基于间接超声征象的胎儿胼胝体完全缺如的危险因素分析及预测模型初步建立

Risk factors analysis and preliminary establishment of a predictive model for fetal complete agenesis of the corpus callosum based on indirect ultrasound imaging

目的:分析胎儿胼胝体完全缺如(CACC)的间接超声征象并初步建立预测模型。方法:选择产前超声发现透明隔腔(CSP)细小或缺如的463例胎儿作为研究对象,以磁共振检查结果作为诊断标准,根据是否CACC分为CACC组(107例)和非CACC组(356例),通过单因素及多因素二元Logistic回归分析,初步建立基于间接超声征象的胎儿CACC的预测模型。结果:差异性分析结果显示CACC组的孕周、侧脑室增宽、泪滴样侧脑室、第三脑室上抬、脑中线囊性结构、脑沟回浅平的超声征象发生率均高于非CACC组。多因素二元Logistic回归分析结果显示侧脑室增宽及泪滴样侧脑室是诊断CACC的独立预测因素。建立的预测模型为Logit(P)=-4.149+1.613×侧脑室增宽+4.440×泪滴样侧脑室,该模型具有较好的区分度(曲线下面积为0.951,95%CI:0.925~0.977)和校准度(χ^(2)=0.237 P=0.888)。结论:当产前超声发现CSP细小或缺如时,如同时合并侧脑室增宽、泪滴样侧脑室时CACC的可能性大,所建立的预测模型具有较好的预测价值,可为基层超声医生诊断完全型胼胝体缺如提供重要的诊断依据。

Objective To analyze the indirect ultrasound signs of fetal complete agenesis of the corpus callosum(CACC)and establish a prediction model.Methods A total of 463 fetuses with small or absence of cavum septum pellucidum(CSP),using the results of magnetic resonance(MR)as the diagnostic criteria,into CACC group(n=107)and non-CACC group(n=356),Univariate and multivariate Logistic regression analysis were used to establish a prediction model for CACC based on indirect ultrasound signs.Results Multivariate binary Logistic analysis showed that the lateral ventricular dilation and the lateral ventricle with teardrop like were independent risk factors for diagnosis CACC.The established risk prediction model was Logit(P)=-4.149+1.613×the lateralventriculardilation+4.440×the lateral ventricle with teardrop like,which had good discrimination(area under the curve=0.951,95%CI:0.925-0.977)and calibration(χ^(2)=0.237,P=0.888).Conclusion The risk prediction model constructed based on prenatal ultrasound with small or absence of CSP can provide important diagnostic basis for grassroots ultrasound doctors to diagnose CACC.