Published ahead of print on January 26, 2006, doi:10.1164/rccm.200506-925OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200506-925OC
The Effect of Helium and Oxygen on Exercise Performance in Chronic Obstructive Pulmonary DiseaseA Randomized Crossover TrialRespiratory Medicine, Royal Hallamshire Hospital; Department of General Practice and Primary Care, ScHARR; Biomedical Science, University of Sheffield, Sheffield; Clinical Sciences, University Hospital Aintree, Liverpool; St. George's, University of London, London; and BOC Ltd., Guildford, United Kingdom Correspondence and requests for reprints should be addressed to E.A. Laude, Ph.D., Biomedical Sciences, University of Sheffield, Addison Bldg, Western Bank, Sheffield, S10 2TN UK. E-mail: e.a.laude{at}sheffield.ac.uk
Rationale: Breathing supplemental oxygen reduces breathlessness during exercise in patients with chronic obstructive pulmonary disease (COPD). Replacing nitrogen with helium reduces expiratory flow resistance and may improve lung emptying. Combining these treatments should be independently effective. Objectives: Study the effect of changing oxygen or helium concentration in inspired gas during exercise in patients with stable COPD. Methods: In 82 patients (mean age, 69.7 yr; mean FEV1, 42.6% predicted), we measured endurance shuttle walking distance, resting and exercise oxygen saturation, and end-exercise dyspnea (Borg scale) while patients breathed Heliox28 (72% He/28% O2), Heliox21 (79% He/21% O2), Oxygen28 (72% N2/28% O2), or medical air (79% N2/21% O2). Gases were administered using a randomized, blinded, crossover design via a face mask and an inspiratory demand valve. Results: Breathing Heliox28 increased walking distance (mean ± SD, 147 ± 150 m) and reduced Borg score (–1.28 ± 1.30) more than any other gas mixture. Heliox21 significantly increased walking distance (99 ± 101 m) and reduced dyspnea (Borg score, –0.76 ± 0.77) compared with medical air. These changes were similar to those breathing Oxygen28. The effects of helium and oxygen in Heliox28 were independent. The increase in walking distance while breathing Heliox28 was inversely related to baseline FEV1 breathing air. Conclusion: Reducing inspired gas density can improve exercise performance in COPD as much as increasing inspired oxygen. These effects can be combined as Heliox28 and are most evident in patients with more severe airflow obstruction.
Key Words: chronic obstructive pulmonary disease • exercise capacity • helium • oxygen Chronic obstructive pulmonary disease (COPD) is associated with impaired exercise capacity, which contributes significantly to a reduced quality of life in these patients (1). Several physiologic mechanisms limit exercise performance in COPD, but abnormal lung mechanics predominate. Unlike in healthy subjects, end-expiratory lung volume increases during exercise in patients with COPD (2). This relates to the intensity of self-reported breathlessness during exercise and is thought to result from expiratory flow limitation during tidal breathing (3). In general, treatments improve lung emptying or decrease ventilatory requirement during exercise. Bronchodilator drugs reduce dynamic hyperinflation (4) and surgical lung volume reduction decreases static lung volumes (5). Both increase exercise capacity. Breathing supplementary oxygen during exercise significantly increases self-paced walking distance (6) and endurance time during cycle ergometer endurance exercise by reducing ventilatory demand (7–9). Similar changes occur during pulmonary rehabilitation where oxygen breathing can increase the ability to undergo training (10). An alternative therapeutic approach would be to change the physical characteristics of the inspired gas by replacing nitrogen with the lower density gas helium. This should reduce airway resistance by decreasing turbulent flow (11) and improve respiratory gas exchange (12). This approach has been used with some benefit in the intensive care unit (13, 14), but the use of helium/oxygen gas mixtures (heliox) to increase exercise capacity has produced conflicting results in the small numbers of patients with COPD of varying severity studied (14–18). We hypothesized that replacing nitrogen with helium would increase exercise capacity and reduce exercise-induced breathlessness in patients with stable COPD by a mechanism different to that operating when breathing supplementary oxygen. Hence, the effects of combining the two treatments would be independent of each other. Because expiratory flow limitation and the resulting turbulent air flow at high levels of ventilation will be most marked in those with severe COPD, we also hypothesized that the effects of changing the gas density would be greatest in the most severely obstructed patients. To test these hypotheses, we conducted a randomized, crossover, factorial trial to allow us to identify the independent contribution of each gas to dyspnea and exercise capacity improvement and have sufficient power to carry out subgroup analyses based on disease severity. Data from this study have been previously published in abstract form (19, 20).
