Feasibility of Submillisievert CT of the Skeletal Pelvis Using Iterative Reconstruction: A Human Cadaver Study

OBJECTIVE. The purpose of this study is to investigate the feasibility of submillisievert CT of the skeletal pelvis of human cadavers using a standard-dose protocol and four different reduced-dose protocols reconstructed with filtered back projection (FBP) and iterative reconstruction (IR). MATERIALS AND METHODS. The pelvis of 25 human cadavers was repeatedly examined using different reduced-dose CT (RDCT) protocols with decreasing reference tube current-exposure time products (RDCT protocol 1, 80 mAs; RDCT protocol 2, 60 mAs; RDCT protocol 3, 40 mAs; and RDCT protocol 4, 10 mAs) and a tube voltage of 120 kV. A standard-dose CT (SDCT) protocol (reference tube current-exposure time product, 100 mAs; tube voltage, 120 kV) used for the same cadavers served as the reference. Raw data were reconstructed using FBP and two increasing levels of IR (IR levels 4 and 6). The image quality and diagnostic acceptability of images of the anterior pelvic ring, acetabulum, and posterior pelvic ring including the sacroiliac joints were evaluated on a 5-point scale. A mixed-effects model for repeated measures was performed. RESULTS. The image quality of all anatomic structures was rated as diagnostically acceptable for all protocols reconstructed with IR, except for 11 cadavers that were imaged using RDCT protocol 4. For reconstructions with FBP, image quality was generally rated lower and was diagnostically acceptable only for images obtained using SDCT and RDCT protocol 1 and 2. RDCT protocol 3 with IR was the RDCT protocol with the largest reduced dose still allowing diagnostically acceptable image quality for all anatomic structures in all cadavers. Compared with SDCT, the RDCT protocols resulted in significantly reduced mean (± SD) effective radiation doses (SDCT, 2.0 ± 0.7 mSv; RDCT protocol 1, 1.6 ± 0.6 mSv; RDCT protocol 2, 1.2 ± 0.4 mSv; RDCT protocol 3, 0.8 ± 0.3 mSv; and RDCT protocol 4, 0.3 ± 0.1 mSv; p = 0.001). CONCLUSION. Diagnostically acceptable submillisievert CT of the skeletal pelvis is feasible using IR. To adhere to the ALARA (as low as reasonably achievable) principle, submillisievert pelvic CT protocols combined with IR should be implemented as part of routine clinical practice.