Abstract
We evaluated for possible predictors of radiation-induced changes (RICs) after gamma knife radiosurgery (GKRS) for arteriovenous malformations (AVMs). We identified the nidal component within AVMs to analyze the correlation between the volume of brain parenchyma within the 50% isodose line (IDL) and RICs. We retrospectively reviewed patients with AVMs who underwent a single-session of GKRS at our institution between 2007 and 2017 with at least a 2-year minimum follow-up. Follow-up magnetic resonance images were evaluated for newly developed T2 signal changes and the proportions of nidus and intervening parenchyma were quantified. A total of 180 AVM patients (98 males and 82 females) with a median age of 34 years were included in the present study. The overall obliteration rate was 67.8%. The median target volume was 3.65 cc. The median nidus and parenchyma volumes within the 50% IDL were 1.54 cc and 2.41 cc, respectively. RICs were identified in 79 of the 180 patients (43.9%). AVMs associated with previous hemorrhages showed a significant inverse correlation with RICs. In a multivariate analysis, RICs were associated with a higher proportion of brain parenchyma within the 50% IDL (hazard ratio (HR) 169.033; p < 0.001) and inversely correlated with the proportion of nidus volume within the 50% IDL (HR 0.006; p < 0.001). Our study identified that a greater proportion of brain tissue between the nidus within the 50% IDL was significantly correlated with RICs. Nidus angioarchitectural complexity and the absence of a prior hemorrhage were also associated with RICs. The identification of possible predictors of RICs could facilitate radiosurgical planning and treatment decisions as well as the planning of appropriate follow-up after GKRS; this could minimize the risk of RICs, which would be particularly beneficial for the treatment of incidentally found asymptomatic AVMs.
Highlights
arteriovenous malformations (AVMs) contain no pathological tissue and the delivered dose is concentrated on a confined target volume [19]; it is not possible to exclude brain parenchyma from radiation exposure, which can lead to radiation injury to the intervening brain parenchyma within the isodose line (IDL) [20]
The incidence of Radiation-induced changes (RICs) observed following gamma knife radiosurgery (GKRS) treatment of AVMs is higher than The incidence of RICs observed following GKRS treatment of AVMs is higher than that seen following GKRS treatment of tumors, which demonstrates that RICs do that seen following GKRS treatment of tumors, which demonstrates that RICs do occur due to focal radiation injury
We found that the RIC group had more complex AVM angioarchitectures than the non-RIC group (Table 2), and higher Virginia Radiosurgery AVM Scale (VRAS), SM grade, and Pollock–Flickinger score were associated with the development of RICs (Table 3) as well as the grade of the RICs (Table 4) in the univariate analysis, the grading scales were not included in the multivariate analysis
Summary
Target volume has been suggested as an important predictor of RICs [10,11,14,17,21,22], the incidence of RICs in large AVMs is not consistently higher than that of smaller AVMs. AVMs that are widespread and intermingled with brain parenchyma are more likely to receive a higher dose of radiation than AVMs with compact vascular structures and little intervening brain tissue [20]. We evaluated possible predictive factors associated with the development of RICs after GKRS for AVMs. We attempted to identify nidal components within AVMs to analyze the correlation between the volume of brain parenchyma within the 50% IDL and RICs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
