Purpose: To assess the benefit of dual-energy (DE) over single-energy (SE) CT and determine optimal tube potential settings across a range of phantom sizes in routine non-contrast abdominal scans.
Methods: Images of the Catphan 500 phantom low-contrast module were acquired and reconstructed using standard clinical abdominal protocols on a dual-source (Siemens SOMATOM Force, Erlangen, Germany) CT scanner. Phantom scans were repeated with 30- and 35-cm rings added to evaluate SE and DE dose modulation. Six reference dose levels (0.75–24 mGy CTDIvol 32cm) were tested at four SE kVp settings (80, 100, 120, 140) and five DE kVp settings (70/Sn150, 80/140, 80/Sn150, 90/Sn150, 100/Sn150). Dose modulation (CAREDose4D) was enabled to adjust tube current based on phantom attenuation. The water-equivalent diameter (WED) of each phantom was determined. For each WED, size-specific dose estimates (SSDE) and image contrast-to-noise ratios (CNR) were computed.
Results: Tube current modulation behaved similarly in SE and DE scans. SSDE was consistent across SE and DE kV settings for a given CTDIvol and WED. DE acquisitions (mixing ratio 0.6) consistently produced greater CNR than SE acquisitions. Dose efficiency (CNR per mGy SSDE) was consistently greater for DE acquisitions by 46–89%. Dose efficiency was also greater for lower kV settings when imaging smaller phantom diameters with either SE or paired DE acquisitions. As phantom diameter increased, low-kV settings became less dose-efficient and produced artifacts.
Conclusion: For equivalent image CNR, non-contrast DE scans can reduce CT dose by approximately 40% over SE scans. For smaller patients, reduced kV settings should also be considered for greater dose efficiency and improved tissue contrast. These results indicate a dose-saving benefit to routine DE abdominal scanning. Contrast performance in vivo may differ from the low-contrast phantom data reported here. Additional benefits of low-contrast detectability in virtual monoenergetic images are being evaluated.