Abstract 024: Metabolomics and Incident Hypertension in African Americans from the Atherosclerosis Risk in Communities (ARIC) Study

Abstract

Introduction: Development of hypertension is influenced by genes, environmental effects and their interactions, and the human metabolome is a measurable manifestation of gene x environment interaction. We explored the metabolomic antecedents of hypertension in a sample of African Americans (AAs) in the ARIC study. Methods: We examined 896 (565 females, aged 45-64 years) baseline normotensives, whose serum metabolome was measured by gas chromatography-mass spectrometry/ liquid chromatography-mass spectrometry. The analyses presented here focus on 204 metabolites having high 4-6 week intra-individual repeatability (reliability coefficients ≥ 0.60). We used principal-components analyses (PCA) to identify patterns in these metabolomic data. Weibull parametric models considering interval censored data were used to assess the hazard ratio of developing hypertension. The analyses adjusted for age, gender, leisure-time physical activity, smoking status, alcohol intake, diabetes status, BMI, and estimated glomerular filtration rate in the full model. An additional model adjusted for the previous variables plus baseline systolic and diastolic BP. We used a modified Bonferroni correction accounting for the correlations among metabolites to determine statistical significance (p <3.9×10-4) in the analysis of association between an individual metabolite and hypertension. Results: During 10-years of follow-up, 38% of baseline normotensives developed hypertension (N=344; incidence rate: 63.94 per 1000 person-years). Each SD of baseline 4-hydroxyhippurate, a product of gut microbial fermentation, was associated with 17% higher risk of hypertension (p=2.5×10-4), which remained significant after adjusting for baseline BP (p=3.8×10-4). PCA showed three interpretable metabolomic patterns: sex steroids, alpha amino acids and branched-chain amino acids. The sex steroids pattern mainly consisted of androstenediol and pregnendiol disulfate, and it was significantly associated with risk of incident hypertension independent of all factors in the full model (highest versus lowest quintile HR, 1.72; 95% CI, 1.05 to 2.82; p for trend=0.03). Stratified analyses suggested this association was consistent in both genders. Conclusions: In our study of ARIC AAs with 204 reliable metabolites, one metabolite (4-hydroxyhippurate) and one pattern of sex steroids were identified as biomarkers of incident hypertension independent of measured traditional risk factors. Metabolomic analyses such as these may identify novel biomarkers or pathways in the etiology of hypertension and point to new prevention strategies.