Paul Harrod-Kim MD, Presenter: Nothing to Disclose

Joseph Pulitano RT, Abstract Co-Author: Nothing to Disclose

Mark Orloff MD, Abstract Co-Author: Nothing to Disclose

David Lee Waldman MD, PhD, Abstract Co-Author: Nothing to Disclose

To analyze the impact of vessel size, opacification and method of reconstruction, maximum intensity projection (MIP) vs. volume rendering (VR), on CTA in evaluating live liver donors.

From June through December 2003, 15 patients underwent biphasic CT and DSA. Each exam was interpreted independently. A multi-detector (4-row)CT exam was performed with 150 mL of contrast. The arterial phase was obtained with bolus-tracking (collimation 2.5 mm, reconstruction interval 1.25 mm). The portal venous phase was then obtained (collimation 5 mm, reconstruction interval 2.5 mm). Post-processing was performed with a commercially available software program; MIP and VR images were made two weeks apart. CTA was compared to DSA by radiologists and a transplant surgeon for their ability to detect: 1) the artery supplying segment IV 2) hepatic artery variants 3) portal and hepatic vein variants 4) the degree of arterial conspicuity. The axial CT images were used as the basis for venous anatomy.

CTA sensitivity and specificity for hepatic artery variants was 85.7% and 100 % respectively. CTA had 100% sensitivity/specificity for hepatic and portal vein variants. CTA sensitivty for the segment IV artery was 69.2%. Significant differences were present in segment IV artery sizes (median 1.8 mm vs. 1.3 mm, P = .0093) and relative opacification (median (HU) 199.5 vs. 118.5 P = .043) between those exams in which the artery was visualized and in which it was not, based on the Mann Whitney test.MIP sensitivity was higher for the segment IV artery (69.2% vs. 46.2%, VR). Detection of arterial, hepatic and portal vein variants was identical between the two. The Chi-2 test demonstrated no significant difference in arterial conspicuity (P = 0.0921). There were subjectively improved areas of visualization with MIP.

CTA accurately depicts hepatic and portal venous anatomy. MIP had higher sensitivity for small arteries but was otherwise identical to VR. Arterial visualization is limited by vessel size and degree of opacification. Due to this limitation, CTA with thinner collimation and increased detector number is needed to become the sole modality for evaluating the vessels of living liver donors.

Harrod-Kim, P, Pulitano, J, Orloff, M, Waldman, D, An Analysis of the Accuracy and Limitations of CT Angiography in Evaluating the Hepatic Vessels of Live Liver Donors.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4403369.html