Pharmacologic treatments for chronic obstructive pulmonary disease: a mixed-treatment comparison meta-analysis.

STUDY OBJECTIVE To assess the comparative efficacy of pharmacologic agents for the maintenance treatment of chronic obstructive pulmonary disease (COPD). DESIGN Traditional and mixed-treatment comparison (MTC) meta-analyses of randomized controlled trials. PATIENTS A total of 31,020 patients with COPD from 43 trials. MEASUREMENTS AND MAIN RESULTS A systematic literature search of various databases (through October 2007) was performed to identify randomized controlled trials of long-acting beta(2)-agonists, tiotropium, inhaled corticosteroids, and/or combination therapy with an inhaled corticosteroid and a long-acting beta(2)-agonist in patients with COPD. Forty-three trials were included. Both meta-analyses were used to evaluate the occurrence of one or more episodes of COPD exacerbation, overall mortality, and patient withdrawal rates. With MTC analysis, long-acting beta(2)-agonists, tiotropium, inhaled corticosteroids, and combination inhaled corticosteroid-long-acting beta(2)-agonist therapy each decreased the odds of having an exacerbation by 16%, 31%, 15%, and 24%, respectively, compared with placebo. Moreover, tiotropium use reduced the odds of having at least one exacerbation by 18% compared with long-acting beta(2)-agonists and by 19% compared with inhaled corticosteroids alone. Each of the four drug classes was associated with significant odds reductions in patient withdrawals (26-41%) compared with placebo, and both tiotropium and combination therapy significantly decreased the odds of patient withdrawals compared with long-acting beta(2)-agonists or inhaled corticosteroids alone. Only combination therapy was associated with a mortality benefit, showing a 29% reduction compared with placebo and a 25% reduction compared with long-acting beta(2)-agonists alone. Compared with combination therapy, tiotropium use reduced exacerbations by 9% and increased mortality by only 4%. These findings did not demonstrate significant changes in the sensitivity or subgroup analyses, which were performed to evaluate the effect of heterogeneity among the included studies. CONCLUSIONS Combination inhaled corticosteroid-long-acting beta(2)-agonist therapy was associated with the greatest positive effect on outcomes in patients with COPD. Of the bronchodilator monotherapies, tiotropium was associated with lower odds of having a COPD exacerbation or withdrawal from a study compared with long-acting beta(2)-agonists.