A magnetic resonance imaging scoring system of joint-space narrowing in rheumatoid arthritis showed "a very high" agreement with computed tomography scores and may become a useful tool in rheumatoid arthritis clinical trials after further validation, judging from data presented by Dr. Uffe Møller Døhn.
In a small study, which was conducted to validate the OMERACT-RAMRIS MRI JSN scoring system in the wrists and metacarpophalangeal (MCP) joints, there was a very high agreement between the joint-space narrowing scores on MRI and CT and moderate agreement between scores on MRI and x-ray, said Dr. Møller Døhn of Copenhagen University Hospital at Glostrup at the annual European Congress of Rheumatology. In addition, there was "high to very high" inter- and intrareader reliability, particularly for the wrist joints.
Courtesy Uffe Møller Døhn
An MRI scoring system of joint-space narrowing in rheumatoid arthritis may become a useful tool in rheumatoid arthritis clinical trials after further validation.
An OMERACT (Outcome Measures in Rheumatology) initiative, this scoring system is being developed to provide a more precise and sensitive method of measuring joint space damage in patients with rheumatoid arthritis (RA), but it needs to be validated through comparisons to other imaging methods.
To evaluate the degree of agreement with CT and x-ray scores, this study assessed MRI and CT images of the wrist and the second to fifth metacarpophalangeal (MCP 2-5) joints of 14 people with RA and one healthy control, who were from a clinical trial. Three readers assessed the images twice, and a single reader scored x-rays using the Sharp-Van der Heidje method, said Dr. Møller Døhn, who is in the center for rheumatology and spine diseases at the hospital.
Courtesy Uffe Møller Døhn
The MRI scores "were very highly correlated" with CT scores.
The MRI scores of joint space narrowing "were very highly correlated" with CT scores, when comparing the wrist and MCP scores both separately and combined: Using intraclass correlation coefficients (ICCs) as a measure of agreement between scores and scorers, the MRI and CT scores for joint space narrowing were 0.94 for the MCP joints, 0.92 for the wrist, and 0.92 for the wrist and MCP joints combined. But the ICCs for the x-ray joint space narrowing scores were lower: With MRI scores, the ICCs were 0.49 for the MCP 2-5 joints and 0.55 for the wrist. With CT scores, the ICCs were 0.56 for the MCP 2-5 joints and 0.43 for the wrist.
"The most important next step is to test the scoring system in a longitudinal setting, in order to investigate the sensitivity to change," Dr. Møller Døhn said in an interview before the congress. "Before the system can be implemented as an outcome measure in clinical trials, we need to know if it is more sensitive than other methods that are already available. If it turns out that [joint space narrowing] assessment of several joints on x-ray is just as good as - or better than - MRI, then it does not add information to what we already use today."
Dr. Møller Døhn reported that he had no relevant financial disclosures.