Red Blood Cell Transfusion Characteristics and Morbidity or Mortality in Very-Low-Birth-Weight Infants

Importance: Few studies have evaluated whether modifiable aspects of red blood cell (RBC) transfusions are associated with recipient outcomes in very-low-birth-weight (VLBW) infants. Objective: To determine whether blood donor, RBC modifications and storage, or transfusion thresholds and characteristics are associated with serious adverse outcomes in VLBW infants undergoing transfusion. Design, Setting, and Participants: Transfusion in Preterm Infants was a prospective birth cohort study that recruited VLBW infants (<1500 g at birth) between April 1, 2019, and December 31, 2023, at 5 university-affiliated and 3 community birth hospitals in the US. Electronic medical record data linking blood donor and component data to infants were obtained and linked with Vermont Oxford Network outcome data, with additional outcome review by site. The analysis was completed in January 2026. Exposures: RBC transfusion and transfusion characteristics, evaluated up to the first outcome event. Main Outcomes and Measures: The primary outcome was a composite outcome of severe intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), late-onset sepsis, severe bronchopulmonary dysplasia (BPD), retinopathy of prematurity, or death (and secondary individual outcomes), with follow-up through 90 days or death. Blood donor characteristics (sex, age, and hemoglobin), anticoagulant preservative solution and/or unit characteristics, transfused volumes, transfusion number, and infants' pretransfusion hemoglobin values were evaluated using multivariable generalized estimating equation regression models to account for correlation within hospital, adjusted for illness severity. Results: The study enrolled 2605 VLBW infants, and 1283 (586 [45.7%] female; 713 [55.6%] <27 weeks' gestational age) received RBC transfusion. Median pretransfusion hemoglobin level was associated with higher odds of the composite outcome (odds ratio [OR] per 1 g/dL increase, 1.15; 95% CI, 1.07-1.25; P < .001). In contrast, use of additive solution (AS)-1 or AS-5 vs the reference of citrate phosphate dextrose adenine or citrate phosphate dextrose as the anticoagulant preservative solution was associated with lower odds of the composite outcome (OR, 0.72; 95% CI, 0.53-0.97; P = .03). There were no significant associations with other examined modifiable factors and the composite outcome. Among secondary outcomes, anticoagulant preservative solution was associated with lower risk of BPD (AS-1 and AS-5, OR, 0.48; 95% CI, 0.33-0.69; AS-3, OR, 0.65; 95% CI, 0.56-0.77) and higher risk of NEC (AS-3, hazard ratio [HR], 5.33; 95% CI, 1.22-23.29). Transfusion dose was associated with higher risk of mortality (HR per 5 mL/kg, 1.25; 95% CI, 1.17-1.35), and donor age was associated with a lower risk of mortality (HR for age ≥60 years, 0.63; 95% CI, 0.44-0.92). Shorter postirradiation storage duration (<1 day) was associated with lower risk of NEC (HR, 0.44; 95% CI, 0.22-0.89). Female donor sex was associated with lower risk of IVH (HR, 0.60; 95% CI, 0.36-0.99). Conclusions and Relevance: In this cohort study of VLBW infants, pretransfusion hemoglobin and anticoagulant preservative solution were associated with a composite of morbidities and mortality, unlike other modifiable blood banking practices. For individual outcomes, select donor and blood banking factors were identified that may be modifiable targets for further evaluation.