Abstract
Observational studies have produced inconsistent findings regarding the relationship between Alzheimer's disease (AD) and cerebral small vessel disease (CSVD) risk. Residual confounding and potential reverse causality are inevitable in such conventional observational studies. We tried to examine the causal relationship between AD and CSVD-related phenotypes using genetic methods. Genetic instruments for each AD and CSVD-related phenotypes (cerebral microbleeds, white matter hyperintensity, and lacunar stroke) were derived from large-scale genome-wide association studies. In this study, two-sample Mendelian randomization (MR) tested potential causal associations between AD and CSVD-related phenotypes, followed by a colocalization analysis to corroborate MR findings and explain possible mechanisms. Using univariable MR, we observed that genetic liability to AD was associated with an increased risk of cerebral microbleeds (CMBs) [odds ratio (OR) = 1.149; 95% confidence interval (CI) = 1.070-1.235, P < 0.001], and a modest increase in white matter hyperintensities (WMHs) volume (β = 0.031 mm(3), 95% CI = 0.009-0.054 mm(3), P = 0.005). In multivariable MR, the causal effect of genetic liability for AD on CMBs and WMHs remained after adjusting for risk factors, with the estimate across the IVW method. Colocalization results provided evidence for a shared causal variant between AD with CMBs (PPH4 = 0.996) and WMHs (PPH4 = 0.657), suggesting that the MR estimates were not confounded by linkage disequilibrium. Our MR analyses provided robust evidence for the causal effects of genetic liability for AD on an increased risk of CMBs and WMHs. More work is warranted to confirm the mechanisms of association between AD and CSVD.