Objectives To identify the MRI parameter that best separates healthy controls from early RA patients. Patterns of enhancement from conventional MRI and dynamic contrast enhanced (DCE)-MRI of healthy controls and of early RA patients with active disease undergoing treatment were compared.

Methods Conventional and DCE-MRI of the wrist and the 2nd-5th metacarpophalangeal (MCP) joints was performed once in 26 healthy controls, age 47.5 (24-64) years (median (range)), and before and after 6 and 12 months of DMARD treatment in 14 early RA patients (age 39 (26-63) years; disease duration <6 months), using a 1.0T Siemens Impact whole body MRI unit (Siemens, Erlangen, Germany). DCE-MRI was obtained in 1 preselected coronal slice covering the wrist and the 2nd-5th MCP-joints. The dynamic sequences were analyzed using Dynamika v. 4.2 (Image Analysis Ltd., Leeds, UK): Regions of interest were drawn delineating the individual joints. The number of enhancing voxels (Nvoxel), the average initial rate of enhancement (IRE, %/s), the average maximal enhancement (ME), MExNvoxel and IRExNvoxel were calculated for the combined examined areas (wrist and 2nd-5th MCP-joint). The conventional contrast-enhanced MR-images were evaluated by the OMERACT RA MRI (RAMRIS) synovitis score.

Results 16 of 26 (61.5%) healthy controls and 13 of 14 (92.9%) RA patients at baseline showed enhancement (Number of enhancing voxels (Nvoxel) median 3 and 362, respectively). Healthy controls values were lower than baseline RA values for all parameters (Mann-Whitney, all p<0.001 (except for ME p=0.003)), Table 1. At baseline only 1 patient had a baseline RAMRIS below the 95th percentile of the controls, the numbers were 3 for Nvoxel, MExNvoxel and IRExNvoxel and 9 and 10 for IRE and ME, respectively. RAMRIS, Nvoxel, IRExNvoxel (trend) but not IRE, ME, and MExNvoxel decreased from baseline to 6 months after start of treatment (Wilcoxon signed rank test, p=0.003, 0.038, 0.050, 0.26, 0.374, and 0.051).

Conclusions RAMRIS synovitis scoring of conventional contrast-enhanced MRI and, to a lesser extent 1 slice DCE-MRI analysis of number of enhancing voxels, best discriminated early RA patients from healthy controls. Both parameters decreased during treatment, ie. showed sensitivity to change.

Disclosure of Interest M. B. Axelsen Grant/Research support from: Abbott, B. Ejbjerg: None Declared, M. Hetland: None Declared, K. Hørslev-Petersen Grant/Research support from: Abbott, U. Lauridsen: None Declared, O. Majgaard: None Declared, M. Boesen Consultant for: Image Analysis, O. Kubassova Shareholder of: and director of Image Analysis, H. Bliddal Grant/Research support from: Abbott, M. Østergaard Grant/Research support from: Abbott