Dec 2, 2014 - discriminate normal vs. abnormal liver biopsy in candidates for live liver donation. Juan F. Gallegos-Orozco,. 1. Alvin C. Silva,. 2. Mashal J.
Abdominal Imaging
ª Springer Science+Business Media New York 2014 Published online: 2 December 2014
Abdom Imaging (2015) 40:795–802 DOI: 10.1007/s00261-014-0310-y
Magnetic resonance elastography can discriminate normal vs. abnormal liver biopsy in candidates for live liver donation Juan F. Gallegos-Orozco,1 Alvin C. Silva,2 Mashal J. Batheja,3 Yu-Hui Chang,4 Kathleen L. Hansen,5 Dora Lam-Himlin,6 Giovanni De Petris,6 Bashar A. Aqel,5 Thomas J. Byrne,5 Elizabeth J. Carey,5 David D. Douglas,5 David C. Mulligan,7 Annelise M. Silva,8 Jorge Rakela,5 Hugo E. Vargas5 1
Division of Gastroenterology, Hepatology and Nutrition, University of Utah School of Medicine, 30 North 1900 East, SOM 4R118, Salt Lake City, UT 84132, USA 2 Department of Radiology, Mayo Clinic Arizona, Phoenix, AZ, USA 3 Division of Gastroenterology, ‘‘Carl T. Hayden’’ VA Medical Center, Phoenix, AZ, USA 4 Department of Biostatistics, Mayo Clinic Arizona, Phoenix, AZ, USA 5 Division of Gastroenterology and Hepatology, Mayo Clinic Arizona, Phoenix, AZ, USA 6 Department of Laboratory Medicine and Pathology, Mayo Clinic Arizona, Phoenix, AZ, USA 7 Department of Surgery, Yale School of Medicine, New Haven, CT, USA 8 Arizona State University, Tempe, AZ, USA
Abstract Purpose: The aim of this study was to define liver shear stiffness by magnetic resonance elastography (MRE) that distinguishes normal from abnormal liver biopsy, especially when steatosis ‡20%, among potential live liver donors. Methods: Baseline clinical, laboratory, imaging, MRE, and liver biopsy results were recorded. Using MRE, hepatic shear stiffness in kilopascals (kPa) was measured and compared to liver biopsy. Comparison between groups was done using v2 or Fisher’s exact test for categorical variables and Wilcoxon test for continuous variables. Receiver operating characteristic (ROC) curve was calculated to assess diagnostic accuracy. Statistical significance was set at p < 0.05. Results: 38 healthy adults were included. Liver biopsy was normal in 27 and abnormal in 11. ROC curve for MRE defined optimal cutoff at 2.6 kPa (sensitivity 0.72, specificity 0.85, AUC 0.81) to distinguish these 2 groups. Hepatic steatosis ‡20% on biopsy is a contraindication for liver donation in our center. We evaluated the ability of MRE to distinguish this degree of steatosis: 8 persons had steatosis ‡20% and were excluded from donation. ROC curve for Correspondence to: Juan F. Gallegos-Orozco; email: juan.gallegos@hsc. utah.edu
MRE defined optimal cutoff at 2.82 kPa (sensitivity 0.88, specificity 1, AUC 0.98) to identify this group. Conclusions: Liver stiffness measured by MRE, even in the absence of liver fibrosis, can be useful in differentiating normal from abnormal liver histology, and most importantly in patients under evaluation for live liver donation, can very accurately distinguish those with complicated hepatic steatosis ‡20%, our cutoff for donation. In the future, MRE might provide supplementary information to make liver biopsy unnecessary in the donor evaluation process. Key words: Live liver donation—Elastography— Liver stiffness—Liver steatosis—Live donor liver transplantation
Every year in the United States, there are approximately 16,000 patients listed for liver transplantation, while approximately 6,500 transplants are performed yearly—this means there is a large gap between patients in need of a life-saving surgery and the number of patients that actually get the procedure each year. In an effort to increase the number of donors available, several strategies have been devised by the transplant community. One such strategy has
796
been live liver donation, in which the right lobe of the liver from a healthy adult is resected and transplanted into a patient with end-stage liver disease on the transplant list [1]. Despite being a relatively common procedure in Asia, adultto-adult living donation is less commonly performed in the United States. For example, from a total of 6,455 liver transplants performed in the US in 2013, only 252 were LDLT (4%) (http://optn.transplant.hrsa.gov consulted on 8/22/2014). However, living donor liver transplantation (LDLT) has become an integral part of our transplant program. Since 2001, 138 LDLT have been performed at our institution (mean 10 LDLT/year), accounting for 17% of the transplants performed since the inception of the program (138 of 808 liver transplants) (http://optn. transplant.hrsa.gov consulted on 8/22/2014). It is estimated that 3 potential donors need to be evaluated for every successful LDLT, which means that in our program we would have to evaluate 30–40 persons each year [2]. Due to the potential morbidity and small risk of mortality (~ 