Development and Internal Validation of a Risk Prediction Model for Acute Cardiovascular Morbidity in Preeclampsia.

Area of Science:

• Obstetrics and Gynecology
• Cardiology
• Maternal-Fetal Medicine

Background:

• Preeclampsia poses a significant short-term risk for adverse cardiovascular events in pregnant and postpartum women.
• Developing a predictive tool is crucial for managing these risks.

Purpose of the Study:

• To develop and internally validate a risk assessment tool for predicting acute cardiovascular morbidity in women diagnosed with preeclampsia.
• To identify key predictors of cardiovascular complications in this population.

Main Methods:

• A retrospective cohort study including 4171 women with preeclampsia at delivery (2007-2017).
• Development and internal validation of a predictive model using bootstrapping.
• Evaluation of model discrimination (AUC) and calibration.

Main Results:

• The final model included 8 variables, with gestational age, maternal age, and prior cesarean delivery being significant predictors.
• The model achieved an area under the receiver operating characteristic curve of 0.72, indicating good predictive ability.
• The tool demonstrated adequate calibration and performed well on internal validation.

Conclusions:

• A validated risk prediction tool can identify women with preeclampsia at high risk of acute cardiovascular morbidity.
• External validation of this tool could enable timely interventions to prevent adverse cardiovascular outcomes during pregnancy and postpartum.