Long-Acting Bronchodilator Initiation in COPD and the Risk of Adverse Cardiopulmonary Events: A Population-Based Comparative Safety Study.

Abstract

Long-acting bronchodilators, including long-acting beta₂-agonists (LABA) and the anticholinergic tiotropium, are recommended as initial maintenance therapy in COPD. Studies to date have been limited in size and reported ambivalent results on the comparative risk of cardiovascular, cerebrovascular, and pulmonary adverse events between these two long-acting bronchodilators. Moreover, little information is available for the period when treatment is first initiated, a time when subjects may be especially at risk. We identified a cohort of new users of long-acting bronchodilators between 2002 and 2012, age 55 or older, from the United Kingdom's Clinical Practice Research Datalink. Patients initiating tiotropium were matched on high-dimensional propensity scores and prior inhaled corticosteroid use with patients initiating LABAs, and followed for 1 year for the occurrence of acute myocardial infarction, stroke, heart failure, arrhythmia, and pneumonia. A total of 26,442 tiotropium initiators were matched to 26,442 LABA initiators, mainly single inhalers combined with inhaled corticosteroids. The hazard ratio of acute myocardial infarction associated with tiotropium initiation, relative to LABA initiation, was 1.10 (95% CI, 0.88-1.38), whereas for stroke it was 1.02 (95% CI, 0.78-1.34), for arrhythmia 0.81 (95% CI, 0.60-1.09), and heart failure 0.90 (95% CI, 0.79-1.02). The incidence of pneumonia was significantly less with tiotropium (hazard ratio, 0.81; 95% CI, 0.72-0.92). COPD treatment initiation with tiotropium compared with LABA does not increase cardiovascular risk in the first year of treatment. The risk of pneumonia is higher with LABA, a likely effect of the inhaled corticosteroids present in many LABA inhalers used in real world clinical practice.