Prediction of preeclampsia and induced delivery at <34 weeks gestation by sFLT-1 and PlGF in patients with abnormal midtrimester uterine Doppler velocimetry: a prospective cohort analysis.

BACKGROUND: Women with bilateral abnormal uterine artery Doppler velocimetry (UtADV) are at increased risk for an adverse pregnancy outcome. This study aimed to determine if additional assessment of midtrimester angiogenic factors improves the predictive accuracy of Doppler results for various outcome parameters.

METHODS: Women with a bilateral abnormal UtADV, which was defined as a postsystolic incision and/or an increased pulsatility index greater than the 95th centile, and a singleton pregnancy were prospectively recruited between 19 + 0 and 26 + 6 weeks of gestation. Maternal serum levels of placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFLT-1) were measured with a fully automated immunoassay and their ratio was calculated.

RESULTS: Angiogenic factors could predict the development of preeclampsia (PE), as well as induced delivery at <34 weeks of gestation, but failed to predict the development of normotensive intrauterine growth restriction. Twelve (24.0%) of the 50 recruited women developed PE. Nine of these patients had early-onset disease (<34 + 0 weeks). Six (12.0%) patients were delivered at <34 + 0 weeks. The most useful test results in the prediction of PE and induced delivery at <34 + 0 weeks were observed using the sFLT-1/PlGF >95th centile ratio with a sensitivity, specificity, positive predictive value, and negative predictive value (NPV) of 66.7%, 89.5%, 66.7%, and 89.5% for PE, and 85.7%, 86.1%, 50.1%, and 97.4% for induced delivery, respectively. Positive and negative likelihood ratios were 6.33 (95% CI 2.31-17.38) and 0.37 (95% CI 0.17-0.84) for PE, and 6.14 (95% CI 2.76-13.69) and 0.17 (0.03-1.02) for induced delivery, respectively. Corresponding odds ratios were 17.0 (95% CI 3.5-83.0) and 37.0 (95% CI 3.8-363.9), respectively.

CONCLUSIONS: Measurement of angiogenic factors improves the specificity of an abnormal UtADV for prediction of PE. Compared with prediction of PE an abnormal sFLT-1/PlGF ratio revealed higher sensitivity for prediction of induced delivery at <34 + 0 weeks. The NPV of 97% will help to reassure most patients with an abnormal UtADV and a normal sFLT-1/PlGF ratio.