Higher hemoglobin transfusions do not improve survival without impairment in ELBWIs
Click Here to Manage Email Alerts
Higher hemoglobin thresholds for red-cell transfusions in extremely-low-birth-weight infants with anemia did not improve their survival without neurodevelopmental impairment, researchers reported in the New England Journal of Medicine.
“Limited data suggest that higher hemoglobin thresholds for red-cell transfusions may reduce the risk of cognitive delay among extremely-low-birth-weight infants with anemia,” Haresh M. Kirpalani, MD, MSc, a neonatologist at Children’s Hospital of Philadelphia, and colleagues wrote.
To further assess the effect of this practice, Kirpalani and colleagues enrolled 1,824 infants with a mean birth weight of 756 g within 48 hours after delivery in an open, multicenter trial and randomly assigned them in a 1:1 ratio to receive a red cell transfusion at higher or lower hemoglobin thresholds until 36 weeks of postmenstrual age or until discharge.
Randomization was made by birth weight — those weighing less than 750 g, and those weighing between 750 and 1,000 g.
The authors stated that their main objective was to assess the differences between the groups in the outcome of death or neurodevelopmental impairment at 22 to 26 months of age. Kirpalani said that in a previous study, he and his colleagues found that a primary analysis did not show any benefits for blood transfusions by discharge.
“When we looked at the follow-up for 2 years, in a secondary analysis, it indicated that there may be benefits in maintaining a higher hemoglobin,” Kirpalani told Healio. “But because it was a secondary analysis, we needed to be sure because it has huge implications. That's why we did this study.”
Primary outcome data was available for 1,692 patients (92.8%), the researchers reported. Among 845 infants in the higher transfusion threshold group, 423 (50.1%) either died or survived with neurological impairment compared with the 422 (49.8%) of 847 infants in the lower threshold group (adjusted RR = 1; 95% CI, 0.92-1.1).
“Clinicians do not need to jump straight into giving a blood transfusion, unless the hemoglobin is very different from the ranges of what we studied,” Kirpalani said. “It's quite OK, as far as we can tell, for babies who are not undergoing surgery, because that was a period when we allowed the clinicians to go off an algorithm for the randomized transfusions. But there is no harm that comes to these babies from allowing the hemoglobin to come down, and they don't need necessarily to be topped up with transfusions all the time.”
The authors reported that in sensitivity analyses that accounted for missing primary outcome data, results were “materially unchanged,” no matter if the outcomes were assumed to be events (aRR, 1.01; 95% CI, 0.93-1.09) or nonevents (aRR, 1; 95% CI, 0.91-1.1).
“I think it's very important to get across that the outcome for preterm babies [has] improved over the years since neonatal intensive care started,” Kirpalani said. “But there are many practices which are completely untested. In fact, the parents who helped us to do the study by enabling us to take consent from them and allow a randomized study have helped enormously in the field.”