Background: Remote-access perfusion and robotics have enabled totally endoscopic closure of atrial septal defect (ASD) and patent foramen ovale (PFO). We report on a stepwise approach to a totally endoscopic procedure.
Methods: Seventeen patients (median age, 39 years; range, 21-55 years) underwent limited-access ASD or PFO closure. As a preparative step, the operation was carried out through minithoracotomy in 11 patients. In parallel, experience with robotic surgery was gained with totally endoscopic coronary artery bypass grafting procedures. After performance of ASD closures in dry-laboratory models using the da Vinci telemanipulation system, 6 patients were operated on in a totally endoscopic fashion.
Results: With the endoscopic approach, significant learning curves were noted for cardiopulmonary bypass time y(min) = 226 - 41 * ln(x) (P = .03) and aortic cross-clamp time y(min) = 134 - 42 * ln(x) (P = .01) (x = number of procedures). There was no hospital mortality, and no residual shunts were detected at postoperative echocardiography. Median ventilation time was 9 hours (range, 0-18 hours) for the minithoracotomy group and 6 hours (range, 4-19 hours) for the totally endoscopic group. Median intensive care unit stay was 20 hours (range, 18-24 hours) and 18 hours (range, 18- 120 hours), respectively.
Conclusions: The implementation of robotic totally endoscopic closure of ASD or PFO in a heart surgery program seems to be safe. An intermediate step of performing the operations through minithoracotomy, adapting to remote access perfusion systems, and gaining experience in other robotic cardiac surgical procedures seems worthwhile. Learning curves are apparent, and adequate defect closure does not seem to be compromised by the totally endoscopic approach.