Modified robotic atrial septal defect repair on beating heart: a nine-year single-centre experience with learning-curve analysis and comparison with thoracoscopy

Objective To evaluate a modified robot-assisted technique for atrial septal defect (ASD) repair on beating heart and to compare perioperative outcomes with thoracoscopic repair. Methods 198 patients with ASD underwent modified ASD repair using the Da Vinci robotic surgical system on beating heart in the Department of Cardiovascular Surgery of the First Affiliated Hospital with Nanjing Medical University. Cumulative sum (CUSUM) analysis was used to build learning curves for operation time and cardiopulmonary bypass (CPB) time. Patients operated after both curves had plateaued formed the robotic group (n = 117); 113 patients treated thoracoscopically during the same period served as the comparison group. Results The learning curves for operation time and CPB time were 81 cases and 66 cases respectively. ASD diameter and pulmonary artery systolic pressure were higher in the robotic group, which also contained seven patients with partial anomalous pulmonary venous connection (PAPVC) and one with anomalous drainage of the inferior vena cava. CPB time was shorter in the robotic group. Operation time, mechanical ventilation time, intensive care unit stay, postoperative hospital stay and 24-hour chest drainage did not differ. Echocardiography at 3, 30 and 90 days and at one year showed no residual shunt or tricuspid regurgitation in either group. Conclusions The modified robot-assisted technique for beating-heart ASD repair is safe and reproducible. Once the learning curve is complete, CPB time is shorter than with thoracoscopy despite greater anatomical complexity, and the technique offers a valid minimally invasive option for defects that are unsuitable for device closure.