Analysis of the Efficacy of Red Blood Cell Transfusion in Very Low Birth Weight Infants and Construction of a Prediction Model

Abstract This study aimed to identify key determinants influencing red blood cell (RBC) transfusion efficacy in very low birth weight (VLBW) infants and develop a clinically applicable predictive model. We conducted a retrospective analysis of 993 eligible very low birth weight (VLBW) infants who received leukoreduced suspended RBC transfusions in the Neonatal Intensive Care Unit of Women and Children’s Hospital of Ningbo University between January 2020 and December 2023. Comprehensive data collection encompassed infant characteristics, clinical parameters, and donor information.Using logistic regression analysis, we identified five independent predictors for ineffective RBC transfusion: birth weight, postnatal age at transfusion, weight during transfusion, transfusion volume, and pre-transfusion hemoglobin levels. These variables were incorporated into a nomogram prediction model developed using R software. The model demonstrated strong predictive performance, with calibration curves showing excellent agreement between predicted and observed outcomes (AUC = 0.799). Our findings provide valuable insights into transfusion efficacy in VLBW infants and offer a practical tool for clinical decision-making.The validated nomogram enables individualized risk assessment, facilitating precision transfusion practices and potentially improving outcomes for this vulnerable patient population. This study represents a significant step toward optimizing transfusion strategies in neonatal care.