The Risk Factors for Discrepancy After Endoscopic Submucosal Dissection of Gastric Category 3 Lesion (Low Grade Dysplasia)

Abstract

Treatment with endoscopic submucosal dissection (ESD) for gastric category 3 lesion (low grade dysplasia, LGD) diagnosed by endoscopic forceps biopsy (EFB) is controversial. The purpose of the present study was to validate the use of ESD for gastric LGD diagnosed by EFB and to evaluate predictable factors for pathologic upgrade diagnosis to category 4 (high grade dysplasia, HGD) or 5 (early gastric cancer, EGC) lesions. Between November 2008 and October 2011, a retrospective analysis of a prospective database was conducted at a single tertiary referral center. A total of 218 ESD procedures were carried out for gastric LGD lesions identified by EFB. The under-diagnosis rate by EFB and the predictable factors for upgrade diagnosis to category 4 or 5 lesions were analyzed. Pathologic discrepancy between EFB and surgical resection was 20.1% (44/218). Thirty eight lesions (17.4%) were diagnosed HGD or EGC by ESD. Gastric HGD lesions were 14 cases (6.4%) and EGC lesions were 24 cases (23 mucosal and 1 submucosal cancer) (11.0%). Multivariate analysis revealed that lesion diameter more than 1cm (OR 3.496 [95% CI 1.375-8.849]), surface redness (OR 6.493 [95% CI 2.557-16.666]) and nodular surface (OR 2.762 [95% CI 1.237-6.172]) were significant risk factors. Endoscopic resection can be recommended if a LGD lesion has risk factors such as a size of 1cm or greater, surface redness or surface nodulariy. For lesions without the risk factors, follow-up endoscopy may be recommended.