Abstract
Background: Arteriovenous malformation (AVM) is one of the cerebrovascular diseases that bear a high risk of hemorrhage. The treatment modalities include microsurgical resection, endovascular embolization, stereotactic radiosurgery, or combinations that vary widely. Several large series have been reported, while data from Asian populations were few. The aim of this study was to examine the efficacy of linear accelerator stereotactic radiosurgery (LINAC SRS) for the treatment of intracranial AVMs, to evaluate the hemorrhage rate and to analyze associated factors. Methods: One hundred and sixteen patients with AVM were treated with LINAC SRS in a single institute between September 1994 and May 2005 and were retrospectively evaluated. The demographics of patients, clinical characteristics of AVM, the treatment modalities, and the parameters of the LINAC SRS were analyzed. Delayed toxicity and hemorrhage rate after treatment were also evaluated. The AVM obliteration and bleed rates were calculated using the Kaplan-Meier method and Cox regression analyses. Results: The efficacy rate with total obliteration after treatment was 81.9% (95 of 116 patients). The median interval to achieve total obliteration was 49 months. Microsurgical resection combined with SRS for residual AVMs achieved better obliteration rates compared to SRS alone (statistically significant, p = 0.001), while no significant difference was found between the embolization group and the group with no prior treatment (p = 0.895). The Spetzler-Martin grade of AVM is a relative factor of obliteration, higher grades resulting in a worse outcome (p = 0.009). Obliteration was significantly influenced by AVM volume in univariate analysis (p = 0.034), and volume <5 cm3 contributed to improved obliteration (p = 0.01). There was no statistically significant difference in the hemorrhagic rate and the complication rate between ruptured and unruptured AVMs, while the unruptured group had a higher obliteration rate (p = 0.024). The annual hemorrhage rate after LINAC SRS treatment was 1.9%. The bleeding rate was 3.3% in the first year after radiosurgery, 2.1% in the second year, 1.9% between the second and fifth year, and 1.5% between the fifth and tenth year. Patients with hemorrhagic events before radiosurgery appeared to have a higher rebleeding risk during the latency period. Twenty-three patients (19.8%) had late adverse effects with regard to posttreatment radiological follow-up, but only 1 (0.8%) had newly developed neurological deficits. Conclusion: LINAC SRS achieved a high obliteration rate and reduced the risk of hemorrhage effectively in ruptured and unruptured intracranial AVMs. Prior microsurgical resection provided better outcome, while embolization showed no benefit. Adverse effects after treatment are acceptable and require long-term follow-up.