The effect of gastrectomy on the subsequent development of esophageal cancer was investigated. Duodenogastroesophageal reflux is thought to be common in patients after distal gastrectomy, but whether this contributes to the development of esophageal cancer in such patients is controversial. We retrospectively evaluated 153 patients who underwent subtotal esophagectomy for thoracic esophageal cancer between January 2002 and July 2005. They were divided into two groups, according to whether or not they had previously undergone a gastrectomy: group 1, comprising 14 patients who had undergone gastrectomy and group 2, comprising 139 patients who had not. Clinical profiles of the patients were obtained from the medical records and the whole resected esophagus was histopathologically examined. The interval between gastrectomy and esophagectomy in group 1 was significantly shorter in the patients who had undergone gastrectomy for gastric cancer (10.5 +/- 4.2 years) than in those who had undergone gastrectomy for a peptic ulcer (28.9 +/- 3.0 years). The interval was also somehow shorter in the patients for whom anastomosis had been performed by Billroth I (21.3 +/- 5.6 years) compared with Billroth II (29.7 +/- 3.2 years), although the difference did not reach its statistical significance (P = 0.11). Moreover, the proportion of lower third tumors in patients after gastrectomy was significantly higher compared with that of the patients with intact stomach. These findings suggest that a history of gastrectomy is associated with more lower-third squamous cell esophageal carcinoma.