Interventions for bronchiectasis: an overview of Cochrane systematic reviews

Emma J Welsh, David J Evans, Stephen J Fowler, Sally Spencer

Bronchiectasis is a chronic respiratory disease characterised by abnormal dilatation of the bronchi, and presents typically with a chronic productive cough (or chronic wet cough in children) and recurrent infective exacerbations. It significantly impacts daily activities and quality of life, and can lead to recurrent hospitalisations, severe lung function impairment, respiratory failure and even death. To provide an overview of the efficacy and safety of interventions for adults and children with bronchiectasis from Cochrane reviews.To identify gaps in the evidence base that will inform recommendations for new research and reviews, and to summarise information on reported outcomes and make recommendations for the reporting of standard outcomes in future trials and reviews. We included Cochrane reviews of non-cystic fibrosis (CF) bronchiectasis. We searched the Cochrane Database of Systematic Reviews. The search is current to 11 February 2015. We also identified trials that were potentially eligible for, but not currently included in, published reviews to make recommendations for new Cochrane reviews. We assessed the quality of included reviews using the AMSTAR criteria. We presented an evidence synthesis of data from reviews alongside an evidence map of clinical trials and guideline data. The primary outcomes were exacerbations, lung function and quality of life. We included 21 reviews but extracted data from, and rated the quality of, only nine reviews that reported results for people with bronchiectasis alone. Of the reviews with no usable data, two reviews included studies with mixed clinical populations where data were not reported separately for people with bronchiectasis and 10 reviews did not contain any trials. Of the 40 studies included across the nine reviews, three (number of participants nine to 34) included children. The studies ranged from single session to year-long studies. Each review included from one to 11 trials and 28 (70%) trials in the overview included 40 or fewer participants. The total number of participants included in reviews ranged from 40 to 1040. The age range of adult participants was from 36 to 73 years and children ranged from six to 16 years. The proportion of male participants ranged from 21% to 72%. Where reported, mean baseline forced expiratory volume in one second (FEV1) ranged from 1.17 L to 1.66 L and from 47% to 88% predicted. Most of the reviews had search dates older than two years.We have summarised the published evidence as outlined in Cochrane reviews, but it was not possible to draw definitive conclusions. There was inconclusive evidence on the use of long-term antibiotics and nebulised hypertonic saline for reducing exacerbation frequency and evidence that human deoxyribonuclease (RhDNase) increases exacerbation frequency. Improvements in lung function were reported for inhaled corticosteroids (ICS) though this was small and not clinically relevant. Evidence of benefit for hyperosmolar agents and mucolytics was inconclusive. There was limited evidence of improvements in quality of life with airway clearance techniques and physical therapy but evidence of benefit for hyperosmolar agents was inconclusive. Secondary outcomes were not clearly reported in all trials in the included reviews. Improvements in dyspnoea, wheeze and cough-free days were reported for small trials of ICS and LABA (long-acting beta2-agonsts)/ICS and cough reduction was also reported for a small bromhexine trial. Reduction in sputum production was reported for long-term antibiotics and airway clearance techniques but evidence of benefit for hyperosmolar agents was inconclusive.Adverse events were included as outcomes in seven reviews. The review of long-term (four weeks to one year) prophylactic courses of antibiotics reported significantly more cases of wheeze (Peto odd ratio (OR) 8.56, 95% confidence intervals (CI) 1.63 to 44.93), dyspnoea (12 versus three, P value = 0.01) and chest pain (seven versus zero, P value = 0.01) from the same trial (74 participants) but no differences in occurrence of diarrhoea, rash or number of withdrawals. In the review of mucolytics versus placebo, relevant outcomes were not reported for erdosteine comparisons and no significant adverse effects were reported for bromhexine, though adverse events were associated with RhDNase (OR 28.19, 95% CI 3.77 to 210.85, 1 study). Of the remaining five reviews, adverse events were not reported in the single trials included in the ICS review or the physical therapy review and the impact of adverse events in the single trial included in the inhaled LABA/ICS combination versus ICS review were unclear. The reviews of short-term courses of antibiotics and inhaled hyperosmolar agents reported no significant differences in occurrence of adverse events. Fewer admissions to hospital were reported for long-term antibiotics, but this outcome was not reported in all reviews. No reviews reported differences in mortality, but again this outcome was not included in all reviews.We did not explicitly include antibiotic resistance as an outcome in the review, but this was unclear in the Cochrane reviews and evidence from other trials should be considered.We rated all reviews as high quality (AMSTAR), though opportunities for improved reporting (e.g. summary of findings and GRADE evaluation of the evidence) were identified for inclusion in future updates of the reviews. However, the majority of trials were not high quality and confidence in the effects of treatments, therefore, requires additional evidence from larger and more methodologically robust trials. We evaluated the overall coverage of important topics in bronchiectasis by mapping the quality of the current evidence base against published guidelines and identifying high priority areas for new research on; use of short-course and long-term antibiotics, ICS and oral corticosteroids, inhaled hyperosmolars, mucolytics, and use of airway clearance techniques. This overview clearly points to significant opportunities for further research aimed at improving outcomes for people with bronchiectasis. We have highlighted important endpoints for studies (particularly exacerbations, quality of life and lung function), and areas of clinical practice that are in most urgent need of evidence-based support (including long-term antibiotics, ICSs and mucolytics).As the evidence is confined to small trials of short duration, it is not currently possible to assess the balance between the benefits and potential harms of treatments for bronchiectasis.