Abstract

OBJECTIVES: Historically, the nature of association between chronic kidney disease (CKD) and gouty arthritis has been unclear. The goal of the present study was to test the hypothesis that CKD is an independent risk factor for developing incident gout. DESIGN: Patients were from the original Framingham Heart Study cohort. Using Cox proportional hazard models we estimated the HR of CKD to incident gout among men and women separately after adjusting for age, alcohol consumption, smoking, hypertension, diabetes and body mass index. SETTINGS: Patients were all from Framingham, Massachusetts, USA. PARTICIPANTS: Excluding patients who had CKD in the first visit from this study, 2159 men and 2558 women were selected covering a 54-year period (1948-2002). RESULTS: There were 371 incident cases (231 men and 140 women) of gout over the follow-up of 140,421 person-years. Incidence rates of gout per 1000 person-years for participants with and without CKD were 6.82 (95% CI 5.10 to 9.10) and 2.43 (2.18 to 2.71), respectively. In multivariable Cox models, CKD was associated with gout, with a HR of 1.88 (1.13 to 3.13) among men and 2.31 (1.25 to 4.24) among women. Additional analyses using alternate definitions for CKD and cross-sectional study did not change the results. Sensitivity analysis suggested that the observed findings might be an underestimate of the true relative risk. CONCLUSIONS: The present study provides epidemiological evidence to support the notion that CKD is a risk factor for gout.