Serum Uric Acid and Type 2 Diabetes

Abstract

Objectives: In recent years, serum uric acid has emerged as a possible risk factor for type 2 diabetes mellitus (T2DM). It remains unclear if this is independent of other well-known risk factors related to the metabolic syndrome. This retrospective epidemiologic study attempts to clarify the relationship between uric acid and T2DM, as well as to assess uric acid as a predictor for future diabetes development. Methods: Data was collected by the NIDDK biennial study from Pima Indians in Arizona over several decades. A cross sectional analysis using multivariate logistic regression and a survival analysis using a Cox proportional hazards model were created. Sex and body mass index (BMI) were hypothesized to create significant interactions with other variables. Interactions were confirmed by log likelihood tests, so the data was analyzed stratifying by sex. An interaction term between body mass index and uric acid was also included in analyses performed in women, as it was found to be significant in women only. Results: The cross sectional analysis showed that men with diabetes are significantly more likely to be older (OR=1.033, p<0.0001), have a higher BMI (OR=1.117, p<0.0001), mean arterial pressure (MAP) (OR=1.020, p=0.0024), cholesterol (OR=1.003, p=0.003), and lower uric acid (OR=0.625, p<0.0001) than men without diabetes. Uric acid levels did not correlate with diabetes status in women, but an interaction between uric acid and BMI was significant (p=0.0094). A goodness of fit test of the models comparing predicted to observed outcomes were significant with an R-squared value of almost 0.90 in both sexes. The survival analysis in women demonstrated that BMI (p=<0.0002) and uric acid (p=0.0209) both confer risk for diabetes development, and a significant interaction between BMI and uric acid exists with a negative parameter estimate. A nested analysis of the effect of uric acid assessed in BMI quartiles demonstrated an increased risk in normal to moderately overweight individuals, and a hazard ratio under 1 in more obese individuals. Results in men demonstrated no significance of uric acid (p=0.6571). Conclusion: The relationship between serum uric acid and diabetes varies significantly by sex, and BMI appears to have a confounding relationship with uric acid, especially in women. Uric acid is lower in men with current diabetes, confirming previous studies, which may be due to renal hyperfiltration or induction of uric acid as an antioxidant response to diabetes. In women, elevated uric acid confers higher risk of future development of T2DM. Why this was not shown in men is unclear. Utilizing uric acid in clinical practice as a screening tool is limited by interactions between uric acid and other metabolic risk factors, specifically BMI, as well as variations influenced by diet and renal function.