AB1101 THE PERFORMANCE OF COMPUTED TOMOGRAPHY FOR ASSESSING SACROILIITIS IN EARLY SPONDYLOARTHRITIS: WHICH STRUCTURAL LESIONS ARE MOST CONTRIBUTIVE TO THE DIAGNOSIS?

Abstract

Background:Computed tomography (CT) of sacroiliac joints (SIJ) is usually considered as a third-line modality in the diagnosis of spondyloarthritis (SpA), as it shows only structural lesions. According to 2015 EULAR recommendations, CT may be helpful if conventional radiograph is inconclusive and MRI cannot be performed [1]. However, its contribution for the early assessment of sacroiliitis is still little-studied.Objectives:The purpose of this study was to assess the performance of CT scan for detecting sacroiliitis and the contribution of the different structural lesions to the early diagnosis of SpA.Methods:This cross-sectional prospective monocentric study included consecutive patients, aged over 16, consulting for symptoms suggestive of SpA between February 2014 and 2017. Patients with a pelvic radiograph showing a confirmed sacroiliitis (making CT unnecessary to confirm the diagnostic) were not included. Eligible patients underwent CT of SIJ and the images were reviewed by 2 experimented musculoskeletal radiologists blinded to clinical findings. Then, 2 experienced rheumatologists analyzed clinical, biological and radiological data and classified patients into 2 groups: confirmed non-radiographic spondyloarthritis (nr-SpA) or no SpA. This classification was considered as the gold standard of this study.Results:A total of 60 patients, 45 women and 15 men, were included in this study. Forty-six patients were assessed as confirmed nr-SpA (76.6%) and 14 patients as no SpA (23.4%). Mean age at the diagnosis was 39.5±10.8 years [17-59], and the disease duration from first symptoms was 48.6 ± 40 months [6.6-180]. Chronic back pain met the Calin criteria in 37 patients (61.7%), the Berlin criteria in 46 patients (76.7%) and the ASAS criteria in 32 patients (53.3%).Among the patients diagnosed as confirmed SpA, 71.7% were determined to have sacroiliitis at CT. As shown by the following table, the most sensitive lesion was subchondral bone sclerosis and the most specific one was new bone formation at entheses:TableConfirmed SpANo SpAPSe (%)Sp (%)PPV (%)NPV (%)Interruption of subchondral bone67.4%28.6%0.01 (S)67.471.488.640subchondral bone erosions65.2%35.7%0.05 (S)65.264.385.736Subchondral bone sclerosis78.3%28.6%0.001 (S)78.371.49050new bone formation at entheses10.9%7.14%0.68 (NS)10.992.983.324.1CT conclusion: sacroiliitis71.7%28.6%0.04 (S)71.771.489.243.5CT: computed tomography, SpA: Spondyloarthritis, Se: Sensitivity, Sp: Specificity, PPV: Positive Predictive Value, NPV: Negative Predictive ValueConclusion:The relatively limited place of SIJ CT can be explained by the lack of investigations evaluating its real contribution in detecting sacroiliitis. In this study, we found that the main SI structural lesions were significantly associated to the final diagnosis of SpA with a sensitivity up to 78.3% and a specificity up to 92.9%. The main limitation of this technique remains the large radiation dose delivered to the gonads, especially with the commonly used axial position for CT. However, recent studies found that oblique coronal CT imaging seems to be a very promising technique, since it allows visualization of the subtle radiologic signs of sacroiliitis while delivering less radiation dose [2].References:[1]Mandl P, Navarro-Compán V, Terslev L, et al (2015) EULAR recommendations for the use of imaging in the diagnosis and management of spondyloarthritis in clinical practice. Ann Rheum Dis 74:1327–1339.https://doi.org/10.1136/annrheumdis-2014-206971[2]Li SG, Liu X, Zhou H, Zhang Q (2018) Interrater reliability and radiation dosage of oblique coronal computed tomography. Br J Radiol 91:6.https://doi.org/10.1259/bjr.20150700Disclosure of Interests:None declared