Uric acid and cardiovascular disease risk reclassification: Findings from NHANES III

Abstract

Background: The studied associations between serum uric acid (sUA) and cardiovascular disease (CVD) events have been controversial. We sought to evaluate the association between sUA and CVD mortality, including its ability to reclassify risk in a multiethnic nationally representative population free of clinical CVD and diabetes at baseline. Methods: The study cohort included 11,009 adults enrolled as a part of the National Health and Nutrition Examination Survey (NHANES) III. Multivariate Cox proportional hazard analysis was performed to evaluate sUA as a predictor of CVD and coronary heart disease (CHD) mortality. Discriminative and recalibrative properties of sUA for CHD deaths were also assessed over traditional CVD risk factors. Net reclassification index (NRI) was calculated by comparing regression models incorporating traditional CVD risk factors with and without sUA. Results: sUA was not predictive of either CVD mortality [model 4: hazards ratio (HR) 1.06, 95% confidence interval (CI) 0.96-1.16, p=0.27] or CHD mortality (model 4: HR 1.06, 95% CI 0.94-1.19, p=0.32). Addition of sUA to traditional CVD risk factors resulted in no significant increment in c-statistic, receiver-operating characteristics-area under curve, absolute NRI (0.5%, 95% CI -1.9 to 2.9%, p=0.68), or intermediate NRI (2.5%, 95% CI -1.6 to 6.6%, p=0.24) for prediction of hard CHD deaths. Conclusions: sUA was not an independent predictor of both CVD and CHD mortality. Ethnicity did not influence the association of sUA with CVD mortality. Furthermore, sUA did not add to risk assessment beyond traditional CVD risk factors.