Skeletal muscles in chronic obstructive pulmonary disease: Deconditioning, or myopathy?


Peter D. Wagner, Division of Physiology, University of California, San Diego, 9500 Gilman Drive, DEPT 0623A, La Jolla, CA 92093-0623, USA. Email:


Abstract:  In recent years, COPD has become increasingly thought of as a systemic disease affecting many tissues and organs in addition to the lungs. The skeletal muscles in particular have been the target of much research focusing on whether the universally observed exercise limitation reflects a systemic myopathic effect of COPD, or simply the consequences of extreme, long-term inactivity. In this paper, the evidence is reviewed for COPD patients without loss of muscle mass and who are not taking systemic steroids. While altered levels of antioxidant defences (lower), circulating inflammatory biomarkers (higher) and anabolic hormones (lower) have been found in COPD, cause and effect remains to be established for the link of inflammation/oxidative stress to muscle dysfunction. Other evidence used to propose a myopathic state (early lactate release, reduced power output, lower metabolic enzyme capacities, greater phosphocreatine breakdown and slower phosphocreatine restoration after exercise, and altered fibre type distribution) also occur in normal subjects who are extremely inactive. Furthermore, intense small muscle mass training can normalize small muscle function in these patients. Based on these data, it remains to be shown that the muscles in COPD patients without loss of muscle mass are myopathic. The interesting discussion about systemic effects of COPD should not get in the way of systematic muscle training, which has been shown to be an effective component of rehabilitation.


It is common knowledge that patients with COPD have reduced exercise capacity. All agree that airways obstruction is the major reason, because this limits the ability to breathe during exercise. Normally, ventilation increases relatively at least as much as does metabolic rate, and this allows the arterial inline image and inline image to remain at or near resting levels up to very high levels of exercise intensity.1 However, in COPD, ventilation is limited and this in turn reduces exercise capacity.

COPD as a systemic disease

Over the past few years, several investigators2–4 have proposed that COPD is a systemic disease and that one manifestation of its systemic nature is a series of changes in locomotor skeletal muscles that taken together constitute a myopathy. If by ‘systemic’ is meant manifestations outside the lungs, there is little doubt that COPD is a systemic disease, especially in some patients. A certain percentage of COPD patients develop cachexia with severe skeletal muscle wasting that even very sedentary normal individuals do not have. Others develop sleep-disordered breathing, which is known to have multiple systemic manifestations. Yet others develop cor pulmonale with right heart failure, which also causes manifestations outside the lungs. In addition, the regular use of corticosteroids in some patients produces multiple systemic side-effects. Those who receive new lungs by transplant often develop systemic manifestations attributable to their immunosuppressive drug therapy.

Muscle dysfunction versus myopathy

Even those COPD patients without loss of muscle mass can be shown to exhibit features that some have called evidence of a systemic disease with myopathy. Altered muscle redox status,5 changes in muscle structure,2,4,6 changes in muscle oxidative enzyme levels,7 slow restoration of muscle phosphocreatine stores after exercise,8 early lactate appearance during exercise,9 increases in circulating pro-inflammatory cytokines10,11 and reductions in hormones such as testosterone and insulin-like growth factor (IGF) that normally support muscle structure and function10,12 have all been invoked as arguments for myopathy in such patients.

There is no argument against the claim that some patients with COPD have systemic disease leading to what all would agree amounts to a myopathy. Those with cachexia, those with steroid-induced myopathy and those with immunosuppression-induced myopathy following transplant are undeniable examples.

The purpose of this review is to critically examine whether evidence supports, as has been claimed, the existence of myopathy in that majority group of COPD patients with reduced exercise capacity but who do not show muscle wasting.

Even in this group of patients, there is no disagreement that there is evidence of abnormal muscle function. The fundamental question is whether this represents anything more than the results of long-term physical inactivity. Does the impairment of function reflect an intrinsic state of muscle pathology, or is the impaired function just reflective of the long-standing abnormal environment in which the muscle is placed?—very inactive, subject to a low inflowing arterial inline image, and in an older age group susceptible to peripheral vascular disease.

Why is the question even important?

