Abstract
BACKGROUND: Studies have suggested that the symptoms of knee osteoarthritis (OA) are rather weakly associated with radiographic findings and vice versa. Our objectives were to identify estimates of the prevalence of radiographic knee OA in adults with knee pain and of knee pain in adults with radiographic knee OA, and determine if the definitions of x ray osteoarthritis and symptoms, and variation in demographic factors influence these estimates. METHODS: A systematic literature search identifying population studies which combined x rays, diagnosis, clinical signs and symptoms in knee OA. Estimates of the prevalence of radiographic OA in people with knee pain were determined and vice versa. In addition the effects of influencing factors were scrutinised. RESULTS: The proportion of those with knee pain found to have radiographic osteoarthritis ranged from 15-76%, and in those with radiographic knee OA the proportion with pain ranged from 15% - 81%. Considerable variation occurred with x ray view, pain definition, OA grading and demographic factors CONCLUSION: Knee pain is an imprecise marker of radiographic knee osteoarthritis but this depends on the extent of radiographic views used. Radiographic knee osteoarthritis is likewise an imprecise guide to the likelihood that knee pain or disability will be present. Both associations are affected by the definition of pain used and the nature of the study group. The results of knee x rays should not be used in isolation when assessing individual patients with knee pain.