Dosimetric Measurements of An N-Butyl Cyanoacrylate Embolization Material for Arteriovenous Malformations
Z E Labby*, D A Roberts, N Chaudhary, J J Gemmete, A S Pandey, The University of Michigan Hospital & Health Systems, Ann Arbor, MISU-E-T-415 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: The therapeutic regimen for cranial arteriovenous malformations (AVMs) often involves both stereotactic radiosurgery and adjunct endovascular embolization. Embolization agents often contain radiopaque contrast to assist the neurointerventionalists, leading to concerns regarding the dosimetric effects of these contrast agents. This study investigated dosimetric properties of n-butyl cyanoacrylate (n-BCA) plus lipiodol (NPL) with and without tantalum powder.
Methods: The NPL was provided cured from the manufacturer (Codman Trufill; Codman & Shurtleff, Inc., Raynham, MA) with and without tantalum. Attenuation measurements were made for the samples with and without tantalum powder and compared to the attenuation of a water substitute using a 6MV photon beam. Effective linear attenuation coefficients (ELAC) were derived from the attenuation measurements and measured sample thicknesses. Interface effects were investigated using a parallel plate ion chamber placed at set distances below fixed samples. Finally, Hounsfield units were measured using our department's standard stereotactic radiosurgery CT protocol.
Results: For a 3cm square field, the ELAC was 0.0428 1/cm and 0.0474 1/cm for NPL without and with tantalum respectively, compared to 0.0461 1/cm for the water substitute. Non-density corrected dose calculations would result in errors of -0.33% and 0.13% per cm thickness of NPL without and with tantalum, respectively. Interface effects compared to water were less than 1.3% for both embolization materials. CT values at 120 kVp were 2082 HU and 2358 HU for NPL without and with tantalum, respectively.
Conclusion: The dosimetric properties of NPL are very close to those of water for therapeutic energies. Therefore, treating the entire intracranial space as uniform in composition will result in typically less than 1% dosimetric errors. However, due to the high attenuation of both embolization agents for diagnostic energies, artifacts may occur that necessitate MR imaging to accurately identify the spatial extent of the region to be treated.