Survival study of natural orifice translumenal endoscopic surgery for rectosigmoid resection using transanal endoscopic microsurgery with or without transgastric endoscopic assistance in a swine model

Background: The feasibility of transanal rectosigmoid resection with transanal endoscopic microsurgery (TEM) was previously demonstrated in a swine nonsurvival model in which transgastric endoscopic assistance also was shown to extend the length of colon mobilized transanally. Methods: A 2-week survival study evaluating transanal endoscopic rectosigmoid resection with stapled colorectal anastomosis was conducted with swine using the transanal approach alone (TEM group, n = 10) or a transanal approach combined with transgastric endoscopic assistance (TEM + TG group, n = 10). Gastrotomies were created using a needleknife and balloon dilation, then closed using prototype T-tags. Outcomes were evaluated and compared between the groups using Student's t-test and Fisher's exact test. Results: Relative to the TEM group, the average length of rectosigmoid mobilized in the TEM + TG group was 15.6 versus 10.5 cm (p < 0.0005), the length of the resected specimen was 9 versus 6.2 cm (p < 0.0005), and the mean operative time was 254.5 versus 97.5 min (p < 0.0005). Intraoperatively, no organ injury or major bleeding was noted. Two T-tag misfires occurred during gastrotomy closure and four small staple line defects requiring transanal repair including one in the TEM group and three in the TEM + TG group (p = 0.2). Postoperatively, there was no mortality, and the animals gained an average of 3.4 lb. Two major complications (10%) were identified at necropsy in the TEM + TG group including an intraabdominal abscess and an abdominal wall hematoma related to T-tag misfire. Gastrotomy closure sites and colorectal anastomoses were all grossly healed, with adhesions noted in 60 and 70% and microabscesses in 50 and 20% of the gastrotomy sites and colorectal anastomoses, respectively. Conclusions: Natural orifice translumenal endoscopic surgery (NOTES) for rectosigmoid resection using TEM with or without transgastric endoscopic assistance is feasible and associated with low morbidity in a porcine survival model. Transgastric assistance significantly prolongs the operative time but extends the length of the rectosigmoid mobilized transanally, with a nonsignificant increase in complication rates related to gastrotomy creation.