Article Text
PDF
Abstract
Background and Purpose Left ventricular diastolic dysfunction (LVDD) is a key pathophysiological mechanism in heart failure with preserved ejection fraction (HFpEF) but its environmental determinants are poorly understood. Environment-wide association studies (EWAS) provide a comprehensive method to test a variety of exposures across the human environment and life-course in a high-throughput, unbiased manner. We conduct the first life-course EWAS for LVDD.
Methods Participants were from the Medical Research Council (MRC) National Survey of Health and Development (NSHD, the British 1946 birth cohort) who had echocardiographic data recorded at age 60-64 years. LVDD (outcome) was defined by the presence of ≥2 abnormal echocardiographic parameters out of: left atrial volume index, E/e’, septal e’, lateral e’, and E/A, with normal cut-off values obtained from the American Society of Echocardiography guideline for LVDD diagnosis. 326 life course environmental factors (exposures) were investigated for their association with LVDD. Significant factors were identified using a logistic regression model adjusting for sex, body mass index and socioeconomic position (SEP), and a false discovery rate of 5%. Interactions between individual exposures were appraised using exposome correlation globes and matrices, and a principal component analysis.
Results A total of 1616 participants were included (50.4% men, 21.4% with LVDD). We discovered 26 factors independently associated with LVDD (p≤ 0.05) (figure 1). Significant factors from 0-18 years included childhood cognition (odds ratio [OR]: 0.83; 95% confidence interval [CI]: 0.69-0.99), quality of home conditions (OR: 0.90; 95% CI: 0.83-0.97), crowding of childhood dwelling (OR: 1.17; 95% CI: 1.05-1.30) and father’s SEP (OR: 0.74; 95% CI: 0.56-0.98). Childhood cognition displayed inter-domain positive correlations with father’s SEP and housing quality. From 19-44 years, significant factors were reading test performance (OR: 0.96; 95% CI: 0.93-0.99), oily fish consumption (OR: 0.99; 95% CI: 0.98-1) and canned fruit consumption (OR: 0.99; 95% CI: 0.98-1.00). Other significant factors were systolic blood pressure from 45-59 years (OR: 1.01; 95% CI: 1.00-1.02), and tissue plasminogen activator (OR: 1.04; 95% CI: 1.01-1.08) and urine creatinine from 60-64 years (OR: 0.96; 95% CI: 0.93-1). (Figure 2) illustrates the correlations between all exposures relevant to ages 0-18 years.
Manhattan plot displaying multivariable environment-wide associations with LVDD. Y-axis displays the -log10(p-value) of the logistic regression coefficient foc each factor tested. X-axis indicates the exposure domains. 26 factors above the red line are significant at the p ≤ 0.05 level. bpprobJems = blood pressure problems, DBP = diastolic blood pressure, GHQ1162 = taking longer to do things, GHQ12.62 = doing things well, GHQ13 = satisfied with tasks, GHQ2162 = considering yourself a wotlhless person, SBP = systolic blood pressure, tPA = tissue plasminogen activator
Exposome correlation globes for all exposures relevant to ages 0–18. Domains (key is on the right) and correlations are colour coded. Each line represents a correlation between two exposures. Correlations were analysed used Spearman’s rank correlation coefficient. Irti = lower respiratory tract infection
Conclusion We have unmasked and rediscovered several exposures throughout the life course that associate with LVDD in later life, including dietary factors and cognition in childhood and young adulthood. Exposures identified in this study merit multi-cohort validation and have the potential to inspire more holistic public health efforts to tackle the emerging epidemic of HFpEF.
Conflict of Interest None