Evaluating the Role of MRI in Early Rheumatoid Arthritis Diagnosis

Title:

Evaluating the Role of MRI in Early Rheumatoid Arthritis Diagnosis

Author(s):

L. G. Suter, R. S. Braithwaite, L. Fraenkel. Yale University, New Haven, CT

Purpose: The use of magnetic resonance imaging (MRI) in rheumatoid arthritis (RA) management is increasing rapidly. Our objective was to evaluate the evidence supporting the use of MRI for diagnosis of early RA.

Methods: We used Cochrane Collaboration systematic review guidelines and employed a broad search strategy using medical subject headings (MeSH) “exp Magnetic Resonance Imaging/” and “exp Arthritis/”. We searched Ovid, Cochrane and CINAHL databases. We manually searched bibliographies of included studies and review articles, abstracts of relevant journals, and meeting proceedings of professional societies. We included studies of any duration that 1) prospectively examined the ability of MRI findings to predict an RA diagnosis among patients with symmetric inflammatory hand or wrist arthritis, 2) used ACR 1987 revised criteria and/or clinical assessment by a rheumatologist as the diagnostic gold standard, and 3) reported adequate information to calculate sensitivity and specificity.

Results: The search strategy is shown in Figure 1. Seven studies, comprising a total of 262 patients with a mean follow-up of 1.5 years, met inclusion criteria. Two studies compared periarticular enhancement on MRI to the ACR 1987 criteria; one evaluated OMERACT RAMRIS scores; one evaluated a prediction score of clinical synovitis, MRI edema and/or erosions, and serologic tests; two compared MRI and bone scintigraphy; and one compared MRI synovitis with edema or erosions to anti-CCP antibodies. Sensitivity and specificity of early MRI findings for eventual RA diagnosis ranged from 64-100% and 50-87%, depending upon the clinical and radiographic features included. The one abstract to directly compare MRI to anti-CCP antibodies found that MRI was more sensitive (100% vs. 23%), but less specific (78% vs. 100%) than anti-CCP antibodies. The only study to report validated OMERACT RAMRIS scoring noted that an OMERACT erosion score of 15 or greater had a sensitivity of 64% and specificity of 70%.

Conclusions: Despite rapidly rising MRI utilization in early RA, data evaluating MRI for the diagnosis of early RA are limited by inconsistent MRI scoring and application of radiographic and clinical features. Data also suggest significant false positive rates for MRI, although the highest false positive rates are in populations with other rheumatic diseases for whom there are fewer anticipated adverse consequences to misdiagnosis. Use of the validated OMERACT RAMRIS scoring system and a uniform approach to combining radiographic and clinical information would significantly improve our understanding of the diagnostic role of MRI in undifferentiated inflammatory arthritis.

Disclosures:

L.G. Suter, None; R.S. Braithwaite, None; L. Fraenkel, None.