Proper evaluation of potential patients is the keystone to a successful pulmonary rehabilitation programme.1 The so-called functional approach which relates toimpairment (the physiological deficit),disability (total effect of impairment on the patient’s life), and handicap (the social disadvantages) as part of the comprehensive programme of care is one approach. It is useful not only for monitoring the patient’s functional status, but it enables the rehabilitation team to set and to achieve goals to improve the quality of the patient’s life.

Although an exercise test may help to uncover other coexisting diseases, it is generally necessary to assess the patient’s exercise tolerance and to evaluate possible blood gas changes which cannot be predicted from baseline lung function tests.2 The exercise test is also used to establish a safe and appropriate prescription for subsequent training. Cardiopulmonary exercise testing has been found to be useful in the evaluation of exercise tolerance in patients with dyspnoea and cardiopulmonary diseases. Compared with a clinical laboratory approach, it allows the detection of an underestimated circulatory component causing exercise limitation.3 Graded exercise testing has been shown to be useful in diagnosing patients with chronic unexplained dyspnoea although it was not sensitive in distinguishing a cardiac disease from deconditioning.4Variables measured and/or monitored during testing should include workload, heart rate, electrocardiogram, arterial oxygenation, and symptoms. Blood gas sampling during exercise adds a significant degree of complexity to testing. Non-invasive techniques such as cutaneous oximetry of arterial oxygen saturation are useful for continuous monitoring but should not be relied on for precise assessment of arterial oxygenation because of their limited accuracy.5Other measurements such as analysis of expired gas to calculate variables such as oxygen consumption (V˙o ₂) may be performed, depending on the interest and expertise of the referring physician, laboratory personnel, and programme staff. Assessment of exercise is best made using the type of exercise that will be employed in training—for example, treadmill testing for a walking exercise training programme; however, results from one type of exercise test can be translated to similar forms of exercise.6 Indeed, although regular bicycle exercise was unfamiliar and generated the greatest lactate response in patients with chronic obstructive pulmonary disease (COPD), peak V˙o ₂ was the same during cycle and treadmill exercise.7

Laboratory tests are the gold standard measurements. However, “field tests” can provide a useful assessment of task performance when laboratory facilities are unavailable. Timed walking tests can be used to measure exercise capacity indirectly following rehabilitation, particularly when limited resources are available. Measuring the distance covered during a walking test is considered a simple and reproducible way to determine exercise tolerance in patients with chronic lung disease. The first test proposed involved measuring the distance covered over 12 minutes (12MWD, 12 minute walking distance). It was shown that the distance walked was related to peak V˙o ₂. The test has been progressively shortened to the now most frequently used 6MWD and, more recently, a 2MWD test has been proposed.8-10 The 6MWD has been shown to predict survival in patients with COPD and heart failure.11 ,12 It has been shown that, in patients with stable COPD, the smallest difference in 6MWD distances that was associated with a noticeable difference in patients’ subjective comparison ratings of their walking ability was 54 m.13 The main advantages of walking tests are simplicity, minimal resource requirements (a corridor and a supervisor), and general applicability. The main disadvantages of these tests are patient and supervisor susceptibility to motivation, their non-standardised nature, and their dependence on a single quantitative measure of distance covered. Although walking tests are capable of meeting stringent test-retest criteria, the plethora of circumstances in which testing takes place limits comparison of the magnitude of various rehabilitation treatment results from different centres. Where facilities are limited the timed walking test remains a simple method for assessing exercise capacity in individual patients provided that reproducibility of the measurement is demonstrated. The effects of learning on initial walks need to be taken into account.14 ,15 Furthermore, the lack of control of the workload in the 6MWD does not define the type of activity in any individual.

The basic function of walking is one of the five major life activities (with breathing, hearing, seeing, and speaking). Significant impairment in at least one of these would constitute a disability. For walking to be impaired and considered a disability with regard to independent ambulation, qualitative (dynamic balance sufficient for safety to prevent falls) and quantitative (velocity and distance tolerated) parameters need to be described and compared with realistic functional demands in the community. A review of the literature reports a normal walking velocity of 83 m/min.16 These velocity values are comparable to the functional velocity needed (79 m/min) to cross a traffic intersection from kerb to kerb with the crossing signal lit, and to the velocity determined for urban pedestrians. In addition, 600 m was the longest distance found necessary for a person to walk to visit a common commercial destination in the community.17Thus distance and velocity criteria for walking both require consideration for arriving at or defining a person’s level of independence regarding functional mobility. A review of the literature shows that the distances walked in 12 minutes by patients with COPD with different degrees of airflow obstruction ranged from 315 to 1215 m. Nevertheless, studies do not usually report the number of rest stops or the total rest time taken during performance of the walking tests, so it is difficult to interpret accurately the degree to which the major life activity of walking is significantly impaired for independent community ambulation.

