An investigation of backscatter factors for kilovoltage x-rays: a comparison between Monte Carlo simulations and Gafchromic EBT film measurements.

Backscatter factors are important parameters in the determination of dose for kilovoltage x-ray beams. However, backscatter factors are difficult to measure experimentally, and tabulated values are based largely on Monte Carlo calculations. In this study we have determined new backscatter factors by both experimental and Monte Carlo methods, and compared them with existing backscatter factors published in the AAPM TG-61 protocol. The purpose of this study is twofold: (1) to evaluate the overall effectiveness of using Gafchromic EBT film for backscatter factor measurements and (2) to determine whether existing Monte Carlo-calculated backscatter factors need to be updated. We measured backscatter factors using Gafchromic EBT film for three field sizes (2, 4 and 6 cm diameter cones) and three kilovoltage beam qualities, including 280 kVp for which similar measurements have not previously been reported. We also present new Monte Carlo-calculated backscatter factors obtained using the EGSnrc/BEAMnrc code system to simulate the Pantak kilovoltage x-ray unit used in our measurements. The results were compared with backscatter factors tabulated in the AAPM TG-61 protocol for kilovoltage x-ray dosimetry. The largest difference between our measured and calculated backscatter factors and the AAPM TG-61 values was found to be 2.5%. This agreement is remarkably good, considering that the AAPM TG-61 values consist of a combination of experimental and Monte Carlo calculations obtained over 20 years ago using different measurement techniques, as well as older Monte Carlo code and cross-section data. Furthermore, our Monte Carlo-calculated backscatter factors agree within 1% with the AAPM TG-61 values for all beam qualities and field sizes. Our Gafchromic film measurements had slightly larger differences with the AAPM TG-61 backscatter factors, up to approximately 2% for the 6 cm diameter cone at a beam quality of 50 kVp. The largest difference in backscatter factors, of 2.5%, was found between Monte Carlo-calculated and Gafchromic film-measured data for the 100 kVp x-ray beam with the 4 cm diameter cone. The differences in backscatter factors between the three data sets (measurements, calculations and published values) are all within the uncertainties from our Gafchromic film measurements and Monte Carlo calculations. Our results demonstrate the suitability of using Gafchromic EBT film to measure equipment-specific backscatter factors for kilovoltage x-ray beams over the entire energy range and also confirm that backscatter factors published in kilovoltage dosimetry protocols still remain valid.