High serum uric acid (UA) levels are associated with numerous vascular risk factors, and vascular disease, that predispose patients to cognitive impairment, yet UA is also a major natural antioxidant and higher levels have been linked to slower progression of several neurodegenerative disease. In-order to test the association between UA and subsequent cognitive performance among patients that carry a high vascular burden, UA levels were determined by calorimetric enzymatic tests in a sub-cohort of patients with chronic cardiovascular disease who previously participating in a secondary prevention trial. After an average of 9.8±1.7 years, we assessed cognitive performance (Neurotrax Computerized Cognitive Battery) as well as cerebrovascular reactivity (CVR) and common carotid intima-media thickness (IMT). Among 446 men (mean age 62.3±6.4 yrs) mean UA levels were 5.8±1.1 mg/dL. Adjusted linear regression models revealed that low UA levels (bottom quintile) were associated with poorer cognitive performance. Adjusted differences between the bottom quintile and grouped top UA quintiles were (B coefficient±SE) -4.23±1.28 for global cognitive scores (p = 0.001), -4.69±1.81 for memory scores (p = 0.010), -3.32±1.43 for executive scores (p = 0.020) and -3.43±1.97 for visual spatial scores (p = 0.082). Significant difference was also found for attention scores (p = 0.015). Additional adjustment for impaired CVR and high common carotid IMT slightly attenuated the relationship. Stronger UA effect on cognitive performance was found for older (age>65) patients with significant age interaction for global cognitive score (p = 0.016) and for executive (p = 0.018) and attention domains (p<0.001). In conclusion, we demonstrate that low UA levels in patients with preexisting cardiovascular disease are associated with poorer cognitive function a decade later. These findings lend support to the hypothesis that oxidative stress may be involved in the pathogenesis of age-associated cognitive impairment.