Scholarlink

Purpose <p>To develop and validate a prenatal prediction model for aortic coarctation (CoA) using morphologic, hemodynamic, and fetal growth parameters to enhance diagnostic accuracy and guide clinical decision-making.</p> Method <p>Eighty-three fetuses with suspected CoA were retrospectively analyzed. Key prenatal predictors were analyzed using multivariable logistic regression to construct a nomogram. Model performance was evaluated via area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis (DCA).</p> Results <p>Of the 83 fetuses, 28 (33.7%) were postnatally confirmed with CoA. The final model identified abdominal-to-head circumference ratio × 100% (<i>β</i> = 0.90, 95% CI 0.22–1.58), maximum aortic arch z-score (<i>β</i> = − 0.85, 95% CI − 1.50 to − 0.19), ventricular septal defect (OR = 1.85, 95% CI 1.02–3.53), and abnormal atrial hemodynamics (OR = 0.73, 95% CI 0.38–1.39) as significant predictors. The model achieved an AUC of 0.86 (95% CI 0.78–0.94), with calibration plots demonstrating strong agreement between predicted and observed probabilities. DCA confirmed clinical utility across a wide threshold range.</p> Conclusions <p>This nomogram enhances CoA prediction by integrating structural and functional ultrasound markers. It offers strong diagnostic performance and practical value for prenatal risk stratification, potentially reducing false positives and unnecessary interventions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hemodynamic and developmental biomarkers enhance prenatal coarctation prediction: a validated multiparametric ultrasound model

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月17日

查看全文
  • Yu Wang,
  • Shuhua Luo,
  • Weiqiang Ruan,
  • Nan Guo

摘要

Purpose

To develop and validate a prenatal prediction model for aortic coarctation (CoA) using morphologic, hemodynamic, and fetal growth parameters to enhance diagnostic accuracy and guide clinical decision-making.

Method

Eighty-three fetuses with suspected CoA were retrospectively analyzed. Key prenatal predictors were analyzed using multivariable logistic regression to construct a nomogram. Model performance was evaluated via area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis (DCA).

Results

Of the 83 fetuses, 28 (33.7%) were postnatally confirmed with CoA. The final model identified abdominal-to-head circumference ratio × 100% (β = 0.90, 95% CI 0.22–1.58), maximum aortic arch z-score (β = − 0.85, 95% CI − 1.50 to − 0.19), ventricular septal defect (OR = 1.85, 95% CI 1.02–3.53), and abnormal atrial hemodynamics (OR = 0.73, 95% CI 0.38–1.39) as significant predictors. The model achieved an AUC of 0.86 (95% CI 0.78–0.94), with calibration plots demonstrating strong agreement between predicted and observed probabilities. DCA confirmed clinical utility across a wide threshold range.

Conclusions

This nomogram enhances CoA prediction by integrating structural and functional ultrasound markers. It offers strong diagnostic performance and practical value for prenatal risk stratification, potentially reducing false positives and unnecessary interventions.