FRI0577 Mri-bone marrow oedema combined with low-dose ct scanning perform optimally in the diagnosis of axial spondyloarthritis. (12th June 2018)

• Record Type: Journal Article
• Authors: Ye, L.; Brown, M.A.; Chen, D.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 812
• Page End: 812
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.6071
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: The evidence of sacroiliitis on radiography and MRI have been regarded as crucial for diagnosis of ankylosing spondylitis (AS) and axial spondyloarthritis (axSpA). Low-dose CT (ldCT) has advantages on sacroiliitis has not been taken into consideration until now. Objectives: To compare the performance of plain radiography, ldCT, and MRI for facilitating the diagnosis of sacroiliitis in radiographic negative axial spondyloarthritis (nr-axSpA) or AS from suspected axSpA. Methods: Patients presenting with chronic back pain (>3 months duration) were recruited and assessed by axSpA-experienced rheumatologists. Patients not meeting the ASAS clinical criteria for axSpA were recruited as controls, and divided into non-inflammatory and inflammatory groups on the basis of presence of inflammatory back pain and/or CRP/ESR elevation. The ASAS clinical axSpA or modified New York AS criteria were used to define nr-axSpA or AS cases respectively. Clinical variables, pelvic radiography, sacroiliac joint (SIJ) ldCT (voltage 120 kV, electricity 70mAs) and SIJ MRI (SPAIR, T1-weighted, T2-weighted sequences) were obtained in one week period. All images were read by two expert musculoskeletal radiologists independently and differences in initial reads resolved by consensus. Plain radiographic and ldCT SIJ images were graded 0–4 according to the modified New York criteria. MRI-BMO was defined by the presence of periarticular or subchondral BMO lesions. Fat deposition, sclerosis, erosions or bony ankylosis were regarded as signs MRI-structural lesions. Subjects were considered MRI positive if they had either BMO or structural lesions. Results: 130 patients were included in the study and had SIJ radiography and ldCT, of whom 71 additionally had SIJ MRI. 28 (39.3% female, 36% B27 prevalence) non-inflammatory controls, 24 (33.3% female, 50% B27 prevalence) inflammatory controls, 34 (26.5% female, 86.3% B27 prevalence) nr-axSpA and 44 (36.4% female, 95.3% B27 prevalence) AS cases were recruited in total. Positive imaging results according to clinical diagnoses are given in the table. These findings show that ldCT had much higher sensitivity than radiography for nr-axSpA and similar specificity. Whilst no non-inflammatory control was positive for ldCT and only one for radiography, three inflammatory controls were positive for ldCT but negative for radiography, with two being negative by MRI, and all being HLA-B27 negative, suggesting that they may be false positives. MRI-BMO had the highest sensitivity for nr-axSpA, but had lower specificity, with 31% of non-inflammatory controls being positive for this modality, and lower sensitivity for AS. MRI-structural had intermediate performance with slightly lower sensitivity to ldCT for both AS and nr-axSpA, but lower specificity than ldCT, and lower sensitivity for nr-axSpA than MRI-BMO. Conclusions: ldCT and MRI-BMO examination are more sensitive than either plain radiography or MRI-structural assessment, but MRI overall is less specific than plain radiography or ldCT. The relative position of these imaging modalities in screening patients for axSpA needs to be reconsidered in the light of these findings, also taking into account the costs involved. Reference: [1] Van der Linden S, et al. Arthritis Rheum1984;27:361–8. [2] Rudwaleit M, et al. Ann Rheum Dis2009;68:1520–7. Acknowledgements: We thank radiologists for data collection and images reading. Disclosure of Interest: None declared