For two reasons. First, future therapeutic targets in COPD will address ever more specific manifestations. If indeed there is myopathy, it would direct research to explore muscle pathogenetic mechanisms in search of abnormal pathways and specific therapeutic targets. If there is no myopathy, scarce research resources would be better directed to other areas. Second, at present there is precious little we can do for the lungs of patients with COPD. Formal rehabilitation programmes have been shown to improve quality of life in COPD patients.13 Regular exercise is a key part of these programmes. We should look for every motivation possible to encourage the training of skeletal muscles. If the problem is just detraining, there is no psychological barrier to push training. On the other hand, a myopathy would signify a likely worse prognosis and could reduce enthusiasm for effective muscle training. There is good evidence that trained muscle requires less pulmonary ventilation than untrained muscle, in large part because of less lactate production at any given workload. This reduces dyspnoea and enables higher workloads even in the face of unchanged lung disease per se. Add to that the possibility of weight reduction, and the ventilatory cost of exercise would fall even more.

Why has the question not yet been settled?

In large part because investigators have used methods of exercise evaluation that do not allow one to establish the presence or absence of myopathy. In particular, employing large muscle mass exercise (treadmill, cycle) takes the patient to maximal ventilation but not to maximal muscle effort. Another problem is finding the right normal subjects as controls—not simply matched by age, gender and size but most importantly by the very low levels of physical activity typical of COPD patients. When otherwise matched controls simply do not engage in regular exercise, one usually finds much more activity than in COPD patients, and so the patient data will always look abnormal. To some extent, the problem exists because the markers of inactivity and of myopathy considerably overlap. And because some biological markers such as hormones, cytokines and redox state may differ between patients and controls, thus does not prove cause and effect.


Table 1 lists 10 lines of evidence that have been used to argue the case. We will examine these one at a time.

Table 1.  Types of evidence cited to decide the issue
  1. IGF, insulin-like growth factor; PCr, phosphocreatine; test., testosterone; TNF, tumour necrosis factor.

1.Poor relationship between exercise and FEV1
2.Reduced peak power output
3.Early blood lactate accumulation
4.Reduced muscle oxidative enzyme capacity
5.Greater PCr breakdown; slower PCr recovery
6.Muscle histology: type II fibres ↑, capillaries ↓
7.Circulating inflammatory cytokines (TNF, IL-6)
8.Reduced anabolic hormones (test., IGF-1)
9.High reactive O2 species; impaired antioxidant defence
10. Irreversibility with training

There is a poor relationship between FEV1 and exercise capacity

As shown by Jones et al. many years ago, this statement is undoubtedly true.14 But ability to breathe is not the only determinant of exercise capacity. Determinants of muscle O2 delivery (arterial inline image, [Hb], cardiac output and leg blood flow) also contribute, as will orthopaedic factors, body mass variability and levels of daily activity. How could one expect a good correlation between just one determinant (FEV1) and exercise capacity when there are so many other factors involved? One needs to perform multivariate analyses to identify the importance of each factor. Thus, the argument (that, because FEV1 alone is a poor predictor of exercise capacity, there must be a myopathic component) is weak.

Patients with COPD have a low peak power output

Of course they do. They cannot breathe enough to allow them to reach power outputs that normal subjects can produce. But if you use the strategy of studying exercise of a muscle mass small enough not to require the patient to breathe maximally, patients can reach the same peak muscle inline image as do normal subjects. This was shown by Richardson et al., whose data appear in Figure 1.15 Even those who argue the case for a myopathy admit that muscle function when studied outside the constraint of maximal ventilation may be normal. Thus, Debigare et al. found that the contractile properties of the vastus lateralis were preserved,16 and Bernard et al. found that muscle strength per unit cross-sectional area was not impaired.17

Figure 1.

Peak one-leg inline image in patients with COPD and in activity-matched controls. Top panel: during two-legged cycling when exercise is limited primarily by ability to breathe; lower panel: the same patients on the same day, during one-legged knee extensor exercise when ventilation is only about 60% of maximal. The lower panel shows that in the absence of ventilatory limitation, peak leg inline image is not impaired (in the absence of muscle wasting). From the study by Richardson et al.15

Early blood lactate accumulation

All agree that blood lactate levels start rising at lower work rates in patients than in normal subjects.9 But this is very well known to be a manifestation of the detrained state. Exercise training of patients with COPD delays lactate appearance in exercise,18 just as is well known to occur in normal subjects. This is in fact one of the keys to successful training in both circumstances. It is also well known that arterial hypoxaemia elevates blood lactate at a given exercise load in normal subjects. Thus, early lactate appearance is not at all specific to myopathic states and simply cannot be taken as evidence of myopathy.

