Background

There is no evidence demonstrating the feasibility of colorectal natural orifice transluminal endoscopic surgery (NOTES) resection with currently available endoscopic instrumentation.

Objective

This study aimed to evaluate the responsiveness of surgical endoscopists to simulated transanal NOTES sigmoidectomy training.

Design

Participants were trained in simulated NOTES sigmoidectomy by using disposable abdominal trays with tattooed sigmoid cancer in a hybrid simulator.

Setting

Endoscopy simulation laboratory in a university hospital.

Interventions

NOTES sigmoidectomy included 8 steps performed transanally with 2 colonoscopes, endoscopic scissors, and clip applier: (1) colonoscopic viscerotomy with a balloon; (2) retroperitoneal dissection; (3) left ureter identification, inferior mesenteric vessels division; (4) colonoscopy; (5) splenic flexure mobilization; (6) left side of the colon/rectal mobilization; (7) transanal specimen transection; (8) extracorporeal colorectal anastomosis.

Main outcome measurements

Responsiveness was defined as a change in performance over time and assessed comparing baseline testing with unmentored final testing. Content-valid measures included the length of the specimen, the distance of the anastomosis from the anal verge, and the proximal and distal resection margins and operating time (minutes).

Results

Four participants performed 21 resections. Tumor distance from the anal verge was 29.2 cm (range 26-2.5 cm). Operating time overall (127.5 vs 74 minutes, P = .068), viscerotomy (17.5 vs 9 minutes, P = .197), colonoscopy (4.5 vs 3.5 minutes, P = .655), flexure mobilization (19.5 vs 10 minutes, P = .144), colon mobilization (20 vs 14.5 minutes, P = .461), specimen extraction (9.5 vs 8.5 minutes, P = .465), and anastomosis (14 vs 11 minutes, P = .715) times improved.

Limitations

Ceiling effects because of fixed anatomy.

Conclusions

Simulated NOTES sigmoidectomy training affected responsiveness of surgical endoscopists with a 42% reduction in operating time.