Percutaneous closure of atrial septal defects using an Amplatzer Septal Occluder (ASO) has recently become the procedure of first choice. However, when ASO deployment is difficult, procedures may be prolonged and complications may occur. We investigated a method for identifying cases in which ASO deployment would be difficult. After retrospectively identifying 70 patients (age: 4.1-70.4 years; body weight: 15.6-77.3 kg) who underwent atrial septal defect closure using an ASO in 2007 or later, we classified them into three groups: Group A, implantation by a conventional approach; Group B, implantation by a right upper pulmonary vein approach, and Group C, change from a conventional to a right upper pulmonary vein approach. Characteristics of the groups were compared. Individually, none of the investigated characteristics was suitable for identifying difficult cases. Furthermore, we observed no consistent trends between aortic rim deficiency and ASO diameter, or between SG/IAS angle, which is the angle formed by the super stiff guidewire (SG) and the intra-atrial septum (IAS). However, the ASO diameter divided by the diameter of the left atrium (ASO/LA) correlated with the SG/IAS angle in Group C. Using this correlation, ASO implantation is predicted to be difficult in patients with an ASO/LA (%) ratio exceeding the (SG/IAS angle) × 1.44 + 48.1, which represents the 95th percentile of Group C. Graphing the SG/IAS angle and the ASO/LA ratio can identify cases in which ASO implantation may be difficult. We consider this method of selecting an approach to be extremely useful for avoiding various risks.