A nomogram for predicting the risk of bronchopulmonary dysplasia in preterm infants: a prospective multicenter study

Background Bronchopulmonary dysplasia (BPD) is a common and serious complication in very preterm infants, with potential long-term adverse effects. This study aimed to analyze the perinatal and postnatal risk factors for BPD in preterm infants, and to develop and validate a risk model for early postnatal risk stratification based on information available within the first 7 days of life. Methods Data on preterm infants with gestational age < 32 weeks were prospectively collected from 28 hospitals in Shenzhen from January 2022 to December 2023. Neonates admitted to the NICU of Peking University Shenzhen Hospital and Shenzhen People’s Hospital were designated as the external validation cohort, whereas those from the other 26 NICUs were randomly divided into training and internal validation cohorts in an 8:2 ratio. BPD was the primary outcome and was diagnosed based on the 2018 revised criteria. Predictive factors were identified from the training cohort using logistic regression, and a nomogram model was constructed. Model performance was evaluated in terms of discrimination, calibration, and potential clinical utility. Results A total of 1,336 preterm infants were included, with 801 in the training cohort, 214 in the internal validation cohort, and 321 in the external validation cohort. Among them, 259 infants (19.39%) were diagnosed with BPD. The nomogram model was based on six predictive factors: gestational age, birth weight (per 100 g increase), occurrence of preterm premature rupture of membranes (PPROM), antenatal corticosteroid (ACS) administration, early onset sepsis, and invasive mechanical ventilation (IMV) within the first seven days of life. The area under the curve (AUC)s were 0.812, 0.783, and 0.810 in the training, internal validation, and external validation cohorts, respectively, indicating acceptable discrimination. Calibration showed reasonable agreement between predicted and observed outcomes. Decision curve analysis suggested potential net benefit across a range of threshold probabilities, although clinical interpretation remains exploratory. Conclusion This nomogram provides early postnatal risk stratification of BPD based on routinely available clinical information within the first week of life and may help identify preterm infants at higher risk.