Hepatobiliary 1

Monday 12 May 2014

Sonia Colombo Serra¹, Pernille Rose Jensen², Luigi Miragoli¹, Magnus Karlsson², Claudia Cabella¹, Luisa Poggi¹, Luca Venturi³, Fabio Tedoldi¹, and Mathilde Lerche^2
1Centro Ricerche Bracco, Bracco Imaging SpA, Colleretto Giacosa, Torino, Italy, ²Albeda Research, Copenhagen, Copenhagen, Denmark, ³CEIP, University of Torino, Colleretto Giacosa, Torino, Italy

Hepatocellular carcinoma (HCC) is most often found in cirrhotic livers and can be hardly diagnosed relying on anatomical information alone. Magnetic resonance is commonly used to identify anatomical lesions, but can also provide information on cellular metabolism by an emerging method that uses ¹³C labeled hyperpolarized molecules. This study aimed at probing the potential of hyperpolarized [1,3-¹³C₂]ethyl acetoacetate as a metabolic marker of HCC. A liver cancer implanted in rats is diagnosed due to a higher substrate-to-product ratio in tumors than in healthy livers demonstrating that the metabolism of the liver isoform of carboxyl esterase can be monitored in vivo.

Elhamy R Heba¹, Claude B Sirlin², Tanya Wolfson³, Anthony Gamst³, Rohit Loomba⁴, and Michael S Middleton^2
1Radiology, UCSD, san diego, CA, United States, ²Radiology, UCSD, San Diego, CA, United States, ³UCSD, San Diego, CA, United States, ⁴Internal medicine, UCSD, San Diego, CA, United States

This study was done to compare the agreement of different MRI-PDFF liver fat quantification analysis methodologies in 580 adults with known or suspected non-alcoholic fatty liver disease (NAFLD), using MRS-PDFF as reference standard. We found that 3-, 4-, 5- and 6-echo MRI accurately quantify hepatic PDFF in adults, with 3-echo MRI informally showing the closest and 2-echo MRI the worst agreement.

Masami Yoneyama¹, Masanobu Nakamura², Taro Takahara³, Thomas Kwee⁴, Yukihisa Takayama⁵, Akihiro Nishie⁵, Atsushi Takemura², Yasutomo Katsumata², Makoto Obara², Satoshi Tatsuno¹, and Seishi Sawano^1
1Yaesu Clinic, Tokyo, Japan, ²Philips Electronics Japan, Tokyo, Japan, ³Tokai University School of Engineering, Kanagawa, Japan, ⁴University Medical Center Utrecht, Netherlands, ⁵Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

This study introduced a novel approach to differentiate hypointense nodules on gadoxetic acid-enhanced hepatobiliary phase images by use of a non-balanced spin-echo SSFP (T2FFE) sequence. T2FFE typically provides T2-weighted contrast, but theoretically has high sensitivity to T1 shortening effects due to gadoxetic acid. This approach might particularly be useful for distinguishing metastases and hemangiomas.

Arthur Peter Wunderlich¹, Steffen Klömpken¹, Holger Cario², Markus Juchems¹, and Meinrad Beer^1
1Dept. for Diagnostic and Interventional Radiology, Univ.-Clinic Ulm, Ulm, Germany, ²Pediatry, Univ.-Clinic Ulm, Ulm, Germany

The liver of iron overloaded patients was scanned with breathhold multicontrast GRE sequences at 1.5 T and 3 T. To address high liver iron content (LIC) resulting in high R₂* values we used short echo spacing resulting in out-of-phase signal for fat and water. Therefore, liver fat fraction had to be included as free parameter for data fit. Four different flip angles lead to sufficient measurement data reliably above image noise at the cost of T₁ as additional fit parameter. Data were fit to theory by Levenberg-Marquardt algorithm. Dependence of R₂* on LIC was compared between field strengths.

Jia Ning¹, Sheng Xie², Lei Wang², Hua Guo¹, Xihai Zhao¹, Chun Yuan^1,3, and Huijun Chen^1
1Center for Biomedical Imaging Research & Department of Biomedical Engineering, Tsinghua University, Beijing, China, ²Department of Radiology, China-Japan Friendship Hospital, Beijing, China, ³Department of radiology, University of Washington, Seattle, WA, United States

Liver function is an important and sensitive indicator for liver damage. In this study, we proposed a new hemodynamic model to better describe the behavior of the Gd-EOB-DTPA by adding a directly measured contrast excretion term from hepatic vein. Validations with serum liver functional parameters were carried out for the proposed model and original model. Model parameters Ki is found significantly correlated with the concentration of preALB and ALB in serum. The proposed dual-input one output two-compartment pharmacokinetics model for dynamic Gd-EOB-DTPA enhanced liver MRI is capable of localized liver function evaluation in-vivo.fibrosis can envolve to cirrhosis and finally end up with hepatic carcinoma. Various studies have shown that fibrosis of level 1 and level 2 can be reversible [2], early diagnosis of liver fibrosis is critical for treatment and recovery. Dynamic contrast enhanced MR imaging is one of the most powerful protocols to evaluate the condition of liver. Gd-EOB-DTPA is a new type of hepatocyte specific contrast agent [3] for liver function evaluation. It has the advantages of specific binding with hepatocytes, and excretion through the hepatic vein, kidney and biliary. According to the properties of this contrast agent, we propose a new hemodynamics model to describe the perfusion of the liver. It is supposed that the parameters of the model can give an instruction of hepatic function and provide useful information for diagnosis of liver fibrosis.

