A monthly roundup of key articles in other journals

Abstract

Outcomes after implementing restrictive blood transfusion criteria in extremely premature infants. Knee D, Knoop S, Davis AT, et al., J Perinatol 2019;39:1089-97. A restrictive threshold for red blood cell (RBC) transfusion in low-birth-weight infants is associated with decreased rates of secondary morbidities without influencing mortality. Anemia is common in very low birth-weight infants; however, there is a lack of consensus around the hemoglobin (Hb) and hematocrit (Hct) thresholds that should be implemented for RBC transfusion. Studies in adult patients indicate improved safety and efficacy with restrictive thresholds, prompting evaluation in younger populations. Knee et al. performed a retrospective, single-center assessment of the impact of RBC transfusion thresholds on clinical outcomes in very low birth-weight infants (weight <1500 g). A liberal transfusion strategy was employed at a large neonatal intensive care unit in 2012-2013, before switching to a restrictive approach for 2014-2015. Transfusions were guided by the combination of Hct levels and respiratory support. For infants transfused using a liberal approach (n = 384), 15 mL/kg RBCs were transfused for a Hct of 40% or less while on 35% or greater oxygen via ventilator, nasal continuous positive airway pressure (NCPAP), or noninvasive positive pressure ventilation (NIPPV); for an Hct of 30% or less while on any form of oxygen support; or for a Hct of 20% or less. For infants transfused using the restrictive approach (n = 382), 15-20 mL/kg RBCs were transfused for an Hct of 30% or less if on 40% or greater oxygen (mean arterial pressure [MAP] >7); 20 mL/kg RBCs were transfused for an Hct ≤25% on <40% oxygen (MAP ≥7); for an Hct of 20% or less on supplemental oxygen or NCPAP/ventilator/NIPPV, MAP less than 7 and symptomatic; or for an Hct of 18% or less. The primary outcome, all-cause mortality, did not differ between liberal and restrictive approaches (6.3% vs. 6.8%; p = 0.755). Predictably, the number of infants who received transfusions was lower in the restrictive group (39.8% vs. 60.2%; p < 0.001), as was the number of units transfused (2 vs. 4; p < 0.001). The Hct nadir was lower in the restrictive group; however, Hct levels at discharge were comparable. Days of intravenous fluids and total parenteral nutrition were lower for the restrictive group (12 vs. 18 days; p < 0.001). Prespecified secondary outcomes included the rate of sepsis, which was lower in the restrictive group (11.5% vs. 18.0%; p = 0.012), as was a diagnosis of necrotizing enterocolitis within 48 hours of transfusion (2.4% vs. 5.2%; p = 0.039). In infants transfused using the restrictive approach, there were lower rates of periventricular leukomalacia (1.3% vs. 6.3%; p < 0.001) and retinopathy of prematurity (38.0% vs. 46.1%; p = 0.015). These findings support restrictive transfusion protocols in infants. Nonetheless, in contrast to the present study, a small randomized controlled trial showed an increased rate of neurological morbidities using a restrictive threshold. Larger trials currently under way will provide further enlightenment.