Background Conventional radiography is the most common used method to detect structural lesions in the sacroiliac joints (SIJ) in axial SpondyloArthritis (axSpA). However, reliability is a problem. Besides inflammatory lesions on MRI, which are used to define a positive MRI in the ASAS criteria, structural lesions are visible on MRI.
Objectives To investigate the impact of replacement of x-rays by structural lesions on MRI on the ASAS axSpA classification of patients.
Methods Patients in the SPACE cohort (chronic back pain: ≥3 months, ≤2 years, onset <45 years) with (suspicion of) axSpA underwent MRI and X-rays of the SIJs. Three well-calibrated readers, blinded for patient characteristics, read all available baseline MRI-SI (ASAS definition) and X-SI (mod New York). MRI-SI and X-SI were considered positive if 2/3 readers agreed. In addition, MRI T1-weighted images (viewed simultaneously with STIR) were assessed on the presence of structural lesions (in each of the quadrants of each SIJ). Lesions were marked present if 1 lesion was seen on ≥2 consecutive slices. Mean scores of 2 out of 3 agreeing readers (based on ASAS definition) were used. Earlier, we proposed cut-offs to define a positive MRI-SI based on structural lesions (MRI-SI-struct) based on <5% presence among no-SpA patients. These proposed cut-offs are: erosions ≥3, fatty lesions ≥3, fat lesions and/or erosions ≥5. Patients were classified according to the ASAS axSpA-criteria and grouped in the different arms, using the various definitions of MRI-SI-struct instead of X-SI.
Results For this analysis 294 patients with complete imaging data were included. Using the cut-off for fatty lesions and/or erosions ≥5, classification did not change in 275+3+8 (286) patients (97.3%) (table). In the remaining 8 patients, 5 patients (1.7%) would not be classified axSpA if only MRI-SI-struct was performed, while 3 patients (1.0%) would be additionally classified as axSpA. Furthermore, an additional 8 patients would be classified by different arms. Very similar results were found when replacing X-rays by the definition of a positive MRI-SI-struct for the presence of fatty lesions or erosions alone (both cut-off of >3), and for the scores of the 2 readers individually.
Conclusions The replacement of x-rays by assessment of structural lesions on MRI does not lead to a different ASAS axSpA classification in the majority of the patients;. These data are promising, however, this is in a cohort of patients with a relatively low number of patients fulfilling the axSpA criteria and in the early phases of the disease. Therefore these data need to be confirmed in patients with established disease.
Disclosure of Interest None declared