Background Recently ultrasonography (US) has been prevalent as a valid and objective modality of joint examination in patients with rheumatoid arthritis (RA). However, few studies have reported the agreement between subjective joint symptoms and US synovitis at each joint level.
Objectives The aim of this study was to examine and compare the concordance between joint symptom/tenderness/swelling and US synovitis.
Methods Fifty patients with RA (84% female, the median age, disease duration and disease activity score of 28 joints (DAS28) were 69 years, 2.4 years, and 3.8, respectively) were asked for the self-evaluation of joint symptoms including pain and considerable stiffness in (proximal) interphalangeal (IP/PIP), metacarpophalangeal (MCP), wrist, elbow, shoulder, knee and ankle joints. Those joints were evaluated by physical examination for the presence of tenderness and/or swelling, and also by US for the presence of synovitis defined as gray-scale score ≥2 or power Doppler signal score ≥1.
Results In a total of 1494 evaluated joints, joint symptoms, tenderness and swelling were observed in 288 joints (19.3%), 182 joints (12.2%), and 220 joints (14.7%), respectively, and US synovitis was observed in 317 joints (21.2%).The overall concordance rate with US synovitis was the least for joint tenderness (κ=0.30) when compared with joint symptoms (κ=0.39) or swelling (κ=0.43), irrespective of the evaluated joint except for the elbow (Table 1). Moreover, the percentage of inflamed joint clinically detected only by joint symptom, tenderness, or swelling was 4.0%, 0.5% and 1.8%, respectively, when US synovitis was defined as a gold standard.
Conclusions Joint examination for the presence of swelling is crucial, and that for the presence of tenderness can be excluded from clinical joint assessment, or it may be better substituted by patient-reported joint symptoms. A recent proposal of the modification of composite activity measures such as simplified disease activity index (SDAI) by removing tender joint count (and patient's global assessment) in order to improve the correlation with magnetic resonance imaging (MRI) synovitis may, at least in part, support our results.
Backer J, et al. Arthritis Rheum 2014;66:794-802
Disclosure of Interest None declared