Development and validation of a predictive model for preeclampsia: a retrospective cohort study

Area of Science:

• Obstetrics and Gynecology
• Reproductive Medicine
• Maternal-Fetal Medicine

Background:

• Preeclampsia (PE) poses significant risks to maternal and fetal health.
• Early identification of PE is crucial for timely intervention and improved outcomes.
• Existing predictive models may require refinement for broader clinical application.

Purpose of the Study:

• To develop and validate a predictive nomogram for preeclampsia (PE).
• To identify independent risk factors for PE in a high-risk population.
• To inform the development of early intervention strategies in clinical practice.

Main Methods:

• Data from 2063 women at medium or high risk for PE were analyzed.
• Placental growth factor (PlGF)-based testing was utilized.
• Logistic regression modeling and five-fold cross-validation were employed to build and validate the predictive model.
• Model performance was assessed using AUROC, calibration curves, decision curves, and clinical impact curves.

Main Results:

• 108 cases of PE were identified among 2063 women.
• Independent risk factors included BMI, MAP, sFlt-1/PlGF ratio, adverse pregnancy history, family history, previous PE, chronic hypertension, autoimmune disease, and PCOS.
• The predictive model achieved an AUROC of 0.883 in the training set and 0.862 in the validation set.
• The model demonstrated good sensitivity and specificity, with favorable calibration and clinical utility.

Conclusions:

• A predictive nomogram for PE based on common, interpretable features demonstrates desirable efficacy.
• The developed nomogram can effectively identify women at risk for PE.
• This tool can inform the development of specialized preventive protocols and early intervention strategies in clinical practice.