The causal and mediated relationships between major depressive disorder (MDD), post-traumatic stress disorder (PTSD), anxiety disorders and premature mortality: Evidence from genome-wide association study

Abstract

Background

Observational studies have reported that patients with major depressive disorder (MDD), post-traumatic stress disorder (PTSD), and anxiety disorders have significantly higher rates of aging-related illnesses and premature mortality than the general population. However, previous studies have not systematically assessed the possible causal relationship between MDD, PTSD, and anxiety disorders and premature mortality (aging and lifespan) at the genetic level, or searched for possible mediators between them.

Methods

Summary-level GWAS data for MDD (N = 480,359), PTSD (N = 206,655), and anxiety disorders (N = 31,060) were all obtained from the Psychiatric Genomics Consortium (PGC). The GWAS data for premature mortality were obtained from two different sources for the verification of the results: with aging (mvAge, N = 1,958,774) from Zenodo and human lifespan (N = 1,012,240) from Edinbugh DataShare. The summary-level GWAS data of potential mediators were obtained from FinnGen and UK Biobank databases. We first utilized two-sample Mendelian randomization (MR) analysis to evaluate the causal relationship between MDD, PTSD, anxiety disorders and both aging (mvAge) and lifespan. Subsequently, mediation MR analysis was conducted to search for potential mediating factors. Then, we employed linkage disequilibrium score regression (LDSC) to further estimate these genetic correlations based on the initial MR results. Finally, a series of mediation MR analyses were conducted to better understand the relationship between MDD and lifespan.

Results

We found that MDD was negatively associated with both mvAge and lifespan in two-sample MR analysis, with consistent results across the two GWAS databases representing premature mortality (mvAge: β (95%CI): −0.064(−0.087–0.041), p < 0.001; human lifespan: β (95%CI): −0.081 (−0.138–0.024), p = 0.005). The causal relationship remained significant after false discovery rate (FDR) correction. The results of LDSC suggested a genetic correlation between MDD, mvAge and lifespan (mvAge: r_g (95%CI): −0.421(−0.479–0.364), p < 0.001; lifespan: r_g (95%CI): −0.175 (−0.238–0.116), p < 0.001). In mediation MR analysis, we found that coronary atherosclerosis, hypertension and major coronary heart disease events were mediators between MDD and lifespan.

Conclusion

Our comprehensive findings provide robust genetic evidence that individuals with MDD are at significantly increased risk of premature mortality. This heightened risk is closely associated with a greater susceptibility to coronary atherosclerosis, hypertension, and major coronary heart disease events, which in turn further accentuate their propensity for accelerated aging.