Detecting Erosions in Sacroiliac Joints in axSpA: MRI vs CT

In comparison with computed tomography, data showed that the volume-interpolated breathhold examination MRI technique is more sensitive than T1-weighted MRI for detecting structural damage in axial spondyloarthritis.

Compared with computed tomography (CT), magnetic resonance imaging (MRI) was more sensitive for detecting articular erosions at the sacroiliac joint (SIJ) in patients with axial spondyloarthritis (axSpA), with the volumetric interpolated breath-hold examination (VIBE) MRI sequence superior to the T1-weighted (T1W) MRI sequence, according to a report published in The Journal of Rheumatology.

Although CT is considered the gold standard for viewing SIJ anatomy, MRI allows imaging of both structural and inflammatory changes without radiation exposure. The VIBE MRI technique is particularly useful for detection of cartilaginous erosions, and therefore especially helpful for diagnosis of early active axSpA, before bony changes detectable by CT are visible. Investigators sought to compare MRI techniques with CT scanning for the identification of such lesions and diagnosis of axSpA.

A total of 109 patients with axSpA (mean age, 45.3 years; 67.9% male; 82.3% human leukocyte antigen-B27 positive; mean disease duration, 4.8 years; mean Bath Ankylosing Spondylitis Disease Activity Index, 4.9; mean C-reactive protein, 1.9 mg/dL) were enrolled in a retrospective image analysis. All participants had CT and MRI (T1W and VIBE) scanning performed, with images read by 2 experienced blinded readers. Erosion structural damage was evaluated according to Assessment of Spondyloarthritis international Society definitions and was recorded using SIJ quadrants (n=872 for each imaging technique).

Related Articles

More erosions were detected by VIBE MRI (199 quadrants) compared with T1W MRI (n=182; P =.031) and CT (n=153; P <.001).  

Assuming a CT gold standard for comparison, VIBE MRI demonstrated superior sensitivity to T1W MRI (71.2% vs 63.4%), but similar specificity (87.3% vs 88.2%). Linear regression analysis revealed that compared with CT, both VIBE MRI (β=0.384; P <.001) and T1W MRI (β=0.369; P <.001) saw erosion occurrence significantly associated with younger patient age; a direct comparison of the 2 MRI techniques did not yield the same correlation with age (β=0.127; P =.224).

Reader agreement was excellent for positive and negative findings and between CT and MRI analyses (intraclass correlation coefficients, 0.997 and 0.979, respectively).

“Clearly, the diagnostic and prognostic relevance of erosions detected by CT vs by VIBE-MRI and T1W-MRI deserves more study, especially in order to better understand the benefits and limitations of new techniques such as VIBE-MRI,” noted the authors.

“Although there were erosions seen by CT which were not detected by MRI techniques, VIBE detected the highest number of [SIJ-quadrants] with structural damage and erosions when all techniques were directly compared,” they concluded.

They recommended that future research involve healthy control patients without axSpA to enable the direct comparison of SIJ imaging techniques in patients with and without disease.


Baraliakos X, Hoffmann F, Deng X, Wang Y, Huang F, Braun J. Detection of erosions in the sacroiliac joints of patients with axial spondyloarthritis using the magnetic resonance imaging VIBE technique [published online February 15, 2019]. J Rheumatol. doi:10.3899/jrheum.181304