Comparison of Definitions for Bronchopulmonary Dysplasia: A Cohort Study.

We aimed to compare the definitions of National Institute of Child Health and Human Development (NICHD) for bronchopulmonary dysplasia (BPD) for determining the incidences, and predicting late death and respiratory outcome. This retrospective cohort study included infants born at<32 weeks' gestation who survived up to 36 weeks' postmenstrual age (PMA). Infants were classified as having BPD or no BPD per thedefinitions of NICHD 2001 and 2018. The incidences of BPD were 49 and 32% according to the 2001 and 2018 NICHD definitions. Gestational age, birth weight and intubation after birth were associated with BPD by both definitions. The NICHD 2018 definition displayed similar sensitivity (100%) and negative predictive value (100%), and higher specificity (70 vs. 52%) for predicting death after 36 weeks' PMA; a higher specificity (72 vs. 53%), comparable negative predictive value (77 vs.76%), but lower sensitivity for predicting adverse respiratory outcome within 12 months corrected age compared with the NICHD 2001 definition. The NICHD 2018 definition is as powerful as the 2001 definition for predicting late death and seems to be a better indicator for long-term respiratory outcome. The use of supplemental oxygen or oxygen plus respiratory support should be considered while predicting both late death and long-term respiratory outcome.