Airway clearance devices for cystic fibrosis: an evidence-based analysis.

OBJECTIVE: The purpose of this evidence-based analysis is to examine the safety and efficacy of airway clearance devices (ACDs) for cystic fibrosis and attempt to differentiate between devices, where possible, on grounds of clinical efficacy, quality of life, safety and/or patient preference.

BACKGROUND: Cystic fibrosis (CF) is a common, inherited, life-limiting disease that affects multiple systems of the human body. Respiratory dysfunction is the primary complication and leading cause of death due to CF. CF causes abnormal mucus secretion in the airways, leading to airway obstruction and mucus plugging, which in turn can lead to bacterial infection and further mucous production. Over time, this almost cyclical process contributes to severe airway damage and loss of respiratory function. Removal of airway secretions, termed airway clearance, is thus an integral component of the management of CF. A variety of methods are available for airway clearance, some requiring mechanical devices, others physical manipulation of the body (e.g. physiotherapy). Conventional chest physiotherapy (CCPT), through the assistance of a caregiver, is the current standard of care for achieving airway clearance, particularly in young patients up to the ages of six or seven. CF patients are, however, living much longer now than in decades past. The median age of survival in Canada has risen to 37.0 years for the period of 1998-2002 (5-year window), up from 22.8 years for the 5-year window ending in 1977. The prevalence has also risen accordingly, last recorded as 3,453 in Canada in 2002, up from 1,630 in 1977. With individuals living longer, there is a greater need for independent methods of airway clearance. AIRWAY CLEARANCE DEVICES: THERE ARE AT LEAST THREE CLASSES OF AIRWAY CLEARANCE DEVICES: positive expiratory pressure devices (PEP), airway oscillating devices (AOD; either handheld or stationary) and high frequency chest compression (HFCC)/mechanical percussion (MP) devices. Within these classes are numerous different brands of devices from various manufacturers, each with subtle iterations. At least 10 devices are licensed by Health Canada (ranging from Class 1 to Class 3 devices). EVIDENCE-BASED ANALYSIS OF EFFECTIVENESS:

RESEARCH QUESTIONS: Does long-term use of ACDs improve outcomes of interest in comparison to CCPT in patients with CF?Does long-term use of one class of ACD improve outcomes of interest in comparison to another class of ACD in CF patients?

LITERATURE SEARCH: A comprehensive literature search was performed on March 7, 2009 using OVID MEDLINE, MEDLINE In-Process and Other Non-Indexed Citations, EMBASE, the Cumulative Index to Nursing & Allied Health Literature (CINAHL), the Cochrane Library, and the International Agency for Health Technology Assessment (INAHTA) for studies published from January 1, 1950 to March 7, 2009.

INCLUSION CRITERIA: All randomized controlled trials including those of parallel and crossover design,Systematic reviews and/or meta-analyses. Randomized controlled trials (RCTs), systematic reviews and meta-analyses

EXCLUSION CRITERIA: Abstracts were generally excluded because their methods could not be examined; however, abstract data was included in several Cochrane meta-analyses presented in this paper;Studies of less than seven days duration (including single treatment studies);Studies that did not report primary outcomes;Studies in which less than 10 patients completed the study.

OUTCOMES OF INTEREST: Primary outcomes under review were percent-predicted forced expiratory volume (FEV-1), forced vital capacity (FVC), and forced expiratory flow between 25%-75% (FEF25-75). Secondary outcomes included number of hospitalizations, adherence, patient preference, quality of life and adverse events. All outcomes were decided a priori.

