Thoracotomy repair has long been considered the gold standard for the repair of esophageal atresia but is associated with potential musculoskeletal complications which may result in long term morbidity for the patient. thoracoscopy repair offers better visualization of the posterior mediastinal structures, while limiting the surgical trauma. However, studies have shown that the incidence of anastomotic leakage and anastomotic stricture in thoracoscopic repair is not significantly lower than thoracostomy repair. Robotic repair had shorter anastomotic time, lower incidence of anastomotic leakage and stricture, and lower unplanned readmission rate than the thoracotomy repair. However, there were no randomized controlled trials to verify the effectiveness of three procedures. The objection was to compare the difference between robotic repair and thoracoscopic repair, and thoracotomy repair in intraoperative parameters and postoperative complications in EA neonates.

Esophageal atresia (EA) and tracheoesophageal fistula (TEF) is one of the most common congenital malformations of esophagus, with an incidence of 1/2500~1/4500. The condition can be an isolated malformation or may be associated with other congenital anomalies. Since Dr Cameron Height performed the first successful primary repair of a neonate with EA/ TEF in 1941, many advances in surgical technique and neonatal care have steadily improved survival rates of babies within the EA/TEF spectrum. Survival of infants with esophageal atresia has increased over time since the first successful repair and the overall survival exceeds 90%. Commonly, esophageal atresia is repaired via a right posterolateral thoracotomy and more recently muscle sparing thoracotomy has become an alternative to the traditional muscle cutting approach. Open repair has long been considered the gold standard for the repair of esophageal atresia but is associated with potential musculoskeletal complications which may result in long term morbidity for the patient.

The first successful thoracoscopic surgery of a child with EA was reported in 1999. Compared to thoracotomy repair the proposed main advantage of thoracoscopic repair is that it offers better visualization of the posterior mediastinal structures, while limiting the surgical trauma. However, studies have shown that the incidence of anastomotic leakage and anastomotic stricture in thoracoscopic repair is not significantly lower than thoracostomy repair, thoracoscopic repair also offers concerns with more complicated anesthesia, limited workspace, and difficultly controlling the vascular structures. Especially, suturing within such a small, closed space has been considered a major technical difficulty.

Robotic repair was first reported by Meehan in 2009, followed by several case reports. The reported reasons for conversion mainly focused on the incompatibility between the robotic trocar's size and the intercostal space's width, and the technical challenges due to instrument collisions in the extremely limited thoracoscopic space. The intercostal space of neonates is highly narrow, and the thoracic diameter is only 8 cm. These are the two key technical issues to be addressed in this study. According to the existing robotic system setup standard for adults, the distance required between trocar ports in robots is usually at least 8 cm to ensure sufficient operating space and avoid instrument collisions. Even for the new generation of robots, this minimum distance requires 5-6 cm. Huge robotic trocars used in EA neonates fail to meet the standard for conventional operating port distances. There have study designed an asymmetric port distribution technique in which the third and eighth intercostal ports are 3 cm and 5 cm away from the camera port. The surgeons primarily manipulated the inner-articulating part of the robotic arms within the thoracic cavity, avoiding instrument collisions outside. Moreover, the setup of the trocars ensures that the robotic arms can reach the main operating area. When combined with instruments of 7 degrees of freedom, the mobilization and anastomosis of the esophagus could be completed easily, breaking through the narrow space restriction of thoracic cavity. Inserting 8-12 cm trocars into tiny intercostal space was another technical challenge. The results shown the robotic repair had shorter anastomotic time, lower incidence of anastomotic leakage and stricture, and lower unplanned readmission rate than the thoracotomy repair.

An international survey from 2014 highlighted the need for consensus on the optimal surgical treatment of EA. However, a detailed understanding of whether thoracoscopic repair or robotic repair offers advantages in terms of health outcomes, safety, and efficacy for providers compared to thoracotomy repair is still lacking. Several reviews are opinion-based or obscured by institutional/personal experiences. Herewith, we designed a comprehensive study and focused on evaluating the difference between robotic repair and thoracoscopic repair, and thoracotomy repair in intraoperative parameters and postoperative complications in EA neonates.

Inclusion Criteria:

• Type C EA neonates with short esophageal gap length (less than 3 vertebral bodies), mini-invasive repair, and a successful one-stage anastomosis were included.

Exclusion Criteria:

• Patients with respiratory distress requiring assisted ventilation, long esophageal gap length, multistage surgery, other types EA (type A/B/D/E) or surgical contraindications were excluded.

Gestational age less than 35 weeks and birth weight less than 2kg were excluded.