OBJECTIVE The elemental diet (ED) is a formula to support nutritional status without increasing chylous burden. This study evaluates the efficacy of early ED feeding after pancreatoduodenectomy (PD). METHODS A prospective phase II study of consecutive patients who underwent PD with early ED feeding was conducted. Patient backgrounds, surgical outcomes, and ED feeding tolerability were compared with a historical cohort of 74 PD patients with early enteral feeding of a low residue diet (LRD). RESULTS The ED group comprised 104 patients. No patient in the ED group discontinued enteral feeding because of chylous ascites (CAs), whereas 17.6% of the LRD group experienced refractory CAs that disrupted further enteral feeding. The CAs rate was significantly decreased in the ED group compared with the LRD group (3.9% and 48.7%, respectively; P < 0.001). There was no significant difference in the incidence of major complications (ED: 17.3%, LRD: 18.9%; P = 0.844). Postoperative prognostic nutritional index was similar between the 2 groups ( P = 0.764). In multivariate analysis, enteral feeding formula, and sex were independent risk factors for CAs (LRD: P < 0.001, odds ratio, 22.87; female: P = 0.019, odds ratio, 2.78). CONCLUSIONS An ED reduces postoperative CAs of patients undergoing PD in the setting of early enteral feeding.