The impact of device size choice on closure results was analyzed in 138 (101 females, 37 males; age 0.5-84.0 years) consecutive patients who underwent transcatheter closure of the secundum atrial septal defect (ASD) using the Amplatzer septal occluder (ASO). The balloon stretched diameter (SD) of ASD was 19.5 +/- 7.2 mm in 123 patients with single defects, and 20.4 +/- 6.6 mm for the largest defects in 15 patients with multiple ASDs. The difference (delta) between ASO size chosen for closure and the stretched diameter of the defect was calculated and divided into groups: A (delta < -2 mm); B (delta -2.0 to -0.1 mm); C (delta = 0); D (delta 0.1-2.0 mm) and E (delta > 2 mm). The results demonstrated that immediate and 24-hour complete closure rates were significantly higher in patients in groups C and D (P < 0.001). However, at 6-month follow-up, the complete closure rates were similar in patients of groups A-D, while patients of group E had a lower closure rate of 75%. The complication rates were similar in all groups. In conclusion, a choice of a device size identical to or within 2 mm larger than the SD of the defect should be used to maximize the closure rates of ASD using the ASO.