Background Device closure of atrial septal defect (ASD) has emerged as a treatment modality for the past 3 decades and has changed the natural history of ASD compared to that of surgical closure. Early intervention in ASD retards the geometrical and electrical remodelling of the atrium that contributes to the development of atrial tachyarrhythmias. We studied the incidence of atrial arrhythmias in patients undergoing surgical and device closure of ASD. Methods We did this retrospective observational study at a tertiary referral centre, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram, Kerala. Patients of all age groups undergoing surgical and device closure of ASD between 1 January 2003 and 31 December 2008 were included to compare the incidence and nature of atrial arrhythmias and also analyse the difference in new-onset atrial arrhythmias between the two arms. Results Of 277 patients, 144 with surgical closure and 133 with device closure were followed up for 10-15 years. A larger number of men underwent surgical closure (41.7%) compared to device closure (25.6%). The mean (SD) follow-up was 12.6 (3.7) years in the surgical closure group and 10.9 (2.6) years in the device closure arm. There were a larger number of patients with atrial tachyarrhythmias in the surgical closure group (6.3%) compared to the device closure group (0.8%) at baseline (p=0.02). A significantly larger number of patients had atrial fibrillation in the surgical closure group (5.6%) compared to the device closure group (0.7%) at baseline (p=0.003). Pulmonary hypertension at baseline was present in 38.9% of patients in the surgical closure group and in 23.3% of patients in the device closure group (p=0.006). New-onset atrial arrhythmias occurred in patients ≥30 years of age (p=0.006) and exclusively in patients with pulmonary hypertension in the surgical group (3.7%) and in the device closure group (6.6%). This was statistically significant in the device closure group (p=0.05) but not in the surgical closure group (p=0.13). The incidence of new-onset arrhythmias was not statistically significant in both groups. Conclusions Atrial arrhythmias were significantly more common in patients who underwent surgical or device closure at ≥ 30 years of age and in patients with pulmonary hypertension. There was no difference in new-onset atrial arrhythmias between the surgical and device closure groups. Our study results suggest that surgical or device closure before 30 years of age and before the development of atrial arrhythmias may be beneficial with respect to the development of atrial arrhythmias.