Abstract Purpose: Preventing harmful hyperglycemia is important in critical illness. However, insulin therapy increases the risk of hypoglycemia. In patients with diabetes, isomaltulose-based enteral formula (IF) feeding has been shown to reduce glycemia. This randomized controlled crossover study was conducted to determine whether IF feeding improves glycemia in postoperative critically ill patients. Material and Methods: Eight patients who developed hyperglycemia (N150 mg/dL) after esophagectomy were included. Patients were randomized to either the IF or the standard feeding formula (SF) arm. After 16 hours of administration of randomized formula and 8 hours of washout, patients crossed over to the other formula for the next 16 hours. Continuous glucose measurement using STG-22 (Nikkiso, Tokyo, Japan) was performed during the trial. Results: Maximum blood glucose concentration was 181 mg/dL with IF, significantly lower than the 206 mg/dL with SF (P = .001). Mean glycemia during feeding periods was 162 mg/dL with IF, significantly lower than the 176 mg/dL with SF (P = .0001). Seven (87.5 ) patients taking SF exceeded 180 mg/dL compared with 3 (37.5 ) patients taking IF (P = .005). This effect was seen without any risk of hypoglycemia and complication. Conclusions: Isomaltulose-based enteral formula might be useful for safer glycemic control in postoperative critically ill patients. Further study to determine clinical benefit of IF feeding is justified.