![]() | Only 14 pages are availabe for public view |
Abstract Objective: To compare the clinical assessment of overall inflammatory activity in patients with rheumatoid arthritis (RA) with grey scale and power Doppler (PD) ultrasonography (US). Methods: Ninety four consecutive patients with RA were included. Demographic and clinical data, C reactive protein (CRP) level, and erythrocyte sedimentation rate (ESR) were recorded for each patient. The presence of tenderness, swelling, and a subjective swelling score from 1 to 3 were independently assessed by two rheumatologists, who reached a consensus in 60 joints examined in each patient. All patients underwent a US examination by a third blinded rheumatologist, using PD. US joint effusion, synovitis, and PD signal were graded from 1 to 3 in the 60 joints. Joint count and joint index for effusion, synovitis, and PD signal were recorded. A 28 joint count for clinical and US variables was calculated. Interobserver reliability of the US examination was evaluated by a fourth blinded rheumatologist. Results: US showed significantly more joints with effusion (mean 15.2) and synovitis (mean 14.6) than clinical examination (mean 11.5, p<0.05). A significant correlation was found between joint count and joint index for swelling, US effusion, synovitis, and PD signal. The 28 joint count for effusion, synovitis, and PD signal correlated highly with the corresponding 60 joint counts. US findings correlated better with CRP and ESR than clinical measures. Interobserver reliability was better for US findings than for clinical assessment. Conclusion: US is a sensitive method for assessing joint inflammatory activity in RA, complementary to clinical evaluation. Twenty-five patients with known or suspected tears of the rotator cuff in 26 shoulders underwent MR imaging. All patients also underwent arthrography or surgery. MR visualized abnormalities consistent with a tear in 20 of the 22 tears diagnosed by arthrography or surgery. In most cases, tears were seen as regions of increased signal intensity within the cuff on long-TR pulse sequences, although two cases simply showed an almost complete absence of normal cuff. The MR appearance of the two cases with partial tears was similar to that of full-thickness tears. Of the four cases with normal arthrograms, one case had MR findings consistent with a tear. We conclude that MR has good potential for the noninvasive diagnosis of rotator cuff tears. Its ultimate role in this diagnosis must await prospective studies comparing its accuracy with that of sonography, CT, and arthrography. |