenosine, than larger vessels [52, 584]. On the other hand, myogenic responses are more prominent in intermediate-sized arterioles relative to small arteries [60, 85, 108, 109, 316, 449]. Flow-induced vasodilation is greatest in large arterioles, less so in smaller arterioles, with small arteries demonstrating even smaller responses to equivalent levels of shear stress [89, 252, 266, 269, 349, 350, 634, 635]. These segmental differences in the relative responses to differing stimuli indicate arteriolar function is differentially controlled along the microvascular network, providing for integrated interactions of local control mechanisms to regulate skeletal muscle blood flow when metabolism is increased. These observations have led to the proposal that small arterioles preferentially vasodilate in response to increased metabolite production in exercising skeletal muscle. Metabolic vasodilation of these distal arterioles lowers intraluminal pressure in upstream intermediate-sized arterioles possessing strong myogenic responses, causing them to vasodilate [350]. As a consequence of metabolic and myogenic vasodilation in small- and intermediate-sized arterioles, respectively, blood flow rate increases throughout the microvascular network, thereby inducing shear stress-dependent vasodilation in proximal larger arterioles [350]. As noted above, metabolic vasodilators also act to reduce norepinephrine release at sympathetic nerve terminals (functional sympatholysis) in the vascular walls.

4.9.1. Convective Delivery

Total oxygen delivery to the tissues depends on the blood flow rate supplying skeletal muscle and the concentration of oxygen in the arterial blood, which collectively contribute to convective oxygen delivery (equals blood flow times arterial oxygen concentration). Red blood cell hemoglobin serves as the oxygen carrier in blood, together with a small amount of dissolved oxygen. The oxygen-carrying capacity of hemoglobin refers to the maximal amount of oxygen that can be bound on hemoglobin, whereas the oxygen-carrying capacity of blood depends on hemoglobin concentration in each red blood cell and hematocrit.

The proportion of hemoglobin that is bound to oxygen (percent saturation) is determined by the partial pressure of oxygen (PO2 or oxygen tension) in arterial blood. In normal healthy individuals, hemoglobin is nearly fully saturated with oxygen at an arterial PO2 of approximately 100 mm Hg. Because the ability of respiratory system to exchange oxygen is not normally rate-limiting in exercise, the oxygen-carrying capacity of blood does not limit convective oxygen delivery. Since the arterial oxygen concentration is relatively constant in the short term (except in some mammals which may contract their spleens to expel erythrocytes into the circulation in response stress), the moment-to-moment regulation of oxygen delivery is accomplished primarily by changes in vascular tone. Changes in arterial oxygen concentration that occur over time may also occur in individuals who ascend to altitude or who inject erythropoietin (blood doping), both of which increase hematocrit. This increases the oxygen-carrying capacity of blood, owing to increased hemoglobin concentration. The opposite occurs when hematocrit falls. Women experience alterations in hematocrit which coincide with the menstrual cycle. These complicating factors aside, matching blood flow rate and oxygen-carrying capacity (convective oxygen supply) to oxygen demand by the tissues is essential for maintaining muscle homeostasis, with changes in vessel caliber being the primary mechanism to alter convective delivery on a moment-to-moment basis [641].

4.9.2. Diffusion of Oxygen

After convective delivery to the exchange vessels by the flowing blood, oxygen must diffuse from the capillaries to the mitochondria of skeletal muscle cells. Significant microvascular oxygen flux also occurs across the walls of smaller arterioles [144]. The difference between exchange vessel and skeletal muscle myocyte oxygen tension provides the driving force for the diffusive flux of oxygen from the blood to the mitochondria. The microcirculatory parameters that have the greatest impact on oxygen delivery via diffusion include red cell transit time through exchange vessels, number of red cells per length of capillary, the relationship between red cell transit time and surface area available for gas exchange, microvascular oxygen content, and functional capillary density (number of open capillaries).

Even in the face of maximal increases in blood flow rate to exercising skeletal muscle, red cell transit time does not limit diffusive flux of oxygen. Thus, in addition to the blood-cell PO2 gradient, microvascular surface area available for exchange and the capillary-to-cell diffusion distance are also important determinants of diffusive oxygen flux. Both of these latter factors are influenced by the number of perfused capillaries (perfused capillary density). In quiescent skeletal muscle, approximately 25% of capillaries are open to flow at any given time. This affords the opportunity to increase perfused capillary density during exercise as a means to increase diffusive oxygen flux.

As noted earlier, capillary density varies in accord with the oxidative capacities of the muscle fibers they supply. Slow twitch oxidative fibers can be perfused by approximately 3 times as many capillaries/fiber as compared to fast-twitch glycolytic fibers. The spatial arrangement of capillaries also varies by fiber type, with the vast majority of these microvessels running parallel to the long axis of white muscle fibers [205, 206, 219, 491, 532]. In contrast, capillaries follow a more tortuous course around red muscle fibers, forming loops that encircle these fibers as the vessels course along the long axis of the fibers. This increases the surface area and decreases diffusion distance for oxygen exchange in these oxidative fibers.