0.2%) associated with live liver donation, we are very careful in donor selection [3–5]. The selection process includes a rigorous medical, surgical, and psychosocial evaluation. If no contraindications are present, then a liver biopsy is performed as the final step of the process. Liver biopsy carries its own risk of complications to the potential donor [6, 7], and if a complication at the time of the biopsy occurred, it potentially could lead to delays in the liver transplant or even cancelation of the donation process altogether. Hence, a non-invasive evaluation of the liver among potential live donors would avoid the risk of a liver biopsy. As part of the donor evaluation, the person undergoes a detailed anatomic assessment of the liver which includes a combination of CT cholangiography and MR angiography, which provides the surgeon a comprehensive understanding of the arterial, venous, and biliary anatomic variants of the donor and allows for calculation of residual liver volume, all necessary elements to plan the LDLT. Magnetic resonance elastography (MRE) is an emerging technique to assess tissue stiffness [8, 9], and when applied to the liver, increased stiffness has not only been associated with fibrosis but also with inflammation such as is observed in non-alcoholic steatohepatitis [10–12]. In our center, we routinely perform a liver biopsy in all potential donors who have otherwise been deemed good candidates from a medical, surgical, and psychosocial standpoint and who do not have any anatomic abnormalities that would preclude donation. If the liver has any evidence of liver fibrosis or macrovesicular steatosis ‡20%, then that person is excluded from donation. The aim of the present study was to define liver shear stiffness by MRE and compare it to the liver biopsy results in healthy adults undergoing evaluation for living liver donation and assess how well hepatic shear stiffness distinguished normal vs. abnormal liver biopsy particularly as it pertains to hepatic steatosis ‡20%, our cutoff for donation.
J. F. Gallegos-Orozco et al.: MRE in live liver donors
Materials and methods The Mayo Clinic institutional review board approved this study, and written informed consent was obtained from all participants. We included healthy adults undergoing preoperative evaluation as live liver donors between January 2008 and January 2010. Relevant demographic, clinical, laboratory, imaging (CT and MRI), MRE, and liver biopsy results were collected prospectively as part of the evaluation process (MRE and liver biopsy were performed within 35 days of each other), and the outcomes for each potential donor were also recorded. Data collected included age, race, gender, body mass index (BMI), liver function panel (total protein, albumin, total bilirubin, alkaline phosphatase, AST, and ALT), INR, serum creatinine, and hepatitis B and C viral serologies.
MRE technique All exams were performed on a 1.5 T MR (Signa, GE Healthcare, Milwaukee, Wisconsin) with 8-channel surface coil. MRE requires additional hardware and software to the standard MR exam [13]. To create hepatic waves, a passive acoustic driver is positioned over the right upper quadrant and connected to an active driver outside the MR suite that generates low-frequency (60 Hz) vibrations (Fig. 1). These vibrations are imaged with a modified phase contrast MR sequence, which are subsequently processed by an inversion algorithm to produce a quantified map of hepatic stiffness (elastogram). For this study, two observers (with a combined MRE experience >4 years and blinded to clinical and pathology information) performed measurements within the regions of interest (ROI) on all liver donor elastogram images in consensus. Mean shear hepatic stiffness measurements in kilopascals (kPa) were obtained from 4 slice acquisitions through the liver, using a single ROI that encompassed the central area of a liver slice acquisition, avoiding larger vessels, areas of wave interference, artifacts, and left lateral segment (Fig. 2).
Liver histopathology All specimens obtained from the percutaneous liver biopsies were processed at the pathology laboratory at the Mayo Clinic in Arizona. Standard techniques for hematoxylin and eosin and Masson trichrome staining were applied to assess fibrosis. Fibrosis staging was assessed using classification systems developed for the specific liver disease etiologies including the METAVIR system [14] or the Brunt et al. classification for nonalcoholic fatty liver disease [15]. Two experienced liver pathologists who were blinded to clinical information and MRE results interpreted each biopsy.
797
J. F. Gallegos-Orozco et al.: MRE in live liver donors
Fig. 1.
Magnetic resonance elastography (MRE) set-up. Reprinted from [10].
inflammation), or according to the extent of macrovesicular steatosis present (
