Abstract:

Objective To explore the predictive value of platelet and leukocyte parameters for bronchopulmonary dysplasia (BPD) in preterm infants and to construct a logistic regression prediction model. Methods Medical records of preterm infants with gestational age ≤32 weeks admitted to the Neonatology Department and NICU of the First Affiliated Hospital of Xinxiang Medical University and Jincheng People's Hospital from October 2019 to October 2024 were retrospectively analyzed. Infants were divided into BPD and non-BPD groups based on BPD diagnosis. General characteristics, platelet parameters, and leukocyte parameters at 24 hours, 1 week, 2 weeks, and 4 weeks post-birth were compared between groups. Variables with statistical significance were included in multivariable logistic regression analysis to establish a predictive model, which was validated using the receiver operating characteristic (ROC) curve. Results A total of 258 preterm infants were included, and 93 (36%) were included in the BPD group. In comparison of general characteristics, infants in the BPD group had a significantly higher male-to-female ratio, longer length of stay and duration of non-invasive and invasive mechanical ventilation, younger gestational age, lower birth weight and 1-min and 5-min Apgar scores, higher proportion of pulmonary surfactant (PS) use and incidence of comorbidities (e.g., patent ductus arteriosus, retinopathy of prematurity, persistent pulmonary hypertension in newborns) than the non-BPD group (all P<0.05). In comparison of platelet and white cell parameters, infants in the BPD group had significantly lower platelet count at 2 weeks, platelet distribution width at 24 h, lymphocyte count at 24 h, and platelet quality index at 2 weeks, but higher mean platelet volume at 1 week, neutrophil count at 4 weeks, neutrophil to lymphocyte ratio at 24 h, 2 weeks, and 4 weeks, and platelet to lymphocyte ratio at 24 h than the non-BPD group (all P<0.05). A logistic regression model incorporating significant general characteristics and 24-hour platelet/leukocyte parameters was constructed. The model demonstrated an area under the curve (AUC) of 0.783(95%CI: 0.724-0.842) for predicting BPD. Conclusion Platelet and leukocyte parameters at 24 hours post-birth serve as early predictive biomarkers for BPD. The logistic regression model based on these parameters exhibits high predictive accuracy.

Key words: bronchopulmonary dysplasia, premature infants, platelet parameters, white blood cell parameters, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio