Middle cerebral artery occlusion (MCAO) models have become well established as the most suitable way to simulate stroke in experimental studies. The high variability in the size of the resulting infarct due to filament composition, rodent strain and vessel anatomy makes the setup of such models very complex. Beside controllable variables of homeostasis, the choice of anesthetics and the grade of ischemia and reperfusion played a major role for extent of neurological injury. Transient MCAO was induced during either isoflurane or ketamine/xylazine (ket/xyl) anesthesia with simultaneously measurement of cerebral blood flow (CBF) in 60 male Wistar rats (380-420 g). Neurological injury was quantified after 24 h. Isoflurane compared with ket/xyl improved mortality 24 h after MCAO (10 vs. 50 %, p = 0.037) and predominantly led to striatal infarcts (78 vs. 18 %, p = 0.009) without involvement of the neocortex and medial caudoputamen. Independent of anesthesia type, cortical infarcts could be predicted with a sensitivity of 67 % and a specificity of 100 % if CBF did not exceed 35 % of the baseline value during ischemia. In all other cases, cortical infarcts developed if the reperfusion values remained below 50 %. Hyperemia during reperfusion significantly increased infarct and edema volumes. The cause of frequent striatal infarcts after isoflurane anesthesia might be attributed to an improved CBF during ischemia (46 ± 15 % vs. 35 ± 19 %, p = 0.04). S-100β release, edema volume and upregulation of IL-6 and IL-1β expression were impeded by isoflurane. Thus, anesthetic management as well as the grade of ischemia and reperfusion after transient MCAO demonstrated important effects on neurological injury.