Colonic Mucosal Resection of Significant (>1 cm) Sessile and Non-Polypoid Colorectal Neoplasms: Long Term Experience of a United States Endoscopy Unit. Tonya Kaltenbach, Anamika Maheshwari, Daniel Ouyang, Shai Friedland, Roy Soetikno Division of Gastroenterology, Veterans Affairs Palo Alto Health Care System, Stanford University School of Medicine, Stanford CA

Abstract

Introduction: Endoscopic mucosal resection (EMR) is a standard technique that is necessary for the resection of sessile and nonpolypoid (flat and depressed) colorectal neoplasms. Historically, the resection of large (>2 cm) sessile or non-polypoid colorectal neoplasms has been considered contraindicated based on concerns of incomplete removal and high incidence of malignancy. To date, there is no published United States outcomes data on colonic mucosal resection. We studied the neoplastic recurrence following EMR of colorectal polyps. Methods: Retrospective review of prospectively collected consecutive data on patients who underwent EMR for a sessile or non-polypoid colorectal neoplasm larger than 1 cm by a single endoscopist in order to assess long-term recurrence rates. A standardized inject and cut EMR technique using saline was performed in all patients. Argon plasma coagulation was applied to residual tissue accordingly. On surveillance colonoscopy, diluted indigo carmine chromoscopy was used to confirm the presence of innominate groves at the prior EMR scar site. Results: We identified 231 significant colorectal lesions: 106 sessile (Is), 118 flat (IIa) and 7 depressed (IIc) that underwent attempted EMR. The mean size of the lesions was 17 mm (range 10 - 70 mm). Histologic findings were 9 with submucosal invasion, 18 with carcinoma in-situ, and 166 with adenoma. 22 lesions were surgically referred due to advanced histology or indeterminant margins, and 32 hyperplastic lesions did not require follow up colonoscopy. 26 lesions did not receive surveillance colonoscopy due to change of care facility, comorbidities, interval death or patient refusal. Thus, on 151 surveillance colonoscopies, we identified remnant tissue at the prior EMR site in 5% (7/151) and 2% (3/151) of cases, at the first and second surveillance colonoscopy, respectively. Conclusion: EMR is a safe and effective treatment in the resection of significant (>1 cm) sessile and non-polypoid colorectal neoplasms in an American population. As non-polypoid lesions become increasingly recognized, this technique should be applied as the first line of therapy.