PET imaging of skeletal muscle

Skeletal muscle

Skeletal muscles comprise almost 30-40% of the total body mass in healthy individuals, and are therefore important in regulation of the metabolic homeostasis. Skeletal muscle accounts for the majority of insulin-stimulated glucose disposal. Muscle perfusion is intrinsically linked to cardiovascular system, as muscle can increase its metabolism up to 100 fold during extreme work.

Skeletal muscle accounts for 40-50% of total body water and about 1/3 of interstitial fluid volume (Bhave & Neilson, 2011).

Skeletal muscle cells (myocytes, myofibers) may be up to 30 cm long, with multiple nuclei beneath the cell membrane (sarcolemma). Capillaries are embedded in grooves of the sarcolemma, fastened by collagenous support which keeps them open even during muscle contraction. Bundles of collagen fibrils support the myocytes and nerves. Fascia is the fibrous connective tissue that surrounds skeletal muscles, important in force transmission to adjacent structures; mechanical load regulates contractile myofibroblasts that are found in fascia, and rapid changes in fluid characteristics alter the stiffness of fascia (Schleip et al., 2018). Muscle Extracellular matrix (ECM) constituents are mainly produced by fibroblasts, residing in the ECM, but other cell types, including myoblasts and pericytes, can also produce ECM proteins. Cytoplasm of the muscle cell (sarcoplasm) is mostly occupied by myofibrils, tightly packed filamentous contractile proteins. Most of the mitochondria are located in the space between myofibrils and sarcolemma (mitochondrial reticulum), close to the capillary grooves. Perivascular end of mitochondria are specialized in generation of membrane potential, while the end closer to myocytes is specialized in generation of ATP (Glancy et al., 2015).

Sarcolemma contains ion channels that, when opened, allow the diffusion of Na+ and K+ across the membrane, generating action potential that is transmitted along the myocyte membrane. Action potential starts at neuromuscular junction, where a motor neuron and myocyte meets. A single neuron is attached to multiple, even hundreds, myocytes, in order to stimulate the contraction of many myocytes simultaneously (motor unit). Neuromuscular junction is part of the cholinergic system, using acetylcholine (ACh) to transmit the signal across synaptic cleft. Endoplasmic reticulum in myocytes is called sarcoplasmic reticulum (SR), and it has a crucial role in muscle contraction: action potential in the sarcolemma opens Ca2+ channels in the SR, and the released Ca2+ is bound to troponin C, which promotes the shortening of the myofibrils. Ca2+-ATPase then pumps Ca2+ back into the SR.

Skeletal muscles contain combination of three myocyte types: fast oxidative fibres, slow oxidative fibres, and fast glycolytic fibres. Oxidative (aerobic) fibres contain myoglobin, high number of mitochondria, and have more capillaries, while glycolytic fibres contain only little or no myoglobin and few mitochondria. The fast fibres use have fast twitch speed, but also use ATP quickly. Fast oxidative fibres can also produce ATP quickly, but fast glycolytic fibres cannot, and therefore have low resistance to fatigue. Slow oxidative fibres are highly resistant to fatigue, since they can replenish ATP as it is consumed.

Muscle stem cells (satellite cells) are located between the basal lamina and the plasma membrane of myocytes. Stem cells activate in response to muscle injury and mechanical load, starting to proliferate, differentiate, and then fuse into existing myofibers or with each other to form new full muscle cells. Fibroblasts are required in this process. Stem cells are also involved in the maintenance and regeneration of the neuromuscular junctions. Pericytes reside in the microvascular basement membrane, and participate in vascular remodelling and ECM protein production. Fibro-adipogenic progenitors (FAPs) can differentiate into either fibroblasts or adipocytes, and while FAPs normally assist in muscle tissue regeneration after injury, they can also proliferate and produce the marked amounts of fibrotic and/or adipose tissue infiltration in muscles (Chapman et al., 2016).

Quadriceps muscle is frequently used in exercise studies since one-legged knee extension exercise is easy to accomplish in laboratory. In normal subjects quadriceps muscle mass was found to be 2.40±0.17 kg (range 1.36-3.40 kg) (Blomstrand et al., 1997). Richardson et al (1999) assumed that active muscle mass in knee extension exercise was 2.5 kg and in cycle exercise 7.5 kg. For PET studies at rest, femoral muscles are usually scanned, but lumbar muscle could be equally suitable (Yokoyama et al., 2013).

Skeletal muscle produces and releases metabolites, myokines, miRNAs, and exosomes, which can regulate the function of other tissues and organs, including liver, adipose tissue, pancreas, and heart (Barlow & Solomon, 2018). Muscle contraction induced factors are linked to the benefits of exercise (Febbraio, 2017).

