Abstract
In this study, we assessed the precision and repeatability of the daily patient positioning for three distinct immobilization devices used for head-and-neck patients undergoing RapidArc radiation therapy using cone beam computed tomography (CBCT). An analysis was conducted on the accuracy of patient setup for three distinct immobilization devices, resulting in 1204 CBCT images for 189 patients in total. Using a typical posifix supine headrest and five fixation point podcast-plus-thermoplastic masks, the first group of 39 patients (125 CBCTs) was immobilized. The identical method was used to immobilize the second group of 19 patients (158 CBCTs) in the same posture (supine), and AccuFormTM custom headrests were employed as an added measure. Over 65% of the patients in the third group had a double shell positioning system (DSPS) covering their entire head and neck. Patient-alignment-accuracy or couch shifts in the vertical, longitudinal, and lateral directions from CT-CBCT fusions were recorded from ARIA. Our results showed that in 90% of the anteriorposterior (AP), 90% of the superior-inferior (SI), and 92.7% of the right-left (RL) population in the first group, patient-alignment-accuracy or couch shifts were within 2 mm. For 99.4% (AP), 100% (SI), and 98.7% (RL) of the second group's total population, patient-alignment-accuracy was within 2 mm. In the third group, it was within 2 mm for 92.1% (AP), ~89% (SI), and 93.3% (RL) of the total population. In conclusion, a significant improvement was seen with the application of a mouth bite and a tailored backrest cushion to the five fixation point posicast mask. In addition, significant improvement in the alignment of the lower neck area was observed with the use of DSPS. Virtually 100% of the head-and-neck patients were aligned within an accuracy of 3 mm, which is the PTV margin in our department.