© 2015 Osteoarthritis Research Society International. Objective: To identify the independent relation of synovitis with incident radiographic knee osteoarthritis (OA) after adjusting for other structural factors known to cause synovitis. Design: We examined MRIs from knees that developed incident radiographic OA from the Multicenter Osteoarthritis Study (MOST) and compared these case knees with controls that did not develop OA. We examined baseline MRIs for knees developing OA at any time up to 84 months follow-up. We scored lesions in cartilage, meniscus, bone marrow and synovitis. Synovitis scores were summed (0-9) across three regions, suprapatellar, infrapatellar and intercondylar region, each of which was scored 0-3. After bivariate analyses examining each factor's association with incidence, we carried out multivariable regression analyses adjusting for age, sex, BMI, alignment and cartilage and meniscal damage. Results: We studied 239 case and 731 control knees. In bivariate analyses, cartilage lesions, meniscal damage, synovitis and bone marrow lesions were all risk factors for OA. After multivariable analyses, synovitis was associated with incident OA. A higher synovitis score increased the risk of incident OA (adjusted OR per unit increase 1.1; (95% CI 1.0, 1.2, P = .02)), but increased risk was associated only with synovitis scores of ≥3 (adjusted OR 1.6; 95% CI 1.2, 2.1, P = .003). Conclusions: Synovitis, especially when there is a substantial volume within the knee, is an independent cause of OA.