Objective: To evaluate the intraobserver and interobserver reproducibility of B-mode and power Doppler (PD) sonography in patients with active long-standing rheumatoid arthritis (RA) comparatively with clinical data.
Methods: In each of 7 patients being considered for a change in their RA treatment regimen, 7 healthcare professionals examined the 28 joints used in the Disease Activity Score 28-joint count (DAS28). Then 7 sonographers examined each of the 7 patients twice, using previously published B-mode and PD grading systems. The clinical reference standard was presence of synovitis according to at least 4/7 examiners. The sonographic reference standard was at least grade 1 (ALG1) or 2 (ALG2) synovitis according to at least 4/7 sonographers. Interobserver reproducibility of sonography was assessed versus the sonographer having the best intraobserver reproducibility. Agreement was measured by Cohen's kappa statistic.
Results: Intraobserver and interobserver reproducibility of B-mode and PD used separately was fair to good. Agreement between clinicians and sonographers at all sites using B-mode, PD, and both was 0.46, 0.37, and 0.36, respectively, for grade 1 synovitis; and 0.58, 0.19, and 0.19 for grade 2 synovitis. The number of joints with synovitis was smaller by physical examination (36.7%) than by B-mode with ALG1 (58.6%; p < 0.001). The number of joints with synovitis was higher by physical examination than by PD with both ALG1 (17.8%; p < 0.0001) and ALG2 (6.6%; p < 0.0001).
Conclusion: PD findings explain most of the difference between clinical and sonographic joint assessments for synovitis in patients with long-standing RA.