AJR:197, October 2011
Diana Litmanovich, Denis Tack, Pei-Jan P. Lin, Phillip M. Boiselle, Vassilios Raptopoulos, Alexander A. Bankier
OBJECTIVE. We compared phantom organ doses delivered to breast, lung, and pelvis by five protocols using current dose reduction methods for routine chest CT and pulmonary CT angiography.
MATERIALS AND METHODS. We measured the radiation dose to an anthropomorphic phantom using 64-MDCT with metal oxide semiconductor field effect transistor (MOSFET) detectors in the breast (skin and parenchyma), the lungs, and the pelvis (upper and lower). We compared the following five protocols: protocol 1,120 kVp, automatic dose modulation, 120-320 mA; protocol 2,120 kVp, automatic dose modulation, 60-200 mA; protocol 3,100 kVp and fixed dose of 200 mA; protocol 4, 120 kVp, automatic dose modulation, 200-394 mA; and protocol 5, 80 kVp and fixed dose of 120 mA. Organ doses in milligrays and as a percentage of the volume CT dose index (CTDIvol) were compared using the analysis of variance for repeated measurements.
RESULTS. Protocol 1 delivered the highest breast dose (mean ± SD, 15.8 ± 1.8 mGy; 110.5% of CTDIvol). A decrease in breast radiation of more than 50% was achieved with protocol 3 (4.8 ± 1.8 mGy; 91.7% of CTDIvol) compared with protocol 4 (13.1 ± 5.5 mGy; 87.0% of CTDIvol) (p = 0.003). The lung received the highest organ dose regardless of the protocol (protocol 4: 21.5 ± 1.7 mGy; 142.5% of CTDIvol). Pelvic radiation was low regardless of protocol and did not exceed 0.2 mGy (up to 3.7% of CTDIvol; p = 0.118-0.999).
CONCLUSION. The results of this anthropomorphic phantom study showed substantial and significant variation in radiation doses to the breast and lungs depending on the scanning protocol used with the potential for over threefold dose reduction.