BACKGROUND Perfusion computed tomography (PCT) is part of acute stroke protocol in many hospitals; however, its clinical utility is still being disputed. Beyond its use in core and penumbra estimation, there is also a question about PCT role in stroke mimics diagnosis. Case series or small, retrospective studies showed equivocal results. This is the first published prospective, comparative study on PCT in differentiating stroke and seizure in acute setting. METHODS Patients with acute focal neurologic deficits and without acute ischemic lesions on routine CT underwent PCT and electroencephalography (EEG) within 12 hours after symptom onset. Perfusion parameters were set up as asymmetry indices for corresponding regions of brain hemispheres. EEG findings were assigned to 1 of 5 classes. Neurologic examination was performed using the National Institutes of Health Stroke Scale (NIHSS). Follow-up noncontrast computed tomography was performed on the third day after symptom onset. If no CT changes appeared, magnetic resonance diffusion-weighted imaging was conducted. RESULTS Final diagnosis was hemispheric ischemic stroke in 17 patients and focal neurologic deficits in the course of seizures (post- and intraictally) in 12 patients. Those groups were significantly different only in one single PCT parameter-time to peak (TTP)-in the lateral part of the middle cerebral artery territory. Analyzed groups were not significantly different in the NIHSS scores and the EEG evaluation. CONCLUSIONS TTP may stay relatively when seizure is a cause of focal neurologic deficits, but not stroke. Further, large, prospective studies are necessary to verify the results.