Background: Preoperative weight loss has been shown to be a prognostic factor for many cancers. However, whether preoperative weight loss has clinical significance in patients with esophageal cancer is still controversial. Methods: A total of 2,174 Chinese patients underwent radical resection of esophageal cancer from 2000 to 2008 were included in our study. Patients were divided into two group: no weight loss (-) and weight loss (+), according to whether they had weight loss compared with their usual weight at diagnosis. The influence of preoperative weight loss on disease-free survival (DFS) and overall survival (OS) was estimated using the Kaplan-Meier method and Cox proportional hazard models. Results: weight loss (+) was significantly associated with age (P=0.001), alcoholism (P<0.001), tumor location (P=0.003), pT category (P=0.003), pN category (P=0.001). Patients of group weight loss (+) had significantly poorer DFS (Mean: 63.3 months (m) vs 76.8 m, P<0.001) and OS (67.4 m vs 83.3 m, P<0.001) than the no weight loss (-) group. In the final multivariate survival analysis with adjustment for covariates, we found that the weight loss (+) group had a 19% higher risk of death (HR=1.19, 95%CI: 1.07-1.33, P=0.002) and had a 13% higher risk of disease progression (HR=1.13, 95%CI: 1.01-1.25, P=0.027), respectively, than the no weight loss (-) group. Subgroup analysis indicated that the association with preoperative weight loss and better DFS or OS was observed in patients with esophageal squamous cell carcinoma (ESCC) and early pathological stage (I-II). Conclusion: Preoperative weight loss is associated with shorter OS and DFS, which means poor postoperative prognosis in esophageal cancer patients.