Patients We studied patients with a diagnosis of COPD (21) confirmed by an FEV1/FVC ratio less than 0.7, an FEV1 less than 80% predicted, and limited bronchodilator reversibility. All patients complained of exertional dyspnea, defined by a Borg score (22) after exercise of 3 or more, and had no history of recent exacerbation. All patients were ex-smokers who continued their usual medication throughout the study. The protocol was approved by the local research ethics committees, and patients gave written, informed consent.
Measurements Dyspnea at rest and on exercise was rated using the modified Borg category scale (22) and a 100-mm visual analog scale (VAS). SaO2 and heart rate were measured continuously with a pulse oximeter (Minolta PulseOx3i; DeVilbiss, Wollaston, UK). Values before and 5 min after breathing the test gas mixture together with the lowest SaO2 and maximum heart rate during exercise were recorded. Tympanic temperature (IVAC Corporation, San Diego, CA) was measured for 5 min at rest, before and after breathing the gas, and at end exercise.
Test Gases
Protocol
At Visit 1, the patients practiced the incremental shuttle walking test breathing room air and rated their dyspnea with the Medical Research Council breathlessness scale (26). At Visit 2, the incremental shuttle walking test was repeated, breathing medical air, to identify the subsequent ESW speed. Patients also performed a practice ESW breathing medical air. At Visits 3 and 4, patients were randomly assigned to receive two of the four gas mixtures. Two ESW tests were performed 40 min apart at each visit, one with each test gas mixture.
Statistical Analysis The study sample size was established from previous studies using breathlessness as an outcome (27–29), 24 patients per center being required for full randomization.
Patients The progress of patients through the study is summarized in Figure 1, and the characteristics of the 82 patients randomized to receive the test gases are shown in Table 1. There was no order or visit effect with the exception of SaO2 that showed a small within-visit variation.
There was no significant change in FEV1, baseline Borg, or VAS scores measured while breathing room air before the walking test at each visit. Walking distance, breathing medical air, was reproducible between tests (r = 0.81, p < 0.001). The mean value for ESW at Visit 2 was 258 m (95% CI, 207–310 m); at postrandomization, it was 257 m (95% CI, 201–314 m). Breathlessness at end-exercise breathing medical air was also reproducible: the mean Borg score at Visit 2 was 4.75 (95% CI, 4.06–5.44); at postrandomization, it was 4.66 (95% CI, 4.42–4.91). No adverse effects were reported from breathing heliox gas mixtures. Tympanic temperature did not change.
Distance Walked
Exercise capacity (expressed as log distance walked) breathing medical air correlated with baseline FEV1 (Pearson correlation, r = 0.36), but when breathing Heliox28, this relationship was lost (r = 0.04). Oxygen28 and Heliox21 showed an intermediate picture (r = 0.32 and 0.25, respectively). The percentage increase in exercise capacity relative to the distance walked while breathing medical air is shown in Figure 2 as a box plot. When breathing Oxygen28 and Heliox21, the improvement relative to medical air was normally distributed, but there was a skewed distribution of improvement in walking distance when breathing Heliox28 (Figure 2). Some patients had a very large percentage of improvement relative to breathing medical air when breathing Heliox28. This observation was compatible with our hypothesis that patients with more severe obstruction may show the greatest benefits.
To explore this, we used analysis of covariance to model the relationship between baseline FEV1 and percentage change in distance walked for all gases relative to that individual's walking distance when breathing medical air. The calculated regression coefficients were used to express the results of this analysis graphically (Figure 3). The size of improvement with Heliox21 and Heliox28 was negatively correlated with baseline FEV1 (p < 0.001), indicating that patients with more severe airways obstruction showed the proportionately greatest increase in exercise capacity when breathing a gas mixture based on helium rather than nitrogen. Oxygen28 did not have this effect, a similar improvement in walking distance occurred irrespective of baseline spirometry.
Symptoms Dyspnea. Patients' dyspnea ratings at end-exercise using the modified Borg scale were significantly different between all four gases. Breathing Heliox28 had the lowest score and medical air the highest (Table 2). The VAS scores gave similar results although the scores for Heliox21 and Oxygen28 did not differ significantly (Table 2).
Limb fatigue.
Oxygen Saturation The minimum oxygen saturation during endurance walking differed significantly between gas mixtures (Table 2). When comparing the degree of exercise-induced desaturation, values for medical air (8.8%; 95% CI, 7.6–10.0%) and Heliox21 (7.7%; 95% CI, 6.7–8.8%) were similar, but there was significantly less desaturation breathing either Heliox28 and Oxygen28 (5.2%; 95% CI, 4.2–6.1%) and (6.2%; 95% CI, 5.1–7.3%), respectively.
