Evaluation of first trimester maternal serum inhibin-A for preeclampsia screening.

BACKGROUND

International professional organizations recommend aspirin prophylaxis to women screened high risk for preterm preeclampsia (PE) in the first trimester. The UK Fetal Medicine Foundation (FMF) screening test for preterm PE using mean arterial pressure (MAP), uterine artery pulsatility index (UTPI) and placental growth factor (PlGF) was demonstrated to have lower detection rate (DR) in Asian population studies. Additional biomarkers are therefore needed in Asian women to improve screening DRs as a significant proportion of women with preterm and term PE are currently not identified.

OBJECTIVES

To evaluate maternal serum inhibin-A at 11-13 weeks as an alternative to PlGF or as an additional biomarker within the FMF screening test for preterm PE.

STUDY DESIGN

This is a nested case-control study using pregnancies initially screened at 11-13 weeks for preterm PE using the FMF triple test in a non-intervention study conducted between December 2016 and June 2018. Inhibin-A levels were retrospectively measured in 1,792 singleton pregnancies, 112 (1.7%) with PE matched for time of initial screening with 1,680 unaffected pregnancies. Inhibin-A levels were transformed to multiple of the expected median (MoM). The distribution of log10 inhibin-A MoM in PE and unaffected pregnancies and the association between log10 inhibin-A MoM and gestational age (GA) at delivery in PE were assessed. The screening performance determined by area under receiver operating characteristic curves (AUC) and detection rates (DRs) at a 10% fixed false positive rate (FPR), for preterm and term PE was determined. All risks for preterm and term PE were based on the FMF competing risk model and Bayes theorem. Differences in AUC (ΔAUC) between different biomarker combinations were compared using the Delong test. McNemar's test was used to assess the off-diagonal change in screening performance at a fixed 10% FPR after adding inhibin-A or replacing PlGF in the preterm PE adjusted risk estimation model.

RESULTS

Inhibin-A levels in unaffected pregnancies were significantly dependent on GA, maternal age and weight and were lower in parous women with no previous history of PE. Mean log10 inhibin-A MoM in any-onset PE (p<0.001), preterm (p<0.001) and term PE (p = 0.015) pregnancies were all significantly higher than that of unaffected pregnancies. Log10 inhibin-A MoM was inversely but not significantly correlated (p = 0.165) with GA at delivery in PE pregnancies. Replacing PlGF with inhibin-A in the FMF triple test reduced AUC and DR from 0.859 and 64.86% to 0.837 and 54.05%, the ΔAUC was not statistically significant. AUC and DR when adding inhibin-A to the FMF triple test were 0.814, 54.05% and the -0.045 reduction in AUC was statistically significant (p = 0.001). At a fixed 10% FPR, replacing PlGF with inhibin-A identified 1 (2.7%) additional pregnancy but missed 5 (13.5%) pregnancies which subsequently developed preterm PE identified by the FMF triple test. Adding inhibin-A missed 4 (10.8%) pregnancies and did not identify any additional pregnancies with preterm PE.

CONCLUSION

Replacing PlGF by inhibin-A or adding inhibin-A as an additional biomarker in and to the FMF triple screening test for preterm PE does not improve screening performance and will fail to identify pregnancies that are currently identified by the FMF triple test.

Informations de copyright

Copyright: © 2023 Moungmaithong et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Déclaration de conflit d'intérêts

LCP has received speaker fees and consultancy payments from Roche Diagnostics and Ferring Pharmaceuticals. LCP and DS have received in‐kind contributions from Roche Diagnostics, PerkinElmer, Thermo Fisher Scientific, and GE Healthcare. This does not alter our adherence to PLOS ONE policies on sharing data and materials. Other co-authors report no conflicts of interest.