Background/Purpose: While knee osteoarthritis (KOA) leads to cumulative damage to a diarthrodial joint there are also more dynamic processes that fluctuate throughout the disease process (e.g., bone marrow lesions [BMLs], effusion-synovitis). To monitor the different constructs of structural progression, we recently validated 2 magnetic resonance (MR)-based composite scores. First, the cumulative cartilage damage score represents hyaline cartilage damage throughout the knee, relates to radiographic severity, and reflects the damage attributable to KOA over the course of the disease. Secondly, the disease activity score is a composite of BMLs and effusion-synovitis volumes that relates to knee pain and reflects a patient’s current state of disease and symptoms. It is unknown if these composite scores predict poor clinical outcomes (e.g., knee replacement). Hence, we assessed adults without radiographic KOA to determine if baseline and 1-year change in cumulative cartilage damage or disease activity predicted knee replacement over the subsequent 9 years.

Methods: We performed a secondary analysis using existing MR-based data from a sex-matched nested case-control study of 3 groups from the Osteoarthritis Initiative without radiographic KOA at baseline (Kellgren-Lawrence (KL)< 2): 1) accelerated KOA: developed KL 3 or 4 within 48 months; 2) typical KOA: increase in KL grade within 48 months; 3) no KOA: no change in KL grade within 48 months.  We quantified tibiofemoral cartilage damage, BML volume, and effusion-synovitis volume with semi-automated programs. All MR-based measures were normalized to bone width and standardized, so all measurements were on the same scale. The cumulative cartilage damage score was the sum of standardized cartilage damage for the medial and lateral tibia and femur. The disease activity score was the sum of the standardized volumes of effusion-synovitis (single volumetric measure) and BML (4 locations: medial and lateral tibia and femur). The outcome was knee replacement (partial or total) that was reported or observed between the 1- and 9-year follow-up ( >96% adjudicated). For the primary analyses we combined the 3 groups and used logistic regression models to assess if baseline or 1-year change in cumulative cartilage damage or disease activity predicted knee replacement. We performed a sensitivity analysis limited to those who developed accelerated KOA.

Results: The groups were mostly female (63%) and overweight, 33% reported frequent knee pain within a year of baseline, 23% developed radiographic KOA during the first year (KL >1), and 19 people received a knee replacement between the 1- and 9-year follow-up. Greater baseline and 1-year change in disease activity, but not cumulative cartilage damage, was statistically associated with greater chance of receiving a knee replacement (Table). These findings were consistent among adults who developed accelerated KOA.

Conclusion: Prior to the onset of radiographic KOA, disease activity was associated with a knee replacement over the subsequent 9 years. This supports the construct validity of disease activity, which relates to knees symptoms that are the main reason a person would receive a knee replacement.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-novel-composite-score-reflecting-disease-activity-predicts-future-knee-replacements-data-from-the-osteoarthritis-initiative/