Research article
MRI bone oedema scores are higher in the arthritis mutilans form of psoriatic arthritis and correlate with high radiographic scores for joint damage
-
* Corresponding author: Fiona M McQueen f.mcqueen@auckland.ac.nz
1 Department of Molecular Medicine and Pathology, University of Auckland, Park Road, Auckland 1010, New Zealand
2 Department of Rheumatology, Auckland District Health Board, Greenlane West, Auckland 1051, New Zealand
3 Department of Rheumatology, Copenhagen University Hospitals at Hvidovre and Gentofte, Kettegård alle 30, Hvidovre, DK-2650, Denmark
4 Department of Radiology, Auckland City Hospital, Grafton Rd, Auckland 1010, New Zealand
5 Department of Epidemiology and Biostatistics, University of Auckland, Morrin Road, Auckland 92019, New Zealand
6 Department of Medicine, University of Otago Wellington, Mein St, Wellington 6021, New Zealand
7 Department of Rheumatology, QE Health, Whakaue St, Rotorua 3010, New Zealand
Arthritis Research & Therapy 2009, 11:R2 doi:10.1186/ar2586
Published: 6 January 2009Abstract
Introduction
The aim of this study was to investigate the magnetic resonance imaging (MRI) features of bone disease in the arthritis mutilans (AM) form of psoriatic arthritis (PsA).
Methods
Twenty-eight patients with erosive PsA were enrolled (median disease duration of 14 years). Using x-rays of both hands and feet, 11 patients were classified as AM and 17 as non-AM (erosive psoriatic arthritis without bone lysis)by two observers. MRI scans (1.5T) of the dominant hand (wrist and fingers scanned separately) were obtained using standard contrast-enhanced T1-weighted and fat-saturated T2-weighted sequences. Scans were scored separately by two readers for bone erosion, oedema and proliferation using a PsA MRI scoring system. X-rays were scored for erosions and joint space narrowing.
Results
On MRI, 1013 bones were scored by both readers. Reliability for scoring erosions and bone oedema was high (intraclass correlation coefficients = 0.80 and 0.77 respectively) but only fair for bone proliferation (intraclass correlation coefficient = 0.42). MRI erosion scores were higher in AM patients (53.0 versus 15.0, p = 0.004) as were bone oedema and proliferation scores (14.7 versus 10.0, p = 0.056 and 3.6 versus 0.7, p = 0.003 respectively). MRI bone oedema scores correlated with MRI erosion scores and X-ray erosion and joint space narrowing scores (r = 0.65, p = 0.0002 for all) but not the disease activity score 28-C reactive protein (DAS28CRP) or pain scores.
Conclusions
In this patient group with PsA, MRI bone oedema, erosion and proliferation were all more severe in the AM-form. Bone oedema scores did not correlate with disease activity measures but were closely associated with X-ray joint damage scores. These results suggest that MRI bone oedema may be a pre-erosive feature and that bone damage may not be coupled with joint inflammation in PsA.