This study was designed to identify the roentgenological changes of the knee that correlated with an unfavourable outcome when treated with an intra-articular knee viscoelastic supplementation. A prospective cohort of 60 patients receiving a standard course of intra-articular knee viscoelastic supplementation with a commercial uncross-linked hyaluronic acid derivative of an intermediate molecular weight were studied. Follow up was for 12 weeks post-treatment with clinical improvement measured using the Western Ontario and McMasters Universities Osteoarthritis Index. X-Rays of the relevant knee were viewed and graded for the severity of joint space, osteophyte, tibial spine, sclerosis, cyst formation, alignment and general severity by an observer blinded to the outcome of the treatment. There were no appreciable differences noted in the age, sex, length of follow up, pre-treatment, the severity of symptoms before treatment and number of intra-articular injections given per course in each X-ray category identified. There was a significant amount of improvement in patients with a minor loss of medial and lateral joint space in all outcome measures. Minimal changes in tibial spine and global appearance also indicated a positive outcome in stiffness, pain and overall improvement. Thus, patients with moderate to severe osteoarthritic changes in joint space on roentgenological examination would not significantly benefit from intra-articular knee viscoelastic supplementation. In addition, we feel that changes in the tibial spine and global appearance are not reviewed consistently enough to be included as part of our recommendation. As such, we conclude that only patients with a minimal to mild loss in joint space on roentgenological examination should form part of the target group who are likely to benefit from intra-articular knee viscoelastic supplementation.