ISMRM 23rd Annual Meeting & Exhibition • 30 May - 05 June 2015 • Toronto, Ontario, Canada

Combined Educational & Scientific Session

Quantitative Biomarkers in Liver MRI: How to Use Them in the Real World


ORGANIZERS: Lorenzo Mannelli, M.D., Ph.D., Ivan Pedrosa, M.D., Scott B. Reeder, M.D., Ph.D. & Edwin J.R. van Beek, M.D., Ph.D., M.Ed., FRCR

Monday 1 June 2015

This two-hour course focuses on the implementation of quantitative MRI biomarkers for diffuse hepatic disease in clinical practice. State-of-the-art approaches for quantification of hepatic steatosis, iron overload and hepatic fibrosis will be described. Emphasis will be made on understanding the basic principles applied to these quantification methods as well as understanding how to interpret the results of these MRI examinations and how to include these results in an MRI report. Didactic presentations and scientific abstracts on each area of interest will be presented to enhance the educational and scientific value of this course on each specific area of interest.

Target Audience
This course is aimed at radiologists, imaging scientists and MR technologists who wish to review the state-of-art MRI methods for quantification of diffuse hepatic disease, and to learn about incorporating these methods in routine clinical practice.

Educational Objectives
As a result of attending this course, participants should be able to:
• Illustrate the clinical role of MRI biomarkers of liver disease;
• Demonstrate how to implement these methods into clinical protocols;
• Convey practical approaches to interpret quantitative biomarkers of liver disease; and
• Integrate approaches to reporting quantitative biomarkers of liver disease.

Moderators: Catherine D. G. Hines, Ph.D., Kartik S. Jhaveri, M.D.
10:45   Introduction
10:48   Liver Fat Quantification - Seriously, Who Cares?
Mustafa Shadi R. Bashir, M.D.
11:03 0085.   Systematic comparison between modified Dixon MRI techniques, MR spectroscopic relaxometry, and different histologic quantification methods in the assessment of fatty liver disease
Guido Matthias Kukuk1, Alois Martin Sprinkart1, Wolfgang Block1, Holger Eggers2, Jürgen Gieseke1,3, Kanishka Hittatiya1, Patrick Kupczyk1, Julian Luetkens1, Rami Homsi1, Vera Keil1, Michael Meier-Schroers1, Milka Marinova1, Asadeh Lakghomi1, Dariusch Hadizadeh1, Hans Heinz Schild1, and Frank Träber1
1University of Bonn, Bonn, NRW, Germany, 2Philips Research Europe, Hamburg, Hamburg, Germany, 3Philips Healthcare, Best, NL, Netherlands

The availability of various non-invasive and invasive techniques for liver fat quantification raises the question for comparability and congruence of quantitative results. Therefore we performed a systematic comparison of mDixon MRI techniques, MRS and two different histologic methods for liver fat quantification. Our results demonstrate that six-echo-mDixon, MRS and semi-automatic histologic fat quantification provide the most robust and congruent data. The performance characteristics and systematic differences of the various fat quantification methods should be considered in clinical routine and in further studies when comparing results obtained from biopsy with quantitative MRI or MRS.

11:15 0086.   Multi-site, multi-vendor validation of accuracy, robustness and reproducibility of fat quantification on an oil-water phantom at 1.5T and 3T
Diego Hernando1, Mustafa R. Bashir2, Gavin Hamilton3, Jean M. Shaffer2, Samir D. Sharma1, Claude B. Sirlin3, Keitaro Sofue2,4, Nikolaus M. Szeverenyi3, Takeshi Yokoo5,6, Qing Yuan5, and Scott B. Reeder1,7
1Radiology, University of Wisconsin-Madison, Madison, WI, United States, 2Radiology, Duke University, Durham, NC, United States, 3Radiology, University of California, San Diego, San Diego, CA, United States, 4Radiology, Kobe University Graduate School of Medicine, Kobe, Hyogo, Japan,5Radiology, University of Texas Southwestern Medical Center, Dallas, TX, United States, 6Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, United States, 7Medical Physics, University of Wisconsin-Madison, Madison, WI, United States

Chemical shift-encoded techniques for MRI-based quantification of triglyceride concentration have shown great promise for assessment of nonalcoholic fatty liver disease. However, direct validation of these techniques in phantoms across multiple sites, vendors, platforms and field strengths has yet to be performed. In this work, we address this gap by scanning the same oil-water phantom sequentially at four sites and comparing proton density fat-fraction measurements obtained with a common reconstruction algorithm. Our results demonstrate excellent accuracy, robustness and reproducibility of PDFF measurements in a phantom across three vendors, four sites, two field strengths, and eight magnets.

