Transplant Outcomes and Immunosuppression


1666

To transplant or not to transplant - Lessons learned from 20 years global collaboration in liver transplantation for hereditary transthyretin amyloidosis

Bo-Göran Ericzon1, Henryk E. Wilczek1, Marie Larsson1, Arie J. Stangou3, Priyantha Wijayatunga2, Ole Suhr2;

1Transplantation Surgery, Karolinska University Hospital, Stockholm, Sweden; 2Department of Medicine, Umeå university Hospital, Umeå, Sweden; 3NHS Amyloid Treatment Programme, Queen Elizabeth Hospital, Birmingham, United Kingdom

Until recently, liver transplantation (Ltx) was the only available treatment for familial amyloidotic polyneuropathy (FAP), but in the last years several pharmaco-therapeutic approaches have emerged hoping to halt disease progression. In the present study, Ltx as the golden standard of treatment is evaluated in a 20 years perspective by analysis of the FAP world transplant registry (FAPWTR). From April 1990 until December 2010, data from 78 liver transplant centers in 19 countries have been accumulated. Approximately 125 liver transplants are performed yearly worldwide. The Registry holds a total of 1940 patients undergoing 2127 Ltx. 561 patient deaths were reported. Eighty-eight Ltx were performed in combination with a heart or a kidney. Patients undergoing combined Ltx were generally older than those only subjected to Ltx and with a non-Val30Met mutation. The overall 20 year survival after tx, all mutations included, was 55.3%. Expected mortality rate decreased on average by approximately 4% per year between 1 990 and 201 0. In a multivariate analysis modified body mass index (mBMI), early onset of disease, disease duration before Ltx and Val30Met versus non Val30Met were independent significant factors for survival after Ltx. Survival of patients with onset of disease after age 50 was significantly reduced when compared to early onset patients, and most pronounced in the male subgroup. Thus, expected mortality rate in late onset male patients was 137% of that of late onset female patient mortality (p<0.05). Early onset patients (all mutations) had an expected mortality rate of 38% of that of the late onset group (p<0.01). Furthermore, Val30Met patients had an expected mortality rate of 61% of that of non Val30Met patients (p<0.01). With each year of increase in duration of disease before Ltx, the expected mortality increased by 1 1% (p<0.01). Overall, with each unit of increase in mBMI at Ltx, the expected mortality decreased by 0.12% (p<0.01). Twenty-two percent of the death causes were related to cardiovascular disease, thus significantly more common than usually seen in Ltx for end stage liver disease. Conclusion: Long term survival after Ltx for FAP is excellent. A good nutritional status, short duration of disease at the time of Ltx and early onset of disease were significant independent factors for survival. Val30Met patients had significantly better outcome when compared to non-Val30Met patients. The risk of delaying Ltx by testing alternative treatments needs consideration.

Disclosures:

The following people have nothing to disclose: Bo-Göran Ericzon, Henryk E. Wilczek, Marie Larsson, Arie J. Stangou, Priyantha Wijayatunga, Ole Suhr

1667

Comparative Effectiveness Analysis of Transplant Strategies For Liver Transplant Candidates With Renal Impairment

Yaojen Chang1, Lorenzo Gallon2, Colleen L. Jay2, Bing Ho2, Kirti Shetty1, Josh Levitsky2, Talia B. Baker2, Daniela P. Ladner2, John J. Friedewald2, Michael M. Abecassis2, Gordon Hazen2, Anton I. Skaro2;

1Georgetown University, Washington, DC; 2Northwestern University, Chicago, IL

BACKGROUND: The aim of this study was to assess outcomes of three different transplant strategies for end stage liver disease (ESLD) patients receiving pre-transplant renal replacement therapy (RRT) for acute kidney injury. The transplant strategies investigated in the simulation model included (1) liver transplant alone (LTA), (2) simultaneous liver kidney (SLK) transplant, and (3) LTA followed by subsequent kidney transplant alone (KTA) for recipients who do not recover native renal function. METHODS: This study was performed under implementation of available transplant strategies within a Markov discrete-time state transition model. In each simulation trial, a patient randomly traverses through the decision model to experience events (e.g. native renal recovery) and move between different health states until death or the end of the simulation period. RESULTS: Among three transplant strategies for ESLD patients receiving pre-transplant RRT for 30 days or less in the base case model, the SLK transplant recipients had better survival rates than the LTA recipients with/without subsequent KTA (83% vs. 77% vs. 74% at the first post transplant year), when using standard criteria donor (SCD) kidneys and where the waiting time was the same. The survival of LTA recipients eligible for KTA under the current kidney allocation policy and SLK recipients using SCD kidneys stratified by duration of pre-transplant RRT (< 30 days, 30–59 days, 60–89 days, and > 90 days) and MELD score group (21–30 and > 30) showed that survival advantage at the first post transplant year of SLK transplants using SCD kidneys over LTA in the MELD score 21–30 and > 30 categories increased from 7% to 28% and from 7% to 30%, respectively, when the pre-transplant RRT duration was extended from 30 days to 120 days. Given more selective behavior in SLK transplant using an extended criteria donor (ECD) or SCD kidney, we respectively adopted 50% and 200% of national median waiting time between 2003–2007 for MELD > 30 for ECD and SCD kidney in the model. We found that the survival rates of SLK transplant candidates using ECD kidneys exceeded those using SCD kidneys (72% vs. 61% at 1 year). CONCLUSIONS: A survival advantage for SLK transplant compared to LTA for ESLD patients undergoing pre-transplant RRT is clearly demonstrated. As the pre-transplant RRT duration was extended, the survival advantage from SLK transplant increased. Longer waiting times associated with SLK transplant could compromise outcomes particularly in the higher MELD strata where waitlist mortality occurs in excess of the survival differential between use of a SCD and ECD kidney.

Disclosures:

Kirti Shetty - Grant/Research Support: Ikaria, Novartis, Onyx-Bayer, Hyperion; Speaking and Teaching: Merck-Schering Plough, Salix, Gilead, Onyx

Josh Levitsky - Grant/Research Support: Salix, Novartis; Speaking and Teaching: Gilead, Salix, Novartis

Michael M. Abecassis - Management Position: Transplant Genomics, Inc

The following people have nothing to disclose: Yaojen Chang, Lorenzo Gallon, Colleen L. Jay, Bing Ho, Talia B. Baker, Daniela P. Ladner, John J. Friedewald, Gordon Hazen, Anton I. Skaro

1668

MicroRNA 125b is associated with hepatocellular carcinoma recurrence after liver transplantation

Claudio De Vito1, Christophe Bourdeaux4, Delucinge-Vivier Céline3, Christian Toso2, Jan P. Lerut4, Christine Sempoux5, Gilles Mentha2, Laura Rubbia-Brandt1;

1Division of Clinical Pathology, Geneva University Hospitals, Geneva, Switzerland; 2Department of surgery, Geneva University Hospitals, Geneva, Switzerland; 3Genomics Platform, University of Geneva, Geneva, Switzerland; 4Department of Abdominal and Transplantation Surgery, Cliniques Universitaires Saint-Luc, Brussels, Belgium; 5Department of Pathology, Cliniques Universitaires Saint-Luc, Brussels, Belgium

Introduction: Liver transplantation (LT) using Milan criteria (MC) provides the best option of cure for cirrhotic patients with hepatocellular carcinoma (HCC). Despite accurate selection 10–15% of patients develop HCC recurrence after LT supporting the identification of new markers to better select transplant candidates. MicroRNAs (miRNAs) are small non-coding RNA that regulate gene expression at the post-transcriptional level and are involved in many liver diseases including HCC. The aim of this study is to identify a miRNA signature related to HCC recurrence after LT. Methods: From 1 998 to 201 0, 54 patients from two liver transplant centers (Geneva and Brussels) were retrospectively included in this study. 27 patients who developped HCC recurrence after LT and 27 matched patients for age, gender, number and size of tumors, MC and year of LT were selected. MiRNA expression profile was performed using Nanostring technology and selected miRNAs were validated by qRT-PCR. Results: 69 miRNAs were statistically differentially expressed between the two groups among which 48 and 21 were respectively up-regulated and down-regulated in recurrent HCC compared to non recurrent HCC. Mir-125b, a known tumor suppressor miRNA in HCC, was identified to be down-regulated in recurrent HCC and validated by qRT-PCR. Moreover preliminary data suggests that mir-125b could discriminate early from late recurrent HCC with a lower expression in early recurrent HCC. Conclusion: In this study we identified that mir-125b is associated with HCC recurrence and could discriminate early from late recurrence. An assessment of mir-125b in HCC prior to LT might provide an additional marker for a better selection of liver transplant candidates and to reduce HCC recurrence.

Disclosures:

Jan P. Lerut - Grant/Research Support: Fresenius Biotech, Astellas

The following people have nothing to disclose: Claudio De Vito, Christophe Bourdeaux, Delucinge-Vivier Céline, Christian Toso, Christine Sempoux, Gilles Mentha, Laura Rubbia-Brandt

1669

Complement Binding Donor Specific Antibodies are Associated with Late Graft Dysfunction Following Pedi-atric Liver Transplantation

Laura J. Wozniak1, Michelle J. Hickey2, Tiffany Smith2, Gertrudes Aguas2, Giovanni Lopez2, Yael Korin2, Robert S. Venick1, Sue V. McDiarmid1, Elaine F. Reed2;

1Pediatric Gastroenterology, UCLA, Los Angeles, CA; 2Pathology and Laboratory Medicine, UCLA, Los Angeles, CA

Background: Increasing evidence suggests that donor-specific HLA antibodies (DSA) following liver transplantation (LTx) are associated with graft dysfunction. Complement binding DSA have been shown to be more pathogenic in kidney and heart transplantation. Our aim was to assess the impact of complement binding DSA on graft outcomes in long-term pediatric LTx survivors. We hypothesized that complement binding DSA would be associated with de-novo autoimmune hepatitis (d-AIH) and/or episodes of late acute or chronic rejection. Methods: Serum was collected at routine clinic visits from 50 pediatric LTx recipients a mean of 1 1.8±5.3 years post-LTx. Samples were blinded and antibody detection against HLA-A, B, DR, and DQ was performed using single-HLA-antigen-coated beads (SAB) and Luminex multiplex technology (LABScreen; One Lambda). For patients with detectable HLA antibodies, donor-recipient mismatched HLAs were compared, and +DSA were defined as a normalized value >1000 MFI. Serum from patients with +DSA was also tested for complement binding using commercially available SAB-C1q kits (C1qScreen; One Lambda). C1q+DSA were defined as a normalized value >1000 MFI. Descriptive statistics and multivariate logistic regression were performed with results reported as means and odds ratios (OR). Results: +DSA were detected in 28/50 (56%) of patients, with the majority directed against HLA Class II DR or DQ loci (40% and 52%, respectively). Patients with +DSA had a mean of 1.8 (range 1–4) DSA with a mean MFI of 12004±8948. Patients with +DSA were younger at LTx (2.4±2.9 vs 5.4±5.4 years, p=0.01). There were no differences in other clinical variables including sex, primary diagnosis, history of re-LTx, and early acute rejection. C1q+DSA were detected in 14/28 (50%) of patients with +DSA. C1q+DSA were only detected in patients with +DSA MFI of > 10000. All C1q+DSA were directed against Class II loci (18% DR, 82% DQ). C1q+DSA had a mean MFI of 16468±5288. C1q+DSA was the only statistically significant multivariate predictor of d-AIH (OR 7.3, p=0.04) and late acute rejection (OR 8, p=0.03). Patients with C1q+DSA also showed a trend towards higher mean total bilirubin (2.4±5.2 vs 0.7±0.4 mg/dL, p=0.1) and a greater number of late rejection episodes (0.8±1 vs 0.4±0.8, p=0.08). Two of 3 patients with chronic rejection were noted to have C1q+DSA. Conclusion: The results of this study support our hypothesis that complement binding DSA are associated with late graft dysfunction following pediatric LTx. Class II DSA, and in particular DQ-DSA, are more common in patients with late graft dysfunction suggesting that the ability of DSA to fix complement is clinically relevant.

Disclosures:

The following people have nothing to disclose: Laura J. Wozniak, Michelle J. Hickey, Tiffany Smith, Gertrudes Aguas, Giovanni Lopez, Yael Korin, Robert S. Venick, Sue V. McDiarmid, Elaine F. Reed

1670

Excess mortality beyond 1 year after liver transplantation - a Nordic multicenter population-based study

Fredrik Åberg 1,2, Mika Gissler3,4, Tom H. Karlsen5, Bo-Göran Eric-zon6, Aksel Foss7, Allan Rasmussen8, William Bennet9, Michael Olausson9, Helena Isoniemi1;

1Transplantation and Liver Surgery Clinic, Helsinki University Hospital, Helsinki, Finland; 2Deparment of Gastroenterology, Helsinki University Hospital, Helsinki, Finland; 3Information Department, National Institute for Health and Welfare, Helsinki, Finland; 4Nordic School of Public Health, Gothenburg, Sweden; 5Department of Transplantation Medicine, Division of Cancer, Surgery and Transplantation, Oslo University Hospital, Rikshospitalet, Oslo, Norway; 6Division of Transplantation Surgery, CLINTEC, Karolinska University Hospital, Huddinge, Sweden; 7Sec-tion for Transplantation Surgery, Department of Transplantation Medicine, Division of Cancer, Surgery and Transplantation, Oslo University Hospital, Rikshospitalet, Oslo, Norway; 8Department of Surgical Gastroenterology and Liver Transplantation, Rigshospitalet University of Copenhagen, Copenhagen, Denmark; 9Department of Surgery, University of Gothenburg, Sahlgrenska University Hospital, Gothenburg, Sweden

With growing number of liver transplantations (LTs) performed and with nearly 90% nowadays surviving 1 year post-transplant, a major challenge is shifting towards improving survival beyond the first year. Understanding the causes for premature death is a prerequisite for improving long-term outcome. Overall and cause-specific mortality rates among all LT patients in the Nordic countries transplanted in 1 985–2009 and who survived 1 year post-operatively were divided by expected rates in the general population adjusted for age, sex, calendar time, and country to yield standardised mortality ratios (SMRs). Patient data came from the Nordic LT Registry, and general population data from national cause-of-death statistics. Among 3299 patients (23 691 person-years of follow-up, mean 8.2 yrs, SD 0.6 yrs), 763 deaths occurred. Patient survival >1 year remained unchanged over time and a considerable gap of 20% at 10 years was noted to the expected survival in the matched general population. SMR >1 year post-LT was 2.4 (95% CI 2.2–2.6), and SMR for death before age 75 was 5.8 (95% CI 5.4–6.3). SMR was higher for children than adults. Significantly elevated cause-specific SMRs were noted for infection (SMR 76, 95% CI 59–96), malignancy (SMR 5.1, 4.4–5.8), liver disease (SMR 12, 8–16), genitourinary disease (SMR 35, 24–50), and suicide and intentional self-harm (SMR 3.0, 1.6–5.0), but not for cardiovascular disease (SMR 2.2, 0.7–5.0). According to primary LT indication, all-cause SMR for death before age 75 was highest among patients transplanted due to hepatocellular carcinoma (HCC) (SMR 33, 95% CI 26–41), followed by alcoholic liver disease (ALD) (SMR 18, 14–22) and hepatitis C (HCV) (SMR 12, 10–16). Indication-based variation in cause-specific SMRs emerged, with acute liver failure and HCV showing the highest SMRs for infection, HCC for malignancy, and ALD for liver disease (mostly disease recurrence). Although cardiovascular disease presented the second most common cause of death overall, SMR for cardiovascular disease was significantly elevated only in primary biliary cirrhosis, ALD and patients aged 45–59 at LT. Transplant-specific causes of death such as rejection, biliary complications and death within 3 months after re-transplantation constituted 16% of all late deaths. Cause of death was unknown in 5%. In conclusion, mortality >1 year after LT is considerable higher than in the general population and has not improved overtime. The high relative mortality from infections, genitourinary disease, liver disease, suicide, and malignancies merits increased attention in future research and clinical patient follow-up.

Disclosures:

The following people have nothing to disclose: Fredrik Åberg, Mika Gissler, Tom H. Karlsen, Bo-Göran Ericzon, Aksel Foss, Allan Rasmussen, William Bennet, Michael Olausson, Helena Isoniemi

1671

Antibody Mediated Rejection As a Significant Contributor to Previously Unexplained Early Liver Allograft Loss

Jacqueline G. O'Leary1, Hugo Kaneku2,5, Anthony J. Demetris3, John D. Marr1, S. M. Shiller4, Brian M. Susskind1, Glenn W. Tilery4, Paul I. Terasaki5, Goran Klintmalm1;

1Annette C & Harold C SImmons Transplant Institute, Baylor University Medical Center, Dallas, TX;2University of California, Los Angeles, Los Angeles, CA; 3Pathology, University of Pittsburgh, Pittsburgh, PA; 4Pathology, Baylor University Medical Center, Dallas, TX; 5Terasaki Foundation Laboratory, Los Angeles, CA

Aim: To determine the prevalence of early liver allograft loss [defined as death or retransplantation because of a non-functioning graft <90 days after liver transplant (LT)] caused or contributed to by preformed donor specific HLA antibodies (DSA). Methods: From 1/1989 to 7/2010 337 patients experienced graft loss within 90 days of ABO compatible LT. Patients were excluded for the following reasons: lack of a pre-LT serum sample (n=51), death with a functioning graft (n=86), known cause of graft loss or death (n=123), and lack of biopsies (n=1 7). The remaining 60 patients experienced unexplained graft loss and were included in the study. A pre-transplant serum sample was analyzed for DSA, and post-reperfusion and follow-up tissue specimens were re-reviewed (initially blinded to DSA results) to determine whether antibody-mediated rejection (AMR) contributed to liver allograft loss using relatively stringent criteria that included: 1) DSA in serum 2) histopathological evidence of diffuse microvascular endothelial cell injury; 3) strong C4d pos-itivity in at least one of the tissue samples; and 4) reasonable exclusion of other causes of injury that might cause similar findings. Results: 33% of patients with a pre-transplant DSA pattern of bead saturation after serial dilutions (1:81) developed AMR, and 1 00% of patients with diffuse microvascular endothelial cell C4d staining on post-reperfusion biopsy developed AMR. One additional multiparous female developed, what appeared to be, a strong "recall" response resulting in combined AMR and acute cellular rejection (ACR) causing graft failure. Definite evidence of AMR was not detected in the remaining sensitized patients, but a contribution of AMR to allograft failure was possible (at least two but not all criteria were fulfilled) in three additional patients, 2 of whom also had at least one pretransplant DSA with bead saturation after serial dilutions. In conclusion, liver allograft recipients with high mean fluorescence intensity (MFI) preformed DSA despite serial dilutions had a 33% change of definite AMR and an additional 33% change of possible AMR often in combination with ACR leading to graft loss. In addition, 100% of patients that showed strong and diffuse microvascular C4d staining on post-reperfusion biopsy experienced definite AMR. These new criteria may be used to identify patients at highest risk for DSA mediated liver allograft injury and loss.

Disclosures:

Jacqueline G. O'Leary - Consulting: Vertex, Gilead

Anthony J. Demetris - Consulting: Abbvie, DCL/Novartis, Omnyx

Goran Klintmalm - Advisory Committees or Review Panels: Novartis; Grant/Research Support: Astellas, Novartis, Pfizer, Opsona, Quark

The following people have nothing to disclose: Hugo Kaneku, John D. Marr, S. M. Shiller, Brian M. Susskind, Glenn W. Tilery, Paul I. Terasaki

1672

Influence of the cytochrome P450 3A5*3 genotype in a graft liver on the occurrence of acute cellular rejection after living-donor liver transplantation

Miwa Uesugi1, Satohiro Masuda1, Haruka Shinke1, Yuki Nish-ioka1, Kazuo Matsubara1, Yasuhiro Fujimoto2, Toshimi Kaido2, Shinji Uemoto2;

1Department of Clinical Pharmacology and Therapeutics, Kyoto University Hospital, Kyoto, Japan; 2Divisions of Hepato-Pancreato-Biliary, Transplant and Pediatric Surgery, Department of Surgery, Kyoto University Hospital, Kyoto, Japan

[Background and Aim] Tacrolimus, a widely used immunosup-pressive agent after liver transplantation, is mainly metabolized by cytochrome P450 (CYP) 3A4 and 3A5 in both the liver and the intestine. In liver transplantation, it is important to suppress the onset of acute cellular rejection to avoid graft loss. In this study, we aimed to clarify the association between the CYP3A5*3 genotype and the occurrence of acute cellular rejection in recipients of living-donor liver transplantation (LDLT). [Methods] Adult and pediatric Japanese patients receiving de novo LDLT (n = 286) were enrolled in this study after providing written informed consent. Cases of ABO incompatibility were excluded. We retrospectively analyzed the trough concentration/daily dose (C/D) ratio of tacrolimus, CYP3A5*3 genotype in both donors and recipients, and acute cellular rejection after LDLT. This study was conducted in accordance with the Declaration of Helsinki and its amendments and was approved by Kyoto University Graduate School and the Faculty of Medicine Ethics Committee. [Results] The allele frequency of the CYP3A5*1 genotype was 0.207 in the graft liver and 0.219 in the native intestine. Patients with the CYP3A5*3/*3 genotype (CYP3A5 defect) in the native intestine rather than in the graft liver had an approximately 2-fold higher median C/D ratio of tacrolimus compared to patients with the CYP3A5 expresser (CYP3A5*1/*1 and *1/*3) for 5 weeks after surgery (P < 0.001, Mann-Whitney U test). Patients grafted with the CYP3A5*1 genotype-carrying liver showed a significantly higher frequency of acute cellular rejection between postoperative days 14 and 26 after LDLT than those with the CYP3A5*3/*3 genotype (15.5% vs. 8.1%, P = 0.049, Log-rank test). However, the CYP3A5 genotype in the native intestine was not associated with the occurrence of acute rejection. These results suggest that the CYP3A5 genotype in the graft liver affects the hepatic concentration of tacrolimus and its immunosuppressive effect in recipients of LDLT. [Conclusion] The CYP3A5 genotype in the graft liver could be a marker to predict the risk of post-transplantation acute cellular rejection.

Disclosures:

The following people have nothing to disclose: Miwa Uesugi, Satohiro Masuda, Haruka Shinke, Yuki Nishioka, Kazuo Matsubara, Yasuhiro Fujimoto, Toshimi Kaido, Shinji Uemoto

1673

Primary Sclerosing Cholangitis Recurs Earlier After Living Donor Liver Transplantation

Mansour G. Alghanem1, Andrew L. Mason1, Vincent G. Bain1, Mang M. Ma1, Norman Kneteman2, Kelly W. Burak3, AldoJ. Mon-tano-Loza1;

1Division of Gastroenterology & Liver Unit, University of Alberta, Edmonton, AB, Canada; 2Department of Surgery, University of Alberta, Edmonton, AB, Canada; 3Liver Unit, University of Calgary, Calgary, AB, Canada

Background/Aims: The natural history of patients with primary sclerosing cholangitis (PSC) after liver transplant may be complicated by recurrence of the disease. Outcomes after liver transplantation in patients with PSC could be different according to the type of transplant. This study analyses the outcomes and evidence of recurrent PSC (rPSC) after liver transplantation in patients with living-related donor compared to cadaveric donor transplants. Methods: We retrospectively evaluated 90 patients who received a liver transplant secondary to PSC at the University of Alberta Hospital. rPSC was defined according to Mayo clinic criteria. Patients were followed until graft loss, death or time of last visit. Results: Seventy-three were males (81%), and mean age at liver transplant was 44±1 years (range, 8–72 years), and 46 patients had inflammatory bowel disease at the time of the transplant (37 ulcerative colitis and 9 Crohn's disease). Seventy patients (78%) received a cadaveric transplant, and 20 (22%) received a living-related transplant. Twenty-five patients (28%) developed rPSC during follow-up. Overall median time for rPSC was 181 ±1 1 months (95% CI, 159–204), and probability of rPSC was 8% at 5-years, and 21% at 10-years. Median time for recurrence in living-related was shorter when compared with cadaveric transplant (90±3 vs. 1 87±8 months), and the probability of rPSC at 10-years was 15% in cadaveric transplants vs. 62% in living-related transplants (P=0.009). By Cox regression analysis having a living related-transplant was associated with higher risk of rPSC (HR, 4.54; 95% CI 1.31–15.73; P=0.02). There was a statistical trend towards lower median graft survival in living-related vs. cadaveric donor transplants (79±14 vs. 191±1 1 months, P=0.09), and there were no significant differences in overall median survival (87±13 vs. 249±38 months, P=0.2). Conclusions: Patients that received a living-related transplant have a higher risk of rPSC, when compared to cadaveric transplant recipients. Recurrent PSC may cause impairment in graft survival; however, overall patient survival is comparable.

Disclosures:

Andrew L. Mason - Grant/Research Support: Abbott, Gilead

Vincent G. Bain - Advisory Committees or Review Panels: Astellas, Novartis, Merck, Astellas, Boehringer Ingelheim

Kelly W. Burak - Advisory Committees or Review Panels: Gilead, Gilead, Gilead, Gilead, Janseen; Grant/Research Support: Bayer, Bristol Myers Squibb, Genen-tech, Bayer, Bristol Myers Squibb, Genentech, Bayer, Bristol Myers Squibb, Genentech, Bayer, Bristol Myers Squibb, Genentech, Boehrihnger Ingelheim; Speaking and Teaching: Gilead, Astellas, Merck, Roche, Gilead, Astellas, Merck, Roche, Gilead, Astellas, Merck, Roche, Gilead, Astellas, Merck, Roche

The following people have nothing to disclose: Mansour G. Alghanem, Mang M. Ma, Norman Kneteman, AldoJ. Montano-Loza

1674

Risk Factors of Neurologic Events after Liver Transplantation: Towards a Prognostic Risk Score Assessment

Federico Piñero1, Manuel Mendizabal1, Diego T. Arufe1, Rodolfo E. Quiros2, Victoria Marquevich3, Maria P. Raff a1, Ariel Gonzalez Campaña1, Mariano Barreiro1, Martín Fauda1, Oscar C. Andri-ani1, Luis G. Podesta1, Marcelo O. Silva1;

1Liver Unit, Hospital Uni-versitario Austral, Pilar, Argentina; 2Infectious Diseases, Hospital Universitario Austral, Pilar, Argentina; 3Critical Care Unit, Hospital Universitario Austral, Pilar, Argentina

Background: Major neurologic events (MNE) are common after liver transplantation (LT). Controversy exists regarding risk factors related to these events. Aim: To identify risk factors related to MNE after LT and propose a prognostic score. Methods: 262 consecutively transplanted adult patients in our LT unit (12/2001–12/2012) were included. MNE was defined as: impaired consciousness (coma or stupor), seizures, focal neurologic deficit (excluding tumors), visual impairment or blurred speech. Univariate and stepwise multivariate logistic regression analysis (Wald test) were performed (OR, 95% CI, P<0.05). From corresponding OR a prognostic score was made (0–5 points). To assess our model's prognostic accuracy and discrimination power we performed a ROC curve analysis (AUROC) and Hosmer-Lemeshow test. Survival analysis was made with Kaplan-Meier. Results: A total of 38 MNE events in 32 patients were recorded during the study period (stupor n=5, coma n=3, focal neurologic signs n=8, visual or blurred speech n=5 and seizures n=17) with a cumulative incidence of 12.2%. Permanent neurological impairment was present in 15 patients (46.9%). In univariate analysis, factors associated with development of MNE were: presence of pre-LT moderate-severe ascites (P=0.002), pre-LT creatinine (P=0.048), Living donor LT (P=0.05), post-LT hypomagnesemia (P=0.001), delta Sodium ≧12 mEq/l (P=0.032), cold ischemia time >8 hs (P=0.06), use of fluconazole (P=0.023) and immunosuppression with Tacrolimus (P=0.05). Independent variables related to MNE and corresponding assigned points were pre-LT ascites OR 3.22 (1.36–7.60; P=0.007) 2 points, delta sodium ≧12 mEq/l OR 2.76 (0.94–8.53; P=0.077) 1 point and hypomagnesemia OR 3.71 (1.64–7.60; P=0.002) 2 points. This score had a Hosmer-Lemeshow test of P=0.968 and AUROC of 0.74 (0.64–0.83: P<0.0001) with a best cut-off >3 points: 99% Sensibility and 90% Specificity for predicting a MNE. Overall patient survival rate was lower in those who presented MNE (62.5% vs 77.4%; P= 0.066), mean follow-up of 4.5 years. Conclusions: The Predicting MNE Score identifies patients at higher risk of neurologic events following LT; neurotoxic drugs (calcineurin inhibitors, fluconazole) in these high-risk patients should cautiously use. Larger studies are needed to further confirm these clinical results.

