Background development of tophi has been associated with increasing level of serum urate (Gutman AB, A&R 1973), whereas other factors have not been clearly stated, such as gender, comorbidities or concomitant medications. Current EULAR (Zhang W, ARD 2006) and ACR (Khanna D, AC&R 2012) recommendations and guidelines support initiation of urate-lowering medications depending on renal function, presence of tophi, or number of episodes of acute inflammation, but not to risk factors for developing severe gout, including serum urate levels.

Objectives to explore risk factors associated with the development of tophi in patients with gout.

Methods analysis of a prospective cohort of patients with gout that systematically captures general data, comorbid conditions and concomitant medications, including urate-lowering medications and doses, along with clinical data including time from onset of symptoms, joint involvement and severity. Survival analysis was with time from onset of symptoms to first clinical evaluation as time exposed to disease was used to select variables (Kaplan-Meier, log-rank test). Continuos variables were categorized into quartiles for analysis. Variables with a plausible association (p<0.2) were selected for analysis with stepwise multivariant Cox regression. The cohort study was approved by the Ethics and Clinical Investigation Board.

Results data from 1,005 patients until Dec 2013 were available fro analysis; missing data were <1,5% at most for any variable; there were 93% males, 32% with subcutaneous tophi, 39% had received urate-lowering medications, 36% showed >4 joints ever involved. Comorbid conditions and medications included 30% on diuretics, 47% hypertension, 20% diabetes, 46% hyperlipidemia, 30% vascular event. Time from onset of symptoms to first rheumatology evaluation was 6.8±7.1 years (interquartile range 2-10).

Variables associated with the presence of tophi in univariant analysis were: urate level, gender, hypertension, diuretics, diabetes, vascular event and previous treatment with urate-lowering medications. Multivariant analysis showed that the presence of tophi was independently associated with the following variables (risk ratio, 95%CI): female gender (1.85, 1.17-2.92), diuretic use (1.65, 1.3-2.13), and baseline serum urate (1.08, 1.01- 1.16, as per 50 mcmol/l increment). Previous treatment with urate-lowering medications was associated with lower risk of developing tophi (0.61, 0.48-0.77).

Conclusions in this cohort, baseline serum urate level was not the unique factor associated to increased risk of tophi. Female gender (might be associated to increased prevalence of osteoarthritis) and concomitant use of diuretics (might be associated to greater persistence in the highest range of serum urate level) are also associated with increased risk, whereas previous urate-lowering therapy (although mostly not to target) is associated to reduced risk. This profile of gout patients may also deserve early urate-lowering treatment.

Disclosure of Interest F. Perez-Ruiz Consultant for: Astra-Zeneca, Menarini, Metabolex, Novartis, Pfizer, SOBI, Speakers bureau: Menarini, E. Castillo: None declared, S. Chinchilla: None declared, A. Herrero-Beites: None declared

DOI 10.1136/annrheumdis-2014-eular.5335