# Maternal inflammatory biomarkers and neonatal characteristics as predictors of bronchopulmonary dysplasia in preterm infants: a retrospective cohort study

## TL;DR

This study finds that a mother's blood cell ratio before delivery, along with neonatal factors, can predict bronchopulmonary dysplasia in preterm infants.

## Contribution

Identifies maternal platelet-to-white blood cell ratio as a novel predictor of BPD in preterm infants.

## Key findings

- Lower maternal platelet-to-white blood cell ratio independently predicts BPD in preterm infants.
- A nomogram integrating maternal PWR and neonatal factors shows strong discrimination (AUC = 0.866) for BPD prediction.
- Early and late-onset sepsis and mechanical ventilation duration are significant predictors of BPD.

## Abstract

Maternal peripartum inflammation may contribute to bronchopulmonary dysplasia (BPD) in preterm infants, yet readily available hematologic predictors are underexplored. We evaluated whether the maternal platelet-to-white blood cell ratio (PWR) within 24 h before delivery, together with neonatal clinical variables, predicts BPD.

We conducted a retrospective cohort study at a tertiary NICU in Southwest China, including all eligible mother–infant pairs with preterm birth (<37 weeks) between June 2020 and June 2024. BPD was defined per the 2018 NICHD criteria. Maternal demographics, obstetric complications, hematology (WBC, PLT, PWR, NLR, and PLR), and neonatal characteristics (gestational age, birth weight, sex, Apgar score, SGA, sepsis, and respiratory support) were abstracted from electronic records. Variables with p < 0.10 in univariate analysis were entered into multivariable logistic regression. Model performance was assessed by AUC, Hosmer–Lemeshow test, bootstrap calibration, and a nomogram was constructed.

Among 345 pairs, 117 infants (33.9%) developed BPD. Mothers of BPD infants had higher WBC and lower PLT, yielding a lower PWR (median 18.9 vs. 23.1, p < 0.001). In multivariable analysis, lower maternal PWR independently predicted BPD (aOR = 0.94 per unit increase; 95% CI, 0.91–0.98; p = 0.002). Additional independent predictors included chorioamnionitis (aOR = 1.94; 1.01–3.71; p = 0.046), male sex (aOR = 1.61; 1.00–2.59; p = 0.049), early-onset sepsis (aOR = 2.45; 1.04–5.77; p = 0.040), late-onset sepsis (aOR = 2.86; 1.47–5.56; p = 0.002), and mechanical ventilation >7 days (aOR = 6.74; 3.34–13.62; p < 0.001). Gestational age and birth weight were protective (per week: aOR = 0.68; 0.58–0.79; p < 0.001; per 100 g: aOR = 0.86; 0.78–0.94; p = 0.001). The final model showed strong discrimination (AUC = 0.866; 95% CI, 0.827–0.905) and good calibration (Hosmer–Lemeshow χ2 = 10.833, p = 0.211; bootstrap slope = 0.98). At the optimal probability threshold (0.763), sensitivity was 77.2% and specificity was 80.9%. A nomogram enables individualized risk estimation.

A lower maternal PWR measured immediately before delivery, alongside established neonatal risk factors, independently predicts BPD in preterm infants. This simple hematologic marker, integrated into a nomogram, may facilitate early identification and targeted prevention strategies in high-risk dyads.