Disclosures:

Marcelo O. Silva - Advisory Committees or Review Panels: Schering Plough, Vertex, Abbott; Board Membership: BMS; Grant/Research Support: Schering Plough, BMS; Speaking and Teaching: Schering Plough, BMS, MSD

The following people have nothing to disclose: Federico Piñero, Manuel Mendiz-abal, Diego T. Arufe, Rodolfo E. Quiros, Victoria Marquevich, Maria P. Raffa, Ariel Gonzalez Campaña, Mariano Barreiro, Martín Fauda, Oscar C. Andriani, Luis G. Podesta

1675

Impact of Donor Quality and Early Allograft Dysfunction on Post-Liver Transplant Kidney Outcomes

Ranjeeta Bahirwani1, D. A. Herbst1, Kimberly A. Forde1, Orapin Tanapanpanit1, Anthony Guzman2, Peter L. Abt2;

1Gastroenterol-ogy, University of Pennsylvania, Philadelpia, PA; 2Surgery, University of Pennsylvania, Philadelphia, PA

Attempts at increasing the donor pool due to limited organ supply have led to the utilization of higher risk donors for liver transplantation; higher Donor Risk Index (DRI) is associated with Early Allograft Dysfunction (EAD). The aim of this study was to assess the impact of donor organ quality and early graft dysfunction on the development of post-transplant acute kidney injury (AKI) as well as chronic kidney disease (CKD). METHODS: A retrospective analysis of 544 liver transplant recipients at the University of Pennsylvania from 2/2/2005 to 6/6/2010 was performed to determine the incidence of post-transplant AKI and CKD and the impact of DRI and EAD on these outcomes. RESULTS: Mean patient age was 53.2 years (±10.9); 73.2% patients were male, 73.1% Caucasian, and 54.4% had Hepatitis C as the etiology of liver disease. Median MELD score at transplant was 25 (22, 30) and mean serum creatinine was 1.64 mg/dl (±1.2). Seventeen patients (3.1%) required pre-transplant renal replacement therapy (RRT) for a median of 1 3 days. Median DRI was 1.5 (1.2, 1.8), 30% patients had DRI > 1.7. Early Allograft Dysfunction (defined as serum bilirubin > 10 mg/dL, INR > 1.6, or AST/ALT > 2000 IU/L within the first 7 days post-transplant) occurred in 154 patients (28.3%) and was predicted by a DRI > 1.7 (OR 1.60, p 0.02). AKI (defined by the AKIN criteria) occurred in 67.7% of transplant recipients; 14% had stage 3 AKI (creatinine increase of 300% or > 4 mg/dl with an acute rise of at least 0.5 mg/dl). Fifty-two patients (9.6%) required RRT acutely post-transplant for a median of 7 days. Median post-transplant estimated glomerular filtration rates (eGFR) at 1, 3 and 5 years were 56, 60, and 61 ml/min respectively; 27.5% of patients developed stage 4 CKD (eGFR < 30 ml/min), 1 8.3% achieved an eGFR < 20 ml/min, and 7.7% required chronic post-transplant RRT for end-stage renal disease (ESRD). EAD was predictive of post transplant AKI (OR 1.63, p 0.02) as well as ESRD (OR 1.9, p 0.05). DRI was not predictive of post- transplant AKI (p 0.32) or CKD (p 0.48). CONCLUSIONS: Early Allograft Dysfunction is a predictor of post-transplant acute kidney injury as well as end stage renal disease. Poor donor quality is associated with early allograft dysfunction but does not predict acute or chronic kidney disease post-transplant.

Disclosures:

The following people have nothing to disclose: Ranjeeta Bahirwani, D. A. Herbst, Kimberly A. Forde, Orapin Tanapanpanit, Anthony Guzman, Peter L. Abt

1676

MicroRNA expression profiling to determine quality of Donation after Cardiac Death (DCD) Liver Grafts

Shirin E. Khorsandi, Siamak Salehi, Wayel Jassem, Hector Vilca-Melendez, Andreas Prachalias, Parthi Srinivasan, Nigel Heaton;

Institute of Liver Studies, King's College Hospital, London, United Kingdom

Purpose DCD liver grafts are marginal organs their use is associated with a higher risk of primary non function (PNF) or initial poor graft function (IPGF). To give insight into pathophysiology the aim of this work was to determine if a microRNA (miRNA) expression profile was able to discriminate between DCD grafts of varying early function and outcome. Methods Three DCD groups were identifed. PNF group (n=8) retransplanted within a week, good functional outcome group (n=7) AST < 1000 IU/L and graft dysfunction group/IPGF (n=8) AST ≧ 2500 IU/L. miRNA was isolated from FFPE trucut post perfusion biopsies. Affymetrix GeneChip® miRNA 2.0 Arrays (Santa Clara, CA, USA) were used. Expression data analysis was based on ANOVA statistics in the Partek Genomic Suite software (Partek Inc., ST Louis, MI, USA). A statistical significance level of p<0.05 was used to identify differentially expressed miRNA in comparison of the three groups. To validate microarray results real time quantitative PCR (RT-qPCR) was performed on all samples for miRNA species identified as being significant. Ct (threshold cycle) values were normalised using the average Ct of the endogenous control to generate ΔCT and fold-change to reference sample of normal liver (2ΔΔCT) for relative expression analysis. Data was then analysed by one way non parametric ANOVA to detect significance (p<0.05) between the three groups. If significance was found the Bonferroni correction for multiple comparisons was applied. Results From the array analysis 16 miRNA species (miR-107, miR-378, miR-23b and miR-122_st, miR-103, miR-125b, miR-24, miR-let-7a, miR-191, miR-194, miR-296–5p, miR-455–3p, miR-940, miR-let-7d, miR-22 and miR-155) were identified as being significantly different (p<0.05) and of potential biological interest. miR-1 55 and miR-940 had the highest relative expression across all samples. The miRNA that was found to have a significant differential expression between the three groups was miR-22. From computational biology analysis miR-22 was predicted to influence signalling pathways that impact protein turnover, metabolism and apop-tosis/cell cycle. Conclusion Based on this analysis microRNA expression patterns have a low diagnostic potential in discriminating DCD graft quality. The miRNA species (miR-22), that appeared to define whether a poor quality graft had the potential to recover (IPGF) to one that would not (PNF), are those that contribute to the 'survival response' by minimising cellular replacement processes and halting apoptotic initiation of proliferation to conserve energy.

Disclosures:

Nigel Heaton - Advisory Committees or Review Panels: Novartis, Roche; Speaking and Teaching: Astellas

The following people have nothing to disclose: Shirin E. Khorsandi, Siamak Salehi, Wayel Jassem, Hector Vilca-Melendez, Andreas Prachalias, Parthi Srini-vasan

1677

Six minute walk distance predicts waitlist mortality more accurately than sarcopenia in liver transplant candidates

Anitha Yadav1, Yu-Hui H. Chang2, Alvin C. Silva3, Bashar Aqel1, Thomas J. Byrne1, David D. Douglas1, Hugo E. Vargas1, Elizabeth J. Carey1;

1Mayo Clinic, Pheonix, AZ; 2Health Sciences Research, Mayo Clinic AZ, Pheonix, AZ; 3Radiology, Mayo Clinic AZ, Pheonix, AZ

Background Sarcopenia (loss of skeletal muscle mass) is commonly seen in liver transplant (LT) candidates and is associated with poor outcomes. Six minute walk distance (6MWD), a simple test of global functional capacity, also predicts mortality in LT candidates. Recently, the European Working Group on Sarcopenia in Older People (EWGSOP) developed consensus diagnostic criteria for sarcopenia that includes functional status (gait speed) and muscle mass. There are no studies evaluating whether sarcopenia, as defined by EWGSOP criteria, help to predict outcomes in LT candidates. Aim To compare the clinical utility of three tests: 6MWD, muscle mass and EWGSOP-defined sarcopenia (gait speed <0.8m/s and muscle mass) in the prediction of death on wait list in LT candidates Methods Data collected included: age, gender, etiology of liver disease, Model for End-stage Liver Disease (MELD) score. Muscle mass was assessed by measuring total psoas area (TPA) on pre-LT cross-sectional imaging. TPA was stratified to: 1 604mm2 (Q1), 2060mm2 (median) and 2572mm2 (Q3). The Likelihood ratio (LR) for death on wait list was calculated for 3 different tests: 1) 6MWD (<250m) 2) TPA (<1604mm2) 3) EWGSOP proposed sarcopenia (gait speed <0.8m/s AND TPA <1604mm2) The values for the LRs for predicting death on wait list for all three tests were compared. The analyses were conducted using likelihood ratio as a measure of diagnostic accuracy. Prediction for the death on wait list is considered conclusive with LRs of > 1 0 Results A total of 277 cirrhosis patients were included. The mean age was 55 ±9 and 137 (63.1%) were males. HCV (45%) was the most common etiology of liver disease. The mean MELD score was 14.6. The mean 6MWD was 378.9±127.6m and mean TPA was 2131.8 ± 657.3mm2. The LRs for all three different tests are shown in table 1. Conclusions 6MWD, TPA and EWGSOP-defined sarcopenia predict death on the waitlist in LT candidates. However, 6MWD <250m has highest LR for identifying patients likely to die on waitlist compared to other two tests. Incorporating functional status into the definition of sarcopenia, as proposed by the EWGSOP criteria is promising, but the current criteria are less helpful than the 6MWD alone in predicting mortality in LT candidates.

Table 1. Likelihood ratios for all 3 tests
tetotLikelihood ratiop-value
6MWD <250m19.00<0.0001
TPA(<1604mm2)10.840.0044
EWGSOP-defined sarcopenia (Gait speed <0.8m/s AND TPA <1604 mmit17.01<0.000l

Disclosures:

Hugo E. Vargas - Advisory Committees or Review Panels: Eisai; Grant/Research Support: Merck, Gilead, Idenix, Novartis, Vertex, Janssen, Bristol Myers, Ikaria, Abbott

The following people have nothing to disclose: Anitha Yadav, Yu-Hui H. Chang, Alvin C. Silva, Bashar Aqel, Thomas J. Byrne, David D. Douglas, Elizabeth J. Carey

1678

OPI is superior to BAR, MELD, D-MELD and DRI to predict mortality and morbidity after liver transplantation

Felix J. Bäuerlein, Reinhart Zachoval;

Medical Policlinic II, Medical Centre of Ludwig-Maximilians-University Munich - Campus Grosshadern, Munich, Germany

Purpose Allocation for liver-transplantation (LTx) in The UNOS and EUROTRANSPLANT regions is based on the Model for End-Stage Liver Disease (MELD) score. The MELD-score was developed to predict 3 months mortality on the waiting list but seems to be suboptimal to predict survival after LTx. The value of other scores: Donor-Risk-Index (DRI), Organ-Patient-Index (OPI) and the Balance-Of-Risk (BAR) score - to better predict the outcome has yet to be determined. Methods We analyzed 533 liver transplantations (including 73 Retransplantations and 82 high-urgent transplantations) at our center from 01/1997 until 12/201 1 and compared the potency to predict post LTx mortality and morbidity between BAR-, MELD-, D-MELD-, DRI-, OPI-Score. Morbidity was measured as length-of-stay on ICU and the overall in-hospital stay in the first 12 months after LTx. Results 67% men, mean age at LTx 50 yrs (15–67), indication for LTx alcohol 25%, HCV 18%, HCC 12%, acute liver failure 1 1%, cholestatic liver disease 11%, HBV 7%, others 14%. Mortality: All score cutoffs were able to discriminate 1yr-transplant-sur-vival: BAR(<16/≧16) 76.1% - 63.3% (p=.015), MELD(<30/>30) 82.3% - 73.5% (p=.030), D-MELD(<1600/≧1600) 76.0% - 61.7% (p=.004), DRI(<1.4/1.4–1.6/>1.6) 84.4% - 75.0% - 66.9% (p<.001), OPI(<2.0/2.0–2.5/≧2.5) 86.5% - 76.8% - 64.4% (p<.001). OPI best identified patients with early death after LTx. The power to identify patients with high mortality risk fades during the 1 0yr-followup for MELD and BAR-score, but remains significant for D-MELD(<800/800-1600/≧1600) 60.8% - 44.4% -31.1% (p<.001), DRI(<1.4/1.4-1.6/>1.6) 65.9% - 49.0% -38.2% (p<.001) and OPI(<2.0/2.0-2.5/>2.5) 66.5%-50.3% - 37.1 % (p<.001). Morbidity: Kaplan Meier analysis showed a correlation between length-of-stay on ICU and score-points for BAR, MELD, D-MELD and OPI (for all: p<.001), but not for DRI. Median stay on ICU and [AUROC for ICU>21d] was BAR(<16/≧16) 7d - 15d [.67], MELD(<30/>30) 7d - 12d [.65], D-MELD(<1600/≧1600) 7d - 12d [.64], OPI(<2.0/>2.0) 6d -10d [.64]. The median in-hospital stay in the first 12 months was significantly associated with the score only in OPI(<2.0/2.0–2.5/>2.5) 50d - 67d - 76d (p<.001). Conclusion In the time of extreme organ shortage two aspects of liver allocation have to be considered - disease severity reflected by the MELD score and patient survival after transplantation for which an ideal score has yet to be found. Our findings indicate that OPI performs best to predict the 12 months morbidity and more important one/ten year survival.

Disclosures:

The following people have nothing to disclose: Felix J. Bäuerlein, Reinhart Zachoval

1679

Regional differences in the outcome of Liver Transplantation for Hepatocellular Carcinoma beyond Milan Criteria

Victor I. Machicao1, Sachin Batra1, Shivang Mehta1, Kevin J. Dasher1, Jen Jung Pan1, Wasim A. Dar2, MarkJ. Hobeika2, J. S. Bynon2, Michael B. Fallon1;

1Gastroenterology, Hepatology and Nutrition, University of Texas Health Science Center at Houston, Houston, TX; 2Immunology and Organ Transplantation, University of Texas Health Science Center at Houston, Houston, TX

Background: Liver transplantation (LT) for hepatocellular carcinoma (HCC) beyond Milan criteria (UNOS-T3) is performed upon candidate approval by regional review board (RRB). Since 2005 UNOS Region 4 RRB criteria for HCC beyond Milan (R4-T3) has been standardized to a single <6 cm lesion or up to 3 lesions (largest <5 cm) and total tumor diameter <9 cm. Similarly since 2007 UNOS Region 5 RRB criteria is limited to UCSF Criteria for HCC beyond Milan (R5-T3). The aim of our study is to compare the outcome after LT for UNOS-T3 lesions according to UNOS region, and compared with the outcome of HCC Milan criteria (UNOS T2 lesions) within each Region. Methods: We reviewed LT data from UNOS STAR Database in adults transplanted from 01/2005 to 03/2010. We excluded all recipients listed as status 1, multiorgan transplant or re-transplantation or LT performed before 01/2007 in Region 5 (n=75) since R5-T3 criteria was not used before that time. Multivariate cox regression analysis was performed to determine regional differences in mortality. Results: A total of 4844 LT forT2 lesions and 1257 LT for T3 lesions were performed during study period. The demographic and clinical variables of each cohort were comparable (Table 1). Patient survival curve for T3 lesions were comparable to T2 lesions transplanted within each region. Patient survival for R4-T3 were significantly better in region 4 compared to other UNOS regions (HR 0.67, 95%CI: 0.45–0.99, p<0.05), whereas survival of R5-T3 was comparable to other regions. This effect was independent of recipient age and race, donor age, waiting time, and MELD at time of LT. Dropout and death rate while listed were similar among regions for HCC individuals. Conclusion: Outcome after LT in T3 lesions is similar to T2 lesions. Outcome after LT for R4-T3 lesions is better compared to outcome of UNOS-T3 lesions in other regions. A standardized criterion used for selection of T3 lesions could be a potential explanation of our findings. The implementation of these criteria does not affect dropout or mortality rates among HCC patients within the UNOS region.

Clinical and Demographic Features of T3 lesions by UNOS region

VariablesR4-T3(n=l68)R5-T3(n=345)Other UNOS Region-T3 (n=751)
  1. * (p<0.05), compared to Other UNOS Region

Age58 ±758 ±757 ±7
Male sex132(79%)269 (78%)617 (82%)
Caucasian race102(61%)156(45%)530(71%)
MELD at LT14 ± 7 *13 + 7*I2±5
HCV (+)89 (66%)230 (67%)472 (68%)
Waiting time (mo)8± 1418 ± 25 *10± 18

Disclosures:

Victor I. Machicao-Advisory Committees or Review Panels: Gilead Sciences Inc, Vertex Pharmaceuticals

Michael B. Fallon - Advisory Committees or Review Panels: Bayer/Onyx; Grant/Research Support: Ikaria Therapeutics, Gilead, ANADYS, Mochida, Eaisi, Research Triangle Institute

The following people have nothing to disclose: Sachin Batra, Shivang Mehta, Kevin J. Dasher, Jen Jung Pan, Wasim A. Dar, Mark J. Hobeika, J. S. Bynon

1680

Higher risk of graft failure and biliary complications for donation after cardiac death liver transplantation in primary sclerosing cholangitis: Analysis of the UNOS database

Vinay Sundaram, Steven D. Colquhoun, Walid Ayoub, Nicholas N. Nissen, Andrew S. Klein, Tram T. Tran;

Cedars-Sinai Medical Center, Los Angeles, CA

Aim: Disease specific outcomes are unknown with use of donation after cardiac death (DCD) livers. We assessed the effect of DCD livers on graft outcomes in patients transplanted for primary sclerosing cholangitis (PSC). Methods: We analyzed the United Network for Organ Sharing (UNOS) database for patients age 1 8 or older, who underwent orthotopic liver transplant (OLT) from 2002–2012. Patients who underwent living donor transplant, multi-organ transplant or had a previous transplant were excluded. We used Student's t-test and Mann-Whitney U test for continuous variables and Chi-square test and Fisher's exact test for categorical variables. Multivariate logistic regression was performed to determine the odds of graft failure. Graft survival was compared using Kaplan-Meier analysis and log-rank test. Results: Of 41,018 patients receiving OLT, 1,667 (4.1%) had PSC and 39,351 (95.5%) did not. Among PSC patients, 75 (4.5%) received DCD grafts, while 1,592 (95.5%) received donation after brain death (DBD) grafts. Among non-PSC patients, 1,868 (4.8%) received DCD grafts, while 37,483 (95.3%) received DBD grafts. PSC patients receiving DCD grafts received younger donor livers (32.6 vs 41.1 years, p<0.001) and more Caucasian donor livers (90.7% vs 72.7%, p<0.001) but were otherwise similar to PSC patients receiving DBD grafts. PSC patients receiving DCD grafts had higher prevalence of overall graft failure (37.3% vs 20.4%, p=0.001) and graft failure from biliary complications (47.4% vs 13.9%, p=0.002), compared to PSC patients receiving DBD grafts. However, among all transplanted patients, multivariate logistic regression demonstrated PSC patients receiving DCD grafts did not have higher risk of graft failure (Table 1). Among all DCD transplanted patients, Kaplan-Meier analysis showed PSC patients did not have shorter 5-year graft survival than non-PSC patients (p=NS). Conclusion: Although overall graft failure for DCD transplants was not different between PSC and non-PSC patients, PSC patients receiving DCD grafts had both a higher risk of graft failure from biliary complications and overall graft failure than PSC patients receiving DBD grafts.

Multivariate logistic regression analysis for risk of graft failure

 Odds ratiop-value95% Confidence Interval
  1. * = model adjusted for covariates

PSC0.71>0.0010.58–0.86
DCD1.67>0.00l1.39–2.00
PSC-DCD interaction2.31NS0.91–5.95
Warm ischemia time*1.00NS0.99–1.00
Cold ischemia time*1.04>0.0011.03–1.05
Donor age*1.020.0021.01–1.02
Caucasian donor*0.880.0020.81–0.95

Disclosures:

Vinay Sundaram - Advisory Committees or Review Panels: Salix; Speaking and Teaching: Salix

Tram T. Tran - Advisory Committees or Review Panels: Gilead, Bristol Myers Squibb, Vertex; Consulting: Gilead; Grant/Research Support: Bristol Myers Squibb; Speaking and Teaching: Bristol Myers Squibb, Gilead, Vertex

The following people have nothing to disclose: Steven D. Colquhoun, Walid Ayoub, Nicholas N. Nissen, Andrew S. Klein

1681

Prevalence of Donor Specific Antibodies in Liver Transplant Recipients with Unexplained Liver Graft Abnormalities is Extremely High and Significantly Higher than in Control Groups: Role of Antibody-mediated Rejection in Long Term Graft Dysfonctions

Ainad Ghandir1, Maryvonnick Carmagnat3, Elie Serge Zafrani2, Jean Philippe Mavier1, Dominique Desvaux2, Monika Hurtova1, Cyrille Feray1, Ariane Mallat1, Caroline Suberbielle3, Christophe Duvoux1;

1hepatology, Hopital Henri Mondor, Créteil, France; 2pathology, Hopital Henri Mondor, Créteil, France; 3Immunology and Histocompatibility, Hopital Saint Louis, Paris, France

Introduction: Chronic antibody-mediated rejection (AMR) which is well-defined in kidney transplantation, is much less documented after liver transplantation (LT). The aim of this study was to analyze the prevalence of donor specific antibodies (DSA) in LT recipients on the long term according liver graft function. Patients and methods: Inclusion criteria : consecutive outpatient LT recipients diagnosed between 01 and 07/2012 with unexplained LFT abnormalities or unexplained liver fibrosis as assessed by Fibroscan, where tested for DSA (G0). 4 control groups were designed after matching on age and time post LT: pts with normal LFT and no reLT (G1), pts with abnormal LFT and no reLT (G2), pts with abnormal LFT and reLT (G3) and pts with HCV recurrence (G4). G1-G4 referred to control pts overall. Exclusion criteria: ischemic cholangiopathy, infection with HBV or HEV and patients transplanted for autoimmune hepatitis. By definition there were no HCV pts in G0. Results: A total of 91 pts (males:67, females :24, age: 47,2 +/- 11,3 yrs) including 22 patients with unexplained LFT abnormalities (G0), 31, 19,3, 1 6, pts in G1, G2, G3, G4, respectively were investigated, 9.6+/-6.2 yrs after LT. Indications for LT were: alcoholic cirrhosis 23 (25.3%),HCV 23 (25.3%)PCS 13(14.3 %),HCC 4(4,4%), FH 8(8.8%), HBV 6(6.6%),NASH 6 (6.6%), PBC 4 (4.4%) and others 4(4.4%). Anti HLA class I and class II antibodies were found in 86 and 1 00%, respectively in G0, and in 62 and 70% in G1-G4. DSA were found in 95.5% in G0, anti class II in all cases, with a median MFI of 11973 [IQR: 7142–1 7084] (score > 8 in all cases). DSA were found in 50.7 % in G1-G4 (p<0.001 vs G0) with a median MFI of 3443 [1043–12626] (p=0,02 vs G0), score > 8, 6 and 4 in 50%, 28%, 22% (p<0.001 vs G0). The prevalences of DSA in G1, G2, G3 and G4 were 45.2, 52.6, 33.3, and 68.8% respectively (p=0.003 + p<0.001 G0 vs G1, p=0.002 G0 vs G2, p=0.03 G0 vs G3, p=0.065 G0 vs G4). Conclusion: This study shows that 1-Prevalence DSA (mainly anti class II) is extremely high (95%) and significantly higher than in control groups (50%) in patients with unexplained liver graft dysfunction on the long term 2-MFI of DSA in patients with unexplained liver graft dysfunction are significantly higher than in control groups These findings highly uggest the responsibility of DSA and therefore the involvement of a late humoral rejection process in late unexplained allograft dysfunctions 3-In HCV positive pts, the prevalence of DSA was also found high and raises the issue of DSA as a contributor to fibrosis progression.

