Article Text

AB0964 Synovial Inflammation Analysed by 3 Tesla Magnetic Resonance Imaging in Etanercept-Treated Patients with Rheumatoid Arthritis Indicates Persistent Disease Activity despite of Clinical Remission
  1. M. Oleszowsky1,
  2. M. Marinova2,
  3. W. Willinek3,
  4. M. Seidel1
  1. 1Department of Onkology, Hematology of and Rheumatology, University Hospital of Bonn, Medizinische Klinik III
  2. 2Department of Radiology, University Hospital of Bonn, Bonn
  3. 3Department of Radiology, Neuroradiology, Sonography and Nuclear Medicine, Krankenhaus der Barmherzigen Brüder Trier, Trier, Germany


Background Treatment efficacy of conventional disease-modifying anti-rheumatic drugs (DMARDs) is insufficient in some patients with rheumatoid arthritis (RA). These patients require TNF-a antagonists that demonstrate excellent remission rates confirmed by clinical scores and conventional radiographs. Magnetic resonance imaging (MRI) is a more detailed diagnostic tool and furthermore, the 3 Tesla technique provides a higher spatial resolution compared to 1.5 Tesla. Especially early and thus prognostically relevant subchondral bone lesions might be detected more precisely using 3 Tesla MRI.1

Objectives The aim of this study was to evaluate a reduction of joint inflammation in RA patients treated with DMARDs as compared to etanercept (ETA) over a period of 12 months.

Methods This epidemiologic clinical pilot MRI study examined 10 ACR/EULAR-positive RA patients with inadequate response to two DMARDs with an indication for ETA using the RA MRI system (RAMRIS). The control group consisted of 10 patients adequately responding to DMARDs. 3 Tesla MRI (PDSPAIR; T1 with and without contrast agent) using a multi-channel hand coil was performed at baseline, four (only ETA) and 12 months following initiation of ETA. RAMRIS scoring included analysis of synovitis, effusion, subchondral erosion, and also bone marrow edema. Additional analysis consisted of the visual analogue scale (VAS) and the Disease Activity Score 28-joint count (DAS28) at baseline and after 12 months in DMARD patients and at baseline, four and 12 months in ETA-patients.

Results In DMARD-patients, the VAS averaged 21.0±11.3 and 20.2±24.6 (p=0.9). In ETA-patients, the VAS decreased from 46.3±7.9 (baseline) to 23.9±7.1 (p=0.04) and 24.0±6.3 (p=0.04). In DMARD-patients, the DAS28 remained at 2.1±0.6 and 2.9±1.0 (p=0.06). In ETA-patients, however, it decreased from 3.8±0.4 (baseline) to 2.8±0.3 (p=0.08) and then significantly to 2.5±0.3 (p=0.01). In DMARD-patients, the RAMRIS did not change between baseline and 12 months (11.0±2.3 vs. 11.8±2.8, respectively, p=0.8). Similarly, in ETA-patients, the RAMRIS did not change significantly (28.9±5.0 vs. 25.8±4.7, p=0.7 and 24.6±4.5, p=0.5). In contrast, the RAMRIS was significantly higher at baseline in ETA-patients as compared to DMARD-patients (28.9±5.0 vs. 11.0±2.3 p=0.005) and after 12 months (24.6±4.5 vs. 11.8±2.8, p=0.03). Separate analysis of synovitis, erosion and bone marrow edema did not change after 12 months.

Conclusions The data show a significant higher RAMRIS and thus more actively diseased patients prior to ETA treatment as compared to DMARD-patients at baseline and after 12 months despite of adequate DAS28 remission (<2.6). Persistence of 3 Tesla MRI synovial inflammation and DAS28 remission are thus separate entities. MRI scoring before starting a treatment may indicate the requirement for anti-TNF-a agents.

  1. Wieners G, et al. Eur Radiol. 2007; 17(8):2176–82

Acknowledgement In part supported by Pfizer.

Disclosure of Interest M. Oleszowsky Grant/research support from: In part supported by Pfizer., M. Marinova: None declared, W. Willinek: None declared, M. Seidel: None declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.