First- and second-trimester tests to predict stillbirth in unselected pregnant women: a systematic review and meta-analysis

Background Several biophysical and biochemical tests have been proposed to predict stillbirth but their predictive ability remains unclear. Objective To assess the accuracy of tests performed during the first and/or second trimester of pregnancy to predict stillbirth in unselected women with singleton, structurally and chromosomally normal fetuses through use of formal methods for systematic reviews and meta-analytic techniques. Search strategy Electronic databases, bibliographies and conference proceedings. Selection criteria Observational studies that evaluated the predictive accuracy for stillbirth of tests performed during the first two trimesters of pregnancy. Data collection and analysis Two reviewers selected studies, assessed risk of bias and extracted data. Summary receiver operating characteristic curves, pooled sensitivities, specificities and likelihood ratios (LRs) were generated. Data were synthesised separately for stillbirth as a sole category and for specific stillbirth categories. Main results Seventy-one studies, evaluating 16 single and five combined tests, met the inclusion criteria. A uterine artery pulsatility index >90th centile during the second trimester and low levels of pregnancy-associated plasma protein A (PAPP-A) during the first trimester had a moderate to high predictive accuracy for stillbirth related to placental abruption, small-for-gestational-age or pre-eclampsia (positive and negative LRs from 6.3 to 14.1, and from 0.1 to 0.4, respectively). All biophysical and biochemical tests assessed had a low predictive accuracy for stillbirth as a sole category. Conclusions Currently, there is no clinically useful first-trimester or second-trimester test to predict stillbirth as a sole category. Uterine artery pulsatility index and maternal serum PAPP-A levels appeared to be good predictors of stillbirth related to placental dysfunction disorders.