Background Diffusion weighted imaging (DWI) is a new Magnetic Resonance Imaging (MRI) sequence proposed for spondyloarthritis (SpA) diagnosis. Whether it is more useful than the traditional short tau inversion recovery (STIR) sequence in disease diagnosis had not been evaluated.

Objectives By comparing with traditional STIR sequence in a group of back pain patients newly referred to rheumatology clinics, we evaluated the usefulness of DWI in SpA diagnosis at different stages.

Methods All new patients referred to the rheumatology clinics with persistant back pain were recruited. DWI and STIR MRI were performed. Conventional radiographs of the pelvis were assessed according to the modified New York criteria for ankylosing spondylitis. Bone marrow edema (BME) and active sacroiliitis according to the ASAS definition were evaluated in STIR and DWI by two independent observers.

Results One hundred and thirty-three patients were recruited. Ninety patients (67.7%) had a clinical diagnosis of SpA. Average back pain duration was 8.5±8.9 years. The presence of Human Leukocyte Antigen (HLA) B27 was found in 42.9% of the study population. Inter-observer correlations were excellent (STIR 95.4%, p<0.001; DWI 69.5%, p<0001). DWI was found to be comparable to STIR in disease diagnosis (sensitivity DWI 34.1% vs STIR 34.3%; specificity DWI 85% vs STIR 93.8%) and when applied to the Assessment of SpondyloArthritis international Society (ASAS) criteria for axial SpA (sensitivity DWI 78.9% vs 79.5%; specificity DWI 75.0% vs 78.8%). DWI is better than STIR in non-radiographic axial SpA group (sensitivity DWI 37.8% vs STIR 33.8%; specificity DWI 85.3% vs STIR 95.6%). In the group with disease duration less than 3 years, the two images showed improved sensitivity (sensitivity DWI 34.5% vs STIR 41.4%; specificity DWI 84.6% vs STIR 88.5%).

Conclusions DWI is useful in SpA diagnosis especially in non-radiographic axial SpA group. Both DWI and STIR are especially useful for early disease diagnosis.

Disclosure of Interest None declared