BACKGROUND Computed tomography (CT) and magnetic resonance imaging (MRI) are tools of investigation in acute stroke. We wondered if the additional information offered by MRI outweighs the disadvantage of its longer scanning duration for systemic thrombolysis. METHODS Two hundred ninety-four consecutive patients (66 ± 13 years) were subjected to thrombolyis between 1999 and 2004. Inclusion criteria were ischemic infarction, scoring at entry and discharge with the NIH stroke scale and modified Rankin scale, systemic thrombolysis within 3 hours after symptom onset, multimodal MRI or standard CT. Subgroup analysis of 42 patients compared standard CT with CT and CT angiography. RESULTS Patients were similarly affected on admission (P > .1). At discharge, 6 days after stroke onset, the patients investigated with MRI were less impaired than those investigated with standard CT (P < .05). Symptomatic hemorrhage was rare in both groups. Also, patients investigated with CT and CT angiography were less impaired at discharge than those with standard CT (P < .02). A multifactorial regression showed that systolic blood pressure, glucose level and initial neurological impairment determined the neurological outcome at discharge. CONCLUSIONS Systolic blood pressure, glucose level and neurological impairment but not the imaging modality determined the neurological outcome following systemic thrombolysis in the 3-hour window.