Huff and puff of exercise for airway clearance in cystic fibrosis: how clear is the evidence?

Abstract

Cystic fibrosis (CF) lung disease is characterised by reduced mucus clearance, resulting in retention of thick mucus, creating a nidus for chronic infection and progressive lung damage. A cornerstone of the medical management of CF lung disease is daily airway clearance therapy (ACT), tailored to the individual, such as active cycle of breathing technique (ACBT), autogenic drainage, positive expiratory pressure (PEP) and oscillating PEP devices, postural drainage, percussion and highfrequency chest wall oscillations. There is widespread agreement among people with CF and healthcare professionals that exercise may also improve airway clearance and many people with CF already use exercise as a substitute for ACT. Whether exercise is a satisfactory substitute for traditional ACTs in people with CF has been identified as a top 10 research priority, yet none of the many systematic reviews on the various ACTs have answered this question, as in the five systematic reviews only one study was eligible for inclusion, and the systematic review on exercise training did not specifically address airway clearance. People with CF are encouraged to regularly exercise, due to the known benefits on exercise capacity and potential improvements in lung function and quality of life. If exercise could adequately replace traditional ACTs for people with CF, it may also reduce treatment burden, which was another of the top 10 research priorities. In this issue of Thorax, Ward et al reported their systematic review addressing the question ‘is exercise a satisfactory substitute for traditional ACTs in people with CF?’, which examines the effects of exercise compared with no intervention and exercise compared with traditional ACTs. This was a welldesigned and rigorously conducted review, prospectively registered on PROSPERO, followed the Preferred Reporting Items for Systematic Reviews and MetaAnalyses guidelines and reported the quality of the evidence according to Grading of Recommendations Assessment, Development and Evaluation ratings for the primary outcomes. The review included 12 studies with 187 participants of wideranging demographics and disease severity (7–48 years and 18%–113% of predicted FEV1). Interventions included any form of physical exercise (eg, walking, cycling) and traditional ACTs (eg, ACBT, PEP). All studies were short duration (ranging from a single session up to 2 weeks) with small sample sizes (n≤32) and 11 of the 12 studies were randomised crossover trials. The outcome measures chosen reflected both physiological measures of respiratory function and airway clearance, as well as exacerbation frequency, healthrelated quality of life and participant preference. Only one study included more than a single session of each intervention, so there were no data for longer term outcomes such as exacerbation frequency and quality of life. Also, due to significant heterogeneity of the interventions (both exercise and traditional ACTs), metaanalyses were limited. The authors did, however, succinctly combine subtypes of exercise into single forest plots for some outcomes, which allows the reader to visualise findings for exercise as a whole. There was lowquality evidence that exercise did not improve spirometry lung function (FEV1 and FVC) compared with no intervention and moderate quality evidence that exercise did improve peak expiratory flow, which could potentially improve cephalad movement of mucus. Trends or significant improvements were reported with exercise compared with no intervention for mucus clearance, expectorated sputum weight and ease of expectoration. When comparing exercise to traditional ACTs, however, the results were not consistent between studies. There was low to verylowquality evidence for the effect of exercise compared with traditional ACTs on spirometry lung function (FEV1, FVC and FEF25–75), with most studies reporting no significant difference between the interventions. A pattern emerged when looking at airway clearance—studies that included huffing during both exercise and traditional ACTs had no significant differences in mucus clearance or expectorated sputum weight, whereas the studies that had significantly less mucus clearance or expectorated sputum weight with exercise compared with traditional ACTs included huffing with traditional ACT but not during exercise. The main limitation of this systematic review was the paucity of evidence—a low number of shortduration trials with very small sample sizes and significant variances in study design and interventions that prevented pooling of results for metaanalyses. It is likely some of the studies were underpowered, so the review cannot exclude small differences between exercise and traditional ACTs that may not have been detected. The clinical implications from this systematic review are to recommend including huff with exercise if the intention is to improve mucus clearance and that it is possible a single session of exercise (with huff) may have similar effects as traditional ACTs. As none of the included studies, however, went for more than 2 weeks, there is no evidence to answer the question whether exercise can act as a substitute for traditional ACTs in the medium or long term. Going forward, in order to answer this highly relevant clinical question with meaningful outcomes (eg, exacerbation frequency and quality of life), it is necessary to conduct a longterm randomised controlled noninferiority trial (of at least 12 months’ duration) comparing exercise to traditional ACTs. Considering the wellestablished benefits of exercise in the management of people with CF, it would not be ethical nor wise to omit exercise from routine clinical care. Instead, it would be advisable to have one arm of the parallel trial to withdraw traditional ACT (being replaced by exercise plus huff) and the other arm to maintain routine management (ie, ACT and usual exercise). This idea appears to have support from both people with CF and healthcare professionals. A recent survey reported 48% of patients with CF skip their ACT if they have exercised and 73% of healthcare professionals would support a trial of this design. It would also be important to accurately objectively measure the adherence, intensity and quality of both the exercise and ACTs, such as with portable activity monitors and inline devices that measure pressure and flow. Finally, in the new era of CF medical management, it would be important to stratify for genotype, CFTR Faculty of Medicine and Health, University of Sydney, Sydney, New South Wales, Australia Respiratory Medicine, Royal Prince Alfred Hospital, Sydney, New South Wales, Australia