The failure of adjuvant therapy to significantly improve the prognosis of patients undergoing esophago-gastrectomy for cancer may be because of poor patient selection. We sought prognostic factors that would identify those patients who could benefit from adjuvant therapy. Data on 15 possible prognostic factors were prospectively collected on 225 patients undergoing esophago-gastrectomy at a single institution, and univariate and multivariate analyzes performed. T, N, M and overall UICC stage, differentiation, involvement of the circumferential resection margin and number of metastatic of lymph nodes were identified as significant prognostic factors by univariate analysis. Multivariate analysis revealed that the completeness of resection (R-category), ratio of metastatic to total nodes resected and the presence of vascular invasion were independently significant prognostic factors. Following R0 or R1 resection, patients with a metastatic to total lymph node ratio > 0.2 and /or the presence of vascular invasion have a poor prognosis, and the effects of adjuvant therapy in these patients should be studied.