Objectives : This study was conducted to investigate the associations of non alcoholic fatty liver with metabolic syndrome and the serum carotenoids. Methods : This study was conducted in a general hospital in South Korea from November, 2004 to August, 2005. The study subjects were 350 sampled persons who were aged from 40 years and older (males : 180, females : 170). They were grouped into the normal, mild and severe groups according to fat accumulation in their livers, as determined by ultrasonography. We analyzed the association between non alcoholic fatty liver and metabolic syndrome by multiple logistic regression analysis and we analyzed the association between non alcoholic fatty liver and the serum carotenoids by a general linear model(ANCOVA). Results : After adjustment for the effect of potential covariates, the prevalence of metabolic syndrome was associated with fat accumulation in the liver (p trend <0.001). If the odds ratio of normal group is 1.00, then that of the mild group is 2.80 (95% C.I=1.17-6.71) and that of the severe group is 7.29 (95% C.I=2.76-19.30). The prevalence of metabolic alterations fitting the criteria of metabolic syndrome, according to the class of fat accumulation in the liver, was significantly increased, except for criteria of high blood pressure, a large waist circumference and low HDL (high density lipoprotein) cholesterol level (p trend <0.001). The level of serum ${eta}$-carotene was decreased according to the class of fat accumulation in the liver (p trend=0.036), but the levels of serum ${alpha}$-carotene, lycopene, ${eta}$-cryptoxanthin and lutein were not decreased. Conclusions : This study shows that non alcoholic fatty liver was associated with metabolic syndrome and with the serum ${eta}$-carotene level.