• Exercise

Limitation of exercise capacity plays a central role in the life of the patient with chronic obstructive pulmonary disease (COPD), both as a marker of well-being1 and as an indicator of a poor prognosis.2 Our ability to characterise this crucial aspect of disease has grown rapidly in the last decade and with this so has our understanding of the many complex reasons for exercise impairment. It has long been recognised that the maximum ventilation during exercise is related to the initial FEV₁ (forced expiratory volume in 1 s), with several formulae being developed to predict this. It was accepted that an inability to sustain a high level of ventilation would limit exercise performance in COPD, although exactly why this happened was uncertain. In the last decade there has been compelling evidence that changes in the operating lung volumes during exercise lead to mechanical limitation of inspiration and hence of tidal volume, which is associated with the sensation of breathlessness.3 4 Dynamic hyperinflation is a very consistent finding in COPD and can even occur early in the natural history of COPD, at least in symptomatic people.5 However, not all patients are limited exclusively by breathlessness on exertion, and data from the McMaster group in the 1990s pointed out that many patients were limited by a feeling of heaviness or fatigue in their legs, either along with breathlessness or dominating this sensation.6 As a result, attention began to turn to other factors, such as co-morbid cardiac disease and the possibility that skeletal muscle itself was not normal in COPD.7

One way of exploring the impact of lung disease on exercise performance is to study people before and after treatments designed to improve lung mechanics in some way. Bronchodilators improve operating lung volumes in subjects at rest and have been shown to increase exercise performance measured in a variety of ways in COPD. In a series of studies, Professor O'Donnell has shown that bronchodilators, both anticholinergics and β-agonists, improve exercise capacity and that this is best related to falls in the end-expiratory lung volume—that is, an increase in inspiratory capacity at rest and during exercise.8–10 Not all people benefit from this treatment, and we have previously described paradoxical responses where lung function appears to improve at rest but the response of the patient is abnormal and exercise tolerance worsens.11 How widespread such effects are has not been documented so far. Other groups found that bronchodilators were less effective when patients' exercise performance was limited by the development of quadriceps muscle fatigue.12 These studies, and many others, have contributed to an active debate about the mechanisms limiting exercise performance in COPD which has recently been summarised.13

One important issue is the potential interaction between the effort made during respiration in patients with COPD and the cardiac output. Patients with COPD develop marked intrathoracic pressure swings which relate to some degree to the extent of abdominal muscle activation at rest and during exercise.14 There are now data suggesting that patients who show paradoxical movement of the lower ribcage are more likely to develop breathlessness on exercise, while those without this feature show a different pattern of chest wall hyperinflation and may be limited by leg fatigue.15 Thus the behaviour of the respiratory muscles and hence the intrathoracic pressures may be relevant to exercise performance. Direct measurements of cardiac output and peripheral muscle performance are inevitably difficult in these circumstances. The group from Brazil who report new findings in this issue of Thorax (see page 588)16 have previously established that there is a mismatch between oxygen uptake in the body in general and the rate at which oxygen in consumed within exercising skeletal muscle.17 Moreover, when subjects were breathing heliox, a gas mixture which reduces the mechanical work of breathing and also improves exercise performance in COPD,18 then the matching of oxygen delivery and oxygen uptake to the leg muscles of the patient with COPD was improved.19 Similar findings were seen when patients exercised while using proportional assist ventilation to decrease respiratory muscle work.20 Now Berton and colleagues have extended these observations to the clinically more relevant area of bronchodilator treatment.

They studied 12 patients with severe COPD (FEV₁ 38.5% predicted) without resting arterial hypoxaemia in a series of constant work-rate cycle ergometer exercise tests conducted at 70–80% of a previously determined peak exercise performance. They did so before and after the patients had received the combination of ipratropium bromide and salbutamol by inhalation or an identical placebo. Their main outcomes were the speed with which whole body oxygen uptake changed during exercise and how this related to the change in oxygenation within the vastus lateralis muscle using near-infrared spectroscopy and in cardiac output measured by impedance cardiography. This comprehensive analysis allowed them to look at metabolic and cardiac output changes during exercise in COPD as well as the kinetics of peripheral muscle de-oxygenation. These complex studies produced some interesting results. As expected, the bronchodilators increased the exercise endurance time in constant work-rate exercise by almost 40% of the baseline performance, and the authors showed that there was less dynamic hyperinflation which they measured by the change in inspiratory capacity during exercise. The bronchodilators produced small but statistically significant improvements in both FEV₁ and resting inspiratory capacity (130 and 390 ml, respectively), although only 7 of the 12 subjects would have met the conventional criteria for bronchodilator responsiveness. When the bronchodilators were administered there was an improvement in the speed with which the oxygen uptake and cardiac output reached the steady state, and there was a slowing of the desaturation which occurred in the exercising muscle measured as a change in the ratio of deoxygenated haemoglobin to myoglobin. Mismatch between delivery and oxygen consumption was improved by the bronchodilator drugs, and the greatest improvements in cardiac output kinetics were seen in the patients who showed the biggest reduction in dynamic hyperinflation postbronchodilator.

These elegant physiological studies have applied new methodologies which directly measure key variables that determine exercise performance in COPD. Inevitably the number of participants is small and the data apply to those with relatively severe disease. There is continuing controversy about the use of impedance cardiography to monitor cardiac output in patients with hyperinflated lungs, although the group from Sao Paulo have provided further evidence for its appropriateness in these studies. In practical terms, more direct and invasive methods of measuring cardiac output would pose formidable challenges even for determined investigators and patient volunteers, although data such as these may emerge in due course. Similarly, there has been some uncertainty about exactly what near-infrared spectroscopy is recording and how representative it is of overall muscle oxygenation. Despite these issues the data here represent a considerable technical achievement and point to a clinically relevant effect of bronchodilators on tissue oxygen delivery as well as lung function. How important these effects are in the overall limitation of exercise performance and why oxygen delivery is limited will be debated, although a direct effect of hyperinflation on cardiac function output seems increasingly likely. Whatever the mechanisms are, these data have implications for the way we view the symptomatic treatment in COPD. The effects of bronchodilators on cardiac function represent an additional potentially beneficial effect which could explain why in some circumstances, far from increasing cardiac mortality, use of long-acting inhaled bronchodilators is associated with a lower cardiac event rate.21 It is also relevant to ask our patients why they are limited when they exercise, and recognise that developing tiredness in the legs is not just a sign of being unfit but may be a direct effect of lung disease on their muscle function. We should no longer be surprised when they tell us their legs are less tired after taking their inhalers because, as usual, the patient has got it right before the doctor can explain why this happens.