On the Monte Carlo Calculation of the Air-Kerma Strength for Brachytherapy Sources

Purpose: To study in detail how the various quantities entering the calculation of the air-kerma strength of brachytherapy sources affect the final value of this key quantity. Specifically, this study examines the role of the electrons emitted by the β radioisotopes used in brachytherapy sources, the effect of using μen/ρ instead of μtr/ρ, how the size of the bin used to determine the detected photon fluence modifies the final value of the air-kerma strength, the dependence of this value on the considered threshold photon energy, and how the level of detail with which simulations are carried out may affect the results of the calculations. Methods: Using the Monte Carlo code penelope, the air-kerma strength SK is calculated according to TG-43U1 protocol. Various simulations with different detail levels, electron transport parameters, and cut-off energies are conducted in order to elucidate how the various ingredients entering the definition of SK affect its final value. Calculations are carried out for two sources: the 60Co Flexisource Co-60 HDR and the 192Ir microSelectron mHDR-v2r. Results: Fully neglecting electron production and transport modified SK by 0.10%(2.0σ) for the Flexisource Co-60 HDR and 0.29%(14.1σ) for the microSelectron mHDR-v2r. The effect was larger in the case of the latter because electrons emitted by 192Ir are on average more energetic than those emitted by 60Co. Other ingredients of the calculation produced changes of the same order or bigger. For example, substituting μtr/ρ by μen/ρ reduced SK by 0.39%(7.9σ) and 0.08%(3.9σ) for the Flexisource and microSelectron, respectively. Increasing the size of the energy bins employed to score the photon fluence rate from 1 keV to 10 keV increased SK by 0.22%(4.4σ) for the Flexisource, while for the microSelectron a reduction of 0.49%(23.6σ) was obtained. Air composition also plays a role, producing changes of up to 0.16% when considering 40% moist air instead of dry air; a similar effect of up to 0.10% was generated by changing the low-energy photon threshold used to calculate the air-kerma rate. Finally, no differences significant within 1.7σ were found when using fully detailed or mixed electron transport schemes to carry out the simulations. Conclusions: Though electrons may seem not to affect kerma calculated via Monte Carlo simulation in an overall analysis of the problem, the more detailed study of their role carried out in the present work indicates that they can produce significant effects in the obtained air-kerma strength that cannot be neglected. Other ingredients must also be considered, as their contributions to air-kerma strength may sum up to produce an overall effect of ∼1%.