ObjectiveTo conduct a Mendelian randomization (MR) analysis to elucidate the potential causal relationship between sarcopenia (SA) and knee osteoarthritis (KOA). MethodsThree SA-related traits were selected as exposure factors from the summary data of the genome-wide association studies database (IEU GWAS). KOA and hospital-diagnosed osteoarthritis of the knee (osteoarthritis of the knee hospital diagnosed) were chosen as outcome factors. The inverse variance-weighted (IVW) method was employed as the primary analytical approach to evaluate the causal relationship between SA and KOA. Heterogeneity tests, sensitivity analyses, and pleiotropy analyses were conducted to validate the reliability of the results. ResultsThe MR results indicated a substantial causal relationship between genetically predicted appendicular muscle mass(OR=1.079, 95%CI 1.015 to 1.147, P=0.015 5), walking speed(OR=0.157, 95%CI 0.101 to 0.248, P<0.001). No significant causal relationship was found between grip strength and KOA (OR=1.318, 95%CI 0.933 to 1.859, P=0.116 6), and the sensitivity analysis results did not exhibit horizontal pleiotropy. ConclusionSA may have a causal relationship with KOA, and appendicular muscle mass and walking speed may be risk factors for the occurrence and development of KOA.