P271 An analysis of short-term repeat MRI scans of vertebral corner lesions in suspected early axSpA: defining the prevalence and evolution of clinically significant spinal lesions without concurrent SIJ changes on imaging

Abstract

Abstract Background MRI offers an enhanced opportunity to detect early spinal changes of axial spondyloarthritis (axSpA), by identifying characteristic inflammatory and structural lesions, so called Romanus lesions. These include bone marrow oedema lesions on the vertebral corners and fatty replacement of these lesions, both highly suggestive features of axSpA. Current evidence suggests that treatment of these lesions requires early biologic therapy, hence early identification is imperative. We evaluate the prevalence and variation of vertebral corner lesions on short-term repeat MRI scans in patients with suspected early axSpA. Methods 109 MRI scans were performed at baseline and at 4, 8 and 12-weeks on 30 patients with suspected axial spondyloarthritis, who fulfilled the ASAS inflammatory back pain criteria, and had normal sacroiliac joints (SIJs) on antero-posterior pelvis radiographs. The protocol included sagittal T1 and short-tau inversion recovery of the cervico-thoracic spine and thoracolumbar spine. Results 29 patients completed the study (66% were male, 72% HLA-B2-positive). All patients had ≥1 clinical spondyloarthritis (SpA) feature and 86% had ≥2. 13 patients used NSAIDs regularly over the 12-week study period. Overall, 40 corner lesions were present in participants at baseline scanning. 67 new vertebral corner lesion changes occurred at different time points over the follow-up period compared to baseline. 43 changes were new or worsening lesions, while 24 changes were an improvement or resolution of a lesion. 48.6% (14/29) of patients had a minimum of 1 corner lesion present at baseline. 78.5% (11/14) of patients with baseline corner lesions experienced either a decrease/improvement or increase/progression of spinal corner lesions. 20.7% (6/29) of patients demonstrated transient corner lesions at baseline or follow-up with resolution by the 12-week scan (likely artefact). 5/29 patients met spinal imaging criteria suggestive of AS (3 at baseline, 1 only transiently at 1 month, and 1 which persisted from interval scanning). At 12-weeks, 13.8% of patients had at least 3 concomitant baseline or de-novo vertebral corner lesions present (minimum number needed for diagnostic significance). 75% of these patients did not have evidence of concomitant SIJ changes (10.3% of all patients). HLA-B27 status, gender, NSAID use, and number of SpA features were not associated with corner lesion development or improvement. Conclusion Approximately half of all patients who meet ASAS criteria for inflammatory back pain, but do not meet ASAS criteria for axSpA, demonstrated at least 1 vertebral corner lesion on MRI scan at baseline, which may represent artefact or a prelude to future disease progression. 13.8% of patients had at least 3 concomitant baseline or de-novo vertebral corner lesions present on MRI at the conclusion of 12-weeks of follow-up. In cases of suspected axSpA with negative SIJ MRI imaging, 10.3% of patients had significant spinal evidence of axSpA on MRI, highlighting the importance of spinal imaging and monitoring as part of the diagnostic work-up for axSpA. Disclosures S. Chatterjee None. H. Marzo-Ortega None. D. McGonagle None. A. Bennett None. R. Sengupta None.