RESUMO
PURPOSE: Immobilization devices that improve the setup reproducibility of pelvic cancer patients and that provide comfort to patients during radiotherapy were designed and the feasibility of the devices was evaluated. MATERIALS AND METHODS: A customized device was designed to immobilize a knee, thigh, and foot of a patient. Sixty-one patients with prostate cancer were selected and were divided into two groups-with or without devices. The setup errors were measured with respect to bony landmarks. The difference between digitally reconstructed radiographs (DRR) and simulation films, and the differences between DRR and portal films were measured. RESULTS: The left-right (LR), anterior-posterior (AP) and craniocaudal (CC) errors between the DRR and simulation films were 1.5+/-0.9 mm, 3.0+/-3.6 mm, and 1.6+/-0.9 mm, respectively without devices. The errors were reduced to 1.3+/-1.9 mm, 1.8+/-1.5 mm and 1.1+/-1.1 mm, respectively with the devices. The errors between DRR and portal films were 1.6+/-1.2 mm, 4.0+/-4.1 mm, and 4.2+/-5.5 mm, respectively without the devices and were reduced to 1.0+/-1.8 mm, 1.2+/-0.9 mm, and 1.2+/-0.8 mm, respectively, with the devices. The standard deviations among the portal films were 1.1 mm, 2.1 mm, and 1.0 mm at each axis without the devices and 0.9 mm, 1.6 mm and 0.8 mm with the devices. The percentage of setup errors larger than 3 mm and 5 mm were significantly reduced by use of the immobilization devices. CONCLUSION: The designed devices improved the setup reproducibility for all three directions and significantly reduced critical setup errors.