Objective: To determine if quantitative hand images obtained from an office-based MRI extremity scanner reliably distinguish patients with rheumatoid arthritis from controls.
Methods: The hands of 39 patients suffering from rheumatoid arthritis were imaged using a small bore, 1.0 Tesla Magnetic Resonance Imager. Non-contrast images of the metacarpophalangeal joints and wrist joints were evaluated using a method based on the validated rheumatoid arthritis magnetic resonance imaging system (RAMRIS). The extent and degree of synovitis, bone edema and bone erosions was assessed. Derived scores were compared with the corresponding scores for groups of younger (n=14) and older (n=27) controls with no signs or symptoms of joint disease.
Results: The mean (+/-standard error) total joint scores were 0.3+/-0.2 for young controls, 11.5+/-2.4 for older controls and 34.1+/-6.0 for the patients with rheumatoid arthritis. The greatest difference between rheumatoid patients and older controls was observed for synovitis with scores that were greater by a factor of almost 6.5. Scores for erosions and edema were factors of 2.9 and 2.3 greater in rheumatoid arthritis than in controls. The relationship between scores for the same joints on the dominant and non-dominant sides was generally stronger than the relationship between the metacarpophalangeal and wrist joints of the same hand.
Conclusion: These observations indicate that scoring of hand images obtained from a small bore, office based, 1.0 Tesla MR imager have clinical validity and may be used to distinguish patients with rheumatoid arthritis from aged matched controls.