OBJECTIVETo evaluate the feasibility and safety of intracorporeal Roux-en-Y esophagojejunostomy via the transoral anvil(OrVil) by mini-laparotomy anastomosis during laparoscopic total gastrectomy (LTG) for gastric cancer.

METHODSFrom March 2010 to December 2016, 414 consecutive gastric adenocarcinoma patients underwent either intracorporeal Roux-en-Y esophagojejunostomy (n=43) via the OrVil or extracorporeal circular anastomosis (n=371) via auxiliary incision during LTG. After generating propensity scores with six covariates, including gender, age, body mass index (BMI), neoadjuvant chemotherapy, tumor location, and tumor size, 43 patients undergoing OrVil method (OrVil group) were matched with 43 patients undergoing extracorporeal circular anastomosis approach (extracorporeal anastomosis group). Operation-associated parameters and safety were compared between the two groups.

RESULTSBoth groups were balanced regarding baseline variables (all P > 0.05). The total operative time [(235.6±49.8) minutes vs. (221.1±46.5) minutes, t=1.397, P=0.166] and anvil insertion time [(10.0±3.2) minutes vs. (10.6±4.5) minutes, t=-0.671, P=0.504] were not significantly different between the two group, whereas the duration of reconstruction and the mean length of minilaparotomy [(48.3±12.0) minutes vs. (55.9±12.3) minutes, t=-2.899, P=0.005; (5.6±0.6) cm vs. (8.1±2.2) cm, t=-7.118, P=0.001] in the OrVil group were significantly shorter. The number of retrieved lymph nodes, mean blood loss and proximal resection margin were not significantly different between two groups (all P > 0.05). As a whole, OrVil group had advantages over extracorporeal anastomosis group during the postoperative recovery course. The time to liquid intake [(3.7±1.8) days vs. (6.2±7.2) days, t=-2.236, P=0.030], time to fluid diet [(4.8±2.3) days vs. (7.2±7.1) days, t=-2.013, P=0.048], and time to semi-fluid diet [(6.7±2.9) days vs. (10.2±9.6) days, t=-2.245, P=0.029] were significantly shorter in the OrVil group. The first ambulatory time, time to first flatus and length of hospital stay were not significantly different between two groups(all P>0.05). The morbidity of intraoperative complication [7.0%(3/43) vs. 4.7%(2/43), χ²=0.000, P=1.000] and postoperative complication [30.2%(13/43) vs. 20.9%(9/43), χ²=1.484, P=0.223], and even the distribution of severity (χ²=0.013, P=0.990) between the two groups were not significantly different. The incidence of anastomotic leakage (AL) was 9.3% (4/43) and 18.6% (8/43) in the OrVil group and extracorporeal anastomosis group respectively without significant difference (χ²=1.550, P=0.213). Multivariate analysis showed that the OrVil anastomosis was not a risk factor of AL(HR=0.663, 95%CI:0.120-3.674, P=0.638).

CONCLUSIONSIntracorporeal esophagojejunostomy using the OrVil system is more minimally invasive and convenient to operate without increasing the risk of operation-related complication. Thus it may be a potential safe approach to optimize the reconstruction for LTG.