Deep Learning-Based Comparison of Knee Minimum Joint Space Width in Patients with Rheumatoid Arthritis and Osteoarthritis Before Total Knee Arthroplasty.

Abstract

While osteoarthritis (OA) and rheumatoid arthritis (RA) can necessitate total knee arthroplasty (TKA), the mechanisms and radiographic patterns of joint space narrowing (JSN) differ. Deep learning (DL) enables compartment-specific measurement of minimum joint space width (mJSW). This study compared JSN patterns between patients with RA and OA undergoing TKA and evaluated associations between mJSW and RA disease activity. In this retrospective study, 409 RA patients undergoing TKA (2000-2021) were age- and sex-matched with OA patients. A validated DL model quantified medial and lateral mJSW from AP knee radiographs at two time points: early (>2 years before TKA) and late (≤2 years before TKA). The rate of JSN was calculated in 302 patients with paired radiographs. Mixed-effects models adjusted for BMI and alignment. RA disease activity, serology, and medications were recorded. RA demonstrated narrower lateral mJSW than OA at the late time point (5.6 vs. 7.0 mm, P<0.01), with comparable medial values. RA exhibited uniform bicompartmental JSN, whereas OA patients showed medial narrowing. RA had a faster mean lateral ΔmJSW (0.11 mm/year; P = 0.01), while OA had a faster mean medial ΔmJSW (0.21 mm/year; P<0.001). Longer RA duration and higher inflammatory markers were negatively associated with lateral mJSW. Seronegative RA showed faster lateral ΔmJSW in the surgical knee than seropositive patients (0.12 vs 0.06 mm/year; P = 0.03). Patients with RA demonstrated diffuse, symmetric cartilage loss contrasting with medialpredominant narrowing in patients with OA. Automated DL-based measurement enables scalable, compartment-specific assessment of compartment-specific patterns of joint degeneration.