Volume-Staged Radiosurgery for Large Arteriovenous Malformation: Retrospective Analysis of 19 Cases

Introduction The effectiveness of Gamma Knife surgery (GKS) for small arteriovenous malformations (AVMs) is well known. However, for large AVMs, the prescribed dose should be decreased to reduce the risk of radiation damage, but it leads to a decrease in nidus obliteration rates. Therefore, it is very difficult to achieve complete obliteration of large AVMs in a single treatment, and methods using multiple irradiation such as volume-staged stereotactic radiosurgery (VS-SRS) have been suggested. We retrospectively reviewed our results of VS-SRS for large AVMs to assess the efficacy of VS-SRS. Methods Nineteen patients with AVMs of ≥10 ml and who consented to VS-SRS were treated by this surgical strategy and retrospectively analyzed. We excluded AVMs that were too large such as those >40 cc to avoid severe radiation damage. The components were divided mainly in the vertical direction, and each component was irradiated with a marginal dose of 18 Gy. Each irradiation was performed at intervals of 3-6 months, and the components with main feeders were irradiated first, and the components that included the main drainer were irradiated last. We tried to keep V18 to <10 ml if possible. The follow-up after GKS was performed by MRI every 6 months, and cerebral angiography was performed to confirm complete nidus obliteration, but if the patient refused, it was judged on the basis of MRI findings. Results Nineteen patients with a mean age of 40.2 years underwent VS-SRS. Each compartment was irradiated at 3--16 month (median, 3 months) intervals. The mean initial AVM volume was 19 ± 5.6 ml. Fourteen patients received two-stage radiosurgery and five received three-stage radiosurgery. The median target volume was 9.1 ml at stage 1, 9.0 ml at stage 2, and 10.1 ml at stage 3. The median margin dose was 18 Gy at each stage. The mean follow-up after the last stage of radiosurgery was 3.9 (1-11.4) years. Complete obliteration was confirmed by angiography in six patients, and by magnetic resonance angiography in one patient. The cumulative obliteration rates were 30.7% and 58.2% at 3 and 5 years following VS-SRS, respectively. The cumulative hemorrhage rates were 7.1% and 22.1% at 3 and 5 years, respectively. MRI showed T2-weighted prolongation in 15 patients (78.9%). Of these 15 patients, four were symptomatic (epilepsy in all) and two underwent surgical removal of symptomatic expanding hematomas. Conclusions In our experience, VS-SRS offers a viable treatment strategy in patients with large AVMs. Further optimization of the dose and volume at each stage is required.