Background/Purpose:
Early diagnosis and early detection of therapeutic response within the so called “window of opportunity” in rheumatoid arthritis (RA) are essential for superior clinical and functional outcome. Magnetic Resonance Imaging (MRI) is the only modality for allowing simultaneous assessment of all the structural changes involved in RA such as synovitis, cartilage damage, bone marrow edema pattern, bone erosion and tendonsynovitis. Fast and reliable MRI quantitative methods are significant to evaluate disease activity and assist management for RA patients.
The objectives of this study were to investigate the reliability of multiple computer-aided and manual quantifications for MRI synovitis, bone marrow edema pattern, erosion and semi-quantitative cartilage loss of wrists with RA.
Methods:
Wrist imaging was performed with 3T MRI in 16 patients with RA and 3 healthy controls. Computer-aided semiautomatic quantifications for synovitis volume and perfusion (maximum enhancement (E) and steepest slope (S)), bone marrow edema pattern (BMEP) volume, signal intensity and perfusion (E and S), and erosion dimension were measured using in-house developed software, and compared with the OMERACT-RAMRIS classification. In addition a semi-quantitative MRI wrist cartilage loss score system was developed. Intra-Class Correlation (ICC) coefficients were calculated to examine intra- and inter-operator reproducibility; Spearman correlation coefficients were calculated between MRI quantifications, RAMRIS, and radiographic Sharp van der Heijde score.
Results:
The intra- and inter-observer ICC were 0.97 and 0.89 for synovitis volume quantification, respectively. The intra- and inter-reader ICC for erosion dimension measurement were 0.91 and 0.93, respectively. The synovitis volume, BMEP volume and signal intensity, and erosion dimension were significantly correlated with the corresponding RAMRIS sub-scores (r from 0.727 to 0.900). Regarding perfusion, no significant correlation was found between synovitis perfusion, BMEP perfusion and RAMRIS scores except for between synovitis E, synovitis volume and synovitis RAMRIS (r = 0.928, p = 0.002; r = 0.798, p = 0.017), and between synovitis and BMEP perfusion E (r = 0.900, p = 0.037). The intra- and inter-reader ICC of the MRI cartilage loss score were 0.98 and 0.99 respectively. The MRI cartilage loss score was significantly correlated with the hand Sharp van der Heijde joint space narrowing score (r = 0.635, p = 0.008).
Conclusion:
Multiple computer-aided and manual quantitative methods can be used to quantify MRI pathologies in RA and showed excellent reliability compared with RAMRIS. The quantitative methods have the potential to be more sensitive for detecting early and subtle changes than conventional scoring systems, and thus may be helpful with early diagnosis and providing critical monitoring after treatment.
Disclosure:
H. Yang,
None;
J. Rivoire,
None;
M. Hoppe,
None;
J. B. Imboden,
None;
T. M. Link,
None;
X. Li,
None.
« Back to 2013 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/computer-aided-and-manual-quantifications-of-mri-synovitis-bone-edema-erosion-and-semi-quantitative-cartilage-loss-in-rheumatoid-arthritis/