IMPORTANCE Reduced death and neurodevelopmental impairment among infants is a goal of perinatal medicine. OBJECTIVE To assess the association between surgery during the initial hospitalization and death or neurodevelopmental impairment of very low-birth-weight infants. DESIGN, SETTING, AND PARTICIPANTS A retrospective cohort analysiswas conducted of patients enrolled in the National Institute of Child Health and Human Development Neonatal Research Network Generic Database from 1998 through 2009 and evaluated at 18 to 22 months' corrected age. Twenty-two academic neonatal intensive care units participated. Inclusion criteria were birth weight 401 to 1500 g, survival to 12 hours, and availability for follow-up. A total of 12 111 infants were included in analyses. EXPOSURES Surgical procedures; surgery also was classified by expected anesthesia type as major (general anesthesia) or minor (nongeneral anesthesia). MAIN OUTCOMES AND MEASURES Multivariable logistic regression analyses planned a priori were performed for the primary outcome of death or neurodevelopmental impairment and for the secondary outcome of neurodevelopmental impairment among survivors. Multivariable linear regression analyses were performed as planned for the adjusted mean scores of the Mental Developmental Index and Psychomotor Developmental Index of the Bayley Scales of Infant Development, Second Edition, for patients born before 2006. RESULTS A total of 2186 infants underwent major surgery, 784 had minor surgery, and 9141 infants did not undergo surgery. The risk-adjusted odds ratio of death or neurodevelopmental impairment for all surgery patients compared with those who had no surgery was 1.29 (95% CI, 1.08-1.55). For patients who had major surgery compared with those who had no surgery, the risk-adjusted odds ratio of death or neurodevelopmental impairment was 1.52 (95%CI, 1.24-1.87). Patients classified as having minor surgery had no increased adjusted risk. Among survivors who had major surgery compared with those who had no surgery, the adjusted risk of neurodevelopmental impairment was greater and the adjusted mean Bayley scores were lower. CONCLUSIONS AND RELEVANCE Major surgery in very low-birth-weight infants is independently associated with a greater than 50% increased risk of death or neurodevelopmental impairment and of neurodevelopmental impairment at 18 to 22 months' corrected age. The role of general anesthesia is implicated but remains unproven. Copyright © 2014 American Medical Association. All rights reserved.