Background MRI determined bone marrow oedema (BMO) lesions in the sacroiliac joints (SIJ) and spine have both diagnostic and prognostic value in axial spondyloarthritis (axSpA). It is unclear how quickly BMO may change on MRI and MRI scans are often normal in early axSpA. There is no clear evidence to guide the role of repeat MRI in the early diagnosis of axSpA.
Objectives To identify:1. If negative MRI scans in patients with clinically suspected axSpA may become positive within a 12-week period. 2. The natural change in BMO MRI lesions over 12-weeks in patients with IBP
Methods 29 patients, <40yrs, fulfilling ASAS Inflammatory Back Pain (IBP)criteria, with normal SIJ x-rays were recruited from 2 sites in the UK. Participants had MRI scans at baseline, 4, 8, and 12 weeks (total 109 scans). The MRI protocol included sagittal T1 and STIR cervico-thoracic and thoraco-lumbar spine and coronal oblique T1 and STIR sequences of the SIJs. NSAIDs were allowed if either continued at stable doses or discontinued throughout the study. Scans were read by consensus by 2 experienced readers unaware of clinical details or date sequence using the Leeds MRI Scoring system. In addition, each scan was scored as either positive or negative according to the ASAS definitions for the SIJ and spine.
Results 28 patients completed the study. Male n=18 (66%), HLA-B27+ve n=21 (75%), mean symptom duration 5 (±2.1) years. Regular NSAIDs were used by n=13 (46%) of patients. Mean baseline BASDAI 4.8 (±0.7) and mean baseline CRP 2.7 mg/dl (±1.7). 10 (36%) met the ASAS clinical criteria for axSpA prior to any imaging. 64% (n=18) of the patients were ASAS (spine and SIJ) negative at baseline and remained so during the entire study. 25% (7) were ASAS SIJ positive, 0.04% (1) were ASAS spine positive and 7% (2) were ASAS (spine and SIJ) positive. A total of 4 (14%) patients changed from ASAS axSpA imaging classification criteria negative to positive (spine (n=3) or SIJ (n=1)) during the study. All of these met ASAS clinical arm classification criteria at baseline.
At any time point, 15 patients had 212 lesions in the spine: 185 vertebral corners (98 inflammatory and 87 fatty), 15 inflammatory endplates (3 patients over all time points) and 12 inflammatory posterior elements (3 patients over all time points).
At baseline, there were 22 inflammatory and 18 fatty corner lesions in the spine. 12 inflammatory corner lesions resolved between any 2 time points. 26 new spinal inflammatory corner lesions appeared at any time point when the baseline MRI was normal.
At baseline, BMO was present in 38 (33%) SIJ quadrants. Between any 2 time points, 16 (14%) BMO quadrant lesions resolved and new BMO lesions appeared in 6 (5%) SIJ quadrants.
Conclusions These findings do not support the routine use of rescanning in new onset IBP cases suspected of having SpA. In a patient with severe IBP and high BASDAI with an initial negative MRI scan, a repeat scan has a 14% chance of changing the diagnosis which would thus make an otherwise ineligible patient suitable for potentially effective therapy
Disclosure of Interest None declared