Objective: This two-sample Mendelian randomization (MR) study aimed to examine the potential causal association of telomere length (TL) with the risk of osteoarthritis (OA).Method: The summary-level data for OA was derived from the United Kingdom Biobank cohort, including 50,508 individuals of European descent. Eighteen single nucleotide polymorphisms associated with TL were identified as instrumental variables from the most up-to-date TL genome-wide association study (GWAS) involving over 78,592 individuals of European descent. Based on the GWASs data, MR was performed using established statistical analysis methods including the inverse variance weighted, weighted median, MR-Egger, and MR pleiotropy residual sum and outlier.Results: Genetically determined TL was not associated with the risk of total OA (IVW odds ratio [OR] = 1.00, 95% confidence interval [CI] = 0.83, 1.21). In subgroup analyses stratified by OA site, no evidence in favor of association between genetically determined TL and knee OA was found (IVW OR = 1.18, 95% CI = 0.89, 1.58). However, using WM method, we observed a limited protective effect of longer TL on the risk of hip OA (OR = 0.60, 95% CI = 0.36–0.99), whereas the results of the IVW (p = 0.931) and MR-PRESSO (p = 0.932) showed that TL had no effect on hip OA.Conclusions: This study does not support a causal association between TL and total OA. A potential protective association between longer TL and hip OA, though possible, remains less certain.