Background: Uric acid is detrimental to the kidneys in animal models. However, its role in human diabetic nephropathy has not been extensively studied. This study evaluated the association between serum uric acid and urinary albumin-to-creatinine ratio (ACR) among patients with type 2 diabetes mellitus in Taiwan.
Methods: A total of 343 patients (144 men and 199 women), aged 62.8 +/- 10.8 years and not using uric acid-lowering agents, diuretics, or alcohol, were recruited. Serum uric acid and urinary ACR were determined. Normoalbuminuria, microalbuminuria, and macroalbuminuria were defined as ACR <30.0, 30.0 to 299.9, and > or =300.0 microg/mg, respectively.
Results: The respective uric acid levels for normoalbuminuria (N= 166), microalbuminuria (N= 130), and macroalbuminuria (N= 47) were 5.2 +/- 1.6 mg/dL, 5.6 +/- 1.9 mg/dL, and 6.7 +/- 2.1 mg/dL (P < 0.001). The mean +/- SD (minimum-maximum) values of uric acid for the first to the fourth quartile were 3.4 +/- 0.6 (1.7-4.2), 4.9 +/- 0.4 (4.3-5.4), 6.0 +/- 0.3 (5.5-6.5), and 8.1 +/- 1.2 (6.6-12.2), respectively. Prevalence of abnormal albuminuria (microalbuminuria plus macroalbuminuria) for the respective quartiles were 38.4%, 51.2%, 50.6%, and 66.3% (P trend <0.01). In men, uric acid correlated positively with triglycerides and natural logarithmic [ln (ACR)] (gamma= 0.168, P < 0.05). In women, uric acid correlated positively with triglycerides, ln (ACR) (gamma= 0.277, P < 0.01) and body mass index (borderline significant P < 0.1), but negatively with calculated creatinine clearance. The standardized regression coefficient for ln (ACR) and the odds ratio for abnormal albuminuria for every 1 mg/dL increment of uric acid after adjusting for calculated creatinine clearance and other confounders were 0.138 (P < 0.05) and 1.183 (1.025-1.364), respectively. The results after excluding 127 cases with a history of hypertension were similar.
Conclusion: Serum uric acid is an independent correlate of urinary ACR in Taiwanese patients with type 2 diabetes mellitus.