Chronic obstructive pulmonary disease (COPD) is a highly prevalent process, characterized by lung involvement and by the presence of numerous systemic manifestations and/or extrapulmonary comorbidities.1 These include nutritional disturbances and skeletal muscle dysfunction,2,3 which are in turn interelated and determine per se worse prognosis.4,5 Dysfunction of the respiratory muscles leads to difficulties in ventilation, especially when requirements are increased. Meanwhile, limb muscle dysfunction impairs both exercise capacity and numerous daily-life activities.6 Although some causal factors are common to the various muscle groups of the body, others are more specific.3 This is the case of changes in ventilatory mechanics that mainly affect the respiratory muscles, and the loss of condition derived from less physical activity, that mainly affects the lower-limb muscles.3 In addition, the cellular and molecular mechanisms that contribute to dysfunction and associated structural and metabolic findings can be either general or specific to a muscle group.7 Other consequences of the above factors and mechanisms include changes in muscle bioenergetics, which becomes less efficient, at least in the case of lower-limb muscles.8 However, metabolic studies are associated with a number of important problems. On the one hand, in vivo studies require aggressive instrumentalization of the patient (e.g. catheterization of vessels entering and leaving the muscle),8 and is not always technically possible. On the other hand, in vitro studies require the immediately processing of biopsy samples.9 Therefore, few studies have evaluated the metabolic changes occurring in muscles with difficult access, such as breathing muscles, and most of these are performed using invasive procedures such as biopsy.10–12 However, greater understanding of the use of energy substrates in the muscles of COPD patients may be useful in the management of muscle dysfunction and changes in body composition and nutritional status often associated with the disease.

Imaging techniques have advanced dramatically in recent years. One such technique, positron emission tomography (PET), provides guidance on the origin of various injuries based on the rate of uptake and utilization of key metabolites. This technique was used in 2005 to assess the activity of thoracic and abdominal muscles of COPD patients, and more qualitative activity was observed among healthy subjects.13 A few years later, Osman et al. (2011) confirmed these findings by combining PET with computed tomography (CT) in inspiratory muscles.14 We hypothesized that this combined technique might be useful for in vivo assessment with low invasiveness of simultaneous metabolic activity of respiratory and peripheral muscles, since structural and metabolic findings in biopsies of COPD patients are largely divergent.3,7 Current radiotracer uptake measurement techniques will also allow more accurate observations. The aim of this study, therefore, was to obtain a general but quantitative evaluation of the synchronous metabolic activity of various muscle groups in COPD patients compared with that of patients with normal lung function.

The most relevant findings of this study are: (1) The combined use of CT-PET with 18F-FDG can be useful to simultaneously assess the glucose metabolism of different muscle groups (respiratory and peripheral) in vivo and in COPD patients; (2) at rest and under calm breathing conditions, heterogeneous uptake in the different muscle groups studied was observed, being higher in the main inspiratory muscle, the diaphragm, both in healthy subjects and COPD patients; (3) under the foregoing circumstance metabolic activity of the respiratory muscles of COPD patients is known to be higher than that of control subjects, and this was quantitatively confirmed, with a similar trend being observed the quadriceps; and (4) in COPD patients there is a direct relationship between the degree of air trapping and metabolic activity of both respiratory muscles and quadriceps.

The results obtained confirm that the metabolic activity at rest is different in different muscle groups, logically being higher in the respiratory muscles, since they contract regularly. In this regard, oxygen consumption of these muscles at rest is known to represent about 2%–5% of the whole organism in healthy subjects, but increases to more than double in well-nourished COPD patients.20,21 The use of oxygen in the aerobic pathways involves a proportional degradation of metabolic substrates such as glucose to obtain energy. It is therefore interesting to confirm that metabolic activity in the respiratory muscles of COPD patients is higher than that of healthy subjects, and a similar trend is observed in the peripheral muscles. Furthermore, the level of muscle metabolism is proportional to variables representative of changes in the respiratory drive. All this is broadly consistent with previous studies using other techniques.8,20,21 On the one hand, we know that mechanical system loads increase the work of breathing, and this leads to increased oxygen and energy consumption. Calculation of the former corresponds to approximately 1–2ml per extra ventilation liter in healthy subjects, and to 3–6ml/l in COPD patients.20,21 Moreover, several studies have shown that the structural and metabolic changes that occur in the muscles of COPD patients vary according to contractile groups; they all show functional impairment, but to a different degree.22 Furthermore, the respiratory muscles combine greater structural injury with increased presence of inflammatory cells and cytokines, oxidative stress, and apoptosis signs23,24; but they also exhibit more favorable changes such as increased capillarity, key aerobic pathway enzyme activity, and percentage of predominantly oxidative fibers.12,25–27 A novel and intriguing finding of this study is the higher radiotracer uptake observed in the crural diaphragm, which has a more anchoring function, compared to the costal portion, which acts mainly as a plunger. No previous studies have measured the differential energy consumption between different portions of the diaphragm, although this might be related to the known heterogeneity of its fiber composition, the regional glycogen content, or the different oxidative capacity of fast fibers.28,29