Disclosures:

Christophe Duvoux - Advisory Committees or Review Panels: Novartis, Roche, Novartis, Roche, Novartis, Roche, Novartis, Roche; Speaking and Teaching: Astellas, Astellas, Astellas, Astellas

The following people have nothing to disclose: Ainad Ghandir, Maryvonnick Carmagnat, Elie Serge Zafrani, Jean Philippe Mavier, Dominique Desvaux, Monika Hurtova, Cyrille Feray, Ariane Mallat, Caroline Suberbielle

1682

Relationship between CYP3A5, ABCB1 and immune genetics and tacrolimus concentrations, and the occurrence of nephrotoxicity and rejection following liver transplantation

Janet Coller1, Jonathan Tuke1, Jeyamani Ramachandran2, Alan J. Wigg2, Matthew Doogue3;

1Pharmacology & School of Mathematical Sciences, University of Adelaide, Adelaide, SA, Australia; 2Hepatology, Flinders Medical Centre, Adelaide, SA, Australia; 3Clinical Pharmacology, Flinders University, Adelaide, SA, Australia

Background: Survival of liver transplants is dependent on preventing organ rejection with immunosuppressants, such as tacrolimus, while avoiding drug toxicity. Tacrolimus is metabolised by the cytochrome P450 3A (CYP3A) subfamily of enzymes and transported by the P-glycoprotein efflux transporter (coded by the ABCB1 gene). Aim: To investigate the effect of genetic variability in drug metabolism (CYP3A5), drug transport (ABCB1, P-glycoprotein) and immune factors of both recipient and donor on tacrolimus concentrations and clinical outcomes in a cohort of liver transplant patients. Methods: Thirty-two tacrolimus treated recipients were recruited (mean age 52 yr; 79% male; written informed consent was obtained and study protocol adhered to the Declaration of Helsinki; no donor organs were obtained from executed prisoners or institutionalized persons). Both recipient and donor tissue was genotyped forCYP3A5*3 (carriers termed as non-expressors), ABCB1 haplotype (loci at position 61, 1199, 1236, 2677 and 3435) and immune factors (cytokines, receptors, and innate immune mediators). Dose-adjusted tacrolimus concentrations, and incidence of nephrotoxicity and biopsy proven rejection over the follow up period were compared between genotype groups using Mann-Whitney U-tests or Kruskal-Wallis tests, and odds ratios for numerical and categorical data, respectively. Logistic regression of genetic data was performed for clinical outcomes. Results: Dose-adjusted tacrolimus concentrations (μg/L) were altered similarly by recipient and donor CYP3A5 expression; CYP3A5 non-expressor recipient / donor pairs had the highest concentrations (2.7), the pairs with either non-expressor recipient / expressor donor or expressor recipient / non-expressor donor had intermediate concentrations, (1.25 and 1.1, respectively), and the expressor recipient / donor pair had the lowest concentration (0.94) (P = 0.023). No significant effect of recipient or donor ABCB1 haplotype on dose-adjusted tacrolimus concentrations was seen (P = 0.18). Neither CYP3A5 expression nor ABCB1 haplotype impacted on the incidence of nephrotoxicity or biopsy proven rejection (P = 0.24 to 1.00). Logistic regression found a strong association between donor IL-6R and IL-1 0 genotype and the risk of nephrotoxicity (receiver operating characteristic curve AUC = 0.93). Conclusion: Tacrolimus dose-adjusted concentrations are dependent on both recipient and donor CYP3A5 genotype, whereas ABCB1 haplotype does not have a major effect. Recipient immune genetics was associated with nephrotoxicity and should be further investigated as a potential predictor of nephrotoxicity from calcineurin inhibitors.

Disclosures:

The following people have nothing to disclose: Janet Coller, Jonathan Tuke, Jeyamani Ramachandran, Alan J. Wigg, Matthew Doogue

1683

Frailty Measurements in Patients Awaiting Liver Transplantation Predict Post-Operative Complications

Elizabeth C. Verna1, Christine Chan1, Rita M. Abdelmessih1, Jospeh Pisa1, Thresiamma Lukose1, Saravanan K. Krishnamoorthy2, RobertS. Brown1;

1Department of Medicine, Columbia University Medical Center, New York, NY; 2Department of Radiology, Columbia University Medical Center, New York, NY

Background: Liver transplant (LT) outcomes are difficult to predict with current tools. Frailty assessment may enhance our ability to risk stratify pre-LT patients. Methods: Outpatients listed for LT were prospectively enrolled from 12/1 1–4/1 3. The validated Fried Frailty Criteria including slowness (timed walk), weakness (grip strength), weight loss, exhaustion, and physical activity domains were used. Frailty was defined as score >3/5. Liver Disease Quality of Life (LDQOL), Short Physical Performance Battery (SPPB) and psoas muscle diameter (PMD) were measured. Primary outcomes were overall, pre-LT and post-LT mortality. Secondary outcomes were a composite of post-LT complications (death, re-operation, infection) and LT hospital-ization length of stay (LOS). Results: 61 patients were enrolled, mean age 59, 79% male. 25 (42%) met criteria for frailty. 35 (57%) had LT in the median follow up of 286 days. Frail subjects were more likely to be older, female, have diabetes, and non-HCV related cirrhosis. They had higher MELD at listing and LT, and a trend towards more CKD, portal hypertension, ascites and encephalopathy. Frail subjects had significantly lower mean grip strength (23 v. 34 kg, p<0.001), walking speed (6.2 v. 4.5 sec/15ft, p<0.001), SPPB scores (5.0 v. 7.0, p<0.001), PMD (30 v. 35mm, p=0.049) and scores on HDQOL domains including symptoms (p<0.001), disease effects (p=0.02), concentration (p=0.02), memory (p=0.02), and health distress (p=0.01). 8 (13%) patients died (5 pre-LT, 3 post-LT). Frailty was not significantly associated with overall (p=0.64), pre-LT (p=0.91) or post-LT (p=0.69) mortality, but was predictive of post-LT complications (p=0.04), and median LOS (14 v. 9.5 days, p=0.003). Walking speed was predictive of overall (p=0.004) and pre-LT mortality (p=0.006), and grip strength was predictive of post-LT complications (p=0.005), with a trend towards overall (p=0.1 1) and post-LT (p=0.06) mortality. SPPB and PMD scores were numerically lower but not significantly predictive of overall, pre- or post-LT mortality, but PMD was significantly lower in those with post-LT complications (p=0.003). In multivariable analysis PMD (HR 0.83 p=0.006), SPPB score (HR 0.51, p=0.04) and walking speed (HR 0.55, p=0.09) were the most predictive of post-LT complications in this small cohort. Conclusions: Frailty tools, particularly physical performance (SPPB, walking speed, grip strength) and sar-copenia (PMD), are predictive of important outcomes including mortality, post-LT complications, and LOS, and may improve our ability to risk-stratify wait list patients. Larger studies are needed to develop and validate highly predictive disease-specific models.

Disclosures:

Thresiamma Lukose - Grant/Research Support: Cangene

Robert S. Brown - Consulting: Salix, Janssen, Vertex; Grant/Research Support: Gilead, Merck, Vertex, AbbVie, Salix, Janssen, BI; Speaking and Teaching: Genentech, Gilead, Merck

The following people have nothing to disclose: Elizabeth C. Verna, Christine Chan, Rita M. Abdelmessih, Jospeh Pisa, Saravanan K. Krishnamoorthy

1684

Switching from CNI-based immune suppression to Sirolimus improves survival after liver transplantation in patients with Hepatitis C virus (HCV) infection. A single centre, 15-year experience

Meera Shah1, Arun N. Shankar1, Ian Gee1, Kathryn L. Nash1, Matthew Hoare1,3, Paul Gibbs2, Graeme J. Alexander1;

1Dept. of Medicine, University of Cambridge, Addenbrooke's Hospital, Cambridge, United Kingdom; 2Dept. of Surgery, University of Cambridge, Addenbrooke's Hospital, Cambridge, United Kingdom; 3CRUK Cambridge Institute, University of Cambridge, Cambridge, United Kingdom

The use of Sirolimus-based immune suppression in liver graft recipients in general and in HCV-infected recipients in particular remains contentious. There is evidence that Sirolimus retards hepatic fibrosis and may be of benefit in preventing HCC recurrence. It is used routinely in some centres to manage CNI-induced renal impairment. However, concerns persist regarding an increased risk of hepatic artery thrombosis, graft loss and death with Sirolimus prescribed de novo. We describe 195 patients undergoing first liver transplantation for HCV over 15 years. 1 18 switched from CNI-based immune suppression to Sirolimus (median 14 months) after transplantation, largely because of HCV-related hepatic fibrosis (69%) or renal impairment (14%); 77 remained on CNI-based immune suppression throughout follow-up without evidence of hepatic fibrosis or renal impairment. Sirolimus therapy resulted in side effects in 41% of cases requiring withdrawal in 21%. There were 86 episodes of acute rejection; 81% occurred on Tacrolimus, 8% on Sirolimus and 6% during the switch from CNI to Sirolimus. The mean duration of Sirolimus therapy was 3.8 years. When analysed by intention to treat, patients switched to Sirolimus had improved survival (p <0.0001) when compared to those maintained on CNI therapy without hepatic fibrosis or renal impairment, an effect which was only manifest after 2 years therapy. Conclusion: HCV-positive transplant recipients with hepatic fibrosis or renal impairment that switched from CNI to Sirolimus therapy had better survival 2 years and beyond after the introduction of Sirolimus. These data suggest that switching to Sirolimus was both safe and effective, but provide no clear indication of the mode of action of Sirolimus.

Patient survival according to immunosuppression analysed by intention to treat

image

Disclosures:

The following people have nothing to disclose: Meera Shah, Arun N. Shankar, Ian Gee, Kathryn L. Nash, Matthew Hoare, Paul Gibbs, Graeme J. Alexander

1685

Radio-anatomic correlation for hepatocellular carcinoma treated with liver transplantation : impact on survival

Laetitia Lecoq1, Jérôme Gendre2, Nathalie Sturm3, Christian Letou-blon4, Marie Noelle Hilleret1, Jean-Pierre H. Zarski1, Vincent Leroy1;

1Hépato-gastro-entérologie, Centre Hospitalo Universitaire, La Tronche, France; 2Service de radiologie, Centre Hospitalo Universitaire, La Tronche, France; 3Service d'anatomo-pathologie, Centre Hospitalo Universitaire, La Tronche, France; 4Service de chirurgie digestive, Centre Hospitalo Universitaire, La Tronche, France

Milan criteria are widely used for the selection of patients eligible for liver transplantation (LT) and organ allocation policy. Extended criteria such including UCSF, up-to-seven and alpha-foeto protein scores have been more recently proposed. Although Milan criteria were originally defined on explant analyses, scores are used before LT on imaging. Therefore, the objectives of this study were 1) to evaluate the reproducibility of these criteria between pre-operative imaging and explant histology, and 2) to determine its impact on HCC recurrence and survival after LT. One hundred consecutive patients transplanted for HCC in a single center between 2005 and 2012 were included. A neo-adjuvant chemo-embolization (TACE) was performed in 54 patients. A comparison was made between the last imaging done within 3 months before LT (CT scan or MRI) and histology, on a total of 322 nodules. Repeated reading on imaging was made by an expert radiologist. A significant correlation between imaging and histology was observed for the total tumour size (r=0.71, p<0.001) and the number of nodules (r=0.77, p<0.001). Ninety six nodules, mostly of small size (mean 1 1.2mm) were seen on histology but not on imaging. There was a good concordance for Milan criteria between imaging and histology (78.8%) but lower than for UCSF criteria (82.2%), up to seven criteria (84.5%) and AFP score (89%). Discordances were mainly due to the presence at histology of nodules not seen on imaging. We also evaluated the impact of chemo-embolization. The mean tumour destruction, according to mRECIST criteria, was 60% on the imaging, which was superior to the percentage of necrosis found on pathology (48%). Among the patients who had a complete tumoral response on imaging, only 28% actually had a total necrosis on histology. Eleven recurrences were observed with a median follow-up of 33 months after LT, with a 5 year recurrence free survival of 67.9%. Scores evaluated on histology had always a better predictive value than that evaluated on imaging. Patients who were initially outside the scores but achieving downstaging had the same risk of recurrence than those initially inside the scores (5 year recurrence free survival 71.0% vs 69.2% vs 53.3% for patients remaining outside criteria after TACE). This study validates the use of viable tumour on imaging in the selection of patients for LT. However nearly 20% of patients outside Milan, and 1 0–15% for other scores on imaging were eventually inside the scores on histology with an excellent post-LT prognosis. These data challenge the use of too restrictive organ allocation policy, and highlight the need of further prognostic studies.

Disclosures:

Jean-Pierre H. Zarski - Advisory Committees or Review Panels: BMS, Gilead, Janssen Cilag, BMS, Gilead, Janssen Cilag; Consulting: Roche, Scherring Plough, Novartis, Roche, Scherring Plough, Novartis; Speaking and Teaching: Siemens

Vincent Leroy - Board Membership: roche, merck, gilead, bms, roche, merck, gilead, bms, roche, merck, gilead, bms, roche, merck, gilead, bms; Consulting: jansen, jansen, jansen, jansen; Grant/Research Support: roche, gilead, bms, roche, gilead, bms, roche, gilead, bms, roche, gilead, bms; Speaking and Teaching: bms, merck, gilead, roche, bms, merck, gilead, roche, bms, merck, gilead, roche, bms, merck, gilead, roche

The following people have nothing to disclose: Laetitia Lecoq, Jérôme Gendre, Nathalie Sturm, Christian Letoublon, Marie Noelle Hilleret

1686

Urinary neutrophil gelatinase-associated lipocalin as a biomarker for tacrolimus-induced acute kidney injury in recipients of living-donor liver transplantation

Haruka Shinke1, Ayami Tsuchimoto1, Miwa Uesugi1, Kazuo Mat-subara1, Yasuhiro Fujimoto2, Toshimi Kaido2, Shinji Uemoto2, Motoko Yanagita3, Satohiro Masuda1;

1Department of Clinical Pharmacology and Therapeutics, Kyoto University Hospital, Kyoto, Japan;2Divisions of Hepato-Pancreato-Biliary, Transplant and Pedi-atric Surgery, Department of Surgery, Kyoto University Hospital, Kyoto, Japan; 3Department of Nephrology, Kyoto University Hospital, Kyoto, Japan

[Background and Aim] Tacrolimus-induced acute kidney injury (AKI) is a serious problem after living-donor liver transplantation (LDLT). Several urinary biomarkers have been reported to date for detecting AKI. Herein, we aimed to identify which marker is best for the detection of tacrolimus-induced AKI after LDLT. [Methods] A total of 23 patients were enrolled in this study after providing written informed consent. Urine samples were collected on postoperative day (POD) 1, before tacrolimus administration, and on POD 7, 14, and 21. The 7 most recently identified biomarkers were measured using ELISA. The diagnosis of AKI was based on the discretion of the attending physicians or nephrologists. This study was conducted in accordance with the Declaration of Helsinki and its amendments and was approved by Kyoto University Graduate School and the Faculty of Medicine Ethics Committee. [Results] Urinary neutrophil gelatinase-associated lipocalin (NGAL), monocyte chemotactic protein-1, and liver-type fatty acid binding protein levels were significantly higher in the AKI group than in the AKI-free group. However, the concentrations of interleukin-1 8, osteopontin, cys-tatin C, and clusterin did not differ between the 2 groups. Composite area under the receiver operating characteristics curve (AUC-ROC) analysis showed that NGAL had high sensitivity for detecting tacrolimus-induced AKI after LDLT (AUC-ROC = 0.94, 95% confidence interval, 0.89–1.00). Furthermore, the urinary concentration of NGAL on POD 7 could predict AKI development on POD 8–14. The cutoff value of urinary NGAL levels on POD 7 was evaluated by ROC analysis as 62.6 ng/mg creati-nine. [Conclusion] The urinary concentration of NGAL would be the obvious biomarker among the 7 molecules to detect and predict tacrolimus-induced AKI after LDLT. Our findings suggest a decrease in the dose of tacrolimus to prevent further deterioration of kidney function when the urinary NGAL level on POD 7 exceeds the cutoff value.

Disclosures:

The following people have nothing to disclose: Haruka Shinke, Ayami Tsuchimoto, Miwa Uesugi, Kazuo Matsubara, Yasuhiro Fujimoto, Toshimi Kaido, Shinji Uemoto, Motoko Yanagita, Satohiro Masuda

1687

Cyclosporine and tacrolimus have different inhibitory effects on toll-like receptor signalling post liver transplantation

Rohit Sawhney1,2, Jessica Howell1,2, Adam Testro1, Narelle A. Skinner3, Dilip Ratnam4, Peter W. Angus1,2, Paul Gow1,2, Kumar Visvanathan2,3;

1Liver Transplant Unit, Austin Health, Melbourne, VIC, Australia; 2Medicine, The University of Melbourne, Melbourne, VIC, Australia; 3St Vincent's Hospital, Melbourne, VIC, Australia; 4Monash Medical Centre, Melbourne, VIC, Australia

Background: Toll-like receptors (TLRs) play a key role in transplantation biology, including graft rejection, infection risk and recurrence of diseases such as hepatitis C. The effect of immunosuppression on TLR function post liver transplantation is unknown. This study explores the effect of calcineurin inhibitors on TLR function in PBMCs post liver transplantation. Methods: Peripheral blood mononuclear cells (PBMCs) from 1 1 3 post liver transplant patients and 13 healthy controls were stimulated with TLR specific ligands LPS (TLR4), P3C (TLR2), PIC (TLR3), R848 (TLR7/8) and CpG (TLR9) for 24 hrs. PBMCs from five healthy controls were also cultured with therapeutic concentrations of cyclosporine and tacrolimus. Cytokine production was then measured using ELISA and flow cytometry. Results: PBMCs from patients on calcineurin inhibitors post liver transplant produced less IL-6 and TNFα in response to TLR2, TLR4 and TLR7/8 stimulation compared with healthy controls (TLR2 IL-6 p=0.019, TNFα p=0.012; TLR4 IL-6 p=0.021, TNFα p=0.010; TLR7/8 IL-6 p=0.020, TNFα p=0.019). Both NK CD56bright and dim cells from patients on calcineurin inhibitors also produced less IFNγ with TLR7/8 stimulation compared with healthy controls (NK CD56bright p=0.002; CD56dim p=0.004). Similar findings were demonstrated in healthy PBMCs cultured with cyclosporine (PBMC TLR2 IL-6 p=0.005; TLR4 IL-6 p=0.029, TNFα p=0.028; TLR7/8 IL-6 p=0.024, TNFα p=0.010; NK CD56dim cell TLR7/8 IFNγ p=0.025). Healthy PBMCs cultured with tacrolimus had impaired TLR4-mediated IL-6 and TNFα by PBMCs (IL-6 p=0.01 7, TNFα p=0.009). Conclusion: Patients on calcineurin inhibitors have impaired inflammatory cytokine production in response to TLR2, TLR4 and TLR7/8 stimulation compared with healthy controls. Impaired TLR function post liver transplant has important repercussions for risk of infections, graft rejection and disease recurrence such as HCV infection. The innate immune suppressive profiles of cyclosporine and tacrolimus may therefore guide choice of immunosuppression therapy to improve disease outcomes.

Disclosures:

Peter W. Angus - Grant/Research Support: Gilead Sciences

The following people have nothing to disclose: Rohit Sawhney, Jessica Howell, Adam Testro, Narelle A. Skinner, Dilip Ratnam, Paul Gow, Kumar Visvanathan

1688

Comparison of doxorubicin-eluting bead transarterial chemoembolization (DEB-TACE) with hyperselective transarterial chemoembolization (hsTACE) before liver transplantation

Julie Mayer2, Alexis Laurent3, Julien Calderaro4, Amel Soualmia1, Alain Luciani2, Jeanne Tran Van Nhieu4, Charlotte E. Costentin1, Daniel Azoulay3, Ariane Mallat1, Christophe Duvoux1, Hicham Kobeiter2, Thomas Decaens1;

1Department of Hepatology, Henri Mondor Hospital, Creteil, France;2Department of Radiology, Henri Mondor Hospital, Creteil, France; 3Department of Surgery, Henri Mondor Hospital, Creteil, France; 4Department of Pathology, Henri Mondor Hospital, Creteil, France

Aim: To compare doxorubicin-eluting bead transarterial chemoembolization (DEB-TACE) with hyperselective transarterial chemoembolization with lipiodol and doxorubicin (hsTACE) in patients awaiting liver transplantation for hepatocellular carcinoma (HCC). Methods: 16 patients evaluated for liver transplantation for HCC and treated with DEB-TACE were carefully matched with 16 patients treated with hsTACE during the waiting time of liver transplantation. The evaluation of the effectiveness of treatment was performed according to modified RECIST criteria by EASL. Treatment efficacy was also assessed by analysis of the liver explant. Data are expressed as mean ± SD and median (range). Survival curve were compared according to Kaplan Meier method. Results: At baseline, the median number of HCC was 2 in both groups, with a mean diameter of the largest tumor of 30 ± 3 mm for DEB-TACE and 33 ± 3 mm for hsTACE groups. The median waiting time was 7.0 ±1.1 months and 7.5 ±1.3 months, respectively with a median number of cure of 1 (IQR:1–2) and 2 (IQR:1–2). The best radiological response was not significantly different between groups with 4 complete responses in both groups, 5 versus 3 partial responses, 3 versus 4 stable diseases and 1 versus 2 progressive diseases in DEB-TACE and hsTACE, respectively. Radiological response was not assessable in 3 patients in DEB-TACE and 2 patients in hsTACE groups because the waiting time was too short. On liver explant analysis, the median number of tumor was 2 in both groups, and the mean diameter of the largest tumor was 32.8 mm in DEB-TACE vs 34.1 mm in hsTACE. The mean percentage of tumor necrosis was 49.5 ±8.9 and 57.4 ± 8.9 (ns). The percentage of necrosis was higher than 95% in 13 tumors in DEB-TACE group and in 15 tumors in hsTACE group. Tumor differentiation did not differ between group, as well as micro-vascular invasion, satellite tumor and macro-vascular invasion. Five-year overall survival was 78.6 vs 1 00% in DEB-TACE and hsTACE group respectively (p=0.55). Conclusion: In this study, we were not able to identify any difference in between DEB-TACE and hs-TACE in term of radiological response, pathological response or clinical response, suggesting that DEB-TACE efficacy is more related to embolization tech-nic rather than the drug release system.

Disclosures:

Christophe Duvoux - Advisory Committees or Review Panels: Novartis, Roche, Novartis, Roche, Novartis, Roche, Novartis, Roche; Speaking and Teaching: Astellas, Astellas, Astellas, Astellas

The following people have nothing to disclose: Julie Mayer, Alexis Laurent, Julien Calderaro, Amel Soualmia, Alain Luciani, Jeanne Tran Van Nhieu, Charlotte E. Costentin, Daniel Azoulay, Ariane Mallat, Hicham Kobeiter, Thomas Decaens

1689

Detection of alcohol consumption post liver transplantation using determination of alcohol marker

Martina Sterneck1,4, Hilke Andresen-Steichert2, Gregor von Rothkirch1, Karl-Heinz Schulz1, Eik Vettorazzi3, Bjoern Nashan1;

1Department of Hepatobiliary and Transplant Surgery, University Medical Center Hamburg Eppendorf, Hamburg, Germany; 2Department of Legal Medicine, University Medical Center Hamburg Eppendorf, Hamburg, Germany; 3Department of Medical Biometry and Epidemiology, University Medical Center Hamburg Eppendorf, Hamburg, Germany; 4Department of Gastroenterology, University Medical Center Hamburg Eppendorf, Hamburg, Germany

Early detection of alcohol abuse in OLT recipients is essential to offer patients support and counseling to prevent organ damage. Here the diagnostic value of alcohol marker determination for assessment of alcohol consumption was evaluated. Methods: In 104 transplant recipients presenting to the outpatient clinic (31 patients with underlying ALD and 73 with non-ALD) blood ethanol (EtOH), methanol (MeOH), CDT, urine ethyl glu-curonide (uEtG) and hair ethyl glucuronide (hEtG) levels were determined. The results were compared with patients' self-reports in a questionnaire and physicians' assessments. Results: By physicians' assessments 22% of patients were suspected to consume potentially harmful amounts of alcohol, while only 6% of patients admitted consumption of > 2 drinks/week. Alcohol markers revealed alcohol consumption in 17% (18/104) of patients. In all but 2 of 1 8 cases (89%) consumption was detected by a positive hEtG, while CDT, EtOH, MeOH and uEtG were positive in 1 1%, 6%, 0% and 39% of the 1 8 cases with positive alcohol markers, respectively. In 61% of patients hEtG was the only positive marker. Compared to patients with non-ALD, patients with ALD significantly more often stated consumption of > 2 drinks/week (ALD 13% vs non-ALD 3%; p=0.04), were significantly more often suspected to consume alcohol by their physicians (ALD 35% vs non-ALD 16%; p=0.03), had significantly more often positive alcohol markers (35 % vs 10%; p=0.001) and significantly more often a positive hEtG result (32% vs 8%; p=0.002). Furthermore mean proximal hEtG concentration was significantly higher (248.6 pg/mg + 293.7 vs 25.1 pg/mg + 14.3; p=0.049) in patients with ALD. In all patients admitting daily alcohol consumption (3/3) hEtG was positive. But in addition also 42% (1 1/26) of patients admitting only occasional (3 patients 2–6 drinks/ week; 8 patients < 1 drink/week) and 3% (2/75) of patients denying any alcohol consumption, hEtG was positive indicating an alcohol intake of > 10 g per day over the last months. In 12/23 (52%) of patients suspected to possibly consume potentially harmful amounts of alcohol by the physician, the alcohol markers tested positive, while 7% (6/81) of those not suspected had a positive result (PPV 66%, NPV 87%). Conclusion: Recipients with ALD consume significantly more alcohol post transplant compared to patients with non-ALD. Hair-EtG was the best marker for detection of alcohol consumption. We recommend yearly screening of all patients with ALD and with suspected alcohol consumption according to the physician.

