Jehan F Ghany MBChB,BA, Presenter: Nothing to Disclose

Ally Rosen MD, Abstract Co-Author: Nothing to Disclose

Karen M Lee MD, Abstract Co-Author: Nothing to Disclose

Pamela Anastasia Argiriadi MD, Abstract Co-Author: Nothing to Disclose

William Louie Simpson MD, Abstract Co-Author: Nothing to Disclose

This feasibility study was designed to use dual energy CT technology to create virtual non-contrast (VNC) images of the liver to investigate whether such images would be equivalent to true non-contrast images. If so, then VCN images could then be created and used in lieu of the non-contrast phase of a triple phase liver study thus decreasing ionizing radiation exposure in these patients.

22 patients with hepatocellular carcinoma (HCC) undergoing routine follow-up triple phase imaging were selected who had a previous history of trans-arterial chemoembolization (TACE) with retained embolization material on their most recent study. They were imaged on a GE Discovery CT750HD scanner with dual energy capability. VCN images were created at a workstation. The visibility of the residual embolization material on the VCN images was compared to that of the routine non-contrast images using a 3 point scale (1= not visible; 2= partially visible; 3 = fully visible) by three independent readers. Density of the embolization material on the routine non-contrast images was also measured and correlated with the visibility rating. The same process was performed for 10 control subjects who had a history of liver disease with or without HCC undergoing routine follow-up triple phase imaging.

Of the 22 patients, 15 (68%) received a visibility rating of 1 by all readers. The remaining 7 (32%) were rated as 2 by all readers. No VNC scan was rated as 3 by any reader. No VNC scan received different individual ratings by the three readers. For the 7 VNC scans that were rated as 2, the average density measured 362 – 1084 Housfield units (HU). For the 15 VNC scans rated as 1, the average density measured 75 - 625 HU. Thus, the threshold density for a lesion to be visible on a VNC scan is 362 – 625 HU. All 10 (100%) control subjects had VCN scans rated as 1.

VNC images created with dual energy technology in patients who have undergone TACE with retained embolization material cannot substitute for non-contrast images on a triple phase scan. However, VNC images are equivalent in patients who have not undergone TACE or have no residual embolization material.

Creation of VNC images of the liver which can replace non-contrast images on a triple phase scan in patients who have not had TACE thereby decreasing radiation dose.

Ghany, J, Rosen, A, Lee, K, Argiriadi, P, Simpson, W, Use of Virtual Non-contrast Imaging of the Liver to Decrease Radiation Dose in Triple Phase Studies.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034443.html