Retrospective cohort study of neonatal blood transfusion in China

Blood transfusion therapy is extremely important for certain neonatal diseases, but the threshold for neonatal blood transfusion is not the same in different countries. Until now, clinical studies to determine the suitable threshold for newborns in China are lacking. Therefore, it is of high importance to establish a multi-center cohort study to explore appropriate transfusion thresholds for newborns in China.

This retrospective cohort study investigated neonatal blood transfusion therapy administered from January 1, 2017 to June 30, 2018, with the aim of evaluating the effect of restricted and nonrestricted blood transfusion on neonatal health. The subjects were enrolled in 46 hospitals in China. A total of 5669 neonatal cases were included in the study. Clinical diagnosis and transfusion treatment of these neonates were collected and the data were retrospectively analyzed. The neonates were followed up 1 week and 1 month after leaving the hospital. The newborns' and their mothers' data were collected containing 280 variables in the database. The primary outcome of the study was mortality, and the secondary outcomes were complications, hospital stays, NICU hospital stays and hospital costs.

Results from the < 1500 g group showed that there was a higher mortality rate in the restricted transfusion group (11.41%) when compared with the non-restricted transfusion group (5.12%) (P = 0.000). Among the secondary outcomes, the restricted transfusion group had fewer costs. Results from the 1500-2500 g group showed that the mortality rates of the restricted and non-restricted transfusion groups were 3.53% and 4.71%, respectively, however there was no statistical significance between the two groups (P = 0.345). Among the secondary outcomes, the restricted transfusion group had fewer hospital stays, NICU hospital stays and hospital costs. The incidence of necrotizing enterocolitis was lower in the restricted transfusion group (OR, 2.626; 95% confidence interval [CI], 1.445 to 4.773; P = 0.003). The results from the ≥ 2500 g restricted transfusion group suggested that the mortality rate of (3.02%) was significantly lower than that of non-restricted transfusion group (9.55%) (P = 0.000). Among the secondary outcomes, the restricted transfusion group had fewer hospital stays and hospital costs. The incidence of retinopathy of prematurity was lower in the restricted transfusion group (OR, 4.624; 95% confidence interval [CI], 2.32 to 9.216; P = 0.000).

Current transfusion protocols for newborns weighing less than 1500 g may be inappropriate and lead to higher mortality. The current transfusion threshold performed better for the other two weight groups.