Pelvic Treatment Setup Differences Between Skin Markers-Based and Bony References Using Brainlab Exactrac
A Castro*, L Bardella, J Panichella, F Erlich, D Batista, Centro de Oncologia Rede D'Or, Rio de Janeiro, BrazilSU-E-J-49 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To quantify the translational and rotational setup differences between laseres on skin markers and bony references through X-ray Brainlab Exactrac system for patients under pelvic treatment.
Methods: 12 patients were initially localized via Exactrac and then received the isocenter laser projection marks on their skin. Over the course of treatment patients were daily localized by localization lasers and skin marks and then a pair of Brainlab Exactrac X-rays were taken and registered to planning Digital Reconstructed Radiographies (DRRs) in order to calculate the translational and rotational patient shifts. 6D Robotic couch were used to apply shifts. Individual and population statistical analysis of those shifts were performed.
Results: The average mean shifts for all directions (lateral, longitudinal and vertical) were close to zero, -0.6mm, -0.3mm and 1.6mm respectively. The average translational standard deviations were: lateral 2.7mm (ranging from 1.9 to 4.6mm), longitudinal: 2.3mm (1.0 to 4.4mm), and vertical: 3.5mm (1.3 to 5.0mm). For all patients studied the mean lateral angular shift (pitch) was always positive (mean 1.1°, ranging from 0° to 3.8°), while the overall longitudinal angular shift (roll) and vertical rotation were -0,1° (SD=0,6°, varying from -0,7° to 1.3°) and 0.0° (SD=0,4°, -0,5° to 0,9°) respectively.
Conclusion: The low value of overall mean translational shift indicates there is no significant systematic error for the studied patients. Anterior Posterior direction is the most sensitive to daily differences between lasers and Exactrac. Recurrent positive mean lateral angular shift for all patients might be related to table top sagging and its relation to patient weight needs to be further investigated. Patient localization by skin markers may lead to significant differences compared to bony localization and such discrepancies must be considered for PTV margins. Patient positioning via bony references with Exactrac may potentially reduce PTV margins and consequently side-effects.