Abstract
BACKGROUND: Nutritional support is pivotal in managing critically ill patients. Enteral nutrition, which preserves intestinal mucosal barrier function and modulates immune-metabolic homeostasis, is the preferred nutritional support strategy. However, the optimal timing for initiating EN remains controversial: some studies advocate early initiation (within 24-48 h), while others suggest delayed initiation (beyond 48 h), resulting in inconsistent clinical practices and conflicting guideline recommendations. This highlights the need for high-quality evidence to clarify the optimal EN initiation window. OBJECTIVE: This network meta-analysis aims to systematically compare the effects of five EN initiation timings on key clinical outcomes in critically ill patients, rank their efficacy, and identify the optimal initiation window, thereby providing evidence-based guidance for clinical practice. METHODS: We conducted a network meta-analysis following the PRISMA-NMA statement, including randomized controlled trials (RCTs) comparing the five EN initiation timings in critically ill adults. Databases (PubMed, Embase, Web of Science, Cochrane Library) were searched from inception to September 2024. Data analysis was performed using Stata 16.0 and RevMan 5.4, with efficacy ranked via the surface under the cumulative ranking curve (SUCRA). RESULTS: Fifteen RCTs were included, involving five EN initiation timings: <24 h, 24-48 h, 48-72 h, 72-96 h, and >96 h. No statistically significant differences in mortality or ICU length of stay were observed between any two timings. However, SUCRA ranking showed that 24-48 h initiation had the highest probability of reducing mortality (SUCRA = 67.0), while >96 h initiation was most likely to shorten ICU length of stay (SUCRA = 75.2). CONCLUSION: Initiating EN beyond 96 h was associated with the shortest ICU stay, while initiating it within 24-48 h may be associated with lower mortality. Based on the current limited evidence, the 24-48 h window may be relatively optimal for improving clinical outcomes in critically ill patients. However, given the small number of included studies and low to moderate evidence quality, this conclusion requires validation in large, high-quality RCTs. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024581390. identifier PROSPERO (CRD42024581390).