First-Trimester Prediction of Preeclampsia in Nulliparous Women at Low Risk

OBJECTIVE: To identify clinical characteristics and biochemical markers in first-trimester samples that would possibly predict the subsequent development of preeclampsia. METHODS: We conducted a multicenter observational study in 2,434 nulliparous women at low risk to identify biomarkers that possibly predict preeclampsia. Clinical history, complete blood count, and biochemical markers were assessed in the first trimester. The trophoblast and angiogenesis markers ADAM-12, pregnancy-associated plasma protein-A, placental protein 13, placental growth factor, soluble fms-like tyrosine kinase-1, and endoglin were measured in a case-control subset of 174 women with preeclampsia and 509 women in the control group. RESULTS: Univariable analysis revealed maternal age, race, marital status, years of education, source of medical payment, prenatal caregiver, body mass index (BMI, calculated as weight (kg)/[height (m)] 2), and systolic blood pressure at enrollment were significantly associated with preeclampsia. Mean platelet volume was greater at enrollment in women who later had development of preeclampsia (median 9.4 compared with 9.0 femtoliter (fl); P=.02). First-trimester concentrations (multiples of the median) of ADAM-12 (1.14 compared with 1.04; P=.003), pregnancy-associated plasma protein-A (0.94 compared with 0.98; P=.04), and placental growth factor (0.83 compared with 1.04; P<.001) were significantly different in women who had development of preeclampsia compared with women in the control group. The optimal multivariable model included African American race, systolic blood pressure, BMI, education level, ADAM-12, pregnancy-associated plasma protein-A, and placental growth factor, and yielded an area under the curve of 0.73 (95% confidence interval 0.69-0.77) and a sensitivity of 46.1% (95% confidence interval 38.3-54.0) for 80% specificity. CONCLUSION: A multivariable analysis of clinical data and biochemical markers in the first trimester did not identify a model that had clinical utility for predicting preeclampsia in a nulliparous population at low risk. (Obstet Gynecol 2012; 119: 1234-42) DOI: 10.1097/AOG.0b013e3182571669