OBJECTIVES To investigate the feasibility and safety of non-robotically assisted totally thoracoscopic closure for atrial septal defect (ASD) on perfused beating hearts. METHODS Twenty-four patients (8'45 years, mean 14.4 ± 18.7) underwent ASD closure on beating hearts by a totally thoracoscopic approach without the aid of a robotic surgical system. Additional 72 patients undergoing totally thoracoscopic ASD closure on cardioplegic arrested hearts were selected as a control. Cardiopulmonary bypass (CPB) was achieved peripherally. The aorta was not cross-clamped in the study group but it was cross-clamped in the control group. RESULTS ASD closure was successful in all study and control group patients without in-hospital mortality or major complications. The total duration of operation (76 ± 9 vs. 98 ± 6 min, P = 0.012), CPB time (32 ± 5 vs. 48 ± 4 min, P = 0.001), duration of intensive care stay (10.0 ± 5.1 vs. 19.2 ± 4.2 h, P = 0.003) and post-operative hospital stay (4.5 ± 0.8 vs. 5.0 ± 1.1 days, P = 0.045) in the study group were shorter than in the control group. There was no statistically significant difference in the proportion of patients requiring in-operation blood transfusion between study group and control group (25.0 vs. 36.1%, P = 0.226). Follow-up transthoracic echocardiography on Day 5 and Day 30 showed no residual shunts in study or control group patients. CONCLUSIONS Non-robotically assisted totally thoracoscopic closures of ASD on perfused beating hearts are feasible and safe. These procedures are associated with a shorter operation time and a shorter hospital stay than in surgeries on cardioplegic arrested hearts.