Early vitamin AD supplementation reduces transfusion needs in preterm infants: a retrospective cohort study

Introduction Anemia of prematurity (AOP) is a common condition, often necessitating transfusion therapy. Although vitamin AD deficiency has been linked to anemia pathogenesis, the effect of supplementation timing on transfusion needs remains unclear. This study investigates the association between the initiation timing of vitamin AD supplementation and the number of transfusions received in preterm infants. Methods In this retrospective cohort study, we included 226 preterm infants (gestational age <37 weeks) who received both blood transfusions and vitamin AD supplementation in a neonatal intensive care unit (2018–2022). The association was assessed using Spearman correlation. Segmented Poisson regression compared transfusion incidence rates before and after supplementation, controlling for hospitalization timeline. A conditional Cox model (Anderson-Gill) evaluated the impact on subsequent transfusion risk, adjusting for covariates. Patients were stratified by supplementation timing (≤14 vs. >14 days), with propensity score matching applied to balance groups. Independent factors were identified using generalized linear models. Results The timing of vitamin AD initiation was positively correlated with transfusion number [Spearman’s ρ = 0.200(0.068–0.326), p = 0.003], an association that remained significant in subgroups including infants 28–37 weeks and those without complications. Multivariable analysis identified supplementation timing as an independent risk factor for transfusion frequency [ β = 0.016(0.007–0.025), p < 0.001]. Segmented Poisson regression showed a significantly lower incidence rate ratio after supplementation [IRR = 0.139(0.098–0.199)]. The conditional Cox model revealed no immediate change in risk [ HR = 10.858(0.725–162.663), p = 0.084] but a significant time-dependent protective effect [ HR per log(day+1) = 0.381(0.160–0.911), p = 0.03], with the hazard ratio becoming protective by day 5 and reaching 0.32 (68% risk reduction) by day 28. After propensity score matching, no significant difference in transfusion frequency was found between the ≤14-day and >14-day groups [median (IQR): 2 (1.5–3) vs. 2 (2–4), p = 0.544]. Conclusion This study suggests an association between earlier vitamin AD supplementation and reduced transfusion needs in preterm infants, supported by a time-dependent protective effect. The specific optimal timing requires further prospective investigation.