Laparoscopic Versus NOTES Rectosigmoid Resection Using Transanal Endoscopic Microsurgery (TEM) in a Swine Survival Model

Patricia Sylla, MD, Min-Chan Kim, MD, Abdulmetin Dursun, MD, Liliana Bordeianou, MD MPH, Ifode Ajari, MD, Sevdenur Cizginer, MD, Brian Turner, MD, Denise W Gee, MD, Mari Mino-Kenudson, MD, William R Brugge, MD, David W Rattner, MD. Massachusetts General Hospital

INTRODUCTION: Our group previously demonstrated the feasibility and safety of NOTES transanal endoscopic rectosigmoid resection in a swine survival study using TEM alone or in combination with transgastric endoscopic assistance to extend the length of colon mobilized. Transanal endoscopic rectosigmoid resection using TEM with stapled coloanal anastomosis (TEM, n=15) was prospectively compared to laparoscopic sigmoid resection with stapled colorectal anastomosis (Lap, n=15) in a swine survival study.

METHODS: NOTES transanal rectosigmoid procedures were performed as previously described and laparoscopic rectosigmoid resection was performed using 4 trocars. All anastomoses were evaluated endoscopically. Animals were survived for 2 weeks and necropsy findings including histological evaluation of the anastomoses were recorded. Operative and postoperative outcomes were evaluated and compared between the groups using Fisher’s Exact and Student’s T tests.

RESULTS: The mean operative time was 83 minutes (range, 55-175) in the TEM vs. 57 minutes (range, 41-105) in the Lap group ( p=0.006). There were no differences in the average length of colon resected transanally in the TEM group (7cm, range 5.5-10.5) vs. transabdominally in the Lap group (7.6 cm, range 6-10.5, p=0.268). No intraoperative organ injury or significant bleeding was noted in either group and all stapled anastomoses were intact. All Lap animals vs. none in the TEM group required narcotics postoperatively in addition to NSAIDS (p<0.001). Postoperatively, TEM animals passed stool at an average of 2 days (range, 1-5) vs. 3.8 days in the Lap group (range, 2-7, p=0.004). One TEM animal developed progressive renal failure from distal urethral obstruction and was sacrificed on postoperative day 13. One Lap animal developed a port-site hernia on postoperative day 1 that required surgical repair. Twelve animals (40%) developed obstructive symptoms (2 TEM vs. 10 Lap animals, p=0.003) requiring either digital or endoscopic dislodgement of a hematoma at the anastomosis on day 3-5 following procedures. At necropsy, TEM animals gained an average 3.1 kg (range, -3 to +10) vs. 2.6 kg (range, -2 to +8) in the Lap group (p=0.6). Staple lines were located an average 3.3 cm from the anal verge in the TEM group (range, 2.5-4.5) vs. 14.6cm (range, 9.5-25) in the Lap group (p<0.001). Histopathology analysis demonstrated healthy granulation tissue at all anastomoses with microabscesses around the staples in 5/15 TEM vs. 5/15 Lap colorectal specimens.

CONCLUSIONS: Relative to the laparoscopic approach, NOTES rectosigmoid resection using TEM is safe in a porcine survival model and associated with faster recovery of bowel function. A high incidence of symptomatic hematoma at the stapled anastomoses was noted in the laparoscopic group which may be related to a more significant ileus.