Muscle atrophy

Loss of muscle mass (atrophy) occurs with ageing (primary sarcopenia). Muscle atrophy can also be caused by reduced physical activity, malnutrition, or many diseases (secondary sarcopenia). For instance, in patients with chronic kidney disease sarcopenia is common, and associated with reduced mitochondria and increased uncoupling of oxidative phosphorylation (Rao et al., 2018). Denervation of skeletal muscle results in atrophy where muscle fibers are replaced by connective and white adipose tissue (myosteatosis). Duchenne muscular dystrophy is one of genetic neuromuscular disorders which also lead to fat accumulation in muscles. While intramuscular fat content is higher in old subjects than in young, the percent increase may be caused by lower muscle mass, not necessarily by increased fat mass (Yoshiko et al., 2017). In subjects with COPD the intramuscular fat content (quadriceps muscle) was 52% in high BMI group and 34% in low BMI group, with no difference in fat-free muscle volumes (Vivodtzev et al., 2017). Sarcopenic obesity is a very common syndrome in old subjects, showing sarcopenia, insulin resistance, and obesity (Cleasby et al., 2016). Insulin increase protein synthesis and inhibits protein catabolism in muscle. Several adipokines are related to insulin resistance in muscle. Testosterone and growth hormone levels are lower in obesity.

In PET imaging studies the large fat depots can be omitted, based on CT images, but the connective tissue and fat cells entwined between muscle fibres will affect the PET results, and need to taken account in the interpretation of results.

Idiopathic inflammatory myopathies

Idiopathic inflammatory myopathies (IIM) are a heterogeneous group of rare, chronic autoimmune diseases. The most common types are dermatomyositis (DM), polymyositis (PM), necrotizing autoimmune myopathy (NAM), and sporadic inclusion body myositis (sIBM) (Malik et al., 2016). The muscle inflammation may lead to fibrosis, atrophy, and fat accumulation, visible with CT, MR, and PET (Schiffenbauer, 2014). Amyloid-β marker [11C]PIB uptake is increased in the gastrocnemius muscles in inclusion body myositis (Maetzler et al., 2011).


In healthy muscle, vasculature is highly organized (Kusters and Barrett, 2016). Arterioles branch to terminal arterioles, which surround muscle fibers at about 1 mm intervals. Each arteriole supplies 15-20 capillaries, which run parallel to muscle fibre, and, with numerous anastomoses, form a mesh of capillary network. In contracted state of the muscle the tortuosity of capillaries is reduced. Terminal arterioles with their capillary network form microvascular units. Capillaries have variable length (0.02-1 mm) and diameter (2-8 µm). Also inter-capillary red blood cell (RBC) flux and velocity vary over an order of magnitude (Poole et al., 2013). Venules are located between terminal arterioles. Lymph vessels may follow either arterioles or venules.

Terminal arterioles can increase the capillary blood flow through vasodilation and smooth muscle cell driven vasomotion (Kusters and Barrett, 2016). During light exercise, vasomotion may cease. Angiotensin II increases the flow, and serotonin decreases it, directing blood flow instead to slightly larger vessels located in the connective tissue (Zhang et al., 2005); this may be considered as non-nutritive flow (Clark et al., 2006). Non-nutritive flow may account for more than over half of the total blood flow (a href=””>Zhang et al., 2005; Newman et al., 2007). Sub-maximal exercise and hypoxia can produce a compensatory vasodilation, and increased blood flow as compared to exercise during normoxic conditions (Casey & Joyner, 2011). Endothelial cells and red blood cells participate in the local control of blood flow (Jensen, 2009; Kusters and Barrett, 2016).

Exercise training induces vascular adaptation through arteriogenesis and angiogenesis, remodelling and enlargement of arteries and arterioles, and improved control of vascular resistance. Different exercise types affect red and white muscle areas differently.

Insulin and mixed-meal challenge increases muscle perfusion and blood volume. Hyperglycemia, induced by oral glucose challenge (OGC) or oral glucose tolerance test (OGTT) decreases muscle blood flow despite increased insulin concentration (Russell et al., 2018).

Atherosclerosis causes progressive narrowing of arteries in many organs. Peripheral arterial disease (PAD) affects most commonly the circulation of legs, causing claudication, and possibly critical limb ischemia (CLI). Angiogenetic factors are released in response to ischemia, stimulating the development of collateral vessels. Chronic venous disease (CVeD) is very common in the lower limbs.

Superficial blood flow reduction can be detected using Laser Doppler perfusion imaging. Quantitative measurement of nutritive perfusion in skeletal muscle muscle is possible with [15O]H2O PET.