Disclosures:

Bjoern Nashan - Advisory Committees or Review Panels: Novartis, Bristol-Myer Squibb; Speaking and Teaching: Novartis, Bristol-Myer Squibb

The following people have nothing to disclose: Martina Sterneck, Hilke Andresen-Steichert, Gregor von Rothkirch, Karl-Heinz Schulz, Eik Vettorazzi

1690

Liver Histology among Stable Pediatric Liver Transplant Recipients: A Prospective Multicenter Study

John Bucuvalas1, George V. Mazariegos2, Anthony J. Demetris3, Katharine Spain4, Sandy Feng5;

1Pediatric Liver Care Center, Cincinnati Children's Hospital, Cincinnati, OH;2Department of Surgery, University of Pittsburgh, Pittsburgh, PA; 3Department of Pathology, University of Pittsburgh, Pittsburgh, PA; 4Federal Systems Division, Rho, Chapel Hill, NC;5Department of Surgery, University of California San Francisco, San Francisco, CA

BACKGROUND: We rigorously describe liver histology in a well-characterized cohort of stable pediatric liver transplant (LT) recipients. METHODS: Screening liver biopsies (bxs) were performed for a multicenter, prospective trial of immunosuppres-sion (IS) withdrawal for subjects <6y at LT, >4y post-LT, with ALT/GGT <50IU/l and no acute or chronic rejection within 2y. Subjects were excluded if LT was for autoimmune disease, HCV or HBV. A central pathologist blindly assessed bxs for por-tal/perivenular (pv) inflammation and fibrosis, interface activity and bile duct damage. RESULTS:49 (24M/25F) deceased/living donor (29D/20L) LT recipients underwent liver bx. Age at bx was 1 1.4±3.3y, age at LT was 1.7±1.5y, and time since LT was 10±3y. LT indications were biliary atresia (n=26), acute liver failure (n=3), tumor (n=3) and other (n=17). Primary IS at LT discharge was tacrolimus (n=36), cyclosporine (n=10) and other (n=3). Ten subjects were induced with basiliximab or ATG. Baseline ALT and GGT were 27±9 and 17±6IU/l. Ishak fibrosis score 0/1/2/3 was seen in 7/21/8/13 subjects; none/mild/moderate pv fibrosis was seen in 17/22/10 subjects; none/mild interface activity was seen in 41/8 subjects (Figure). Abnormal histologic features were correlated. Ishak stage 3 correlated with moderate portal inflammation (p=0.015), moderate pv inflammation (p=0.013), mild interface activity (p=0.001), and moderate pv fibrosis (p=0.0001). Portal and pv fibrosis did not correlate with donor, recipient, LT/graft factors, rejection history or ALT/GGT. Mild interface activity correlated with increased ALT (p=0.042). CONCLUSIONS: Histologic features of chronic allograft injury and fibrosis are common among stable pediatric LT recipients but minimally correlate with donor, recipient, LT/graft, or clinical parameters. Our findings emphasize the independent importance of liver histology to assess long-term allograft health.

image

Disclosures:

Anthony J. Demetris - Consulting: Abbvie, DCL/Novartis, Omnyx

The following people have nothing to disclose: John Bucuvalas, George V. Mazariegos, Katharine Spain, Sandy Feng

1691

Data mining model validated by Scientific Registry of Transplant Recipients database could identify patients with favorable prognosis following liver transplantation for hepatocellular carcinoma

Tomohiro Tanaka1,2, Masayuki Kurosaki4, David Grant1,3, Paul D. Greig1,3, Leslie B. Lilly1,2, Namiki Izumi4, Morris Sherman2;

1Liver Transplant Unit, Multi-Organ Transplant Program, University Health Network, University of Toronto, Toronto, ON, Canada; 2Depart-ment of Medicine, Division of Gastroenterology, University of Toronto, Toronto, ON, Canada; 3Department of Surgery, University of Toronto, Toronto, ON, Canada; 4Department of Gastroenterology and Hepatology, Musashino Red Cross Hospital, Musashino, Japan

Background: The recurrence of hepatocellular carcinoma (HCC) following liver transplantation (LT) is a big concern on evaluating LT candidates with HCC. Several selection criteria for pretransplant HCC have been proposed, although the optimal cutoff of each value for such criteria remains uncertain. Methods: To build a predictive model for recurrent HCC, we performed data mining analysis on patients who underwent LT from 1 996 to 201 0 at University Health Network and were followed at least for 2.5 years or until the recurrence of HCC (n=246). The model was externally validated using a cohort of an first liver transplant registered in the Scientific Registry of Transplant Recipients (SRTR) database (n=9800, receiving LT from May 2003 to July 2012). Results: Among 246 patients, 14.6% (n=36) experienced recurrent HCC within 2.5 years post-LT. Using serum AFP and characteristics of HCC (largest tumor size, number of the tumor and total tumor diameter [TTD]), the risk prediction model for a 2.5-year HCC recurrence identified 3 subgroups; low- (TTD < 4 cm and AFP <73 ng/ml), intermediate- (TTD <4 cm and AFP >73 ng/ml) and high risk group (TTD >4 cm), with a recurrence rate of 4.4, 19.0 and 28.9%, respectively. The cumulative 5-year HCC recurrence rate post-LT were 5, 19 and 35 % in low-, intermediate- and high risk group, respectively (p<0.001). In addition, when compared to low risk group, intermediate- and high risk group showed significantly inferior patient survival (p=0.002). The reproducibility of the model was validated through SRTR database; the 5-year patient survival rate was significantly better in the low risk group than others (74.5% versus 64.2%, p<0.001). This yardstick was revealed to efficiently predict posttransplant survival (hazard ratio 1.53, confidence interval 1.4–1.7, P<0.001), independent of age, gender, MELD score or etiology of liver disease of the recipients. Even when selecting patients within Milan criteria (77.4%, n=7583), patients with low risk showed significantly better posttransplant survival rate than other groups (74.1% vs 63.0% at 5 years, respectively, p<0.001). Conclusions: Grouping LT candidate with pre-LT HCC by the cutoffs of TTD 4 cm and AFP 73 ng/ml which were unearthed by data mining analysis efficiently classify patients according by the posttransplant prognosis. This yardstick could be useful to identify LT candidates with particularly better outcome following LT, among patients within Milan criteria.

Disclosures:

Namiki Izumi - Speaking and Teaching: MSD Co., Chugai Co., Daiichi Sankyo Co., Bayer Co., Bristol Meyers Co.

Morris Sherman - Consulting: Gilead, Merck, Bayer, Arqule, Celsion, Boehringer Ingelheim, Janssen, Abbvie

The following people have nothing to disclose: Tomohiro Tanaka, Masayuki Kurosaki, David Grant, Paul D. Greig, Leslie B. Lilly

1692

Novel sets of genes are correlated with acute cellular rejection after pediatric living-donor liver transplantation

Haruka Shinke1, Miwa Uesugi1, Yuki Okuda1, RIsa Taniguchi2, Kazuo Matsubara1,2, Eri Ogawa3, Atsushi Yoshizawa3, Shinya Okamoto3, Toshimi Kaido3, Shinji Uemoto3, Satohiro Masuda1,2;

1Department of Clinical Pharmacology and Therapeutics, Kyoto University Hospital, Kyoto, Japan;2Department of Pharmacy, Kyoto University Hospital, Kyoto, Japan; 3Divisions of Hepato-Pancreato-Biliary, Transplant and Pediatric Surgery, Department of Surgery, Kyoto University Hospital, Kyoto, Japan

[Background and Purpose] Despite improvements in immuno-suppressive therapy, acute cellular rejection (ACR) remains an important issue in graft loss and patient mortality after living-donor liver transplantation (LDLT). In this study, we aimed to clarify the molecular characteristics in the graft liver immediately before transplantation (zero-biopsy). [Methods] A total of 75 pediatric recipients of LDLT at Kyoto University Hospital were enrolled after written informed consent was provided by their parents. Twelve patients were selected because their backgrounds were quite similar except for ACR status (with/without). The event-free patients were classified as the control group, while those patients who experienced ACR confirmed by pathological examination within 2 weeks after LDLT were classified as the event group. The zero-biopsy gene expression profiles were determined using microarray analysis (Whole Human Genome Microarrays, Agilent Technologies). This study was conducted in accordance with the Declaration of Helsinki and its amendments and was approved by Kyoto University Graduate School and the Faculty of Medicine Ethics Committee. [Results] The systemic clearance of tacrolimus was not significantly different between the 2 groups, suggesting that pharmacokinetic factors did not affect the occurrence of ACR. Using healthy graft liver at zero-biopsy, microarray analysis indicated that 224 genes showed greater than 2-fold increased expression in the event group than in the control group. The interleukin (IL)-1-mediated signaling pathway was the most relative to these genes, followed by the endothelial nitric oxide synthase activity pathway and IL-6- or IL-1 7-mediated signaling pathway. There were common genes among the 3 pathways, such as inhibitor of nuclear factor kappa-B kinase subunit γ (IKK-γ) and activator protein-1 (AP-1). The expression levels of IKK-γ and AP-1 in the event group were increased 2.4- and 2.1-fold compared to the control group, respectively. [Conclusion] The high expression levels of IKK-γ and AP-1 in the zero-biopsy graft liver are suggested to associate with the pathogenesis of ACR after LDLT.

Disclosures:

The following people have nothing to disclose: Haruka Shinke, Miwa Uesugi, Yuki Okuda, RIsa Taniguchi, Kazuo Matsubara, Eri Ogawa, Atsushi Yoshizawa, Shinya Okamoto, Toshimi Kaido, Shinji Uemoto, Satohiro Masuda

1693

Hepatic Toxicity from Rifampin Prophylaxis for Latent Tuberculosis in Liver Transplant Candidates: A Case-Control Study

Allison J. Kwong1, Jennifer C. Lai1, Jennifer L. Dodge2, Bilal Hameed1, Francis Y. Yao1,2;

1Medicine, University of California, San Francisco, San Francisco, CA; 2Surgery, University of California, San Francisco, San Francisco, CA

Background: Prophylaxis for latent tuberculosis infection (LTBI) is recommended for liver transplant (LT) candidates to prevent reactivation of TB after LT, but the prevalence of LTBI and the risk of hepatotoxicity from chemoprophylaxis have not been well characterized. Methods: The study cohort included consecutive patients ≧18 years of age who were listed for primary LT between March 2009 and March 2013. Exclusion criteria were acute liver failure as the indication for LT or a history of active TB. Cases were those with LTBI defined as a positive PPD or quantiferon test without active TB. Controls were patients without LTBI, matched by age, gender, diagnosis of hepatocel-lular carcinoma, and baseline MELD score. Our protocol for chemoprophylaxis in patients with LTBI was rifampin 300 mg twice daily for 4 months while on the LT waiting list. The primary outcome of this study was hepatotoxicity defined as an increase in total bilirubin by >2 mg/dL or an increase in MELD score by >5 points within 6 months after initiating rifampin. The risk of hepatotoxicity was compared between cases and controls with chi-square p values accounting for matched data, and predictors of hepatotoxicity were assessed in LTBI cases by logistic regression. Results: Of 989 patients listed for LT during the study period who met inclusion criteria, 607 (61%) completed PPD or quantiferon testing. Among them, 101 patients (17%) were diagnosed with LTBI: 74% were male; 35% were Asian, 25% were white, 23% were Hispanic, 8% were Arab or Middle Eastern, 6% were black, and 4% were of other races. Thirty-two of 1 01 patients received rifampin for LTBI chemoprophylaxis. Compared to 162 matched controls, 38% of cases versus 1 8% of controls had an increase in total bilirubin by >2 points (p=0.006). An increase in MELD score of >5 points was observed in 41 % of cases and 1 8% of controls (p=0.004). The risk of hepatotoxicity, based on an increase in both total bilirubin by >2 points and MELD by >5 points, was 38% in the rifampin group versus 1 1% in the controls (p<0.001). Rifampin was discontinued prematurely in 47% patients, mainly due to hepatotoxicity. In an exploratory analysis, only age (odds ratio 0.88 per year, 95%-CI 0.79–0.99, p=0.03) predicted hepatotoxicity from rifampin, whereas gender, baseline MELD score, and etiology of liver disease did not. Conclusion: We observed a 17% prevalence of LTBI among LT candidates. Compared to matched controls, rifampin use is associated with a significantly greater risk of hepatotoxicity. Re-appraisal of the most appropriate regimen and timing of chemoprophylaxis for LTBI based on its risks and potential benefits is needed.

Disclosures:

The following people have nothing to disclose: Allison J. Kwong, Jennifer C. Lai, Jennifer L. Dodge, Bilal Hameed, Francis Y. Yao

1694

Programmed liver biopsies after orthotopic liver transplantation indicate occurrence of non-alcoholic steato-hepatitis during a 5-year follow up

Mona Petri1, Eva Müller1, Arno Schad4, Maria Hoppe-Lotichius2,3, Sandra Koch1, Jörn M. Schattenberg1, Tim Zimmermann1, Jens Mit-tler2, Peter R. Galle1, Hauke Lang2, Gerd Otto3, Arndt Weinmann1, Martin F. Sprinzl1;

1I Medical Department, University Mainz, Mainz, Germany; 2Derpartment of Surgery, University Mainz, Mainz, Germany; 3Department of Transplant Surgery, University Mainz, Mainz, Germany; 4Institute of Pathology, University Mainz, Mainz, Germany

Background: Metabolic deterioration regularly occurs after orthotopic liver transplantation (OLT) and may affect hepatic pathology and outcome. Following a previous study which identified incipient non-alcoholic liver disease (NAFLD) in context of a post OLT metabolic syndrome (1), we investigated whether early histological changes are predictive for occurrence of steatohepatitis (NASH) during long-term follow up. Material and Methods: Patients (N = 144) receiving OLT between 1 998 and 2007 at a single transplant center were retrospectively analyzed over a period of 5 years post OLT. OLT with a complete clinical history and a programmed 1- and 5-year liver biopsy were included. Histology was reviewed for NAS score and liver fibrosis by an experienced pathologist. Diagnosis of NAFLD required a stenosis >5% of hepatocytes and NASH was identified by a minimum NAS score of >3 points. Diagnosis of NASH further required exclusion of ongoing viral hepatitis. Results: Eligible OLT patients received programmed 1-and 5-year biopsies after a median interval of 382 days and 1832 days post OLT, respectively. 1-year biopsies identified NAFLD in 38.2% (N=55/144) and NASH in 4.9% (N=7/144) of our cohort. Following 5-year biopsies showed an increase of NAFLD to 52.1 % (N=75/1 44) and of NASH to 12.5% (N = 1 8/144) during follow up. The proportion of steatotic hepatocytes within liver sections followed this trend and increased from an average of 7.67% to 12.34% (p = 0.002). Mild NAFLD findings at the 1-year biopsy correlated with the development of NASH during follow up (P = 0.006). However, NAFLD 1-year after OLT was not predictive for the development of hepatic fibrosis. Conclusion: Metabolic changes after OLT impact the pathology of liver grafts as shown by programmed liver biopsy. Mild NAFLD findings 1-year after OLT correlated with NASH after 5 years and indicated progressive liver disease during long-term follow up. However these findings were not associated with marked fibrosis, which probably shows a later onset in context of post OLT NASH. (1) MF. Sprinzl et al. Metabolic syndrome and its association with fatty liver disease after orthotopic liver transplantation.Transpl Int. 2013 26(1):67–74.

Disclosures:

Peter R. Galle - Advisory Committees or Review Panels: Bayer, BMS, Lilly, Daiichi, Jennerex; Consulting: Medimmune; Grant/Research Support: Roche, Lilly; Speaking and Teaching: Bayer, BMS

Arndt Weinmann - Speaking and Teaching: Bayer Healthcare

The following people have nothing to disclose: Mona Petri, Eva Müller, Arno Schad, Maria Hoppe-Lotichius, Sandra Koch, Jörn M. Schattenberg, Tim Zimmermann, Jens Mittler, Hauke Lang, Gerd Otto, Martin F. Sprinzl

1695

Can Cardiopulmonary Reserve Predict Post-Liver Transplant Survival and Hospital Stay

Mohammad U. Malik1, Stuart D. Russell2, Shashank Garg3, Carol B. Thompson4, Aliaksei Pustavoitau5, Andrew M. Cameron6,1, Ahmet Gurakar1,6;

1Gastroenterology/Liver Transplant, The Johns Hopkins Hospital, Baltimore, MD; 2The Johns Hopkins Heart and Vascular Institute, The Johns Hopkins Hospital, Baltimore, MD; 3Internal Medicine, Johns Hopkins University-Sinai Hospital Program in Internal Medicine, Baltimore, MD; 4Johns Hopkins Biosta-tistics Center, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; 5Johns Hopkions Anesthesiology and Critical Care Medicine, The Johns Hopkins Hospital, Baltimore, MD; 6 Surgery/Transplant surgery, The Johns Hopkins Hospital, Baltimore, MD

Background: Cardiopulmonary reserve may influence liver-transplant outcomes. While cardiac function has been previously studied, there is a scarcity of literature for pulmonary function and its association with liver-transplant outcomes. This study explores pulmonary and cardiac indices to predict post-transplant mortality and hospital stay in liver transplant recipients. Methods: After IRB approval, the charts of patients who had undergone cadaveric liver transplantation at Johns Hopkins Hospital, between January, 2009 to March, 2013 were retrospectively reviewed. Collected information included left ventricular ejection fraction (LVEF), right ventricular systolic pressure (RVSP) and dobutamine stress echo (DSE) indices etc. The pulmonary variables included forced vital capacity (FVC), total lung capacity (TLC)(% predicted) etc. In addition, liver and renal function indices were also recorded. Patients with limited graft survival and pediatric patients were excluded. Results: 165 patients were included in the study. There were 1 1 1 males (M: F=2.1).A post-operative mortality of 1 8.2% was observed over a median follow-up time of 1.8 years (range 0.008–4.09). The one-year survival of the cohort was 86%. The median hospital stay was 13.5 days (range 6–1 97).Patients who died had higher anion gap (p=0.003), RVSP (p=0.04) and inability to reach 85% predicted heart rate on DSE (p=0.01). The univari-ate Cox regression revealed age, anion gap (AG), RVSP, DSE and FVC to be predictive of post-transplant mortality. In multi-variate analysis, BMI (HR: 0.65, 95%CI: 0.47–0.87), AG (HR: 1.93, CI: 1.12–3.31), RVSP (HR: 1.31, CI: 1–1.7) and FVC (HR: 0.91, CI: 0.83–0.99) were predictive of mortality. An RVSP > 40 was associated with higher mortality (HR: 3.3, CI: 1.29–8.4) RVSP of ≧ 42 was associated with prolonged postoperative stay (log-rank test p=0.05). A FVC >70% was associated with lower mortality (HR: 0.25, CI: 0.09–0.69) Conclusion: Transplant recipients with limited cardiorespiratory reserve are at higher risk of post-transplant mortality. These patients could develop complications that may prolong their post-operative hospital stay. Optimal liver transplant outcomes may be achieved after triaging and appropriately managing patients with limited cardiorespiratory reserve.

image

Disclosures:

Ahmet Gurakar - Advisory Committees or Review Panels: Gilead, Gilead; Grant/Research Support: BMS, BMS

The following people have nothing to disclose: Mohammad U. Malik, Stuart D. Russell, Shashank Garg, Carol B. Thompson, Aliaksei Pustavoitau, Andrew M. Cameron

1696

The evolving use of higher risk grafts is associated with an increased incidence of acute kidney injury after liver transplantation

Joanna A. Leithead1,2, Bridget Gunson2, Paolo Muiesan1, James W. Ferguson1;

1Liver Unit, Queen Elizabeth Hospital, Birmingham, United Kingdom; 2NIHR Biomedical Research Unit and Centre for Liver Research, University of Birmingham, Birmingham, United Kingdom

The growing discrepancy between supply and demand for liver transplantation has necessitated a greater use of higher risk grafts. Donation after Cardiac Death (DCD) liver transplant recipients have an increased frequency of acute kidney injury (AKI). Consequently, graft injury has been implicated in the pathogenesis of renal dysfunction in this setting. Our aim was to examine the effect of the evolving use of marginal grafts on the incidence of AKI after liver transplantation. Methods: A single-centre study of 1 152 patients undergoing first-single-organ liver transplantation for chronic liver disease 01/2000–12/201 1. To assess the impact of the evolution of graft quality over time; donor/graft/recipient variables were compared over three 4-year time periods: 01/2000–12/2003, 01/2004–12/2007, 01/2008-12/2011. Exclusion criteria: pre-trans-plant renal replacement therapy, death by postoperative day 7. Definitions: peri-operative AKI - creatinine>2 times baseline (KDIGO 2012 guidelines stage 2/3); chronic kidney disease -eGFR <60ml/min/1 .73m2. Results: Recipient preoperative eGFR improved over the follow-up period (p<0.001), and the median postoperative day 1 (p<0.001), 2 (p<0.001) and 3 (p<0.001) tacrolimus trough levels fell. There was a progressive reduction in cold (p<0.001) and recipient warm (p<0.001) ischaemic times. However, the proportion of DCD grafts increased (p<0.001), and there was a progressive increase in donor age (p<0.001) and donor body mass index (BMI) (p=0.01 3) in the whole donation after brain death (DBD) grafts. The mean donor risk index was 1.60, 1.65 and 1.90 in 01/2000-12/2003, 01/2004-12/2007 and 01/2008-12/2011, respectively (p<0.001). Patient survival improved (log-rank p=0.003). The incidence of AKI increased over the 3 study periods (01/2000-12/2003, 28.4%; 01/2004-12/2007, 31.4%; 01/2008-12/2011, 39.1%; p=0.006). AKI was an independent risk factor for chronic kidney disease (HR 1.32; 95% CI 1.08–1.61, p=0.007). On multivariate analysis, variables associated with AKI included: DCD liver (OR 2.04, p=0.011), donor age >60 years (OR 1.60, p=0.021), donor BMI >30 (OR 1.82, p=0.006), donor serum sodium >165 mmol/l (OR 2.73, p=0.037) and increasing recipient warm ischaemic time (OR 1.02, p=0.046). Conclusion: The increasing use of marginal liver grafts is associated with an increased incidence of AKI. These findings emphasise the need for renal injury minimising strategies in liver transplant recipients.

Disclosures:

James W. Ferguson - Advisory Committees or Review Panels: Astellas, Novartis

The following people have nothing to disclose: Joanna A. Leithead, Bridget Gun-son, Paolo Muiesan

1697

Accuracy of transient elastography (FibroScan®) in evaluation of liver fibrosis after liver transplantation: comparison with liver biopsy

Bianca Della Guardia1, Andréia S. Evangelista1, Guilherme Felga1, Lidiane V. Marins2, Marcio D. Almeida1;

1Liver Transplantation, Hospital Israelita Albert Einstein, Sao Paulo, Brazil; 2Anato-mopathology, Hospital Israelita Albert Einstein, Sao Paulo, Brazil

Introduction: Liver biopsy (LB) is often performed in liver transplant patients to determine the evolution of the graft, especially in the case of diseases with potential for relapse, such as hepatitis C. Despite being the reference method, LB has a risk of complications. Transient elastography (TE) has been shown to be a good noninvasive method compared to LB to determine the degree of liver fibrosis in various diseases. Objective: To determine the accuracy and the cutoffs of TE in liver transplant patients compared with the LB. Patients and methods: LBs were evaluated in liver transplant performed at the Hospital Israelita Albert Einstein between 201 0 and 2012. TE was performed at the time of LB. The median value of ten successful measurements was kept as a representative of the liver stiffness, with interquartile range less than 30% of the median value. The results were expressed in kiloPascals (kPa). The METAVIR classification was used for fibrosis graduation in LB. ROC curves with 95% of confidence intervals were used to assess the performance TE. The definition of cutoff points was made to ensure specificity of > 90% for all fibrosis stages (F0-F4). Results: LB was performed in 267 patients. TE failed in 8 (3%) due BMI>30. The optimal liver stiffness cut-off value and diagnostic performance are summarized in Table 1. Discussion: TE demonstrated good performance for each degree of fibrosis in liver transplant recipients. TE can be considered an alternative to LB in post-liver transplantation for fibrosis staging.

Table 1: TE performance measures

fibrosiscutoffnsensibilityspecificityPPVNPV
  1. PPV (positive predictive value); NPV (negative predictive value)

F>18,11200,510,900,980,20
F>212,3530,430010,720,76
F>315,13)0,590,90013)0,96
F=41.60,1281.601.900,111.60

Disclosures:

The following people have nothing to disclose: Bianca Della Guardia, Andréia S. Evangelista, Guilherme Felga, Lidiane V. Marins, Marcio D. Almeida

1698

Complications of Deceased Donor Liver Transplant: A Meta-Analysis of Reported Incidences

Lisa McElroy, Amna Daud, Ashley E. Davis, Brittany Lapin, Talia B. Baker, Michael M. Abecassis, Jane L. Holl, Daniela Ladner;

Center for Healthcare Studies, Northwestern University Feinberg School of Medicine, Chicago, IL

Purpose: To understand the magnitude of the burden of postoperative complications on liver transplant recipients. Methods: A systematic review of the literature was performed to identify studies that reported incidence of postoperative complications in deceased donor liver transplant recipients. Articles were selected for inclusion if they reported an incidence of postoperative complications in deceased donor liver transplant recipients based on human studies. Articles were excluded if they were not available in English, were published prior to January 1, 2002, or originated from a location other than the United

States, Australia, Canada or the United Kingdom. Data from the selected manuscripts were independently extracted from each study by 3 reviewers (LM, AD, AED). Disconcordant decisions were solved by a consensus among the investigators with guidance from a fourth reviewer (DPL). Incidence rates were calculated as percentages, and heterogeneity was assessed using the Cochran Q chi-square test and the I2 statistic. Results: Our original search criteria returned 8159 studies. After preliminary exclusions and supplemental searches, 134 studies were selected for full text review. In total, our meta-analysis includes 29,227 deceased donor liver transplant recipients from 75 studies performed between 2002 and 2012. The pooled incidence rate of biliary leak and stricture was 0.079 (95% CI = 0.055 to 0.107) and 0.125 (95% CI = 0.099 to 0.154) respectively. The pooled incidence of hepatic artery thrombosis was 0.045 (95% CI=0.031 to 0.062), and postoperative hemorrhage was 0.083 (95% CI = 0.046 to 0.131). The estimated mean incidence of acute renal failure was 0.299 (95% CI = 0.220 to 0.385). The pooled incidence of intra-abdominal abscess was 0.044 (95% CI = 0.01 9 to 0.078). We found evidence of statistically significant heterogeneity for all but one complication (Q statistic, p<0.001; I2 range 80.1–99.0%). Conclusions: Our results show that although the absolute incidence of major postoperative complications varies, further efforts are needed to standardize the monitoring of postoperative morbidity in liver transplant recipients. Improved knowledge of the incidence of these complications will allow for more accurate estimation of the effect of interventions aimed at improving patient safety.

Disclosures:

Michael M. Abecassis - Management Position: Transplant Genomics, Inc

The following people have nothing to disclose: Lisa McElroy, Amna Daud, Ashley E. Davis, Brittany Lapin, Talia B. Baker, Jane L. Holl, Daniela Ladner

1699

Long term Patient and Graft-Related Outcomes after Liver Transplantation for Primary Sclerosing Cholangitis

Siddharth Singh1, Jayant A. Talwalkar1, J. Eileen Hay1, John J. Poterucha1, Gregory J. Gores1, Michael R. Charlton1, Charles B. Rosen2, Julie Heimbach2, Russell H. Wiesner1;

1Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN; 2Transplant Surgery, Mayo Clinic, Rochester, MN

Background: Liver transplantation (LT) is the only effective treatment option for patients with end-stage liver disease due to PSC. Aims: To determine whether clinical outcomes have improved over time among subjects undergoing liver transplantation for PSC. Methods: We identified all patients who underwent LT for advanced stage PSC for non-cholangiocarci-noma indications at Mayo Clinic, Rochester from 1998–2008, with follow-up through 2012. Information on long-term patient and allograft survival, recurrence of PSC (rPSC) (based on Mayo Clinic criteria), acute and chronic rejection, as well as associated IBD was extracted and compared with similar outcomes in 150 consecutive patients who had undergone LT for PSC from 1985–1996. One-, 5-, 10- and 15-yr event-free survival was estimated using Kaplan-Meier curves. Results: 101 patients underwent LT for PSC (mean age, 48 yrs; 62 males). The follow-up period was a median (IQR) 8.4 (5.5–10.7) yrs. When compared to the time period 1985–1996, significant improvements in long-term patient and graft-survival were observed in the recent era (Table 1). Solid-organ cancer and sepsis were leading causes of mortality after LT in the current time period in contrast to sepsis and operative mortality in the previous analysis. Twenty-five patients (25 %) developed rPSC over a mean follow-up of 8.8 yrs as compared to 20% over a mean follow-up of 4.6 yrs in the previous cohort (p=0.150). When comparing 1998–2008 to 1985–1996, the frequency of steroid-responsive rejection (50.5% v. 68.7%; p<0.001), steroid-resistant acute cellular rejection (6.9% v. 21.3%; p<0.001) and chronic ductopenic rejection (3.0% v. 8.0%; p=0.006) was significantly lower in the current era. Notably, tacrolimus-based immunosuppression (94% v. 21.3%; p<0.001) was more common in the current era. The frequency of associated IBD (78% v. 79%; p=0.974), pre-transplant colec-tomy (31.2% v. 43.7%; p=0.917) and post-transplant colec-tomy (1 9.7% v. 11%; p=0.467) were comparable across time periods. Conclusions: Long-term patient and graft-survival have improved even further for patients with PSC undergoing LT. These differences may be related to improved operative techniques, switching to tacrolimus-based immunosuppression, and earlier recognition and treatment of complications.

