Purpose: There is currently no consensus on the most accurate method for Gamma Knife output calibration, and significant variability can result from various â€˜acceptableâ€™ methodologies. The purpose of this work was to compare dose calibration formalisms for the Leksell Gamma Knife (GK) Icon unit using commercially available ionization chambers and solid phantoms.
Methods: Absorbed dose to water calibrations based on the TG-21 formalism and a modified TG-51 formalism were performed using two ionization chambers (Exradin A26, PTW 31010) and two spherical phantoms (solid water, lucite). Modifications to the TG-51 formalism were made in accordance with literature recommendations (Alfonso et. al, 2008), whereby a correction factor was applied to account for differences in field size, irradiation and phantom geometry, and phantom material. The modified TG-51 formalism was used only for the solid water phantom because appropriate correction factors for lucite are not available. As a verification, thermoluminescent dosimeters (TLDs) were irradiated and independently processed by a third-party.
Results: The absorbed dose rate to water between all measurements showed a maximum difference of 3.8%. Measurements using the same chamber, but different formalisms and phantom materials, yielded results with a spread of less than 1%. The results for the PTW 31010 chamber were 3% â€“ 4% lower than that of the Exradin A26 chamber which is most likely due to the volume averaging effect. The TLD-measured machine output was approximately 3% less than the average of measurements made with the A26 chamber using the TG-21 and modified TG-51 formalisms, however the TLD measurements carry an associated 3% uncertainty.
Conclusion: GK output measurements were performed by a variety of methods, and significant variation in the measured output, up to 3.8%, was observed. As there is currently no community consensus, it is critical for these differences be communicated to end-users of the Gamma Knife.