Studies have correlated elevated plasma factor VIII (FVIII) with thrombosis; however, it is unclear whether elevated FVIII is a proinflammatory biomarker, causative agent, or both. We raised FVIII levels in mice and measured the time to vessel occlusion (TTO) after ferric chloride-induced injury. Compared with control (saline-infused) mice, elevated FVIII had no effect after longer (3-minute) carotid artery injury, but it shortened the TTO after shorter (2-minute) injury (P < .008). After injury, circulating thrombin-antithrombin (TAT) complexes were lower after short versus long injury (P < .04), suggesting short treatment produced less coagulation activation. TAT levels in FVIII-infused mice were higher than in controls after short, but not longer, injury. Accordingly, elevated FVIII had no effect on in vitro thrombin generation or platelet aggregation triggered by high tissue factor, but it increased thrombin generation rate and peak (2.4- and 1.5-fold, respectively), and it accelerated platelet aggregation (up to 1.6-fold) when initiated by low tissue factor. Compared with control mice, elevated FVIII stabilized thrombi (fewer emboli) after short injury, but it had no effect after longer injury. TTO and emboli correlated with TATs. These results demonstrate dependence of FVIII activity on extent of vascular injury. We propose elevated plasma FVIII is an etiologic, prothrombotic agent after moderate but not extensive vascular damage.