The present study aimed to determine the changes of RDW values in preterm infants born at < 30 weeks’ gestation and the relationship of RDW with BPD. The mean RDW value at birth were 16.2 ± 1.4 % and increased in a week of life. Thereafter, no significant change occurred until a month of life. The RDW values of infants born at < 28 weeks’ GA were lower than those of infants born at 28–29 weeks’ GA. However, the RDW values increased in both groups, did not differ significantly after 1 weeks of life, and remained similar levels for a month. The RDW values did not differ significantly between the BPD groups.
Although it is known that the RDW levels of newborns are higher than those of children or adults due to active erythropoiesis and physiologic reticulocytosis,[26, 27] Determining the normal range of RDW in newborns is challenging because they are changing physiologically during perinatal period and are easily influenced by various conditions including GA. Therefore, the normal range of RDW values of preterm infants have not yet been established. Garofoli et al. reported that RDW and GA were negatively correlated.[28] A study of a multihospital dataset showed similar results that the upper reference was higher in preterm infants than in term and late preterm infants (23% vs 20%), although the authors mentioned of a selection bias in which more sick babies might be included in the preterm group.[26] Tonbul et al. and Desai et al. reported that mean RDW was the highest at 32–34 weeks’ gestation in the both studies and suggested that it was secondary to active erythropoiesis in the third trimester.[27, 29] In the present study of preterm infants with a lower GA than in previous studies, the RDW levels at birth were higher in the infants at 28–29 weeks’ gestation than the infants born at < 28 weeks’ GA. On the other hand, Alur et al. reported no significant difference in the RDW values among preterm infants born at 23–25 weeks, 26–28 weeks, 29–31 weeks’ GA.[30] Looking into the data, we noticed that the RDW in 26–28 weeks’ GA and 29–31 weeks’ GA groups were higher than that in the 23–25 weeks’ GA group, although these were no significant differences. It appears that erythropoiesis increased in the fetal period and peaked at 32–34 weeks’ gestation.
The RDW values in the < 28 weeks’ and ≥ 28 weeks’ GA groups increased during the first week of life, while the gap of RDW level between the groups disappeared. Christensen et al.[26] explained that the increase of RDW in the early days of life in preterm infants was secondary to that induced by a previous RBC transfusion. However, the RDW showed similar trends in the present study, although the infants who received RBC transfusions in the first week of life were excluded. The increase in RDW seems to be partially due to physiologic anemia of preterm accompanied by reticulocytosis. White blood cell counts and C-reactive protein levels were higher in the < 28 weeks’ GA than in the ≥ 28 weeks’ GA group. It seems relevant that early preterm labor was induced by perinatal inflammation/infection.[31]
In adult studies, a number of studies were reported in terms of the role of RDW as indicators of severity or predictors of outcome for various diseases including sepsis, respiratory disease, cardiovascular disease and critical illness.[3, 6, 8, 9, 12–17, 32–35] Even though the mechanism of RDW increase was not fully determined, it was suggested that chronic hypoxia, malnutrition and inflammation cause an increase in RDW values.[4, 5, 36] Hypoxia induces erythropoietin release, which leads to release immature reticulocyte into circulation.[5] Injured RBC by inflammation aggravates the progress of disease through decrease of oxygen transfer to organs and tissues.[4]
RDW values were higher in patients with COPD than healthy people[33] and associated with its severity and outcomes including mortality and readmission rates.[8, 32–34] The pathophysiology of BPD and COPD are similar.[37] Both are resulted of the impairment of alveolarization/vascularization after inflammation and oxidative stress that can present as chronic hypoxia. However, in this study, the RDW values measured in the first month of life were not related to development of BPD. It appeared that other clinical conditions affected the relationship during the period from RDW measurement to BPD presentation. Garofoli et al. reported that the RDW at 1 month of life was higher in the BPD group than in non-BPD group, whereas the RDW within the first 3 days of life did not differ between groups.[28] However, the GA range was too broad and other contributing factors were not considered in the previous study.
There were some limitations in this study. First, it could not evaluate whether the infants with an evaluated RDW were likely to develop BPD because the normal RDW range has not yet been established in preterm infants yet. Second, chorioamnionitis was not evaluated as a perinatal factor because pathologic exam data of placenta were not available.
In conclusion, the RDW values at birth were higher in infants born at 28–29 weeks’ GA than in those born at < 28 weeks’ GA. The RDW values of both groups increased in the first week of life and remained similar for the first month of life. Studies about the RDW values for preterm infants are rare and the present study, to our knowledge, is the first to evaluate changes in RDW within the first month of life for preterm infants. The RDW values in the first month of life were not associated with BPD development independent on severity. Thus, the usefulness of RDW as a predictor of BPD remains unknown and requires further studies.