Céline Giraudeau¹, Jean-Luc Daire¹, Matthieu Lagadec¹, Sabrina Doblas¹, Catherine Pastor², and Bernard Van Beers^1
1INSERM Centre de Recherche Biomédicale Bichat Beaujon,CRB3 U773, Université Paris Diderot, Sorbonne Paris Cité, Clichy, France, ²Laboratoire de Physiopathologie Hépatique et Imagerie Moléculaire,Hôpitaux Universitaires de Genève, Geneva, Switzerland

We have previously shown that the hepatic functional parameters (hepatic extraction fraction (HEF) and mean residence time (MRT)) obtained with deconvolution analysis at gadoxetate-enhanced MRI correlate with the expression of the oatp/mrp transporters in liver cirrhosis. The aim of this study was to assess if a multicompartmental model of hepatocytic transport gives further insight into the expression of the oatp/mrp transporters in liver cirrhosis. Both methods were evaluated in normal and cirrhotic rats through calculation of Akaike criteria and multiple regression analysis. The superiority of a multicompartmental relative to a deconvolution model remains here to be proven.

Debra E. Horng^1,2, Diego Hernando¹, and Scott B. Reeder^1,2
1Radiology, University of Wisconsin-Madison, Madison, WI, United States, ²Medical Physics, University of Wisconsin-Madison, Madison, WI, United States

Accurate fat quantification using chemical shift-encoded techniques requires correction for R2* (=1/T2*) decay. Correction for a common R2* for both fat and water (“single-R2*”) has been shown to be accurate for FF quantification in patients without iron overload. In this work, the accuracy of single-R2* correction is assessed in patients with iron overload. Fat-fraction and R2* were measured in 42 subjects at 1.5T and 3.0T, with both imaging and spectroscopy. The R2* of fat and water are very similar even in the presence of iron overload, showing that fat quantification with single-R2* correction is accurate even for patients with iron overload.

Takeshi Yokoo^1,2, Qing Yuan¹, Julien Senegas³, Andrea Wiethoff^2,4, and Ivan M Pedrosa^1,2
1Radiology, UT Southwestern Medical Center, Dallas, TX, United States, ²Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, TX, United States, ³Philips Research Laboratories, Hamburg, Germany, ⁴Philips Research North America, Briarcliff Manor, NY, United States

Patients with severe hepatic iron overload are at risk for developing end-stage liver disease and in greatest need for therapy. An R2*-based measure has been proposed as a surrogate for liver iron, but its estimation is challenging in severe iron overload due to rapid signal decay. In this series of simulation, phantom, and human studies, we compared R2* estimation performance of several existing methods: linear least squares, nonlinear least squares (NLS), weighted NLS, NLS with constant noise offset, and Rician-noise based. Our results show that Rician-noise based method is clinically feasible and may be necessary to accurately estimate R2* for severe iron overload.

Amelia Wnorowski¹, Flavius Guglielmo¹, Laurence Parker¹, Sandeep Deshmukh¹, Patrick O'Kane¹, Christopher Roth¹, and Donald Mitchell^1
1Radiology, Thomas Jefferson University Hospital, Philadelphia, PA, United States

Gadoxetate disodium is a useful MRI contrast agent for the diagnosis of hepatocellular carcinoma in cirrhosis. Its use is limited by poor hepatic and biliary enhancement in some patients. The purpose of this study was to identify predictors of poor enhancement, which was assessed in 91 cirrhotic patients and compared to various laboratory parameters, liver length and degree of ascites. A correlation matrix was performed, followed by multiple and logistic regressions. Several variables independently correlated with enhancement. However, the first variable chosen, either albumin or MELD score, accounted for almost all of the variance in the model.

Nathan Artz¹, William Haufe², Tanya Chavez², Gavin Hamilton², Michael Middleton², Jeff Schwimmer³, Diego Hernando¹, Ann Shimakawa⁴, Jonathan Hooker², Claude Sirlin², and Scott Reeder^1,5
1Radiology, University of Wisconsin, Madison, WI, United States, ²Radiology, University of California, San Diego, CA, United States, ³Pediatrics, University of California, San Diego, CA, United States, ⁴Global Applied Science Laboratory, GE Healthcare, Menlo Park, CA, United States, ⁵Medicine, University of Wisconsin, Madison, WI, United States

The purpose of this work is to examine the 1.5T vs 3T reproducibility of hepatic proton density fat fraction (PDFF) measurements. Obese patients were scanned using three distinct quantitative MR techniques on the same day at both field strengths. Regression analysis was used to compare co-localized 1.5T and 3T PDFF measurements. All three techniques demonstrated very good agreement between 1.5T and 3T liver PDFF measurements, indicating that PDFF quantification is reproducible across field strengths.