BACKGROUND: Using a large data set, we investigated the impact of the number of resected and involved lymph nodes on overall survival for patients with esophageal cancer. METHODS: From the National Oncology Database, esophageal cancer cases with data available on the total number of resected and involved nodes as well as other variables were evaluated as it relates to overall survival by multivariate analysis using Cox proportional hazards method. Patients with 0, exactly 1 or 1-3 positive nodes were separately studied to determine the association between the number of lymph nodes resected and overall survival. RESULTS: From 1969 to 2002, 3,144 (17%) of 18,390 esophageal cancer cases with complete data were identified. Increasing number of involved nodes predicted poorer outcome (P < 10(-6)). Results from studying patients with 0, exactly 1 or 1-3 positive nodes showed that survival improved with increasing number of nodes analyzed up to 12. Three-tier nodal grouping with increasing risk of death were identified, 0, 1-3, and >or=4 positive nodes (P < 10(-5)). CONCLUSIONS: The pathological assessment of minimal 12 lymph nodes provides sufficient prognostic information. Three-tier nodal grouping is suggested for the next version of AJCC staging system for esophageal cancer.
BACKGROUND: Using a large data set, we investigated the impact of the number of resected and involved lymph nodes on overall survival for patients with esophageal cancer. METHODS: From the National Oncology Database, esophageal cancer cases with data available on the total number of resected and involved nodes as well as other variables were evaluated as it relates to overall survival by multivariate analysis using Cox proportional hazards method. Patients with 0, exactly 1 or 1-3 positive nodes were separately studied to determine the association between the number of lymph nodes resected and overall survival. RESULTS: From 1969 to 2002, 3,144 (17%) of 18,390 esophageal cancer cases with complete data were identified. Increasing number of involved nodes predicted poorer outcome (P < 10(-6)). Results from studying patients with 0, exactly 1 or 1-3 positive nodes showed that survival improved with increasing number of nodes analyzed up to 12. Three-tier nodal grouping with increasing risk of death were identified, 0, 1-3, and >or=4 positive nodes (P < 10(-5)). CONCLUSIONS: The pathological assessment of minimal 12 lymph nodes provides sufficient prognostic information. Three-tier nodal grouping is suggested for the next version of AJCC staging system for esophageal cancer.