BACKGROUND AND AIMS: To clarify optimal therapeutic strategies for early gastric cancers without vestigial remnant or recurrence, we evaluated the benefits of endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) according to tumor size and location. METHODS: From January 2000 to December 2007, a total of 328 gastric lesions were treated using conventional EMR, while 572 lesions were treated by ESD. Patients who underwent surgery on the upper gastrointestinal tract before EMR or ESD were excluded from the study. We compared tumor size, location and rates of complete resection, curative resection, postoperative bleeding, perforation and local recurrence between EMR and ESD according to tumor situation. RESULTS: Overall local complete resection rate (EMR, 64.2%; ESD, 95.1%) and overall curative resection rate (EMR, 59.5%; ESD, 82.7%) were significantly higher in ESD than in EMR. No significant differences were seen in complication rates between EMR and ESD. Local recurrence was detected in 13 lesions (4.0%) of the EMR group during follow up. In contrast, no local recurrence was detected in the ESD group. For lesions 5 mm or less in diameter, complete resection rate in the EMR group was not significantly inferior to that in the ESD group at any location. However, rates were overwhelmingly better in the ESD group than in the EMR group for lesions more than 5 mm in diameter, regardless of location. CONCLUSION: We concluded that lesions exceeding 5 mm in diameter should be treated by ESD, although a high resection rate is obtained also with EMR for lesions of 5 mm or less in diameter.
BACKGROUND AND AIMS: To clarify optimal therapeutic strategies for early gastric cancers without vestigial remnant or recurrence, we evaluated the benefits of endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) according to tumor size and location. METHODS: From January 2000 to December 2007, a total of 328 gastric lesions were treated using conventional EMR, while 572 lesions were treated by ESD. Patients who underwent surgery on the upper gastrointestinal tract before EMR or ESD were excluded from the study. We compared tumor size, location and rates of complete resection, curative resection, postoperative bleeding, perforation and local recurrence between EMR and ESD according to tumor situation. RESULTS: Overall local complete resection rate (EMR, 64.2%; ESD, 95.1%) and overall curative resection rate (EMR, 59.5%; ESD, 82.7%) were significantly higher in ESD than in EMR. No significant differences were seen in complication rates between EMR and ESD. Local recurrence was detected in 13 lesions (4.0%) of the EMR group during follow up. In contrast, no local recurrence was detected in the ESD group. For lesions 5 mm or less in diameter, complete resection rate in the EMR group was not significantly inferior to that in the ESD group at any location. However, rates were overwhelmingly better in the ESD group than in the EMR group for lesions more than 5 mm in diameter, regardless of location. CONCLUSION: We concluded that lesions exceeding 5 mm in diameter should be treated by ESD, although a high resection rate is obtained also with EMR for lesions of 5 mm or less in diameter.