Stereotactic radiosurgery with or without embolization for intracranial dural arteriovenous fistulas.

BACKGROUND Treatment options for dural arteriovenous fistulas (DAVFs) have expanded with the application of stereotactic radiosurgery (SRS). OBJECTIVE To assess the role of SRS with or without embolization, we reviewed our entire DAVF experience. METHODS Between 1991 and 2006, 40 patients with 44 DAVFs underwent Gamma knife SRS. Twenty-eight patients had upfront SRS before or after embolization and 12 patients underwent delayed SRS for recurrent or residual DAVFs after initial embolization. The median patient age was 60 years (range, 29-90). DAVFs were diagnosed in 7 patients after they sustained an intracranial hemorrhage. The median SRS target volume was 2.0 mL (range, 0.2-8.2 mL) and the median marginal dose was 21.0 Gy (range, 15-25 Gy). RESULTS At a median follow-up of 45 months (range, 23-116 mo), a total of 28 patients (harboring 32 DAVFs) had obliteration confirmed by imaging. We found a 83% obliteration rate in patients who had upfront SRS with embolization and a 67% obliteration rate in patients who only had SRS. One patient died of an intracerebral hemorrhage 2 months after SRS. Cavernous carotid fistulas were associated with higher rates of occlusion (P = .012) and symptom improvement (P = .010) than were transverse-sigmoid sinus-related fistulas. CONCLUSION When upfront SRS is possible in conjunction with embolization, successful DAVF obliteration is possible in most patients, especially those with carotid cavernous fistulas. SRS should target the entire fistula regardless of whether it precedes or follows embolization. In selected patients with a small-volume, low-risk DAVF, SRS alone is an effective treatment option in most patients.