It is important to recognize that opening capillaries to flow not only increases the surface area available for exchange but also decreases diffusion distance over which oxygen must flux in moving from the bloodstream to the muscle fibers [47, 48, 207]. From the geometric organization of skeletal muscle capillaries, it has become apparent that surface area increases in direct proportion to the number of open capillaries (N), while diffusion distance is inversely proportional to the square root of N [207]. Thus, the diffusive flux of oxygen in skeletal muscle is proportional to N^3/2, which predicts that oxygen diffusion rate increases 8-fold for a 4-fold increase in perfused capillary density [207].

Recruitment of non-perfused capillaries occurs in association with the hyperemia that accompanies exercise [317–319, 321]. It is thought that terminal arterioles control the number of perfused capillaries, while larger arterioles and resistance arteries regulate vascular resistance and blood flow [47, 105, 145, 199, 257, 439]. This concept arose because terminal arterioles exhibit vasomotion (rhythmic contraction/relaxation cycles), which could account for changes in perfused capillary density and contribute very little to total microvascular resistance [161, 165, 166, 411, 477]. However, capillary perfusion is not controlled on a single capillary basis, but rather occurs by recruitment as groupings of capillaries (microvascular units, networks, or modules) at the initiation of muscular activity [105, 116, 199, 265, 581, 584, 665]. It appears that capillary blood flow, erythrocyte distribution, and red cell concentration are coordinated within and among these microvascular units to equalize oxygen supply [47, 48, 583]. This coordination is facilitated by conducted (propagated) vasodilator responses, which also function to minimize perfusion heterogeneity by reducing the potential for more dilated arterioles to steal flow from less dilated microvessels [606–608]. As a consequence of vasodilation of resistance arteries and arterioles, capillary recruitment, and increased oxygen extraction, oxygen delivery is matched to demand in exercising skeletal muscle.

The increased blood flow to active skeletal muscle might be expected to reduce the time it takes for erythrocytes to traverse exchange vessels, which could impose constraints on oxygen release from hemoglobin during their shortened transit time [256]. However, the concomitant opening of non-perfused microvascular units increases cross-sectional area and effective capillary volume, which prevents dramatic reductions in red cell transit time. Thus, convective shunting at high arterial red cell velocities does not limit tissue oxygenation.

4.9.3. Mitochondrial Oxygen Utilization

The third process that governs oxygen consumption in skeletal muscles and the final stage of oxygen delivery from the blood to the muscle fibers is the use of oxygen by mitochondria [205]. Oxygen removes reducing equivalents from cytochrome a₃, the terminal oxidase in the respiratory chain. Thus, the rate of mitochondrial oxygen consumption is determined by cell PO2 and the concentration of reduced cytochrome a₃ [205].

During exercise, the extraction of oxygen from blood coursing through exchange vessels lowers venous oxygen content compared to quiescent muscle. As a consequence, the arterio-venous oxygen content difference increases with activity. Thus, as venous PO2 decreases with muscular activity, oxygen uptake decreases because the pressure gradient driving oxygen flux from the capillary blood to the mitochondria decreases. This implies a limitation for oxygen diffusion along this path as a determinant of maximal oxygen consumption. The main barriers to the diffusion oxygen include the thin layer of plasma between the red cells and the capillary walls, the endothelium, and the interstitial spaces. This relates to the fact that myoglobin serves to maximize the transcapillary gradient for PO2, but minimizes intracellular PO2 gradients [99, 192, 193, 255]. This hemeprotein exhibits 50% saturation with oxygen at PO2 of ∼5 mm Hg, allowing myoglobin to stabilize tissue PO2 at levels well below capillary blood PO2. Thus, cell PO2 and myoglobin saturation are low but remarkably uniform throughout a given cell and were not related to distance from a capillary [255]. This stabilization of cell PO2 by myoglobin minimizes intracellular gradients in oxygen tension and indicates that the major diffusion limitations occur in moving oxygen from capillary blood to the muscle cell, and not from the myocyte membrane to the mitochondria.

The oxidative capacity of contracting muscle cells does not appear to limit maximal oxygen consumption in normal healthy individuals [259, 479, 480, 491, 561–564, 641, 698, 699]. This concept is supported by the observation that increasing arterial oxygen concentration raises maximal oxygen consumption in contracting muscle fibers [304]. Thus, there is not normally a metabolic limitation to maximal oxygen consumption. On the other hand, perfusion and diffusion shunting produced by perfusion heterogeneities in skeletal muscle may limit maximal oxygen uptake [521, 649]. However, mismatching between oxygen consumption and blood flow is reduced by increasing oxygen uptake with exercise [41, 249].