Context: The risk associated with serum uric acid (SUA) levels within the normal range is unknown, especially among lean and apparently healthy adults.
Objective: Evaluating whether high-normal SUA levels, 6.8 mg/dL and below, are associated with an increased diabetes risk, compared with low-normal SUA.
Design and setting: This was a cohort study with 10 years of followup involving all clinics of the largest nationally distributed Health Maintenance Organization in Israel.
Participants: Participants included 469,947 examinees, 40-70 years old at baseline, who had their SUA measured during 2002. We excluded examinees who had hyperuricemia (SUA > 6.8 mg/dL), impaired fasting glucose, overweight or obesity and chronic cardiovascular or renal disorders. The final cohort was composed of 30 302 participants.
Interventions: Participants were followed up to a new diagnosis of diabetes during the study period.
Main outcome measures: Odds ratio of developing diabetes among participants with high-normal baseline SUA were compared with low-normal (2 ≤ uric acid < 3 and 3 ≤ uric acid < 4 in women and men, respectively).
Results: In a logistic regression model adjusted for age, body mass index, socioeconomic status, smoking, baseline estimated glomerular filtration rate, and baseline glucose, SUA levels of 4-5 mg/dL for women were associated with 61% increased risk for incident diabetes (95% confidence interval, 1.1-2.3). At the highest normal levels for women (SUA, 5-6 mg/dL) the odds ratio was 2.7 (1.8-4.0), whereas men had comparable diabetes risk at values of 6-6.8 mg/dL (hazard ratio, 1.35; 95% confidence interval, 0.9-2.1).
Conclusions: SUA levels within the normal range are associated with an increased risk for new-onset diabetes among healthy lean women when compared with those with low-normal values.