OBJECTIVE: This study aimed to elucidate the bidirectional causal relationships between Alzheimer's disease (AD), cerebral small vessel disease (CSVD), and the effect of inflammatory cytokines on AD and CSVD using Mendelian randomization (MR). METHOD: We employed publicly available summary-level data from genome-wide association studies for AD, CSVD, and 91 inflammatory cytokines. Genetic variants strongly associated with each risk factor were selected as instrumental variables. The inverse variance weighted (IVW) method was primarily used for causal inference, with sensitivity analyses including MR-Egger and weighted median estimators. RESULTS: MR analysis revealed that genetically predicted CSVD significantly increased the risk of AD (odds ratio [OR] = 1.035, 95% CI, 1.015-1.056, P = 0.001). Conversely, AD did not significantly influence CSVD risk (OR = 0.878, 95% CI, 0.701-1.100, P = 0.257). Among inflammatory cytokines, Axin1 (OR = 1.082, 95% CI, 1.009-1.159, P = 0.026) and bNGF (OR = 1.061, 95% CI, 1.001-1.125, P = 0.048) increased AD risk, while CD5 (OR = 0.937, 95% CI, 0.887-0.991, P = 0.022) and CXCL11 (OR = 0.951, 95% CI, 0.912-0.992, P = 0.019) decreased AD risk. FGF19 (OR = 0.560, 95% CI, 0.405-0.773, P <
0.001) and TNFSF14 (OR = 0.744, 95% CI, 0.580-0.954, P = 0.020) were protective against CSVD. CONCLUSION: Our findings suggest that CSVD may increase AD risk, while specific inflammatory cytokines exhibit differential associations with these conditions. Targeting vascular health and inflammation may offer promising therapeutic avenues for managing neurodegenerative diseases.