The short and long term effects of exercise training in non-cystic fibrosis bronchiectasis – a randomised controlled trial

Bronchiectasis unrelated to cystic fibrosis (non-CF bronchiectasis) is a chronic respiratory condition characterised by bronchial dilatation secondary to airway inflammation, infection and dysfunction of mucociliary clearance [1]. Common features include persistent cough with chronic sputum production, dyspnoea and fatigue [2, 3]. This clinical profile is associated with increased anxiety and depression, reduced health-related quality of life (HRQOL) and impaired exercise tolerance [3, 4]. Patients with non-CF bronchiectasis frequently experience acute exacerbations [5], which are an independent predictor of progressive decline in respiratory function and a poorer prognosis [6, 7]. While the global prevalence of non-CF bronchiectasis is unknown, it is associated with rising hospitalisation rates amongst the older population [8‐10]. This growing healthcare utilisation and economic burden [11, 12] emphasises the need for effective medical and physiotherapy treatment approaches.

Current guidelines for the treatment of non-CF bronchiectasis recommend pulmonary rehabilitation, with the aim of improving exercise tolerance and HRQOL [13]. Pulmonary rehabilitation often incorporates self-management strategies to promote treatment adherence and this approach has been advocated for bronchiectasis [14]. There is limited evidence of the effects of pulmonary rehabilitation in non-CF bronchiectasis, with two retrospective studies demonstrating similar benefits in exercise capacity and HRQOL as those observed in patients with chronic obstructive pulmonary disease (COPD) [15, 16]. Two prospective studies showed short-term improvements in exercise capacity and HRQOL with a combination of endurance, strength training and inspiratory muscle training [17] or regular airway clearance therapy (ACT) [18]. These studies incorporated formal education as part of the clinical approach and had a follow up period of three months; it is not clear if benefits persist in the longer term.

Health benefits are also assessed through quality-adjusted life years (QALYs), which quantify the benefit gained from an intervention by measuring the change in HRQOL over time. In COPD, pulmonary rehabilitation (exercise training and education) has been associated with fewer acute exacerbations and improved QALYs [19]. Whether similar effects are achieved in bronchiectasis is unknown.

Therefore, the aims of this study were to 1) determine the short and long-term effects of exercise training and review of ACT on exercise capacity and dimensions of HRQOL and 2) evaluate the effect of this intervention on the incidence of acute exacerbations over a 12-month follow up period.

Between April 2009 and February 2011, a total of 144 patients were screened (Figure 1), 36 were excluded prior to randomisation for not meeting the inclusion criteria and a further 23 declined to participate. The subject flow reflected the recommendations from the Consolidated Standards of Reporting Trials (CONSORT) statement [38]. A total of 42 were randomised to the exercise training group and 43 to the control group. All participants had been previously prescribed ACT. For those in the intervention group, no changes to ACT regimens were required by any participant over the eight week duration and the average VAS sputum volume rating did not change over the intervention period (p = 0.56). Adherence to exercise diary completion during the 8 week program in the exercise group was 29%. All control group participants reported no change to their usual exercise patterns between baseline assessment and at 9 weeks follow up.

One participant in the exercise group withdrew from the study during the intervention phase due to an illness unrelated to bronchiectasis. Data for this participant were not available at nine weeks, six or 12 months. There was one death in the control group from respiratory failure. A total of 38 (90%) participants completed the exercise training, with reasons for non-completion outlined in Figure 1. Reasons for not undertaking outcome measures at the nine week assessment were respiratory illness (n = 3), fractured thoracic spine which was unrelated to exercise training (n = 1) or declined to attend (n = 4). Baseline characteristics of the 85 participants are shown in Table 1, with no difference between groups at baseline in any measures. According to the MMRC Dyspnoea grade, the extent of functional impairment in exercise capacity was mild with 64 (75%) participants reporting dyspnoea only on strenuous activity (MMRC Grade 1). There was no change in FEV₁ (p = 0.26) or FVC% predicted (p = 0.43) over the study duration in either group.

The effect of exercise training on maximal exercise capacity is shown in Figure 2. There was a significant interaction between group and time for ISWD (p = 0.005). The ISWD improved in the exercise group following training, with a mean difference (95% CI) of 62 m (24 to 101 m) compared to the control group. However, this improvement was not sustained at either six or 12 months. For the CRDQ, the exercise group demonstrated a significant reduction in dyspnoea (p = 0.009) and fatigue (p =0.01) compared to the control group immediately following training (Figure 3), but these improvements were not sustained when assessed at six and 12 months. There was no significant change in emotional function or mastery.

Of the secondary outcomes, there was a significant improvement in functional exercise tolerance following training, with a mean change (95% CI) in 6MWD of 41 m (19 to 63 m) compared to the control group, but this improvement was not sustained over the longer term. There were no significant differences between groups in any of the LCQ domains or in anxiety or depression immediately following intervention or at follow up (Table 2).

The adherence rate to symptom diary completion was 82% in the exercise group and 77% for the control group. A total of 12 participants in the exercise group and 18 in the control group experienced at least one exacerbation over the 12 months follow up. There was a longer median time to first exacerbation in the exercise group of 8 months (95% CI 7 to 9 months), compared to control group of 6 months (5 to 7 months), with log rank of 0.49 (95% CI 0.01 to 0.97), p = 0.047 (Figure 4). Exercise training reduced the number of exacerbations over 12 months (Table 3), with a trend towards fewer exacerbations requiring antibiotics (p = 0.061). There was no significant difference between groups in the number of days or exacerbation days requiring antibiotics for each exacerbation. The relative risk of exacerbation was 0.69 (95% CI 0.49 to 0.98), indicating there was less risk of exacerbation in the exercise group compared to the control group.

