Effect of liquid embolic agents on Gamma Knife surgery dosimetry for arteriovenous malformations. Clinical article.

OBJECT

The effectiveness of Gamma Knife stereotactic surgery to obliterate brain arteriovenous malformations (AVMs) may be diminished by the preoperative adjunctive use of endovascular liquid embolic agents. The purpose of the present investigation was to determine if commercially available liquid embolic agents reduce the radiation dose to the target because of attenuation of the (60)Co beam.

METHODS

The apparent linear attenuation coefficients for 120- to 140-keV radiographs in embolized regions were retrieved from CT scans for several patients with AVMs who had undergone embolization procedures with liquid embolic agents to reduce nidal volumes. Based on these coefficients and a virtual model of Gamma Knife surgery (GKS) with basic ray tracing, the authors obtained the path lengths and densities of the embolized regions. The attenuation of (60)Co beams was then calculated for various sizes and positions of embolized AVM regions and for the number of beams used for treatment. Published experiments for several high-atomic-number materials were used to estimate the effective (60)Co beam attenuation coefficients for the N-butyl cyanoacrylate (NBCA, suspended in ethiodized oil) and ethylene vinyl alcohol copolymer (EVOH, with suspended micronized tantalum powder, Onyx) used in the AVM embolizations. Dose reductions during GKS were calculated for a theoretical model based on the CT-documented apparent linear attenuation coefficients and for the (60)Co energy attenuation coefficient. Dose measurements were obtained in a phantom study with EVOH for comparison with the estimates generated from the two attenuation coefficients.

RESULTS

Based on CT (keV) apparent attenuation coefficients, the authors' theoretical model predicted that the cumulative effect of either of the embolic agents decreased the number of kilovoltage photons in an embolized nidus by -8% to -15% because of the increased atomic number and density of NBCA and Onyx. However, in using the effective attenuation coefficient for the (60)Co energies as is used in GKS, the authors' theoretical model yielded only a 0.2% dose reduction per beam and a < 0.01%-0.2% dose reduction in total. These theoretical results were validated by measurements in a head phantom containing Onyx.

CONCLUSIONS

Dose reduction due to attenuation of the (60)Co beam by the AVM embolization material was negligible for both NBCA and EVOH because of the high-energy (60)Co beam.