External validation of the preeclampsia predictive model of the Hospital Clinic of Barcelona in a second-level unit in Guayaquil, Ecuador

Abstract

Introduction: The objective of this present study was to evaluate the operational capacity of the Hospital Clinic of Barcelona's predictive algorithm for detecting preeclampsia (PE) risk in a cohort of pregnant women in a secondary care level setting in Guayaquil, Ecuador.

Methods: A retrospective cohort of 304 pregnant women between 2011 and 2019 was analyzed. The algorithm used a combination of biomarkers, including sFlt-1, PlGF, and uterine artery Doppler, to predict the development of PE. The algorithm's performance was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive and negative predictive values.

Results: The algorithm demonstrated a high predictive capacity with an AUC of 0.924, a sensitivity of 88.46%, and a specificity of 91.37%. Implementing a cut-off point of 0.75 optimized the identification of PE cases. However, using biomarkers could detect only 15% to 30% of cases that would not be identified in high-risk patients.

Conclusions: The predictive algorithm effectively detects PE in Ecuador's secondary care level setting. The study suggests the advantage of stratifying women with biomarkers in high-risk groups to optimize precision without applying universal screening, thus adapting resources to local needs.