To examine predictive factors for abdominal obesity or metabolic syndrome, we investigated the association of plasma fatty acid composition, estimated desaturase activity, and nutrient intakes, with abdominal obesity or metabolic syndrome in Japanese males. Clinical characteristics, the fatty acid composition of plasma cholesteryl esters, and energy and nutrient intakes were analyzed in 3 groups: metabolic syndrome (MS, n=24), abdominal obesity (OB, n=43), and control (n=27). The estimated desaturase activities were calculated by the ratio of 16:1n-7/16:0, 18:3n-6/18:2n-6, and 20:4n-6/20:3n-6 in plasma cholesteryl esters as surrogates of the measure of the delta 9, delta 6, delta 5 desaturase (D9-16D, D6D and D5D) activities, respectively. Plasma fatty acid composition did not differ significantly between the OB group and the control group. The MS group had higher levels of palmitoleic, oleic, and γ-linolenic acids, but a lower level of linoleic acid than the control. Stronger D6D activity and weaker D5D activity were observed in the OB group. A higher level of D9-16D activity as well as a higher level of D6D activity and a lower level of D5D activity was observed in the MS group. A logistic regression analysis showed that the low D5D activity and high D9-16D activity were predictive of the development of abdominal obesity from controls (odds ratio=0.39, p<0.05) and metabolic syndrome from abdominal obesity (odds ratio=2.44, p<0.05), respectively. In the multiple linear regression analysis, D5D activity positively correlated with the intake of eicosapentaenoic acid (EPA). In conclusion, the estimated D5D activity was a predictive factor for abdominal obesity and the estimated D9-16D activity was a predictive factor for developing metabolic syndrome from abdominal obesity in Japanese male subjects. Dietary intake of EPA would play an important role in preventing abdominal obesity and the development of metabolic syndrome.