Objectives: To study the prevalence of knee osteoarthritis in a group of former top-level football players and to assess the condition's impact on joint function and structure, compared with a control group., Materials and Methods: A cross-sectional, descriptive study was performed on a group of male former top-level football players (group G1), aged over 45 and with no history of knee trauma, arthritis, arthropathy or surgery. A second group of otherwise matched nonsporting subjects (group G2) was compared with the first group. For each subject, we specified age, weight, height, body mass index (BMI), dominant foot, the presence of knee axis deviation, the presence of pain and functional impairment. The pain level was assessed using a visual analogue scale (VAS). The functional assessment was performed using the Arabic version of the Lequesne index. Moreover, for each former player, we specified a number of sporting parameters, including those related to their playing career. The diagnosis of knee osteoarthritis was made using standard radiological and clinical criteria. The radiological severity of knee osteoarthritis was assessed using the Kellgren and Lawrence classification. We compared the two groups in terms of the frequency of knee osteoarthritis, the severity of pain and disability and the severity of structural impairment., Results: Our study included two groups of patients: a group of 50 former football players (G1) with a mean age of 49.2. Overweight was noted in 40 subjects. The mean number of training hours a week was 14+/-3.5 during their professional career and 2.5 during their retirement. Half of the sportsmen had taken part in more than 200 matches. Knee axis deviation was observed in 29 former players (i.e. 58% of the cases) and 27 displayed genuvarum. A group of 50 nonsporting volunteers (G2) was matched to the G1 group in terms of age, BMI and frequency of axis deviation. Knee osteoarthritis was more common in the football players than in the nonsporting subjects (80% versus 68%), although the difference was not statistically significant. Whereas pain was noted in only six footballers (with an average VAS score of 25.4+/-6.3mm) and was observed in 50% of controls (with a mean VAS score of 39.2mm+/-7.3) (P=0.001). Disability was recorded in six sportsmen, with a mean Lequesne score of 0.38+/-1.27. Disability was more frequent (23 subjects) and more intense (with a mean Lequesne score of 1.71+/-3.2) in the nonsportsmen (P=0.001). The Kellgreen and Lawrence radiological classification revealed that 57.5% of the sportsmen had scores of III or IV, compared with just 29.4% in the control group., Conclusion: Knee osteoarthritis is common in male football players. However, our study shows that the condition is less painful and less likely to cause functional disability (but paradoxically more destructive) than in nonsportsmen.