This study aimed to identify predictive factors and to evaluate appropriate treatments for recurrence of esophageal cancer after curative esophagectomy. About 166 consecutive patients, who underwent curative esophagectomy, were enrolled between April 1994 and March 2003. Recurrence was classified as loco-regional or distant. Logistic regression analysis was used to identify predictive factors for recurrence. Prognostic factors were evaluated by Log-rank test and Cox proportional hazard regression analysis. The disease-specific 5-year survival was 56.8%. Recurrence was observed in 72 patients (43.4%), with 64 of these occurring within 3 years. The number of metastatic lymph nodes and lymphatic invasion independently predicted recurrence. There were significant differences in time to recurrence and survival time between loco-regional, distant recurrence, and combined recurrence. The 5-year survival time in patients with recurrence was 11.9%, and median survival time was 24 months. There was also a significant difference in survival after recurrence between treatment methods (no treatment vs chemo-radiotherapy, p=0.0063; chemotherapy, p=0.0247; and radiotherapy, p<0.0001). Meticulous, long-term follow-up is particularly necessary in patients with four or more metastatic lymph nodes to achieve early detection of recurrence. Randomized controlled trials should be used to develop effective modalities for each recurrence pattern to improve therapeutic outcomes.