Early postoperative endoscopy for predicting anastomotic leakage after minimally invasive esophagectomy: A large-volume retrospective study

Background Anastomotic leakage is one of the most severe adverse events of minimally invasive esophagectomy for esophageal cancer. Early postoperative endoscopy is considered to be the most objective means to diagnose anastomotic leakage, but its safety is questioned by clinicians. This study aimed to evaluate the safety and effectiveness of early postoperative endoscopy in predicting anastomotic leakage. Methods Patients who underwent minimally invasive esophagectomy (from January 2017 to June 2021) in our center were identified and divided into early postoperative endoscopy and control groups according to whether they underwent early postoperative endoscopy within 72 hours after surgery. Propensity score matching was used to balance baseline characteristics. The incidence of postoperative adverse events was compared between the 2 groups, risk variables for anastomotic leakage were identified using logistic regression, and abnormal endoscopic findings related to anastomotic leakage occurrence were explored. Results A total of 436 patients were enrolled, of whom 134 underwent early postoperative endoscopy. One hundred and thirty-two pairs were matched by propensity score matching, and baseline characteristics were well-balanced. Both before and after propensity score matching, early postoperative endoscopy did not increase the incidence of postoperative adverse events (chyle leak, hypoproteinemia, pneumonia, etc) and in-hospital mortality. Notably, the incidence of anastomotic leakage (9.8% vs 22.7%) and the length of mean postoperative hospital stay (17.6 vs 20.9 days) was significantly decreased in the early postoperative endoscopy group. Finally, based on the findings under early postoperative endoscopy, we found that gastric graft ischemia is related to a higher incidence of anastomotic leakage (P = .023). Conclusion Early postoperative endoscopy does not increase postoperative adverse events after minimally invasive esophagectomy and may guide early prediction and intervention strategies for anastomotic leakage in patients undergoing minimally invasive esophagectomy.