Does spontaneous cartilage thickening occur in osteoarthritic knees?: Data from IMI-APPROACH and the OAI

Abstract

Objective: Determine whether spontaneous cartilage thickening (i.e. without specific intervention) occurs in patients with or at high risk of knee osteoarthritis (OA) and compare characteristics with knees exhibiting cartilage thinning. Design: 1447 knees from IMI-APPROACH and OAI with radiographic and MRI data at baseline, one year, and two years were included. Cartilage thickening was defined as an increase in both radiographic average whole-joint minimum joint space width (mJSW) and MRI mean cartilage thickness over the total femorotibial subchondral bone area (ThCtAB) according to per-patient fitted linear regression lines over the two-year period. Knees with decreases in both were categorized as thinning. Characteristics in structural and clinical parameters between both groups were compared. Results: Cartilage thickening was observed in 203 knees (14%). Patients with thickening were younger (median 59 vs 64 years, p<0.001), more often female (69% vs 59%, p=0.011), and had lower KL grades (32% KL grade 0 or 1 vs 21%, p<0.001). Over two years, thickening knees showed an increase in mJSW (+0.11mm/year) and ThCtAB (+0.03mm/year) while thinning knees showed a decrease (-0.15mm/year and −0.05 mm/year, respectively). No significant differences were found in BMI or weight change. Fewer hyaluronic acid injections were observed during follow-up in the thickening group (p=0.045). Sensitivity analyses, using alternative definitions of thickening based on JSW and MRI thresholds, confirmed these findings. Conclusions: Spontaneous cartilage thickening occurs in a relevant proportion of knees with or at risk of knee OA. This intrinsic thickening warrants further investigation into the mechanisms and clinical relevance.