Since the first report of thoracoscopic repair of EA/TEF in 1999, thoracoscopic surgery for EA/TEF has gradually been proven to be safe and feasible. Compared with thoracotomy, thoracoscopic surgery can allow easier esophageal dissection with excellent visualization and can avoid the thoracotomy incision which is associated with more pain and skeletal deformities. Several reports have assessed the safety and effectiveness of thoracoscopic surgery. Elbarbary et al. showed comparable outcomes between the thoracoscopic and the open method for short-gap type-C EA/TEF.  Yamoto et al. demonstrated that the thoracoscopic approach was favorable and safe for EA/TEF repair in carefully selected patients. A meta-analysis carried out in 2016 concluded that there were no significant differences between thoracoscopy and thoracotomy groups with respect to anastomotic leaks and strictures. In addition, patients who had received thoracoscopic surgery were extubated faster, started oral feeding earlier, and stayed in the hospital for a shorter period of time. However, it was noted that their operative time was longer.
The reported rate of conversion from thoracoscopy to thoracotomy surgery in literature ranges from 4–44%. In the present study, 6 patients (9.68%) were converted to thoracotomy surgery due to decreased oxygen saturation or high esophageal anastomosis tension. All conversions took place before 2016 as shown in Supplementary Table 2. With more experience, none of the thoracoscopic procedures had to be converted to open surgery anymore. In addition, compared to before 2016, the duration of operation time decreased significantly, and there was a remarkable reduction in postoperative leakage from 58–19% (Supplementary Table 2). We believe that increased surgical expertise and the technical adjustments led to this reduction in postoperative leakage. As this study has demonstrated, there is a clearly learning curve, but upon overcoming the learning curve and mastering the intracorporeal knotting, the operative time of thoracoscopy was comparable to thoracotomy.
In the present study, we found that the duration of mechanical ventilation after surgery of thoracoscopy was longer than that of thoracotomy. The main reason was the improvement of perioperative management system which enabled more patients to be admitted to the intensive care unit for mechanical ventilation after thoracoscopy surgery. Mechanical ventilation helps avoiding the impact of autonomous respiration and swallowing on the esophageal anastomosis. However, we found no significant differences in thoracoscopy versus thoracotomy in regard to duration of stay in the intensive care unit after surgery, length of time before starting liquid diet and hospitalization length.
Wu Y et al. analyzed the outcomes that are universally considered to be indicators of the effectiveness of the thoracoscopic EA/TEF repair, namely, conversion to thoracotomy surgery, the rates of complications, anastomotic leaks and strictures. In their meta-analysis, they demonstrated statistically insignificant differences in all the parameters considered. In the present study, there were no significant differences in the rates of pneumothorax, anastomotic leakage, anastomotic stricture, and recurrent tracheoesophageal fistula between the 2 groups. To a certain extent, thoracoscopic surgery reduced the incidence of anastomotic leakage and increased the incidence of anastomotic stricture in this study. However, 74 patients with anastomotic leakage were all healed after fasting, parenteral nutrition support, and insertion of a gastric tube. It has been reported that endoscopic balloon dilatation is an effective method for the treatment of anastomotic stricture[12, 13]. In our study, patients received balloon dilatation due to esophageal stricture, and the median number of dilations for both groups was 6. As for the higher incidence rate of anastomotic stricture in the thoracoscopic group, we analyzed that it might be related to the more stitches and the smaller proximal pouch opening during the thoracoscopic anastomosis. At present, in our clinical work, we are still improving the thoracoscopic technique to further improve the safety of surgery and reduce postoperative complications.
There were several limitations in our study. On the one hand, a high rate of loss to follow-up (39/190, 20.53%) has an impact on the outcome. We found that 38 patients (38/39, 97.44%) were lost-to follow up before standardized EA/TEF treatment and management procedures (complete follow-up procedures were established in 2016 in our hospital). In order to create a more effective postoperative long-term care and treatment, it is fundamental to ensure a standardized, multidisciplinary follow-up that must continue until adulthood. On the other hand, the criteria for decisions related to surgical procedures and perioperative and postoperative management were impacted by the experience and improvement of the treatment plan, as well as the establishment of clinical teams and changes in practice over time. Thus, a prospective randomized controlled trial is needed to explore the differences in clinical outcomes between thoracoscopy and thoracotomy surgery for Gross type C EA/TEF in the furture.