Rate of eventPatientSurvivalGraft Survival
 1998–20081985–19961998–20081985–1996
1 -year98.0%93.7%94.1%83.4%
5-year95.0%86.4%90.1%79.0%
10-year87.1%99.0%85.1%60.5%
15-year84.1%-83.2% 

Disclosures:

Jayant A. Talwalkar - Consulting: Lumena; Grant/Research Support: Intercept, Salix, Gilead

Gregory J. Gores - Advisory Committees or Review Panels: Bayer, Chugia, Dai-ichi, Generon, Conatus, IntegraGen

The following people have nothing to disclose: Siddharth Singh, J. Eileen Hay, John J. Poterucha, Michael R. Charlton, Charles B. Rosen, Julie Heimbach, Russell H. Wiesner

1700

Sequential liver transplantation using familiar amy-loidotic polyneuropathy livers: one solution for organ shortage or a new problem?

Suzana Calretas, Nuno Silva, José Ferrão, Luis Tomé, Antonino Barros, Dulce Diogo, Ana Eufrasio, Fernando M. Pinto, Luciane Pereira, Carlos Seco, Helena Vieira, Carlos Bento, Emanuel Fur-tado;

Unidade de Transplantação Hepática Pediátrica e de Adul-tos, Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal

Familiar Amyloidotic polyneuropathy (FAP) is an inherited autossomal dominant systemic disease caused by a point mutation in transthyretin (TTR), the most common TTR Met30. It becomes clinically apparent in the third decade and inexorably progresses to death in 1 0 to 15 years. As the liver is the main source of the circulating mutant form of TTR, orthotopic liver transplantation was initiated in 1990 as a treatment for FAP. Besides the production of the abnormal protein, no other liver functional or anatomic disturbance is present. In 1995, Fur-tado et al performed the first domino liver transplant, using one FAP liver, and since then, a large number was done worldwide. It was expected that the TTR Met30 producing grafts in genetically non-affected patients would need at least the same time to develop symptoms. Time is telling us something different from what was expected. In our Liver Unit 81 sequential liver transplants were made. To date, 6 patients were retransplanted for iatrogenic amyloidotic polyneuropathy. All men, mean age at the sequential liver transplant 48.8 years. The indication was alcoholic cirrhosis (2), hepatitis B cirrhosis (1), hepatocellular carcinoma (1), chronic rejection after immunosuppression weaning for Kaposi syndrome meanwhile cured (1), and liver metastasis of neuroendocrine tumor (1). Fist symptoms appeared in average in 80 months (4,5 to 8,5 years), mainly diarrhea and sensitive neuropathy. Two had rhythm disturbances that required pacemaker. Electromyography was abnormal in all. Retransplant was performed in average 132 months (±1 1 years) after the sequential liver transplant. By that time, mean age was 60 years (53–69), mean body mass index 26,5 Kg/m2. All of them where received a cadaveric liver, using piggy-back technique. They were discharged from ICU in 2 to 5 days and from the ward in 15 days (9–21). Explanted livers showed normal architecture (5), fibrosis F1 (1), and all of them had amyloid deposition. One patient died from septic shock three months later; the other 5 are alive (follow-up time from 1 to 26 months). Conclusions: The use of FAP livers is an effective alternative to expand the pool of donors, allowing to treat patients that who otherwise would not. However, the appearance of the disease in non-genetically susceptible individuals is earlier than originally deducted from the natural history of the disease (although not universal). Strict criteria should be used when selecting the receptors, and they should be kept close watch.

Disclosures:

The following people have nothing to disclose: Suzana Calretas, Nuno Silva, José Ferrão, Luis Tomé, Antonino Barros, Dulce Diogo, Ana Eufrasio, Fernando M. Pinto, Luciane Pereira, Carlos Seco, Helena Vieira, Carlos Bento, Emanuel Fur-tado

1701

Hepatic hemodynamic disturbances in cirrhotic children: correlation to PELD and portal vein complications after liver transplantation

Catherine de Magnée1, Karlien Carbonez2, Dana Dumitriu3, Renaud Menten3, Philippe Clapuyt3, Thierry Pirotte4, Francis Vey-ckemans4, Raymond Reding1;

1Pediatric surgery and transplantation, St Luc University Hospital, Brussels, Belgium; 2Pediatric cardiology, St Luc University Hospital, Brussels, Belgium; 3Pediatric radiology, St Luc University Hospital, Brussels, Belgium; 4Pediatric anesthesiology, St Luc University Hospital, Brussels, Belgium

Background: Liver hemodynamic disturbances and portal vein (PV) complications are frequently observed in pediatric liver transplantation (LT). We hypothesized that cirrhosis in children is associated with liver hemodynamic features which could lead to vascular complications after LT. Methods: We retrospectively analyzed 1 98 LT recipients (median age: 1.6 year; range: 0.2–17.5), including 130 biliary atresia (BA). Pre-LT hemodynamics were studied using Doppler ultrasound, and correlated with the incidence of post-LT vascular complications. In children with BA-related PV hypoplasia, the technique used for PV reconstruction was correlated with the incidence of post-LT PV complications. Moreover, intraoperative flowmetry of liver hemodynamics was studied prospectively at LT in 10 cirrhotic children (median age: 1.0 year; range: 0.5–14), including 8 BA. The gradient between PV pressure and central venous pressure (PVP-CVP) was also invasively measured to estimate the severity of portal hypertension in the native liver during the LT procedure. Results: At retrospective analysis, hepatic arterial resistance index > 1 at pre-LT ultrasound was associated with a higher rate of PV complications (p=0.041). In children with BA-related PV hypoplasia, the portoplasty technique alleviated the extra risk of PV complications (6.2% with portoplasty versus 1 9.3% without portoplasty). The prospective hemodynamic studies showed that: (1) pediatric cirrhosis was associated with a reduction of pre-LT total liver flow of 50% (median: 50.4 ml/min/100gr of liver; range: 1 9–1 1 0.5), compared to expected values (100 ml/min/1 00gr); (2) this reduction of total liver flow was correlated with PELD (r=0.32); (3) PVP-CVP gradient was high (median: 14.5 mmHg; range:-7–26), and was also correlated with PELD (r=0.54). Conclusions: Pediatric cirrhosis is associated to severe liver hemodynamic disturbances, which are correlated with PELD. Pre-LT hemodynamic assessment may help to predict, and efficiently manage liver vascular anomalies in pediatric LT, including BA-related PV hypoplasia.

Disclosures:

The following people have nothing to disclose: Catherine de Magnée, Karlien Carbonez, Dana Dumitriu, Renaud Menten, Philippe Clapuyt, Thierry Pirotte, Francis Veyckemans, Raymond Reding

1702

Why do patients die after liver transplantation: Part II ?

Kimberly Daniel, Michael R. Lucey, Jens C. Eickhoff;

Univ. of Wisconsin School of Medicine and Public Health, Madison, WI

More patients are surviving liver transplantation (LT) for many years. We hypothesized that the causes of death (COD) would change as LT recipients achieve extended survival. Previously, we analyzed death patterns in the UNOS LT database from 9/30/1987 to 9/3/2010 (Hepatology 2011, 54: 651A). AIM: to gain further insight into COD after LT, we reviewed the COD in a LT cohort from a US academic medical center (AMC). Methods: the AMC cohort was recruited from July 1984 until December 2010, with follow up until 4/26/2012. COD was recorded in most cases at the time of death. Where no COD was recorded in the database, we reviewed the patient record and ascribed a COD. COD were bundled into 9 categories. Results: In the AMC cohort, there are 1458 LT patients, 1324 had a single transplant, 121 had 2, 1 1 had 3, 1 each had 4 and 5 transplants. 734 (50.3%) recipients died. 205 (27.9%) deaths occurred within 1 year, 193 (26.3%) >1<5 years, and 336 (43.8%) at >5 years after LT. In contrast, the distribution of deaths in UNOS dataset was as follows: year 1: 12,860 deaths (43.1%), years 2–5: 8,921 deaths (29.9%),> 5 years: 8,061 deaths (27%). In the UNOS dataset, the frequency of 'unknown' as a COD was 13.8%, and increased with longer survival: 20% for those dying in year 6; 25% after year 1 0. In the AMC cohort, 84 (1 1.4%) of CODs were unknown, including 20% after 5 years. In both cohorts, the pattern of COD in relation to length of survival was similar for death due to infection, which was the most common COD in year 1 (37.6% AMC; 28% UNOS) after which infection as a primary COD declined, and for cardiovascular COD, which was relatively stable across all survival intervals. The pattern for malignant COD differed in the two groups. In the UNOS dataset the pattern was as follows, 5% in year 1, > 1 < 5 years 20% and declining to 14% for all years thereafter. In contrast, in the AMC cohort, there were 13/205 (6.5%) malignant deaths in year 1, increasing in the > 1 < 5 years to 53/193 (27.5%), and remaining significant in the subsequent years: 76/336 (22.6%). Further analysis indicated that the differences, particularly for deaths occurring after 5 years, were due to de novo tumors, such as lung or colon, rather than recurrent hepatoma or cholangiocarcinoma. Conclusion: an AMC LT cohort had a greater proportion of deaths occurring after longer survival than the national dataset. In this setting, the fatal consequences of de novo malignancies became apparent. Addressing the risk factors underlying tumor development and maintaining appropriate screening for these cancers is imperative for the long-term care of this at risk population.

Disclosures:

Michael R. Lucey - Grant/Research Support: Vertex, Abbvie, Gilead, Salix; Speaking and Teaching: Roche

The following people have nothing to disclose: Kimberly Daniel, Jens C. Eickhoff

1703

Predictors and Survival of Redo Simultaneous Liver-Kidney Transplantation versus Redo Liver Alone Transplantation

Neehar D. Parikh, Brittany Lapin, Talia B. Baker;

Comprehensive Transplant Center, Northwestern University, Chicago, IL

Background: Redo liver transplantation (rLT) due to allograft failure is often a complex and technically difficult operation with high risks of complications and questionable long-term outcomes. The addition of a simultaneous kidney transplant to the rLT increases the potential for complications, however this has not been systematically studied. Objective: In this study, our aim was to identify the predictors, incidence, and survival of patients undergoing redo simultaneous liver-kidney (rSLK) transplantation versus rLT alone. Methods: We conducted a secondary analysis of the OPTN database from 2002–2012 of patients who underwent redo transplantation. We stratified patients by rLT and rSLK and compared recipient and donor characteristics of both the first and second transplantation. In order to determine predictors of rSLK vs rLT, we constructed a stepwise regression analysis adjusting for recipient and donor variables. We also analyzed unadjusted and adjusted graft and patient survival after rLT and rSLK using a Cox-regression model. Results: During the study period, there were a total of 255 rSLK transplants and 2,744 rLTs. Predictors of time to rSLK vs. rLT were post LT dialysis (HR: 6.33 95% CI: 3.69–10.84), history of coronary artery disease (HR: 3.23 95% CI: 1.33–7.86), status 1 listing (OR: 4.11 95% CI: 1.29–13.13), primary donor cold ischemia time (HR: 0.91 95% CI: 0.84–0.99), and creatinine at primary LT (HR: 1.18 95% CI: 1.04–1.35). rSLK patients had a one and 5-year graft survival of 70.7% and 54.6%, respectively, compared to rLT recipients who had rates of 67.6% and 51.1%, respectively (p=0.58). There were no differences in adjusted patient or graft survival rates between rSLK and rLT recipients. In a subanalysis excluding those patients listed as a status 1 after the first transplantation, the survival remained equivalent between rLT and rSLK recipients. Conclusions: The majority of redo LT conducted in the US is rLT alone, however predictors for need for rSLK include post LT renal failure, coronary artery disease, and elevated creatinine at time of primary LT. Outcomes after rLT are poor with 5-year patient survival at approximately 60% and graft survival at approximately 50%. There was, however, no differences in survival rates between rLT and rSLK repeat transplants.

Disclosures:

The following people have nothing to disclose: Neehar D. Parikh, Brittany Lapin, Talia B. Baker

1704

Late presentation in the MELD era is still a risk for death: A competing risks analysis of patients presenting for liver transplantation

Jennifer T. Higa1, Kenneth Mukamal1, Marina Anastopoulos2, Susan McDermott2, Michael P. Curry1,2;

1Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA; 2Transplant Institute, Beth Israel Deaconess Medical Center, Boston, MA

Liver Transplant (LT) is an effective treatment for end stage liver disease.Prioritization for LT is based on medical urgency rather than waiting time. We compared risk of death with that of listing for LT and identified factors associated with death or listing using a competing risks analysis (CR) in patients referred for LT. We studied a cohort of patients referred for LT at a single center from 1 /1 /02 to 7/31/12. Patients were excluded for acute liver failure,re-transplant and those with incomplete records.

Data were retrieved from electronic medical records and the Social Security Death Index.Covariates included demograph-ics,insurance,employment,education,marital status, etiology of ESLD, substance use,comorbidities,ascites,encephalopathy and HCC. Of the 916 patients evaluated 64% were listed,19% unlisted/alive at end of study and 1 7% died before listing.At presentation,57% had MELD <15,34% MELD 15–24,and 9% MELD >25. There were 31 % who underwent LT,40% who died (with or without listing) and 29% alive at close of study.Cox and CR analyses identified several independent variables associated with listing and death prior to listing (Table 1);hazard ratios (HR) did not differ between the two methods.The HR associated with MELD score >25 was significantly greater for death than for listing in CR analyses (HR 15.29 vs. 5.22, p=0.001).Encephalopathy at presentation was more strongly associated with death (HR 1.41, 95%CI 0.76–1.09) than with listing (HR 0.97, 95%CI 0.97–2.06;p difference=0.04).The HR associated with MELD score > 25 did not differ for outcomes of transplantation versus death prior to transplantation (p=0.79). Across all MELD scores,patients are more likely to get listed than die without listing. However, when compared with a MELD score < 15,a MELD score > 25 at presentation had a significantly stronger association with death than listing for LT.Encephalopathy at presentation is also more strongly associated with death than listing.Even in the era of liver allocation based on medical urgency,patients should ideally be referred prior to onset of encephalopathy or MELD > 25.

Table 1: Competing risks analysis

 Outcome Death HR/95%CIOutcome Listed HR/95%CI
  1. CVD=cardiovascular disease

MELD 15–241.88/1.27,2.791.53/1.26.1.85
MELD > 2515.29/7.85,29.765.22/3.55,7.67
Alcoholic liver disease0.60/0.38,0.940.79/0.62,1.00
Age1.34/1.07,1.69nrS
Employed0.47/0.27,0.791.37/1.12,1.67
HCC2.15/1.39,3.321.71/1.35,2.16
No InsurancerS0.20/0.05,0.77
Private insurance1.76/1.18,2.60US
CVDNS0.64/0.47,0.86

Disclosures:

Michael P. Curry - Grant/Research Support: Gilead Sciences, Mass Biologics, Merck, Ikaria; Stock Shareholder: Achilion, Idenix

The following people have nothing to disclose: Jennifer T. Higa, Kenneth Mukamal, Marina Anastopoulos, Susan McDermott

1705

Hyperglycaemia within 14 days of liver transplantation predicts new onset diabetes after transplantation (NODAT)

Anna Turner2,1, Aidan Pierce2,1, Hugo Farne1, Stephen Kriese1, Debbie Shawcross1,2, Nigel Heaton1, Michael A. Heneghan1;

1Institute of Liver Studies and Transplantation, King's College Hospital, London, United Kingdom; 2King's College London, London, United Kingdom

Background and Aims: New onset diabetes after transplantation (NODAT) is an important complication of liver transplantation associated with greater graft rejection, sepsis, renal impairment and biliary complications. The aetiology of NODAT is unknown but recognized risk factors include calcineurin inhibitor immunosuppression and hepatitis C infection. There is limited evidence that hyperglycaemia within 14 days post-transplant may also contribute to NODAT. Methods: We retrospectively studied 148 consecutive patients who received a liver transplant in 2009 at King's College Hospital. We gathered demographic and biochemical data, pre-transplant UKELD and MELD scores, aetiology of liver disease, daily glucose kinetics, total daily insulin requirement in the 14 days post-transplant, requirement for diabetes medications at 1 year, immunosup-pression therapy (including steroids), intensive care and total hospital stay. These variables were analysed to identify those predictive of NODAT using Fisher's exact test, unpaired t test and logistical regression. Results: Of the 148 patients identified, 58 had complete blood glucose data and were included in the analysis. 11/58 had pre-transplant diabetes; 7/58 developed NODAT and 40 did not. Key results are shown in the table. Of note, hyperglycaemia in the 14 days post-transplant was associated with NODAT (p=0.003), as did hepatitis C (p=0.01 8) and lower average plasma tacrolimus concentration on days 1–6 post-transplant (p=0.005). Conclusion: This is the largest study reported to date to demonstrate that hyperglycaemia within the first 14 days post-transplant is predictive of NODAT, in a population given moderate doses of steroid post-transplant (20mg prednisolone daily). Hepatitis C predisposes to NODAT but contrary to previous reports higher tacrolimus concentrations did not confer increased susceptibility to NODAT. Whilst the pathophysiology remains unclear, being vigilant for this important complication and starting hypergly-caemic management in a timely fashion may reduce the incidence of and morbidity associated with NODAT.

Results

Median (range)NODAT (n = 7)Non-NODAT, Non-DM (11=40)P
  1. 1Fisher's exact test

  2. 2Unpaired t test

Male5190.4161
Age (years)53 (38–59)54(19–73)0.8102
Hyperglycaemia in 14 days post-transplant (>15-<20mmol/L)550.0031
Hepatitis C450.0181
Tacrolimus concentration (g/L, average day 1–6)4.0 (2.6–6.6)7.0(3.7–17.9)0.0052

Disclosures:

Debbie Shawcross - Advisory Committees or Review Panels: Norgine; Speaking and Teaching: Norgine

Nigel Heaton - Advisory Committees or Review Panels: Novartis, Roche; Speaking and Teaching: Astellas

The following people have nothing to disclose: Anna Turner, Aidan Pierce, Hugo Farne, Stephen Kriese, Michael A. Heneghan

1706

Cystatin C and eGFR can be utilised in the prediction of cardiovascular (CV)and non cardiac death post ortho-topic liver transplantation (OLT)

Norma C. McAvoy1, Alasdair W. Hay2, Graham McKillop3, David E. Newby4, Peter C. Hayes5,1;

1Department of Hepatology, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom; 2Dept of Anaesthetics, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom; 3Dept of Radiology, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom; 4Centre for Cardiovascular Sciences, University of Edinburgh, Edinburgh, United Kingdom; 5University of Edinburgh, Edinburgh, United Kingdom

Background:OLT is a valuable resource with increasing demand.Accurate CV risk stratification is crucial in OLT assessment patients to allow the successful recipient to get maximal benefit.Cystatin C (Cys C) is a potent inhibitor of lysosomal cathepsin proteinases.Traditionally Cys C has been regarded as a biomarker of renal dysfunction but it has been recently shown to be directly involved in the artherosclerotic process. Aim:To determine the relationship with Cys C & eGFR in prediction of cardiac & non cardiac death post OLT. Methods:Sin-gle centre,prospective observation study.Cardiac events recorded prospectively.SPSS version 18 used for statistical analysis. Results:97 patients recruited with mean age of 53yrs.Median follow up 48 mths. 16 patients died during follow up with 26 cardiac events occuring. Initial univariable analysis suggested that Cys C was a better predictor of CV death post OLT than eGFR.In contrast,a stronger association between eGFR & non-cardiac death post OLT was seen.Cox proportional hazards model with an outcome measure of all cause mortality post OLT suggested an interaction (effect modification) between eGFR & Cys C (p=0.058).We combined the two risk factors into a simple four-category factor using a low/high cut-off for both variables.The Kaplan-Meier survival analysis using this grouping is shown in figure 1. (Group 1: GFR<55 ml/min & Cys C>1.4 μg/ml. Group 2: GFR<55 & Cys C<1.4 Group 3: GFR>55 &Cys C>1.4. Group 4: GFR>55 & Cys C<1.4.) Con-clusion:This study reports for the first time the potential of combining Cystatin C & eGFR as a powerful predictor of survival after OLT.

Figure 1 .KM survival post OLT in Cys C & eGFR groups (Log Rank p<0.001)

image

Disclosures:

Peter C. Hayes - Advisory Committees or Review Panels: Roche, Roche, Jannsen, Gilead, ?ONO, Norgine; Grant/Research Support: Novartis; Speaking and Teaching: Roche, MSD, Roche, MSD, Pfiser, Gore, Falk, Ferring

The following people have nothing to disclose: Norma C. McAvoy, Alasdair W. Hay, Graham McKillop, David E. Newby

1707

Renal Dysfunction in Hepatitis C Virus and Hepatocellu-lar Carcinoma Patients after Liver Transplantation: Preliminary Results of the Italian Cross-sectional, Multicenter Study (surf) Cohort

Maria F. Donato3, Marco Senzolo1, Rossella Brusa7, Alfonso Gale-ota Lanza5, Anan J. Bastiampillai3, Denis Dissegna4, Giorgio Rossi3, Michele Imparato5, Giulia Magini2, Paolo De Simone6, Patrizia Burra1;

1Multivisceral Transplant Unit, Gastroenterology, University Hospital of Padua, Padua, Italy; 2Gastroenterology Unit, A.O. Papa Giovanni XXIII, Bergamo, Italy; 3General Surgery and Liver Transplantation Unit, Fondazione Cà Granda Ospedale Mag-giore Policlinico, Milan, Italy; 4Medical Liver Transplantation Unit, Internal Medicine,, University Hospital of Udine, Udine, Italy; 5Gas-troenterology Unit, AORN "A. Cardarelli, Naples, Italy; 6Hepato-biliary Surgery and Liver Transplantation, University Hospital of Pisa, Pisa, Italy; 7Novartis Pharma, Novartis Farma S.p.A., Varese, Italy

Introduction: Chronic renal disease (CKD) may affect up to 25% of liver transplant (LT) recipients at 10 years. Calcineurin inhibitors (CNI's) as immunosuppressants, lower renal function at transplantation, and hepatitis C virus (HCV) infection seem to be risk factors. Scanty information is available on the contributing role of hepatocellular carcinoma (HCC) to progression of CKD after LT. To investigate the role of HCV and HCC in the development of CKD after LT, a sub-analysis of the maintenance, adult and consenting cohort of the italian, multicenter SURF study was performed. Methods: Data were derived from an observational, cross-sectional survey (SURF study) across 15 Italian LT centers. Inclusion criteria called for adult (>1 8 years) recipients enrolled between 6 months and 5 years after primary LT. Information on renal function, immunosuppression, and co-morbidities were retrieved from medical records. Primary objective was the prevalence of patients with eGFR<60 mL/min/1.73m2. Herein we present an interim analysis of the first 753 enrolled patients. Results: From March, 2012 to January, 2013, patients was consecutively enrolled (planned 1000 patients), and 724 were evaluable (males 74.9%; median age 52.8 years; median time since LT29.1 months). Indication to LT was hepatitis C virus-related in 300 patients (41.4%), while HCC was present in 292 (40.3%). At transplantation, HCV patients were similar to non-HCV for eGFR <60 mL/min/1.73m2 (13.3% vs. 15.7%), and comparable to the total population (14,6%); as expected, HCC patients had better renal function at transplant (eGFR <60 mL/min/1.73m2 = 7.8% in HCC vs. 21.6% in non-HCC, p<.0001), after adjusting for HCV as co-indication to LT. At enrolment, HCV patients were more frequently on cyclosporine (CyA) (39%, p<0.0001) and showed similar renal function deterioration as non-HCV (eGFR <60 mL/min/1.73m2 = 28.2% vs. 23.3%), and similar to the total cohort (25.4%). Similarly, HCC patients showed a decrease in renal function vs. non-HCC (eGFR <60 mL/min/1.73m2 = 23.5% vs. 28.8%), after adjusting for HCV. Conclusion: In this large, cross-sectional cohort of Italian LT patients, HCV and HCC recipients surviving 6 months to 5 years after surgery show a decline in renal function similar to the overall population in the study and should undergo similar preventative policies.

Disclosures:

Rossella Brusa - Employment: Novartis

Paolo De Simone - Consulting: Novartis, Novartis

The following people have nothing to disclose: Maria F. Donato, Marco Senzolo, Alfonso Galeota Lanza, Anan J. Bastiampillai, Denis Dissegna, Giorgio Rossi, Michele Imparato, Giulia Magini, Patrizia Burra

1708

Endoscopic therapy for bile leaks after liver transplantation: An analysis of two high-volume transplant centers

Oriol Sendino1, Ryan Law2, Domingo Balderramo1, Josep Llach1, Todd H. Baron2, Andres Cardenas1;

1Institut de Malalties Digestives, Hospital Clinic, Barcelona, Spain; 2Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN

Abstract Bile leak after liver transplantation (LT) is commonly treated with endoscopic retrograde cholangiopancreatography (ERCP); however, there is scarce data in regards to the optimal strategy with this technique. We aimed to examine the role of ERCP in LT recipients with bile leaks at two large institutions. Methods We reviewed all ERCPs performed in LT recipients with bile leak and duct-to-duct biliary anastomosis at two high-volume transplant centers (Hospital Clinic-Barcelona and Mayo Clinic-Rochester) from 2000 to 2013. Information on clinical and endoscopic outcome was obtained from electronic health records and endoscopy databases from both institutions. Pre-operative and endoscopic variables were evaluated as factors determining outcomes after ERCP. Results A total of 74 patients were included (39-Hospital Clinic and 35-Mayo Clinic). Location of bile leak was: anastomosis (45%), T-tube (32%), intra-hepatic (11%), cystic remnant (9%) and cut surface (3%). Forty-two (56.7%) patients underwent ERCP with sphinctero-tomy and plastic stent placement and 32 patients (43.2%) underwent sphincterotomy alone. Complete resolution was obtained in 93% of the stent group vs. 56% of the sphincterotomy group (p=0.01). Resolution occurred in <3 months in 71% of the stent group vs. 31% of the sphincterotomy group (p=0.01). There was no difference in 3 month survival among both groups. Percutaneous transhepatic therapy and surgery were required in 5% and 7%, respectively in the stent group vs. 12% and 40% respectively in the sphincterotomy group (p=0.01). The only predictive factor of bile leak resolution was stent placement. Complications of ERCP occurred in 8% (4 mild pancreatitis, 2 mild bleed). Conclusion ERCP is a safe and effective therapy for post-LT bile leaks. Combination therapy with sphincterotomy and plastic stent is highly successful and more effective than sphincterotomy alone. These results indicate that ERCP with sphincterotomy and plastic stent placement is the treatment of choice for bile leaks in LT recipients with duct-to-duct biliary anastomosis.

