Intra-articular corticosteroids remain widely used for symptomatic treatment of peripheral joint osteoarthritis (OA). Several studies in knee OA have indicated a significant benefit compared with placebo, although the effect appears to last for only 1 to 3 weeks. Two placebo-controlled studies have confirmed these findings. A further uncontrolled study has examined the effect of intra-articular corticosteroid at the hip. Attention is increasingly focused on a possible disease-modifying role for steroids in OA. Although a beneficial effect can be demonstrated in some animal models, caution should be exercised when extrapolating to human cartilage. Osteoarthritis is increasingly viewed as a phasic condition in which organ damage occurs intermittently. An ability to detect these phases of increased disease activity, perhaps with new imaging or biochemical techniques, could lead to a more rational approach to the use of intra-articular steroids in OA.