BMI and lipidomic biomarkers with risk of gestational diabetes in pregnant women

Abstract Objective The study aimed to identify BMI‐related lipids and to explore the role of lipids linking BMI and gestational diabetes mellitus (GDM). Methods Plasma lipidome, height, and weight were measured in early pregnancy among 1008 women. Pearson correlation analyses and least absolute shrinkage and selection operator regression (LASSO) were performed to identify BMI‐associated lipids. Based on these lipids, a lipid score was created using LASSO, and its association with GDM risk was evaluated by conditional logistic regression. The causal relationships between BMI and lipids were tested by Mendelian randomization analysis with genotyping data. Mediation analysis was conducted to evaluate the mediating effect of lipids on the association of BMI with GDM. Results Of 366 measured lipids, BMI was correlated with 28 lipids, which mainly belong to glycerophospholipids and glycerolipids. A total of 10 lipid species were associated with BMI, and a lipid score was established. A causal relationship between BMI and lysophosphatidylcholine 14:0 was observed. The lipid score was associated with a 1.69‐fold increased risk of GDM per 1‐point increment (95% CI: 1.33‐2.15). Furthermore, BMI‐associated lipids might explain 66.4% of the relationship between BMI and GDM. Conclusions Higher BMI in early pregnancy was associated with altered lipid metabolism that may contribute to the increased risk of GDM.