Andersson lesions of whole spine magnetic resonance imaging compared with plain radiography in ankylosing spondylitis

The objective of this study was to identify the characteristics of Andersson lesions using whole spine magnetic resonance imaging (MRI) compared with plain radiography in ankylosing spondylitis (AS). A total of 62 patients with AS who had undergone whole spine MRI and plain radiography were retrospectively enrolled in this study. We compared the number of discovertebral units (DVUs) with Andersson lesions with clinical and radiographic indices such as erythrocyte sediment rate (ESR), C-reactive protein (CRP), the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Bath Ankylosing Spondylitis Functional Index (BASFI), and modified Stoke Ankylosing Spondylitis Spine Score (mSASSS). Fifty-three patients (85.5 %) by whole spine MRI and 23 patients (37.1 %) by plain radiography had at least one Andersson lesion. We found 129 DVUs with Andersson lesions (11.1 %) by MRI and 35 DVUs by plain radiography over all the spine levels. Andersson lesions by MRI were most commonly detected at the lower thoracic spine (from T7-8 to T12-L1). Among the 151 total Andersson lesions by whole spine MRI, 41 were identified as central disc type, 26 as anterior peripheral disc type, 44 as posterior peripheral disc type, and 40 as diffuse disc type. However, the number of Andersson lesions did not correlate with ESR, CRP, BASDAI, BASFI, or mSASSS (p > 0.05 for all). Our study indicates that the presence of Andersson lesions in patients with AS is clearly underestimated. MRI is a superior technique for detecting early Andersson lesions compared with plain radiography.