Abstract
Unlike other complications among very low birth weight infants (VLBW), the incidence of bronchopulmonary dysplasia (BPD) has not decreased substantially, partly because of the different definitions of BPD applied by different researchers. In this retrospective cohort study, we aimed to compare the 2018 revised definition and the 2001 consensus definition of BPD proposed by the National Institute of Child Health and Human Development (NICHD), as well as to identify which definition better predicts severe respiratory morbidities or death. We included 417 infants born at a gestational age <32 weeks and classified them as having BPD or without BPD based on the two definitions, with a final follow-up at 18-24 months. We performed between-group comparisons of death and respiratory outcomes. Statistical analyses were performed using descriptive statistics, comparative tests, and receiver operating characteristic curves. The mean ± standard deviation gestational age and birth weight of the 417 eligible infants were 29.1 ± 1.4 weeks and 1186.6 ± 197.8 g, respectively. Among the included infants, five and three infants died before and after 36 weeks of post-menstrual age (PMA), respectively, with 68 and 344 infants evaluated at discharge and 36 weeks' PMA, respectively. We diagnosed 163 (39.1%) and 70 (16.8%) infants with BPD according to the 2001 and 2018 NICHD definitions, respectively. The 2001 NICHD definition displayed a higher sensitivity (0.60 vs. 0.28), better negative predictive value (0.89 vs. 0.85), and larger area under the receiver operating characteristic curve (0.66 vs. 0.57), but a lower specificity (0.65 vs. 0.87) and worse positive predictive value (0.26 vs. 0.31), than the 2018 definition for serious respiratory morbidity or mortality at a corrected age of 18-24 months. Compared with the 2018 NICHD definition of BPD, the 2001 NICHD consensus definition may result in more cases of false-positive or unclassified severity. However, it may be a better indicator of severe respiratory morbidities or death during the first 18-24 months. Nevertheless, there is a need for future studies to assess the validity of the new diagnostic criteria.