Xiaobing Fan PhD, Presenter: Nothing to Disclose

Chad Haney PhD, Abstract Co-Author: Nothing to Disclose

Devkumar Mustafi PhD, Abstract Co-Author: Nothing to Disclose

Marta A. Zamora BS, Abstract Co-Author: Nothing to Disclose

Erica Markiewicz BA, Abstract Co-Author: Nothing to Disclose

Gregory Stanislaus Karczmar PhD, Abstract Co-Author: Research Consultant, Perceptive Informatics, Inc Research Consultant, BioClinica, Inc

Gillian Maclaine Newstead MD, Abstract Co-Author: Consultant, Naviscan, Inc Consultant, Bayer AG Spouse, stockholder, Hologic, Inc

Accurate measurement of the arterial input function (AIF) is critical for quantitative analysis of dynamic contrast enhanced (DCE) MRI data. The ‘reference tissue method’ infers a local AIF based on contrast media uptake in the reference tissue, and this avoids errors associated with direct arterial measurements. However, this approach relies on literature values for Ktrans and ve. Here we estimate Ktrans and ve for a reference tissue by comparing its contrast media concentration to the plasma concentration during the washout phase.

DCEMRI of Copenhagen rats with prostate tumors in the hind limb was performed on a 4.7 T scanner. Small vessels in the leg with slow blood flow were selected to determine plasma concentration of contrast media. A region in leg muscle was selected as reference tissue. The ve of muscle was determined by comparing contrast concentration in muscle and plasma at equilibrium, i.e. at long times after contrast injection. Using the calculated ve, we found the Ktrans that gave the best fit to the muscle data beginning at 1 minute after injection, using the two compartment model, and the directly measured plasma contrast media concentration during the washout phase. Ktrans and ve of muscle, and the muscle concentration of contrast media as a function of time were used to determine the complete local AIF using the conventional reference tissue method.

The average Ktrans and ve for muscle were 0.14±0.06 (min-1) and 0.19±0.05, respectively (n=8). The AIF calculated from the muscle reference tissue had peak amplitude of 2.77±1.19 mM, which was much higher (p < 0.002) than the peak amplitude measured directly from the vessel (0.69±0.07 mM).

The new approach described here produced Ktrans and ve values that are consistent with literature values. However the results suggest large intra-subject variation and this demonstrates the need to perform these calculations on a subject-by-subject basis. The amplitudes of the calculated AIF’s were much larger than the AIF amplitude measured directly from blood vessels – demonstrating the potential advantage of reference tissue measurements.

This technique can easily be applied to clinical DCEMRI data to perform quantitative analysis, which may improve diagnostic capability.

Fan, X, Haney, C, Mustafi, D, Zamora, M, Markiewicz, E, Karczmar, G, Newstead, G, Using a Reference Tissue Plus Blood Vessel to Derive More Accurate Arterial Input Function for Analyzing DCEMRI Data.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9009196.html