gms | German Medical Science

44. Kongress der Deutschen Gesellschaft für Rheumatologie, 30. Jahrestagung der Deutschen Gesellschaft für Orthopädische Rheumatologie, 26. Jahrestagung der Gesellschaft für Kinder- und Jugendrheumatologie

31.08. - 03.09.2016, Frankfurt am Main

Which is the most reliable imaging method for detection of structural changes in the sacroiliac joints of patients with ankylosing spondylitis? A cross-sectional study comparing MRI, CT and conventional radiographs

Meeting Abstract

  • Xenofon Baraliakos - Rheumazentrum Ruhrgebiet, Herne
  • Florian Hoffmann - Rheumazentrum Ruhrgebiet, Herne
  • Xiaohu Deng - Chinese PLA General Hospital, Department of Rheumatology, Beijing, China
  • Yanyan Wang - Chinese PLA General Hospital, Department of Rheumatology, Beijing, China
  • Huang Feng - Chinese PLA General Hospital, Department of Rheumatology, Beijing, China
  • Jürgen Braun - Rheumazentrum Ruhrgebiet, Herne

Deutsche Gesellschaft für Rheumatologie. Deutsche Gesellschaft für Orthopädische Rheumatologie. Gesellschaft für Kinder- und Jugendrheumatologie. 44. Kongress der Deutschen Gesellschaft für Rheumatologie (DGRh); 30. Jahrestagung der Deutschen Gesellschaft für Orthopädische Rheumatologie (DGORh); 26. Jahrestagung der Gesellschaft für Kinder- und Jugendrheumatologie (GKJR). Frankfurt am Main, 31.08.-03.09.2016. Düsseldorf: German Medical Science GMS Publishing House; 2016. DocSP.39

doi: 10.3205/16dgrh115, urn:nbn:de:0183-16dgrh1151

Published: August 29, 2016

© 2016 Baraliakos et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Background: Structural changes (erosions, sclerosis and ankylosis) assessed by conventional radiographs (CR) or CT are characteristic for AS. Direct comparisons of CR, CT and MRI have not been performed to date. We compared the reliability of CR and MRI vs. CT for detection of structural changes in SIJs of AS patients.

Methods: Complete sets of MRI, CT and CR of SIJs of 69 AS patients and 49 age- and gender-matched controls in whom CTs had been performed for other reasons than back pain were analyzed. Two readers evaluated the images independently and blinded to patient´s characteristics. Assessment of lesions was performed based on SIJ-quadrants (SQ).

Results: The mean age of AS patients was 44.6 years, 72.5% were male, the mean time since diagnosis was 4.8±5.8 years (range 1-14 years). In total, 552 SQ (276 pairs) were analyzed.

Erosions were found in 131 (23.7%) SQ by CR, 141 (25.5%) by CT and in 167 (30.3%) SQ on T1-MRI. Agreement for erosions was seen for 64 SQ assessed by CR/CT, 100 SQ by CT/MRI and 70 SQ by CR/MRI, with 48.9% of SQ detected by CR also seen on CT and 45.4% detected on CT also seen by CR. The corresponding numbers for CT/MRI were 70.9% and 59.9% and for CR/MRI 53.5% and 41.9%, respectively. Disagreement for erosions was found in 144 (26.1%), 108 (19.6%) and 158 (28.6%) SQ, respectively.

Sclerosis was seen in 86 SQ on CR (15.6%), 91 SQ on CT (16.5%) and 63 on T1-MRI (11.4%) and ankylosis was seen in 91 SQ pairs on CR (33,3%), 130 SQ pairs on CT (47,1%) and 106 SQ pairs on MRI (38,4%), with similar agreement and disagreement rates.

In controls, 392 SQ were analyzed. Erosions were found in only 19 (4.8%) SQ and sclerosis in 23 (5.9%) SQ, while no patient showed ankylosis.

Conclusion: Erosions and ankylosis are more common than sclerosis in SIJs of AS patients and rarely seen in controls. The agreement between methods was rather limited. Compared to CT, less erosions were detected by MRI and CR when only erosions agreed on were counted. CT and MRI were more reliable than CR for the detection of ankylosis.