<![CDATA[Introduction: Early enteral nutrition (EEN) is recommended in critical care, yet its association with mortality remains debated due to heterogeneous evidence across populations and comparators. This systematic review aims to evaluate the association between EEN and mortality in adult ICU patients, identify condition-specific effects, and determine factors modifying this relationship. Methods: A systematic review was conducted on 80 studies including randomized controlled trials, etc examining EEN (initiation within 24-48 hours of ICU admission) compared with delayed EN, parenteral nutrition (PN), or standard care in adult ICU patients. Mortality outcomes were extracted alongside population characteristics, comparators, and effect modifiers. Results: In general ICU populations, large RCTs (CALORIES, n=2400; NUTRIREA-2, n=2410) found no mortality difference between early EN and early PN (RR 0.97, p=0.57; 37% vs 35%, p=0.33). However, meta-analyses comparing EEN specifically against delayed EN demonstrated significant mortality reduction (OR 0.45, 95% CI 0.21-0.95, p=0.038). Condition-specific benefits emerged: sepsis (OR 0.59, 95% CI 0.37-0.94, p=0.03), burns (OR 0.36, 95% CI 0.18-0.72, p=0.003), traumatic brain injury (RR 0.35, 95% CI 0.24-0.50), and severe acute pancreatitis (aOR 0.44, 95% CI 0.20-0.96). Hemodynamic status significantly modified effects: EEN benefited patients with transient or low-to-moderate vasopressor requirements but not those with persistent severe shock (OR 1.28, p=0.485). Methodological quality assessment revealed that positive findings were predominantly from smaller, higher-bias trials. Discussion: The mortality association with EEN is comparator-dependent and population-specific. When compared with early PN, EEN shows no survival advantage, indicating the route of early nutrition is less critical than timely nutritional delivery. Conversely, EEN reduces mortality compared with delayed EN, supporting early initiation over delayed strategies. Biological plausibility is supported by immune modulation (Th17/Treg regulation), reduced infectious complications, and gut barrier preservation. Hemodynamic stability represents a critical treatment effect modifier, with benefits confined to resolving or moderate shock. Conclusion: EEN is associated with reduced mortality compared to delayed EN in specific ICU populations including sepsis, burns, traumatic brain injury, and severe acute pancreatitis, but not when compared with early PN in general ICU populations. Hemodynamic status should guide clinical decision-making. Future research should prioritize large, low-bias trials in stratified populations.]]>
