Background Clinical and radiological presentation of diffuse idiopathic skeletal hyperostosis (DISH) and axial radiographic spondyloarthritis (axSpA) may overlap in the elderly. To date, it is not known whether MRI may help to discriminate between those two diseases.
Objectives 1) To describe the MRI findings in DISH patients and 2) to assess the proportion of DISH patients whose MRI findings would fulfill the Assessment of SpondyloArthritis international Society (ASAS) criteria for a positive MRI of axSpA  and the factors associated with such fulfillment.
Methods This retrospective study involved all DISH patients (diagnosed with DISH according to the rheumatologist) who had a spine or sacroiliac joints (SIJ) MRI performed between January 2009 and December 2014. Demographics, and disease characteristics were collected by review of the medical files. Available X-rays and MRI were analyzed and blindly scored by an experimented reader (modified Stoke AS Spine Score (mSASSS) and modified New York (mNY) grades for spine and SIJ Xrays and the Spondyloarthritis Research Consortium of Canada (SPARCC) score for both spine and SIJ MRI).
Results Fifty-three DISH patients (mean age ± SD 65.6 ± 13.0 years, 71.7% male patients) were included in the analysis, with 31 SIJ MRI, 52 spine MRI and 37 spine or SIJ Xrays. On SIJ Xrays, 8 (19.5%) patients presented with sacroiliitis according to the mNY criteria. On spine Xrays, mean mSASSS score was 6.48 (± 8.86) and 40 (97.6%) patients had osteophytes with an angle of ≥45°, and up to 14 (34.15%) with an angle <45°. Mean SPARCC score of the spine was 18.3 ± 23.5. Thirty five (67.3%) patients had at least one fatty corner, and 41 (78.9%) patients had osteophytes ≥45° while only 2 (3.88%) with <45°. Thirty (57.7%) patients met the ASAS definition of a spine MRI suggestive of axSpA, but only 6 (19.4%) patients with an available SIJ MRI had a sacroiliitis according to ASAS criteria, and none of them had ≥3 erosions on the SIJ. Severe intervertebral disc lesions (Pfirmann score ≥4/5) were found in 38 (73.1%) patients. No factor was found to be significantly associated with fulfilment of ASAS criteria for a positive MRI of the Spine for axSpA.
Conclusions Inflammatory lesions of the spine are also present on the MRI of patients with DISH and more than a half of DISH patients presented with inflammatory lesions of the spine that would meet the ASAS criteria for a spine MRI suggestive of axSpA. However only few patients would meet the ASAS definition of a positive MRI for SIJ, suggesting that MRI of the SIJ but not of the spine might allow the differential diagnosis of DISH vs. axSpA in the eldery.
Hermann KG et al. Assessment in SpondyloArthritis international S. Descriptions of spinal MRI lesions and definition of a positive MRI of the spine in axial spondyloarthritis: a consensual approach by the ASAS/OMERACT MRI study group. Ann Rheum Dis. 2012;71(8):1278–88.
Disclosure of Interest None declared