A Composite Risk Scoring Model for Predicting Adverse Perinatal Outcomes in Patients with Pre-Eclampsia- A Pilot...

Area of Science:

• Obstetrics and Gynecology
• Maternal-Fetal Medicine
• Clinical Risk Stratification

Background:

• Early identification of pregnancies at high risk for adverse fetal outcomes (e.g., small-for-gestational-age, preterm birth, stillbirth) is crucial for timely interventions.
• Existing methods for risk stratification require validation and improvement for clinical utility.
• Development of accurate and accessible risk assessment tools is a priority in prenatal care.

Purpose of the Study:

• To pilot the development and internal validation of a novel point-based Fetal Risk Score (rFRS₅) using ten routine maternal parameters.
• To compare the performance of the rFRS₅ against a simpler, four-item Fetal Risk Score (FRS red).
• To assess the discriminatory and calibration capabilities of both risk scores.

Main Methods:

• A retrospective cohort study involving 118 pregnant women was conducted.
• Points were assigned to maternal parameters including age, BMI, blood pressure, and laboratory values.
• Performance was evaluated using ROC AUC for discrimination and Brier score for calibration, with 5-fold cross-validation.

Main Results:

• The rFRS₅ achieved an ROC AUC of 0.80 and a Brier score of 0.219.
• The FRS red achieved an ROC AUC of 0.82 and a Brier score of 0.220.
• Both scores achieved 100% sensitivity for ruling out risk (NPV 100%). The FRS red demonstrated superior specificity (94%) and PPV (50%) for ruling in risk compared to rFRS₅ (68% specificity, 40% PPV).

Conclusions:

• Both the rFRS₅ and FRS red scores effectively stratify fetal risk in pregnancy.
• The simpler FRS red score exhibited superior performance for ruling in adverse fetal outcomes, particularly in resource-limited settings.
• Further validation in larger, prospective studies is recommended to confirm these findings and clinical applicability.