ABSTRACT
BACKGROUND AND STUDY AIMS: Endoscopic submucosal dissection (ESD) has been reported to produce excellent treatment results for early gastric cancer. In terms of lesions that previously met the criteria for endoscopic mucosal resection (EMR), there is now controversy about which of the two methods is superior, and whether the two methods are comparable. PATIENTS AND METHODS: A total of 177 patients (202 lesions) with early gastric cancer who met the guidelines for EMR and who underwent either EMR or ESD were studied. The rates of en bloc resection, complete resection, local recurrence, and complications were compared between EMR and ESD. RESULTS: The overall en bloc and complete resection rates were lower in patients undergoing EMR than in those undergoing ESD (en bloc: 53.8 % vs. 94.3 %, P < 0.001; complete: 37.5 % vs. 92.6 %, P < 0.001). The overall 5-year recurrence-free rate was lower in the EMR group than in the ESD group (82.5 % vs. 100 %; P < 0.001). However, with regard to the tumor size, the two groups did not differ in en bloc ( P = 1.0) or complete resection rate ( P = 0.8) for tumors < or = 5 mm and in 5-year recurrence-free rate ( P = 0.19) for tumors < or = 10 mm. The mean time required for resection was longer for ESD than for EMR ( P < 0.001). Perforation and bleeding requiring blood transfusion occurred in a small percentage in the ESD group, but in none in the EMR group. CONCLUSION: In this study, EMR was comparable to ESD for the millimeter-sized lesions. We suggest that such small lesions might be well suited to treatment with EMR.