The recent Pulmonary Rehabilitation Joint ACCP/AACVPR evidence based guidelines18 recommend that patients with COPD who undergo a programme of lower extremity exercise training consistently improve measures of exercise tolerance without evidence of adverse outcome. This recommendation is based on the results of 14 randomised controlled trials that have appeared in the peer reviewed literature. Outcome measures in these studies included timed walking tests, incremental treadmill and stationary bicycle protocols, and constant work rate treadmill or cycle studies. Effort dependent outcome measures in such studies may be influenced by the motivation of participants. Of the nine studies that employed timed walking tests, all but one reported significant increases. Although it has been recommended that timed walking tests should be repeated up to three times at baseline to achieve reproducible results, none of these eight studies followed this procedure. In the five studies that used incremental treadmill protocols only one reported an increased peakV˙o ₂, and in the six studies in which incremental cycle ergometer protocols were used again only one reported an increase in peak V˙o ₂. Seven studies reported the responses to submaximal treadmill or cycle ergometer exercise. In three of these studies the duration of exercise tolerated was increased. The overall conclusion of these guidelines was that timed walking distances and treadmill walking protocols are dependent on effort, motivation, and strategy. Improvement in these measures can be due to the psychological effects of a training programme and cannot be taken as clear evidence of a physiological training effect.

In this issue of Thorax Revillet al 19 introduce the endurance shuttle walk test (ESWT) which is a standardised, constant work load, field walking test. It is derived from and complements the incremental shuttle walk test (ISWT).20 It is reasonable to view these tests as the field equivalents of the symptom limited laboratory exercise test and a constant workload laboratory examination. The ISWT measures maximal capacity (potential) while the ESWT examines the ability to use that capacity. The rationale for development is that changes in maximal capacity are unlikely to be demonstrated by therapeutic interventions with chronic lung disease. However, changes in submaximal utilisation of maximal capacity are more likely to be seen but currently cannot be assessed in a standardised and reproducible way. The work rate of the ESWT is set at 85% of maximal capacity which is obtained from a prior ISWT. This workload is apparently high, but in keeping with the recognised ability of patients with COPD to achieve high relative workloads. The chosen workload is also a compromise between intensity and the duration of the test which is open ended. This aspect may be a potential disadvantage since an endurance test may go on for too long unless the intensity is sufficiently high. The need for a prior ISWT is another disadvantage but may obviate the requirement for a second practice ESWT. According to their proposers, both types of shuttle walk tests are simple in concept and easy to conduct in practice; the subjects usually have no difficulty in understanding what is required and the end point is quite clear. Like other walking tests, the ESWT may be particularly suited to patients with chronic lung or cardiac disease in whom disability is evident. It will not be appropriate for normal subjects since the required walking speeds will be too high. The value of the ESWT lies in the sensitivity with which it can detect the effect of rehabilitation. The effect size is many times greater than a symptom limited test of capacity and is more likely to reflect real life improvements in activity. The ESWT also has the potential to form a standardised but sensitive instrument to examine the effect of other interventions on functional capacity such as drugs, oxygen, and pressure support.

However, standardised maximal and submaximal exercise tests and related endurance tests (treadmill, cycloergometer, measurement of exhaled gas, etc) already exist and it is hard to prefer these field tests when an exercise laboratory is available. The use of this proposed test involves the use of cassettes and appears to be more complex than previously used field tests. Revill et al 19 imposed a 20 minute time limit for the faster endurance speed; with this limitation 40% of the patients reached the 20 minute limit of the 75% ESWT. Following a rehabilitation programme the mean time achieved during the ESWT was 11.2 minutes, seven of the 21 patients achieving >17 minutes. This is far longer than the duration of a 6MWD. An analysis of time spent on the test and a cost/benefit ratio in comparison with standardised endurance tests and other field tests should therefore be undertaken in future research.

Opinion appears to be divided between those who are interested in scientific research in pulmonary rehabilitation and those who have the practical responsibility for provision of the service. Some practitioners claim that there is a need to make rehabilitation as simple and as practical as possible in order for it to develop widespread acceptance. Their concern is that it will become a barrier to widespread uptake if the measures proposed for evaluation are too demanding technically or administratively.

Measurements of ventilatory, circulatory, and metabolic adaptations to exercise during “in the field” protocols—such as walking or stair climbing or upper limb exercise—should be encouraged, particularly in severely limited patients. Abnormalities observed “in the field” are likely to reflect more closely symptoms and limitations referred by the patients during day life activities. Laboratory tests such as cycle ergometry, however, remain the gold standard to assess maximal exercise tolerance. By using an incremental exercise protocol, as currently recommended by the ERS Task Force on Clinical Exercise Testing,21 it is possible to have a good indicator of factors that eventually limit exercise tolerance and to identify the work rates that the subject can tolerate easily (moderate) or sustain with difficulty (heavy). The integration of information obtained with laboratory and “in the field” protocols represents the best way to address questions relative to the patient’s ability to sustain specific tasks (short bouts of heavy exercise, endurance exercise, exercise that involves different muscle groups). Walking tests have been proposed for assessing exercise tolerance in severely disabled patients. The information obtainable with the shuttle test may be considered of greater value than other unpaced walking tests, particularly if ventilatory and metabolic measurements are obtained.22