Christopher Changchien BS, Abstract Co-Author: Nothing to Disclose

Arian Mashhood BS, Abstract Co-Author: Nothing to Disclose

Yauk Lee MD, Abstract Co-Author: Nothing to Disclose

Emmanuil Smorodinsky MD, Abstract Co-Author: Nothing to Disclose

Irene W Cruite MD, Abstract Co-Author: Nothing to Disclose

Joshua Chia-Hwa Chen BS, Presenter: Nothing to Disclose

Takeshi Yokoo MD, PhD, Abstract Co-Author: Nothing to Disclose

Michael E. Schroeder BS, Abstract Co-Author: Nothing to Disclose

Gavin Hamilton PhD, Abstract Co-Author: Research support, General Electric Company

Tanya Wolfson MS, Abstract Co-Author: Nothing to Disclose

Mark Bydder PhD, Abstract Co-Author: Research grant, General Electric Company

Michael Simca Middleton PhD, MD, Abstract Co-Author: Research Consultant, Siemens AG Research Consultant, Confirma, Inc Grant, General Electric Company Stockholder, General Electric Company Grant, Bayer AG Research Consultant, Merge Healthcare Research Consultant, Allergan, Inc Research Consultant, Merck & Co, Inc Research Consultant, Naviscan, Inc Speaker, SenoRx, Inc Speaker, C. R. Bard, Inc

Claude B. Sirlin MD, Abstract Co-Author: Research grant, Bayer AG Research grant, General Electric Company Research grant, Bracco Group Contract, Isis Pharmaceuticals, Inc Contract, Pfizer Inc Speaker Bureau, Bayer AG Consultant, Bayer AG Medical Advisory Board, Bayer AG Consultant, Merck & Co, Inc Medical Advisory Board, General Electric Company

To quantify total liver fat volume (TLFV), assess its intra-exam and inter-exam variability, and correlate it with total liver fat fraction (TLFF) using MRI

15 subjects (9 men, 6 women, mean age 32 yrs, range 14-57) with known or suspected fatty liver disease were enrolled in this IRB approved, HIPAA- compliant, prospective, pilot study. Axial images of the entire liver were acquired using a SGRE sequence with low flip angle (10°) to minimize T1 effects. Six serial opposed-phase and in-phase echoes (1.1, 2.3, 3.4, 4.6, 5.7, 6.9 ms) were obtained in a single breath hold. Each subject had three examinations, thirty minutes apart. Sequence was acquired three times in each examination. Three readers segmented the liver by drawing along the liver margin on all slices using MATLAB (MathWorks, Natick, MA). Readers also drew the largest-possible circular region-of-interest (ROI) in each Couinaud segment. Liver segmentations and ROIs were completed on an out-of-phase image and replicated on the other TE images. Proton density fat fraction was calculated using a previously described model (Yokoo et al. Diagnostic and Fat-Grading Accuracy at MR Imaging. Radiology 2009; 251:67-76). Per-voxel fat volume was calculated as the product of per-voxel fat fraction and voxel volume. TLFV was defined as the sum of all per-voxel fat volumes. TLFF was calculated as the mean of fat fractions from segmental ROIs

Fat volumes ranged from 18.1mL to 618.9mL (M=233.1mL, SD=166.1mL). Fat fractions ranged from 1.5% to 33.5% (M=13.6%, SD= 8.9%). TLFV intra-examination variability 0.92%, inter-examination variability was 0.37%, and ICC was 0.995. TLFF intra-examination variability was 1.69%, TLFF inter-examination variability was 0.07%, and ICC was 0.999. TLFV showed high correlation with TLFF (r=0.940, p<0.05)

TLFV and TLFF are repeatable within and between examinations, and TLFV highly correlates with TLFF

Total liver fat volume is a novel metric that may prove useful in evaluation of fatty liver disease. Further work is necessary to determine its utility.

Changchien, C, Mashhood, A, Lee, Y, Smorodinsky, E, Cruite, I, Chen, J, Yokoo, T, Schroeder, M, Hamilton, G, Wolfson, T, Bydder, M, Middleton, M, Sirlin, C, Quantification of Total Liver Fat Volume Using MRI-determined Proton Density Fat Fraction and Whole Liver Segmentation on MR at 3.0 T.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034522.html