The MCP2 and the wrist plus two tendons are the most affected and responsive joints/tendons out of the US7 score in patients with rheumatoid arthritis – an observational study

Background: There is no international consensus on an optimal ultrasound score for monitoring of rheumatoid arthritis (RA) on patient-level yet. Our aim was to reassess the US7 score for the identification of the most frequently pathologic and responsive joint/tendon regions to optimize it and contribute to an international consensus. Furthermore, we aimed to evaluate the impact of disease duration on the performance of the score. Methods: RA patients were assessed at baseline and after 3 and 6 months of starting/changing DMARD therapy by the US7 score in greyscale (GS) and power Doppler (PD). The frequency of pathologic joint/tendon regions and their responsiveness to therapy were analyzed by Friedman test including the comparison of palmar vs. dorsal regions (Chi-square test). The responsiveness of different reduced scores and the amount of information retained from the original US7 score were assessed by standardized response mean/linear regression. Analyses were also performed separately for early and established RA. Results: A total of 435 patients (N=138 early RA) were included (56.5 (SD 13.1) years old, 8.2 (9.1) years disease duration, 80% female). The dorsal wrist, palmar MCP2, extensor digitorum communis (EDC) and carpi ulnaris (ECU) tendons were most frequently affected by GS/PD synovitis/tenosynovitis (wrist: 45%/43%; MCP2: 35%/28%; EDC: 30%/11% and ECU: 25%/11%) and significantly changed within 6 months of therapy (all p≤0.003 by GS/PD). The dorsal vs. palmar side of the wrist by GS/PD (p<0.001) and the palmar of the finger joints by PD (p<0.001) were more frequently pathologic. The reduced US7 score (GS/PD: dorsal MCP2, dorsal wrist, EDC and ECU, only GS: palmar MCP2) showed therapy response (SRM 0.433) after 6 months and retained 76% of the full US7 score’s information. No major differences between the groups of early and established RA could be detected. Conclusions: The wrist, MCP2, EDC and ECU tendons were most frequently pathologic and responsive to therapy in both early and established RA and should therefore be included in a comprehensive score for monitoring RA patients on patient-level.