FRI0480 Scoring syndesmophytes on ct spine images of patients with radiographic axial spondyloarthritis from the sensitive imaging of axial spondyloarthritis (SIAS) cohort. (15th June 2017)

• Record Type: Journal Article
• Authors: Bruin, F de; Berg, R van den; Baraliakos, X; Braun, J; Gaalen, F van; Reijnierse, M; Heijde, D van der
• Is Part Of: Annals of the rheumatic diseases. Volume 76(2017)Supplement 2
• Page Start: 669
• Page End: 669
• Publication Date: 2017-06-15
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2017-eular.3533
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Low dose Computed Tomography (CT) could potentially be more sensitive than radiographs in the follow up of patients with axial spondyloarthritis (axSpA). Objectives: Our first goal was to develop a scoring method for the spine and test its interreader variability and frequency of progression in the various parts of the spine. Methods: Patients (pts) of the Sensitive Imaging of Axial Spondyloarthritis (SIAS) cohort fulfilled the modified New York (mNY) criteria, had ≥1 syndesmophyte in cervical or lumbar spine on radiographs and ≥1 inflammatory lesion on MRI-spine. Baseline and 2 years follow up low-dose CT scans (approximately 2–3 mSV) were performed. Syndesmophytes were scored in the coronal and sagittal planes for all "corners" per view, thus scoring 8 'quadrants' per vertebral unit (VU). Syndesmophytes were scored as absent (score 0), <50% of the intervertebral disc height (IVDH) (score 1), ≥50% of the IVDH but no bridging (score 2) or as "bridging the IDVH" (score 3). Score range 0–552. Two readers, blinded for clinical information and time order, scored the CT scans paired by patient. Scores of the 2 readers were compared using heatmaps. Interreader variability was assessed by (two-way average) ICC and smallest detectable change (SDC) for the whole spine and per segment. Results: In total, 56 pts (84% male, mean age 48.6) were included. The heatmap (fig 1 ) presents the similarity in basseline and change scores per reader per VU (VU on the y-axis). Ankylosis (score 3) is most prevalent in the thoracic spine, and syndesmophytes in general are least prevalent in the cervical spine. Both readers use almost the entire possible range, pick up a similar magnitude of change and have high ICCs. Most change is present in the thoracic spine. A change ≥SDC is seen in 33.9% of the patients. Conclusions: A fine-grained scoring system for CT spine was developed. Scoring syndesmophytes with good accuracy on CT images was feasible, picking up changes over a 2-year period in a high percentage of patients, especially in the thoracic spine, which is insufficiently visible on radiographs. Disclosure of Interest: None declared