Hemorrhage risk of cerebral arteriovenous malformations before and during the latency period after GAMMA knife radiosurgery.

BACKGROUND AND PURPOSE To evaluate the hemorrhage rates of cerebral arteriovenous malformations (AVM) and the risk factors of hemorrhage before and after Gamma Knife radiosurgery (GKS). METHODS The annual hemorrhage rate was calculated as the number of hemorrhages divided by the patient-years at risk. Characteristics of patients and AVM related to hemorrhagic or nonhemorrhagic presentation were evaluated by logistic regression. Risk factors predicting AVM hemorrhage during the period from the diagnosis to GKS of AVM and during the latency period after radiosurgery were evaluated using Cox regression hazards model. RESULTS The annual hemorrhage rate before GKS was 2.0% assuming patients were at risk for hemorrhage since their birth. The hemorrhage rate calculated between the diagnosis and GKS of AVM was 6.6% and reduced to 2.5% after GKS until obliteration of the AVM. Although small and deep nidi and those with deep and single draining veins tended to present themselves with hemorrhage, only nidi with single draining veins and those ruptured before were more likely to bleed once the AVM had been diagnosed. These factors no longer predisposed the nidus to a rupture after radiosurgery and the only predicting factor for hemorrhage was a low radiosurgical prescription dose to the margin of nidus. CONCLUSIONS The AVM hemorrhage rate seems to reduce after GKS. After radiosurgery, none of the patients or nidus-related risk factors remained relevant to the occurrence of hemorrhage. The nidus treated with a high radiosurgical dose is less likely to bleed.