Is it time to retire the CTDI for CT quality assurance and dose optimization?

To the Editor, Ultimately, computed tomography quality assurance and dose optimization CT QA/DO have the goal of reducing, as far as practical, radiation-induced cancer risks in the examined population. In turn, these cancer risks are determined by the organ doses to which individuals are exposed. Thus it is logical for the quantities measured in CT QA/DO to bear as direct relationships to organ doses as is reasonably practical. As discussed here, the dose descriptors currently used for CT QA/DO, the CT dose index CTDI and its subsequent modifications, bear an increasingly distant relationship to organ doses. The technology does now exist, however, to directly and routinely measure organ doses from helical CT scans in realistic anthropomorphic phantoms, and it is thus suggested here that such measurements now represent a more logical basis for CT QA/DO than do CTDI measurements. Historically, because of the unique pattern of dose delivery inherent in CT scans, a variety of CT-specific dose descriptors have been developed, based on the concept of the CTDI. The CTDI descriptor was pragmatic in origin, and not intended as a surrogate of risk, but rather was designed to measure, in a homogeneous phantom, an average central dose associated with a multiple contiguous fixedcurrent axial-mode scan. The CTDI was necessarily based on single-slice dose measurements, having been formulated at a time when CT scanners were slow and with limited anode heat capacity for multislice measurements. While the CTDI has not failed in its original, quite limited, goal, as CT technology has advanced—for example to multislice variable pitch helical scanning—modified CTDI indices have been required, such as CTDI14T, CTDI100, CTDIw, CTDIvol, and dose length product DLP . While the CTDI was not originally intended as a surrogate of organ dose or risk, these modifications to the CTDI concept are, in part, motivated by a desire to use the CTDI in this way. As CT technology develops, still more modifications to the CTDI concept are and will be needed. For example, as multislice scanners utilize increasingly broad beams, the 100-mm ion chamber now used in CTDI measurements will no longer collect all the scatter from a singleslice profile. Another recent development is the advent of continuous automated tube current modulation along the z direction i.e., continuous current modulation to compensate for changes in attenuation by different organs along the patient axis ; in this situation, individual organ doses will no longer be scalable from the overall mean mAs.