Pancreatic dual-source dual-energy CT: Is it time to discard unenhanced imaging?

Abstract

To compare pancreatic virtual unenhanced (VUE) and true unenhanced (TUE) images and to calculate the potential dose reduction by omitting the conventional unenhanced scan. Fifty-one patients with known or suspected pancreatic masses underwent contrast-enhanced computed-tomography (CT) during unenhanced and portal venous phases acquired in single-energy (SE) mode, and pancreatic parenchymal phase acquired in dual-energy (DE) mode. The image quality (IQ) and image noise (IN) of TUE and VUE images were evaluated. The effective dose of a combined DE/SE dual-phase protocol was compared with that of a theoretical standard SE triple-phase protocol. The results were tested for statistical significance using the Cohen's k, the Wilcoxon's signed rank test, and the paired t-test; p-values of less than 0.05 were considered significant. Mean TUE and VUE IQ were 1.5±0.6 and 1.6±0.6 (k=0.891), with no significant difference (p>0.05). Mean TUE and VUE IN were 12.3±1.6 and 10.3±1.5HU, and resulted significantly different (p<0.001). Mean effective doses for a combined DE/SE dual-phase protocol and SE triple-phase protocol were 8.9±2.4mSv (range 4.8-16.2mSv) and 12.1±3.1mSv (range 6.4-21.1mSv). The calculated mean dose reduction achievable by omitting the unenhanced scan was 26.7±9.7% (range 10-46.1; p<0.001). VUE images are feasible for pancreatic abdominal CT. A combined DE/SE dual-phase protocol permits a significant reduction in dose exposure to patients.