Are tests for predicting pre-eclampsia good enough to make screening viable? A review of reviews and critical appraisal.

The aim of this article is to review the accuracy of tests purported to be predictive of pre-eclampsia, a major cause of maternal and perinatal mortality and morbidity worldwide. A review of systematic reviews was done. A total of 219 studies were evaluated for the accuracy of 27 tests for predicting pre-eclampsia. Study quality assessment and data abstraction were performed using piloted proformas. Bivariate meta-analyses were used to synthesize data. Levels of sensitivity and specificity were measured. There were deficiencies in many areas of methodology including blinding, test description, and reference standard adequacy. No test had a high level of both sensitivity and specificity of greater than 90%. Where multiple studies were available, only BMI > 34, alpha-fetoprotein, fibronectin (cellular and total), and uterine artery Doppler (bilateral notching) measurements reached specificity above 90%. Only Doppler (any/unilateral notching, resistance index, and combinations) measurements were over 60% sensitive. Studies were of variable quality and most tests performed poorly. Further research should focus on tests which offer much higher levels of sensitivity than tests currently available. High sensitivity is a more useful attribute in early detection of pre-eclampsia than specificity because consideration of benefits, harms and costs indicates a much greater preference for minimizing false negatives than false positives, although the ideal would be to avoid both.