Background and Purpose— Tissue factor (TF) expression is increased in inflammatory atherosclerotic plaques and has been related to their thrombogenicity. Blood-borne TF has been also demonstrated to contribute to thrombogenesis. However, few studies have evaluated the association of circulating levels of TF with stroke. We investigated the association of baseline circulating levels of TF with stroke events occurred in the European Prospective Investigation into Cancer and Nutrition-Italy cohort. Methods— Using a nested case–cohort design, a center-stratified random sample of 839 subjects (66% women; age range, 35–71 years) was selected as subcohort and compared with 292 strokes in a mean follow-up of 9 years. Blood samples were collected at baseline in citrate, plasma was stored in liquid nitrogen and TF was measured by ELISA (IMUBIND, TF ELISA, Instrumentation Laboratory, Milan, Italy). The odd ratios and 95% confidence intervals, adjusted by relevant confounders (covariates of TF) and stratified by center, were estimated by a Cox regression model using Prentice method. Results— Individuals in the highest compared with the lowest quartile of TF plasma levels had significantly increased risk of stroke (odds ratio _{IVvsI quartile} , 2.01; 95% confidence interval, 1.25–3.23). The association was independent from several potential confounders (odds ratio _{IVvsI quartile} , 1.91; 95% confidence interval, 1.15–3.19). No differences were observed between men and women. The increase in risk was restricted to ischemic strokes (odds ratio _{IVvsI quartile} , 2.13; 95% confidence interval, 1.10–4.12; fully adjusted model), whereas high levels of TF were not associated with the risk of hemorrhagic stroke (odds ratio _{IVvsI quartile} , 1.12; 95% confidence interval, 0.49–2.55; fully adjusted model). Conclusions— Our data provide evidence that elevated levels of circulating TF are potential risk factors for ischemic strokes.