OBJECTIVE: In this Mendelian randomisation (MR) study, we investigate the causal effect of metabolically different adiposity subtypes on osteoarthritis. METHODS: We performed two-sample MR using summary-level data for osteoarthritis (10,083 cases and 40,425 controls) from genome-wide association using UK Biobank, and for site-specific osteoarthritis from arcOGEN consortium. We used three classes of genetic instruments, which all increase body mass index but are associated with different metabolic profiles (unfavourable, neutral and favourable). Primary analysis was using inverse variance weight (IVW), with additional sensitivity analysis from different MR methods. We further applied a nonlinear MR using UK Biobank data to understand the nature of the adiposity-osteoarthritis relationship. RESULTS: Greater metabolically unfavourable and metabolically neutral adiposity were associated with higher odds of osteoarthritis (IVW: OR 1.56 (95%CI 1.31 to 1.85) and 1.60 (1.15 to 2.23), respectively). The estimate for the association between metabolically favourable adiposity and osteoarthritis was similar, although with notable imprecision (1.55, 0.70 to 3.41). Using site-specific osteoarthritis, metabolically unfavourable, neutral and favourable adiposity were all associated with higher odds of knee osteoarthritis (1.44, 1.04 to 1.98; 2.28, 1.04 to 4.99; and 6.80, 2.08 to 22.19, respectively). We found generally consistent estimates with wider confidence interval crossing the null from other MR methods. The nonlinear MR analyses suggested a nonlinear relationship between metabolically unfavourable adiposity and osteoarthritis (Pnonlinear =0.003). CONCLUSION: Metabolic abnormalities did not explain the association between greater adiposity and the risk of osteoarthritis, which might suggest that the association is largely due to a mechanical effect on the joints.