Our aim was to study whether there is causal association between serum uric acid and metabolic syndrome (MetS). A cross-sectional study was performed, including a total of 27,009 subjects (23,345 subjects having uric acid data) from the Dongfeng-Tongji Cohort study. The MetS was defined by the International Diabetes Foundation criteria of 2005. Association analysis was performed by logistic regression. A genetic risk score was calculated by adding the uric acid increasing alleles in two SNPs (rs11722228 in SLC2A9 and rs2231142 in ABCG2) which were identified from our genome-wide association study on uric acid levels. The causal association was examined by mendelian randomization analysis. Among a middle- and old-age Chinese population, serum uric acid concentrations were strongly associated with the risk of MetS and its several components (P < 0.0001). The effects were stronger in women than in men. Despite the lack of statistical significance, both SNPs exhibited a trend with increased MetS risk (rs11722228, OR = 1.06, 95 % CI 0.99-1.14; rs2231142, OR = 1.02, 95 % CI 0.95-1.10), consistent with their increasing uric acid effects. Each additional uric acid increasing allele in the genetic risk score was associated with 3 % increased MetS risk (OR = 1.03, 95 % CI 0.98-1.09; P = 0.23). Further adjustment for serum uric acid attenuated the trend of individual SNP and genetic risk score with increased MetS risk (all OR < 1.0). These findings suggested that serum uric acid was associated with MetS risk in a middle- and old-age Chinese population. Whether this association was causal remained to be investigated in the future studies.