基于产前超声指标和妊娠期高血压的列线图模型预测低体质量儿不良结局的价值

目的 探讨基于产前超声指标和妊娠期高血压（PIH）构建的列线图模型在预测低体质量儿不良结局中的价值。方法 选取在我院行产前超声检查且分娩的PIH孕妇102例（PIH组）和正常孕妇121例（对照组），比较两组临床及产前超声检查资料的差异；应用单因素及多因素Logistic回归分析筛选低体质量儿不良结局的影响因素，并建立预测低体质量儿不良结局的列线图模型。采用Bootstrap重抽样法进行内部验证，绘制受试者工作特征（ROC）曲线分析模型的诊断效能；Hosmer-Lemeshow检验并绘制校准曲线评估模型的拟合优度及校准度；决策曲线分析模型的临床获益。结果 PIH组检出低体质量儿33例，对照组检出低体质量儿6例，两组低体质量儿占比比较差异有统计学意义（P<0.01）。单因素及多因素Logistic回归分析显示，PIH、晚孕期胎儿体质量、妊娠32周后脐动脉收缩末期峰值流速与舒张末期峰值流速比值均是低体质量儿不良结局的独立影响因素（OR=6.162、0.997、16.712，均P<0.05）。ROC曲线分析显示，列线图模型预测低体质量儿不良结局的曲线下面积为0.852(95%可信区...

Prediction model of low birth weight infant outcome based on ultrasound parameters and pregnancy-induced-hypertension

Objective To establish a predictive model for the outcome of low birth weight infants based on ultrasound parameters and pregnancy-induced-hypertension. Methods A retrospective study was conducted to include clinical and ultrasound examination data from a sample of 223 pregnant women from January 2019 to December 2021 at Wuhan Central Hospital, and single- and multi-factor regression analyses were performed to screen predictors, and logistic regression was used to construct the prediction model. Bootstrap resampling was used for internal validation, and C statistics, Hosmer?Lemeshow test, and DCA curves were used to evaluate the model, and nomogram were used for model presentation. Results Of the 223 valid samples, 102 (45.7%) were pregnant women with pregnancy-induced-hypertension, 121 (54.3%) were normal, and 39 (16.7%) were low birth weight infants. The variables that were finally included in the prediction model were: presence of pregnancy-induced-hypertension, fetal weight on late pregnancy systemic ultrasound, and S/D of the umbilical artery on routine ultrasound after 32 weeks of gestation. the model had good discrimination and calibration with a C-statistic of 0.852 (95% CI: 0.789~0.914) and Hosmer?Lemeshow test P=0.788. Conclusions We developed a predictive model for predicting the outcome of low birth weight infants using clinical examination data of pregnant women combined with ultrasound parameters. The model has high accuracy and can guide early intervention for adverse pregnancy outcomes and has certain clinical application.