Hyperuricemia is common among patients with hypertension and metabolic syndrome and therefore may be a cause of or result from these comorbid conditions. Few studies, however, have examined the relationship between the presence-absence of hyperuricemia and changes in the estimated glomerular filtration rate (eGFR) using the large cohort of the general population. We examined subjects who participated in two screenings, in 1993 and 2003, in Okinawa, Japan, yielding data on serum creatinine and uric acid levels (N=16,630). eGFR (ml min(-1) per 1.73 m(2)) was calculated using the formula used by the Japanese Society of Nephrology. In both sexes, a uric acid (UA) level >7.0 mg dl(-1) was defined as hyperuricemia (H), and a UA level below that threshold was classified as normouricemia (N). Based on the absence or presence of hyperuricemia in both the 1993 screening and the 2003 screening, we categorized patients into four groups: group 1, N/N; group 2, H/N; group 3, N/H; and group 4, H/H. Multiple regression analysis was performed to estimate the independent effects of several variables on the decline in eGFR. In all groups, an increase in UA from 1993 to 2003 (ΔUA) was a strong independent risk factor for a decline in eGFR than that of the baseline levels of UA, the presence of hypertension, or diabetes. The estimated decline in eGFR per 1 mg dl(-1) increase in UA was 4.19, 1.91, 2.36 and 2.01 ml min(-1) per 1.73 m(2) in groups 1, 2, 3 and 4, respectively. The results suggest that UA has a role in chronic kidney disease (CKD) progression. We have no information on medications used, such as xanthine oxidase, uricosuric drugs and hypotensives; therefore, the impact of hyperuricemia might be underestimated in our analysis. The results suggest that maintaining a normal range of UA is important to maintain eGFR decline in a normal range.