Background: Increased intestinal permeability may lead to sepsis in resected upper gastrointestinal (GI) cancer patients. This study sought to determine whether these patients demonstrated increased intestinal permeability and if early postoperative enteral nutrition would alter this result. Methods: Nineteen patients undergoing complete resection of upper GI malignancy were randomized into two groups: the nonfed group received IV crystalloid, and the fed group started enteral nutrition by jejunostomy on postoperative day (POD) 1. Six nonoperative volunteers were controls. The lactulose/mannitol test was performed on PODs 1 and 5. Ten grams of lactulose and 5 g of mannitol were given, and urine was collected for 6 hours. Results: All patients (nonfed, 1.895 ± 0.34; fed, 0.893 ± 0.24) had elevated lactulose/mannitol ratios on POD 1 vs controls (0.262 ± 0.1; p < .008 and p = .05). These elevated levels returned toward control levels in both groups by day 5 (nonfed, 0.533 ± 0.1, p = .06; fed, 0.606 ± 0.12, p = .08). Conclusions: Major upper GI surgery for malignancy resulted in a significant increase in intestinal permeability on POD 1. With or without enteral nutrition, this measure of intestinal permeability returned to normal on POD 5 in well-nourished patients.