Scholarlink

Background <p>Predicting pre-eclampsia (PE) risk in twin pregnancies enables timely intervention and disease management. We developed a predictive model using five clinically accessible variables to identify high-risk patients for early intervention.</p> Methods <p>This prospective cohort study enrolled 339 twin pregnancies receiving antenatal care in the Maternal and Child Healthcare Hospital of Changning District, Shanghai, China (January 2019–March 2024). General information was collected. Serum alpha-fetoprotein (AFP), unconjugated estriol (uE3), inhibin-A levels, and bilateral uterine artery pulsatility index (UtA-PI) were measured. A nomogram prediction model for PE risk was constructed.</p> Results <p>PE was diagnosed in 43 of 339 women. The predictive model identified lower pre-pregnancy BMI, IVF conception, and dysregulated serum biomarkers (uE3, inhibin A) as key predictors of PE. The nomogram demonstrated moderate discriminative ability (C-index: 0.73 (95% CI: 0.66–0.80)). Calibration curves showed close agreement between predicted and actual probability (<i>P</i> for Hosmer and Lemeshow test was 0.23), suggesting good model fit. Decision curve analysis confirmed its clinical utility.</p> Conclusions <p>Integrating IVF status, chorionicity, early-pregnancy serum biomarkers (inhibin A, uE3) and pre-pregnancy BMI in a model developed from a single-center cohort. This model represents an initial step in risk modeling and hints at clinical translatability.</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

The predictive value of early pregnancy markers for the risk of preeclampsia in women with twin pregnancies

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月20日

查看全文
  • Chao Wang,
  • Meng Yu,
  • Jiangnan Wu,
  • Tingyu Hu,
  • Ting Peng

摘要

Background

Predicting pre-eclampsia (PE) risk in twin pregnancies enables timely intervention and disease management. We developed a predictive model using five clinically accessible variables to identify high-risk patients for early intervention.

Methods

This prospective cohort study enrolled 339 twin pregnancies receiving antenatal care in the Maternal and Child Healthcare Hospital of Changning District, Shanghai, China (January 2019–March 2024). General information was collected. Serum alpha-fetoprotein (AFP), unconjugated estriol (uE3), inhibin-A levels, and bilateral uterine artery pulsatility index (UtA-PI) were measured. A nomogram prediction model for PE risk was constructed.

Results

PE was diagnosed in 43 of 339 women. The predictive model identified lower pre-pregnancy BMI, IVF conception, and dysregulated serum biomarkers (uE3, inhibin A) as key predictors of PE. The nomogram demonstrated moderate discriminative ability (C-index: 0.73 (95% CI: 0.66–0.80)). Calibration curves showed close agreement between predicted and actual probability (P for Hosmer and Lemeshow test was 0.23), suggesting good model fit. Decision curve analysis confirmed its clinical utility.

Conclusions

Integrating IVF status, chorionicity, early-pregnancy serum biomarkers (inhibin A, uE3) and pre-pregnancy BMI in a model developed from a single-center cohort. This model represents an initial step in risk modeling and hints at clinical translatability.