A study on the clinical application of different diagnostic criteria for bronchopulmonary dysplasia / 中国新生儿科杂志

ABSTRACT

Objective:To compare the results of clinical diagnosis and severity grading in preterm infants with bronchopulmonary dysplasia (BPD) using three different diagnostic criteria and the consistency of two new diagnostic criteria.Methods:From January to December, 2020, infants with gestational age <32 w admitted to neonatal intensive care unit of our hospital were retrospectively enrolled in this cohort study. The patients were diagnosed and graded according to the 2001, 2018 and 2019 criteria of BPD. Chi-square test was used to compare the differences of BPD diagnostic rate and mortality rate using three criteria and Kappa coefficient test was used to compare the consistency between the two new criteria of 2018 NICHD and 2019 NRN.Results:A total of 231 preterm infants were enrolled, including 130 males (56.3%) and 101 females. 9 patients were dead. According to 2018 NICHD criteria, 97 cases (42.0%) were diagnosed with BPD, including 16 gradeⅠ, 44 grade Ⅱ, 31 grade Ⅲ and 6 grade ⅢA. The remaining 134 cases were not BPD (58.0%). No significant differences existed ( P>0.05) among the diagnostic rates of 2001 criteria (112/231, 48.5%), 2018 criteria (97/231, 42.0%) and 2019 criteria (91/231, 39.4%). For grade Ⅲ BPD, the diagnostic rate of 2001 criteria was significantly higher than the 2018 criteria (including grade Ⅲ and grade ⅢA, 16.0%) and 2019 criteria (6.5%) and the diagnostic rate of 2018 criteria was also significantly higher than 2019 criteria ( P<0.05). No significant differences existed in the overall mortality rate of BPD among three criteria ( P>0.05), however, the case mortality rate of grade Ⅲ BPD of 2001 criteria (3.9%) was significantly lower than 2018 criteria (24.3%) and 2019 criteria (20.0%) ( P<0.05). The 2018 and 2019 criteria were highly consistent in the overall diagnostic rate of BPD (Kappa value = 0.946), the positive consistency rate was 93.8% (95% CI 85.5%~97.5%) and the negative consistency rate was 100.0% (95% CI 96.5%~100.0%). But the consistency of severity grading for BPD was weak (Kappa value = 0.597) between the two criteria. Conclusions:The 2001 NICHD BPD criteria is no longer valid because it tends to overdiagnose severe BPD, thus underestimate the case mortality. The 2018 NICHD criteria is comprehensive and detailed and the 2019 NRN criteria is simple and practical. The two new criteria are highly consistent in the overall diagnosis of BPD, but the consistency of severity grading is weak.