Beta-adrenergic stimulation may increase heart rate and the potential for cardiac arrhythmias. The effect of inhaled long-acting beta2-agonists (LABAs) on these outcomes was evaluated in patients with chronic obstructive pulmonary disease (COPD) in 2 double-blind randomized clinical trials. The pretreatment arrhythmia occurrence frequency in these patients was also described. In this analysis, 24-hour Holter monitoring data were pooled from 2 identically designed Phase III trials. Patients were randomized to LABA treatment or placebo for 12 weeks: a) nebulized arformoterol 15 microg BID, b) 25 microg BID, or c) 50 microg QD; d) salmeterol metered dose inhaler 42 microg BID; or e) placebo. The 24-hour Holter monitoring was performed pretreatment and at Weeks 0 (first day of dosing), 6, and 12. We assessed the proportion of patients with each of 4 arrhythmias: atrial tachycardia, atrial fibrillation/flutter, and "nonsustained"; (4-10 beats) and "sustained"; (>10 beats) ventricular tachycardia. There were 5226 Holter recordings in 1429 treated patients. At baseline, there was a low frequency of occurrence of atrial fibrillation/flutter (0.1%), nonsustained ventricular tachycardia (3.1%), and >10 beat ventricular tachycardia (0.3%). Atrial tachycardia occurred frequently (41.8%). The proportion of patients with treatment-emergent atrial tachycardia ranged from 27% to 32% and was non-significantly higher, by approximately 2%-5% (p = 0.70), in the LABA groups compared with the placebo group. The rates of the other more serious arrhythmias did not increase with LABA treatment and were similar to placebo. All treatment groups (LABA and placebo) had consistent small decreases from baseline in mean 24-hour and maximum hourly heart rate. In conclusion, in this large cohort of COPD patients with no or stable cardiac comorbidities, a high proportion ( approximately 40%) of patients were observed to have atrial tachycardia before treatment, which increased by 2%-5% with LABA treatment. More serious arrhythmias were infrequent and did not increase with inhaled LABA therapy. LABA administration did not increase mean heart rate.