Background/Purpose: Little is known about the prevalence of structural lesions (erosions, syndesmophytes, fatty lesions) on MRI of the spine (MRI-spine) in patients with chronic back pain. Therefore the purpose of our study was to investigate the prevalence of structural lesions on MRI-spine in patients with chronic back pain.

Methods: Patients with back pain (≥3 months, ≤2 years, onset <45 years) recruited from 5 participating centres in Europe were included in the SPondyloArthritis Caught Early (SPACE)-cohort. Patients underwent MRI-spine scored independently by 3 blinded, well-calibrated readers for presence of erosions, fatty lesions and (bridging) syndesmophytes on T1-weighted images (STIR images viewed simultaneously). Fatty lesions suggestive of spondylitis were scored when visible on ≥2 consecutive slices. For erosions and syndesmophytes, suggestive of spondylitis, the presence on ≥1 slice was sufficient.  Agreement of 2/3 readers was used. The prevalence of structural lesions was calculated based on several cut-offs.

Results: All patients with complete MRI-spine data were included for analysis (n=306). All patients were grouped according to the ASAS axSpA criteria; imaging arm (fulfilling the modified New York criteria (mNY+) and not fulfilling the modified New York criteria (mNY-)) and clinical arm, possible SpA (patients not fulfilling the ASAS axSpA criteria but with 1 SpA feature with a LR+ of ≥6 or 2 SpA features with a LR+ <6) and no-axSpA. Prevalence of fatty lesions was the highest of all structural lesions in the spine in all subgroups. With a cut-off of ≥3 fatty lesions, the ‘false-positive’ rate in the no-axSpA group was low (6.5%) with still a considerable prevalence in the imaging and radiographic axSpA groups. The prevalence of erosions and syndesmophytes was generally low (see the table).

Conclusion: The prevalence of fatty lesions on MRI-spine is the highest of any structural lesions in all subgroups. Fatty lesions with a cut-off of ≥3 are primarily present in patients fulfilling the ASAS axSpA criteria, infrequent in possible axSpA patients and no-axSpA patients. Erosions and syndesmophytes in the spine are equally present in all groups. Fatty lesions are the only type of structural lesions that can potentially discriminate between patients with and without axSpA.

	ASAS axSpA, n=126			Possible axSpA, n=116	No-axSpA, n=64
	Imaging-arm		Clinical-arm, n=54
	mNY+ n=26	mNY- n=46
Fatty lesions ≥1	9 (34.6%)	18 (39.1%)	8 (14.8%)	19 (16.4%)	8 (12.5%)
Fatty lesions ≥2	5 (19.2%)	14 (30.4%)	6 (11.1%)	12 (10.3%)	7 (10.9%)
Fatty lesions ≥3	4 (15.4%)	11 (23.9%)	3 (5.6%)	10 (8.6%)	3 (4.7%)
Fatty lesions ≥4	4 (15.4%)	9 (19.6%)	2 (3.7%)	5 (4.3%)	2 (3.1%)
Fatty lesions ≥5	4 (15.4%)	7 (15.4%)	2 (3.7%)	4 (3.4%)	2 (3.1%)
Fatty lesions ≥6	3 (11.5%)	5 (10.9%)	2 (3.7%)	3 (2.6%)	2 (3.1%)
Fatty lesions ≥7	3 (11.5%)	4 (8.7%)	2 (3.7%)	3 (2.6%)	2 (3.1%)
Erosions ≥1	5 (19.2%)	11 (23.9%)	9 (16.7%)	20 (17.2%)	10 (15.6%)
Erosions ≥2	2 (7.7%)	6 (13.0%)	5 (9.3%)	10 (8.6%)	7 (10.9%)
Erosions ≥3	1 (3.8%)	2 (4.3%)	2 (3.7%)	3 (2.6%)	4 (6.3%)
Erosions ≥4	0	1 (2.2%)	0	3 (2.6%)	0
Erosions ≥5	0	0	0	1 (0.9%)	0
Syndesmophyt ≥1	5 (9.3%)	10 (21.7%)	1 (3.8%)	17 (14.7%)	7 (10.9%)
Syndesmophyt ≥2	1 (1.9%)	2 (4.3%)	0	8 (6.9%)	3 (4.7%)
Syndesmophyt ≥3	1 (1.9%)	0	0	2 (1.7%)	2 (3.1%)

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/prevalence-of-structural-lesions-seen-on-mri-spine-in-patients-with-possible-axial-spondyloarthritis-axspa-included-in-the-space-cohort/