胎儿生长受限围产期不良结局预测模型的构建

Establishment of prediction model of adverse perinatal outcome of fetal growth restriction

目的:分析胎儿生长受限(FGR)发生围产期不良结局的相关危险因素并建立列线图预测模型。方法:回顾分析2017年1月至2022年11月在广西医科大学第一附属医院符合纳入标准的269例FGR孕产妇的临床资料,根据是否发生复合围产期不良结局分为不良结局组和非不良结局组,采用LASSO-logistic回归分析,筛选FGR围产期不良结局的独立危险因素,构建列线图预测模型,并利用受试者工作特征(ROC)曲线下面积(AUC)、校准曲线及决策曲线评估预测效能。结果:269例FGR孕产妇中,80例(29.7%)发生复合围产期不良结局。LASSO-logistic回归分析结果显示,诊断FGR孕周越小、合并子痫前期或慢性高血压并发子痫前期、异常或不典型无应激试验(NST)、舒张期脐血流缺如、胎儿体重增长速度<100g/w是FGR围产期不良结局的独立危险因素(P<0.05)。构建列线图预测模型,采用Bootstrap 1000次重抽样进行内部验证,结果显示模型的区分度(AUC=0.885)、校准度及临床实用性较高。结论:多种因素与FGR围产期不良结局发病相关,本研究构建的列线图模型具有良好的预测效能和临床实用价值。

Objective:To analyze the related risk factors for adverse perinatal outcomes in fetal growth restriction(FGR)and to establish a nomogram prediction model.Methods:A retrospective analysis was conducted on the clinical data of 269 FGR pregnant women who met the inclusion criteria at the First Affiliated Hospital of Guangxi Medical University from January 2017 to November 2022.All patients were used as the training set and were divided into adverse outcome group and non-adverse outcome group based on whether they experienced composite adverse perinatal outcomes.LASSO-logistic regression analysis was used to screen the independent risk factors for adverse perinatal outcomes in FGR and construct a column chart prediction model.The predictive performance was evaluated using the area under the receiver operating characteristic curve,calibration curve,and decision curve analysis.Results:Of 269 FGR pregnant women,80 cases(29.7%)experienced adverse composite perinatal outcomes.The LASSO-logistic regression analysis results showed that a smaller gestational age at diagnosis of fetal growth restriction(FGR),the presence of comorbidities such as preeclampsia or chronic hypertension with preeclampsia,abnormal or atypical non-stress test(NST),absent diastolic umbilical blood flow,and fetal weight gain rate<100g/week were independent risk factors for adverse perinatal outcomes in FGR(P<0.05).A column chart prediction model was constructed,and internal validation was carried out using 1000 bootstrap resamples.The results showed that the model had high discrimination(AUC=0.885),calibration,and clinical utility.Conclusions:Multiple factors are associated with the occurrence of adverse perinatal outcomes in FGR.The column chart model constructed in this study has good predictive ability and clinical value.