1173-P: Metabolomic Signatures of Refined Grains Intake and Risk of Gestational Diabetes

• Published in: Diabetes (New York, N.Y.) Vol. 71; no. Supplement_1
• Main Authors: ZHU, YEYI, ZHENG, SIWEN, NGO, AMANDA, CHEHAB, RANA F., WANG, EMILY Z., FIEHN, OLIVER, FERRARA, ASSIAMIRA
• Format: Journal Article
• Language: English
• Published: New York American Diabetes Association 01.06.2022
• Subjects: Diabetes mellitus; Dietary fiber; Dietary intake; Gestational diabetes; Lipid metabolism; Metabolic pathways; Metabolites; Metabolomics; Prediction models; Pregnancy; Risk factors; Sphingolipids
• ISSN: 0012-1797; 1939-327X
• DOI: 10.2337/db22-1173-P

Previous studies linked refined grains intake to risk of gestational diabetes (GDM) ; however, the underlying mechanisms remain understudied. We aimed to identify and validate refined grains-related metabolites as biomarkers for GDM. In a nested case-control study of 91 women with GDM and 180 non-GDM controls (discovery set) in the Pregnancy Environment and Lifestyle Study (PETALS) cohort, we assessed dietary intake of refined grains by a validated food frequency questionnaire and measured fasting serum untargeted metabolomics at gestational weeks (GW) 10-13. We identified metabolites associated with refined grains intake using multivariable linear regression and examined prospective associations of refined grains-related metabolites at GW 10-13 with risk of GDM in later pregnancy using conditional logistic regression. We further examined incremental predictivity of refined grains-related metabolites selected by 10-fold cross-validated LASSO regression beyond conventional risk factors in the discovery set and a random PETALS subsample (38 GDM, 336 non-GDM) as the validation set. Among 821 known metabolites, 42 (mostly lipids/lipid-like molecules and organic oxygen/acids compounds) were associated with refined grains intake in early pregnancy, of which 17 were significantly associated with GDM risk (all FDR < 0.05) . Adding seven out of 17 metabolites selected by LASSO regression to a conventional risk factor-based prediction model increased the C-statistic from 0.71 (95% CI 0.64-0.77) to 0.77 (0.71-0.83) ; P-for-difference < 0.05. Similar improvement was observed in the validation set [0.77 (0.69-0.86) to 0.81 (0.74-0.89) ; P-for-difference < 0.05]. Refined grains intake may be associated with risk of GDM via perturbed metabolic pathways of glycerolipids, glycerophospholipids, and sphingolipids. Refined grains-related metabolites might help improve prediction of GDM risk and inform diet-related preventive strategies. Further validation studies are warranted. Disclosure Y.Zhu: None. S.Zheng: None. A.Ngo: None. R.F.Chehab: None. E.Z.Wang: None. O.Fiehn: None. A.Ferrara: n/a. Funding National Institutes of Health Building Interdisciplinary Research Careers in Women's Health (BIRCWH) Program (K12HD052163) ; National Institute of Diabetes and Digestive and Kidney Diseases (K01DK120807)