Linear accelerator-based radiosurgery alone for arteriovenous malformation: more than 12 years of observation.

Abstract

Although radiosurgery is an accepted treatment method for intracranial arteriovenous malformations (AVMs), its long-term therapeutic effects have not been sufficiently evaluated, and many reports of long-term observations are from gamma-knife facilities. Furthermore, there are few reported results of treatment using only linear accelerator (LINAC)-based radiosurgery (LBRS). Over a period of more than 12 years, we followed the long-term results of LBRS treatment performed in 51 AVM patients. The actuarial obliteration rates, after a single radiosurgery session, at 3, 5, 10, and 15 years were 46.9%, 54.0%, 64.4%, and 68.0%, respectively; when subsequent radiosurgeries were included, the rates were 46.9%, 61.3%, 74.2%, and 90.3%, respectively. Obliteration rates were significantly related to target volumes ≥4 cm(3), marginal doses ≥12 Gy, Spetzler-Martin grades (1 vs other), and AVM scores ≥1.5; multivariate analyses revealed a significant difference for target volumes ≥4 cm(3). The postprocedural actuarial symptomatic radiation injury rates, after a single radiation surgery session, at 5, 10, and 15 years were 12.3%, 16.8%, and 19.1%, respectively. Volumes ≥4 cm(3), location (lobular or other), AVM scores ≥1.5, and the number of radiosurgery were related to radiation injury incidence; multivariate analyses revealed significant differences associated with volumes ≥4 cm(3) and location (lobular or other). Positive results can be obtained with LBRS when performed with a target volume ≤4 cm(3), an AVM score ≤1.5, and ≥12 Gy radiation. Bleeding and radiation injuries may appear even 10 years after treatment, necessitating long-term observation.