Risk factors for arterial stiffness in a multi-ethnic Asian population: Results from the Health for Life in Singapore Study

Abstract Background Arterial stiffness (AS) measured by pulse wave velocity (PWV) is a well-recognized marker for increased cardiovascular disease risk in Asian and Western individuals. We evaluated factors in relation to AS using the exposure-wide association study (EWAS) approach within the Health for Life in Singapore (HELIOS) study. Subsequently, two-sample Mendelian randomization (MR) analysis was utilized to examine potential causal associations between identified factors and AS. Methods The HELIOS study is a multi-ethnic prospective cohort study that included adults aged 30-85 years. We analyzed 8,880 subjects of Chinese, Malay, and South Asian ethnicity participating in the HELIOS cohort. PWV was measured by Vicorder. We assessed 54 predictor variables using EWAS approach and correction for multiple comparisons was applied using the false discovery rate (FDR < 5%) method. Linear regression models were used to assess the association of each predictor variable with PWV. Inverse-variance-weighted MR method was used to estimate causal effects. Results Of the 54 variables, 41 variables were significantly associated with PWV at FDR <5%. We found age, systolic blood pressure (SBP), diastolic blood pressure, and anthropometric, and body composition measurements were positively associated with PWV. Furthermore, insulin, glycated hemoglobin (HbA1c), glucose, triglycerides, albumin, alanine transaminase, gamma-glutamyl transferase, C-reactive protein, and uric acid were positively associated with PWV. Inverse associations were observed for PWV with high-density lipoprotein cholesterol (HDL-C), potassium, vitamin D, bilirubin, handgrip strength and educational level. These associations appear stronger among Chinese participants than among Malay and South Asian participants. In MR analyses, genetically predicted SBP was positively with AS, while genetically predicted HDL-C and HbA1c were inversely associated with AS. Conclusions Our systematic evaluation provides new knowledge on the complex array of anthropometric, body composition, physiological, biochemical, behavioral, and sociodemographic correlates of PWV. Our MR analysis indicates a potential causal association between SBP, HDL-C, HbA1c and AS.