Dynamic Heterogeneity of Exercising Muscle Blood Flow and O2 Utilization

Shunsaku Koga; Harry B. Rossiter; Ilkka Heinonen; Timothy I. Musch; David C. Poole


Med Sci Sports Exerc. 2014;46(5):860-876. 

In This Article

Abstract and Introduction


Resolving the bases for different physiological functioning or exercise performance within a population is dependent on our understanding of control mechanisms. For example, when most young healthy individuals run or cycle at moderate intensities, oxygen uptake (V̇O2) kinetics are rapid and the amplitude of the V̇O2 response is not constrained by O2 delivery. For this to occur, muscle O2 delivery (i.e., blood flow × arterial O2 concentration) must be coordinated superbly with muscle O2 requirements (V̇O2), the efficacy of which may differ among muscles and distinct fiber types. When the O2 transport system succumbs to the predations of aging or disease (emphysema, heart failure, and type 2 diabetes), muscle O2 delivery and O2 delivery–V̇O2 matching and, therefore, muscle contractile function become impaired. This forces greater influence of the upstream O2 transport pathway on muscle aerobic energy production, and the O2 delivery–V̇O2 relationship(s) assumes increased importance. This review is the first of its kind to bring a broad range of available techniques, mostly state of the art, including computer modeling, radiolabeled microspheres, positron emission tomography, magnetic resonance imaging, near-infrared spectroscopy, and phosphorescence quenching to resolve the O2 delivery–V̇O2 relationships and inherent heterogeneities at the whole body, interorgan, muscular, intramuscular, and microvascular/myocyte levels. Emphasis is placed on the following: 1) intact humans and animals as these provide the platform essential for framing and interpreting subsequent investigations, 2) contemporary findings using novel technological approaches to elucidate O2 delivery–V̇O2 heterogeneities in humans, and 3) future directions for investigating how normal physiological responses can be explained by O2 delivery–V̇O2 heterogeneities and the impact of aging/disease on these processes.


Historically, the interpretation of whole-body physiological responses such as cardiac output, pulmonary gas exchange, or fractional O2 extraction (arterial-mixed venous O2 concentration difference) tacitly presumed homogeneity across body compartments (e.g.,).[80,98–100] Progress in the physiological understanding of control processes and their spatial dynamics, driven, in part, through empirical findings and technological advances, reveals that the whole-body response conceals a remarkable heterogeneity. Specifically, the human body's physiology (and pathophysiology) is governed by ~1028 atoms reacting and interacting as dictated by their own particular chemistry and surrounding milieu. Within this complexity, the level(s) at which determination of heterogeneity becomes scientifically meaningful and provides information about systems control is dependent on the physiological questions posed as tempered by the technology available (see Table 1 ).

This review investigates heterogeneity within the O2 transport system from mouth to muscle mitochondria and addresses the global question as to how body O2 delivery (cardiac output × arterial O2 concentration) is distributed and matched to the requirement for O2 utilization (V̇O2) at the level of individual tissues during exercise. Effective matching distributes O2 among and within muscles (muscle O2 delivery) according to their needs and facilitates high rates of mitochondrial adenosine triphosphate production, making sustained exercise possible. Ineffective matching may slow V̇O2 kinetics and force reliance on substrate-level phosphorylation, rapid expense of finite intramuscular glycogen stores, and accelerate exhaustion. In a relatively fit, healthy young adult from rest to exercise (e.g., cycling or running), cardiac output may increase from ~5 to more than 25 L·min−1 raising body O2 delivery proportionally from 1 to >5 L·min−1.[80] When that O2 is directed effectively among and within the exercising skeletal muscles, 80%–90% of the arterial O2 concentration is extracted, V̇O2 achieves its maximal value (i.e., V̇O2max) of >4 L·min−1, and effluent venous blood returning to the lungs has an O2 concentration of only 2–4 mL O2 per 100 mL. Determining how muscle O2 delivery and V̇O2 are matched and the heterogeneity associated with those processes yields crucial information germane to O2 transport systems control in health and dysfunction in diseases such as heart failure and diabetes. The balance between muscle O2 delivery and V̇O2 is crucial because this sets the microvascular or capillary PO2 and thus the upstream driving pressure for blood–myocyte O2 flux as well as influencing metabolic control via the impact on intramyocyte PO2 and V̇O2 kinetics. Elegant experimental designs and technological advances have afforded unprecedented capabilities to temporally and spatially resolve muscle O2 delivery and V̇O2 and their matching and dynamics at rest and during exercise. However, without further insights, the interpretation of muscle O2 delivery and V̇O2 (and microvascular PO2) is challenging. Specifically, it is straightforward that a low O2 delivery–V̇O2 ratio will result in a high fractional O2 extraction and, consequently, a microvascular PO2 so low that blood–myocyte O2 flux (and both muscle and pulmonary V̇O2 kinetics) will be impaired compromising mitochondrial control. However, to what degree a higher O2 delivery–V̇O2 ratio (and microvascular PO2) is beneficial in terms of optimizing V̇O2 kinetics and metabolic control in a specific region has not been determined. Moreover, that advantage must be considered with respect to potentially impoverishing (i.e., reducing the O2 delivery–V̇O2 ratio) in another spatially distinct muscle region. The resolution of these problems will come from determining the spatial and temporal heterogeneity existent in healthy muscle and the impact of conditions such as priming exercise, exercise training, and aging as well as chronic disease(s).

