Validation of the radiosurgery-based arteriovenous malformation score in a large linear accelerator radiosurgery experience

Abstract

Object The radiosurgery-based arteriovenous malformation (AVM) score (RBAS) is a grading system designed to predict patient outcomes after Gamma Knife surgery for AVMs. This study seeks to validate independently the predictive nature of the RBAS, not only after single treatment but for retreatment, and to assess the overall outcome regardless of number of radiosurgeries. Methods The authors analyzed 403 patients treated with linear accelerator (LINAC) radiosurgery for AVMs between May 1988 and June 2008. The AVM scores were determined by the following equation: AVM score = (0.1 × volume in cm3) + (0.02 × age in years) + (0.3 × location). The location values are as follows: frontal/temporal = 0, parietal/occipital/corpus callosum/cerebellar = 1, and basal ganglia/thalamus/brainstem = 2. Results Testing demonstrated that the RBAS correlated with excellent outcomes after single or repeat radiosurgery (p < 0.001 for both variables). One hundred sixty-two (49%) of 330 patients had excellent outcomes (obliteration without deficit) after a single treatment. Excellent outcomes were achieved in 74, 64, 50, and 11% of patients with AVM scores of < 1.0 (Group 1), between 1.0 and < 1.8 (Group 2), between 1.8 and < 2.5 (Group 3), and ≥ 2.5 (Group 4), respectively. Fifty-one patients (70%) obtained radiosurgical cure and 46 (63%) achieved excellent outcomes after repeat radiosurgery. Of these, 100% achieved excellent outcomes in Group 1, 70% did so in Group 2, 47% in Group 3, and 14% in Group 4. The RBAS correlated with excellent outcomes after overall treatment (p < 0.001). Two hundred seventy-seven patients (69%) obtained AVM obliteration, and 62% achieved excellent outcomes. In Group 1, 87% achieved excellent outcomes, 75% did so in Group 2, 61% in Group 3, and 24% in Group 4. Conclusions The RBAS is a good predictor of patient outcomes after LINAC radiosurgery.