Reduced muscle oxidative enzyme activities

This too is not questioned as an observation.7 But just as for lactate levels, low Krebs cycle oxidative enzymes are the hallmark of the untrained state, and increase substantially with training. Interestingly, muscle cytochrome oxidase activity is inversely related to arterial inline image in COPD patients.19 That is, the levels are higher in those with more severe hypoxaemia, suggesting a compensatory response to reduced O2 availability to augment ATP production, as shown in Figure 2. And Gea et al. found that oxidative enzymes were not reduced in the deltoid (upper arm) muscle of COPD patients.20 This heterogeneity of responses argues against a blanket systemic interference with the oxidative process.

Figure 2.

Cytochrome oxidase levels ascertained from vastus lateralis biopsy in patients with COPD in relation to their arterial inline image. There is a clear association between arterial hypoxaemia and enzyme activity suggesting hypoxaemia stimulates this enzyme in compensation for limited O2 availability. COX, cytochrome oxidase. From the study by Sauleda et al.19

Greater phosphocreatine (PCr) breakdown and slower recovery after exercise

This also goes with state of training. Untrained normal subjects show greater PCr breakdown early in exercise, probably due to slower inline image kinetics and slower PCr recovery rates after exercise, reflecting lower mitochondrial activity. These parameters are returned essentially to normal in COPD patients by intense muscle training.8

Muscle histological changes

Various authors have found different structural consequences of COPD in biopsied leg muscle. Some find fewer capillaries4 but their control subjects were considerably more active than the patients, which may explain the difference. Richardson et al. found no reduction in muscle capillarity using control subjects carefully selected for similar degrees of inactivity as patients (Fig. 3).15 There is one finding that all agree on—the percentage of type II fibres is increased in COPD. This may explain a functional observation—that while muscle peak inline image is not reduced when measured during small muscle mass exercise, the power output produced is slightly less in COPD than normal.15 It is known that type II fibres are less efficient in their use of O2 than are type I fibres. Is this evidence for myopathy? Possibly. However, Larsson and Ansved found in normal subjects undergoing long-term, extreme inactivity that there was also a higher percentage of type II fibres.21 Thus, it is possible that these findings are again the result of extreme, prolonged inactivity.

Figure 3.

Lack of histological changes in the vastus lateralis of patients with COPD, compared with well-selected activity-matched control subjects. Based on data from the study by Richardson et al.15

Increased circulating pro-inflammatory cytokines

Both tumour necrosis factor (TNF)-a and IL-6 have been found to be elevated in circulating blood of patients with COPD. The source of these molecules is not known, but more importantly their functional significance remains to be established. At the present time, this can be considered only as association and not as evidence of a myopathy. That said, TNF-a levels are especially elevated in those COPD patients with cachexia, and this may end up being cause and effect. But in patients without loss of muscle mass, this is far less certain.

Reduced circulating anabolic hormones

Testosterone and IGF-1 levels are found to be reduced in many patients with COPD.10–12 Trials of hormone therapy have not been successful in enhancing exercise capacity,22,23 and once again, while a reasonable hypothesis, whether reductions in levels of such hormones lead to myopathy remains to be seen. As with the cytokine story, cause and effect remains to be established.

Impaired muscle antioxidant defences

Two groups have found evidence for impaired antioxidant levels in COPD patients.3,5 In addition, generation of reactive O2 species may be increased in COPD, which could cause greater inflammation leading to muscle damage. Whether this is associated with myopathy also remains to be clarified. Interestingly, it has been found that in muscle reactive O2 species generation may be part of the signalling cascade that leads to positive adaptation of muscle to training.24,25

Ability to respond to muscle training

It is well accepted that the muscles of patients with COPD can be trained by all criteria commonly used: greater exercise tolerance and inline image, delayed lactate appearance and lower ventilatory cost of exercise, increased oxidative enzyme capacity, faster PCr recovery, increased muscle strength and mass, etc. Even large muscle mass training8,18,26–28 leads to such adaptations, in spite of the low levels of workload imposable because of ventilatory limitation. When small muscle mass training is used, the effects are even clearer. If the structure and function of the muscles of COPD patients can be returned to normal by just 8 weeks or so of training in a disease that likely has been present for well over 10 years in most cases, it is hard to make the case that significant myopathy exists.


This short review has examined evidence used to infer the existence of a myopathic state in the skeletal muscles of patients with COPD. Leaving aside those patients with obvious muscle pathology from cachexia, steroids alone, or steroids and other immunosuppressives after transplant, the evidence for anything other than a high degree of deconditioning is weak. The take-home message is that muscle training, especially small muscle mass training, can be accomplished and is effective in returning muscle function per se to normal or near normal levels. This should bring positive reinforcement to those afflicted by this disease because training, as part of rehabilitation, leads to improved quality of life.