VALIDATION OF FULL PIERS MODEL FOR PREDICTION OF ADVERSE MATERNAL AND FETAL OUTCOMES AT A TERTIARY CARE HOSPITAL

Abstract

Objective: To validate the performance of the fullPIERS (Pre-eclampsia Integrated Estimate of RiSk) model in predicting adverse maternal and fetal outcomes in patients presenting with pre-eclampsia at a tertiary care hospital.

Method: This cross-sectional study was performed at Department of Obstetrics and Gynaecology, Mother and Child Hospital, Pakistan Institute of Medical Sciences,  Islamabad, between February and August 2024. Pregnant women diagnosed with pre-eclampsia and satisfying inclusion criteria were enrolled. We collected comprehensive data for the fullPIERS model, including demographics, clinical signs, and laboratory results. The primary endpoints were adverse maternal outcomes (such as eclampsia, HELLP syndrome, maternal mortality, placental abruption), adverse fetal/neonatal outcomes (including perinatal mortality, stillbirth, neonatal intensive care unit admission, small for gestational age), and a composite of both. The fullPIERS risk calculator was employed to determine an optimal cutoff for predicting these adverse events. The model's predictive accuracy was assessed through sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic  (AUC-ROC) curve.

Results: In this study, 100 women were screened over the course of six months, with 7 cases (7.3%) of preeclampsia (PE) identified. Among the PE cases, 56.7% were preterm, and 74.5% had severe features. The most common complications observed included HELLP syndrome (6.7%), eclampsia (3.8%), and placental abruption (2.4%). The FullPIERS assessment yielded a median score of 1.2% (range 0.45% – 2.3%), with an excellent performance in predicting adverse maternal outcomes (AUC = 0.845, 95% CI: 0.776 – 0.914, p-value < 0.01). For perinatal adverse outcomes, the performance was suboptimal (AUC = 0.699, 95% CI: 0.581 – 0.816, p-value < 0.01). The composite of maternal and perinatal adverse outcomes also showed a suboptimal performance (AUC = 0.804, 95% CI: 0.736 – 0.872, p-value < 0.01). The cutoff value that best predicted maternal adverse outcomes was 2.15%, with a sensitivity of 75% and specificity of 83%.

Conclusion: The FullPIERS model is a valid and reliable tool for predicting adverse maternal outcomes in women with preeclampsia within our tertiary care hospital setting. However, its utility for predicting fetal outcomes was found to be suboptimal, indicating that further refinement may be needed for perinatal risk prediction. Its application could aid in risk stratification and timely clinical interventions, ultimately improving maternal health. Further local validation and potential adjustments to the model may be required to enhance its accuracy, especially for predicting fetal outcomes.