11:27   Iron
Takeshi Yokoo, M.D., Ph.D.
11:42 0087.   Quantitative Ultra-short Echo Time Imaging for Massive Iron Overload Assessment: A Way to Make It Happen - permission withheld
Axel J. Krafft1,2, Ralf B. Loeffler1, Ruitian Song1, Mary E. McCarville1, Matthew D. Robson3, Jane S. Hankins4, and Claudia M. Hillenbrand1
1Radiological Sciences, St. Jude Children's Research Hospital, Memphis, TN, United States, 2Radiology - Medical Physics, University Medical Center Freiburg, German Cancer Consortium (DKTK), Heidelberg, Germany, 3Radcliffe Department of Medicine, University of Oxford, Oxford, United Kingdom,4Hematology, St. Jude Children's Research Hospital, Memphis, TN, United States

Iron overload requires accurate assessment of hepatic iron content (HIC) which can be achieved via R2*-MRI using multi-echo gradient echo (mGRE) imaging. The precision in previous R2*-HIC studies is compromised for high HIC (> 20 mg Fe/g) as conventional mGRE techniques fail due to the rapid T2* signal decay. This limitation could be overcome using 2D ultra-short echo time (UTE) imaging which is, however, sensitive to various system imperfections hampering quantitative MRI. We describe a 2D UTE sequence that allows for precise R2* quantification for detection of iron overload. The sequence was evaluated in phantoms and in vivo.

11:54 0088.   A T2* MRI Prospective Survey on Pancreatic Iron in Thalassemia Major Patients Treated with Deferasirox, Deferiprone and Desferrioxamine
Antonella Meloni1, Gennaro Restaino2, Stefania Renne3, Massimiliano Missere2, Maria Chiara Resta4, Vincenzo Positano1, Daniele De Marchi1, Gaetano Roccamo5, Nicola Romano6, Maria Giovanna Neri1, and Alessia Pepe1
1CMR Unit, Fondazione G. Monasterio CNR-Regione Toscana, Pisa, Italy, 2Dipartimento di Radiologia, Università Cattolica del Sacro Cuore, Campobasso, Italy, 3Struttura Complessa di Cardioradiologia-UTIC, P.O. “Giovanni Paolo II”, Lamezia Terme, Italy, 4Struttura Complessa di Radiologia, OSP. SS. Annunziata ASL Taranto, Taranto, Italy, 5Unità di Prevenzione e Cura delle Mictrocitemie, PO di S. Agata di Militello (ASP-ME), S. Agata di Militello (ME), Italy, 6S.C. Medicina Trasfusionale, AO Arcispedale "S. Maria Nuova", Reggio Emilia, Italy

For the first time, we described the changes in pancreatic T2* values over a follow-up (FU) of 18 months and we evaluated prospectively the effectiveness of the three iron chelators in monotherapy in thalassemia major (TM) patients. At the dosages used in the clinical practice all three chelators in monotherapy allowed a significant reduction in pancreatic iron.

12:09   MR Imaging of Liver Fibrosis
Laurent Huwart, M.D., Ph.D.
12:21 0089.   
Evaluation of Spin-echo based sequences for MR Elastography of Liver with Iron overload
Bogdan Dzyubak1, Yogesh K. Mariappan2, Kevin J. Glaser1, Sudhakar K. Venkatesh1, and Richard L. Ehman1
1Radiology, Mayo Clinic, Rochester, Minnesota, United States, 2Philips Healthcare, Bangalore, Karnataka, India

Magnetic Resonance Elastography (MRE) can measure the shear stiffness of soft tissues and is clinically used for hepatic fibrosis assessment. In liver pathologies resulting in high iron deposition(Ex: hemochromatosis), the typical gradient-echo MRE sequence can fail due to low SNR. To address this issue, two spin-echo based pulse sequences with short echo times have been developed. On patients with no iron overload, these modified sequences provide stiffness values that are in good agreement with the standard MRE, indicating that the current clinically used critical stiffness values can continue to be used for liver fibrosis assessment in patients with iron overload, when using these sequences as well.

12:33 0090.   Analysis of Clinical and Histopathological Changes That Influence Liver Stiffness measured by MR Elastography
Wen-Pei Wu1,2, Ran-Chou Chen2,3, Chen-Te Chou1, Chih-Wei Lee1, Cheng-In Hoi2, Yi-Chun Wang2,4, and Kwo-Whei Lee1
1Radiology, Chang-Hua Christian Hospital, Chang-Hua, Taiwan, Taiwan, 2Biomedical Imaging and Radiological Science, National Yang-Ming Medical University, Taiwan, Taiwan, 3Radiology, Taipei city Hospital, Taipei, Taipei, Taiwan, Taiwan, 4Taoyuan general hospital ministry of health and welfare, Taiwan, Taiwan

Purpose The purpose of our study was to determine the influence of the histological or clinical parameters on liver stiffness (LS) measured by MRE. Methods 354 patients underwent MRE and histological examination during a 3-month interval. Clinical, laboratory data and histological parameters were recorded. The spleen volume was also measured using semiautomatic software. The influence of the parameters on the LS was investigated through uni- and multivariate standard regression. Results The univariate regression analysis showed age, etiology, fibrosis, activity, iron deposition, AST, ALT, total bilirubin, platelet and spleen size were significantly correlated with LS. In multiple regression analysis, three variables independently influenced LS: fibrosis, activity, and spleen size. The parameters together explained 58.7% of the variances of the LS, with fibrosis making the most unique contribution (26.0%). Conclusions The degree of liver fibrosis , activity and spleen size independently and significantly exaggerated LS.

12:45   Adjournment & Meet the Teachers