摘要：Numerous strategies for perioperative nutrition therapy for patients undergoing pancreaticoduodenectomy (PD) have been proposed. This systematic review aimed to summarize the current relevant published randomized controlled trials (RCTs) evaluating different nutritional interventions via a traditional network meta-analysis (NMA) and component network meta-analysis (cNMA). EMBASE, MEDLINE, the Cochrane Library, and ClinicalTrials.gov were searched to identify the RCTs. The evaluated nutritional interventions comprised standard postoperative enteral nutrition by feeding tube (Postop-SEN), preoperative enteral feeding (Preop-EN), postoperative immunonutrients (Postop-IM), preoperative oral immunonutrient supplement (Preop-IM), and postoperative total parenteral nutrition (TPN). The primary outcomes were general, infectious, and noninfectious complications; postoperative pancreatic fistula (POPF); and delayed gastric emptying (DGE). The secondary outcomes were mortality and length of hospital stay (LOS). The NMA and cNMA were conducted with a frequentist approach. The results are presented as odds ratios (ORs) and 95% confidence intervals (CIs). Two primary outcomes, infectious complications and POPF, were positively influenced by nutritional interventions. Preop-EN plus Postop-SEN (OR 0.11; 95% CI 0.02~0.72), Preop-IM (OR 0.22; 95% CI 0.08~0.62), and Preop-IM plus Postop-IM (OR 0.11; 95% CI 0.03~0.37) were all demonstrated to be associated with a decrease in infectious complications. Postop-TPN (OR 0.37; 95% CI 0.19~0.71) and Preop-IM plus Postop-IM (OR 0.21; 95% CI 0.06~0.77) were clinically beneficial for the prevention of POPF. While enteral feeding and TPN may decrease infectious complications and POPF, respectively, Preop-IM plus Postop-IM may provide the best clinical benefit for patients undergoing PD, as this approach decreases the incidence of both the aforementioned adverse effects.
