Endoscopic mucosal resection with full-thickness closure for difficult polyps: a prospective clinical trial.

Background: Large flat polyps may be more amenable to endoscopic resection if an endoluminal method for full-thickness closure were available.

Objective: Assessment of feasibility of endoluminal full-thickness closure.

Design: Prospective, open-label, interventional study.

Setting: Tertiary referral center.

Patients: Patients referred to surgery for endoscopically unresectable polyps.

Interventions: Endoscopic resection of colon polyps with full-thickness closure of the resection site under laparoscopic observation by using a novel needle and T-tag tissue apposition system.

Main Outcome Measurements: Feasibility and efficacy of tissue apposition with the TAS during procedure and safety at 3-month follow-up.

Results: Nineteen patients referred with unresectable polyps at initial colonoscopy were enrolled. Five patients had successful endoscopic polypectomy and did not require closure of the resulting defect. In 6 patients, the polyp could not be resected endoscopically and surgical resection was performed. Use of the TAS was attempted in 8 and successfully deployed in 7 patients; there was 1 device malfunction. Deployment of the tags through the needle could be performed more safely under laparoscopic guidance when the resection site was visible from the peritoneal cavity. The location of the tags could not be safely determined when the needle was directed toward the retroperitoneal or mesenteric site. There were no long-term complications. Colonoscopy at a 3-month follow-up showed normal healed mucosa with the sutures and anchoring devices in place.

Limitations: Small number of patients, single-center feasibility study without control arm.

Conclusions: Full-thickness endoluminal closure of large polypectomy sites in humans is feasible for selected difficult polyps. Closure should be performed with concurrent laparoscopic guidance to maximize safety. (

Clinical Trial Registration Number: NCT00553436.).