Abstract
OBJECTIVE: To identify risk factors and develop an early prediction model for bronchopulmonary dysplasia (BPD) in preterm infants with a gestational age (GA) under 32 weeks. DESIGN: A single-centre, retrospective case-control study. SETTING: A tertiary neonatal intensive care unit in Fujian Province, China. PARTICIPANTS: 762 infants (GA under 32 weeks, survived beyond the corrected GA of 36 weeks) who were hospitalised in our department between January 2018 and December 2023. INTERVENTION: None. MAIN OUTCOME MEASURES: The primary outcome was the diagnosis of BPD at 36 weeks of corrected GA, with diagnostic criteria based on the 2018 criteria established by the National Institute of Child Health and Human Development. RESULTS: A total of 762 infants was enrolled in the study, including 572 in the training cohort, among whom 138 (24.13%) had BPD. Multivariate logistic regression identified the following as independent predictors for BPD development: birth weight (OR=0.997, 95% CI 0.996 to 0.998, p<0.001), chest X-ray score within 24 hours of birth (OR=1.777, 95% CI 1.253 to 2.521, p=0.001), chest X-ray score at day 7 after birth (OR=1.756, 95% CI 1.264 to 2.440, p<0.001) and extended invasive mechanical ventilation to day 7 after birth (OR=4.280, 95% CI 2.149 to 8.526, p<0.001). In training and validation cohorts, the area under the curve was 0.873 (95% CI 0.839 to 0.907) and 0.848 (95% CI 0.778 to 0.917), respectively. In addition, the calibration curves showed good agreement between the column-line graph model and the actual observations. A decision curve analysis indicated a significantly better net benefit of the model. CONCLUSIONS: Higher chest X-ray scores within 7 days of life and extended invasive mechanical ventilation to day 7 after birth were significantly associated with BPD. The model had good predictive performance for predicting BPD using birth weight, chest X-ray scores within 24 hours of birth and at day 7 after life, and the state of invasive mechanical ventilation at day 7 after birth.