新生儿呼吸窘迫综合征预后Nomogram预测模型的构建与评估

Construction and evaluation of Nomogram prediction model for prognosis of neonatal respiratory distress syndrome

目的构建新生儿呼吸窘迫综合征(NRDS)预后Nomogram预测模型,并评估其应用价值。方法收集2017年1月至2021年12月扬州大学附属医院收治的NRDS病例216例,均追踪确诊后30 d内预后情况,依据追踪结果分成预后不良组(n=54)与预后良好组(n=162),采用Logistic回归分析筛选影响NRDS预后的危险因素;采用R软件构建NRDS预后的Nomogram预测模型,并采用ROC曲线、校准曲线进行Nomogram预测模型验证。结果Logistic回归分析结果显示,胎龄、Apgar评分(5 min)、妊娠糖尿病(GD)、新生儿肺出血是NRDS患儿预后不良的独立危险因素(P<0.05)。绘制ROC曲线与校准曲线:ROC曲线下面积为0.906(95%CI:0.863~0.949);校准曲线斜率接近1,且H-L拟合优度检验χ2=9.169,P=0.328。结论基于胎龄、Apgar评分(5 min)、GD、新生儿肺出血等4项因素构建的Nomogram预测模型有助于医护人员个体化预测NRDS患儿预后,能辅助临床进行治疗决策。

Objective To construct a Nomogram prediction model for the prognosis of neonatal respiratory distress syndrome(NRDS)and to evaluate its application value.Methods A total of 216 NRDS cases accepted by our hospital from January 2017 to December 2021 were collected,and their prognosis within 30 days after diagnosis was tracked.According to the tracking results,they were separated into a poor prognosis group(n=54)and a good prognosis group(n=162),the risk factors affecting the prognosis of NRDS were screened by Logistic regression analysis.The Nomogram prediction model of NRDS prognosis was constructed by R software,and the Nomogram prediction model was verified by ROC curve and calibration curve.Results Logistic regression analysis showed that gestational age,Apgar score(5 min),gestational diabetes(GD),and neonatal pulmonary hemorrhage were independent risk factors for poor prognosis in children with NRDS(P<0.05).Drawing ROC curve and calibration curve:The area under the ROC curve was 0.906(95%CI:0.863-0.949),the slope of the calibration curve was close to 1,and the H-L goodness-of-fit testχ2=9.169,P=0.328.Conclusion The Nomogram prediction model constructed based on four factors including gestational age,Apgar score(5 min),GD,and neonatal pulmonary hemorrhage can help individualized prediction of the prognosis of children with NRDS,and can assist clinicians in making treatment decisions.