Heart Rate
This is the first randomized controlled trial to compare the effect of increasing the inspired oxygen concentration with reducing inspired gas density on the exercise performance of patients with stable COPD. In addition, this is the first study to test whether combining these treatments produces an equivalent or an additive effect on exercise duration and dyspnea intensity. Previous investigators have focused on identifying the underlying physiologic mechanisms that explain improved exercise performance while breathing increased oxygen concentrations (8, 9). Those observations have been confirmed in a larger number of patients using field exercise testing (30). The situation with heliox breathing in COPD has been more complicated. A number of small trials in patients with either mild (18) or very severe (17) COPD have used heliox either to explore the influence of expiratory flow limitation on exercise or as a method for unloading the inspiratory muscles (31) rather than primarily to examine its effect on exercise performance. More recently, in a single-blind randomized trial, Heliox21 was found to increase cycle endurance substantially (16). Our data in a randomized double-blind trial confirm the beneficial effect of exercising with supplementary oxygen and helium gas mixtures alone and in combination. These effects were additive in nature, but were influenced differently by the initial severity of airflow obstruction. Earlier studies identified two rather different mechanisms by which oxygen and heliox might work, the former reducing lactate production and diminishing ventilatory drive (8, 9), whereas the latter increases maximum expiratory flow by reducing the pressure required to overcome frictional resistance and the degree of turbulence at high flow rates (16). Higher levels of minute ventilations during exercise have been reported while breathing heliox in both normal subjects and patients with COPD (18, 32). Our data confirm that patients exercising at a constant pace can walk further with a lesser degree of breathlessness when breathing 28% oxygen or a 21% oxygen/79% helium mixture, which reduces the gas density by 0.82 kg/m3 (density: Oxygen28, 1.24 kg/m3; Heliox21, 0.42 kg/m3). The improvements in walking distance and dyspnea were very similar irrespective of the gas mixture used. When a modest increase of inspired oxygen was combined with a similar reduction in gas density (density Heliox28, 0.50 kg/m3), a further improvement in walking distance and reduction in dyspnea occurred. Distance walked was reproducible and the maximum heart rate achieved did not differ between tests, suggesting a comparable degree of cardiac stress on each occasion. However, we noted more between-subject variation in walking distance in those whose exercise performance was better preserved. We overcame this by reporting the data as log walking distance and the geometric mean distances walked derived from this analysis were very different with the different gas mixtures. Raising the inspired oxygen and Heliox21 both increased walking distance significantly. Combining the two inspired gas changes (Heliox28) increased the walking distance further and statistical analysis showed that their effects were independent. This is in keeping with their acting by independent mechanisms as suggested in previous mechanistic studies. Both gas mixtures have been shown to reduce end-expiratory lung volume during exercise, but the time course of the change in lung volume appears to be different (16). Whether changes in this measurement explain the additive effect of combining the two gas modifications should be established in future studies. Breathing Heliox21 did not change resting oxygen saturation, which rose as expected when the inspired oxygen concentration increased. This change mitigated the degree of exercise-induced desaturation, but did not abolish it. Despite the increase in walking distance, desaturation breathing Heliox21 was significantly less than during medical air breathing, but this was still worse than with Oxygen28, which itself was worse than the desaturation breathing Heliox28. This improvement in exercise-related gas exchange may reflect better lung mechanics during exercise or improved oxygen diffusion in the presence of helium (12). In either case, the consequent reduction in ventilatory drive would decrease the degree of dyspnea experienced for any given distance covered relative to the medical air breathing test. We observed significant heterogeneity in the response to treatment, most evident when breathing Heliox28. Because Heliox may lessen the effect of expiratory flow limitation during exercise (32) and because expiratory flow limitation occurs more frequently in patients with a reduced FEV1 (33), we tested whether the relative improvement in exercise capacity was related to the baseline FEV1 recorded breathing room air. In our patients, there was only a modest relationship between medical air breathing exercise distance and baseline spirometry. When breathing heliox gas mixtures, particularly Heliox28, the correlation between walking distance and spirometry was abolished, mainly because of a greater degree of improvement in the exercise performance of patients with the worst lung function. When tested in an analysis of covariance model, we found significant differences between the responses to the different gas mixtures that were dependent on initial lung function. There was evidence of a greater proportionate benefit from heliox mixtures in patients with the worst spirometry. Breathing heliox does not appear to affect the degree of resting expiratory flow limitation (14), but its effect on lung emptying and increasing maximum exercise ventilation may be particularly important in those patients with the worse lung function who exhibit the greatest degree of ventilatory limitation on exercise. In contrast, breathing Oxygen28 produced a similar percentage improvement in performance irrespective of the initial degree of spirometric impairment, in keeping with its known effects on ventilation and lactate production. We adopted