Background Radiologic evidence of sacroiliitis is important for the diagnosis, classification, and management of patients with ankylosing spondylitis (AS)[1]. Conventional radiography (CR) has been the most widely utilized imaging modality for the assessment of sacroiliac involvement in axial spondyloarthritis (AxSpA) because of the low radiation dose, ease of operation, and low expenses. CR is used in both the 1984 modified New York criteria and ASAS criteria for classifying AS and AxSpA but lacks sensitivity for early changes. Though MRI detects early changes of the sacroiliac joints, as well as chronic structural changes, its use is limited by the associated cost, procedural time and certain contraindications. Computed tomography (CT) is a modality that enables visualization of erosions, sclerosis, and new bone formation, with the added benefit of multiplanar cross-sectional imaging.

Objectives We aimed to analyze the diagnostic value of computer tomography (CT) features of sacroiliitis for ankylosing spondylitis in patients with inconclusive CR evidence for sacroiliitis.

Methods In this retrospective monocentric observational study, 50 patients with chronic low back pain (LPB) with a duration longer than three months in the lumbosacral region were included. Using the patient record data were extracted on age, pain duration, and plain radiography. Eligible patients should have had plain radiography without evident radiographic sacroiliitis. All the patients had undergone a CT scan of the sacroiliac joint that a radiologist and rheumatologist evaluated for subchondral osteosclerosis, erosions, joint space narrowing (JSN), subchondral cysts, and ankylosis. Based on the clinical and instrumental findings, including structural changes characteristic of radiographic sacroiliitis, a final decision by a certified rheumatologist was made and a diagnosis of ankylosing spondylitis was set or ruled out.

Results The mean age and duration of LBP of the patients were 44.7 (14.7) years and 63.9 months, respectively. Of the 50 included patients, 28 (56%) were females. The mean values of c-reactive protein and erythrocyte sedimentation rate were 13.9 mg/l and 39.4 mm/h. Subchondral osteosclerosis was found in 44 patients (88%), JSN – in 21 patients (42%), erosions – in 17 patients (34%), subchondral cysts – in 10 patients (20%), ankylosis – in 18 patients (36%). Definite CT data for sacroiliitis was seen in 24 patients (48%) and the diagnosis of ankylosing spondylitis was set in 23 patients (46%) of the study group. The likelihood ratios (LR+) for diagnosis of AS were high for erosions (18.4), subchondral cysts (11), ankylosis (6.9) and low for JSN (LR+ 1.9) and ankylosis (LR+ 1.1).

Conclusion Nearly half of the patients with inconclusive CR evidence for sacroiliitis were diagnosed with AS after CT imaging. Erosions seen on CT increase the likelihood of assuming AS diagnosis. Computed tomography is a useful tool in diagnosing AS, but it is associated with higher ionizing radiation doses.

Reference [1]Mandl P, Navarro-Compán V, Terslev L, Aegerter P, van der Heijde D, D’Agostino MA, Baraliakos X, Pedersen SJ, Jurik AG, Naredo E, Schueller-Weidekamm C, Weber U, Wick MC, Bakker PA, Filippucci E, Conaghan PG, Rudwaleit M, Schett G, Sieper J, Tarp S, Marzo-Ortega H, Østergaard M; European League Against Rheumatism (EULAR). EULAR recommendations for the use of imaging in the diagnosis and management of spondyloarthritis in clinical practice. Ann Rheum Dis. 2015 Jul;74(7):1327-39. doi: 10.1136/annrheumdis-2014-206971. Epub 2015 Apr 2. PMID: 25837448.

Acknowledgements: NIL.

Disclosure of Interests None Declared.