Background Untreated gout transitions from an acute intermittent arthritis to a chronic inflammatory arthritis indicating at some point the inflammation associated with gout does not abate; it is unclear when this inflammatory process starts.

Objectives The aim of this study was to determine the percentage of patients with inter-critical gout who have chronic synovial-based inflammation as evidenced by synovial pannus on a contrast-enhanced MRI of their most involved joint and determine if the presence and/or severity correlates with their serum urate levels.

Methods All patients in this prospective trial had inter-critical gout and received a 3T MRI with and without gadolinium of their index joint (i.e. the joint most often involved with acute attacks of gout). Each subject also had a plain radiograph of the index joint as well as a serum urate, highly sensitive (hs)-CRP, and creatinine obtained on the same day. The MRI and radiograph were read by two musculoskeletal radiologists in an independent and blinded fashion. The primary endpoint was to determine the correlation of serum urate levels with the presence and/or severity (previously utilized monoarthritis MRI grading scale of 1-6) of the synovial pannus on the index joint. The MRI and radiographs were also assessed for erosions, intraosseous tophi, soft-tissue tophi, soft tissue swelling, joint effusions, and bone marrow reactive changes (MRI only). Secondary endpoints included the correlation with the presence and/or severity of synovial pannus with hs-CRP and estimated glomerular filtration rate (eGFR). The other MRI and plain radiograph findings were also assessed for correlation with their serum urate level.

Results 74 subjects were screened; 72 completed the protocol. 65/72 (90%) participants were males (50 Caucasian, 10 African-American, 9 Hispanic, 3 other) with an average age and disease duration of 56.4 years (range 28-78) and 10.1 years (range 0.5-37), respectively. 53/72 (74%) index joints were the first metatarsalphalangeal joint; the average number of attacks in the index joint was 11.4 (+/- 13.6 SD) and 21.8 (+/- 25.8 SD) total attacks in any joint. 39 (54.2%) of the patients were on urate lowering therapy; 15 (20.8%) and 7 (9.7%) were taking colchicine or a NSAID daily, respectively. 63/72 (87.5%; 95% CI of 5.9%) of the subjects had synovial pannus on their MRI with good inter-reader agreement between the two radiologists (kappa 0.74). The mean serum urate level was 7.93 (+/- 2.13 SD). There was no correlation with the presence (p=0.33 [t-test]) or severity (Spearman correlation coefficient 0.12 [p=0.34]) of synovial pannus and serum urate levels. There was also no correlation with the presence (p=0.32) or severity (p=0.30) of synovial pannus and hs-CRP or the presence of pannus and eGFR (p=0.62), but there was a correlation with the severity of pannus and eGFR (p=0.02). Finally, the serum urate levels did not correlate with the presence of any of the other findings on MRI or plain radiograph of the index joint.

Conclusions The overwhelming majority of patients with inter-critical gout have evidence of occult and chronic synovial-based inflammation. However, the presence and severity of this chronic synovial-based inflammation does not appear to correlate with serum urate levels.

Acknowledgements Support provided by Takeda Pharmaceuticals USA, Inc.

Disclosure of Interest J. Carter Grant/research support from: Takeda Pharmaceuticals, USA, Inc, Speakers bureau: Takeda Pharmaceuticals, USA, Inc, M. Patelli: None Declared, S. Anderson: None Declared, N. Prakash: None Declared, R. Aydelott: None Declared, E. Rodriguez: None Declared, H. Bateman: None Declared, A. Sterrett: None Declared, J. Valeriano: None Declared, L. Ricca: None Declared