Increased risk of osteoarthritis has been found among athletes active in different kinds of sports. Knee injury is an established risk factor for knee osteoarthritis. In this population-based case–control study we investigated the risk of knee osteoarthritis with respect to sports activity and previous knee injuries. A total of 825 cases with x-ray-verified femorotibial osteoarthritis were identified at six hospitals in southern Sweden. The cases were matched (age, sex and residential area) with 825 controls from the general population. Mailed questionnaire data on sports activity for more than 1 year after the age of 16, knee injuries and confounding variables (weight, height, heredity, smoking and occupation) were collected and analyzed using logistic regression models. The response frequency was 89%. Among men knee osteoarthritis was related to soccer (odds ratio (OR) 1.6, 95% confidence interval (CI) 1.1–2.2), ice hockey (OR 1.9, 95% CI 1.2–3.0) and tennis (OR 2.0, 95% CI 1.1–3.8) but not to track and field sports, cross-country skiing, and orienteering. After adjustment for confounding variables soccer and ice hockey remained significantly related to knee osteoarthritis, but after adjustment for knee injuries no significant relation remained. The sports-related increased risk for knee osteoarthritis was explained by knee injuries.