Influência das características maternas e pressão arterial média no primeiro trimestre gestacional na predição de pré-eclâmpsia

Abstract

Introduction: According to a formal international recommendation, it is necessary and possible to predict pre-eclampsia to obtain more favourable maternal and neonatal results. Objectives: To assess the accuracy of maternal characteristics and average blood pressure as predictors of pre-eclampsia in the first trimester; to create an algorithm that can predict pre-eclampsia in the local population; and to compare the accuracy of the created prediction model with the models suggested by the National Institute of Clinical Excellence (NICE) and the American College of Obstetricians and Gynaecologists (ACOG). Methods: This is a methodological and technological development study based on two cohort studies conducted between August 2009 and January 2014 in the Fortaleza General Hospital (HGF) and the Gonzaga Mota de Messejana District Hospital (HDGMM). The sample consisted of 733 women who were between 11 and 13 weeks pregnant, of which 55 developed pre-eclampsia (PE). Of these women, 21 had preterm PE and 34 had term PE. Data were obtained from maternal characteristics (MC) and average arterial pressure (MAP). Logistic regression analysis determined the contribution of the tests in predicting early PE. The detection rate (DR) for 5 and 10% of false positives (FPR) and the areas under the ROC curve (AUC) were obtained to measure accuracy. Results: Variables such as prior PE, family history of PE, BMI, and MAP proved useful as predictive parameters and were considered in the calculation of risk. For total PE, the AUC of the MC+MAP model was 0.787 (CI 95% 0.756 - 0.816), and the DR was 22% and 44% for a FPR of 5% and 10%, respectively. For preterm PE, the AUC of the MC+MAP model was 0.842 (CI 95% 0.814 - 0.868), and the DR was 43% and 67% for a FPR of 5% and 10%, respectively. The AUC values (0.562 for ACOG and 0.657 for NICE) for preterm PE undermine the accuracy of these PE prediction models for this population. Conclusion: The association of MC+MAP can appropriately predict PE in the first trimester of pregnancy among the studied population.