BACKGROUND Some observational studies have suggested that serum uric acid (SUA) levels are one of the determinants of the metabolic syndrome (MetS). However, previous studies reported combined results for men and women after adjusting for sex and few studies take body composition into consideration. Therefore, we performed this sex-specific longitudinal study to investigate how baseline SUA levels influence incident MetS, including body composition as an adjusting factor in a large number of subjects. METHODS A total of 14,442 participants (8715 men and 5727 women) participating in a medical health check-up program without diagnosed MetS at baseline were enrolled. Separate analyses were performed for men and women including body composition as a confounding factor. Cox proportional hazards models were used to quantify independent associations between SUA levels and incident MetS. RESULTS During 63,940person-years of follow-up, there were 4215 (2974 men, 1241 women) incident cases of MetS between 2006 and 2012. After adjustments for age, systolic BP, diastolic BP, BMI, eGFR, smoking status, TG, LDL-C, HDL-C, fasting glucose, and proportion of fat-free mass (100-fat mass, %), the hazard ratios (HR) [95% confidence interval (CI)] for incident MetS comparing the second, the third, and the fourth quartiles to the first quartile of SUA levels were 0.862 (0.770-0.965), 1.102 (0.991-1.225), and 1.246 (1.121-1.385) in men (p for trend<0.001), and 1.045 (0.862-1.266), 1.251 (1.050-1.490), and 1.321 (1.109-1.574) in women (p for trend<0.001), respectively. As a continuous variable, in fully-adjusted models, the HRs (95% CI) for incident MetS associated with each increase of 1mg/dl of SUA levels were 1.094 (1.060-1.130) in men (p<0.001) and 1.148 (1.072-1.228) in women (p<0.001), respectively. CONCLUSION We demonstrated that SUA levels are strong and independent predictors of MetS. This relationship remained significant after full adjustments for multiple associated confounders including body composition in both men and women.