In the microcirculation of resting skeletal muscle, hematocrit is much lower than in larger blood vessels because red blood cells travel there faster than plasma. At rest, in almost all capillaries at least plasma is flowing, and in most capillaries there is also RBC flux, although with slow velocity. In working muscle the velocity, hematocrit, and RBC and plasma flux increases. In vivo microscopy has revealed that work or maximal vasodilation does not increase the proportion of capillaries with RBC flow (Poole et al., 2013). Partial pressure of O2 is low but almost uniform inside myocytes of a working muscle. O2 flux is dependent on the density of RBCs close to the myocytes, and usage in mitochondria, but not limited by the diffusion of O2. Oxygen diffusion is very fast because of the small size of O2 molecule and its lipophilicity. Even very low transit times do not limit RBC O2 offloading. However, RBC flux may be absent in about half of capillaries in certain diseases such as chronic heart failure, diabetes, and sepsis.

Near-infrared spectroscopy (NIRS) is commonly used to study the oxygenation status of skeletal muscle, as it allows quantification of deoxygenated and oxygenated haemoglobin; NIRS signals are attenuated by the adipose tissue, and NIRS measurements must be corrected for adipose tissue thickness (Craig et al., 2017). NMR techniques can be used to measure the oxygen saturation of myoglobin, although the precision in early studies may have been suboptimal (Tran et al., 1999). Muscle perfusion can be reduced by 50% before decrease in oxygen saturation can be observed with NIRS (Thomassen et al., 2017).

Oxygen uptake into skeletal muscle can be measured using [15O]O2-PET, and has been used already in 1979 to study the muscle function in peripheral vascular disease (Clyne et al., 1979), and later by Depairon et al (1988a, 1988b, 1988c, 1991, and 1996). Oxygen extraction ratio (OER), measured using [15O]O2, is decreased in subcutaneous tissue in venous ulceration and lipodermatosclerosis (Hopkins et al., 1983; Spinks et al., 1985). The relatively low perfusion combined with high oxygen retention capacity of myoglobin may hamper precise quantification of OER.


Due to its large proportion of body size, muscle is an important storage site of carbohydrates, amino acids, and fat. Fatty acid oxidation is the major source of ATP for skeletal muscle during resting state. During intense but short exercise, muscle glycogen and blood glucose are the preferred fuels. Lactate production, accumulation, and release are increased when oxygen supply to mitochondria is limited, but muscle produces and releases lactic acid and alanine even during normoxia, particularly during rapid glycolysis, because lactate dehydrogenase (LDH) has the highest Vmax of enzymes in the glycolytic pathway (Gladden, 2001). Cycling exercise can reduce the leg muscle glycogen content to ∼1/3 or the initial concentration, with no change in extra- and intracellular water content (Shiose et al., 2018). During prolonged and low-to-moderate exercise, plasma fatty acids are the major energy source. Subjects with higher mechanical efficiency (ME, ratio of mechanical work to metabolic energy expended) have higher muscle fatty acid uptake than subjects with low ME, while muscle perfusion, glucose uptake, and delivery of fatty acids did not differ (Laaksonen et al., 2018).

Glucose is transported across the endothelium via GLUT1, which is expressed in endothelial and smooth muscle cells. GLUT4 transports glucose into myocytes. Insulin induces translocation of GLUT4 from intracellular vesicles to the myocyte plasma membrane, increasing glucose uptake and net glycogen synthesis. Glucose consumption can be measured using [18F]FDG, and glucose transport with [11C]3-O-methylglucose (Ng et al., 2014) or [18F]2-fluoro-6-deoxy-D-glucose (Huang et al., 2012).

Endothelial cells express fatty-acid binding proteins, and myocytes release lipoprotein lipase (LPL). Myocytes contain lipid droplets, but also intermuscular adipocytes exist, increasingly in old age, obese subjects, and in diseases such as type II diabetes, myositis, osteoarthritis, spinal stenosis, and cancer (Aubrey et al., 2014; Correa-de-Araujo et al., 2017). Muscle fatty acid metabolism can be measured using fatty acid tracers such as [1-11C]palmitate and [18F]FTHA.


Inflammation causes increased glucose uptake in the inflamed tissue, often via upregulation of GLUT1 and GLUT3, which can be detected using FDG PET. Also FDG-labelled white blood cells accumulate to the site of muscle inflammation (Pellegrino et al., 2005). TSPO tracers may detect the macrophages, neutrophils and lymphocytes in the inflamed muscle with better specificity than FDG (Wu et al., 2014); however, in another animal study no uptake was observed (Zheng et al., 2016).

Phosphatidylserine tracer [18F]FEN-DPAZn2, developed for cell death imaging, was found to have relatively high uptake in aseptic muscle inflammation model in mice (Liang et al., 2014).

Inflammation also typically leads to vascular leakage and oedema which can be detected using in vitro or in vivo labelled albumin (Niu et al., 2014).

