The aim of this study was to examine whether the combination of abdominal obesity and high serum levels of high-sensitivity C-reactive protein (hsCRP) improves the prediction of new-onset hypertension in the general Japanese population. Participants in the Tanno-Sobetsu study, a prospective cohort study, were enrolled. Of 1516 subjects aged 30 years or older in 2002, those with hypertension or abnormal hsCRP levels were excluded, and the remaining 705 subjects were included in the present analyses. Abdominal obesity (AO) and high hsCRP levels were defined by the Japanese criteria of waist circumference and median hsCRP values, respectively. Subjects were followed up on for a maximum of 4.5 years, and the hazard ratio (HR) for new-onset hypertension was calculated using the Cox proportional hazard model. The HR for new-onset hypertension after adjustment for age, sex, fasting plasma glucose, alcohol intake, smoking, treatment for dyslipidemia and high normal blood pressure at baseline was significantly higher in the group with AO and high hsCRP (HR=1.44, 95% CI: 1.00-2.07) compared with the reference group, a group with no AO and low hsCRP; AO alone or high hsCRP alone was not associated with a significant increase in the HR. Similar trends for the increase in the HR by AO and high hsCRP were observed in separate analyses of men and women, although the differences did not reach statistical significance. Co-presence of AO and a high level of hsCRP is associated with a high risk for new-onset hypertension in the general population.