A comparison of the respiratory signals acquired by different respiratory monitoring systems used in respiratory gated radiotherapy

PURPOSE: Respiratory monitoring systems are used to detect the respiratory phase of patients during the planning and administration of respiratory gated radiotherapy by using four-dimensional computed tomography (4DCT) or 4D positron-emission tomography/CT (4DPET/CT) and the linear accelerator (linac), respectively. Generally, identical respiratory monitoring systems are used for 4DCT, 4DPET/CT, and linac. However, different systems are sometimes used in combination because the accessibility of the respiratory monitoring systems may differ by manufacturer. The combined use of different respiratory monitoring systems in phase-based gating is of concern because the differences in the timing of tags (end-respiration signals algorithmically determined by the respiratory monitoring system), defined by the two systems, may result in phase differences, The purpose of this study is to estimate this difference and evaluate its effect on 4DCT data. METHODS: Ten patients (seven men and three women) with a median age of 75 yr (range: 57-84 yr) were treated by gated stereotactic body radiation therapy between April and December 2009. Two types of respiratory monitoring systems--RPM (Varian Medical Systems) and AZ-733V (Anzai MEDICAL)--were placed on the abdominal surface of the patients, and the respiratory signals were acquired by both systems. The relationship between the amplitude peak and the tag obtained by each respiratory system was analyzed for each patient. Further, the 4DCT images were reconstructed by using the signals obtained from both the RPM and the AZ-733V systems, and the tumor volumes and the tumor centroid positions in the craniocaudal plane were analyzed for each patient. RESULTS: The correlation factor between the respiratory signals from the RPM system and AZ-733V system was 0.990 (range: 0.940-0.994). The amplitude peak of the RPM system corresponded well with that of the AZ-733V system. The median +/- standard deviation of the phase difference for all the patients ranged from -4.3 +/- 7.1% to 3.5 +/- 2.2%. In the case of some patients, differences were noted between the two systems in the estimation of the tumor centroid position and tumor shape. CONCLUSIONS: The estimation of the position of the tumor centroid and tumor shape may vary with the use of different respiratory monitoring systems. This implies that it is preferable to use the same respiratory monitoring system with 4DCT, 4DPET-CT, and linac.