Adverse reactions to metal debris (ARMD), caused by metal-on-metal bearings in total hip arthroplasties and hip resurfacing arthroplasties, is characterized by local severe inflammation necrosis, and fibrin deposition. FDG PET can be used to detect the inflammatory reaction in the hip and gluteal muscle region (Aro et al., 2017).

See also:


Arpino JM, Nong Z, Li F, Yin H, Ghonaim N, Milkovich S, Balint B, O’Neil C, Fraser GM, Goldman D, Ellis CG, Pickering JG. Four-dimensional microvascular analysis reveals that regenerative angiogenesis in ischemic muscle produces a flawed microcirculation. Circ Res. 2017; 120(9): 1453-1465. doi: 10.1161/CIRCRESAHA.116.310535.

Aubrey J, Esfandiari N, Baracos VE, Buteau FA, Frenette J, Putman CT, Mazurak VC. Measurement of skeletal muscle radiation attenuation and basis of its biological variation. Acta Physiol. 2014; 210(3): 489-497.

Barrett EJ, Rattigan S. Muscle perfusion - its measurement and role in metabolic regulation. Diabetes 2012; 61(11): 2661-2668.

Chapman MA, Meza R, Lieber RL. Skeletal muscle fibroblasts in health and disease. Differentiation 2016; 92: 108-115.

Cleasby ME, Jamieson PM, Atherton PJ. Insulin resistance and sarcopenia: mechanistic links between common co-morbidities. J Endocrinol. 2016; 229(2): R67-R81.

Correa-de-Araujo R, Harris-Love MO, Miljkovic I, Fragala MS, Anthony BW, Manini TM. The need for standardized assessment of muscle quality in skeletal muscle function deficit and other aging-related muscle dysfunctions: a symposium report. Front Physiol. 2017; 8:87. doi: 10.3389/fphys.2017.00087.

Glancy B, Hartnell LM, Malide D, Yu Z-X, Combs CA, Connelly PS, Subramaniam S, Balaban RS. Mitochondrial reticulum for cellular energy distribution in muscle. Nature 2015; 523(7562): 617.

van Hall G. The physiological regulation of skeletal muscle fatty acid supply and oxidation during moderate-intensity exercise. Sports Med. 2015; 45(Suppl 1): 23-32.

Hudlicka O. Microcirculation in skeletal muscle. Muscles Ligaments Tendons J. 2011; 1(1): 3-11.

Jensen FB. The dual roles of red blood cells in tissue oxygen delivery: oxygen carriers and regulators of local blood flow. J Exp Biol. 2009; 212(Pt 21): 3387-3393.

Joyner MJ, Casey DP. Muscle blood flow, hypoxia, and hypoperfusion. J Appl Physiol. 2014; 116(7): 852-857. doi: 10.1152/japplphysiol.00620.2013.

Kusters YH, Barrett EJ. Muscle microvasculature’s structural and functional specializations facilitate muscle metabolism. Am J Physiol Endocrinol Metab. 2016; 310(6): E379-E387. doi: 10.1152/ajpendo.00443.2015.

Ng JM, Bertoldo A, Minhas DS, Helbling NL, Coen PM, Price JC, Cobelli C, Kelley DE, Goodpaster BH. Dynamic PET imaging reveals heterogeneity of skeletal muscle insulin resistance. J Clin Endocrinol Metab. 2014; 99(1): E102-E106. doi: 10.1210/jc.2013-2095.

Poole DC, Copp SW, Ferguson SK, Musch TI. Skeletal muscle capillary function: contemporary observations and novel hypotheses. Exp Physiol. 2013; 98(12): 1645-1658. doi: 10.1113/expphysiol.2013.073874.

Rudroff T, Ketelhut NB, Kindred JH. Metabolic imaging in exercise physiology. J Appl Physiol. 2018; 124: 497-503. doi: 10.1152/japplphysiol.00898.2016.

Tanaka S, Ikeda K, Uchiyama K, Iwamoto T, Sanayama Y, Okubo A, Nakagomi D, Takahashi K, Yokota M, Suto A, Suzuki K, Nakajima H. [18F]FDG uptake in proximal muscles assessed by PET/CT reflects both global and local muscular inflammation and provides useful information in the management of patients with polymyositis/dermatomyositis. Rheumatology 2013; 52(7): 1271-1278.

Wu C, Yue X, Lang L, Kiesewetter DO, Li F, Zhu Z, Niu G, Chen X. Longitudinal PET imaging of muscular inflammation using 18F-DPA-714 and 18F-alfatide II and differentiation with tumors. Theranostics 2014; 4(5): 546-555.

Zeller-Plumhoff B, Roose T, Clough GF, Schneider P. Image-based modelling of skeletal muscle oxygenation. J R Soc Interface 2017; 14: 20160992. doi: 10.1098/rsif.2016.0992.


Created at: 2017-09-02
Updated at: 2018-08-23
Written by: Vesa Oikonen