Purpose: Modern linear accelerators (LINACs) are equipped with kV-imaging systems for 2D-images and 3D-cone-beam CT (CBCT). The delivered single kV-CBCT-dose is small compared to the treatment dose but can add up to a cumulative dose of 1-3% for multi-fraction courses according to TG180. Thus, TG142 recommends an annual assessment of the imaging dose. The purpose of this project was to review the current practice, compare approaches suggested in recent literature and standardize annual kV-CBCT dose measurements.
Methods: Current institutional practice is to measure the CBCT dose with the 10cm-long RaySafe X2 pencil-ion-chamber inserted in the center position of a CIRS lung phantom. We introduced the PMMA-CTDI phantom in a satellite-location for all measurements. Dose was measured at 3/6/9/12 o’clock and center positions of a 32cm-wide CTDI phantom to determine max dose heterogeneity for a half-trajectory imaging protocol and at the 12 o’clock and center position of either the head or abdomen phantom for 11 protocols ranging from 80kVp/100mAs to 140kVp/1688mAs using the Raysafe X2, the Standard Imaging A101 pencil-ion-chamber, and the A12 Farmer-ion-chamber. Measurements were repeated with decreased kV-tube-collimator opening as suggested by Varian.
Results: The dose discrepancy due to chamber position was largest between the 6 and 3 o’clock positions (10%) and the average was close to the peripheral measurement at 12 o’clock. All three detectors showed a linear dependency. Reducing CBCT beam collimation increased measurement complexity with only minimally improved agreement between measured and displayed CTDIw values. The standard deviation for dose measurements at different machines reduced from 22%/12% for the lung phantom to 7%/3% (max/mean) for the CTDI phantom.
Conclusion: Dose measurements at 12 o’clock and center position of CTDI-phantom have been defined as our new standard using un-modified treatment imaging protocols; conversion factors for center-only or lung-phantom measurements have been provided.