Controversy exists regarding the potential influence of anemia and blood transfusions on the rate of retinopathy of prematurity (ROP) in premature infants. A prospective, randomized, masked trial was performed to determine the influence of red blood cell transfusion protocol on ROP incidence and severity in a population of high-risk infants.


A total of 50 infants with birth weights <1251 g were divided randomly into two groups beginning on day of life 29. Group 1 (n = 24) received red cell transfusions during the 6-week study period, only if certain symptom-based guidelines were met. Group 2 (n = 26) received red cell transfusions to maintain the hematocrit level above 40% for the entire 6 weeks. Infants were monitored for ROP, growth, and associated morbidity. Serial measurements of serum glucose, lactate, ferritin, total iron-binding capacity, and iron were performed.


ROP occurred in 83% of infants in group 1, and 73% of infants in group 2. There were no statistically significant differences in ROP severity, intraventricular hemorrhage, bronchopulmonary dysplasia, necrotizing enterocolitis, or any of the laboratory values except hemoglobin (10.8 vs 13.2 g/dL) and hematocrit (33.9% vs 41.8%) between the groups. Combining data from both groups, there was no association between hemoglobin or hematocrit ratios and ROP incidence or severity.


A transfusion policy aimed at limiting the amount of blood given to premature infants (symptom-based) during the neonatal period does not impart a significantly different risk for ROP or other associated conditions, than does a policy in which transfusions are given more liberally for replacement purposes.

You do not currently have access to this content.