The health utility scores of the AQOL before intervention (baseline) and 12 months following completion of intervention showed a mean (SD) gain in QALYs from 0.74 (0.22) to 0.88 (0.17) in the exercise group and from 0.74 (0.17) to 0.89 (0.11) in the control group. While there was a significant gain in QALYs over 12 months within both groups (both p < 0.001), the mean difference of 0.004 (95% CI −0.08 to 0.09) between groups was not significant.

To date, this is the only trial of exercise training and regular review of ACT in non-CF bronchiectasis which incorporated 12 months follow up. We have demonstrated short term gains in exercise capacity as well as improvement in symptoms of dyspnoea and fatigue, but these benefits were not sustained beyond program completion. With exercise training, the time to first exacerbation was longer and the number of exacerbations over a 12 month period was reduced. There was no benefit in cough related QOL, change in levels of anxiety or depression or greater improvement in QALYs.

This is the first study to show a reduction in the number of acute exacerbations over the 12 months following exercise training and a longer time to first exacerbation. While the reduced exacerbation frequency in the exercise training group should be interpreted with caution, due to the proportion of participants who were unable to complete 12 months follow up, these findings are encouraging and are similar to case series reports [39]. The mechanism by which exercise training reduced exacerbation frequency is unclear. It could be argued that the review of ACT as necessary in the intervention group may account for the difference, but the familiarity of all participants with ACT prior to study enrolment, the lack of change in ACT regimen for the intervention group and the absence of reinforcement over the follow up suggests this is unlikely. Improved immune function has been demonstrated with moderate intensity exercise in healthy elderly individuals [40]; a similar mechanism may be present in this study. In CF, the increase in expiratory flow and promotion of annular airflow during a single session of exercise has been associated with improved mucociliary clearance, although the long term effects are unknown [41]. We did not collect information related to bacterial colonisation, or other markers in inflammation in this study. Given the significant influence of distal airway colonisation upon exacerbation severity and decline in lung function [7, 42, 43], further examination of the physiological effects of exercise training is needed.

The mean improvement of 64 m in the ISWD and 41 m in the 6WMD is consistent with retrospective findings in this patient population [15, 16] as well as those from a comprehensive pulmonary rehabilitation program complemented by ACT [18]. Although the magnitude of increase in ISWD was less in comparison to pulmonary rehabilitation with inspiratory muscle training [17], it suggests that exercise capacity improvement in bronchiectasis can be achieved using training principles similar to those applied in COPD [14, 26]. While the level of functional impairment prior to intervention was mild compared to some studies [16, 18], the MMRC grades and baseline walking tests were equivalent to other reports of patients included in pulmonary rehabilitation programs [17]. With this magnitude of improvement in patients with mild disease severity and functional limitation, it may be that those with more severe disease and impairment could achieve even greater benefit from this intervention, but this requires further investigation. Our results suggest that the benefits appear to be independent of a comprehensive education program, which has been part of previous work [15, 17, 18]. While the extent of improvement in 6MWD is similar to the described minimal important difference in COPD [44], this has not yet been defined for bronchiectasis. Therefore the clinical significance of the improvements in exercise capacity requires confirmation.

While exercise training was not associated with a greater improvement in QALYs, in contrast to previous reports in COPD [19], the study was not powered for this outcome and the follow up time was short. Health utility is only one dimension of the cost effectiveness and cost utility of this treatment approach. Further exploration of the economic benefit of this intervention, including measures of cost per QALY [19, 35] in bronchiectasis is warranted.

The short-term duration of benefit following exercise training is a common finding in chronic respiratory conditions [14]. Despite monthly telephone follow up and instructions to complete an exercise diary beyond the intervention period, the adherence rate was poor and the absence of a structured exercise program may have impeded the ability to maintain initial improvements. While these results contrast with a retrospective report of pulmonary rehabilitation, which demonstrated that improvement was maintained at 12 months, these effects may have been overestimated, with a significant attrition rate (>50%) in this previous study [15]. The functional decline over time reflects a similar pattern to that observed in COPD and ILD [14, 28], despite reports of a slower rate of decline in lung function in bronchiectasis compared to these conditions [45‐48]. We found that the 6MWD at 12 months was worse than baseline, suggesting that despite the mild disease evident in this sample, exercise capacity has the potential to decline further over time. Reports in COPD have demonstrated that immediately following pulmonary rehabilitation, an increase in physical activity is evident, but the magnitude is small, with suggestions that a longer duration of treatment is necessary to encourage behavioural change [49‐51]. Our results suggest that sustained gains in symptoms and functional capacity may require longer interventions or maintenance programs for people with bronchiectasis.

This is the first prospective study to isolate improvement in dypsnoea and fatigue in people with non-CF bronchiectasis, both key symptoms which contribute to HRQOL [2, 3]. The lack of change in psychological symptoms may be due to the few participants with clinically significant anxiety or depression at baseline. Similarly, the well preserved emotional function and mastery at study commencement in both groups, a contrast to COPD [14], may account for the lack of effect on these measures following exercise training. Finally, all participants had long-standing bronchiectasis and the possibility of a greater tolerance of cough-related symptoms, as reflected by the high baseline scores for the LCQ. These clinical factors combined with similar initial instruction in ACT and the consistent regimen during the study in both groups may account for the negligible effect of exercise training on cough-related QOL.

In conclusion, this study demonstrates that a supervised exercise program of eight weeks is associated with short-term improvement in exercise capacity, dyspnoea and fatigue. Exercise training is associated with a reduction in the frequency of acute exacerbations of bronchiectasis over 12 months, however it appears to have minimal effect on cough related QOL. The significance of these findings for long-term prognosis requires further exploration.