The development of new devices, especially controlled detachable coils, has made the endovascular approach one of the modalities for the treatment of intracranial aneurysms. We describe the treatment and present the results of 35 patients treated by selective occlusion of basilar artery aneurysms in our department during a period of 2 years (November 1992-November 1994). This period of time was chosen to analyze a homogeneous population treated since the introduction of controlled detachable coils and also to be able to have as many follow-up angiographic controls of the treated aneurysms as possible. The clinical presentation was subarachnoid hemorrhage in 32 patients and transient ischemic attack in 1 patient. In another two patients, the aneurysms were incidentally discovered. The majority of the aneurysms were berry aneurysms. The aneurysms were located at the basilar bifurcation (23 patients), at the basilar tip between the posterior cerebral artery and the superior cerebellar artery (5 patients), on the basilar trunk (3 patients), and at the vertebrobasilar junction (4 patients). Endovascular treatment using coils was achieved in 34 patients, using Guglielmi detachable coils (Target Therapeutics, San Jose, CA) in 29 patients and mechanical detachable spirals (Balt, Montmorency, France) in 5 patients. One patient died during the positioning of the first coil into the aneurysmal sac. Twenty-five of 35 aneurysms (73.5%) were completely occluded. Nine aneurysms (26.5%) were only partially (> 90%) occluded. No subsequent bleeding occurred during the follow-up period. Two patients treated in the acute phase of subarachnoid hemorrhage died days or weeks after endovascular treatment because of complications related to the natural history of subarachnoid hemorrhage (vasospasm in one patient and pulmonary complications in the other). In three patients, clotting occurred during the endovascular procedure. In all three patients, occlusion of the aneurysmal sac was achieved despite clotting. Urokinase was administered to two of the three patients. In the remaining patient, no fibrinolytic therapy was initiated. The clinical outcomes were excellent for all three patients. In this study, the morbidity-mortality rate of the endovascular technique is low (3%). If we include complications related to the subarachnoid bleeding, the morbidity-mortality rate remains low (8.8%) Regarding basilar artery aneurysms, endovascular treatment (selective occlusion by controlled detachable coils) is now useful for some patients, especially those with small aneurysms. However, long-term anatomic follow-up is needed to accurately evaluate the role of this treatment modality in the management of basilar aneurysms.