Risk of hemorrhage from an arteriovenous malformation confirmed to have been obliterated on angiography after stereotactic radiosurgery

Abstract

Object. Radiosurgery has been widely adopted for the treatment of cerebral arteriovenous malformations (AVMs) in which the practical endpoint is angiographic evidence of obliteration, presumed to be consistent with elimination of the risk of hemorrhage. To test this unverified assumption, the authors followed 236 radiosurgery-treated AVMs between 1 and 133 months (median 77 months) after angiographic evidence of obliteration. Methods. Four patients experienced hemorrhage between 16 and 51 months after angiographic confirmation of AVM obliteration, and two underwent resection. The histological findings in these patients showed occlusion of the AVM by thickening of the intimal layer with dense hyalinization as well as a small amount of residual AVM vessels and a tiny vasculature. The risks of hemorrhage from these presumaby obliterated AVMs were 0.3% for the annual bleeding risk and 2.2% for the cumulative risk over 10 years. Continuous enhancement of the nidus on computerized tomography (CT) or magnetic resonance (MR) imaging was the only significant factor positively associated with hemorrhage in the statistical analysis (p = 0.0212). Conclusions. Because the study was based on limited follow-up data, its significance for defining predictive features of hemorrhage after angiographic evidence of obliteration is still indeterminable. Nevertheless, disappearance of the AVM on angiography after radiosurgery does not always indicate total elimination of the disease, especially when CT or MR imaging continues to demonstrate an enhancing lesion. The authors therefore recommend continual follow up even after evidence of AVM obliteration on angiography.