A practical grading scale for predicting outcome after radiosurgery for arteriovenous malformations: analysis of 1012...

The authors performed a study to review outcomes following Gamma Knife radiosurgery for cerebral arteriovenous malformations (AVMs) and to create a practical scale to predict long-term outcome.

Outcomes were reviewed in 1012 patients who were followed up for more than 2 years. Favorable outcome was defined as AVM obliteration and no posttreatment hemorrhage or permanent, symptomatic, radiation-induced complication. Preradiosurgery patient and AVM characteristics predictive of outcome in multivariate analysis were weighted according to their odds ratios to create the Virginia Radiosurgery AVM Scale.

The mean follow-up time was 8 years (range 2-20 years). Arteriovenous malformation obliteration occurred in 69% of patients. Postradiosurgery hemorrhage occurred in 88 patients, for a yearly incidence of 1.14%. Radiation-induced changes occurred in 387 patients (38.2%), symptoms in 100 (9.9%), and permanent deficits in 21 (2.1%). Favorable outcome was achieved in 649 patients (64.1%). The Virginia Radiosurgery AVM Scale was created such that patients were assigned 1 point each for having an AVM volume of 2-4 cm(3), eloquent AVM location, or a history of hemorrhage, and 2 points for having an AVM volume greater than 4 cm(3). Eighty percent of patients who had a score of 0-1 points had a favorable outcome, as did 70% who had a score of 2 points and 45% who had a score of 3-4 points. The Virginia Radiosurgery AVM Scale was still predictive of outcome after controlling for predictive Gamma Knife radiosurgery treatment parameters, including peripheral dose and number of isocenters, in a multivariate analysis. The Spetzler-Martin grading scale and the Radiosurgery-Based Grading Scale predicted favorable outcome, but the Virginia Radiosurgery AVM Scale provided the best assessment.

Gamma Knife radiosurgery can be used to achieve long-term AVM obliteration and neurological preservation in a predictable fashion based on patient and AVM characteristics.