Background Few imaging studies have investigated cartilage damage in gout. Magnetic Resonance Imaging (MRI) can produce high quality images of cartilage damage in this disease and also reveal other features of gouty arthropathy.

Objectives To develop and validate a system for quantifying cartilage damage in gout. To test this system for scoring reliability and validity. To explore associations between cartilage damage and other disease manifestations in gout.

Methods 3T-MRI scans of the wrist were obtained in 40 gout patients using a dedicated wrist coil and cartilage-specific sequences (1). MRI cartilage damage was quantified using an adaptation of the radiographic Sharp van der Heidje (SvdH) score. Two readers scored cartilage loss at 7 wrist joints (distal radio-ulnar, radioscaphoid, radiolunate, triquetrum-hamate, lunate-capitate, scaphoid-capitate, and scaphotrapezoid); 0 (normal), 1 (partial narrowing), 2 (complete narrowing) and concomitant osteoarthritis was recorded. Bone erosion and bone marrow edema (BME) were scored using the RAMRIS system and tophi were assessed. The SvdH score was used to assess radiographic erosion and jsn in the same group. Reliability of the MRI cartilage score was determined and correlations with the radiographic jsn score were investigated as were the associations between MRI cartilage damage, bone erosion and BME.

Results A total of 280 joints were scored for MRI cartilage damage (N=40). The gout MRI cartilage score was highly correlated with the total SvdH score and the jsn component (R =0.8 and 0.71 respectively, p<0.001). Reliability for scoring MRI cartilage damage was high (intraobserver, interobserver ICCs =0.87 [0.57-0.97], 0.64 [0.41–0.79] respectively). Reliability and validity of data were retested in the groups given IV contrast, N=28 and not given contrast (impaired renal function), N=12. Reliability was lower in the contrast vs the non-contrast group (interobserver ICCs =0.40 [0.03 - 0.67] vs 0.82 [0.49 - 0.95] respectively). The correlation between the MRI cartilage and SvdH jsn scores was also lower in the contrast group vs the non-contrast group (R =0.66 vs R =0.86 respectively). In the total group, MRI cartilage damage was identified by one or both readers in 40/280 (14%) of joints. Cartilage lesions were focal (Grade 1) at 82% of these sites. The following joints were most frequently involved: distal radioulnar (33% of patients), radiolunate (21%) and radioscaphoid (20%). We repeated the analysis using data for joints where both readers agreed (as to presence or absence of MRI cartilage damage) and this revealed 14 of 229 (6.1%) of sites to be abnormal. Of these, 12 sites (85.7%) had evidence of concomitant OA. Cartilage scores correlated with MRI bone erosion (R =0.57, p<0.001), and tophus size (R =0.52, p=0.001), but not BME scores.

Conclusions MRI can be used to investigate cartilage in gout. The use of non-contrast enhanced sequences leads to optimal reliability and validity of the MRI cartilage score. Cartilage damage was relatively uncommon in gout, focal and frequently associated with OA. Cartilage scores correlated with scores for bone erosions and tophus size but not BME. These findings emphasize the unique pathophysiology of gout.

References

1. McQueen F, Clarke A, McHaffie A, et al. ARD 2010;69:1971-75.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.2275