广东医学2026，Vol.47Issue(3)：395-400,6.DOI:10.13820/j.cnki.gdyx.20252687
新生儿呼吸窘迫综合征风险预测模型的构建及验证
Construction and verification of risk prediction model for neonatal respiratory distress syndrome
摘要
Abstract
Objective To explore the clinical value of constructing a risk prediction model for neonatal respiratory distress syndrome(NRDS)and to validate the performance of this model.Methods A total of 195 neonates admitted to the neonatal intensive care unit of Tangdu Hospital,the Second Affiliated Hospital of Air Force Medical University,be-tween August 2022 and August 2023 were enrolled.According to the occurrence of NRDS,neonates were divided into an NRDS group and a non-NRDS group.Clinical data of neonates and their mothers were collected.Logistic regression a-nalysis was performed to identify independent risk factors associated with NRDS.A risk prediction model was established based on these factors,and its predictive performance was evaluated using receiver operating characteristic(ROC)curves and calibration curves for both internal and external validation.Results The proportions of gestational age＜34 weeks,neonatal asphyxia,intrauterine infection,preeclampsia,Apgar score≤7,fetal distress,postnatal use of pulmonary sur-factant(PS),and antenatal use of glucocorticoids were significantly higher in the NRDS group than in the non-NRDS group(P＜0.05).Logistic regression analysis identified gestational age＜34 weeks and Apgar score ≤7 as independent risk factors for NRDS,while antenatal use of glucocorticoids and postnatal use of PS were protective factors(P＜0.05).The ROC analysis showed that the prediction model achieved an area under the curve(AUC)of 0.877(95％CI:0.823-0.920),with a sensitivity of 90.77％,specificity of 85.38％,and a Youden index of 0.762.The Hosmer-Lemeshow test indicated good calibration(x2=0.734,P=0.318),and the Brier score was 0.141.External validation demonstra-ted an AUC of 0.880(95％CI:0.803-0.936),sensitivity of 90.91％,specificity of 83.33％,Youden index of 0.742,Hosmer-Lemeshow x2=0.906(P=0.271),and a Brier score of 0.147.Conclusion Gestational age＜34 weeks,neonatal asphyxia,intrauterine infection,and fetal distress are independent risk factors for NRDS,while postnatal use of pulmonary surfactant is a protective factor.The risk prediction model constructed based on these factors demon-strates good predictive accuracy and calibration,and may be valuable for early identification and prevention of NRDS.关键词
新生儿呼吸窘迫综合征/影响因素/模型构建/验证Key words
neonatal respiratory distress syndrome/influencing factors/model construction/verify分类
医药卫生引用本文复制引用
王玉娟,赵艳,李桂成..新生儿呼吸窘迫综合征风险预测模型的构建及验证[J].广东医学,2026,47(3):395-400,6.基金项目
陕西省重点研发计划项目(2023-YERE-042) （2023-YERE-042）
