Does extending the body contour into air enhance surface dose calculation accuracy?

Purpose: In radiotherapy treatment planning systems, modelling of superficial dose may be aided by a body contour that is, by default, placed at the outermost air-tissue interface. Here we investigate the accuracy of superficial dose calculated using either the default body contour (DBC) or an extended body contour (EBC) compared to radiochromic film measurements made on a slab phantom and an anthropomorphic phantom.

Methods: Depth dose curves in the superficial region of the slab phantom were measured using stacked radiochromic films and irradiated using static beams delivered from varying incident angles. Surface doses on the anthropomorphic phantom were measured using films taped to the outside and irradiated with dynamic beams targeting the skin. No intervening immobilization devices were between the incident beams and films. Measurements were compared with dose calculations performed with Eclipse 15.6 Acuros XB using both the DBC and EBC. The EBC was generated by extending the DBC out into air by 1 cm.

Results: Differences between the DBC- and EBC-calculated dose were largest within the region ranging from the surface to a depth of 2 mm. Beyond 2 mm, calculations agreed to within 5%. For the slab phantom, the DBC- and EBC-calculated surface doses for 0-degree incident angle were 26% and 44% respectively, while measured dose was 24 ± 3%. At 60-degrees incident angle, DBC- and EBC-calculated surface doses were 33% and 66% respectively, while the measured dose was 43 ± 3%. For the anthropomorphic phantom, the DBC- and EBC-calculated surface doses were 65% and 99% of the prescription dose respectively, while the measured surface dose was 76 ± 3%.

Conclusions: Calculating with the EBC overestimated surface dose, while the DBC underestimated surface dose under conditions of oblique delivery. Overall, the default method of placing the body contour at the air-tissue interface produced a more accurate estimation of surface dose for bare skin.