OBJECTIVETo evaluate the feasibility and the short-term safety of self-pulling and latter transected esophagojejunostomy(SPLT) in totally laparoscopic total gastrectomy (TLTG).

METHODSOne hundred patients with gastric cancer received TLTG-SPLT at General Surgery Department of Huashan Hospital (Fudan University) from June 2014 to January 2017(SPLT group). The clinicopathologic characteristics, surgical and postoperative outcomes were collected retrospectively and compared with the conventional group undergoing TLTG plus overlap or functional end-to-end anastomosis from October 2013 to December 2015. D2 lymph node dissection was regularly performed for all the patients. In SPLT group, a sterile hemp rope was held to ligate and drag down the esophagus to maintain "self-pulling" after the duodenum was transected by the first stapler, allowing the detachment of the posterior mediastinum. Then a hole 2-3 cm above the ligature rope was made on the right-posterior wall of the esophagus. When the mesenteric tension was checked, another hole was made at the anti-mesenteric border of the jejunum 20 cm distal to the ligament of Treitz. A side-to-side esophagojejunostomy (E-J) was then performed between the right-posterior wall of esophagus and the anti-mesenteric wall of the jejunum with the second linear stapler, forming an entry hole. The "latter transection" was applied with the third stapler inserted from the assistant's Trocar, which facilitated the esophagus and the afferent loop jejunum to be simultaneously transected above the level of the entry hole. After that, a side-to-side jejunojejunostomy(J-J) with another 2 staplers was carried out between the afferent loop stump and the Roux limb 40 cm below E-J, in which the E-J entry hole could also work as the entrance for the stapler. The TLTG-SPLT was therefore completed and the specimen was removed through the incision from the umbilical Trocar site.

RESULTSThere were 66 male and 34 female patients in the SPLT group with median age of 64 years. The clinicopathologic baseline data of two groups were comparable(all P>0.05). All the patients underwent operations successfully, and none was converted to open surgery. No positive margin was found in either group. Mean operation duration was (178.2±35.9) minute in SPLT group, including (22.9±7.1) minute of reconstruction, which both were significantly shorter than those in conventional group [(204.4±55.8) minute, P=0.003; (30.5±7.2) minute, P=0.000]. Less blood loss [(74.3±72.5) ml vs. (104.2±71.6) ml, P=0.017] and earlier time to the first flatus [(1.9±1.6) days vs. (2.7±1.3) days, P=0.001] were observed in SPLT group. There were no significant differences in postoperative hospital stay and pathological findings between the two groups(all P>0.05). Postoperative operation-associated complications were found in 7 cases of SPLT group. Of these 7 patients, 1 case developed gastrointestinal bleeding, 3 pancreatic leakage, 2 chyle leakage, who all were discovered within postoperative 1 week and were cured by conservative treatment, while the other 1 case developed anastomotic fistula complicated with peritoneal infection who received laparoscopic exploration and peritoneal scavenge and drainage, then discharged 34 days later. Six patients in conventional group developed postoperative operation-associated complications, including 1 case of anastomotic bleeding, 3 cases of pancreatic leakage, 1 case of chyle leakage and 1 case of peritoneal infection. Morbidity of postoperative operation-associated complication was not significantly different between two groups [7.0%(7/100) vs. 11.5%(6/52), χ=0.414, P=0.520]. Fifty patients from two groups underwent endoscopic examination at postoperative 6-month and 12-month, and no obvious anastomotic stenosis and esophageal reflux were observed.

CONCLUSIONSPLT is a safe procedure with feasibibility in intracorporeal esophagojejunostomy.