Changes In Metabolite Levels After An Oral Glucose Challenge Are Associated With Incident Cardiovascular Disease And Death

Objective: We evaluated whether intra-individual changes in blood metabolites in response to an oral glucose challenge were associated with incident CVD and mortality. Methods: An oral glucose tolerance test (OGTT; 75g glucose) was administered to a subsample of non-diabetic Framingham Heart Study participants (n=361). Metabolite profiling was performed on blood samples drawn before and 2 hours post OGTT. We compared pre- and post-OGTT log2(metabolite levels) using t-tests; for each metabolite with statistically significant changes from pre- to post-OGTT (Δmetabolites), we calculated log2(post/pre) and standardized values to mean=0, standard deviation=1. We constructed multivariable-adjusted Cox models relating each Δmetabolite with CVD and death. Models were adjusted for clinical risk factors of age, sex, baseline metabolite level, diabetes, systolic blood pressure, hypertension treatment, BMI, smoking, and total/HDL cholesterol. Results: Our sample included 42% women, with a mean age of 56±9 years, and body mass index of 30.2±5.3 kg/m 2 . The pre- to post-OGTT change (Δmetabolite) was statistically significant for 170 metabolites (at FDR ≤0.05). A total of 132 CVD events and 144 deaths occurred during follow up (mean 23±5 years for CVD, 26±2 years for death). In Cox models, four Δmetabolites were associated with incident CVD and six Δmetabolites were associated with death, at P<0.05 ( Table ). Notably, baseline metabolite levels were not associated with either outcome in models excluding Δmetabolites. Significant Δmetabolites included those with established roles in cardiometabolic disease (e.g., glutamate, α-ketoglutarate, N-methylmalonamic acid) and metabolites with less defined roles (e.g., glucoronate, lipid species). Conclusion: Intra-individual changes in circulating metabolites in response to an OGTT were associated with CVD and death beyond their resting measures. Dynamic changes in metabolite levels with an OGTT have potential relevance for understanding and predicting CVD risk.