The prediction of outcome after ischemic stroke (IS) is currently based on indirect data from clinical and radiological evaluation. We evaluated the usefulness of serum Tau protein as possible prognostic markers for IS. Fifty-six patients with computed tomography-confirmed IS were enrolled. Blood samples were obtained on days 1, 3, 5, and 10 after stroke onset. Tau and S100BB serum levels were measured by commercially available enzyme-linked immunosorbent assay. Neurological deficits were quantified by the National Institute of Health Stroke Scale on days 1, 3, 5, and 10 of stroke. Functional disability was rated with the Barthel Index and Rankin Scale on days 1, 3, 5, and 10 and additionally 3 months after the stroke. Computed tomography scan was performed to calculate infarct volume on admission to hospital and on day 10 from the diagnosis of IS onset. Tau protein was detected in the serum of 47.8% patients with IS. Patients in whom Tau protein was detected in serum, when compared with patients without Tau protein, developed more severe neurological deficits, had worse functional status measured in the early and late phase of IS, and were found to have larger volume of infarction. However, Tau protein concentrations measured within the early phase of IS did not correlate with degrees of neurological deficit and disability in the early phase and also after 3 months of IS. Detection of Tau protein in the serum of patients with IS but not its concentration can be considered as a bad prognostic factor for the clinical outcome in early and late phase of IS.