Objective:

To evaluate the predictive accuracy of the NICHD bronchopulmonary dysplasia (BPD) outcome estimator for predicting grade 2 and grade 3 BPD and the composite outcome of death or grade 2 to grade 3 BPD at 36 weeks postmenstrual age.

Approach:

• Study Design: Single-centred retrospective cohort study involving neonates with birth weights of 501–1250 g and gestational ages of 23+0 to 28+6 weeks.
• Outcome Measurement: Used C-statistic receiver operating characteristic area under the curve (AUC) to predict outcomes on postnatal days 1, 3, 7, 14, and 28.

Key Findings:

• AUC values were 0.71 on postnatal day 7 and 0.76 on postnatal day 14.
• Prediction accuracy for the composite outcome of death or grade 2–3 BPD at 36 weeks PMA was found to have limited clinical utility.

Interpretation:

The NICHD web-based Neonatal BPD Outcome Estimator (2022) has moderate accuracy in predicting grade 2–3 BPD and the composite outcome of death or grade 2–3 BPD.

Limitations:

• The study was conducted in a single centre, limiting generalizability.
• The predictive performance may vary across different populations and settings.

Conclusion:

The findings indicate the need for cautious interpretation of the estimator's predictions.