a different approach to other studies by using a standardized endurance exercise test that reflects the conditions during exercise outside the laboratory, but still shows a good relationship with endurance exercise testing measured on the treadmill (23). Although this simpler approach restricted the data we could collect, it allowed us to study a larger number of subjects, with a wider range of baseline lung function, on repeated occasions than has been reported previously. This approach also avoided the need for recalibrating equipment between gases and hence unblinding of the investigators. Although shuttle walks are reported to have little "learning effect," we performed two practice tests before the study and used a randomized protocol to minimize any impact of such effects on the comparisons between gases. The analyses showed no sequence or order effects. We did not conduct reproducibility testing to confirm individual improvements with the heliox mixtures, but relied on the randomized blinded design to identify significant changes in group behavior. Expressing our data as walking time rather than distance might have overcome some of the intrinsic variation based on baseline performance, but this did not prove to be the case. We used two different methods of scaling breathlessness because there is no clear consensus as to which has the best measurement properties (34, 35). Grant and colleagues (36) found a clear visit effect with the VAS that was not evident with the Borg score, which we used as our principal measure of dyspnea. Our patients were symptomatic but stable as judged by their dyspnea scores, spirometry, and exercise performance breathing medical air. We found no evidence to support the theoretic concern that the increased thermal conductivity of helium would reduce body temperature. At an individual patient level, the changes in oxygen saturation were generally too small to permit reliable identification of the gases by the investigator. We took care to avoid unblinding the patient and investigator by ensuring that the patient did not speak for at least 2 min after the end of the exercise test. In summary, we believe that there are no experimental factors that may have influenced our findings. Our trial and other studies of supplemental oxygen have all been acute interventions performed under laboratory conditions. Future studies designed to assess the impact of heliox on daily activity will need to address technical issues concerning delivery devices for routine use. In this context, it should be noted that, despite many years of prescription of ambulatory oxygen, there is still no randomized controlled trial evidence for its benefit in daily life. Whether the substantial improvements in exercise performance and dyspnea seen with Heliox28 will translate into more effective use of ambulatory treatment remains to be determined, as will methods to identify individuals who show consistent responses to this treatment. Although the recent American Thoracic Society/European Respiratory Society guidelines have recognized that COPD is a preventable and treatable condition (37), it is still regarded by many as one in which significant improvement is not possible. Our data show that this is not the case. The changes in endurance exercise and the reductions in breathlessness we report while breathing increased inspired oxygen or heliox gas mixtures are substantial, being at least comparable to those achieved with current bronchodilator therapy (38), pulmonary rehabilitation, or even lung volume reduction surgery (39, 40). Recent data suggest that bronchodilator therapy can enhance the effect of pulmonary rehabilitation (41), and future studies should examine whether the same is true if training is undertaken breathing Heliox28, particularly in patients with severe airflow obstruction who may have difficulty training effectively. In selected cases, the availability of heliox therapy might have a dramatic impact on daily exercise performance and health status beyond that possible with ambulatory oxygen alone, provided Heliox28 can be administered in a way patients find acceptable. Our trial has shown that combining different treatment approaches, which modify respiratory physiology, is effective and that reducing inspired gas density by heliox breathing provides a further mechanism that can be exploited therapeutically in COPD.
The authors thank Deirdre Frost, Julie Griffiths, Marion Taylor, Dr. Silia Diamantea, and Dr. Paul Walker for their assistance in collecting patient data; Carrie Seymour for monitoring and collating the data; and Dr. Geoff Lloyd for valuable scientific advice.
This study was supported by BOC Ltd. This article has an online supplement, which is accessible from this issue's table of contents at www.thoracic.org Originally Published in Press as DOI: 10.1164/rccm.200506-925OC on January 26, 2006 Conflict of Interest Statement: E.A.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. N.C.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. C.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. B.D. has been an employee of BOC Ltd. since March 1993. M.J.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.L. was given a grant of £37,500 by BOC to assist with pulmonary rehabilitation. P.W.J. has received fees for speaking at conferences annually from 2000 to 2004 from GlaxoSmithKline (GSK), Boehringer, and Astra Zeneca at less than $5,000 per year per company; he presented as an expert witness for Boehringer in 2002 and currently receives grants from Boehringer and GSK; he has held consultancy contracts with GSK for 6 yr and sat on GSK advisory boards for 7 yr, receiving less than $5,000/yr for each of those activities separately, in each of those years; in 2004, he sat on an advisory board for Novartis and Astra Zeneca—in each case the consultancy fee was less than $5,000. P.M.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form June 15, 2005; accepted in final form January 24, 2006
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
1-antitrypsin deficiency. Am J Respir Crit Care Med 2001;163:936–941.
