Twenty-two trials (n=15,276 patients) were included in the review.
The drop-out rate was 18.5% in the anticholinergic group, 19% in the β2-agonist group and 24.8% in the placebo group. Sensitivity analysis was not carried out since all the trials were judged to be of sufficiently good quality. There was no reported evidence of publication bias.
Inhaled anticholinergics compared with placebo (7 trials).
Statistically significant reductions in risk of withdrawal from the trial for COPD exacerbation and severe exacerbations were reported (respectively) as 40% (RR 0.60, 95% CI: 0.48, 0.75) and 33% (RR 0.67, 95% CI: 0.53, 0.86). The absolute risk reduction for severe exacerbations was approximately 4 cases per 100 patient-years of treatment (NNT=25). The risk of respiratory deaths was significantly reduced in the intervention group by 73% (RR 0.27, 95% CI: 0.09, 0.81). The absolute risk reduction was 0.36% per year (NNT=278).
β2-agonists compared with placebo (13 trials).
Twelve trials assessed salmeterol and formoterol. In the intervention group, a statistically significant reduction in risk of withdrawal from the trial for COPD exacerbation was reported as 19% (RR 0.81, 95% CI: 0.68, 0.95), with no significant effect for hospitalisation. There was an associated increase in respiratory deaths (RR 2.47, 95% CI: 1.12, 5.45). The absolute risk increase was 0.76% per year (NNH=131).
Anticholinergics compared with β2-agonists (7 trials).
β2-agonists showed a statistically significant increase in COPD exacerbations that necessitated withdrawal from the trial (RR 2.02, 95% CI: 1.39, 2.93), and severe exacerbations requiring hospitalisation (RR 1.95, 95% CI: 1.06, 3.59). There was a non significant association between increased respiratory death and β2-agonists in the 2 trials that assessed this outcome. There were no significant differences in severe exacerbations or death when combined anticholinergics and β2-agonists were compared with anticholinergics alone (4 trials).