FRI0636 SUSCEPTIBILITY WEIGHTED SEQUENCES IN MAGNETIC RESONANCE IMAGINGCAN CREATE COMPUTED TOMOGRAPHY-LIKE IMAGES AND IMPROVE THE ACCURACY OF STRUCTURAL LESION DETECTION OF THE SACROILIAC JOINT IN AXIAL SPONDYLOARTHRITIS. (June 2019)

• Record Type: Journal Article
• Authors: Diekhoff, Torsten; Hermann, Kay-Geert; Proft, Fabian; Protopopov, Mikhail; Poddubnyy, Denis; Makowski, Marcus R.
• Is Part Of: Annals of the rheumatic diseases. Volume 78(2019)Supplement 2
• Page Start: 1015
• Page End: 1016
• Publication Date: 2019-06
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2019-eular.6790
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: In patients with axial spondyloarthritis (axSpA) the accurate depiction of structural lesions in magnetic resonance imaging (MRI) is impeded because MRI – while gaining signal from the bone marrow – allows only an indirect depiction of the cortical bone and sclerotic areas. Computed tomography (CT) is considered to be the gold standard as it displays the calcium-containing cortical and trabecular bone substance. However, the exhibited radiation exposure in CT limits its application in the clinical routine. Susceptibility weighted imaging (SWI) is an advanced MRI technique, that is able to visualise calcium due to its magnetic properties and allows a reconstruction of CT-like images without radiation exposure. (1) Objectives: To investigate the diagnostic accuracy of CT-like images generated from SWI for the detection of structural lesions of the SI-joint. Methods: Twentytwo patients with suspicion of or known axSpA were included. All patients underwent a 1.5-Tesla-MRI including T1 and SWI sequences and a low-dose CT of the sacroiliac joints. CT images were reconstructed in 4 mm oblique coronal reconstructions matching the orientation and slice thickness of the MRI. MRI and CT images were scored for erosions, sclerosis and joint space alterations applying the 24 regions method, a modification of the Berlin score. (2) Using CT as standard of reference, sensitivity (SE) and specificity (SP) values were calculated on patients' level. A score of 2 or higher in any quadrant was considered positive. The sumscores of T1 and SWI were compared to CT using Pearson's test. Results: The mean erosion sumscore was 5.5 ± 5.5, 6.3 ± 8.2 and 5.9 ± 9.2 for T1, SWI and CT respectively. The mean sclerosis sumscore was 0.78 ± 1.5, 5.5 ± 6.5 and 6.9 ± 6.9. Five patients had joint space alterations (pseudowidening or ankylosis) in CT. Both, T1 and SWI showed an SE of 100% and a SP of 84.5% for erosions, however, the erosion sumscore corelated better for SWI (Pearson's r: 0.93) than for T1 (r: 0.64). With respect to sclerosis T1 (SE 30%, SP 100%, r 0.42) was inferior to SWI (SE 90%, SP 92%, r 0.81). It was also less sensitive to joint space alterations (T1: SE 40%, SP 94%; SWI: SE 80%, SP 100%). Conclusion: While T1 weighted MRI provides sufficient diagnostic accuracy for the detection of erosions, the CT-like images derived from SWI allow for a more accurate depiction of structural lesion of the SI-joint. SWI provides additional information identifying sclerosis. However, it is not suited to replace T1-weighted imaging because fatty metaplasia of the bone marrow or inside an erosion cavity cannot be detected. References: [1] Bender YY, Diederichs G, Walter TC, et al. Differentiation of Osteophytes and Disc Herniations in Spinal Radiculopathy Using Susceptibility-Weighted Magnetic Resonance Imaging. Invest Radiol. 2016. [2] Diekhoff T, Hermann KA, Greese J, et al. Comparison of MRI with radiography for detecting structural lesions of the sacroiliac joint using CT as standard of reference: results from the SIMACT study. Ann Rheum Dis. 2017. Disclosure of Interests: Torsten Diekhoff Paid instructor for: MSD, AbbVie, Novartis, Canon, but less than 10.000€, Kay-Geert Hermann Paid instructor for: MSD, AbbVie, Novartis, but less than 10.000€, Fabian Proft Grant/research support from: Novartis, Consultant for: yes but less than 10.000, Paid instructor for: yes but less than 10.000, Speakers bureau: yes but less than 10.000, Mikhail Protopopov: None declared, Denis Poddubnyy Grant/research support from: AbbVie, Merck Sharp & Dohme, Novartis, Consultant for: AbbVie, Bristol-Myers Squibb, Janssen, Merck Sharp & Dohme, Novartis, Pfizer, UCB Pharma, Speakers bureau: AbbVie, Bristol-Myers Squibb, Janssen, Merck Sharp & Dohme, Novartis, Pfizer, Roche, UCB Pharma, Marcus R Makowski: None declared