Ischemic stroke is composed of subtypes with variable underlying pathogenesis and studies on ischemic stroke as a whole may inadequately evaluate risk factors, being influenced by subtype distribution among studied population. This study aimed to evaluate risk factors associated with individual ischemic stroke subtypes defined by the Trial of ORG10172 in Acute Stroke Treatment. In a case-control study (290 first-ever ischemic stroke cases and 1160 individually matched controls without stroke) nested within Korean male public servants cohort, a range of potential risk factors measured at periodic health surveys prior to the onset of stroke event were examined using conditional logistic regression analysis. Increased risk for large-artery atherosclerosis was associated with hypercholesterolemia (> or = 6.2 mmol/L), hypertension, and smoking. Increased risk for small-artery occlusion was associated with hypertension, hyperglycemia (> or = 7.0 mmol/L), and frequent alcohol intake. No specific risk factor was identified for cardioembolism. For combined ischemic stroke, hypercholesterolemia, hyperglycemia, hypertension, and smoking were associated with the increased risk, but the relative odds were much smaller than those estimated from subtype analysis. Significance of risk factors evaluated for subtypes, rather than ischemic stroke as a whole, should be reflected in preventive efforts against the burden of ischemic stroke.