Temporal changes in bronchopulmonary dysplasia severity distribution associated with evolving respiratory practices in very preterm infants.

OBJECTIVE: To evaluate temporal changes in respiratory management practices and bronchopulmonary dysplasia (BPD) severity in infants born before 32 weeks of gestation over a 9-year period. DESIGN: Retrospective cohort study. SETTING: Tertiary neonatal intensive care unit. PATIENTS: Infants born before 32 weeks of gestation and admitted between January 2017 and December 2025. EXPOSURE: Birth during one of three predefined clinical eras: pre-COVID (2017-2019), COVID (2020-2022) and post-COVID following relocation to an expanded neonatal intensive care unit (2023-2025). MAIN OUTCOME MEASURES: BPD severity at 36 weeks of postmenstrual age (PMA), mortality before hospital discharge, and the composite outcome of death before hospital discharge or BPD at 36 weeks of PMA. Associations between study era and respiratory outcomes were evaluated using multivariate logistic regression. RESULTS: A total of 1259 infants were included. Respiratory management practices changed across study eras. High-flow nasal cannula exposure increased from 43.7% in era 1 to 70.6% in era 3 (p<0.001), while duration of non-invasive respiratory support increased from 4 to 11 days (p<0.001). Less invasive surfactant administration was introduced predominantly during era 3, and postnatal corticosteroid exposure increased from 16.5% to 36.4% (p<0.001). BPD incidence increased across eras (25.2%, 30.9% and 37.7%; p<0.001), primarily driven by grade 2 disease, whereas grade 3 BPD and mortality remained stable. Compared with era 1, infants in era 3 had higher adjusted odds of moderate-to-severe BPD (adjusted OR 3.29, 95% CI 2.15 to 5.05; p<0.001) and the composite outcome of death before hospital discharge or BPD at 36 weeks of PMA (adjusted OR 2.08, 95% CI 1.47 to 2.94; p<0.001). CONCLUSIONS: Respiratory management practices evolved over time and were accompanied by increased moderate BPD without corresponding increases in severe BPD or mortality. These findings suggest that evolving respiratory support practices and contemporary BPD definitions may influence the observed distribution of BPD severity.