Sleep traits and physical activity mediate the causal association between depression and age-related diseases.

BACKGROUND

A growing body of evidence suggests the relationship of depression with an increased risk of age-related diseases (ARDs). To further understand the genetically predicted causative connections, a Mendelian randomization (MR) approach was conducted.

METHODS

Genetic variants associated with depression were employed as instrumental variables from the PGC and FinnGen consortium, respectively. GWAS summary data for 14 ARDs were derived from recently large consortia. We employed univariable and bidirectional MR analysis, and meta-analysis combining the results from two databases. Importantly, the potential mediation effects and mediated proportions of sleep and exercise traits were evaluated using a two-step mediation MR analysis. Finally, the robustness of all the MR results was confirmed using multiple sensitivity analyses.

RESULTS

In univariable MR analysis, results of combine effect showed that genetically linked depression was causally associated with a range of ARDs, including myocardial infarction (MI) (OR: 1.156, 95%CI: 1.038-1.288, p = 0.008), coronary atherosclerosis (CAS) (OR: 1.008, 95%CI: 1.004-1.013, p < 0.001), peripheral arteriosclerotic disease (OR: 1.002, 95%CI: 1.000-1.003, p = 0.020), obesity (OR: 1.028, 95%CI: 1.001-1.056, p = 0.044), type 2 diabetes (T2D) (OR: 1.112, 95%CI: 1.034-1.197, p = 0.004), and metabolic syndrome (OR: 1.204, 95%CI: 1.064-1.364, p = 0.003). Reverse analyses revealed the causal effect of obesity on depression (OR: 1.110, 95%CI: 1.070-1.152, p < 0.001). In mediation analyses, physical activity and/or sleep traits mediated the causal associations between depression and MI, CAS, and T2D. We further quantified the mediation effects. Sensitivity analyses supported these observations.

CONCLUSION

Depression is causally associated with MI, CAS, peripheral arteriosclerotic disease, obesity, T2D, and metabolic syndrome. Moreover, physical activity and sleep traits, either individually or in combination, appear to mediate the causal associations.