Background/Purpose:

Conventional x-rays of the sacroiliac joints (SIJ) remains the first imaging method in case of suspicion of axial spondyloarthritis (SpA). Moreover, a diagnosis of definite ankylosing spondylitis (AS) according to the modified New York criteria relies on the presence of definite radiographic sacroiliitis (SI). Magnetic resonance imaging (MRI) is a reliable method of detection of active inflammatory changes in the SIJ (active SI) and is potentially able to detect also chronic structural changes (such as sclerosis, erosions, and ankylosis) visible on conventional X-rays. Furthermore, chronic changes might be better visible in MRI because of tomography and MRI is not associated with radiation exposure. However, reliability of MRI in detection of chronic structural changes in SIJ remains unclear.

 This study was aimed at comparing the performance of MRI in comparison to conventional X-rays in detection of chronic structural changes in patients with axial SpA.

Methods:

We included 112 patients with definite axial SpA (68 with AS and 44 with non-radiographic axial SpA – nr-axSpA), for whom sets consisting of an X-rays of the SIJ and an MRI of the SIJ (at least in a T1-weighted sequence) performed at the same time point were available. X-rays of the SIJ were scored according to the modified New York criteria (grade 0 to grade IV) and according to the recently proposed new scoring system [1], which contains separate scores for subchondral sclerosis (score 0-2), erosions (score 0-3), and joint space changes (score 0-5) in each SIJ. MRIs of the SIJ (T1) were scored in the similar way for the same structural changes. In addition, readers were asked to provide an overall impression of the damage extent on MRI according to the scoring system of the modified New York criteria. X-rays and MRIs were scored separately by two trained readers, which were blinded for all clinical data and for the diagnosis of AS or nr-axSpA.

Results:

224 SIJ from 112 patients were available for the analysis. There was a moderate agreement between MRI and X-ray regarding definite subchondral sclerosis scored by both readers (Kappa=0.46, p<0.001), rather low agreement concerning definite erosions (Kappa=0.11, p=0.07), moderate agreement regarding definite joint space abnormalities (Kappa=0.41, p<0.001) and very good agreement regarding joint ankylosis (Kappa=0.85, p<0.001). Importantly, there was a good overall agreement regarding the presence of definite SI: in 84% of the SIJ (128 out of 153) with definite X-rays SI it was also seen in MRI. Interestingly, in 16% of the cases definite SI was seen in X-rays only in 18% of the cases – in MRI only. Furthermore, on the patients’ level, SI fulfilling the modified New York criteria was confirmed on MRI in 81% of the cases (55 out of 68).

Conclusion:

MRI demonstrated good overall performance regarding detection of chronic structural changes in the SIJ and was able to confirm the presence of definite sacroiliitis in more than 80% of the cases.

References

1. Poddubnyy D, et al. Ann Rheum Dis 2012;71(Suppl3):409.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/performance-of-magnetic-resonance-imaging-in-detection-of-chronic-structural-changes-in-sacroiliac-joints-as-compared-to-conventional-x-rays-in-axial-spondyloarthritis/