Background/Purpose: Recent studies tend to demonstrate presence of MRI erosion in Systemic Lupus Erythematosous (SLE) and primary Sjogren’s syndrome (pSS). The objective of this study was to describe MRI characteristics (erosion and osteitis, i.e. bone marrow oedema) of patients with SLE and pSS without association with rheumatoid arthritis (RA) compared to “positive control”,i.e. RA and “negative control” (sex/age- matched healthy controls ).

Methods: pSS, SLE, RA  and HC were prospectively included from 2 university departments (rheumatology and internal medicine) between 2009 and 2011. Inclusion criteria were: 1- for lupus and primary SS : disease duration >2 years, no association with RA  defined as no arthritis, no ACPA, normal X rays of hand and feet. 2-for Healthy controls (HC): no history of tender or swollen joint or rheumatic disease. 3- for RA: established RA with duration>2 years. MRI of MCP2 to 5 and wrist of the dominant hand was performed for all subjects using a dedicated MRI (ESAOTE Cscan 0.2 Tesla) in coronal and axial plans using T1 and STIR sequences, without gadolinium injection, in order to evaluate erosion and osteitis according to the OMERACT definitions. Adaptation of the OMERACT definition for erosion was used in order to differenciate physiological cortical break,i.e. vascular foramen and erosion due to pathological process. MRIs were evaluated by 2 independent readers, blindly to clinical and radiographic data using RAMRIS scores for erosion and osteitis. Statistics were performed using SAS 9.3 software.

Results: 90 subjects were included prospectively in the study: 19 pSS, 21 SLE, 30 RA and 20 HC. 83 % of  RA patients were erosive on Xrays, 86%  and 72 % were RF and ACPA positive. All SLE and pSS patients had normal Xrays of hand and feet (inclusion criteria). All patients and HC had at least one cortical break seen in 2 plans on MRI of hand and wrist without statistical difference between the groups. Frequencies and scores for erosion and for osteitis were statistically different between RA and SLE/pSS while no statistical difference was seen between SLE/pSS and HC (Table). Sensitivity and specificity were respectively for erosion: 0.93 and 0.38, for erosion with grade ≥2: 0.4 and 0.88 and for osteitis: 0.77 and 0.78. A cut-off of RAMRIS erosion was determined at 9 and could discriminate SLE/pss and RA patients (AUC= 0.8007) with good sensitivity (0.68) and specificity (0.84).Erosions in wrist and MCP3 were frequent in all groups, while erosions in MCP2, MCP4 and MCP5 were more frequently observed in RA with statistical difference between the groups.

	RA (n=30)	Lupus (n=21)	SS (n=19)	HC (n=20)	SLE /pSS vs RA	SLE / pSS vs HC
Patient with at least one Cortical break, n(%)	30(100%)	21(100%)	19(100%)	20(100%)	NS	NS
Patient with at least one erosion, n(%)	28(93%)	12(57%)	13(68%)	12(60%)	0.004	1
Patient with at least one erosion ≥grade2, n(%)	22(69%)	9(43%)	7(36%)	5 (25%)	0.019	0.39
Patient with at least one osteitis, n(%)	23(77%)	3(14%)	5(26%)	5(25%)	<.0001	0.74
Patient with at least one erosion and one osteitis, n(%)	22(69%)	3(14%)	5(26%)	5(25%)	<.0001	0.74
RAMRISerosion	21.6±10.34	13.02±5.59	11.56±5.73	10.69±4.54	0.0003	0.40
RAMRISosteitis	2.29±4.69	0.17±0.67	0.11±0.27	0.29±0.79	0.0005	0.63

Conclusion: MRI Cortical break are frequent in SLE and pSS as well as in HC. Erosions may be seen in SLE and pSS but can be distinguished from erosions of RA. Distinction of physiological cortical break and pathological erosion may lead to better interpretation of MRI and further consensus on MRI definition of these 2 items should be to consider.

---

« Back to ACR/ARHP Annual Meeting 2013

ACR Meeting Abstracts - https://acrabstracts.org/abstract/systemic-lupus-erythematosous-and-primary-sjogrens-syndrome-may-display-joint-erosions-on-mri-as-well-as-healthy-control-but-cannot-be-considered-as-erosive-disease-such-as-rheumatoid-arthri/