Objectives: The objective of this study was to evaluate the responsiveness of magnetic resonance imaging (MRI) and ultrasonography (US) compared to conventional measures of disease activity and structural damage in RA patients during the first year of anti-TNFα treatment.
Methods: A cohort of RA patients (N=36, median age 53 years, disease duration 7.6 years and DAS28 5.7) was evaluated by core measures of disease activity, US (one wrist), MRI (one wrist) and conventional radiography (CR, both hands and wrists) at initiation of treatment with anti-TNFα agents and after 3, 6 and 12 months. Responsiveness was assessed by standardized response means (SRM). Accepted thresholds were applied to classify responsiveness as trivial, low, moderate or good.
Results: MRI synovitis (SRM between -0.79 and -0.92) and the MRI total inflammation score comprising synovitis, tenosynovitis and bone marrow edema (SRM between -1.05 and -1.24) were highly responsive. Moderate to high responsiveness was found for MRI tenosynovitis and bone marrow edema, all the composite indices (DAS28, SDAI and CDAI) and the 28-swollen joint count. US displayed low to moderate responsiveness. The MRI erosion score displayed low responsiveness, but was more responsive than CR measures at 3 and 6 months follow-up. MRI and CR measures of annual progression rates of damage performed similarly, and were highly responsive.
Conclusions: The most responsive measure of inflammation when evaluating anti-TNFα medication was a composite measure comprising MRI synovitis, tenosynovitis and bone marrow edema, and this may be a promising outcome measure in clinical studies.