THU0384 IMAGING OF SACROILIAC JOINTS IN EARLY SPONDYLOARTHRITIS: SHOULD WE CHOOSE COMPUTED TOMOGRAPHY OR MAGNETIC RESONANCE IMAGING?

Abstract

Background: In view of the limited utility of pelvic radiography to recognize sacroiliitis in its earliest stages, magnetic resonance imaging (MRI) has been considered as a reliable imaging modality to detect acute subchondral inflammation at disease onset. However, several recent studies reported a high proportion of false-positive of sacroiliac joints (SIJ) MRI. Other studies demonstrated that computed tomography (CT), by its ability to detect early structural damage, may be helpful. Consequently, strengths and limitations of each imaging technique need to be considered when choosing the most appropriate first-line modality in clinical practice. Objectives: We aimed to assess the performance of CT scanning and MRI for detecting sacroiliitis in early SpA by estimating the sensitivity and the specificity of each imaging technique. Methods: Consecutive patients, aged 16 years and over, referred for symptoms suggestive of SpA from February 2014 to February 2017 were enrolled in this cohort. After excluding patients whose conventional radiography showed a confirmed sacroiliitis, eligible patients underwent SIJ CT and/or MRI. The CT and MR images were reviewed by 2 musculoskeletal radiologists blinded to clinical findings. Then, 2 rheumatologists recorded from the clinical files if patients fulfilled the ASAS (Assessment of Spondyloarthritis international Society) classification criteria for SpA. This classification was considered as the gold standard of this study. Results: Fifty-three patients were included: 14 men and 39 women. The mean age was 36 years. The prevalences of HLA–B27 and elevated CRP were 27.6% and 16%, respectively. Fifty-eight percent of the patients (n=31) fulfilled the ASAS criteria for axial SpA. Among these patients, 30 patients underwent SIJ CT and 27 underwent SIJ MRI. Sacroiliitis was visualized by CT in 28 out of 30 patients (93.3%) and by MRI in 17 out of 27 patients (63%). Among the 22 patients who did not fulfill the ASAS criteria for SpA, 21 patients underwent SIJ CT and 20 underwent SIJ MRI. Sacroiliitis was not visualized by CT in any patients and was visualized by MRI in 2 out of 20 patients (10%). Sensitivity, specificity, positive and negative likelihood ratio of CT calculated with ASAS classification as golden standard were respectively estimated at 93.3%, 100%, 100% and 91.3%. Youden index was estimated at 0.93 and Q Yule coefficient at 1. Sensitivity, specificity, positive and negative likelihood ratio of MRI calculated with ASAS classification as golden standard were respectively estimated at 62.9%, 90%, 89.5% and 64.3%. Youden index was estimated at 0.53 and Q Yule coefficient at 0.88. Conclusion: In our study, the evaluation of sacroiliitis by CT, in comparison with MRI, has shown to be more sensitive and more specific. However, other factors should also be taken into account while comparing CT and MRI such as the high radiation exposure of CT scanning in the one hand and the cost and the restricted accessibility of MRI in the other hand. Disclosure of Interests: None declared