Whether hip osteoarthritis (OA) could increase the risk of lacunar stroke (LS) is not well understood. This two-sample Mendelian randomization (MR) study aimed to investigate in depth the effect of genetically predicted hip OA on LS risk. Hip OA-related instrumental variables (IVs) were selected from a genome-wide association study (GWAS) of 393,873 individuals. The summary data of LS were obtained from a GWAS meta-analysis, including 16,030 cases and 248,929 controls. We used the inverse-variance weighted (IVW) as the primary MR analysis method. Moreover, the weighted-median, MR-Egger regression, and the MR pleiotropy residual sum and outlier (MR-PRESSO) test were supplementary methods. The sensitivity analysis was performed using the leave-one-out test. We identified the positive causal relationship between hip OA and the risk of LS (odds ratio [OR] = 1.20, 95% confidence interval [CI]: 1.07, 1.36; p = 0.002 using the IVW method). The weighted median method provided similar results. There was no evidence of directed pleiotropy, and sensitivity analysis results were stable, suggesting the robustness of our study. This study showed a causal effect of hip OA on the risk of LS, and more efforts should be made to explore the potential mechanisms in the future.