Human milk feeding and direct breastfeeding improve outcomes for infants with single ventricle congenital heart disease: Propensity score matched analysis of the NPC-QIC registry

BACKGROUND: Infants with single ventricle (SV) congenital heart disease (CHD) undergo three staged surgeries/interventions, with risk for morbidity and mortality. We estimated the effect of human milk (HM) and direct breastfeeding (BF) on outcomes including necrotizing enterocolitis (NEC), infection-related complications, length of stay (LOS), and mortality. METHODS: We analyzed the National Pediatric Cardiology Quality Improvement Collaborative registry (2016–2021), examining HM/BF groups during stage 1 (S1P) and stage 2 (S2P) palliations. We calculated propensity scores for feeding exposures, then fitted Poisson and logistic regression models to compare outcomes between propensity-matched cohorts. RESULTS: Participants included 2491 infants (68 sites). Estimates for all outcomes were better in HM/BF groups. Infants fed exclusive HM before S1P had lower odds of preoperative NEC (OR=0.37, 95% CI=0.17–0.84, p=0.017) and shorter S1P LOS (RR=0.87, 0.78–0.98, p=0.027). During the S1P hospitalization, infants with high HM had lower odds of postoperative NEC (OR=0.28, 0.15–0.50, p<0.001) and sepsis (0.29, 0.13–0.65, p=0.003), and shorter S1P LOS (RR=0.75, 0.66–0.86, p<0.001). At S2P, infants with any HM (0.82, 0.69–0.97, p=0.018) and any BF (0.71, 0.57–0.89, p=0.003) experienced shorter LOS. CONCLUSIONS: Infants with SV CHD in high HM and BF groups experienced multiple significantly better outcomes. Given our findings of improved health, strategies to increase the rates of HM/BF in these patients should be implemented. Future research should replicate these findings with granular feeding data and in broader CHD populations, and should examine mechanisms (eg, HM components; microbiome) by which HM/BF benefits these infants.