Prediction of early and late pre-eclampsia from maternal characteristics, uterine artery Doppler and markers of vasculogenesis during first trimester of pregnancy

Objective To develop a predictive model for pre-eclampsia using clinical, biochemical and ultrasound markers during the first trimester of pregnancy. Methods This was a nested case-control study within a pre-eclampsia screening project that involved 5367 asymptomatic pregnant women undergoing routine transvaginal uterine artery (UtA) Doppler at 11 + 0 to 13 + 6 weeks. Following exclusions, there were 70 pregnant women who later developed pre-eclampsia and 289 control patients enrolled during the first trimester who had serum or plasma samples taken at enrolment available for the purposes of this study. Of these, 17 pregnancies were diagnosed with early-onset (delivery < 34 weeks) pre-eclampsia and 53 with late-onset (delivery >= 34 weeks) pre-eclampsia. The lowest, highest and mean of left and right UtA pulsatility indices (PI) were calculated. Blood samples were stored at - 84 degrees C until biochemical analysis for markers of vasculogenesis was performed. The distributions of the lowest UtA-PI and the biochemical markers were adjusted for maternal characteristics, expressed as multiples of the median (MoM), and compared between groups. Logistic regression analysis was used to evaluate if any variable was significantly associated with pre-eclampsia. Results Pregnancies that later developed pre-eclampsia were associated with higher maternal prepregnancy body mass index. An increased lowest UtA-PI was significantly associated with both early-and late-onset disease. Placental growth factor (PlGF) MoM was significantly reduced in women who later developed early-or late-onset pre-eclampsia compared with controls (median (interquartile range), 0.69 (0.33-1.46) and 1.10 (0.39-1.56), respectively, vs 1.19 (0.65-1.84), P < 0.05). Different combined models were generated by logistic regression analysis, and the detection rate with a fixed 10% false-positive rate was 47% and 29% for early-and late-onset pre-eclampsia, respectively. Conclusion Pregnancies that later developed early or late pre-eclampsia were characterized by impaired placentation and an anti-angiogenic state during the first trimester of pregnancy. Regression models which include maternal characteristics, UtA Doppler and PlGF can apparently predict approximately half of pregnancies that will be complicated by early-onset pre-eclampsia. We believe more research in several areas is needed to aid in the creation of a better and more population-specific screening test for pre-eclampsia during the first trimester of pregnancy. Copyright. (C) 2012 ISUOG. Published by John Wiley & Sons Ltd.