OBJECTIVES: To assess the efficacy and safety of combined aclidinium bromide and long-acting beta2-agonists in stable COPD.
SEARCH METHODS: We searched the Cochrane Airways Group Specialised Register (CAGR), ClinicalTrials.gov, World Health Organization (WHO) trials portal, United States Food and Drug Administration (FDA) and manufacturers' websites as well as the reference list of published trials up to 12 October 2018.
SELECTION CRITERIA: Parallel-group randomised controlled trials (RCTs) assessing combined aclidinium bromide and LABAs in people with stable COPD.
DATA COLLECTION AND ANALYSIS: We used standard methodological procedures expected by Cochrane for data collection and analysis. The primary outcomes were exacerbations requiring a short course of an oral steroid or antibiotic, or both; quality of life measured by a validated scale and non-fatal serious adverse events (SAEs). Where the outcome or study details were not reported, we contacted the study investigators or pharmaceutical company trial co-ordinators (or both) for missing data.
MAIN RESULTS: We identified RCTs comparing aclidinium/formoterol FDC versus aclidinium, formoterol or placebo only. We included seven multicentre trials of four to 52 weeks' duration conducted in outpatient settings. There were 5921 participants, whose mean age ranged from 60.7 to 64.7 years, mostly men with a mean smoking pack-years of 46.4 to 61.3 of which 43.9% to 63.4% were current smokers. They had a moderate-to-severe degree of COPD with a mean postbronchodilator forced expiratory volume in one second (FEV1) between 50.5% and 61% of predicted normal and the baseline mean FEV1 of 1.23 L to 1.43 L. We assessed performance and detection biases as low for all studies whereas selection, attrition and reporting biases were either low or unclear.FDC versus aclidiniumThere was no evidence of a difference between FDC and aclidinium for exacerbations requiring steroids or antibiotics, or both (OR 0.95, 95% CI 0.71 to 1.27; 2 trials, 2156 participants; moderate-certainty evidence); quality of life measured by St George's Respiratory Questionnaire (SGRQ) total score (MD -0.92, 95% CI -2.15 to 0.30); participants with significant improvement in SGRQ score (OR 1.17, 95% CI 0.97 to 1.41; 2 trials, 2002 participants; moderate-certainty evidence); non-fatal SAE (OR 1.19, 95% CI 0.79 to 1.80; 3 trials, 2473 participants; moderate-certainty evidence); hospital admissions due to severe exacerbations (OR 0.62, 95% CI 0.29 to 1.29; 2 trials, 2156 participants; moderate-certainty evidence) or adverse events (OR 0.95, 95% CI 0.76 to 1.18; 3 trials, 2473 participants; moderate-certainty evidence). Compared with aclidinium, FDC improved symptoms (Transitional Dyspnoea Index (TDI) focal score: MD 0.37, 95% CI 0.07 to 0.68; 2 trials, 2013 participants) with a higher chance of achieving a minimal clinically important difference (MCID) of at least one unit improvement (OR 1.34, 95% CI 1.11 to 1.62; high-certainty evidence); the number needed to treat for an additional beneficial outcome (NNTB) being 14 (95% CI 9 to 39).FDC versus formoterolWhen compared to formoterol, combination therapy reduced exacerbations requiring steroids or antibiotics, or both (OR 0.78, 95% CI 0.62 to 0.99; 3 trials, 2694 participants; high-certainty evidence); may decrease SGRQ total score (MD -1.88, 95% CI -3.10 to -0.65; 2 trials, 2002 participants; low-certainty evidence; MCID for SGRQ is 4 units); increased TDI focal score (MD 0.42, 95% CI 0.11 to 0.72; 2 trials, 2010 participants) with more participants attaining an MCID (OR 1.30, 95% CI 1.07 to 1.56; high-certainty evidence) and an NNTB of 16 (95% CI 10 to 60). FDC lowered the risk of adverse events compared to formoterol (OR 0.78, 95% CI 0.65 to 0.93; 5 trials, 3140 participants; high-certainty evidence; NNTB 22). However, there was no difference between FDC and formoterol for hospital admissions, all-cause mortality and non-fatal SAEs.FDC versus placeboCompared with placebo, FDC demonstrated no evidence of a difference in exacerbations requiring steroids or antibiotics, or both (OR 0.82, 95% CI 0.60 to 1.12; 2 trials, 1960 participants; moderate-certainty evidence) or hospital admissions due to severe exacerbations (OR 0.55, 95% CI 0.25 to 1.18; 2 trials, 1960 participants; moderate-certainty evidence), although estimates were uncertain. Quality of life measure by SGRQ total score was significantly better with FDC compared to placebo (MD -2.91, 95% CI -4.33 to -1.50; 2 trials, 1823 participants) resulting in a corresponding increase in SGRQ responders who achieved at least four units decrease in SGRQ total score (OR 1.72, 95% CI 1.39 to 2.13; high-certainty evidence) with an NNTB of 7 (95% CI 5 to 12). FDC also improved symptoms measured by TDI focal score (MD 1.32, 95% CI 0.96 to 1.69; 2 studies, 1832 participants) with more participants attaining at least one unit improvement in TDI focal score (OR 2.51, 95% CI 2.02 to 3.11; high-certainty evidence; NNTB 4). There were no differences in non-fatal SAEs, adverse events and all-cause mortality between FDC and placebo.Combination therapy significantly improved trough FEV1 compared to aclidinium, formoterol or placebo.
AUTHORS' CONCLUSIONS: FDC improved dyspnoea and lung function compared to aclidinium, formoterol or placebo, and this translated into an increase in the number of responders on combination treatment. Quality of life was better with combination compared to formoterol or placebo. There was no evidence of a difference between FDC and monotherapy or placebo for exacerbations, hospital admissions, mortality, non-fatal SAEs or adverse events. Studies reported a lower risk of moderate exacerbations and adverse events with FDC compared to formoterol; however, larger studies would yield a more precise estimate for these outcomes.