Acurácia dos testes ultrassonográficos para predição de pré-eclâmpsia

Abstract

Objective : To develop a meta-analysis on ultrasonographic methods for preecalmpsia prediction.Methods : The databases MEDLINE, EMBASE, LILACS and SCIELO were searched for selecting studies and data extraction. We assessed study validity by QUADAS II. The statistical analysis was performed using summary operating point for qualitative variables and using summary ROC curves for quantitative variables.Results : In the gestation’s first trimester, ultrasound methods are not applicable for preeclampsia prediction due to low sensitivity. In the second trimester there are three feasible tests. For early preeclampsia prediction (birth < 34 weeks) the uterine arteries mean pulsatility index above 95th centile plus bilateral notching have a low negative likelihood ratio (0.28) and high sensitivity (89%). For preeclampsia prediction at any time of gestation, resistance index may be used. Mean resistance index > 0.58 is applicable for high risk pregnant women with an AUC of 0.804. For pregnant women without risk for preeclampsia, the specificity of the bilateral resitance index > 0.70 was 97%.Conclusion : In the first trimester ultrasound methods can not be used to predict preeclampsia. During the second trimester, the mean pulsatility index above 95th centile has the best predictive power. This test is applicable for low risk patients to predict preeclampsia below 34 weeks. The mean resistance index > 0.58, with sensitivity and specificity around 74%, to predict preeclampsia at any time of gestation in high risk patients, can also be used. For low risk patients and preeclampsia prediction at any time of gestation , all tests had low sensitivities.