Disclosures:

Todd H. Baron - Consulting: Cook Medicine, Olympus, Merit Endotek; Speaking and Teaching: Boston Scientific Corporation, ConMed

Andres Cardenas - Board Membership: Frontline Gastroenterology- BMJ publishing group; Consulting: Uptodate; Stock Shareholder: Limmedx LLC

The following people have nothing to disclose: Oriol Sendino, Ryan Law, Domingo Balderramo, Josep Llach

1709

Preoperative tumour biopsy does not affect the oncologic course of patients with transplantable HCC

David Fuks1, Grazia Fusco1, François Cauchy1, Fédérica Dondéro1, Vanessa Esnault2, Safi Dokmak1, Claire Francoz2, Valérie Vilgrain3, Valérie Paradis4, Francois Durand2, Jacques Bel-ghiti1;

1HPB Surgery, Beaujon Hospital, Clichy, France; 2Hepatology, Beaujon Hospital, Clichy, France; 3Radiology, Beaujon Hospital, Clichy, France; 4Pathology, Beaujon Hospital, Clichy, France

Introduction. Preoperative tumour biopsy (PTB) in patients considered for liver transplantation (LT) may increase the risk of recurrence. The aim of this study was to evaluate whether PTB actually jeopardized the long-term outcomes of patients with HCC eligible for LT. Methods. From 2002 to 2012, all 309

HCC patients listed for LT were analysed from a prospectively maintained database. Preoperative characteristics, modalities of recurrence and disease-free survival were compared between the 80 (26%) patients with preoperative tumour biopsy (PTB+) and the 229 (74%) patients without (PTB-). Results. The two groups (PTB+ vs PTB-) were similar in terms of demography, rates of lesions within the Milan criteria (81% vs. 79%, p=0.67), duration on the waiting list (7.0 vs. 6.9 months, p=0.89), and preoperative treatment (radiofrequency ablation 50% vs. 30%, p=0.06, trans-arterial chemoembolization 45% vs. 37%, p=0.11, and surgery 10% vs. 18%, p=0.11). Dropout during the waiting list related to tumour progression was observed in 31 (1 0%) with no difference between the two groups (6% vs. 11%, p=0.42). Among the 278 (90%) transplanted patients including 75 PTB+, pathological examination of the explants revealed that 1 1 (4%) patients had non-HCC lesions including only one in the PTB+. Median follow-up was 34 months (12–135) and recurrence after LT was observed in 25 (9%) patients with no difference in both groups (9.3% vs. 8.8%). Parietal recurrence was observed in 3; one in PTB+ group and 2 in PTB- group after radiofrequency ablation. On an intention to treat basis, 1-, 3- and 5-year overall (94%, 77% and 71% vs. 82%, 70% and 60%, p=0.07) and disease-free survivals (89%, 77% and 71% vs. 79%, 68% and 60%, p=0.08) were not significantly different between PTB+ and PTB-patients. Conclusion: Preoperative tumour biopsy doesn't influence the oncologic course of HCC patients eligible for LT, and there is no argument to restrict biopsy in doubtful situations.

Disclosures:

Francois Durand - Advisory Committees or Review Panels: Astellas, Novartis; Speaking and Teaching: Gilead

The following people have nothing to disclose: David Fuks, Grazia Fusco, François Cauchy, Fédérica Dondéro, Vanessa Esnault, Safi Dokmak, Claire Fran-coz, Valérie Vilgrain, Valérie Paradis, Jacques Belghiti

1710

Liver transplantation in the morbidly obese: Do the outcomes justify the use of scarce organs?

George Therapondos, David S. Bruce, Humberto E. Bohorquez, Natalie H. Bzowej, Ian C. Carmody, Ari J. Cohen, Nigel Girgrah, Shobha Joshi, Trevor W. Reichman, George E. Loss;

Multiorgan Transplant Institute, Ochsner Clinic Foundation, New Orleans, LA

Background: Morbid obesity is considered a relative contraindication to liver transplantation at many transplant centers. Aims: To compare the survival outcomes of morbidly obese (BMI>40) liver transplant recipients with the survival outcomes of patients with a BMI<40 and to examine their early post-OLT complications and morbidities. Methods: This is a retrospective chart review of adult patients undergoing liver or combined liver/kidney transplant at Ochsner Foundation Hospital between June 1, 2005 and Feb 28, 201 3 who had a minimum follow up of 3 months. 868 adult liver transplants were performed during this period at our center. 65 recipients had a BMI >40 (range 40–56, median 43) and we compared their outcomes with 761 liver or liver/kidney recipients with a BMI<40 (range 12–39, median 27). Results: The obese group consisted of 52% females. Mean recipient age was 52 yrs in the obese group and 53 yrs in the control group. Pre-Tx diagnoses in the obese group compared with the non-obese group were HCV (33% vs 31 %), NASH (30% vs 10%), HCC (16% vs 19% ), alcohol (9% vs 16%) and other (12% vs 24%). MELD scores at the time of OLT were similar between the groups (median 24, obese gp vs 22, control). There were no intra or perioperative deaths in the obese group. 1and 5 year patient and graft survival rates are shown below and were similar between the two groups. Median length of stay in the obese group vs control group was 11 vs 1 0 days. Thirty day re-operation rates were similar for the two groups (14% in obese vs 19% in the control gp). Wound infections and/or dehiscence occurred in 24% of the obese group. 36% of the obese group were diabetic pre-Tx with 40% on insulin. In this group, the prevalence of diabetes increased to 45% post-Tx (86% on insulin). 39% of this group had a history of hypertension pre-Tx with the prevalence postTx rising to 86%. 9% of the obese patients had hyperlipidemia pre-Tx with the rate rising to 17% post-Tx. Only 2 patients had a cardiac history pre-Tx and there were only 3 cardiovascular events noted post-Tx. Conclusion: Liver transplantation in morbidly obese recipients yields excellent short and medium outcomes which are similar to those seen in our non-obese patients. The increased burden of diabetes and hypertension does not appear to impact survival in this group of carefully selected patients. Morbid obesity alone is not a contraindication to liver transplantation.

Paient and graft survival of BMI>40 and BMI<40 groups

 BMI>40  BMI<40 
 1 year5 years 1 year5 years
Patient survival n=6594%85.1%Patient survival n=76193.6%87.2%
Graft survival n=6889.9%81.2%Graft survival n=80090.4%82.8%

Disclosures:

Natalie H. Bzowej - Advisory Committees or Review Panels: Vertex; Grant/Research Support: Genentech, Merck, Gilead Sciences, Vertex, Bristol Myer Squibb, Pharmasset; Speaking and Teaching: Gilead Sciences, Vertex

Nigel Girgrah - Speaking and Teaching: Merck, Bayer, Vertex, Gilead, Merck, Bayer, Vertex, Gilead, Merck, Bayer, Vertex, Gilead, Merck, Bayer, Vertex, Gilead

Shobha Joshi - Advisory Committees or Review Panels: Roche, 3 Rivers Pharmaceuticals; Grant/Research Support: Gilead, Eisai; Speaking and Teaching: Merck, Bristol-Myers Squibb

The following people have nothing to disclose: George Therapondos, David S. Bruce, Humberto E. Bohorquez, Ian C. Carmody, Ari J. Cohen, Trevor W. Reichman, George E. Loss

1711

Extended Follow-up of Pediatric Liver Transplantation Patients Receiving Once Daily Calcineurin Inhibitor

Henry Lin1, Hector Melin-Aldana2, Saeed Mohammad1, Udeme D. Ekong3, Estella M. Alonso1;

1Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, IL; 2Pathology, Northwestern University Feinberg School of Medicine, Chicago, IL; 3Pedi-atrics, Yale University School of Medicine, New Haven, CT

Chronic exposure to immunosuppression (IS) medications is associated with toxicities and safe strategies for IS minimization to once daily calcineurin inhibitor (CNI) have been reported. We report longitudinal results in a cohort of liver transplant (LT) patients minimized to daily CNI monotherapy and assess progression of fibrosis on follow-up liver biopsy. A retrospective chart review of all pediatric LT patients at a single center receiving once daily CNI was performed. Exclusion criteria included prior diagnosis of de novo autoimmume hepatitis or minimization of IS for EBV. Biopsies and biochemical makers (AST, ALT, total bilirubin, direct bilirubin, INR, creati-nine) were obtained at the following time points: pre-mini-mization (PM), 5-year, 7-year, 9-year post-minimization, and most recently (MR) or at time of change in IS. Biopsies were reviewed and graded for inflammation and fibrosis using Ishak and Batt grading systems. Minimization was attempted in 59 patients. Successful minimization to daily CNI monotherapy was defined as normal liver enzymes 1 year after dose change. 44 were successfully minimized, 1 1 failed, and 4 developed de novo autoimmune hepatitis. The successfully minimized patients were followed from 2001 to present and 33/44 patients remained on once daily CNI for at least 5 years (1 6 for 7 years, and 5 for 9 years). Follow-up labs were unavailable for 2 patients. 2 (4.8%) patients developed elevated GGT (>1.5x upper limit of normal) during follow-up, biopsies showed no acute cellular rejection and patients remained on once daily CNI. 25 patients had post-minimization biopsies between 0.73 to 10.6 years on once daily CNI (median 5 years, IQR 3.38 years). Biopsies were preformed in 21 patients for surveillance and in 4 for elevated liver enzymes. All 25 patients had

 AST Mean (SD)ALT Mean (SD)Direct Bilirubin Mean (SD)GGT Mean (SD)Creatinine Mean (SD)eGFR Mean (SD)Biopsy Stage 0–1 (Number)Biopsy >Stage 2 (Number)Biopsy Grade 0–1 (Number)Biopsy >Grade 2 (Number)
PM36.2 (11.4)30.4 (13.3)0.14 (0.18)30.7 (40.1)0.57 (0.18)103 (27)196232
MR29.8 (14.1)31 (24.4)0.16 (0.08)32.2 (41.6)0.7(0.21)101 (28)205241
5 Year33.5 (17.5)31.2 (16.1)0.13 (0.05)29.6 (24.7)0.63 (0.25)112 (47)    
7 Year30 (11.2)26.1 (7.8)0.14 (0.06)24.8 (11.4)0.75 (0.26)98 (29)    
9 Year23 (8.5)23.2 (9.3)0.18 (0.04)43.3 (34.2)0.78 (0.24)94 (26)    

Disclosures:

The following people have nothing to disclose: Henry Lin, Hector Melin-Aldana, Saeed Mohammad, Udeme D. Ekong, Estella M. Alonso

1712

The impact of pre-liver transplant metabolic factors on post-liver transplant survival

Leon Adams1,2, Oscar Arauz2, Peter W. Angus3, Marie Sinclair3, Nicholas A. Shackel4, Linda I. Lin4, Alan J. Wigg5, Sze Pheh Yeap5, Spiro C. Raftopoulos2, Gary P. Jeffrey1,2;

1Medicine and Pharmacology, University of Western Australia, Nedlands, WA, Australia; 2Liver Transplant Unit, Sir Charles Gairdner Hospital, Nedlands, WA, Australia; 3Liver Transplant Unit, The Austin Hospital, Melbourne, VIC, Australia; 4Liver Transplant Unit, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 5Liver Transplant Unit, Flinders Medical Centre, Adelaide, SA, Australia

Metabolic factors such as diabetes mellitus (DM), obesity and hypertension are common and frequently co-exist in subjects undergoing orthotopic liver transplantation (OLT), however their long-term impact on post-OLT survival is unclear. Methods: We performed a retrospective analysis of subjects who underwent OLT in four Australian centers over a 6 year period (2003–09). Body mass index (BMI) was calculated using dry body weight. Exclusion criteria included fulminant liver failure, living-related or multi-organ transplants, re-transplantation, hepatocellular carcinoma outside UCSF criteria or where data regarding metabolic factors was missing. The prognostic significance of metabolic factors was determined using log-rank tests and multivariate Cox-regression analysis. Results: 494 adults, [72% male, 75% Caucasian, mean (SD) age of 51 (9) years] were followed for a mean of 5.5 (2.9) years after OLT. Hepatitis C and alcoholic liver disease were the commonest indications for OLT (31% and 14.6% respectively). Prior to OLT, 18.8% had DM, 9.5% had hypertension. The prevalence of pre-OLT overweight (BMI 25–29.9), obesity (BMI 30–34.9) and morbid obesity (BMI 35+) was 34.4%, 17.6% and 5.2% respectively. During follow-up, 103 subjects died with 1 and 5 year overall survival of 92% and 82% respectively. 1 and 5 year graft survival was 90% and 79%. Overall, pre-OLT DM was independently associated with poorer 5-year post-OLT patient survival (70% vs 84%, p=0.001) and graft survival (71% vs 85%, p=0.04), whereas obesity, hypertension or dyslipidemia did not impact on overall patient or graft survival (p>0.2). The impact of DM however, was limited to obese patients were 5-year survival of subjects with DM was 62% vs. 92% (p<0.001); 5-year survival in non-obese individuals was no different in the presence or absence of DM (77% vs. 86%, p=0.1). Amongst patients with DM, prior abdominal surgery, longer warm ischemic time and number of operative transfused red cells were associated with poorer patient survival (p<0.01 for all) whereas obesity, hypertension and dyslipidemia were not (p>0.2 for all). Morbid obesity, was associated with longer duration of hospitalization (25 vs. 17 days, p=0.005) and a trend towards longer ICU stay (median 4 vs. 3 days, p=0.07), but was not associated with a difference in patient or graft survival (both p=0.2). Conclusion: Obese individuals have reduced post-transplant survival if they have pre-OLT diabetes. Surgical rather than patient metabolic factors however, are associated with reduced survival in patients with pre-transplant DM.

Disclosures:

Peter W. Angus - Grant/Research Support: Gilead Sciences

Gary P. Jeffrey - Advisory Committees or Review Panels: MSD, Novartis

The following people have nothing to disclose: Leon Adams, Oscar Arauz, Marie Sinclair, Nicholas A. Shackel, Linda I. Lin, Alan J. Wigg, Sze Pheh Yeap, Spiro C. Raftopoulos

1713

Safety and Efficacy of Subcutaneous Hepatitis B Immunoprophylaxis using"on demand" Approach: A Single Center Experience

Silvia Pecere1, Antonella Grisolia1, Luisa Pasulo1, Anna Baldan1, Matteo Rota2, Michele Colledan1, Stefano Fagiuoli1;

1Gastroen-terology and Transplant Hepatology, HPG23, Bergamo, Italy; 2Department of Health Sciences, Centre of Biostatistics for Clinical Epidemiology, University of Milan-Bicocca, Monza, Italy

Background:Hepatitis B immunoglobulin(HBIG)administration is the backbone for prophylaxis of HBV re-infection after liver transplantation (LT).Long term effects and efficacy of HBIG are well known only for intravenous(IV)and intramuscular(IM)for-mulations. Aim:to investigate efficacy,safety and feasibility of "on demand" subcutaneous (SC) administration of the new formulation of HBIG BT088 (Zutectra®)LT patients. Materials and methods:a total of 37 LT patients(9F,28M,mean age60±7years)were switched from IV or IM to SC HBIG administration during a period of 2 years from January 201 1 .The conversion to SC administration occurred at a mean time of 44 months from the LT.They were prospectively enrolled and followed up for at least 48 weeks.The dose of HBIG was initially standardized to 1 000I U/week.After a period of stabilization, patients were offered a"on demand"approach,with a targeted level of serum antiHBs of minimum 1 00UI/L. Results:all patients were HBV-DNA negative at the time of transplantation(5spon-taneously-14%,32 under NUCs -86%).All patients were on a combination prophylaxis with HBIG and NUCs.All patients became rapidly independent for the weekly SC self-injection.The treatment was effective in maintaining trough antiHBs levels greater than 1 00 UI/l in all patients. 90%of patients showed a mean HBsAb titer greater than 155 UI/l.Overall, mean values of HBsAb were 262 UI/l(±1 1 8).The mean HBsAb titre prior to switching to SC formulation was 31 8±124, with a mean monthly injection of 5000U/L.No drug related side effects or injection site problems were observed. Conclu-sions:SC HBIG for long term prophylaxis of post LT HBV re-infection has proven to be safe, well-accepted and effective in maintaining the targeted protective anti-HBs levels. Moreover individualization of immuneprophylaxis according to the lowest protective anti-HBs titers is easily applicable with the SC formulation, allowing the exploration of new schedules in order to improve costs while maintaining efficacy.

Disclosures:

Michele Colledan - Advisory Committees or Review Panels: novartis

The following people have nothing to disclose: Silvia Pecere, Antonella Grisolia, Luisa Pasulo, Anna Baldan, Matteo Rota, Stefano Fagiuoli

1714

The Liver May Not Protect Against Kidney Rejection in Simultaneous Liver-Kidney Transplantation

Kathy Nilles, James Krupp, Lorenzo Gallon, Josh Levitsky;

Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL

Introduction: Due to the tolerogenic features of the liver, simultaneous liver-kidney (SLK) recipients are thought to be protected from kidney rejection and typically receive similar immunosup-pression (IS) as liver transplant alone (LTA). However, data to support this practice are not available. Aim: To characterize the incidence, types, and outcomes of rejection, graft function, graft survival, and patient survival in our large SLK population. Methods: Adults > 1 8 yrs undergoing SLK from 1996–2010 were included. IS included tacrolimus, mycophenolate mofetil, and steroid taper; no induction therapy was utilized. Increase in serum creatinine (Cr) or aminotransferases of >50% from baseline prompted liver or kidney biopsy, respectively. Total bilirubin, ALT, and Cr were recorded at 0, 1,3,6, 12, 24, and 36 months. Results: 177 received SLK: age 54.0 ± 10.8 yrs, 60.9 % male, 32.0 % HCV+. Patient survival at 1, 3, and 5 years was 84.2%, 77.0%, and 74.0%. Acute cellular, antibody-mediated, and chronic kidney rejection occurred in 13 (7.3%), 4 (2.3%), and 1 (0.6%). Mean Cr was worse in patients with a history of acute kidney rejection vs. no rejection (p=0.04) at all time points. While kidney graft survival was not different in patients with (72.2%) or without (60.4%) kidney rejection (p =0.38), those with advanced rejection (chronic or acute Banff 2a or 2b) had a trend toward decreased patient (p=0.09) and graft survival (p=0.15). Kidney graft failure occurred in 6 (3.4%). Acute cellular liver rejection occurred in 22 (12.4%) and chronic liver rejection in 5 (2.8%). A trend (p=0.30) toward improved survival in patients with liver rejection (75.0%) occurred, vs those without rejection (65.4%). Overall liver graft survival was 75.0% vs 62.1% in the rejection vs non-rejection group (p=0.19). Liver retransplantation occurred in 4 (2.3%). Chronic rejection was associated with decreased patient (p=0.02) and liver graft (p=0.02) survival. In patients with rejection of both organs, kidney rejection did not precede liver rejection. Conclusions: Over 10% of SLK recipients had kidney rejection and had worse serum Cr over time, indicating that the liver may not be fully protective. Although patient and graft survival were not significantly different, this may be due to small sample size. Our data suggest that SLK patients should receive IS protocols that are more similar to kidney alone than LTA to preserve renal function. In addition, while immune-mediate graft dysfunction is implicated, protocol kidney biopsies in SLK would help elucidate the mechanisms of graft injury.

Disclosures:

The following people have nothing to disclose: Kathy Nilles, James Krupp, Lorenzo Gallon, Josh Levitsky

1715

Reversible Non-Ischemic Cardiomyopathy and Left Ventricular Dysfunction after Liver Transplantation

Maria L. Yataco1, Thomas Difato1, Johannes Bargehr1, Jorge F. Trejo-Gutierrez2, Barry Rosser1, Tushar Patel1, Burcin Taner1, Surakit Pungpapong1, Jaime Aranda-Michel1;

1Transplant, Mayo Clinic Jacksonville, Jacksonville, FL; 2Cardiovascular Diseases, Mayo Clinic Jacksonville, Jacksonville, FL

Background: Reversible non-ischemic cardiomyopathy (NIC) with left ventricular dysfunction (LVD) is a rare complication that has been observed in the immediate postoperative period after liver transplantation (LT). This condition is poorly characterized. Our Aims were to define the prevalence, associated clinical factors and prognosis of this condition. Methods: A retrospective review of all patients who underwent LT at our institution between 2005–2012 was performed to identify all patients who developed NIC and LVD. Results: Seventeen out of 1460 patients (1.2%) developed NIC with moderate to severe LVD. All patients had a negative pre-LT cardiac evaluation. Nine patients (53%) were in ICU prior to LT. Fourteen patients (82%) were severely malnourished. Eleven patients (65%) required CRRT in the perioperative period. Median raw MELD score was 8 in patients with NIC versus 29 in no-NIC patients (p = 0.01). There was no significant difference between NIC and no-NIC patients regarding donor age, donor risk index, cold ischemia time and warm ischemia time. Eight patients presented with cardiac arrest or hypotension during LT, and 1 1 with symptomatic heart failure. Median time of onset was 2 days post LT (0–20). At diagnosis, patients had negative or mild elevated cardiac enzymes and non-specific EKG changes. Echocardio-grams showed global LV hypokinesis and decrease in ejection fraction (EF) from a median of 65% (50–81) to a median of 21% (15–32). Recovery of cardiac function occurred in 16 patients, with a median EF of 44 % (25–65) at the time of discharge. One patient died at day 64 post LT due to ventricular arrhythmia. Median length of hospitalization was 26 days (9–63) in patients with NIC versus 8 days (0–515) in no-NIC patients (p = 0.01). Last echocardiogram showed a significant improvement of LV function with a median EF of 59 % (49–73). The median patient survival after LT was 25 months (2–67) with a minimum follow up period of 6 months. Conclusion: Patients with NIC are frequently critically ill, with high MELD score and severe malnutrition. Potential mechanisms of NIC could include undiagnosed cirrhotic cardiomyopathy, volume overload, stress-cytokine mediated cardiac injury, or heart failure associated with refeeding syndrome in severely malnourished patients. Although NIC is a rare complication of LT, it can cause significant perioperative and postoperative morbidity or mortality. Recognition of individuals at risk of NIC may be beneficial in reducing complications and length of hospitalization.

Disclosures:

Burcin Taner - Advisory Committees or Review Panels: Synergy Pharmaceuticals

Surakit Pungpapong - Grant/Research Support: BMS

The following people have nothing to disclose: Maria L. Yataco, Thomas Difato, Johannes Bargehr, Jorge F. Trejo-Gutierrez, Barry Rosser, Tushar Patel, Jaime Aranda-Michel

1716

Donation after Cardiac Death (DCD) Liver Transplantation (LT): Impact on graft and patient survival and determinants of outcome

Shiva Kumar, Rachel Pedersen, Ajay Sahajpal, Karen Denson, Jef-fery L. Steers;

Liver Transplant Program, Aurora St. Luke's Medical Center, Milwaukee, WI

Donor shortage has prompted increasing use of DCD allografts. However, this has been reported to increase morbidity and mortality following LT. Aims: 1. Compare graft and patient survival following DCD and non-DCD LT. 2. Identify donor and recipient variables that determine outcome following DCD LT. 3. Identify impact of DCD allograft on outcome following combined liver-kidney transplant (LKT). Methods: We conducted a retrospective review of all LT performed at a single transplant center from July 2007 - Feb 2013. DCD transplants were compared to non-DCD LT performed during the same study period. The following donor and recipient variables were analyzed: donor age, warm and cold ischemic time, recipient age, MELD score at LT, HCV status, presence of HCC. Categorical variables were analyzed using Chi-square test and continuous variables using Student t test. Survival comparison was performed using Kaplan-Meier method and multivariate analysis of predictors of outcome using Cox regression method. Results: A total of 162 LT were performed during the study period, of which, 131 were primary single organ LT, 24 LKT with 6 of those being a re-transplanted liver and 4 solitary liver re-LT. 51 / 1 62 (31 %) of these transplants were performed using DCD allografts. The DCD and non-DCD recipients were comparable with respect to age, gender, HCV status, and presence of HCC. 19 (37%) of DCD recipients were HCV positive and 12 (23%) had HCC. The mean MELD at LT was significantly higher in non-DCD recipients compared to DCD (27.2 +/- 8.9 vs. 21.5 +/-7.5, p<0.01.). Patient and graft survival were comparable between non-DCD and DCD LT recipients (p=0.26, p=0.50, respectively). Estimates were as follows: 1 year (86.1% vs 82.2%; 85.3 vs 83.1%), 3 years (78.1 vs 69.2%; 75.5% vs 70.9%) and 5 years (74.0% vs 64.9%; 71.6% vs 66.8%). Mul-tivariable and univariate analysis identified HCV recipients receiving organs from donors>50 years as a significant predictor of graft and patient survival following DCD LT (HR=6.6 (1.7–21.7), p=0.003). Among recipients of LKT from DCD donors, graft survival was comparable to non-DCD LKT (p=0.87) (1 year: 66.7% vs. 65.7 %%). Although there was an observed lower rate of graft survival at 3 years among DCD recipients post LKT, this did not reach statistical significance (55.6% vs. 65.7). Conclusions: Patient and graft survival were comparable among recipients of DCD and non-DCD single organ primary LT. Graft survival was also comparable among recipients of DCD combined LKT. Limiting DCD transplants to non-HCV recipients from younger donors may enable further expansion of the donor pool without compromising graft and patient survival.