SUMMARY OF FINDINGS: Literature searching and back-searching identified 13 RCTs meeting the inclusion criteria, along with three Cochrane systematic reviews. The Cochrane reviews were identified in preliminary searching and used as the basis for formulating this review. Results were subgrouped by comparison and according to the available literature. For example, results from Cochrane meta-analyses included abstract data and therefore, additional meta-analyses were also performed on trials reported as full publications only (MAS generally excludes abstracted data when full publications are available as the methodological quality of trials reported in abstract cannot be properly assessed). Executive Summary Table 1 summarizes the results across all comparisons and subgroupings for primary outcomes of pulmonary function. Only two comparisons yielded evidence of moderate or high quality according to GRADE criteria-the comparisons of CCPT vs. PEP and handheld AOD vs. PEP-but only the comparison of CCPT vs. PEP noted a significant difference between treatment groups. In comparison to CCPT, there was a significant difference in favour of PEP for % predicted FEV-1 and FVC according to one long-term, parallel RCT. This trial was accepted as the best available evidence for the comparison. The body of evidence for the remaining comparisons was low to very low, according to GRADE criteria, being downgraded most often because of poor methodological quality and low generalizability. Specifically, trials were likely not adequately powered (low sample sizes), did not conduct intention-to-treat analyses, were conducted primarily in children and young adolescents, and outdated (conducted more than 10 years ago). Secondary outcomes were poorly or inconsistently reported, and were generally not of value to decision-making. Of note, there were a significantly higher number of hospitalizations among participants undergoing AOD therapy in comparison to PEP therapy. ES Table 1:Summarization of results for primary outcomes by comparison and subgroupingsOutcome or SubgroupNo. of StudiesEstimate of Effectiveness (95% CI)P-valueHeterogeneity (I(2))GRADECCPT vs. PEP     Cochrane          FEV-1          FVC          FEF(25-75%)6640.08 (-1.45 to 1.62)0.38 (-1.56 to 2.23)-0.44 (-3.38 to 2.50)0.910.700.7746%63%36 N/A      Full publications only          FEV-1          FVC          FEF(25-75%)332-0.50 (-3.93 to 2.92)-0.86 (-4.66 to 2.95)-0.12 (-6.22 to 5.98)0.770.660.9777%74%0%  N/A     Long-term, parallel     RCTs only          FEV-1          FVC          FEF(25-75%)111-8.25 (-15.77 to -0.75)-8.74 (-16.03 to -1.45)-3.56 (-13.30 to 6.18)0.030.020.47N/AN/AN/A 1 TrialMODERATECCPT vs. HFCC/MP     Cochrane          FEV-1          FVC     FEF(25-75%)332-1.76 (-4.67 to 1.16)-1.42 (-5.17 to 2.33)0.49 (-2.54 to 3.52)0.240.460.750%70%0%  N/A     Full publications only          FEV-1          FVC          FEF(25-75%)332-2.10 (-5.49 to 1.29)-3.86 (-8.05 to 0.33)0.49 (-2.54 to 3.52)0.230.070.750%0%0% 3 TrialsLOWCCPT vs. AOD     2 of 3 RCTs/Cochrane          FEV-1          FVC          FEF(25-75%)2220.80 (-5.79 to 7.39)6.06 (-2.42 to 14.55)1.26 (-7.56 to 10.09)0.810.160.780%12%0% 3 TrialsLOWAOD vs. PEP     Long-term, parallel     RCTs only/Cochrane          FEV-1          FVC          FEF(25-75%)2220.29 (-4.17 to 4.75)-0.55 (-4.60 to 3.50)0.10 (-4.86 to 5.06)0.900.790.9773%77%0% 2 TrialsMODERATEAOD vs. HFCC/MP     Long-term, parallel     RCTs only/Cochrane          FEV-1          FVC          FEF(25-75%)111-1.6 (-3.44 to 0.24)-1.80 (-4.32 to 0.72)-1.40 (-3.07 to 0.27)0.090.080.16N/AN/AN/A 1 TrialVERY LOWBolding indicates significant differencePositive summary statistics favour the former intervention

ABBREVIATIONS: AOD, airway oscillating device; CCPT, conventional chest physiotherapy; CI, confidence interval; HFCC, high frequency chest compression; MP, mechanical percussion; N/A: not applicable; PEP, positive expiratory pressure

ECONOMIC ANALYSIS: Devices ranged in cost from around $60 for PEP and handheld AODs to upwards of $18,000 for a HFCC vest device. Although the majority of device costs are paid out-of-pocket by the patients themselves, their parents, or covered by third-party medical insurance, Ontario did provide funding assistance through the Assistive Devices Program (ADP) for postural drainage boards and MP devices. These technologies, however, are either obsolete or their clinical efficacy is not supported by evidence. ADP provided roughly $16,000 in funding for the 2008/09 fiscal year. Using device costs and prevalent and incident cases of CF in Ontario, budget impact projections were generated for Ontario. Prevalence of CF in Ontario for patients from ages 6 to 71 was cited as 1,047 cases in 2002 while incidence was estimated at 46 new cases of CF diagnosed per year in 2002. Budget impact projections indicated that PEP and handheld AODs were highly economically feasible costing around $90,000 for the entire prevalent population and less than $3,000 per year to cover new incident cases. HFCC vest devices were by far the most expensive, costing in excess of $19 million to cover the prevalent population alone.

CONCLUSIONS: There is currently a lack of sufficiently powered, long-term, parallel randomized controlled trials investigating the use of ACDs in comparison to other airway clearance techniques. While much of the current evidence suggests no significant difference between various ACDs and alternative therapies/technologies, at least according to outcomes of pulmonary function, there is a strong possibility that past trials were not sufficiently powered to identify a difference. Unfortunately, it is unlikely that there will be any future trials comparing ACDs to CCPT as withholding therapy using an ACD may be seen as unethical at present. Conclusions of clinical effectiveness are as follows: Moderate quality evidence suggests that PEP is at least as effective as or more effective than CCPT, according to primary outcomes of pulmonary function. (ABSTRACT TRUNCATED)