Serum uric acid does not predict incident metabolic syndrome in a population with high prevalence of obesity

Aim: To evaluate whether uric acid (UA) predicts 4-yr incidence of metabolic syndrome (MetS) in non-diabetic participants of the Strong Heart Study (SHS) cohort. Methods and results: In this population-based prospective study we analyzed 1499 American Indians (890 women), without diabetes or MetS, controlled during the 4th SHS exam and re-examined 4 years later during the 5th SHS exam. Participants were divided into sex-specific tertiles of UA and the first two tertiles (group N) were compared with the third tertile (group H). Body mass index (BMI = 28.3 +/- 7 vs. 31.1 +/- 7 kg/m(2)), fat-free mass (FFM = 52.0 +/- 14 vs. 54.9 +/- 11 kg), waist-to-hip ratio, HOMA-IR (3.66 vs. 4.26), BP and indices of inflammation were significantly higher in group H than in group N (all p < 0.001). Incident MetS at the time of the 5th exam was more frequent in group H than group N (35 vs. 28%, OR 1.44 (95% CI = 1.10-1.91; p < 0.01). This association was still significant (OR = 1.13, p = 0.04) independently of family relatedness, sex, history of hypertension, HOMA-IR, central adiposity and renal function, but disappeared when fat-free mass was included in the model. Conclusions: In the SHS, UA levels are associated to parameters of insulin resistance and to indices of inflammation. UA levels, however, do not predict incident MetS independently of the initial obesity-related increased FFM. (C) 2014 Elsevier B.V. All rights reserved.