| Objective:To study a procedure that using an electronic portal imaging device (EPID) with radiopaque fiducial markers embeded in occlusal splint method ('occlusal splint method') for real-time correcting the setup error in nasopharyngeal carcinoma (NPC) intensity modulated radiation therapy (IMRT) in order to reduce the setup systematic error and random error.Methods:This study includes three stages:1) In order to detect the repeatability of the'occlusal splint method', 5 volunteer were chosen to make occlusal splint which embedded three metal markers at the department of stomatology, each volunteer applied with occlusal splint and undergone spiral-CT scanning in different time, then a reproducibility test was performed for the'occlusal splint method'.2) A NPC patient who treated by IMRT was chosen and simulated different translational errors and rotational errors by changing the centric position of irradiation field, couch angle and gantry angle in treatment planning system (TPS, PLATO v2.6.4).The doses of target volume and organ at risk under different setup errors were compared to those of the original plan, and then a correction threshold of setup error was defined as the dose variation.3) 30 NPC patients who treated by IMRT entered into a study of real-time correction of setup error using EPID and the'occlusal splint method'. Patient applying with occlusal splint during irradiation, three metal markers which embedded in the occlusal splint were used as markers for the setup error monitoring. Anterior and lateral isocentric electronic portal images (EPIs) were acquired before radiation and compared to digitally reconstructed radiograph (DRR). If the setup error of X-axis (Right-Left direction), Y-axis (Superior-Inferior direction) or Z-axis (Anterior-Posterior direction) exceeded the correction threshold, real-time correction was performed before radiation and the setup error after correction were detected.Results:1) A reproducibility test of occlusal splint showed that the degree of excursion is little and the'occlusal splint method'had a good stability in different time.2) The irradiation dose loss of GTV (D95), CTV1 (D95) and CTV2 (D95) were 87.2cGy (1.28%) vs. 157.8cGy (2.31%), 64.3cGy (1.04%) vs. 114.8cGy (1.85%) and 52.7cGy (0.94%) vs. 92.5cGy (1.66%) for 2mm and 3mm translational error, respectively. The translational error more than 2mm can obviously increase the irradiation dose of some organs at risk which were more serious than those of target volume. The rotational error more than 3o can obviously cause dose loss of target volume and dose increasing of some organs at risk.3) About 88.3% of all translational setup errors were≤3mm,and 99.1%≤5mm, all rotational setup errors were lower than 2°. 16 patients had translational setup errors more than 2mm in one or two direction which had undergone real-time correction. The standard deviation (SD) of overall setup systematic errors and random errors after correction were obviously lower than those before correction. The estimated margins required for CTV-PTV before correction were calculated according to the van Herk formula was about 3mm in three directions, which was a half after correction.Conclusion:This study shows as follows: 1) It can acquire a good repeatability that radiopaque fiducial markers embeded in'occlusal splint method'applying to the setup error monitoring in NPC IMRT. 2) The translational setup error more than 2mm or rotational setup error more than 3o can reduce the dose of target volume and increase the dose of organ at risk for NPC IMRT. Therefore, we suggest to enhance setup error monitoring for NPC IMRT, real-time correction should be performed if translational setup error more than 2mm or rotational setup error more than 3o.3) Real-time correction of setup error in NPC IMRT using EPID and'occlusal splint method'can reduce the setup systematic error and random error and the estimated margins required for CTV-PTV, improve the setup accuracy. |
PDF Full Text Request