To evaluate whether differences in early nutritional support provided to extremely premature infants mediate the effect of critical illness on later outcomes, we examined whether nutritional support provided to "more critically ill" infants differs from that provided to "less critically ill" infants during the initial weeks of life, and if, after controlling for critical illness, that difference is associated with growth and rates of adverse outcomes. One thousand three hundred sixty-six participants in the NICHD Neonatal Research Network parenteral glutamine supplementation randomized controlled trial who were alive on day of life 7 were stratified by whether they received mechanical ventilation for the first 7 d of life. Compared with more critically ill infants, less critically ill infants received significantly more total nutritional support during each of the first 3 wk of life, had significantly faster growth velocities, less moderate/severe bronchopulmonary dysplasia, less late-onset sepsis, less death, shorter hospital stays, and better neurodevelopmental outcomes at 18-22 mo corrected age. Rates of necrotizing enterocolitis were similar. Adjusted analyses using general linear and logistic regression modeling and a formal mediation framework demonstrated that the influence of critical illness on the risk of adverse outcomes was mediated by total daily energy intake during the first week of life.