Abstract 147: Copeptin and Mid-Regional Pro-ANP in Women With Suspected or Confirmed Preeclampsia: Comparison With the sFlt-1/PlGF Ratio

Abstract

Both arginine vasopressin (AVP) and atrial natriuretic peptide (ANP) are believed to contribute to the pathogenesis of preeclampsia (PE), although current evidence is inconclusive. Here we studied AVP and ANP as biomarkers (measured as their surrogates copeptin and mid-regional pro-ANP, MR-proANP, respectively) for the prediction of PE and adverse pregnancy outcome, making use of a large prospective cohort study in 526 women with suspected or confirmed preeclampsia (age 18 - 48) originally aimed to evaluate the use of sFlt-1/PlGF ratio for the prediction of PE. Women with confirmed PE (n = 134) displayed higher copeptin (median 6 [range 1 - 61] vs. 4 [1 - 42] pM; P<0.01) and MR-proANP (52 [2 - 342] vs. 31 [2 - 298] pM; P<0.001) levels than 250 women with suspected PE. Yet, when subdividing the women according to sFlt-1/PlGF ratio (≥85 versus <85), no differences in copeptin levels were observed (5 vs. 5 pM), while MR-proANP levels remained elevated in women with a sFlt-1/PlGF ratio ≥85 (64 vs. 32 pM; P<0.0001). In accordance with this observation, copeptin correlated only marginally with sFlt-1 (r = 0.17), and not with PlGF, while MR-proANP correlated strongly with both, sFlt-1 (r= 0.30) and PlGF (r=-0.29). To discriminate maternal complications on top of traditional predictors (gestational age, parity, diastolic blood pressure and proteinuria), the sFlt-1/PlGF ratio displayed a higher C-index than copeptin and MR-proANP (0.74, 0.63 and 0.68, respectively, vs. 0.62 for the traditional predictors only), and the same was true for the discrimination of fetal/neonatal complications (C-index = 0.81, 0.76 and 0.78 vs. 0.76). Finally, multiple regression analysis revealed that both copeptin (P=0.015) and MR-proANP (P=0.024) were independent determinants of proteinuria but not blood pressure. In conclusion, both copeptin and MR-proANP levels are elevated in women with PE but have limited value to predict pregnancy complications when compared with the sFlt-1/PlGF ratio. Interestingly, copeptin (but not MR-proANP) effects appear to be largely independent from sFlt-1/PlGF, and predominantly reflect the renal phenotype in this condition.