BACKGROUND: Endoscopic ablation and vagal-sparing esophagectomy offer the potential for reduced morbidity in patients with high-grade dysplasia or early esophageal adenocarcinoma, but neither includes a lymphadenectomy. Although adequate for intramucosal tumors, both are potentially inadequate for patients with submucosal tumor invasion given the high prevalence of nodal metastases with these lesions. Currently there is no test including endoscopic ultrasound that can accurately determine whether a small tumor is confined to the mucosa or has penetrated into the submucosa. The aim of this study was to compare the pathologic depth of invasion by endoscopic mucosal resection with findings and outcome after surgical resection to assess the accuracy and reliability of endoscopic mucosal resection for staging early esophageal adenocarcinoma. METHODS: From 2001 to 2003, 7 patients presented with small, endoscopically visible adenocarcinomas. All underwent endoscopic mucosal resection followed by surgical resection. RESULTS: Analysis of the resected specimens confirmed that the endoscopic mucosal resection had accurately determined the depth of tumor invasion in all patients, and had completely excised the lesion in all but 1 patient (86%). Lymph node dissection was included as part of the resection in 2 patients with submucosal invasion by endoscopic mucosal resection, and a vagal-sparing esophagectomy was used in the 5 patients with only intramucosal tumors. All patients are alive and disease-free at a median follow-up of 7 months. CONCLUSIONS: Endoscopic mucosal resection accurately determines the depth of tumor invasion, and should be used as a staging procedure in patients with early esophageal cancer when therapies that do not include a lymphadenectomy are considered.