We retrospectively examined esophageal cancer patients who received enteral nutrition (EN) to clarify the validity of early EN compared with delayed EN. A total of 103 patients who underwent transthoracic esophagectomy with three-field lymphadenectomy for esophageal cancer were entered. Patients were divided into two groups; Group E received EN within postoperative day 3, and Group L received EN after postoperative day 3. The clinical factors such as days for first fecal passage, the dose of postoperative albumin infusion, differences of serum albumin value between pre- and postoperation, duration of systematic inflammatory response syndrome (SIRS), incidence of postoperative infectious complication, and use of total parenteral nutrition (TPN) were compared between the groups. The statistical analyses were performed using Mann-Whitney U test and Chi square test. The statistical significance was defined as p < 0.05. Group E showed fewer days for the first fecal passage ( p < 0.01), lesser dose of postoperative albumin infusion ( p < 0.01), less use of TPN ( p < 0.01), and shorter duration of SIRS ( p < 0.01). However, there was no significant difference in postoperative complications between the two groups. Early EN started within 3 days after esophagectomy. It is safe and valid for reduction of albumin infusion and TPN, for promoting early recovery of intestinal movement, and for early recovery from systemic inflammation.