Abstract
IMPORTANCE: The optimal type of nutritional support after pancreatoduodenectomy remains controversial. OBJECTIVE: To compare postoperative early enteral nutrition (EEN) with oral nutrition (ON) in patients nutritionally at risk in terms of postoperative morbidity. DESIGN, SETTING, AND PARTICIPANTS: This study was a parallel, open-label, superiority randomized clinical trial that took place at 3 tertiary centers in Switzerland and France. Patients at increased nutritional risk (nutritional risk screening score ≥3) were randomized 1:1 into the EEN or ON groups. Patients in the EEN group received enteral nutrition immediately after surgery via a nasojejunal tube placed intraoperatively and were allowed to have oral food based on the same protocol as the other group. Patients in the ON group were not allowed to have enteral nutrition during hospitalization. In both groups, parenteral nutrition need was standardized. These data were analyzed from February 2025 to April 2025. INTERVENTION: EEN via nasojejunal tube vs ON. MAIN OUTCOMES AND MEASURES: Primary outcome was complications at 90 days postoperatively, defined by the mean comprehensive complication index. Secondary end points were morbidity rates (overall, minor, and major complications), mortality, delayed gastric emptying, pancreatic fistula, postoperative hemorrhage, surgical site infections, infectious complications, pulmonary complications, length of stay, and 90-day readmission. RESULTS: Patients were enrolled from December 15, 2021, through October 8, 2024. A total of 144 patients were included and 142 patients were randomized. After randomization, 24 patients dropped out (17%), leaving 118 patients for analysis (59 in the EEN group and 59 in the ON group). Patients in the EEN group had a lower mean 90-day comprehensive complication index compared with ON patients (mean [SD], 25.5 [21.1] vs 35.8 [25.2]; mean difference 10.3; 95% CI 1.8-18.8; P = .02). At 90 days, overall morbidity rates were 45 of 59 and 51 of 59 in the EEN and ON groups (risk ratio [RR], 1.13; 95% CI, 0.9-1.9; P = .18). No difference was found between both groups regarding specific complications (delayed gastric emptying, pancreatic fistula, postoperative hemorrhage, and surgical site infection). In the EEN group, nasojejunal tube was replaced in 14 patients due to involuntary removal (adverse event). CONCLUSIONS AND RELEVANCE: In this randomized clinical trial, in patients with a nutritional risk screening score of 3 or more, EEN after pancreatoduodenectomy permitted to decrease the burden and incidence of postoperative complications compared with ON. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT05042882.