Predictive values of serum metabolites in early pregnancy and their possible pathways for gestational diabetes: A nested case-control study in Tianjin, China

Aims: To investigate the associations and predictive values of serum metabolites in early pregnancy for later development of gestational diabetes mellitus (GDM), and further explore their metabolic pathways to GDM. Methods: We conducted a 1:1 nested case-control study including 486 pregnant women from Tianjin, China, and collected blood samples at their first registration (median at 10th gestational week). Liquid chromatography-tandem mass spectrometry was used to measure serum metabolites. Orthogonal partial least squares discriminant analysis was used to select specific metabolites associated with GDM, and pathway analysis was used to identify the metabolic pathways related to GDM. Results: A total of 64 serum metabolites were included in this analysis, 17 of which were identified as specific metabolites associated with GDM. Ten metabolites increased and seven metabolites decreased GDM risk. Inclusion of these specific metabolites to the model of traditional risk factors greatly increased the predictive value from 0.69 (95% confidence interval: 0.64–0.74) to 0.92 (0.90–0.95). In addition, we found that glycerophospholipid metabolism, sphingolipid metabolism and primary bile acid biosynthesis were main metabolic pathways related to GDM. Conclusion: We identified a set of serum metabolites and their metabolic pathways in early pregnancy associated with GDM, which provided a theoretical basis for further research on the molecular pathways to GDM and early identification of GDM.