Results of Endoscopic Mucosal Resection for Early Gastric Cancer: Factors Influencing Complete Resection

Kim, Beom Jin; Kim, Jae J; Kim, Ji Hyang; Yang, Jeong Chae; Lee, Jun Haeng; Shim, Sang Kun; Kim, Young Ho; Rhee, Poong Lyul; Paik, Seung Woon; Rhee, Jong Chul; Park, Chul Keun

Abstract

BACKGROUND/AIMS
The aim of the study is to evaluate the results of endoscopic mucosal resection (EMR) for early gastric cancer (EGC) and to investigate the factors with influence the complete resection.
METHODS
We retrospectively analyzed 109 lesions from 108 patients with EGC treated by EMR at Samsung medical center from November 1994 to June 2003. We compared completely resected group with incompletely resected group with regards to size, location, histologic types before and after EMR, methods of procedure, and complication.
RESULTS
The mean size of lesions was 11.3+/-6.5 mm. Eighty two of them were located in the antrum and angle, twenty six in the body, and one in the cardia of stomach. Endoscopically elevated lesions (type I, IIa) were 52 cases and depressed lesions (type IIc) were 40 cases. Histologically curative resection was done in 74 of 109 cases (67.9%). All but one cases have been observed without recurrence for a mean period of 11.1 months. Histologically incomplete resection in 35 cases included 9 positive cancer cell in resection margin, 25 submucosal cancer infiltration, 2 reconstruction failure, 1 lymphatic involvement and 1 signet ring cell type cancer. Complications related to EMR included 9 cases of bleeding and 3 cases of perforation. In comparison of two groups, complete resection rate was significantly higher when tumor was located in the antrum or angle than body or cardia of stomach (p=0.006).
CONCLUSIONS
Our results show that EMR is one effective curative treatment modality in highly selected patient with EGC and location of lesion is an important factor influencing the success of complete resection.