Pediatric Liver Disease


Myeloid dendritic cells maintain Th1 7 lymphocyte responses during the obstructive phase of experimental biliary atresia

Celine S. Lages1, Julia Simmons1, Claire Chougnet3, Kumar S. Shanmukhappa2, Avery Maddox1, Betsy DiPasquale2, Alexander G. Miethke1;

1Pediatric Gastroenterology, Hepatology & Nutrition, Cincinnati Children's Hospital Medical Center, Cincinnati, OH; 2Pediatric Pathology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH; 3Immunology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH

Plasmacytoid dendritic cells (DC) were reported to drive innate injury to biliary epithelium during the early phase of biliary atresia (BA). We hypothesize that myeloid (m)DC coordinate T-cell injury during the obstructive phase of the disease. Methods: Transgenic CD1 1c-DTR GFP (DC reporter) mice were used to study DC in the pathogenesis of rhesus rotavirus induced BA. Immune responses were determined by cytometry, real time RT PCR and Luminex-based quantification of plasma cytokines. Results: Following injection on day 1 of life, the frequency of total DC increased by 6fold in the liver (1.2 % of CD1 1c+/CD3-7AAD- vs 0.2% in saline control, p<0.001) at 12 days post infection (dpi), with particular expansion of the CD11b+ mDC subset (0.4% vs 0.05% in saline, p<0.01). Since CD1 03+ mDC were reported to link innate with adaptive immune responses at inflammatory sites, we compared their frequency in liver and primary lymphoid tissue and found 2-fold increase of CD1 03+ cells in the mDC subset in the liver (1 6.5% of CD103+/mDC vs 8.5% in spleen, p=0.03). Immunohisto-chemistry (IHC) revealed accumulation of GFP+ immune cells in portal tracts, especially around bile duct profiles, at 12dpi. Accumulation of mDC was accompanied by intrahepatic expansion of CD4+ T cells by 4.3fold compared with non-infected controls (p<0.01). Hepatic mRNA concentration for IL17A was increased by 9.4fold between 8 and 12dpi after viral challenge. To investigate the effects of DC on CD4+ cells and Th1 7 responses in the liver after viral challenge, 0.01ng/gram bw of diphtheria toxin (DT) was injected every 48 hours beginning at 3dpi. The number of CD1 1c+ DC was significantly reduced in DT compared with vehicle treated mice at 8dpi (2.5x10^4/100mg tissue vs 5.0 in vehicle group, p<0.01). Anti-GFP-IHC demonstrated near absence of CD1 1 c+ immune cells from portal tracts. DC-depletion reduced CD4+ T cell content in the liver (5.0 vs 14.2x1 0^4/l 00mg tissue in vehicle; p=0.02) and their activation (24.1% vs 39.1% of CD69/CD4 in vehicle, p=0.03) at 12dpi. Hepatic mRNA expression of IL1 7A was down-regulated by 2.3fold (p=0.05), and plasma Th1 7-associated cytokine levels were reduced in DT treated mice compared with controls at 12dpi (IL-17F: 147.2 pg/ml vs 367.5, p=0.04; IL6: 21.3 and 34.5pg/mL, p=0.07, in DT vs vehicle treated mice, respectively). Importantly, diminished hepatic CD4+ cells and Th1 7 responses were associated with reduced plasma total bilirubin levels in DC-depleted mice at 12dpi (6.1 vs 1 1.3mg/dL in control; p<0.01). Conclusions: Myeloid DC maintain hepatic CD4 and Th1 7 responses and regulate cholestatic liver injury during the obstructive phase of experimental BA.


