An atlas on multitudinous risk factors associated with incident hypertension: comprehensive exposome-wide association and wide-angled genetic analyses

Abstract

Abstract Aims Despite numerous risk factors being associated with hypertension, the breadth of research remains constrained, with a notable absence of systematic, data-driven exploration into established and novel factors across a broad spectrum of exposures. This study aims to construct an atlas on known and emerging factors for hypertension through comprehensive epidemiological and genetic analyses. Methods and results We conducted exposome-wide association studies (ExWAS) via Cox regression models on two equally sized datasets for discovery and replication in UK Biobank, a large prospective cohort study. A maximum of 10 806 exposome variables were included in ExWAS and were grouped into 13 categories: genomics, sociodemographic, lifestyle, physical measure, biomarkers, medical history, imaging markers, sex-specific factors, psychosocial factors, cognitive function indicators, local environment, family history, and early life factors. The credibility of epidemiological associations was assessed through meta-analyses. The genetic underpinnings were explored through linkage disequilibrium score regression (LDSC), quantifying global genetic correlation. Two-sample Mendelian randomization (MR) studies were conducted to investigate the causal effects of each exposure on hypertension, with co-analyses undertaken to identify associations supported by both epidemiological and genetic evidence. This study included 214 957 UK Biobank participants, hypertension-free at baseline. In our ExWAS analyses, 964 significant exposome variables were replicated. In meta-analyses, 462 were backed by convincing and highly suggestive evidence. Among 10 765 exposures in LDSC, 1923 had global genetic correlations with hypertension. The MR analyses yielded robust evidence for a causal relationship with 125 phenotypes, probable evidence for 270 phenotypes, and suggestive evidence for 718 phenotypes. Co-analyses identified 146 associations supported by strong epidemiological and genetic evidence. These primarily encompassed traits like anthropometry, lung function, lipids, and factors such as urate and walking pace. This coverage further extended from well-studied factors (like body mass index and physical activity) to less explored exposures (including high light scatter reticulocyte count and age at first live). All study results are compiled in a webserver for user-friendly exploration of exposure-hypertension associations. Conclusion This study provides an atlas on established and novel risk factors for hypertension, underpinned by epidemiological and causal evidence. Our findings present multiple perspectives to prioritize hypertension prevention strategies, encompassing modifiable risk factors like television watching time and walking pace. The study also emphasized the roles of urate in hypertension pathogenesis. Consequently, our study may serve as a critical guide for hypertension prevention and bear significant clinical implications.