Sonography, as compared with clinical assessment, is a sensitive tool for evaluating synovitis in rheumatoid arthritis (RA). However, differences between these assessment tools may depend on how joint activity (i.e., an active joint) is defined. The present study was undertaken to compare clinically active joints with sonographically active joints in patients with RA, applying different sonographic definitions of an active joint.
Sonographic assessment of the finger and wrist joints (total of 11 joints) of each hand was performed in RA patients whose disease was in remission (Clinical Disease Activity Index ≤2.8; n = 60). Gray-scale (GS) and power Doppler (PD) ultrasound signals for synovitis were evaluated on a 4-point scale (grade 0 = none, grade 3 = severe). The sensitivity and specificity of swollen joint counts were investigated using, as reference, increasingly stringent sonographic definitions of an active joint. Sonographic findings were also assessed for correlations with other clinical variables, including the Health Assessment Questionnaire (HAQ) disability index (DI). Followup analyses were performed after 6–12 months.
GS ultrasound signals yielded positive findings for synovitis in 67.2% of the 1,320 joints assessed, and PD ultrasound signals indicated signs of synovitis in 20.4% of the joints assessed. Clinical identification of joint swelling was 100% specific for sonographic joint activity, independent of the stringency of the sonographic definition used; maximum sensitivity of the swollen joint counts was 25% for the most stringent definition (i.e., GS grade 3 and PD grade 3). Furthermore, patients with a higher-grade PD signal (grade 3) showed a higher HAQ DI score compared to patients with lower-grade PD signals (mean ± SD HAQ DI 0.45 ± 0.62 versus 0.20 ± 0.35). A higher grade of PD signal at baseline was found in joints that were assessed as clinically swollen at the consecutive followup visit.
Low-grade PD and GS ultrasound signals may not necessarily reflect the presence of active synovitis in RA joints. High-grade PD signals correlate well with the presence of clinical joint swelling and clinical disease activity, and a higher grade of PD signal is associated with higher degrees of functional impairment.