Disclosures:

Shiva Kumar - Advisory Committees or Review Panels: Vertex; Speaking and Teaching: Bayer Onyx

The following people have nothing to disclose: Rachel Pedersen, Ajay Sahajpal, Karen Denson, Jeffery L. Steers

1717

Coronary Calcification on Routine Chest Computed Tomography Identifies False Negative Stress Echocardio-grams in Liver Transplantation Candidates

Bow Y. Chung, Amit R. Patel, Kirk T. Spencer, Andrew Aronsohn, Sandeep Nathan, John F. Renz, Helen S. Te;

University of Chicago Medicine, Chicago, IL

Background. Significant coronary disease is an important cause of mortality in patients undergoing liver transplantation (LT). Stress echocardiography (SE) is the favored pre-operative test, though its diagnostic performance in this population has not been evaluated in detail. The goals of this study are to determine 1) the diagnostic performance of SE in LT candidates and 2) the additive value of coronary calcification incidentally noted on non-gated chest computed tomography (CT). Methods. We identified 45 patients undergoing LT evaluation who had both coronary angiogram (CATH) and SE performed within one year. A subgroup of 26 individuals incidentally had a chest CT in the same period. CATH was considered abnormal if a coronary stenosis >70% was present. SE was considered abnormal if any wall motion abnormality was present. CT was used to evaluate calcium burden in the left main and 3 major coronary arteries using a qualitative 3 point scale (0=no calcium, 1 =mild, 2=severe, Figure 1). A total calcium value ranging from 0–8 was calculated for each patient by adding the score for each vessel together. Each SE, compared to CATH, was determined to be non-diagnostic (ND), true positive (TP), false positive (FP), true negative (TN), or false negative (FN). Results. 1 8 of the 45 patients had a coronary stenosis >70% (3 LM, 14 LAD, 8 LCx, 6 RCA). 19 SE were ND, predominantly due to failure to achieve target heart rate. Among the 26 individuals who had a CT, FN patients had a significantly higher total calcium value (6.6 ± 1.4) than the TN patients (3.9 ± 2.3, p=0.016). ND patients with abnormal CATH had a total calcium value of 6.7± 2.3; whereas, those without coronary stenosis had a value of 3.5 ± 3.7 (p=0.25). The sensitivity of SE among those with diagnostic studies was 20% and specificity was 81%. Alternately, the CT finding of a single severe calcification in any one artery had 1 00% sensitivity and 1 00% negative predictive value for detecting significant coronary disease. Conclusion. Sensitivity of SE for detecting significant coronary stenosis is suboptimal in LT candidates and incidental CT images should be evaluated for coronary calcification to identify potential false negative SE patients. Future studies are needed to determine the role of formal calcium score testing or coronary CT in patients being considered for LT.

Disclosures:

Helen S. Te - Advisory Committees or Review Panels: Gilead Sciences, Novartis, Idenix; Grant/Research Support: Abbvie

The following people have nothing to disclose: Bow Y. Chung, Amit R. Patel, Kirk T. Spencer, Andrew Aronsohn, Sandeep Nathan, John F. Renz

1718

Validating post-transplant hepatocellular carcinoma recurrence data in the UNOS database

Mariya Samoylova1, Jennifer L. Dodge1, Eric Vittinghoff2, Francis Y. Yao1, John P. Roberts1;

1Surgery, University of California, San Francisco, San Francisco, CA; 2Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA

BACKGROUND: The OPTN/UNOS database is the most comprehensive collection of liver transplantation data, but the quality of these data with respect to hepatocellular carcinoma (HCC) recurrence has not been well-assessed. In this study, we compare observed HCC recurrence rates in the UNOS database to expected ones calculated using a hierarchical model for recurrence adjusted for recipient and tumor characteristics. METHODS: We used the UNOS STAR dataset on adult transplants with initial exception for HCC diagnosis granted between 1/1/06 and 9/30/10 and transplanted in the same time frame. We developed a risk-adjusted Poisson model with patient as unit of analysis, random effects for transplant center, and years of follow-up as offset to predict expected recurrence for each center. Shrinkage estimates of the observed to expected ratio were obtained as best linear unbiased predictions of the center random effects, with model-based standard error. To further investigate the possibility of under-reporting, we imputed expected recurrences within non-HCC deaths. RESULTS: 5034 HCC liver transplant recipients were identified. Death from or diagnosis of recurrence occurred in 6.8% of patients at a median of 1 year (IQR 0.5–1.8) post-transplant. Covariate-adjusted incidence rate ratios (IRR) for HCC recurrence were increased for transplant recipients with 1 tumor >=2cm (IRR 1.63, 95% CI 1.10–2.41, p=0.02) and 2 to 3 tumors >=2cm (IRR 1.82 95% CI 1.04–3.20, p=0.04) versus >1 tumor all <2 cm, ablative therapy vs. none (IRR 1.44, 95% CI 1.15–1.79, p=0.001), AFP > 500 vs. <= 500 (IRR 3.00, 95% CI 2.21–4.09, p<0.001) and increasing donor risk index (IRR 1.97, 95% CI 1.52–2.57, p<0.001). HCC recurrence IRRs were decreased for black versus white race (IRR 0.61, 95% CI 0.29–0.95, p=0.03) and non-cholestatic cirrhosis compared to HCV (IRR 0.48, 95% CI 0.26–0.89, p=0.02). Covariate-adjusted shrinkage estimates of the observed/expected HCC recurrence ratios by transplant center ranged from 0.6 to 1.75 (median 0.97). The 95% confidence intervals for the shrinkage ratios included unity for every center, indicating than none could be unambiguously identified as having lower or higher than expected HCC recurrence. Imputing outcomes for patients with potential unreported recurrence changed the center-specific shrinkage ratios to 0.72 to 1.39 (median 0.98), with no centers having a shrinkage ratio significantly different from one. CONCLUSIONS: Observed HCC recurrence was not significantly lower than expected in any center, suggesting that no systematic under-reporting has occurred. This study validates the OPTN HCC recurrence data, and supports their potential for further analysis

Disclosures:

The following people have nothing to disclose: Mariya Samoylova, Jennifer L. Dodge, Eric Vittinghoff, Francis Y. Yao, John P. Roberts

1719

Racial Disparity in Outcomes of Simultaneous Liver-Kidney Transplants in a Single, Tertiary Care Center in the United States

Brian Kim1,4, Tinsay A. Woreta1, Nyan L. Latt2, Po-Hung Chen1, Berkeley N. Limketkai1, Naudia L. Jonassaint1, Nada Alachkar3, Ahmet Gurakar1;

1Gastroenterology, Johns Hopkins University, Baltimore, MD;2Internal Medicine, Greater Baltimore Medical Center, Baltimore, MD; 3Nephrology, Johns Hopkins University, Baltimore, MD; 4Hepatology, Mount Sinai Hospital, New York, NY

Background: Post-liver transplant renal failure is a common complication that is associated with significant morbidity and mortality. Patients who receive simultaneous liver-kidney transplants (SLK) have improved survival over patients who receive a kidney transplant after a liver transplant. SLK have increased 300% since the institution of the Model for End-Stage Liver Disease (MELD) for liver allocation in 2002. Black patients have been observed to have higher rates of post-transplant mortality and graft failure in liver transplant alone (LTA) and kidney transplant alone (KTA). In this study, we examined the relationship between race/ethnicity and post-transplant outcomes in SLK patients from a single US tertiary center. Methods: We performed a retrospective cohort study of adult simultaneous liver and kidney transplants performed at Johns Hopkins Hospital from January 1992 to January 2013. Multivariate survival analysis was performed to assess predictors of patient mortality. Kaplan-Meier analysis was used to compare patient and graft survival between black vs. non-black patients. Results: A total of 44 patients received SLK from January 1 992 to January 201 3 at Johns Hopkins Hospital. There were 9 black recipients and 35 non-black recipients of SLK. Demographic characteristics including age, gender, and HCV status were similar in both groups. More underlying diabetes was noted in black vs. non-black recipients (77.8% vs. 22.9%, p=0.002). There was a trend towards more black recipients receiving a LTA or KTA prior to the index SLK as compared to non-black recipients (55.6% vs. 22.9%, p=0.055). No differences in liver graft (0% vs. 22.9%, p=0.1 1) and kidney graft rejection were seen (22.2% vs. 17.1%, p=0.72) between the two groups. In multivariate analysis, blacks had lower overall survival (HR 7.28, 95% CI 1.22–43.07) at a median follow-up of 33 months (range 3 months to 21 years). No differences in liver and renal graft survival were seen between black and non-black recipients. Conclusion: In our cohort of SLK patients, black recipients had poorer overall survival as compared to non-black recipients. This difference was not explained by differences in liver and renal graft failures. Black patients were more likely to have underlying diabetes and prior transplants (LTA or KTA), which may have influenced the high mortality seen in our study. Further multi-center prospective studies are needed to elucidate the exact mechanism of increased mortality among black SLK transplant recipients.

Disclosures:

Ahmet Gurakar - Advisory Committees or Review Panels: Gilead, Gilead; Grant/Research Support: BMS, BMS

The following people have nothing to disclose: Brian Kim, Tinsay A. Woreta, Nyan L. Latt, Po-Hung Chen, Berkeley N. Limketkai, Naudia L. Jonassaint, Nada Alachkar

1720

Universal Prophylaxis or Pre-emptive Therapy in the Prevention of Cytomegalovirus (CMV) Disease after Liver Transplantation (LT): A Systematic Review and Meta-analysis

Khalid Mumtaz, Nabiha Faisal, Andrea Morillo, Eberhard L. Ren-ner, Shahid Husain;

Multi organ Transplant, Toronto General Hospital, Toronto, ON, Canada

CMV disease is associated with rejection, allograft failure, and mortality. Strategies to prevent CMV disease in liver allograft recipients include universal prophylaxis (UP) or pre-emptive therapy (PeT) with either ganciclovir (GC) or valganciclovir (Val); it remains debated which of these strategies is the most effective one. Aim: To assess by systematic review and meta-analysis whether UP or PeT (with either GC or Val) is more effective in preventing CMV disease after LT. Methods: We performed an electronic database search of MEDLINE, EMBASE, and the Cochrane Database up to May 201 3. A prior defined eligibility criteria included UP or PeT for CMV infection after LT. Two authors identified/extracted data independently. Dichoto-mous data was expressed as relative risk with 95% CI using a random effects model. Primary outcome was CMV disease, while secondary outcomes were acute cellular rejection (ACR), allograft failure (AF), and mortality. Results: Of 244 references retrieved, 7 (2 RCTs and 5 observational studies) met eligibility criteria. Four studies reported on UP; CMV disease was observed in 8/174 (4.6%) with GC and in 18/216 (8.3%) with Val (RR= 0.52, CI: 0.23–1.19);heterogeneity=0%). Three studies reported on PeT; CMV disease was observed in 1/43 with GC and 1/39 with Val (RR=1.00, CI: 0.07–14.05; heterogeneity not applicable). Mortality with UP was reported in 12/114 (11%) using GC and 21/146 (14%) using Val (RR=0.72, CI: 0.30–1.73), while mortality with PeT was 3/43(7%) using GC and 3/39 (8%) using Val (RR=0.79, CI: 0.18–3.46); heterogeneity= 0%). There was insufficient data on ACR and AF to perform meta-analysis. Conclusions: Risk of CMV disease and mortality is similar with UP and PeT. More data is needed to assess the impact of CMV disease on ACR and AF. An adequately powered randomized control trial is warranted.

Disclosures:

Eberhard L. Renner - Advisory Committees or Review Panels: Vertex Canada, Novartis, Astellas Canada, Rcohe Canada, Gambro; Grant/Research Support: Novartis Canada; Speaking and Teaching: Novartis, Astellas Canada, Roche Canada

The following people have nothing to disclose: Khalid Mumtaz, Nabiha Faisal, Andrea Morillo, Shahid Husain

1721

Determinants Of Recurrent Hepatocellular Carcinoma After Liver Transplantation With Short Wait List Time

Ahmed Abu Shanab1, Zoe Hutchinson1, Linda O'Keefe2, Aiden McCormick1, Raphael B. Merriman1;

1St.Vincent's Hospital, Dublin, Ireland; 2Epidemiology Centre, Department of Obstetrics and Gynaecology, University College Cork, Cork, Ireland

Background: The outcome of orthotopic liver transplantation (OLT) for hepatocellular carcinoma (HCC) depends significantly on pre-OLT radiologic cancer stage. Other factors such as vascular invasion, pre-OLT alpha fetoprotein (AFP) level, the need for down-staging are less well-defined. Short waiting time on the transplant list facilitates expedited OLT and assumed (but unproven) better outcomes though could potentially impede assessment of tumour biology and affect OLT recurrence. We aimed to characterize the factors associated with recurrent HCC (recHCC) after OLT - especially short wait list times in patients who meet usual HCC listing criteria. Methods: Retrospective cohort data analysis of patients with HCC who had OLT in a single, center from 2002–201 1 with a minimum 12 months of follow-up. Pre- and post-OLT variables were studied of patients transplanted for HCC, with and without recHCC including clinical characteristics,, tumour markers, use of locoregional therapy, pre-OLT radiologic staging, waiting times and explant histological stage. Paired and unpaired t-tests were used to compare risk factors for those with and without recHCC. A mul-tivariate logistic regression analysis was applied to variables likely predictive of recHCC. Results: Sixty-nine patients had OLT for HCC from 2002 - 201 1 facilitated by a HCC exception score. Recurrent HCC occurred in 9/69 patients (13%), male (7), mean age (59.78 ± 1.891). The mean medical MELD score was 9.4 for recHCC versus 1 1 in the non recHCC group (p= 0.26). Pre-OLT AFP values, either average or peak prior to OLT were significantly higher in the recHCC group (p = 0.0012 and <0.0001 respt.). Pre-OLT locoregional therapy occurred in 44% with recHCC - however only 1 6% underwent locoregional therapy in the non-recHCC group (p value <0.05). Patients with and without recHCC had similar pre OLT HCC stage of disease - none was downstaged from T3. For those with recHCC, the median waiting time was shorter but not significantly different (50 ± 14 days) as compared to those without recHCC (82 ± 1 0 days), p = 0.23. Explant pathological staging in the recHCC group was T1 (n=3), T2 (n=4) and T3 (n=2) and with lympho-vascular invasion in 44%. The average grade of differentiation was significantly different in recHCC group 2.6 vs. 1.7 in the non-recHCC group (p = 0.003). The best predictor of recurrence was a combination of histologic grade of tumour differentiation and last AFP pre OLT with an AUC estimate of 0.9 but not wait list time. Conclusions: Histologic grade and the last AFP pre-OLT - but not short waiting time on the OLT list - were the most significant factors predictive of recurrence of HCC after OLT.

Disclosures:

The following people have nothing to disclose: Ahmed Abu Shanab, Zoe Hutchinson, Linda O'Keefe, Aiden McCormick, Raphael B. Merriman

1722

Discharge Destination, Caregiver Support, and Clinical Outcomes following Adult Liver Transplantation

Marina Serper1,2, Sean Koppe1, Raymond Kang3, Armen Changelian2, Daniela Ladner2, Josh Levitsky1,2;

1Division of Gas-troenterology and Hepatology, Northwestern University, Chicago, IL; 2Northwestern University Transplant Outcomes Research Collaborative, Northwestern University, Chicago, IL; 3Center for Healthcare Studies, Northwestern University Feinberg School of Medicine, Chicago, IL

Purpose:To study the relationship between discharge destination, caregiver support, and clinical outcomes after liver transplantation (LT). Methods:We studied a sample of 241 LT recipients from '06-'1 1 at a single center in Chicago, IL. Baseline demographics, discharge destination, pre-LT caregiver support, and clinical variables were obtained from medical records. Exclusion criteria were:in-hospital mortality post-LT, discharge to long-term acute care (LTAC). Clinical outcomes were:number of read missions and infections at 12 months, and patient mortality. Multivariate models adjusted for demographics, BMI, MELD, post-LT length of stay (LOS), living donor LT, presence of SLK, comorbidities, discharge destination, and pre-LT caregiver support. Results: The initial sample included 241 patients undergoing LT; 2 died post-operatively, 19 were discharged to LTAC; leaving N=220 for analysis. One third (N=63) were discharged to a facility after LT (skilled nursing, N=34; rehabilitation, N=29). Demographics, clinical characteristics and outcomes by discharge destination are in Table1. Discharge to a facility was associated with significantly more readmissions and infections at 12 months. In multivariate analysis, post-LT LOS and living alone pre-LT were independently associated with higher number of readmissions and infections at 12 months (all p<.05), and increased mortality (p<.01). Con-clusions:Non-home discharge, lack of caregiver support, and LOS are associated with increased morbidity and mortality following LT. Pre-LT assessment of adequate caregiver support and likelihood of post-LT home discharge may enhance patient selection and outcomes.

Baseline Characteristics and Outcomes by Discharge Destination

Variable, n(%)TotalDischargeDestinationP value
 (N=220)Facility (N=63)Home(N=157) 
Age≧65, n (%)108(49.1)34 (54.0)74(47.1)0.36
Male, n (%)138(62.7)38 (60.3)100(63.7)0.64
Race/Ethnicity, n (%)   0.07
White170 (77.6)44 (69.8)126(80.8) 
Black14 (6.4)9(14.3)5 (3.2) 
Hispanic/Latino27(12.3)8(12.7)19(12.2) 
Other8 (3.6)2(3.2)8(5.1) 
MELD score, M (SD)20.5(11.4)27.9(11.5)17.6(10.0)<0.0001
BMI, M (SD)27.4(7.8)26.9(9.1)27.6 (7.3)0.58
Live donor LT, n (%)32(14.5)3 (4.8)29(18.5)<0.01
Simultaneous liver kidney transplant, n (%)28(12.7)18(28.6)10(6.4)<0.0001
# of comorbidities, M (SD)1.5(1.2)1.7(1.3)1.4(1.1)0.06
Living alone pre-LT, n (%)26(11.8)10(15.9)16(10.2)0.24
Post-LT LOS in days, M (SD)16.2(18.3)29.1 (26.2)11.0(10.3)<0.0001
# of readmissions at 12 months, M (SD)1.7(1.7)2.1 (1.5)1.5(1.2)0.02
# of infections at 12 months, M (SD)0.9(1.2)1.5(1.5)0.7(1.0)<0.0001
Patient mortality, n(%)31(14.1)11 (17.5)20(12.7)0.36

Disclosures:

Josh Levitsky - Grant/Research Support: Salix, Novartis; Speaking and Teaching: Gilead, Salix, Novartis

The following people have nothing to disclose: Marina Serper, Sean Koppe, Raymond Kang, Armen Changelian, Daniela Ladner

1723

Evolution of lymphocyte subpopulations after immunosuppression withdrawal in tolerant liver transplant recipients

Rocio Garcia de la Garza1, Juana Merino4, Delia D'Avola1,2, Pablo Sarobe3, Juan J. Lasarte3, Jose A. Delgado4, Virginia Bel-sue3, Leyre Silva3, Mercedes Iñarrairaegui1,2, Bruno Sangro1,2, Iosu Sola5, Fernando Pardo6, Jorge Quiroga1,2, J Ignacio Her-rero1,2;

1Department of Medicine, Liver Unit, Universidad de Navarra, Pamplona, Navarra, Spain; 2CIBERehd, Pamplona, Spain; 3Division of Hepatology and Gene Therapy, Centro para la Investigación Médica Aplicada (CIMA), Pamplona, Spain; 4Department of Immunology, Clinica Universidad de Navarra, Pamplona, Spain; 5Department of Pathology, Clinica Universidad de Navarra, Pamplona, Spain; 6Department of Surgery, Clinica Universidad de Navarra, Pamplona, Spain

Some liver transplant (LT) recipients may develop tolerance (i.e. withdrawal of immunosuppression (IS) without developing rejection). The evolution of lymphocyte subpopulations after IS withdrawal has not been well described. Aim. to study the evolution of lymphocyte subpopulations in LT recipients during reduction of IS. Patients and methods. IS was proggressively reduced in 24 LT recipients after withdrawal or alteration of liver function tests. Fifteen of them (62%) were tolerant and were compared with the 9 non-tolerant patients. Total peripheral lymphocyte count and lymphocyte subpopulations (T, B, NK, CD4+, CD8+ and regulatory T cells (Treg)) were analysed in both groups before reduction of IS, when IS had been reduced at 50% of baseline, and 1 month after complete IS withdrawal (tolerant patients) or at the moment of alteration of liver function tests (non-tolerant patients). Results. In both groups, significant decreases of CD4+ and B lymphocytes, and significant increases of CD8+ and NK cells were found when IS had been reduced at 50% of baseline. These findings persisted at the end of he study. At this last moment, non-tolerant patients had a significant increase of Treg lymphocytes (9.09 vs 5.73; P=0.001), that was not found in tolerant patients (8.13% vs 7.97%; P=0.395). Conclusions. Immunosuppression withdrawal is followed by a recovery of CD8+ and NK subpopulations. Non-tolerant patients had an increase of Treg lymphocytes.

Evolution of lymphocyte subpopulations (baseline versus 50% immunosupprssion reduction)

 Tolerant Non-tolerant 
 Median (basal vs 50%)PMedian (basal vs 50%)P
CD4+ (%)41.5/35.00.04641.7/36.90.018
CD8+ (%)25.2/27.80.01737.0/41.60.004
B lymphocytes (%)9.5 / 8.70.0029.9 / 9.30.013
Natural Killer (%)14.0/20.00.0038.3/11.90.016

Disclosures:

Bruno Sangro - Speaking and Teaching: Sirtex Medical Europe, Bayer Schering Pharma

J Ignacio Herrero - Speaking and Teaching: Roche, Astellas, Novartis; Stock Shareholder: GlaxoSmithKline, Roche, Novartis, Abbott

The following people have nothing to disclose: Rocio Garcia de la Garza, Juana Merino, Delia D'Avola, Pablo Sarobe, Juan J. Lasarte, Jose A. Delgado, Virginia Belsue, Leyre Silva, Mercedes Iñarrairaegui, Iosu Sola, Fernando Pardo, Jorge Quiroga

1724

Outcomes of Liver Transplantation in Hepatitis C-Positive Patients Receiving Rabbit Anti-thymocyte Globulin Induction

Alexa J. Ray1,2, John K. Wright1,3, Seth J. Karp1,3, Chan Y. Chung1,4, Irene D. Feurer1,5, Holly Sybert6, Christie B. Truscott1;

1Vanderbilt Transplant Center, Vanderbilt University Medical Center, Nashville, TN; 2Department of Pharmaceutical Services, Vanderbilt University Medical Center, Nashville, TN; 3Department of Surgery, Vanderbilt University Medical Center, Nashville, TN; 4Department of Medicine, Vanderbilt University Medical Center, Nashville, TN; 5Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN; 6Lipscomb University College of Pharmacy, Nashville, TN

Purpose: Rabbit antithymocyte globulin (rATG) induction therapy has been utilized in orthotopic liver transplantation (OLT) with the goal of delaying initiation of calcineurin inhibitor maintenance immunosuppression as a renal-sparing technique. Evidence-based data are needed to elucidate the effect of rATG induction on clinical outcomes such as post-transplant renal function, patient survival, and complications such as infection and rejection. Additionally, in patients transplanted for hepatitis C virus (HCV) cirrhosis, histologic recurrence of HCV is an outcome of interest. This study evaluated the effects of rATG versus standard methylprednisolone (MP) induction on patient survival, infection-free, rejection-free, and HCV recurrence-free survival, and post-transplant serum creatinine (SCr) in HCV-pos-itive patients. Cost of each induction agent was also assessed. Methods: A single-center retrospective analysis was completed for patients transplanted between November 2006 and May 201 1. Data were analyzed using Kaplan-Meier and Cox regression models. Results: The sample included 106 adults, of whom 33 (31%) received rATG induction immunosuppression. After controlling for the model for end-stage liver disease (MELD) score, pre-transplant serum creatinine, and episodes of infection and rejection, rATG induction was not associated with patient survival (p=0.189). Additionally, rATG induction was not associated with infection-free survival time (p=0.394), rejection-free survival time (p=0.1 75), or HCV recurrence-free survival time (p=0.67). A significant difference was detected with regard to post-transplant trend in serum creatinine between the two patient groups (p<0.001). Drug cost was approximately a hundred-fold higher in the rATG group compared to the MP group. Conclusions: No significant difference in post-transplant patient survival was detected between patients receiving rATG for induction immunosuppression and those receiving standard MP induction therapy. Likewise, there was no significant difference in infection-free survival, rejection-free survival, or HCV recurrence-free survival times between the patient groups. A significant difference was detected with regard to post-transplant trend in SCr between the two patient groups. Notably, it was not possible to control for the comparative incidence of hepatorenal syndrome prior to transplant in this retrospective study. An assessment of costs for these induction agents demonstrates significantly higher drug costs with rATG use, which may not be justifiable to many health systems given its uncertain clinical benefit.

Disclosures:

Seth J. Karp - Patent Held/Filed: Vanderbilt University, Harvard University

The following people have nothing to disclose: Alexa J. Ray, John K. Wright, Chan Y. Chung, Irene D. Feurer, Holly Sybert, Christie B. Truscott

1725

Switch from Calcineurin to mTOR based immunosuppression does not improve long term outcome in kidney function in liver transplant patients

Doris Wagner1, Florian Iberer1, SabineZitta3, Andreas Meinitzer4, Hans Jörg Mischinger2, Alexander R. Rosenkranz3;

1Department of Surgery, Division for Transplantation, Medical University of Graz, Graz, Austria; 2Department of Surgery, Division for General and Visceral Surgery, Medical University of Graz, Graz, Austria; 3Department of Internal Medicine, Divison for Nephrology, Medical University of Graz, Graz, Austria; 4Department for Laboratory Medicine, Medical University of Graz, Graz, Austria

Introduction: To identify an immunosuppressive regimen that preserves renal function while maintaining its efficacy still seems to be an unmet urgent medical need following liver transplantation (OLT). mTOR inhibitors have been proposed to preserve renal functions by estimated glomerular filtration rates (eGFR). The presented study investigated a potential effect of mTORs compared to calcineurin inhibitors (CNI) on renal function after OLT defined by a measured GFR. Patients and Methods: Patients were assigned to two groups - CNI and mTOR. GFR was measured using the inulin clearance (IC). Estimated GFR and proteinuria were determined using the MDRD 4 formula at baseline, 6, 12, 18 and 24 months. Changes in renal function were compared in patients on CNI or mTOR based immunosuppression. Results: The eGFR declined significantly in patients on CNI (81 vs. 61 mL/min1 .73m2, p=0.01) and on mTOR (82 vs 60 mL/min1 .73m2, p=0.01) treatment until study end. The decline was more pronounced in patients on CNI in the first 12 months (CNI: -11% vs. mTOR: -5%, p=<0.01). The IC did not change significantly throughout the study period in CNI (62 vs 61 mL/min1.73m2, p=n.s.) or mTOR (58 vs. 60 mL/min1 .73m2, p=n.s.) patients. Throughout the study period the proportion of patients with proteinuria increased significantly in the mTOR group (50% vs. 62%, p=<0.03) and not in the CNI group (39% vs. 42%, p=n.s.) Discussion: The measured GFR did not confirm protective effect of mTOR based IS that has been proposed by the eGFR.

