RESUMO
BACKGROUND: Preterm neonates are likely to require red blood cell (RBC) transfusion, and extremely low birth weight infants almost invariably receive multiple transfusions. Transfusion-reduction strategies can reduce transfusion rates, and might diminish certain adverse outcomes associated with transfusions. MATERIALS AND METHODS: In a single centre, we retrospectively evaluated RBC transfusion rates among preterm infants ≤32 weeks' gestational age (GA), over a 6-year period before and after adopting national transfusion-reduction strategies. We compared demographic data, adverse events, and outcomes between transfused vs not-transfused neonates. Univariate logistic regression was used to evaluate associations between dichotomous outcomes and number of transfusions, and day of first transfusion. Multivariate logistic regression evaluated the correlation between dichotomous outcomes and transfusion as an independent risk factor. RESULTS: During the 6 years studied, 181 infants born at ≤32 weeks' GA were admitted to our Neonatal Intensive Care Unit of whom 80 (44%) received at least one RBC transfusion. The transfusion rate tended downwards after adopting transfusion-reduction strategies, reaching 31% in 2018. The reduction was largely due to a marked fall in transfusions of neonates born at 29-32 weeks' GA (p<0.001). The number of transfusions received correlated with odds of having intraventricular haemorrhage (IVH) (OR=1.9; 95% CI: 1.3-2.7; p=0.0001) and the duration of oxygen supplementation (rho=0.51; 95% CI: 0.33-0.66; p≤0.0001). In multivariate logistic regression analysis, transfusion was an independent risk factor for IVH (adjusted OR=7.38; 95% CI: 2.24-24.30; p=0.0001). DISCUSSION: The application of national, standardised transfusion-reduction strategies was associated with a lower transfusion rate in neonates born at 29-32 weeks' GA, but was less effective among neonates ≤28 weeks, in whom transfusions appeared to be an independent risk factor for severe IVH.