胎龄≤34周早产儿宫外发育迟缓风险预测模型的构建与验证  

Construction and validation of a risk prediction model for ectopic developmental delay in premature infants with gestational age≤34 weeks

作　　者：郭永琴[1] 豆瑛瑛 李建丽[2] 陈晓旭 张敏 王清萍 张雨竹 GUO Yongqin;DOU Yingying;LI Jianli;CHEN Xiaoxu;ZHANG Min;WANG Qingping;ZHANG Yuzhu(Department of Neonatology,Changzhi Maternal and Child Health Hospital Changzhi 046000,Shanxi,China;Department of Nursing,Changzhi Maternal and Child Health Hospital Changzhi 046000,Shanxi,China)

机构地区：[1]长治市妇幼保健院新生儿科,山西长治046000 [2]长治市妇幼保健院护理部,山西长治046000

出　　处：《护士进修杂志》2025年第6期616-622,共7页Journal of Nurses Training

基　　金：山西省卫生健康委员会科研项目(编号:2023002)。

摘　　要：目的探讨胎龄≤34周早产儿宫外发育迟缓(extrauterine growth restriction,EUGR)的危险因素,以此构建风险预测模型并检验模型效能。方法回顾性分析山西省某三甲医院2020年1月—2023年12月入住新生儿重疗监护室(neonatal intensive care unit,NICU)胎龄≤34周的750例早产儿临床资料,按照7∶3的比例将2020年1月—2023年2月的525例早产儿作为建模集,以出院时体重为评价标准,分为EUGR组和非EUGR组,采用LASSO回归联合多因素logistic回归分析筛选出胎龄≤34周早产儿发生EUGR的独立危险因素,以此构建预测模型;另将2023年3—12月入组的225例早产儿作为验证集。通过受试者工作特征曲线(ROC)下面积(AUC)、Hosmer-Lemeshow检验及校准曲线评估模型的区分度、拟合度,并对模型临床预测效果进行评价。结果建模集525例早产儿中有239例(45.5%)发生EUGR,多因素回归分析结果显示:胎龄、出生体重、健脾法口腔运动干预、静脉营养时间、呼吸系统并发症是胎龄≤34周早产儿发生EUGR的独立危险因素(P<0.05)。建模集AUC为0.929(95%CI:0.906~0.953),验证集AUC为0.913(95%CI:0.866~0.960),表明模型区分度良好;建模集Hosmer-Lemeshow拟合优度检验χ^(2)=15.167,df=8,P=0.056,验证集Hosmer-Lemeshow拟合优度检验χ^(2)=1.902,df=8,P=0.984,表明该模型准确性和一致性较好。验证集灵敏度为87.5%,特异度为92.7%,阳性预测值为88.5%,阴性预测值为92.0%,准确率为90.7%,精确率为88.5%,召回率为87.5%,说明本模型有较好的临床应用效能。结论胎龄、出生体重、健脾法口腔运动干预、静脉营养时间、呼吸系统并发症是胎龄≤34周早产儿发生EUGR的独立危险因素,该列线图模型的建立能够为临床医务人员早期识别高危人群,制定预见性干预管理措施提供有效的评估工具。Objective To investigate the risk factors of extrauterine growth restriction(EUGR)in preterm infants with a gestational age of≤34 weeks,so as to construct a risk prediction model and test the efficacy of the model.Methods The clinical data of 750 preterm infants admitted to the neonatal intensive care unit(NICU)from January 2020 to December 2023 at≤34 weeks gestational age in a tertiary hospital in Shanxi province were retrospectively analyzed,and 525 preterm infants from January 2020 to February 2023 were used as the modeling set according to the ratio of 7∶3,and the weight at discharge was used as the evaluation criterion,and the EUGR group and non-EUGR group were divided into EUGR group and non-EUGR group.LASSO regression combined with multivariate logistic regression analysis was used to screen out the independent risk factors for EUGR in preterm infants with a gestational age of≤34 weeks,and a prediction model was constructed.In addition,225 preterm infants enrolled from March to December 2023 were selected as the validation set.The area under the receiver operating characteristic(ROC)curve(AUC),Hosmer-Lemeshow test,and calibration curve were used to evaluate the discrimination and fit of the model,and the clinical prediction effect of the model was evaluated.Results Among the 525 preterm infants in the modeling set,239(45.5%)developed EUGR,and the results of multivariate regression analysis showed that gestational age,birth weight,spleen-strengthening oral exercise intervention,intravenous nutrition time,and respiratory complications were independent risk factors for EUGR in preterm infants with a gestational age of≤34 weeks(P<0.05).The AUC of the modeling set was 0.929(95%CI was 0.906-0.953),and the AUC of the validation set was 0.913(95%CI was 0.866-0.960),indicating that the model discrimination was good;The Hosmer-Lemeshow goodness-of-fit test of the modeling set wasχ^(2)=15.167,df=8,P=0.056,and the Hosmer-Lemeshow goodness-of-fit test of the validation set wasχ^(2)=1.902,df=8,P=0.984,indica

关 键 词：早产儿 宫外发育迟缓 危险因素 预测模型 

分 类 号：R473.72[医药卫生—护理学]