PLASMA METABOLOMIC PROFILING PREDICTS AORTIC STIFFNESS AND RISK OF INCIDENT CARDIOVASCULAR DISEASE AND MORTALITY AFTER 16 YEARS OF FOLLOW-UP IN THE GENERAL POPULATION

Objective:
The mechanisms and metabolomic profile underlying early vascular ageing have not been explored in the general population. We aimed to investigate the plasma metabolomic profile in aortic stiffness (vascular ageing) and associated risk of incident cardiovascular disease and mortality in the general population.

Design and method:
We analysed 108 metabolites in plasma of 5172 individuals from two prospective population-based cohorts with over 16 years of follow-up. Aortic stiffness was assessed by carotid-femoral pulse wave velocity (PWV). Metabolite-predicted PWV was calculated in the older population based on the LASSO regression model by randomly assigning a training set (80%) and validation set (20%) in the younger cohort. Correlations between metabolite-predicted PWV and traditional cardiovascular risk factors were assessed by Spearman. Cox regression was used to calculate risk of incident cardiovascular disease and mortality, and all-cause mortality. All models were adjusted for age, sex and traditional cardiovascular risk factors.

Results:
Study characteristics are displayed in Table 1. Aortic stiffness was positively associated with following metabolites: long-chain acyl carnitines (C16:0, C16:1 and C13:0), dimethylguanidino valeric acid (DMGV), glutamate and cysteine, while serine was negatively associated with aortic stiffness (Figure 1A). The plasma metabolome predicted aortic stiffness in the validation cohort with good accuracy (R2 0.25, Figure 1B). Metabolite-predicted aortic stiffness was significantly correlated with age (r = 0.68), SBP (r = 0.49), BMI (r = 0.43), and sex (r = 0.20) in both cohorts cross-sectionally. During an average follow-up of 16 years, aortic stiffness predicted by baseline metabolites was associated with increased risk of incident coronary artery disease, incident stroke, cardiovascular mortality, and all-cause mortality in the older population (Figure 1C).

Conclusions:
For the first time, we show that aortic stiffness (vascular ageing) is predicted by altered metabolism of glutamate, acylcarnitine, cysteine, serine, and DMGV in the general population. These metabolic disturbances precede aortic stiffness by several years and are also associated with increased cardiovascular risk factor burden, and future risk of coronary artery disease, stroke, cardiovascular mortality, and all-cause mortality. Our results highlight the relevance of investigating the molecular pathways related to aortic stiffness. Further metabolomics studies are warranted to verify these findings in independent populations.