Neurodevelopmental Outcome of Extremely Low Birth Weight Infants Randomly Assigned to Restrictive or Liberal Hemoglobin Thresholds for Blood Transfusion

Abstract

Extremely low birth weight infants frequently receive red cell transfusions. We sought to determine whether a restrictive versus liberal hemoglobin transfusion threshold results in differences in death or adverse neurodevelopmental outcomes of extremely low birth weight infants. Extremely low birth weight infants previously enrolled in the Preterm Infants in Need of Transfusion Trial, a randomized, controlled trial of low versus high hemoglobin transfusion thresholds, were followed up at 18 to 21 months' corrected age. Erythrocyte transfusion was determined by an algorithm of low (restrictive) or high (liberal) hemoglobin transfusion thresholds, differing by 10 to 20 g/L and maintained until first hospital discharge. The primary composite outcome was death or the presence of cerebral palsy, cognitive delay, or severe visual or hearing impairment. Of 451 enrolled infants, the primary outcome was available in 430. There was no statistically significant difference in the primary outcome, found in 94 (45%) of 208 in the restrictive group and 82 (38%) of 213 in the liberal group. There were no statistically significant differences in preplanned secondary outcomes. However, the difference in cognitive delay (Mental Development Index score < 70) approached statistical significance. A posthoc analysis with cognitive delay redefined (Mental Development Index score < 85) showed a significant difference favoring the liberal threshold group. Maintaining the hemoglobin of extremely low birth weight infants at these restrictive rather than liberal transfusion thresholds did not result in a statistically significant difference in combined death or severe adverse neurodevelopmental outcome.