Identification of distinct vascularization pattern in finger joint synovits in rheumatoid arthritis versus other forms of arthritis by contrast enhanced sonography: a preliminary study.

Background: Rheumatoid arthritis (RA) has among all forms of arthritides the worst outcome. Synovitis in RA is consistent with inflammation, synovial hyperplasia and neovascularization, that correlates with disease activity, aggressiveness and joint destruction. Contrast enhanced ultrasonography (CEUS) has been proven to be a very sensitive method in assessing synovitis in RA, equipollent to magnetic resonance imaging.1,2 The vascularization detected by CEUS in RA synovitis has not yet been compared to other forms of arthritis. Objectives: To investigate by CEUS the pattern of synovial vascularization in RA and other arthritides and to find parameters able to discriminate between both. Methods: 40 outclinic patients with arthritis of finger joints were recruited. 23 patients were affected from RA according to ACR criteria. 6 patients had severe, 11 moderate and 2 low disease activity using Das28 score. Of the 17 non RA patients 11 suffered from psoriatic arthritis, 1 from spondiloarthritis, 1 from osteoarthritis, 2 from connectivitis, 1 from septic arthritis, and 1 from arthritis in celiac disease. 5 patients presented severe, 11 moderate and 1 low disease activity. The most active joint was chosen for CEUS investigation. The hands were water-immersed and steady probe was used to increase resolution of superficial interfaces and to avoid artefacts by movement and inhomogenous gel application. Endovenous bolus injection of Sonovue was performed. For image acquisition contrast tune imaging with low mechanical index was used. Software able to recognise different articular structures and to assess pixel properties of contrast enhancement was devised by our bioengineers. Results: The most active joint of the hands in 23 RA (21 MCF and 2 IFP) and 17 patients affected by other forms of arthritis (12 MCF and 5 IFP) was analysed after contrast administration. The software identified various contrast flow parameters. 8 parameters resulted helpful in discriminating the RA from the non RA group: synovial wash-in and wash-out velocity, synovial peak intensity, fraction of synovial, capsular and peripheral-entheseal pixels, fraction of synovial and peripheral high intensity pixels. The importance for diagnosis of the single parameters was calculated by linear discriminator analysis and summed to an vascularization pattern identifying RA patients. By using this pattern 22 out of 23 RA patients resulted true positive and 1 as false negative compared to clinical diagnosis, whereas true negative results were seen in 16 out of 17 non RA patients and 1 false positive in a patient showing the rheumatoid variant of psoriatic arthritis. The sensibility and specificity was 91% and 94%. The positive predictive value was 95% and the negative predictive value 88%. Conclusion: We found by CEUS a pattern of vascularization in synovitis of rheumatoid arthritis different from that found in other forms of arthritis. In patients with arthritis CEUS represents available application to detect rheumatoid arthritis with worse prognosis and to lead therapeutic decision.