Predictors of disease activity in gout: a 12-month analysis of the ATTACk (Achieving improvement in the management of crystal-induced arthritis) multicentre cohort study

Objective Gout treatment is largely suboptimal in clinical practice. We aimed to assess the predictors of disease-activity at 12 months in a real-life setting. Methods Consecutive patients referred to Rheumatology Units for suspected acute crystal-induced arthritis were enrolled in a multicentre-cohort study. Only patients with clinical diagnosis of gout were eligible. Disease-activity was evaluated by the Patient Acceptable Symptom State (PASS) on a visual analogue scale (VAS, 0=unsatisfactory, 100=satisfactory) at 0 (T0) and 12 months (T12), and the composite score called Gout Activity Score (GAS) calculated on the number of arthritic attacks (flare count), serum uric acid (sUA), cumulative number of tophi, VAS (T12), PtGA (T12). Multivariate linear regression model was performed to assess predictors of gout disease-activity at T12 with PASS and GAS as outcomes. Results: 201 patients had gout (diagnosis on synovial fluid in 45%, tophi in 26%, mean sUA 7.4±1.9 mg/L, 85% with urate-lowering therapy (ULT) in progress/initiated at T0); mean age 63±13 years, 88% men, median (interquartile range) disease duration 2.9 years (0.7-9.4). Follow-up visits were performed in 113 (56%) patients at T12. Mean PASS observed at T0 and at T12 were 38±27 and 74±23, respectively, whereas GAS at T12 was 10±8. A significant association was observed between the presence of tophi and PASS at T12 (-15.3, 95% CI -25.5, -5.2; p=0.003) and GAS at T12 (+4.0, 95% CI 0.6,7.4; p=0.02), adjusted for age, sex, disease duration, sUA