Radiosurgery to reduce the risk of first hemorrhage from brain arteriovenous malformations.

OBJECTIVE: It remains unclear whether or not and to what extent stereotactic radiosurgery can reduce the risk of first intracranial hemorrhage from brain arteriovenous malformations.
METHODS: We performed a retrospective observational investigation of 500 patients with arteriovenous malformations who were treated with gamma knife radiosurgery. The risk of first hemorrhage was analyzed using the Cox proportional-hazards model with age at radiosurgery and angiographic obliteration included as time-dependent covariates. Three periods were defined: from birth to radiosurgery (before radiosurgery); from radiosurgery to angiographic obliteration (latency period); and from angiographic obliteration to end of the follow-up period (after obliteration).
RESULTS: Hemorrhage was documented before radiosurgery in 318 patients (median observation period, 30.0 yr), during the latency period in 11 patients (median observation period, 2.2 yr), and after obliteration in two patients (median observation period, 5.5 yr). Compared with the period before radiosurgery, the risk of hemorrhage decreased by 86% after obliteration (hazard ratio, 0.14; 95% confidence interval, 0.03-0.55; P = 0.005), whereas the reduction observed during the latency period was not statistically significant (hazard ratio, 0.56; 95% confidence interval, 0.31-1.04; P = 0.07). Irrespective of obliteration, the risk of hemorrhage decreased by 62% after radiosurgery (hazard ratio, 0.38; 95% confidence interval, 0.22-0.67; P = 0.001). Similar results were observed when the 33 patients who had undergone previous therapy were excluded from the analysis.
CONCLUSION: Stereotactic radiosurgery significantly reduces the risk of first hemorrhage from brain arteriovenous malformations. The extent of the decrease might be greater if angiography indicates the evidence of obliteration.