Supplemental parenteral nutrition within an enhanced recovery program for open pancreatoduodenectomy for cancer: a pragmatic, multicenter, randomized controlled trial.

Abstract

Background: The role of supplemental parenteral nutrition (SPN) following pancreatoduodenectomy (PD) in the context of an enhanced recovery program is unexplored. This study aimed to determine whether SPN is superior to early oral feeding alone in reducing postoperative complications.

Methods: This pragmatic, multicenter, randomized controlled, trial, across five centers in Italy, enrolled patients aged 18-89 years undergoing open PD for cancer. We excluded patients with an American Society of Anaesthesiology physical status >3 and a preoperative body weight loss of ≥15%. Patients were randomly assigned (1:1) postoperatively to either SPN from day 1 to 5 or no-SPN. All patients were free to begin oral feeding after the operation as desired in the context of a full enhanced recovery after surgery (ERAS) program. The primary outcome was morbidity burden, measured using the comprehensive complication index (CCI). Secondary outcomes included the overall rate of morbidity. Outcomes were assessed up to 90 days postoperatively. Overall, 120 patients per group were required to achieve 80% power and detect at least 30% reduction in the CCI in the SPN group, which was expected to be 23 (median) (interquartile range 21-31). The expected complication rate was 60%, and the type I error rate was set at 5%. Registration at ClinicalTrials.gov (#NCT04438447).

Findings: From June 1, 2022, to December 20, 2023, 405 patients were screened for eligibility and 254 patients were randomly allocated to control (no-SPN; n = 129) or treatment (SPN; n = 125) group. All patients were included in the primary and secondary outcome analysis according to the intent-to-treat principle. The median CCI was 20.9 in both arms (median difference 0 [95% CI: -1.07 to 1.7]). The proportion of patients with at least one complication (CCI >0) was similar in both groups [(29.6% vs 29.2%; risk difference 0.4 (95% CI -11.1 to 7.0)]. The overall 90-day morbidity was 67.4% and 63.2% in the no-SPN arm and SPN arm groups, respectively [risk difference -4.2 (95% CI -16.7 to 8.2)]. In high nutritional risk patients (nutritional risk score ≥3), SPN was not protective against the primary outcome when compared with low-risk patients [OR 1.16 (95% CI 0.71-1.91)].

Interpretation: In an ERAS program emphasizing early postoperative oral feeding, SPN does not affect outcome measures, even in patients at high nutritional risk. However, these results do not apply to severely malnourished patients or with critical comorbidities.

Funding: The Italian Society for Artificial Nutrition and Metabolism (SINPE) and Baxter Italia S.p.A (Rome, Italy).