The clinical trajectory of people with knee osteoarthritis: secondary analysis of a follow-up assessment following a two-year supplemental intervention

Abstract

Purpose: To report the clinical trajectory of physical function in people with knee osteoarthritis (OA) four years following a two-year daily intervention involving daily glucosamine and chondroitin sulfate supplements. Secondary objective is to determine if baseline patellofemoral OA features identified on magnetic resonance imaging (MRI) at baseline predict changes in physical function, pain and quality of life two and four years following a supplement intervention. Methods: This is a secondary analysis of a two-year randomised placebo-controlled trial, the Long-term Evaluation of Glucosamine Sulfate (LEGS) trial. Physical function was evaluated using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) at baseline, two (2011; post-intervention) and four years (2013; two years after formal treatment) with a mailed survey. The minimal clinically important difference of 9.1 was used to identify symptomatic improvement/worsening. Patellofemoral OA features at baseline were evaluated for bone marrow lesions (BMLs), full thickness cartilage loss and osteophytes using the MRI Osteoarthritis Knee Score. Generalised estimating equations determined if patellofemoral OA features were associated with symptomatic improvement/worsening. Results: Of 303 participants, 87% completed two year and 56% completed four year follow-up questionnaires. There were no differences in the number of people followed up between the supplemental groups. From baseline, 41% improved in physical function at two year follow-up, 47% had no change and 13% worsened (Fig 1). From post-intervention, 20% demonstrated further improvement at four years, 60% had no change and 20% worsened. Of people who improved at two-year follow up, 67% had no further changes at four year follow-up (Fig 1). Presence of BMLs in the patellofemoral joint predicted functional improvement at two-year follow-up (OR 0.44, 95% CI 0.24-0.82), and no other patellofemoral features on MRI were associated with clinical changes over time. Conclusions: Majority of people remained the same or improved in physical function, pain and quality of life scores over four years (range=20-60%), indicating that people with knee OA were not likely to functionally decline after an intervention. Most OA features in the patellofemoral joint did not predict symptomatic changes. However, the presence of patellofemoral BMLs predicted improvement in physical function at two-year follow up. Although this finding may be incidental, baseline physical function scores could have potentially influenced the results. However, since baseline physical function scores were adjusted for in the model and the model remained significant, a true effect of patellofemoral BMLs on improved physical function was demonstrated. Overall, the findings from this study demonstrate the importance of understanding the clinical trajectory of people with knee OA. Patients and clinicians need to be made aware of the high proportion of participants who do not have clinical worsening four years following initial start of treatment with supplements, as well as the moderate number of participants likely to demonstrate improvement in physical function either within two or four years.