Jeremy Robert Wortman MD, Presenter: Nothing to Disclose

Urvi Pravin Fulwadhva MD, Abstract Co-Author: Nothing to Disclose

Gregory Aaron Bonci MD, Abstract Co-Author: Nothing to Disclose

Andrew Primak PhD, Abstract Co-Author: Employee, Siemens AG

Heidi Miracle MPH, MS, Abstract Co-Author: Nothing to Disclose

Aaron D. Sodickson MD, PhD, Abstract Co-Author: Research Grant, Siemens AG

The enhancement of normal anatomic structures varies substantially between individuals, confounding efforts to accurately detect pathology on quantitative grounds. Dual energy (DE) CT uniquely allows users to quantify tissue iodine content in a single acquisition. We aimed to evaluate variability of iodine enhancement of abdominal organs using DECT, and to select optimal reference tissues that minimize variability between individuals in normalized enhancement ratios.

76 consecutive portal-venous phase contrast-enhanced abdominal DE CT examinations (Siemens Definition Flash) were post-processed, and DE regions of interest (ROI) placed in iodine overlay images within parenchyma of the renal cortex, liver, adrenal gland, and spleen, as well as vascular structures including the aorta, renal vein, supra-renal IVC, and main portal vein (MPV). Each ROI yielded the attenuation due to iodine enhancement, which was divided by the organ distribution mean to re-scale the tissue enhancement (ENH) distribution to a mean of 1. For each organ, variances were calculated for the raw ENH distribution, and compared with the variances of the normalized ratios obtained by dividing organ ENH by each vascular reference ENH value. The Pitman-Morgan test for differences in variance in paired data was used to compare non-normalized values and normalized ratios.

There was substantial variation in organ enhancement (renal ENH example in Figure 1). For all organs, the variance was significantly reduced through normalization. The best vascular normalizer was found to be the MPV for all organs. Relative reductions in variance were: kidney 38.3% (p<0.0001); liver 17.9% (p=0.007); spleen 46.5% (p<0.0001); adrenal 17.4% (p=0.03). The IVC was consistently the worst normalizer due to substantial variations in enhancement on portal venous phase scans.

Normalization of DE iodine enhancement values using internal controls significantly reduces the observed variation between patients in enhancement of the abdominal organs. This will support future efforts to more accurately differentiate pathology from normal tissues on quantitative grounds.

We demonstrate a method to normalize iodine enhancement from DE CT that reduces physiologic variability between patients and may help to better differentiate pathologic from normal tissue enhancement.

Wortman, J, Fulwadhva, U, Bonci, G, Primak, A, Miracle, H, Sodickson, A, Quantification of Iodine Enhancement Using Dual Energy CT: Internal Normalization Minimizes Physiologic Variation between Patients.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006917.html