The polymorphism of the angiotensin-converting enzyme gene is considered to be associated with increased risk for stroke, but there is a diversity in the results obtained. The genetic involvement of the renin-angiotensin system in stroke also remains unclear. To predict the genetic risk of lacunar infarction, we conducted an association study in an Ohasama population, which is the cohort in a rural region of northern Japan. A total of 134 subjects without major neurological, cardiovascular, or metabolic disorders were recruited. Using brain magnetic resonance imaging, the number of lacunae in each of four brain regions were calculated, and periventricular hyperintensity was classified into five grades. We used the following four candidate gene polymorphisms: angiotensin converting enzyme (ACE)/Insertion(I)-Deletion(D), angiotensinogen (AGT)/M235T, angiotensin II type 1 receptor (AT1)/ A1166C, type 2 receptor (AT2)/C3123A, to examine the association between polymorphisms and the severity of lacunar infarction. AGT/M235T was significantly associated with the number of lacunae in the brain stem, the basal ganglia (P < .05), and whole brain (P < .005) regions. The AT1 polymorphism was also significantly associated with the number of lacunae in the basal ganglia and whole brain regions (P < .05), and with periventricular hyperintensity grade (P < .005) in the younger population. However, ACE and AT2 polymorphisms failed to show an association with either the number of lacunae or the PVH grade. We concluded that AGT and AT1 polymorphisms are independent genetic risk factors for lacunar infarction.