RESUMEN
PURPOSE: To describe and evaluate a method that uses a 3-dimensional (3D) treatment planning system (TPS) to determine the relative dose to the lung, and to study the beam filtration required for lung sparing in translation total body irradiation (TBI). Special dosimetric problems related to moving couch were also considered. MATERIALS AND METHODS: The irradiation technique employed in our hospital is that of patient translation. The patient is positioned on a moving couch passing under a stationary Co-60 beam so that his/her entire body is irradiated. Measurements of basic data at source-skin distance (SSD)=150 cm were used to implement the Co-60 TBI unit to TPS (THERAPLAN plus), which was then used in dose computations. Two stationary, opposed anterior-posterior (40 x 40 cm) fields were employed to irradiate the Alderson phantom. The midline dose to either lung was computed and correction factors (CFs) were obtained that depend on the anatomy and densities of the tissues involved. These factors give the midline lung dose increase relative to the midline dose at the level of the mediastinum. Once the required lung dose was decided, the computed CF was used to estimate the filtration required from the measured broad beam attenuation data. The shielded lung dose distribution could be obtained from the TPS using a transmission corresponding to narrow beam geometry. To verify the TPS computations, measurements using a dosimeter and a diode system were carried out, employing solid water phantoms and the Alderson phantom. RESULTS: For the TPS employed, the computed midline CFs were lower than those measured in simple geometry phantoms for lung densities of 0.2-0.35 g/cm(3), by no more than 2%. For the Alderson phantom studied (lung density of 0.32 g/cm(3)), the computed CF was 1.11, which was 2% higher than the measured value. CONCLUSION: The advantages of a 3D TPS (dose distribution inside the lung, lung dose volume histograms [DVH], accurate attenuator shape from patient's anatomy etc.) allowed to study the lung dose in the Alderson phantom and to estimate the beam filtration required for lung sparing in TBI. The accuracy in lung dose computations, excluding the soft-tissue/lung interface was < or = 5%, which is within the clinical dose requirements. This procedure has been applied to a number of patients prior to their irradiation. Computations and in vivo measurements were in good agreement.