Aims: To explore the maternal metabolic changes of fetal congenital heart disease (FCHD), and screen metabolic markers to establish a practical diagnostic model.
Methods: Maternal peripheral serum from 17 FCHD and 63 non-FCHD pregnant were analyzed by Ultra High-performance Liquid Chromatography-Mass/Mass (UPLC-MS/MS).
Results: In the FCHD and the non-FCHD, 132 metabolites were identified, including 35 differential metabolites enriched in the purine, caffeine, primary bile acid, and arachidonic acid metabolism pathway. Finally, the screened (+/-)9,10-dihydroxy-12Z-octadecenoic acid (AUC = 0.888) and 11,12-epoxy-(5Z,8Z,11Z)-icosatrienoic acid (AUC = 0.995) were incorporated into the logistic regression model. The AUC value of the two-metabolite model was 1.0, superior to proline (AUC = 0.867), uric acid (AUC = 0.789), glutamine (AUC = 0.705), and taurine (AUC = 0.923) previously reported. The clinical decision curve analysis (DCA) showed the highest clinical net benefit of the model, and internal validation by bootstrap shows the robustness of the model (Brier Score = 0.005).
Conclusion: For the prenatal diagnosis of CHD, our findings are of great clinical significance. As an additional screening procedure, the identification model might be used to detect.
Keywords: Biomarker; Congenital heart disease; Diagnostic model; Metabolomics; Pregnancy.
Copyright © 2023 Elsevier B.V. All rights reserved.