Accuracy of dose calculation methods for retracted tissue compensators.

Uncertainties arise in dose calculations involving retracted tissue compensators due to the effects of the compensator upon the scatter component of the dose. Many commercial treatment-planning systems cannot allow directly for the presence of a compensator in isodose calculation or else use simple 2D methods. We present data to test calculation accuracy for a wax compensation system by comparing retraction factors measured along central-axis and off-axis raylines for a variety of compensator shapes, with those derived using effective attenuation coefficient and 3D analytical calculations. The accuracy of using measured uniform-thickness retraction factors for non-uniform shapes and the dose uniformity achievable using a simple compensation system are also discussed. We conclude that the accuracy of simple calculation methods is shape dependent and that calculation errors and dose variations can exceed +/-5% where missing-tissue thickness variations are large. The analytical method is shown to give good agreement with experiment and indicates that it should be possible to adapt algorithms that calculate scatter from the patient for use with compensators.