Background/Purpose: Musculoskeletal ultrasonography (US) is recognized as a useful tool for the diagnosis and monitoring of rheumatoid arthritis (RA), though a standard procedure has not been established. Scanning all joints is ideal but not practical for routine examinations. Therefore, it is important to determine which joints should be examined. Here we investigated the optimal number and combination of joints to be assessed by power Doppler (PD) US in daily practice for RA.

 Methods: PDUS were performed in 24 joints, including all PIP, MCP, bilateral wrist and knee joints in 234 patients with RA. PD signals were scored semiquantitatively from 0 to 3 in each joint, and total PD score-24 was calculated by summing them up as comprehensive assessment. We examined correlations of total PD score-24 with individual joint PD scores and the sum PD scores of 8 sets of arbitrarily selected joint combinations. The sensitivity and negative predictive value (NPV) of the individual sets of selected joint examination for the detection of active synovitis were also evaluated.

 Results: Positive PD signals were more frequently found in bilateral wrist, knee, MCP 2 and 3 joints than the other joints. The individual PD scores of these 8 joints also showed higher correlation coefficients with total PD score-24 (r_s ≥ 0.4). There was no significant laterality in PD scores except for PIP 4 joints (P = 0.037). In wrist, the PD score was significantly lower in each region (radial, medial or ulnar region) as compared to that from the assessment of whole joint (P = 7.6 x 10^-7, 9.8 x 10^-3 and 1.2 x 10^-4, respectively). Among the sum PD scores of the 8 sets of selected joint combinations, that of the combination of 8 joints (total PD score-8), including bilateral MCP 2, 3, wrist, and knee joints, showed the highest sensitivity and NPV (Table 1; 98.1% and 96.2%, respectively). Discrepancy between the comprehensive 24 joint assessment and the selected 8 joint assessment was found in only 3 of 234 patients including 76 patients in US remission. Moreover, total PD score-8 showed very high correlation with the total PD score-24 (Table 1 and Figure 1; r_s = 0.97, P < 0.01). Comparison with other sets of joint combinations revealed that the wrist and knee joints had the stronger impacts on the correlation and the other parameters.

Conclusion: This study suggests that US examination in the selected 8 joints, including the bilateral wrist, knee, MCP 2 and 3 joints, is simple and efficient enough for making the diagnosis, monitoring disease activity, and judging imaging remission of RA in daily practice.