Absorbed dose to water based dosimetry versus air kerma based dosimetry for high-energy photon beams: an experimental study

In recent years, a change has been proposed from air kerma based reference dosimetry to absorbed dose based reference dosimetry for all radiotherapy beams of ionizing radiation. In this paper, a dosimetry study is presented in which absorbed dose based dosimetry using recently developed formalisms was compared with air kerma based dosimetry using older formalisms. Three ionization chambers of each of three different types were calibrated in terms of absorbed dose to water and air kerma and sent to five hospitals. There, reference dosimetry with all the chambers was performed in a total of eight high-energy clinical photon beams. The selected chamber types were the NE2571, the PTW-30004 and the Wellhöfer-FC65G (previously Wellhöfer-IC70). Having a graphite wall, they exhibit a stable volume and the presence of an aluminium electrode ensures the robustness of these chambers. The data were analysed with the most important recommendations for clinical dosimetry: IAEA TRS-398, AAPM TG-51, IAEA TRS-277, NCS report-2 (presently recommended in Belgium) and AAPM TG-21. The necessary conversion factors were taken from those protocols, or calculated using the data in the different protocols if data for a chamber type are lacking. Polarity corrections were within 0.1% for all chambers in all beams. Recombination corrections were consistent with theoretical predictions, did not vary within a chamber type and only slightly between different chamber types. The maximum chamber-to-chamber variations of the dose obtained with the different formalisms within the same chamber type were between 0.2% and 0.6% for the NE2571, between 0.2% and 0.6% for the PTW-30004 and 0.1% and 0.3% for the Wellhöfer-FC65G for the different beams. The absorbed dose results for the NE2571 and Wellhöfer-FC65G chambers were in good agreement for all beams and all formalisms. The PTW-30004 chambers gave a small but systematically higher result compared to the result for the NE2571 chambers (on the average 0.1% for IAEA TRS-277, 0.3% for NCS report-2 and AAPM TG-21 and 0.4% for IAEA TRS-398 and AAPM TG-51). Within the air kerma based protocols, the results obtained with the TG-21 protocol were 0.4-0.8% higher mainly due to the differences in the data used. Both absorbed dose to water based formalisms resulted in consistent values within 0.3%. The change from old to new formalisms is discussed together with the traceability of calibration factors obtained at the primary absorbed dose and air kerma standards in the reference beams (60Co). For the particular situation in Belgium (calibrations at the Laboratory for Standard Dosimetry of Ghent) the change amounts to 0.1-0.6%. This is similar to the magnitude of the change determined in other countries.