A Novel Intraoperative Leak Test Procedure (GAM Procedure) to Prevent Postoperative Anastomotic Leakage in Gastric Cancer Patients Who Underwent Gastrectomy

Background: Anastomotic leakage following a radical gastrectomy is a serious complication of gastric cancer and esophagogastric junction cancer. The benefit of intraoperative leak testing for the prevention of postoperative anastomotic leakage has been controversial. We introduce a new procedure, which combines the techniques of gastroscopy, air, and methylene blue (GAM) for intraoperative leakage testing. Our objective was to evaluate the efficacy and safety of the GAM procedure for intraoperative leak testing and to compare the surgical complications of gastric cancer patients who underwent gastrectomy with and without intraoperative leak testing using the GAM procedure. Materials and Methods: A total of 210 patients who underwent radical gastrectomy for gastric cancer were included. Patients were divided into 2 groups: the intraoperative leak testing group using the GAM procedure (IOLT), and the group for which no intraoperative leak testing was done (NIOLT). Clinical and pathologic characteristics, the incidence of postoperative anastomotic leakage, and other surgical complications were compared between the 2 groups. Results: There were 82 patients in the IOLT group and 82 patients in the NIOLT group after propensity score matching. In the IOLT group, 4 (4.9%) patients were found to have anastomotic discontinuity during the operation; we repaired these anastomotic discontinuities intraoperatively. The incidence of postoperative anastomotic leakage was higher in the NIOLT group compared with the IOLT group, 6 (7.3%) versus 0 (0%), respectively ( P =0.01). The average time of the GAM procedure was 4.99±1.75 minutes. The surgical time was prolonged by 30 minutes in the IOLT group compared with the NIOLT group, 302.2±79.9 versus 272.1±85.2, respectively ( P =0.02). The length of hospital stay, 15.80±4.55 versus 17.00±6.20 ( P =0.16) was reduced in the IOLT group compared with the NIOLT group. The logistic regression model suggested that IOLT, sex, age, Eastern Cooperative Oncology Group, cT stage, tumor diameter, pT stage, pN stage, and Lauren classification were not risk factors for postoperative complication. Conclusions: The GAM procedure of intraoperative leakage testing can effectively reduce the incidence of postoperative anastomotic leakage in gastric cancer patients undergoing gastrectomy.