To develop a simple, relatively noninvasive small-animal model of reversible regional cerebral ischemia, we tested various methods of inducing infarction in the territory of the right middle cerebral artery (MCA) by extracranjal vascular occlusion in rats. In preliminary studies, 60 rats were anesthetized with ketamine and different combinations of vessels were occluded; blood pressure and arterial blood gases were monitored. Neurologic deficit, mortality rate, gross pathology, and in some instances, electroencephalogram and histochemical staining results were evaluated in all surviving rats. The principal procedure consisted of introducing a 4-0 nylon intraluminal suture into the cervical internal carotid irtery (ICA) and advancing it intracranially to block blood flow into the MCA; collateral blood flow was reduced by interrupting all branches of the external carotid artery (EC A) and all extracranial branches of the ICA. In some groups of rats, bilateral vertebral or contralateral carotid artery occlusion was also performed. India ink perfusion studies in 20 rats documented blockage of MCA blood flow in 14 rats subjected to permanent occlusion and the restoration of blood flow to the MCA territory in six rats after withdrawal of the suture from the ICA. The best method of MCA occlusion was then selected for further confirmatory studies, including histologic examination, in five additional groups of rats anesthetized with halothane. Seven of eight rats that underwent permanent occlusion of the MCA had resolving moderately severe neurologic deficits (Grade 2 of 4) and unilateral infarcts averaging 37.6±5.5% of the coronal sectional area at 72 hours after the onset of occlusion. Five rats underwent the same procedure after bilateral vertebral artery occlusion was performed to reduce collateral blood flow. Only two of these five rats survived 72 hours; the neurologic deficits progressed from Grade 2.5 to 3, and the infarcts were larger than after MCA occlusion alone. In two groups of rats, the suture was withdrawn from the ICA to permit reperfusion after 2 or 4 hours of ischemia. Five of 10 rats subjected to 4-hour temporary MCA occlusion and one of six rats subjected to 2-hour temporary MCA occlusion did not survive 72 hours after the onset of occlusion. Infarct areas in surviving rats after 2-hour temporary MCA occlusion were 15.7% smaller than after permanent MCA occlusion, but the neurologic deficit was not significantly reduced by reperfusion. Fatal intracranial hemorrhage occurred in only two of 71 rats after occlusion of the MCA with an intraluminal suture. The results in the six sham-occluded rats showed that occlusion of the extracranial carotid branches, dissection of the cervical ICA, and placement of an intraluminal suture in the ECA did not produce stroke. This model provides a reliable method for studying reversible regional ischemia in rats without craniectomy. (Stroke 1989;20:84-91) T he pathophysiology of cerebral ischemia has been studied extensively in rats with various methods, including multi...