Analysis Of Magnetic Resonance Imaging Scoring System For Facet Joint In Ankylosing Spondylitis

Objectives The aim of this study is to identify radiographic damages at affected facet joint of the spine and also to compare radiographic inflammatory activity between facet joints and spinal disco-vertebral units (DVUs) in ankylosing spondylitis (AS). Methods The whole spine magnetic resonance imaging (MRI) from total 46 AS patients (male =38, 82.6%) were reviewed. Three different methods including the Ankylosing Spondylitis spine MRI-activity (ASspiMRI-a), the Berlin modification of the ASspiMRI-a, and the Spondyloarthritis Research Consortium of Canada (SPARCC) scoring system for spinal inflammation were evaluated. Facet joint scoring system identified bone marrow edema, erosion, and ankylosis at the same level of each 23 DVU, which were scored between 0 to 3 points for both sides at each lesion. Total facet joint score means sum of scores of edema, erosion, and ankylosis at each level. Results Total 131 facet joint lesions, consisting of 98 for bone edema, 18 of bone erosion, and 15 of bony ankylosis, were identified. Most frequent facet joint lesions were detected at levels of C7-T1 (8.4%), T1-T2 (9.9%), L2-L3 (9.2%), and L3-L4 (9.9%) among total 131 facet joint lesions, whereas spinal inflammatory lesions were frequently identified at mid-thoracic spine level especially at T3 to T8. Total facet joint scores were closely correlated with ASspiMRI-a, Berlin, and SPARCC scores (r =0.372, p=0.011; r =0.414, p=0.004; r =0.400, p=0.006, respectively). However, the number of affected spinal joint at each DVU was not associated with that of facet joint lesions, indicating little possibility for concurrence of spinal DUVs and facet joint lesions. Conclusions This study implicates that inflammatory disease activity at facet joint was associated with that at DUVs of spine in AS. Discordance between DUVs and facet joint for concurrence of inflammation was observed. Close attention for facet joint involvement in AS should be needed. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.3418