OP0303 CLEAR DIFFERENCES IN STRUCTURAL MRI LESIONS BETWEEN AXIAL SPONDYLOARTHRITIS PATIENTS AND PATIENTS WITH CHRONIC BACK PAIN, WOMEN WITH POSTPARTUM BACK PAIN, RUNNERS AND HEALTHY SUBJECTS

• Published in: Annals of the rheumatic diseases Vol. 83; no. Suppl 1; p. 117
• Main Authors: De Hooge, M., Kerami, Z., Van de Sande, M. G. H., Van der Heijde, D., Fagerli, K. M., Ramonda, R., Weel-Koenders, A. E. A. M., Landewé, R.
• Format: Journal Article
• Language: English
• Published: Kidlington BMJ Publishing Group Ltd and European League Against Rheumatism 01.06.2024; Elsevier B.V; Elsevier Limited
• Subjects: Ankylosis; Arthritis; Back pain; Bone; Chronic pain; Descriptive Studies; Imaging; Inflammation; Inflammatory diseases; Joint diseases; Lesions; Magnetic Resonance Imaging; Postpartum; Rheumatic diseases; Rheumatology; Sacroiliitis; Scientific Abstracts; Sclerosis; Magnetic Resonance Imaging; Descriptive Studies; Bone; Imaging
• ISSN: 0003-4967; 1468-2060
• DOI: 10.1136/annrheumdis-2024-eular.438

Background:Spondyloarthritis (SpA)-like BME lesions occur in axial (axSpA) and non-SpA subgroups [1]. Structural lesions on MRI of the sacroiliac joints (MRI-SI) have been reported in non-SpA populations but not in direct comparison to patients with axSpA. Cut-off definitions have been proposed trying to better express the distinctive nature of structural axSpA lesions [2,3].Objectives:To investigate the extent and performance of structural lesions on MRI-SI in patients with axSpA, chronic back pain, women with postpartum back pain, runners and healthy subjects.Methods:In this study, 47 patients with diagnosed axSpA and previously assessed sacroiliitis on MRI, and 47 chronic back pain (CBP) patients (irrespective of MRI findings) were selected from the SPACE cohort. These patients were matched with 47 healthy individuals based on age and sex. Further, this study includes 7 women with postpartum back pain and 24 runners. Two well-trained readers blinded and randomly assessed MRI-SI of 172 subjects for erosions, fatty lesions, sclerosis and ankylosis. Readers used a scorings system similar to the Spondyloarthritis Research Consortium of Canada (SPARCC) scoring methodology, based on the presence or absence of individual lesions in four quadrants per SI joint on six consecutive slices covering the cartilaginous part of the joint. This resulted in maximum scores of 48 for sclerosis, 48 for erosions, 48 for fatty lesions and 24 for ankylosis (assessed per half SI joint). Cut-off values proposed by de Hooge et al. (MdH) and the Assessment of SpondyloArthritis international Society Imaging workgroup (ASAS) were tested [2,3]. Finally, concordant presence of inflammatory lesions (previously reported,1) and structural lesions was reported per subgroups. Data is based on reader consensus score presenting lowest number of lesions reported by both readers.Results:Mean age was between 30 and 34 years, 45% of healthy subjects, axSpA, CBP pts and 50% of runners was male. From pts with axSpA 29.5% fulfilled the modified New York criteria. Structural lesions were found in 13% of non-axSpA subjects and 79% of axSpA pts. Erosions were the most commonly seen structural lesion in axSpA pts, CBP pts and women with postpartum back pain at 75%, 9% and 57%, respectively. In the runners subgroup, both erosions and fatty lesions were most common (8%, present in the same pts). Fatty lesions were most common in healthy controls at 6% (Figure 1).For MdH definitions ‘≥3 erosions’ and ‘≥5 fat and/or erosions,’ 43% and 45% of axSpA pts met the criteria without false positives. MdH definition ‘≥3 fatty lesions’ had one false positive in the healthy control subgroup, while 34% of axSpA pts met this definition. ASAS definitions for erosions were fulfilled by 55% (‘erosion on ≥2 consecutive slices’) and 60% (‘≥3 quadrants with erosion’) of axSpA patients. Notably, 2 postpartum individuals and one runner also met these definitions. ‘Fatty lesions on ≥3 consecutive slices’ was met by 36% of axSpA pts with 1 false positive in the healthy control group. Cut-off ‘≥5 quadrants with fatty lesions’ performed identical to MdH cut-off ‘≥3 fatty lesions’.Figure 2 shows a 79% overlap of inflammatory and structural lesions in axSpA pts compared to 4-29% in non-axSpA subjects. In 6% of axSpA pts structural lesions were reported in absence of inflammation, in the non-axSpA groups this is more common.Conclusion:Structural MRI-SI lesions are rare in non-SpA subjects while they are more prevalent in patients with axSpA. Nevertheless, both erosions and fatty lesions differentiate best between axSpA and non-axSpA groups. In addition, the previously proposed MdH cut-off definitions discriminate best between axSpA and non-axSpA groups. Also, in axSpA patients, structural and inflammatory lesions are predominantly concordant in the majority of cases. In non-axSpA patients, both findings occur more frequently in isolation, suggesting a more likely attribution to non-specific mechanisms in these subjects.REFERENCES:[1] Arthritis Rheumatol, 2018. 70(7): p. 1042-1048.[2] Ann Rheum Dis, 2016. 75(7): p. 1308-14.[3] Rheumatology (Oxford), 2021. 60(10): p. 4778-4789.Acknowledgements:NIL.Disclosure of Interests:None declared.