Objective: To compare the prognostic value of 3 diagnostic criteria of bronchopulmonary dysplasia (BPD) in preterm infants with gestational age<32 weeks. Methods: The retrospective cohort study was conducted to collect the clinical data of 285 preterm infants with BPD admitted to the Department of Neonatology, Children's Hospital Affiliated to Zhengzhou University from January 2019 to September 2021, who were followed up regularly after discharge. The primary composite adverse outcome was defined as death or severe respiratory morbidity from 36 weeks of corrected gestational age to 18 months of corrected age, and the secondary composite adverse outcome was defined as death or neurodevelopmental impairment. According to the primary or secondary composite adverse outcomes, the preterm infants were divided into the adverse prognosis group and the non-adverse prognosis group. The 2001 National Institute of Child Health and Human Development (NICHD) criteria, 2018 NICHD criteria, and 2019 Neonatal Research Network (NRN) criteria were used to diagnose and grade BPD in preterm infants. Chi-square test, Logistic regression analysis, receiver operating characteristic (ROC) curve and Delong test were used to analyze the prognostic value of the 3 diagnostic criteria. Results: The 285 preterm infants had a gestational age of 29.4 (28.1, 30.6) weeks and birth weight of 1 230 (1 000, 1 465) g, including 167 males (58.6%). Among 285 premature infants who completed follow-up, the primary composite adverse outcome occurred in 124 preterm infants (43.5%), and the secondary composite adverse outcome occurred in 40 preterm infants (14.0%). Multivariate Logistic regression analysis showed that severe BPD according to the 2001 NICHD criteria, gradeⅡand Ⅲ BPD according to the 2018 NICHD criteria and grade 2 and 3 BPD according to the 2019 NRN criteria were all risk factors for primary composite adverse outcomes (all P<0.05). ROC curve showed that the area under the curve (AUC) of the 2018 NICHD criteria and 2019 NRN criteria were both higher than that of the 2001 NICHD criteria (0.70 and 0.70 vs. 0.61, Z=4.49 and 3.35, both P<0.001), but there was no significant difference between the 2018 NICHD and 2019 NRN criteria (Z=0.38, P=0.702). Multivariate Logistic regression analysis showed that the secondary composite adverse outcomes were all associated with grade Ⅲ BPD according to the 2018 NICHD criteria and grade 3 BPD according to the 2019 NRN criteria (both P<0.05). ROC curve showed that the AUC of the 2018 NICHD criteria and 2019 NRN criteria were both higher than that of the 2001 NICHD criteria (0.71 and 0.71 vs. 0.58, Z=2.93 and 3.67, both P<0.001), but there was no statistically significant difference between the 2018 NICHD and 2019 NRN criteria (Z=0.02, P=0.984). Conclusion: The 2018 NICHD and 2019 NRN criteria demonstrate good and comparable predictive value for the primary and secondary composite adverse outcomes in preterm infants with BPD, surpassing the predictive efficacy of the 2001 NICHD criteria.
目的： 比较支气管肺发育不良（BPD）3种诊断标准对出生胎龄<32周BPD早产儿预后的预测价值。 方法： 采用回顾性队列研究，收集2019年1月至2021年9月郑州大学附属儿童医院新生儿科收治的285例出生胎龄<32周BPD早产儿的临床资料，出院后定期随访。以校正胎龄36周至校正年龄18月龄发生死亡或呼吸系统不良预后定义为主要复合不良结局，发生死亡或神经系统不良预后定义为次要复合不良结局，分别按照主要或次要复合不良结局分为预后不良组和预后正常组。分别应用2001年美国国立儿童健康与人类发育研究所（NICHD）标准（简称2001 NICHD标准）、2018 年NICHD修订标准（简称2018 NICHD标准）和2019年新生儿研究网（NRN）标准（简称2019 NRN标准）对BPD早产儿进行诊断和分级。采用χ2检验、Logistic回归分析、受试者工作特征（ROC）曲线及Delong法分析3种诊断标准对预后的预测价值。 结果： 285例BPD早产儿出生胎龄为29.4（28.1，30.6）周，出生体重为1 230（1 000，1 465）g，其中男167例（58.6%）。285例BPD早产儿完成随访，发生主要复合不良结局124例（43.5%），次要复合不良结局40例（14.0%）。多因素Logistic回归分析显示2001 NICHD标准的重度BPD，2018 NICHD标准的Ⅱ、Ⅲ级BPD和2019 NRN标准的2、3级BPD均是主要复合不良结局发生的危险因素（均P<0.05）；ROC曲线显示2018 NICHD标准及2019 NRN标准的AUC均高于2001 NICHD标准（0.70和0.70比0.61，Z=4.49、3.35，均P<0.001），但2018 NICHD标准与2019 NRN标准相比差异无统计学意义（Z=0.38，P=0.702）。多因素Logistic回归分析显示2018 NICHD标准的Ⅲ级BPD和2019 NRN标准的3级BPD均与次要复合不良结局的发生相关（均P<0.05）；ROC曲线显示2018 NICHD标准及2019 NRN标准的AUC均高于2001 NICHD标准（0.71和0.71比0.58，Z=2.93、3.67，均P<0.001），但2018 NICHD标准与2019 NRN标准相比差异无统计学意义（Z=0.02，P=0.984）。 结论： 2018 NICHD标准和2019 NRN标准对BPD早产儿主要及次要复合不良结局具有较好且相似的预测能力，预测效能均优于2001 NICHD标准。.