The accuracy of the pencil beam convolution and anisotropic analytical algorithms in predicting the dose effects due to attenuation from immobilization devices and large air gaps.

When a photon beam passes through the treatment couch or an immobilization device, it may traverse a large air gap (up to 15 cm or more) prior to entering the patient. Previous studies have investigated the ability of various treatment planning systems to calculate the dose immediately beyond small air gaps, typically less than 5 cm thick, such as those within the body. The aim of this study is to investigate the ability of the Eclipse anisotropic analytical algorithm (AAA) and pencil beam convolution (PBC) algorithm to calculate the dose beyond large air gaps. Depth dose data in water for a 6 MV photon beam, cm²10×10² field size, and 100 cm SSD were measured beyond a range of air gaps (1-15 cm). The thickness of the water equivalent material positioned before the air gap ranged from 0.2 to 4 cm. Dose was calculated with the Eclipse PBC algorithm and AAA. The scattered and primary dose components were calculated from the measurements. The measured results indicate that as the air gap increases (from 1 to 15 cm) the dose reduces at the water surface and that beyond an air gap a secondary buildup region is required to re-establish electronic equilibrium. The dose beyond the air gap is also reduced at depths beyond the secondary buildup region. The PBC algorithm did not predict any reduction in dose beyond the air gap. AAA predicted the secondary buildup region but did not predict the reduction in dose at depths beyond it. The reduction in dose beyond the secondary buildup region was shown to be particularly relevant for air gaps of 5 cm or more when there was a 2 cm of water equivalent material positioned before the air gap. For these cases, where electronic equilibrium is established in the material positioned before the air gap, both algorithms were found to overestimate the dose by 2.0%-5.5%. It was concluded that the dose to depths of up to 15 cm beyond a large air gap is reduced due to a decrease in scattered radiation, produced in the material positioned before the air gap, reaching the point of interest. This effect is not well modeled by the Eclipse AAA and PBC algorithm and may result in dose calculation errors greater than 2.5%. Due to the contribution of other uncertainties in the radiation therapy treatment planning and delivery process, dose calculation errors of this magnitude are not consistent with the recommendation of the International Commission on Radiation Units and Measurements that the absorbed dose to the target volume be delivered with an uncertainty of less than ±5%