Comparison of two diagnostic criteria for bronchopulmonary dysplasia and their predictive value for respiratory outcomes / 中华围产医学杂志

ABSTRACT

Objective:To compare the diagnosis and respiratory outcomes of preterm infants with bronchopulmonary dysplasia (BPD) based on two criteria and to analyze the criteria's predictive value for adverse respiratory system outcomes.Methods:Clinical data of preterm infants with gestational age less than 32 weeks admitted to the First Affiliated Hospital of Sun Yat-sen University from January 2014 to September 2020 were retrospectively analyzed. The diagnosis and classification of all the subjects were based on two definitions developed by the American National Institute of Child Health and Human Development (NICHD) in 2001 and 2018. The severity of pulmonary diseases in eligible infants was quantified with pulmonary scores. The incidence of BPD and respiratory outcomes in infants diagnosed with the two criteria were compared. The logistic predictive model was adopted to analyze the predictive value of the two criteria for adverse long-term respiratory outcomes.Results:The study enrolled 240 infants, among which 138 (57.5%) were male. Infants who met the 2001 and 2018 definitions of BPD accounted for 52.9% (127/240) and 24.0% (57/237), respectively. The rate of pneumonia, mechanical ventilation duration, and overall oxygen duration of the infants who met the 2018 definition were significantly higher or longer than those who met the 2001 definition [84.2% (48/57) vs. 63.7%(79/124), χ2=7.84; 24.0 d (7.0-34.5 d) vs. 9.0 d (4.0-26.0 d), Z=－3.01; 66.0 d (49.0-81.5) vs. 47.0 d (38.0-63.8 d), Z=－3.86, P<0.01]. The incidence of adverse respiratory system outcomes and death of the infants who met the 2018 definition were significantly higher than those who met the 2001 definition [38.6% (17/44) vs. 20.9% (19/91), P<0.05]. Of the 180 infants diagnosed with BPD using the 2001 definition, 67 did not meet the 2018 definition and were all discharged before a corrected age of 45 weeks without oxygen requirement. Only 4.3% (2/47) of those with follow-up data were re-hospitalized twice or more due to pulmonary diseases. The predicted value of the 2018 definition for early death or adverse respiratory outcomes in infants was significantly higher than that of 2001, with the areas under the curves of 0.871 and 0.840, respectively. The area difference under the curve was 0.031 ( Z=2.09, P=0.037). Conclusions:Compared with the one in 2001, the 2018 definition of BPD is stricter, which could be able to identify BPD infants with severer pulmonary status and outcomes and is unlikely to miss cases with adverse respiratory outcomes. The 2018 definition performs well in predicting death or adverse respiratory outcomes in early infancy.