Background Severity of knee joint damage is considered a barrier for effective exercise therapy in patients with knee osteoarthritis (OA), although no study evaluated associations between severity of OA-related knee joint tissue abnormalities and outcome of exercise therapy yet.

Objectives To evaluate associations between severity of OA-related knee joint tissue abnormalities, detected by MRI at baseline, and reductions in activity limitations, reductions in pain, and treatment response (i.e. outcome of exercise therapy) in knee OA patients treated with exercise therapy.

Methods In 99 patients with knee OA, participating a supervised 12-week exercise program, baseline MRI scans of the knee joint were obtained. BLOKS was used to assess severity of multiple MRI features (bone marrow lesions, cartilage loss, osteophytes, effusion, synovitis and meniscal abnormalities). Regression analyses were performed to analyze associations between severity of multiple MRI features (for tibiofemoral and patellafemoral compartment seperately) and reductions in activity limitations (WOMAC physical function, primary outcome), reductions in pain (NRS) and treatment response according to OMERACT-OARSI responder criteria (secondary outcomes).

Results Reductions of 24% and 33% in activity limitations (WOMAC) and pain (NRS), respectively, were found in 99 participants following exercise therapy, while two third (67%) could be considered responder according to OMERACT-OARSI responder criteria. Severity of patellafemoral cartilage loss at baseline was negatively associated (B (95% CI) = -4.7 (-2.5- 5.4), p=.01) with outcome of exercise therapy (i.e. less improvements in activity limitations in patients with moderate-severe patellafemoral cartilage loss). Severity in other MRI features were not found to be associated with outcome of exercise therapy.

Conclusions Severity of knee joint damage does not seem to a barrier for effective exercise therapy in patients with knee OA, except for moderate-to-severe cartilage loss in the patellafemoral compartment. This study suggest that also patients with highest grades of knee OA severity may benefit from exercise therapy. For patients with patellafemoral cartilage defects, exercises may need to be adapted or additional interventions may be necesarry for optimal treatment outcome.

Disclosure of Interest None Declared