This review argues strongly for opening the "black box," which historically has been used—with some predictive success—through either necessity or convenience, to explain pulmonary and muscle V̇O2 (cf.[7] and[34]). Herein, recent advances in technology and approaches to understanding heterogeneity and O2 flux control are sequenced according to the levels(s) at which muscle O2 delivery and V̇O2 heterogeneity are addressed. In the Dynamic Heterogeneity of Exercising Muscle V̇O2 section, a top-down approach is followed (see Fig. 1) where heterogeneity is considered with respect to O2 delivery and V̇O2 among active (i.e., recruited skeletal muscles) and inactive (rest-of-body) compartments and how such behavior is reflected in the pulmonary V̇O2 dynamics. Whereas distinct heterogeneities across limbs (i.e., arms vs legs) are not addressed specifically the impact of such is implicit within the Dynamic Heterogeneity of Exercising Muscle V̇O2 section and the reader is referred to the work of Wray and Richardson,[102] which details important cross-limb vascular heterogeneities in young and older subjects across training states. In the Heterogeneity of Muscle Blood Flow and Microvascular V̇O2 in Animals section, animal investigations using techniques not possible in humans (i.e., radiolabeled microspheres, phosphorescence quenching) unveil how cardiac output can be distributed among and within muscles in the running rat as a function of muscle recruitment, oxidative capacity, and fiber type. Oxygen delivery–V̇O2 matching during electrically stimulated contractions of distinct rat muscles comprised of predominantly slow- or fast-twitch fibers assess the capacity for flexibility of the O2 delivery–V̇O2 relationship. An important cautionary note here is that fiber type and oxidative stratification within and among rodent muscles is generally far more distinct than that within or among human populations (cf. 4, 5 with 22, 50), and this undoubtedly leads to differences in energetic control (e.g.,[31]). It is, however, pertinent that there exists a substantial heterogeneity of fiber types across human locomotor muscles (e.g., soleus 88% Type I vs rectus femoris [RF] 36%), as a function of depth within individual muscles (deeper muscle regions >Type I)[50] and across elite athletic populations (i.e., Type I%, sprinters <30%, distance runners >70%). Thus, although findings in animals cannot tell us what happens in humans, they demonstrate what is physiologically possible and, as such, help frame human experiments and their interpretation. This is the case for investigations of human muscle using positron emission tomography (PET), near-infrared spectroscopy (NIRS), and magnetic resonance imaging (MRI) technology, which are addressed in the final two major sections (i.e., the Muscle Blood Flow Heterogeneity as Assessed by PET section and the Spatial Heterogeneity of Quadriceps Muscle Deoxygenation Kinetics section) of this review. Focus is maintained on the physiological data obtained from each technique placing onus on the reader to refer to the original papers for technical details.

Figure 1.

Rationale for review design and sequence of sections.

As schematized in Figure 1, overall heterogeneities of O2 delivery–V̇O2 matching will be assessed spatially within the whole body (~102 kg) down several orders of magnitude to the muscle microvascular level (10−5 kg). Although a greater resolution of O2 delivery–V̇O2 matching at the level of the individual RBC within a given capillary would doubtless be of great interest, this is currently infeasible in contracting skeletal muscle. Moreover, it is probable that the dominant control mechanisms for such (at least in health) are located upstream of the capillary, and the investigation of these processes is facilitated best by studying O2 delivery–V̇O2 matching within multiple capillary and fiber units, the greatest resolution investigated herein. Throughout this review, the reader is asked to keep in mind that heterogeneities, with respect to O2 delivery, V̇O2, and their matching, at all levels may result from differences in motor unit or muscle recruitment as well as inherent heterogeneities in vascular and/or metabolic control. Techniques for assessing muscle recruitment in vivo are generally crude (e.g., surface EMG) and/or may entail multiple assumptions (e.g., glycogen depletion, blood flow, and V̇O2). The dangers of not accounting for muscle recruitment patterns, for instance, across the human quadriceps muscles, when interpreting O2 delivery–V̇O2 relationships, have been emphasized by the work of Chin et al.[16] and are discussed in the Spatial Heterogeneity of Quadriceps Muscle Deoxygenation Kinetics section. Moreover, it should be appreciated that anatomically, microvascular units (i.e., terminal arteriole and dependent capillaries) are not spatially synchronized with distinct motor units and their fibers.[23] One consequence of this arrangement is that an individual capillary may abut two or more fibers, each with a very different V̇O2, creating broad extremities of micro-mismatch of O2 delivery and V̇O2 within contracting muscle that are hidden from the resolution of present technology.