Major depressive disorder and the development of cerebral small vessel disease: A Mendelian randomization study.

BACKGROUND

Although observational studies indicate a complex, bidirectional association between major depressive disorder (MDD) and cerebral small vessel disease (CSVD), the results are frequently inconsistent. This study investigated the potential correlation of MDD with both CSVD clinical outcomes and radiological markers, utilizing a bidirectional Mendelian randomization (MR) study design.

METHODS

Instrumental variables for MDD were obtained from the latest and largest genome-wide association study (GWAS). For CSVD, we extracted genetic instruments from GWAS datasets corresponding to both clinical outcomes and radiological markers, including intracerebral hemorrhage, small vessel ischemic stroke, white matter hyperintensities volume, mean diffusivity (MD), fractional anisotropy, brain microbleeds, and enlarged perivascular space (PVS). We employed the inverse variance weighting method as the primary analysis, complemented by conducting extensive sensitivity and heterogeneity tests.

RESULTS

In the forward MR analyses, we discovered that the genetically predicted risk of MDD exhibits a potential causal relationship with two CSVD phenotypes demonstrating microscopic white matter (WM) damage: mean diffusivity (β = 0.784, 95 % CI 0.285-1.283, p = 0.002) and WM-PVS (OR = 1.053, 95%CI 1.010-1.097, p = 0.015). A single SNP (rs2232423) was identified as significantly influencing the causal relationship between MDD and WM. After excluding this SNP, our estimated association between MDD and increased MD (β = 0.516, 95%CI -0.001-1.033, p = 0.048) remained. The effects of MDD on WM-PVS passed all the tests for heterogeneity and pleiotropy. Reverse MR analyses showed no evidence of reverse causality between MDD and an altered CSVD risk.

CONCLUSIONS

This study supports a potential causal association between MDD and CSVD-related indicators of impaired WM microstructure. These insights hold promise for improving risk assessment methods in CSVD.