当前位置:首页 / 基于彩色多普勒血流成像参数构建高危妊娠孕妇发生胎儿窘迫的列线图预测模型
论著·临床研究 | 更新时间:2025-08-13
|
基于彩色多普勒血流成像参数构建高危妊娠孕妇发生胎儿窘迫的列线图预测模型
Nomogram prediction model for the occurrence of fetal distress among pregnant women with high⁃risk pregnancy constructed based on color Doppler flow imaging parameters

广西医学 页码:957-961

作者机构:孙惠惠,本科,主治医师,研究方向为妇产科超声。

基金信息:国家自然科学基金(82301933)

DOI:10.11675/j.issn.0253⁃4304.2025.07.07

  • 中文简介
  • 英文简介
  • 参考文献

目的 基于彩色多普勒血流成像(CDFI)参数构建高危妊娠孕妇发生胎儿窘迫的列线图预测模型。方法 回顾性分析916例高危妊娠孕妇的临床资料,所有孕妇在产前均行胎儿脐动脉(UA)、大脑中动脉(MCA)CDFI检查,记录CDFI参数[阻力指数(RI)、收缩期峰值流速与舒张末期流速比值(S/D)、搏动指数(PI)]。根据是否发生胎儿窘迫将高危妊娠孕妇分为发生组(136例)和未发生组(780例)。采用Logistic回归模型分析影响高危妊娠孕妇发生胎儿窘迫的因素,构建列线图预测模型并进行验证及效能评估。结果 发生组胎儿UA的RI、胎儿UA的PI、胎儿UA的S/D、孕妇年龄≥35岁占比、产程延长占比、胎儿出生后1 min Apgar评分≤7分占比高于未发生组,胎儿MCA的RI、胎儿MCA的PI、胎儿MCA的S/D低于未发生组(P<0.05)。Logistic回归分析结果显示,胎儿UA的PI、胎儿UA的S/D、胎儿MCA的PI、胎儿MCA的S/D是高危妊娠孕妇发生胎儿窘迫的影响因素(P<0.05)。基于上述影响因素构建的列线图预测模型的一致性指数为0.853,Hosmer⁃Lemeshow检验显示模型预测发生概率与实际发生概率差异无统计学意义(P>0.05)。受试者工作特征曲线分析显示,该模型预测高危妊娠孕妇发生胎儿窘迫的灵敏度为84.40%、特异度为80.30%,曲线下面积为0.867。结论 胎儿UA的PI、胎儿UA的S/D、胎儿MCA的PI、胎儿MCA的S/D是高危妊娠孕妇发生胎儿窘迫的影响因素,基于这些影响因素所构建的列线图模型预测高危妊娠孕妇发生胎儿窘迫的效能良好。

Objective To construct a nomogram prediction model for the occurrence of fetal distress among pregnant women with high⁃risk pregnancy based on color Doppler flow imaging (CDFI) parameters. Methods The clinical data of 916 pregnant women with high⁃risk pregnancy were retrospectively analyzed. All pregnant women underwent prenatal CDFI examinations of the fetal umbilical artery (UA) and middle cerebral artery (MCA), and CDFI parameters with respect to resistance index (RI), peak systolic/end⁃diastolic ratio (S/D), and pulsatility index (PI) were recorded. According to the presence of fetal distress occurrence, the pregnant women with high⁃risk pregnancy were divided into occurrence group (136 cases) or non⁃occurrence group (780 cases). The Logistic regression model was adopted to analyze factors influencing the occurrence of fetal distress among pregnant women with high⁃risk pregnancy, and a nomogram prediction model was constructed for validation and efficiency evaluation. Results The occurrence group exhibited higher fetal UA RI, PI and S/D, proportion of maternal age ≥35 years, proportion of prolonged labor, and proportion of 1⁃minute post⁃fetus born Apgar score ≤7 compared to the non⁃occurrence group, while lower fetal MCA RI, PI, and S/D as compared with the non⁃occurrence group (P<0.05). The results of Logistic regression analysis revealed that fetal UA PI and S/D, fetal MCA PI and S/D were the influencing factors for the occurrence of fetal distress among pregnant women with high⁃risk pregnancy (P<0.05). The nomogram prediction model constructed based on these influencing factors as above had a C⁃index of 0.853, and the Hosmer⁃Lemeshow test indicated no statistically significant difference between predicted and actual probabilities of the model (P>0.05). The results of receiver operating characteristic curve analysis implied that the model for predicting the occurrence of fetal distress among pregnant women with high⁃risk pregnancy had a sensitivity of 84.40%, specificity of 80.30%, and an area under the curve of 0.867. Conclusion Fetal UA PI and S/D, fetal MCA PI and S/D are the influencing factors for the occurrence of fetal distress among pregnant women with high⁃risk pregnancy. The nomogram model for predicting the occurrence of fetal distress among pregnant women with high⁃risk pregnancy constructed based on these influencing factors demonstrates favorable efficiency.  

106

浏览量

9

下载量

0

CSCD

工具集