Human milk improves neurodevelopment for preterm infants, but relationships between human milk and neurodevelopment for infants with critical CHD are unknown. We aimed to (1) explore associations between human milk/direct breastfeeding and neurodevelopment at 1-year and 2-year follow-up and (2) describe patterns of human milk (maternal, donor) and commercial formula during hospitalisation in the first year of life.This retrospective cohort study included infants who underwent surgery for CHD < 6 months old. The primary outcome was neurodevelopment via Bayley Scales of Infant Development-IV. Analysis included adjusted linear regression for associations between exclusive human milk while inpatient during the first 6 months or any direct breastfeeding while inpatient during the first year of life and 1-year Bayley-IV scores. Models were adjusted for race, insurance type, genetic diagnosis, and length of stay.Of 98 eligible infants, 40% followed up at 1 year; 27% at 2 years. There were differences in follow-up related to demographics (race, ethnicity) and social determinants of health (insurance type, distance from clinic). In adjusted models, infants who directly breastfed had 13.18 points higher cognition (95% CI: 0.84-25.53, p = 0.037); 14.04 points higher language (2.55-25.53, p = 0.018); and 15.80 points higher motor scores (3.27-28.34, p = 0.015) at 1-year follow-up. Infants fed exclusive human milk had 12.64 points higher cognition scores (-0.53-25.82, p = 0.059).Future investigation into nutrition and neurodevelopment in the context of critical CHD is warranted. As neurodevelopmental follow-up becomes standard of care in this population, efforts are needed to mitigate disparities in access to this care.