Metabolomics. 2023 Mar 24;19(4):20. doi: 10.1007/s11306-023-01994-z.

ABSTRACT

INTRODUCTION: Aberrations in circulating metabolites have been associated with diabetes and cardiovascular risk.

OBJECTIVES: To investigate if early and late pregnancy serum metabolomic profiles differ in women who develop prediabetes by two years postpartum compared to those who remain normoglycemic.

METHODS: An NMR metabolomics platform was used to measure 228 serum metabolite variables from women with pre-pregnancy overweight in early and late pregnancy. Co-abundant groups of metabolites were compared between the women who were (n = 40) or were not (n = 138) prediabetic at two years postpartum. Random Forests classifiers, based on the metabolic profiles, were used to predict the prediabetes status, and correlations of the metabolites to glycemic traits (fasting glucose and insulin, HOMA2-IR and HbA1c) and hsCRP at postpartum were evaluated.

RESULTS: Women with prediabetes had higher concentrations of small HDL particles, total lipids in small HDL, phospholipids in small HDL and free cholesterol in small HDL in early pregnancy (p = 0.029; adj with pre-pregnancy BMI p = 0.094). The small HDL related metabolites also correlated positively with markers of insulin resistance at postpartum. Similar associations were not detected for metabolites in late pregnancy. A Random Forests classifier based on serum metabolites and clinical variables in early pregnancy displayed an acceptable predictive power for the prediabetes status at postpartum (AUROC 0.668).

CONCLUSION: Elevated serum concentrations of small HDL particles in early pregnancy associate with prediabetes and insulin resistance at two years postpartum. The serum metabolic profile during pregnancy might be used to identify women at increased risk for type 2 diabetes.

PMID:36961590 | PMC:PMC10038958 | DOI:10.1007/s11306-023-01994-z