Repeat studies in animal models of acute focal ischemia can be compared to incidental studies in the course of ischemic stroke in order to shed light on the development of changes causing ischemic infarcts or recovery of critically perfused tissue. Positron emission tomography (PET) studies of regional cerebral blood flow, cerebral metabolic rate for oxygen, oxygen extraction fraction (OEF), cerebral metabolic rate of glucose and flumazenil (FMZ) binding in the cat middle cerebral artery occlusion (MCAO) model and in patients with acute ischemic hemispheric stroke were reviewed. After permanent MCAO, the development of ‘misery-perfused’ penumbral tissue and its centrifugal conversion into necrosis could be demonstrated, resembling focal pathophysiological changes in patients with ischemic attacks. In the experimental model and in vascular insults in humans, a chance of recovery existed if collateral perfusion developed spontaneously within the first hours. In transient MCAO, reperfusion was only effective in preventing infarction when it was initiated as long as misery perfusion persisted; in these cases tissue was salvaged and large infarcts did not develop. In the other instances when oxygen metabolism broke down, and an increased OEF was no longer seen, reperfusion even at levels above preocclusion had no effect, and large space-occupying infarcts developed. These experimental findings are comparable to the variable outcome after thrombolytic therapy; if reperfusion is achieved within the therapeutic window of tissue viability, large infarcts are prevented and complete or partial recovery can be achieved. In the experimental model of focal ischemia and in human stroke, FMZ can be utilized as a marker of neuronal integrity. If FMZ binding in the cortex is decreased below 4 times the mean value of white matter in the acute stage, permanent infarcts were observed on late CT/MRI; this irreversible damage could not be prevented by thrombolytic therapy. These results demonstrated that PET studies in suitable ischemia models in cats can help to explain various courses and diverging outcomes of acute ischemic stroke. Comparable findings from experimental ischemia and human stroke may affect the selection of appropriate therapeutic strategies.