Disclosures:

The following people have nothing to disclose: Doris Wagner, Florian Iberer, Sabine Zitta, Andreas Meinitzer, Hans Jörg Mischinger, Alexander R. Rosenkranz

1726

10 year longitudinal follow up of consecutive liver transplants for HCV under tacrolimus based triple immunosuppression: Single center experience

Randeep Kashyap, M. Katherine Dokus, Neil G. Kumar, Carlos E. Marroquin, Christopher T. Barry, Mark S. Orloff;

Solid Organ Transplant, University of Rochester Medical Center, Rochester, NY

Background: There are very few studies reporting on the long-term outcomes of liver transplants for HCV+ patients under Tacrolimus based triple immunosuppression therapy. The aim of this study is to examine the long-term outcome of patients transplanted for HCV with at least 1 0 years of follow-up. Methods: 166 primary OLTs for HCV cirrhosis performed between Jan1 993 and jan 2003. and were followed up until Jan 201 3. Patient and graft survivals with causes of death and retrans-plantation, maintenance immunosuppression, and adverse effects were examined. Results: The study population includes 126 males and 40 females comprising 135 patients 18–55 years old (adults), and 31 individuals 55+ years (seniors). The mean follow-up was 96 (median: 1 04) months. The overall 1 0-year actuarial patient and graft survivals were 51.2% and 44%, respectively. At the last follow-up, 79 patients were alive; 66(83.5%) adults, and 1 3 (1 6.5%) seniors. After the first post-OLT year, cardiopulmonary events, recurrence of primary disease, and malignancy were the main causes of death Overall, 26 recipients underwent retransplantation mainly for primary nonfunction, hepatic artery thrombosis, and recurrent primary disease. Twenty-six required post-transplant dialysis, and 4 (2.4%) underwent subsequent kidney transplant. Conclusions: The overall 1 0-year actuarial patient and graft survivals were 51% and 44%, respectively, with better survival among adults. Age-related complications, recurrence of primary disease, and malignancy were the major causes of late graft loss. Graft loss related to immunologic reasons was rare. The prevention of recurrent disease with newer and novel agents, such as the addition of protease inhibitors, will further improve the long-term results of liver transplant for HCV. It will be interesting to follow future long term outcomes of the new HCV therapies with Cyclosporine based immunosuppression.

Disclosures:

Carlos E. Marroquin - Speaking and Teaching: Bayer

The following people have nothing to disclose: Randeep Kashyap, M. Katherine Dokus, Neil G. Kumar, Christopher T. Barry, Mark S. Orloff

1727

Liver transplantation beyond age 65: outcomes and predictors

Tianyan Chen1, Ahsan Alam2, Xianming Tan3, Peter Ghali1;

1Gas-troenterology, McGill University Health Center, Montreal, QC, Canada; 2Nephrology, McGill University Health Centre, Montreal, QC, Canada; 3Biostatistics, McGill University Health Centre, Montreal, QC, Canada

Background and aims: Most transplant centers currently consider age > 65 a contraindication to orthotopic liver transplant (OLT). It is unclear the optimal age of transplantation and whether any pre-transplant variables would predict the outcomes in elderly transplant recipients. Given the aging end-stage liver disease population, this practice should be re-evaluated. Methods: We conducted a retrospective cohort study including 643 adult patients who underwent OLTs from June 1 990 to September 2012 at our University Center. Living donor transplant recipients, multiple solid organ transplantations, and re-transplants were excluded. Patients were stratified into four groups based on their age and pre-transplant renal function: age < 65 (young) and > 65 (old), eGFR < 60 and > 60 mL/min/1.73 m2. Patient and graft survival post transplant were analyzed separately using Kaplan-Meier plots. The covari-ates were calculated MELD score, eGFR, and creatinine. Cox proportional hazards models were used to identify predictors of patient and graft survival. Results: 1 16 out of 643 (18%) patients were age 65 and older at the time of transplantation. With the exception that more patients with alcohol related cirrhosis were present in the young group, the remaining baseline characteristics were similar between the two groups. The medians for age, pre-transplant serum creatinine, and MELD score were 55 (18–64), 102 mmol/L, and 11 in the young, and 68 (65–74), 1 02 mmol/L, 1 1 in the old group, respectively. Older patients with a low eGFR had poor survival. Older patients with preserved renal function (pre-transplant eGFR > 60 mL/min/1.73 m2 or creatinine < 140 mmol/L) appeared to have both better patient and graft survivals till 3 years, and comparable till 5 years post transplant. Beyond 5 years, survival in the elderly was lower than in the young. Multivariate analysis showed recipient age and MELD score were statistically significant predictors for patient survival, whereas pre-transplant serum creatinine, donor age, and non-alcoholic steatohepatitis cirrhosis were for graft survival. Conclusions: Appropriately selected elderly patients (age > 65) had similar 5 year post-transplant outcomes to their younger counterparts. This was especially true in elderly patients with adequate pre-transplant renal function or hepatoma with relatively preserved hepatic function. However, beyond 5 years, survival in this group is lower. Renal impairment and severity of liver disease are independent and more important as short term prognostic factors than recipient age.

Disclosures:

Ahsan Alam - Advisory Committees or Review Panels: Otsuka Canada

The following people have nothing to disclose: Tianyan Chen, Xianming Tan, Peter Ghali

1728

Diastolic dysfunction is associated with post OLT atrial fibrillation

Hye Yeon Jhun1, Maha R. Boktour2, Catherine T. Frenette3, Asham Emad2, Howard P. Monsour1, R. Mark Ghobrial2;

1Internal Medicine, The Methodist Hospital, Houston, TX; 2Surgery, The Methodist Hospital, Houston, TX; 3Hepatology, Scripps, La Jolla, CA

Background: Diastolic dysfunction is commonly observed in patients with end stage liver disease due to hyperdynamic circulation and left ventricular stiffness. However the adverse outcomes after orthotopic liver transplantation (OLT) are poorly defined. The aim of the study is to assess the cardiac complications after OLT in patients with diastolic dysfunction. Methods: Data from 191- primarily OLT recipients between April 2008 and April 2012 were retrospectively reviewed. Recipients with multi-organ transplantation other than liver/kidney (n=8) were excluded. Demographic information, clinical characteristics and post OLT complications of 152 patients who had echocardiogram prior to OLT were analyzed. Recipients with diastolic dysfunction defined by echocardiographic parameters (n=39) were compared to those with normal diastolic function (n=1 13). Results: Mean patient age was 56±10 years, 93 (61%) were male, 96 (63%) were white, and 76 (50%) had HCV. At OLT, median biological MELD was 22 (range 6–49). There were no significant differences between groups for demographic and clinical characteristics. However, recipients with diastolic dysfunction were more likely to be older [mean age was 61 ±6 vs 55±10, p=<0.001], and showed a higher incidence of history of diabetes mellitus pre-OLT [20 (51%) vs.32 (28%), p= 0.01] as compared with controls (OR=2.7; 95% confidence CI, 1.3 -5.6). Both groups exhibited low incidence of complications post-OLT, however recipients with diastolic dysfunction showed more incidence of atrial fibrillation [5 (13%) vs. 4(4%), p= 0.03]. Overall median follow-up was 27.2 months with 1-year patient survival of 93%. Conclusions: Diastolic dysfunction was associated with development of atrial fibrillation after liver transplantation. However there was no significant difference in other cardiac complications. This study implicates that diastolic dysfunction can be an important marker to predict atrial fibrillation after liver transplantation.

Baseline characteristics and outcomes

Study variablesTotal n=152No diastolic dysfunction n=l 13Diastolic dysfunction n=39P value
Age, mean± sd (range)56±10 (21–77)55±10(21–77)61±6 (50–72)0.0006
Male, n(%)93(61)66(58)27 (69)0.23
Race, white, n(%)96 (63)75 (66)21 (54)0.16
HCV seropositive, nC7<076 (50)58(51)18(46)0.58
MELD Median, range22 (6–49)22 (6–48)24 (6–49)1.0
Hx of DM, n(%)52 (34)32 (28)20(51)0.01
Hx of HCC, n(%)67(44)49 (43)18 (46)0.76
Hypotension, n(%)39 (26)28 (25)11 (28)0.67
Arrhythmia, n(%)15 (10)9(8)6(15)0.18
Elevated troponin, n(%)14 (9)11 (10)3(8)0.68
Atrial fibrillation, n(%)9(6)4(4)5(13)0.034

Disclosures:

Catherine T. Frenette - Speaking and Teaching: Onyx, Salix, Gilead

The following people have nothing to disclose: Hye Yeon Jhun, Maha R. Boktour, Asham Emad, Howard P. Monsour, R. Mark Ghobrial

1729

An improved model for preoperative assessment of the risk of recurrence after surgical treatment of hepatocel-lular carcinoma

Nan Jiang, Guo-Ying Wang, Hua Li;

Liver Surgery and Transplant Center,Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China

Background:TNM, Okuda, CLIP and BCLC systems are widely used for assessing prognosis in patients with HCC, but are not tailored to the assessment of recurrence. In the current study, we developed a model to assess, preoperatively, the risk for recurrence in HCC patients after radical surgical treatment. The newly developed system was compared with the TNM, Okuda, CLIP and BCLC systems. Recurrence-free survival(RFS) rate was compared between patients receiving LR(liver resection) vs. LT(liver transplantation). Methods: This retrospective study included 336 HCC patients receiving LR(n=196) or LT(n=140) during a period from 2003 to 201 0.Risk factors were identified using a step-forward approach (univariate analysis followed by multivariate analysis using Cox regression). Independent risk factors identified with this approach were used to construct a model to predict recurrence. Survival data were analyzed using Kaplan-Meier estimation and the log-rank test.The predictive value of this method was evaluated using a receiver operator characteristic curve(ROC) analysis, and compared to that with commonly used methods. Discriminatory ability was evaluated using likelihood-ratio chi-square,liner trend chi-squarestatistic and the Akaike information criterion(AIC). Result: The model included four factors: total tumor volume(TTV), hepatitis B surface antigen(HBsAg)status, Child-Turcotte-Pugh(CTP) class and portal vein tumor thrombus(PVTT). A score of 0 or 1 was assigned to each factor, with a possible highest score of 4 for classification of the patients into A (0–1), B(2), C(3), or D(4) stage. Reccurence free survival(RFS) rate differed significantly across the four stages (3-yr RFS rate at 65.8%, 39.5%, 1 7.0% and 0% for stage A, B, C and D, respectively; p<0.05).The AUC was 0.742 for this newly proposed model, vs. 0.648 with TNM, 0.598 with OKUDA, 0.622 with CLIP, and 0.650 with BCLC. The AIC was 405.077 for the new model, vs, 431.571 with TNM, 450.788 with OKUDA, 443.975 with CLIP and 439.473 with BCLC. 3-year RFS rate was higher for LT than LR in patients in stageA,B (stage A, 67.3%vs.54.6%, p<0.01; stage B, 40.0%vs.30.4%, p<0.05), but stage C (15%vs 0%, p=0.201). All the patients in stage D appeared recurrence. Conclusion: We establish a new simplified model was more discriminant than existing staging systems in classifying patients into different risk groups and was better at predicting recurrence free survival. The patients received LT had higher RFS rate than LR in stage A and B according to the new model.

Disclosures:

The following people have nothing to disclose: Nan Jiang, Guo-Ying Wang, Hua Li

1730

Body Composition In Cirrhotic Patients After Liver Transplantation

Joseph T. Bergerson3, June-Goo Lee4, Alessandro Furlan4, Achuthan Sourianarayanane1,2, Amit D. Tevar2,6, Andrea DiMar-tini5,2, Michael A. Dunn1,2;

1Center for Liver Diseases, University of Pittsburgh, Pittsburgh, PA; 2Starzl Transplant Institute, University of Pittsburgh, Pittsburgh, PA; 3Department of Medicine, University of Pittsburgh, Pittsburgh, PA; 4Department of Radiology, University of Pittsburgh, Pittsburgh, PA; 5Department of Psychiatry, University of Pittsburgh, Pittsburgh, PA; 6Department of Surgery, University of Pittsburgh, Pittsburgh, PA

Background and Aim: Sarcopenia, a strong indicator of decon-ditioning in cirrhosis, predicts mortality awaiting and following liver transplantation. There are no data on the impact of transplantation on progression or recovery from sarcopenia as commonly measured on abdominal CT imaging. We explored whether consistent measurable changes would occur in sarcopenia, visceral and subcutaneous adipose tissue, and bone density, all important determinants of recovery, in 63 patients (38 men, 25 women) transplanted for primary sclerosing cholangitis (PSC), nonalcoholic steatohepatitis (NASH) and alcoholic cirrhosis (AC). Methods: We chose PSC, NASH and AC for initial study because of their low likelihood of early recurrence. We selected patients without the potentially confounding metabolic stresses of major surgery, renal failure or a critical care admission within 1 year after transplant and with archived scans available for comparison before and from 1 to 2 years after transplant. Scans were done for clinical indications including abdominal pain and biochemical abnormalities. Using transverse 5mm CT slices at L3 for segmentation analysis, we measured skeletal muscle index (SMI), visceral fat and subcutaneous fat as areas per height squared, and vertebral bone density as Hounsfield units (HU). We compared data with published standard imaging criteria for sarcopenia and osteoporosis. Results: As shown in the table, sarcopenia (SMI <52.4cm2/m2 men, <38.5cm2/m2 women) was highly prevalent in 70% of patients before and 56% 1 year or more after transplant, and was markedly prominent and sustained in PSC (20 of 22 before and 21 of 22 after transplant). It showed a nonsignificant trend toward improvement in 63% of patients. Osteoporosis (<1 60HU) was present in 60% of patients before and 73% a year after transplant and worsened in 75%. Both visceral and subcutaneous fat was most pronounced before and continued to nonsignificantly worsen after transplant in NASH cirrhosis. Conclusion: We found that major body composition abnormalities are common in cirrhosis and generally fail to resolve or significantly improve a year after transplantation. They appear to be important measurable therapeutic targets for improving post-transplant care. Given the limitations of our selected retrospective dataset, we suggest that prospective evaluation of serial body composition measurements as indicators of recovery and treatment efficacy would help define their clinical value.

 Muscle Mass, cm2/m2Visceral Adipose Tissue, cm2/m2Subcu Adipose Tissue, cm2/m2Bone Density, HU
PSC Pre/Post31.1+11.6/ 33.4±9.743.1±16.6/46.3±22.039.0 ±25.2/ 49.6±27.4157.8±36.3/ 146.7±45.4
NASH Pre/Post49.2±16.0/ 5].3±8.7S9.7±36.0/94.5+39.476.3±36.5 / 87.3±38.1142.5±44.0/ 129.3±38.3
Alcohol Pre/Post43.]±12.4/ 44.9±12.772.1 ±44.6/77.8±40.747.2±36.9 / 62.2±28.8141.2±39.7/ 122.6±46.4

Disclosures:

The following people have nothing to disclose: Joseph T. Bergerson, June-Goo Lee, Alessandro Furlan, Achuthan Sourianarayanane, Amit D. Tevar, Andrea DiMartini, Michael A. Dunn

1731

Portal hypertension in patients with polycystic liver disease listed for liver transplantation is not associated with greater morbidity and mortality

Neil Rajoriya1, Joanna A. Leithead1,3, Sophie Lord1, Bridget Gun-son3, Simon Bramhall2, James W. Ferguson1, Gideon M. Hirschfield1;

1Liver Medicine, Queen Elizabeth Hospital, Bir-migham, United Kingdom; 2Liver Surgery, Queen Elizabeth Hospital, Birmigham, United Kingdom; 3NIHR Biomedical Research Unit and Centre for Liver Research,, University of Birmingham, Birmigham, United Kingdom

Polycystic liver disease (PCLD) is an established indication for liver transplantation, most commonly in the absence of hepatic synthetic dysfunction. Portal hypertension (PH) is however a significant concern in advanced disease. We sought to establish the impact of portal hypertension on wait-list and post transplant morbidity/mortality in patients with PCLD listed for liver transplantation. Methods: A retrospective single-centre study of consecutive patients listed for liver transplantation 01/1990–12/2012. PH at listing was defined as the presence of: documented varices, splenomegaly (without splenic cysts), thrombocytopenia, or ascites without hepatic venous outflow obstruction (HVOO). Results: During the follow-up period 75 patients with PCLD were identified within our practice. Of these, 42.5% were identified as having PH. Within this cohort, 47 patients (62.7%, M:F=5:42) were listed for liver transplantation (single organ, n=35; combined liver-kidney transplantation, n=12). At listing, 19 patients (40.4%) had PH (varices, 2.1%; splenomegaly, 19.1%; thrombocytopenia, 10.6%; ascites, 1 7.0%). An additional 3 patients had ascites in the context of HVOO. The mean listing age (PH, 50.6; nonPH, 47.1 years; p=0.101), median listing MELD (PH, 12; nonPH, 11; p=0.422) and median listing UKELD (PH, 48; nonPH, 46; p=0.344) were similar for PH and nonPH groups. 84.2% of patients with PH had stage 3–5 chronic kidney disease compared to 67.9% of patients without PH (p=0.179). 19.2% and 7.1% of PH and nonPH patients were dialysis dependent, respectively (p=0.317). 34 patients were transplanted by the time of data analysis (72.3%). The median time from listing to transplantation for PH patients was 72 (IQR 34–524) days and for nonPH patients was 139 (IQR 48–390) days (p=0.466). Two PH patients died prior to transplantation and 1 nonPH patient. In the patients who underwent liver transplantation alone, there was no difference in the duration of ITU (PH, 3; nonPH, 2 days; p=0.188) and hospital stay (PH, 9; nonPH, 10 days; p=0.973), or frequency of renal replacement therapy (PH, 2/8; nonPH, 1/14; p=0.121). Similar observations were made in the patients who underwent combined liver-kidney transplantation. Estimated 15-year survival post transplant was 100% and 58.6% for PH and nonPH groups, respectively (log-rank p=0.138). Conclusions: Clinically apparent PH in patients with PCLD listed for liver transplantation is common but does not impact on wait-list and post transplant morbidity and mortality.

Disclosures:

James W. Ferguson - Advisory Committees or Review Panels: Astellas, Novartis

Gideon M. Hirschfield - Advisory Committees or Review Panels: Centocor/J&J, Medigene, Intercept, Falk Pharma ; Consulting: Lumena, Intercept

The following people have nothing to disclose: Neil Rajoriya, Joanna A. Leithead, Sophie Lord, Bridget Gunson, Simon Bramhall

1732

An assessment of Coronary Artery Calcification (CAC) in Orthotopic Liver Transplant (OLT) patients

Norma C. McAvoy1, Alasdair W. Hay2, Graham McKillop3, David E. Newby4, Peter C. Hayes5,1;

1Department of Hepatology, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom; 2Dept of Anaesthetics, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom; 3Dept of Radiology, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom; 4Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, United Kingdom; 5University of Edinburgh, Edinburgh, United Kingdom

Background: Cardiovascular disease is increasing worldwide. For patients being assessed for OLT, it is imperative to accurately stratify cardiovascular (CV) risk, to allow them to receive the maximal benefit from transplantation. Coronary artery calcification (CAC) is a novel and independent predictor of cardiovascular risk, the prevalence of which has been previously described in patients with end stage liver disease (ESLD). We have recently demonstrated CAC as an independent risk factor for cardiac events post OLT. The correlation of CAC with classical and novel CV risk factors in the OLT group is currently unclear. Aim: To determine the relationship of CAC with classical and novel CV markers in patients with ESLD undergoing OLT assessment. Methods: Single centre, prospective observation study. CAC scores were derived by the Agatson method from thoracic CT scans with patients followed up from time of assessment. Results: 199 patients recruited (125M:74F) with mean age 55.12 years (range 23–70 yrs). 121 patients were listed for OLT with 97 undergoing OLT. Median follow up of 41.28 months. CAC was performed in 177 patients with mean CAC of 381.84 (range 0–1017). A significant correlation was identified between the CAC score and age (p=0.000), gender (p=0.001), fasting glucose (p=0.01 8), Cystatin C 0.029, Fram-ingham risk score (p=0.000), PR interval (p=0.005) and number of vessels involved (p=0.000). The mean CAC scores in ALD appeared higher than in non-ALD aetiology; at 608 vs 238 respectively. Conclusion:- This study confirms the high prevalence of occult coronary artery disease in OLT assessment patients and identifies a relationship between CAC scores and age, gender, fasting glucose, Cystatin C, Framingham risk score and number of vessels involved but no relationship seen with other classical CV risk factors.

Disclosures:

Peter C. Hayes - Advisory Committees or Review Panels: Roche, Roche, Jannsen, Gilead, ?ONO, Norgine; Grant/Research Support: Novartis; Speaking and Teaching: Roche, MSD, Roche, MSD, Pfiser, Gore, Falk, Ferring

The following people have nothing to disclose: Norma C. McAvoy, Alasdair W. Hay, Graham McKillop, David E. Newby

1733

Inclusion of Sarcopenia within MELD (MELD-Sarcopenia) and the Prediction of Mortality in Patients with Cirrhosis

Aldo J. Montano-Loza1, Judith Meza-Junco2, Carla M. Prado3, Vickie Baracos2, Michael Sawyer2, Crystal Beaumont2, Mang M. Ma1, Norman Kneteman4, Robert P. Myers5;

1Division of Gas-troenterology & Liver Unit, University of Alberta, Edmonton, AB, Canada; 2Medical Oncology, Cross Cancer Institute, Edmonton, AB, Canada; 3Nutrition, Food and Exercise Sciences, The Florida State University, Tallahassee, FL; 4Surgery, Division of Transplantation, University of Alberta, Edmonton, AB, Canada; 5Liver Unit, University of Calgary, Calgary, AB, Canada

Background/Aims: Since the adoption of MELD by liver transplantation programs in North America, reductions in the number of deaths on the waiting list have been reported. However, MELD has limitations including its failure to include assessments of the nutritional and functional status of cirrhotic patients. In this study, our objective was to evaluate whether the inclusion of muscularity assessment (i.e. sarcopenia) within MELD could improve the prediction of mortality in patients with cirrhosis. Methods: We evaluated 528 cirrhotic patients who were consecutively evaluated for liver transplantation and had a computed tomography (CT) scan at the 3rd lumbar vertebrae within 6 months of evaluation. The skeletal muscle index at the 3rd lumbar vertebra (L3 SMI) was measured by CT and sarcopenia was defined using previously published gender and BMI-spe-cific cutoffs. Using Cox proportional hazards regression, a novel MELD-sarcopenia score was derived and discrimination of MELD and MELD-sarcopenia for mortality were determined using c-statistics for survival data. Patients were followed until death, liver transplantation or the end of follow-up. Results: 357 patients were males (68%); the etiology of cirrhosis was HCV in 38%, alcohol in 21%, cryptogenic/NASH in 25%, autoimmune liver disease in 9%, and HBV in 6%; and 42% had concomitant HCC. By Cox regression analysis adjusted for age, gender, and presence of HCC, both MELD (hazard ratio [HR] 1.06, 95% confidence interval [CI] 1.04–1.08, P<0.001), and the L3 SMI (HR 0.98, 95% CI 0.97–0.99, P=0.003) were associated with mortality. Sarcopenia was present in 235 patients (45%); sarcopenic patients had shorter median survival than patients without sarcopenia (20±2 vs. 58±22 months, log-rank P<0.001). The c-statistic for 3-month mortality was 0.68 (95% CI 0.60–0.76) for MELD and 0.72 (95% CI 0.65–0.79) for MELD-sarcopenia. Corresponding figures for 6-month mortality were 0.65 (95% CI 0.58–0.72) and 0.69 (95% CI 0.63–0.75), respectively. In a subgroup analysis including only patients listed for transplantation (n=363), the c-statistics for 3- and 6-month mortality were 0.72 (95% CI 0.63–0.82) and 0.71 (95% CI 0.63–0.79) for MELD, and 0.76 (95% CI 0.69–0.84) and 0.75 (95% CI 0.68–0.83) for MELD-sarcopenia, respectively. Conclusions: Modification of MELD to include sarcopenia is associated with a modest improvement in the prediction of mortality in patients with cirrhosis. However, prior to the widespread use of MELD-sarcopenia, additional validation in larger cohorts of patients with cirrhosis is necessary.

Disclosures:

Robert P. Myers - Advisory Committees or Review Panels: Roche, Merck, Vertex, Norgine Ltd., GE Healthcare ; Grant/Research Support: Echosens, Roche, Merck; Speaking and Teaching: KNS Canada, Roche, Merck, Vertex

The following people have nothing to disclose: Aldo J. Montano-Loza, Judith Meza-Junco, Carla M. Prado, Vickie Baracos, Michael Sawyer, Crystal Beaumont, Mang M. Ma, Norman Kneteman

1734

Role of Cardiac Calcium Score in identifying occult Coronary artery Disease, among patients undergoing Liver Transplant Work up

Eren Taydas1, Mohammad U. Malik2, Stuart D. Russell2, Matthews Chacko2, Andrew M. Cameron2, Ahmet Gurakar2;

1Internal Medicine, MedStar Franklin Square Medical Center, Baltimore, MD; 2The Johns Hopkins Hospital, Baltimore, MD

Background:Cardiac evaluation is an important part of the liver transplantation work up to achieve higher post-transplant survival rates. Identification of Coronary Artery Disease (CAD) among asymptomatic candidates is done by cardiac stress tests & Transthoracic ECHO. Cardiac catheterization is considered to be the gold standard in investigating for the presence and extent of CAD. Methods:We investigated the role of Cardiac Calcium Score (CCS) to predict the presence of CAD among patients with major cardiac risk factors including DM, HTN and smoking. Following IRB approval, medical records of patients evaluated at Johns Hopkins Liver Transplant Program, between 1/1/2011 & 5/1/2013 were retrospectively reviewed. Results:CCS was obtained on 66 patients out of 1 75, after discussing with consultant cardiologist on the liver transplant team. There were 40 males and 26 females. Mean age was 57.8 ± 0.7 years. Mean MELD was 15.2 ± 6.8. Mean CCS was 379.6 ± 639.8. Cardiac cath was done by a single experienced interventional cardiologist on 16 patients. Four patients had coronary obstruction of <50%, 12 patients had maximum obstruction of >50% (range 50–90%). Mean CCS was 750.3 ± 714 in patients with <49% stenosis and 883.8 ± 659.7 in patients with >50% stenosis. There was no statistical difference between the groups (P > 0.05). The ROC analysis revealed 91% sensitivity and 50% specificity at cutoff of 243 for the identification of occult CAD. Conclusion:Early identification of CAD is important to be able to achieve better post-transplant outcomes. Currently used Stress tests are known to have a lower sensitivity but a higher specificity. CCS above 250 can be considered to select which patients will benefit from further investigation with a cardiac cath. This may eventually eliminate unnecessary procedures in high risk patients with end stage liver disease. More data is needed to verify these findings.

Disclosures:

Ahmet Gurakar - Advisory Committees or Review Panels: Gilead, Gilead; Grant/Research Support: BMS, BMS

The following people have nothing to disclose: Eren Taydas, Mohammad U. Malik, Stuart D. Russell, Matthews Chacko, Andrew M. Cameron

Ancillary