(1143) Maternal Lipidomics for the accurate prediction of Fetal Cyanotic Congenital Heart Defects

Objective: Late diagnosis of cyanotic congenital heart defects (CCHD) significantly increases newborn morbidity and mortality. However, a significant percentage of CCHD cases remain undetected prenatally. Our objectives were therefore to evaluate the accuracy of maternal serum lipidomic markers for the prenatal detection of CCHD and also to elucidate the pathogenesis of these disorders.

Study Design: This prospective case-control study included 40 cases of CHD (22 cyanotic, 18 acyanotic) and 39 unaffected controls. Maternal serum samples were profiled via a combination of Ultra-High-Performance Liquid Chromatography coupled with Mass Spectrometry and Nuclear Magnetic Resonance. The mean of each lipid specie was compared between CCHD vs all others. Logistic regression models were developed for the detection of CCHD. Areas under the receiving operating curve (AUC), sensitivity, and specificity values were calculated. Metabolite Set Enrichment Analysis (MSEA) was also performed to discover the top dysregulated biochemical pathways in CCHD.

Results: In total, 803 lipids and 97 metabolites were profiled in the maternal serum. In CCHD cases, concentrations of ceramides, phosphatidic acids, phosphatidylcholines, and phosphatidylethanolamines were decreased while fatty acids and phosphatidylserines were elevated (p< 0.05) (Fig1). The lipidomics model (Fig2) achieved outstanding predictive accuracy: AUC (95%) = 0.959 (0.946-0.972). This outperformed the accuracy of routinely used standard clinical risk predictors for CHD (diabetes + obesity + hypertension + in vitro fertilization + family history of CHD + alcohol or tobacco exposure). MSEA analysis revealed profound alterations of mitochondrial beta-oxidation of short and long-chain fatty acids involved in lipid metabolism, in CCHD.

Conclusion: Fetal CCHD was associated with significant disturbance of lipid metabolism. Maternal lipid biomarkers were highly accurate for the detection of fetal CCHD. This has the potential for the development of non-invasive pregnancy population screening which could help to reduce mortality and morbidity in CCHD.