MRI Spots Sacroiliac Joint Damage Progression That X-rays Miss in Axial Spondyloarthritis

Magnetic resonance imaging detects changes in structural lesions of the sacroiliac joints in axial spondyloarthritis far more often than radiography, according to a longitudinal comparative study of patients in routine clinical care across four European countries. The work, conducted as part of the EuroSpA Imaging Project and published in Arthritis Research & Therapy, directly compared the two imaging modalities in the same patients at two time points at least one year apart.

The study included 105 patients with axial spondyloarthritis, of whom 65 were male. Their mean age was 38.1 years, with a standard deviation of 10.9 years, and their mean symptom duration was 9.1 years. The average interval between the two imaging time points was 2.6 years. Patients were eligible only if both magnetic resonance images and radiographs of the sacroiliac joints, taken no more than six months apart, were available at both time points.

All images were evaluated centrally and independently by two trained readers who were blinded to the chronological order of the images and to the results of the other modality. The readers applied the Spondyloarthritis Research Consortium of Canada magnetic resonance imaging scores for the sacroiliac joints and the modified New York criteria for radiographs. The researchers then assessed how well the two methods agreed, both at a single time point and in detecting change over time.

At baseline, agreement between radiography and magnetic resonance imaging depended strongly on the lesion type. For erosion detection, agreement was slight to fair, with kappa values ranging from 0.12 to 0.42. For sclerosis, agreement was poor to slight, with kappa values between −0.04 and 0.07. Only for ankylosis, the most advanced structural change, did the two modalities align well, with substantial to almost perfect agreement reflected in kappa values of 0.66 to 0.86.

Agreement on change over time was markedly weaker. Across lesion types, kappa values for change ranged from −0.04 to 0.20, a range the authors characterize as poor to slight. This indicates that the two imaging methods rarely identified the same patients as having progressed or regressed during the follow-up interval.

The divergence in detected change was striking. On radiographs, structural lesions remained largely stable, with six or fewer patients showing a change in scores for each lesion type. In contrast, magnetic resonance imaging detected change in at least one lesion type in at least half of the patients. For sclerosis, the magnetic resonance scores showed positive change in 27 patients and negative change in 30; for erosion, 51 patients showed positive and 38 negative change; and for ankylosis, 14 showed positive and 5 negative change.

The authors also examined how a composite magnetic resonance imaging score of structural damage related to the radiographic grades of the modified New York criteria.

The researchers conclude that, compared with radiography, magnetic resonance imaging detected more changes in sacroiliac joint structural lesions over time, reflecting its ability to capture dynamic changes that are not readily visible on radiographs. Because the study enrolled patients from routine clinical care rather than a controlled trial population, the findings reflect everyday imaging practice, though the observational design and modest sample size mean the results describe detection performance in this cohort rather than establishing clinical consequences of the detected changes.

Subject of Research: A longitudinal comparison of MRI and radiography for detecting progression of sacroiliac joint structural lesions in axial spondyloarthritis.

Article Title: Progression of sacroiliac joint structural lesions in axial spondyloarthritis: a longitudinal comparative study of MRI and radiography among European patients in routine clinical care

Article References: Steimer, A., Micheroli, R., Rodrigues, C. L., Brändli, J., Hadsbjerg, A. E. F., Krabbe, S., Vladimirova, N., Eshed, I., Diekhoff, T., Lambert, R. G. W., Pedersen, S. J., de Hooge, M., Bubová, K., Gregová, M., Nissen, M. J., Möller, B., Závada, J., Snoj, Z., Pintaric, K., … Ciurea, A. (2026). Progression of sacroiliac joint structural lesions in axial spondyloarthritis: a longitudinal comparative study of MRI and radiography among European patients in routine clinical care. Arthritis Research & Therapy. https://doi.org/10.1186/s13075-026-03921-8

Image Credits: AI Generated

DOI: 10.1186/s13075-026-03921-8

Keywords: axial spondyloarthritis, sacroiliac joints, MRI, radiography, structural lesions, longitudinal study, EuroSpA Imaging Project, erosion, sclerosis, ankylosis, modified New York criteria, routine clinical care