Regarding the two previous studies that used PET in the assessment of metabolic muscle activity,13,14 one of the notable developments of this study is the use of a radiotracer uptake quantification technique that allowed greater accuracy. Obviously, a purely qualitative assessment may show major differences between different muscles or individuals, but minor variations cannot be statistically assessed. Another innovation is the distinction between the activities of the costal and crural portions of the diaphragm, whose functions are different (plunger and anchorage of surrounding structures, respectively). Moreover, this study simultaneously assessed limb muscles, in which biopsies from previous studies have shown substantially different findings to those of the respiratory muscles. The simultaneous evaluation of peripheral and respiratory muscles is also interesting, because under certain circumstances both contractile groups seem to compete for blood flow and the consequent supply of metabolic substrates.30

In this study, the quadriceps muscle showed a trend toward increased glucose uptake in COPD patients. This had already been suggested in a previous study showing that this muscle has increased energy consumption in COPD patients on any activity level,8 despite a decreased concentration of aerobic structural elements, accompanied by oxidative stress, signs of lesion and apoptosis.24,31–37 This change in metabolic efficiency (bioenergetics) is not reversed with training, which suggests involvement of factors unrelated to deconditioning.

The upper-limb muscles, meanwhile, show an opposite trend to that observed in the quadriceps. This might be explained by the few structural and metabolic changes caused by COPD in these muscles.9,38 Generally speaking, it could be argued that the known heterogeneity of biopsy findings from different muscle groups is also reflected in the diversity in glucose uptake evidenced in the present study. This shows the diversity of disease impact, which is more extreme in the case of the respiratory muscles. Moreover, the increase in glucose-linked metabolic activity follows a dose–response pattern relative to the mechanical load (represented here by air trapping) sustained by patients. The relationship between increased static lung volume and the metabolic activity of inspiratory muscle (diaphragm in this case) probably depends on both changes in the breathing pattern and the gradual emergence of positive end-expiratory pressure, factors that must be overcome by muscle contraction before generating airflow. This would entail greater demand, and thus higher energy consumption. The expiratory muscle (rectus abdominis) is known to be part of a contractile group acting in COPD patients even at rest, since it is needed to overcome airway expiratory resistance. This activity is further increased by exercise, both to increase expiratory airflow and prevent hyperinsuflation.39 In subjects with normal spirometry, by contrast, glucose uptake was similar to that seen in limb muscles. This is logical, given the inactivity of the latter during the scan and the assumed relative passivity of the former during breathing at rest.40 On the contrary, abdominal muscles are also known to be actively recruited in healthy subjects during exercise (even light), with functional results similar to those mentioned in COPD patients.39

This study was limited to quiet breathing and the burdens of the disease in stable phase. The distribution of metabolic muscle activity and how this changes in COPD patients subject to different levels of respiratory load or general physical activity should also be evaluated. In these situations, several factors may influence metabolic capacity and activity, such as capillarity and blood flow, hemoglobin levels, diffusion of gas in tissue, and the use of oxygen and metabolic substrates, among others. Moreover, some studies show how these variables may behave differently in different muscle groups since, for example, blood flow is known to be reserved to the quadriceps during general exercise at high loads, but perfusion of the respiratory muscles of the thoracic box is restricted.30 Another interesting situation to be explored in the future is exacerbation of the disease. Undoubtedly, both the increased mechanical load and the systemic and pulmonary inflammation accompanying these situations could modify the metabolic activity of various muscle groups.

Potential clinical implications. The direct and linear relationship between mechanical load and muscle energy consumption observed in COPD patients in this study may have implications for patient management. Thus, patients with higher degree of air trapping are probably more susceptible to respiratory muscle failure in ventilatory stress situations. However, this hypothesis, which would involve stepping up monitoring of this subpopulation during severe exacerbations, should be confirmed by future studies involving instrumentally increased ventilatory loads.

Combined use of PET/CT, meanwhile, could become useful in identifying patients at risk of or susceptible to therapeutic interventions (e.g. selective muscle training) due to more or less altered bioenergetics in different muscles, without the need for invasive and technically complex procedures (biopsy, arterial and venous catheterization for measurement of local oxygen consumption). This metabolic characterization could be performed in a clinical setting. Of course, the low invasiveness and ease of use of PET/CT may also pave the way for research into muscle metabolism in respiratory diseases, facilitating understanding of the changes associated with these conditions.

Funded by: SAF2007-62719, SAF2011-26908, 2009SGR393 and CIBERES.

The authors declare no conflict of interest.