Background The ASAS definition of a positive MRI is solely based on inflammation in the sacroiliac joints (SIJ), although spinal inflammatory lesions on MRI suggestive of axial Spondyloarthritis (axSpA) may also occur. It is not well known how often inflammation in the spine occurs in absence of inflammation in the SIJ and consequently what the utility is of including inflammation in the spine in the definition of a positive MRI.
Objectives To analyze the prevalence of spinal inflammation on MRI in patients with chronic back pain (CBP) at baseline and one-year follow-up, and to evaluate the yield of adding MRI-spine as imaging criterion to the ASAS classification criteria for axial SpA.
Methods The SPACE-cohort includes patients with CBP (≥3 months, ≤2 years, onset <45 years) from five participating centres in Europe. All available baseline (BL) and one-year follow-up (FU) MRI of SIJ and spine were scored by 2 well-calibrated readers. MRI-SI were scored according to the ASAS definition. Bone marrow oedema suggestive of axSpA was assessed in the entire spine and only counted if visible on ≥2 consecutive slices. To define a positive MRI-spine, two cut-off values were used: ≥3 inflammatory lesions (ASAS consensus definition) and ≥5 inflammatory lesions (defined as the best cut-off in earlier analyses). Adjudication for the ASAS definition by an experienced reader was performed in case of disagreement and all modalities were considered positive if 2/3 readers agreed.
Results All patients with both MRI-spine and MRI-SIJ available at BL (n=329) and FU (n=168) were included in the analyses. At BL 43/329 (13.1%) of patients had a positive MRI-SIJ, of which 7/43 (16.3%) patients had a positive MRI-spine (ASAS consensus definition, ≥3 inflammatory lesions) and 2/43 (4.7%) if defined by ≥5 inflammatory lesions. Positive MRI-SIJ at FU was seen in 28/168 (16.7%) patients, 14 of which were also positive at BL; MRI-spine positivity was identified in 2/28 (7.1%) and 1/28 (3.6%) patients for the ASAS definition defined by ≥3 and ≥5 inflammatory lesions, respectively. In total, 4 patients had a positive MRI-spine and a negative MRI-SIJ: at BL 2 patients according to the ASAS definition of whom 1 also fulfilled the alternative definition. At FU this was 2 (different patients than at baseline) and 0 patients, respectively. Addition of MRI-spine to the classification criteria by the ASAS definition of ≥3 inflammatory lesions would lead to classification of 3 additional patients via imaging arm, with 1 patient already fulfilling the clinical arm.
Conclusions In this cohort, a positive MRI-spine in the absence of sacroiliitis on MRI was rarely seen Addition of MRI-spine as an imaging criterion to the ASAS criteria had a low yield in number of classifications. Therefore, performing MRI of the spine at either baseline or one-year follow-up is of little value in patients with short duration of CBP and suspicion of axial SpA.
Disclosure of Interest None declared