INTRODUCTION: Since 2003, intravenous thrombolysis with rt-PA for stroke victims has been largely developed in the military hospital of Toulon. We report the results of our practice and compare them with the literature. We also sought to identify predictive factors of favorable outcome after thrombolysis. METHODS: All patients treated with rt-PA for a stroke in the carotid territory between September 2003 and June 2009 were prospectively included. Disability was assessed at 3 months with the modified Rankin Scale (m-RS); outcome was considered unfavorable if m-RS score was above 2. Multivariate analysis was then performed to identify parameters correlating with poor and favorable outcome at 3 months follow-up. RESULTS: One hundred and one patients were included in this study (mean initial National Institute of Health Stroke Scale [NIHSS]: 15.2). 53.4% had a Rankin score higher than 2 at 3 months follow-up. The absence of diabetes mellitus, low NIHSS score on admission, short time from stroke onset to treatment, and prior statin use were identified as independent predictive factors of favorable functional outcome. CONCLUSIONS: After 6 years of activity, our stroke unit has results that appear similar to those of the French and international trials in terms of safety and efficacy. Efficacy of rt-PA in our series is poor for strokes caused by large-vessel atherothrombotic changes and cervical artery dissection due to high incidence of internal carotid thrombosis in these cases. Our studies also suggest that prior statin use may be an independent predictive factor of favorable outcome after thrombolysis.