Retrospective analysis of linac-based radiosurgery for arteriovenous malformations and testing of the Flickinger formula in predicting radiation injury

The aim of the study was to validate the use of linac-based radiosurgery in arteriovenous malformation (AVM) patients and to predict complications using an integrated logistic formula (ILF) in comparison with clinical outcomes. The results of radiosurgery in 92 AVM patients were examined. All patients were treated with linac-based radiosurgery. Of these, 70 patients were followed for 12-45 months (median, 24 months) and were analyzed. The treated volume varied from 0.09 to 26.95 cm(A3) (median, 2.3 cm(A3)) and the median marginal dose was 20 Gy (range, 10.4-22). The median 12-Gy volume was 9.94 cm(A3) (range, 0.74-60.09 cm(A3)). Patients and lesion characteristics potentially affecting nidus obliteration and excellent outcome were evaluated by performing a log-rank test and univariate and multivariate analyses. The risk for radiation injury (RRI) was calculated with an integrated logistic formula. The predictive power of the RRI was assessed by calculating the area under the receiver operating characteristic (ROC) curve. Follow-up magnetic resonance (MR) angiography revealed complete AVM obliteration in 56 of 70 patients. The MR angiography confirmed an obliteration rate of 80%. The annual hemorrhage rate was 1.4% for the first 2 years after radiosurgery and 0% thereafter. The number of patients with an excellent outcome was 48 (68%). Factors associated with better obliteration were higher radiation dose to the lesion margins [12-Gy volume (V12) > 10 cm(A3)], small volume, and a Pollock-Flickinger score less than 1.49; those predicting excellent outcomes were V12 < 10 cm(A3), small volume, and Pollock-Flickinger score less than 1.49, as determined by multivariate analyses. Factors associated with radiation injury were V12 > 10 cm(A3) (p=0.03) and volume greater than 2 cm(A3) (p=0.001), as determined by a univariate analysis. The analyses showed an ROC of 0.66. These data suggest that linac-based radiosurgery is effective. In terms of obliteration, excellent outcomes, and especially radiation injury, V12 and volume should be considered. The Flickinger formula seems to be applicable to Novalis-treated patients, but long-term follow-up is necessary for definite conclusions.