Intracranial stenoses cause 5-10% of all strokes and are increasingly detected by means of modern imaging methods. The stroke danger of high-grade symptomatic stenoses is relatively high, with an annual risk of approximately 10% under medical treatment. Coumadin increases the risk of hemorrhage, and after risk/benefit considerations, antiplatelets should be preferred for antithrombotic therapy. Despite optimized medical treatment, a small group of patients with recurrent symptoms or symptomatic stenoses without adequate collateral supply probably carry higher spontaneous stroke risk and may be considered for intracranial stenting, which itself is associated with procedural risks of up to 10%. Currently published case series show relatively high complication rates as a major drawback of endovascular treatment, mainly strokes after occlusion of perforating branches extending from the stenotic vessel segment or hemorrhagic complications. According to data from smaller feasibility studies, stroke rates in follow-up after successful stenting seem to be low. The average rate of high-grade restenosis with possible indication for reintervention is 10%. Improvements in endovascular treatment aim at reducing vessel wall trauma during balloon angioplasty by underdilatation or the use of self-expanding stents. Until complication rates are dependably reduced to values of 5-6%, indication for endovascular treatment should be restricted to patients without therapeutic alternatives. According to limited data with large variation between different studies, a prospective multicentric registry is proposed for systematic evaluation and further development of the method.