• asthma
• breathing exercises
• complementary medicine

There is considerable lay and professional interest in non-pharmacological treatments for asthma, with reports that up to one third of people with asthma resort to complementary and alternative medicines (CAM).¹ The evidence-based review undertaken for the British guidelines on the management of asthma² found the current evidence for the presence or absence of efficacy of many CAM interventions to be inadequate, and further controlled studies are encouraged. Breathing exercises and yoga have been widely used to treat asthma in Eastern and Western societies for many years, and generally centre on manipulating the respiratory pattern to reduce respiratory frequency and hyperventilation. The Butekyo breathing technique, based on the barely tenable scientific premise that asthma is caused by hyperventilation, makes sweeping claims for effectiveness in asthma.³ In spite of anecdotal reports of benefit given wide coverage in the lay press, the limited scientific scrutiny currently afforded to this technique has indicated more modest improvements in asthma outcomes, with two small controlled studies showing some benefits in symptoms and bronchodilator use although little effect on other measures of asthma severity. A Cochrane review of breathing exercises for asthma⁴ (updated in 2000 and currently undergoing revision) found it was not possible to draw reliable conclusions on the effectiveness of breathing retraining from current published evidence. Since this review, there have been reports in this journal of limited beneficial effects in symptoms and airways hyperresponsiveness to methacholine resulting from yoga breathing exercises,⁵ and of improvements in asthma related quality of life resulting from a community physiotherapy based breathing retraining programme in a subgroup with symptoms suggestive of dysfunctional breathing.⁶ It is still, however, far from clear whether or not breathing exercises can improve asthma outcomes, in which groups they may be effective, or what the mechanism of effect may be. In this issue of Thorax Cooper et al⁷ report a further controlled trial investigating the effectiveness of Butekyo and a device mimicking pranayama yoga exercises, and conclude that the Butekyo method can improve symptoms and reduce bronchodilator use but not by affecting lung function or bronchial hyperresponsiveness.

There is clearly a need for controlled studies on representative patient groups in this area. This study recruited symptomatic adult patients treated with inhaled corticosteroids who had reversible airflow obstruction and hyperresponsiveness to inhaled methacholine. The patients were recruited from an asthma volunteer database and by advertisement, so it is not certain that the study group is necessarily representative of the wider asthma population. The investigators have made efforts to address the real methodological problems that exist in controlling and blinding multifaceted CAM treatments. The study attempted to control for non-specific intervention effects independent of breathing pattern alterations by comparing a variety of asthma outcomes in the Butekyo group with those in the groups using an active and inactive “placebo” pranayama breathing training device. The Butekyo group did, however, receive a higher level of professional contact than the other groups, so it is possible that the effects of professional attention may act as a confounding influence on the improved outcomes reported. Statistically significant improvements in symptom scores were reported for the Butekyo group, although the magnitude and clinical relevance of the improvement was less clear with the unvalidated scoring tool used. Reductions in bronchodilator use were seen in the Butekyo group, although since the Butekyo training process strongly discourages patients from using bronchodilators, the reduction may represent a learning effect and may not be appropriate as a surrogate marker of asthma control in this situation. No differences were seen in other objective outcome measures—including bronchial hyperresponsiveness, lung function, and asthma exacerbations—nor were there any differences in the ability to reduce inhaled corticosteroid dosage, although the study may not have been powered adequately to show this.

If subjective benefits are indeed found in relation to Butekyo and other types of breathing retraining, it is necessary to attempt clarification of the mechanisms of improvement. International consensus has defined asthma as an inflammatory condition characterised by airways hyperresponsiveness and variable airflow limitation,⁸ and subjective improvements need to be related to objective measures of asthma severity. Asthma is a complex disease and the relationship between objective physiological measures and the patients’ subjective experience of their condition is far from simple. It has, for instance, been shown that psychosocial and emotional factors influence asthma symptoms and asthma related health status independently of asthma severity,⁹ and that the relationship between airflow obstruction and symptoms is very weak¹⁰ with some patients experiencing high levels of symptoms in spite of normal or near normal lung function. The question of whether improvements in asthma symptoms associated with breathing retraining may result from indirect effects on emotional or psychological factors cannot be answered from the current evidence. It has been reported that up to 30% of adults with asthma in the community, of all severity levels, may have symptoms suggestive of functional breathing disorders,¹¹ raising the possibility that symptom improvements following breathing retraining interventions may relate to treatment of a co-existent functional problem rather than of asthma per se. This possibility is strengthened by the results of this study, in which subjective symptom improvements are not matched by changes in objective parameters of airways calibre or hyperresponsiveness. Other studies have, however, shown improvements in bronchial hyperresponsiveness in relation to breathing exercises.^5,12 No currently published studies have investigated the effects of breathing retraining on parameters of airway inflammation. Preliminary evidence from an animal model suggests that repeated dry air hyperventilation can result in airways inflammation and hyperreactivity,¹³ so raising a possible link between abnormal breathing and asthma. It is also unclear whether different types of breathing retraining interventions—such as physiotherapist based programmes for treating hyperventilation, different schools of yoga, and the Butekyo method—have similar effects and act by similar mechanisms.

In spite of the various highly effective inhaled and oral pharmacological options available for the treatment of asthma, the goals of asthma management are not currently being met, with high levels of potentially avoidable morbidity revealed in surveys.¹⁴ For a variety of incompletely understood reasons, many of our patients are unable or unwilling to comply with our recommended treatment, and many wish to explore non-pharmacological treatment avenues such as breathing exercises. In contrast to the wealth of high quality evidence informing pharmacological decision making in asthma, often driven by the pharmaceutical industry, there is a paucity of information for rational decision making in non-pharmacological treatments. Although progress is being made in improving this evidence base, large holes still exist in our knowledge and understanding and further controlled studies are required to confirm effectiveness and clarify mechanisms of benefit if found.