Decreased hazard of necrotizing enterocolitis in preterm neonates receiving red cell transfusions

Background and objectives: Many observational studies reporting a temporal association between red cell transfusions (RBCTs) and necrotizing enterocolitis (NEC) in preterm infants fail to take into account RBCTs in infants without NEC. The objective of this study was to investigate the association between RBCTs and NEC in an analytical retrospective cohort study with minimization of selection and measurement bias and controlling for clinical covariates.Methods: Inborn preterm infants [23-32 weeks gestational age (GA)] without major congenital anomalies were eligible. Association of RBCT and modified Bell's Stage 2A NEC was explored using bivariate analyses and verified using multivariable Cox regression.Results: Of 627 eligible infants, 305 neither received RBCT nor developed NEC and 12 developed NEC prior to RBCT. Of 310 infants with RBCT, 27 developed NEC. Compared to infants without NEC, infants with NEC received significantly lower number of RBCTs before diagnosis of NEC (p=0.000). On multivariable Cox regression controlling for clinical covariates, dichotomous RBCT exposure was associated with 60% reduced hazard for NEC.Conclusions: RBCT exposure was associated with decreased hazards for NEC in preterm infants in this study; factors previously reported to be associated with NEC remained statistically significant predictors.