The following people have nothing to disclose: Celine S. Lages, Julia Simmons, Claire Chougnet, Kumar S. Shanmukhappa, Avery Maddox, Betsy DiPasquale, Alexander G. Miethke


Early and Late Liver Transplantation for Children with Biliary Atresia: is there a difference in the outcome? Analysis of the UNOS database

Ronen Arnon1,2, Rachel A. Annunziato3, Guiseppe D'Amelio3, Jaime Chu1,2, Benjamin L. Shneider4;

1RMTI, Mount Sinai Medical Center, New York, NY; 2Pediatrics, Mount Sinai Medical Center, New York, NY; 3Psychology, Fordham University, Bronx, NY; 4Pedi-atrics, Children's Hospital of Pittsburgh of UPMC, Pittsburgh, PA

Biliary Atresia (BA) is the most common indication for liver transplantation (LT) in children.The main indications for LT in these patients are poor bile drainage, failure to thrive, complications of biliary cirrhosis and recurrent cholangitis. Aim: To compare the clinical characteristics and the outcome of first isolated LT (FILT) in children with BA who were transplanted early (below 2 years old) and late (age 3–18 years).Methods: Children with BA who underwent LT between 1/2003 and 6/2011 were identified from the UNOS Standard Transplant Analysis. Results: One thousand five hundred and seventy seven children underwent FILT for BA;1291 (82%) underwent early transplant (ELT) and 286 (1 8%) were transplanted late (LLT).Clinical data at the time of transplant is shown in Table 1 .One and five year patient survival of the ELT and LLT patients was 95.1%, 93.9% and 99.0 %, 98.3% respectively (P value <0.01 for both periods).One and five year graft survival of the ELT and LLT patients was 88.0%, 85.2 % and 94.4%, 92.3% respectively. (P value <0.01 for both periods).Conclusion:The vast majority of transplants for BA occur in children less than 2 years of age. Outcomes in this group are worse than older children with BA highlighting the need for methods to improve bile drainage after the Kasai hepatoportoenterostomy.

Recipient Characteristics*ELT (Ages 0–2)LLT (Ages 3–18)P value
  1. *Mean (SD) or %

Age (years)0.43 (0.62)9.01 (4.43)<0.01
Ethnicity (%) White Hispanic Black Other46.2 22.8 19.5 11.556.3 14.0 17.5 12.2<0.01
BMI (z-score)-0.0% (0.93)0.43(1.17)<0.01
Albumin Gr/L2.97 (0.73)3.19(0.72)<0.0l
Creatinine mg/dL0.28(0.21)0.54 (0.65)<0.01
Bilirubin mg/dL12.97(10.12)6.83 (8.68)<0.01
Transplant Characteristics   
PELD/MELD (median/range)22 (-11–66)15 (-9–50)<0.01
Procedure type (%) Deceased-Whole Deceased-Variant Living47.0 34.9 18.171.7 18.2 10.1<0.0l
Time between listing and transplant (days)115.9(134.6)455.4 (748.8)<0.01


Benjamin L. Shneider - Consulting: Bristol Myers Squibb, Vertex; Stock Shareholder: Bristol Myers Squibb

The following people have nothing to disclose: Ronen Arnon, Rachel A. Annunziato, Guiseppe D'Amelio, Jaime Chu


The Developmental Origins of Biliary Atresia

Richard Kellermayer1, Dorottya Nagy-Szakal1, Ronald A. Harris2, Tamara N. Perreira3, Peter J. Lewindon3,5, Grant A. Ramm3,4, Ross Shepherd1;

1Section of Pediatric Gastroenterology Hepatology and Nutrition, Baylor College of Medicine, Houston, TX; 2Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX; 3Hepatic Fibrosis Group, The Queensland Institute of Medical Research, Brisbane, QLD, Australia; 4Faculty of Health Sciences, The University of Queensland, Brisbane, QLD, Australia; 5Department of Gastroenterology, Royal Children's Hospital, Brisbane, Brisbane, QLD, Australia

Background: Biliary atresia (BA) is the most common cause of cirrhosis and the single most frequent indication for liver transplantation in infants and children. Nevertheless, the cause of BA is yet unknown. Our current understanding is that ancestry and environment collectively determine predisposition to BA. However, neither the critical environmental factors, nor the biological system that those may alter to induce BA are identified. Epigenetic changes are environmentally responsive molecular mechanisms, which can modify gene expression independently from the genetic code. Among the epigenetic processes only DNA methylation has been shown to be stably transmitted through repetitive cell divisions, thereby having the capacity to permanently convey epigenetic information during the lifetime of an individual. Additionally, only DNA methylation has been described to directly communicate environmental exposures to phenotypic outcome in mammals. In the meantime, high-throughput locus specific assessment of hepatic DNA methylation changes in association with BA has not been performed. Therefore, we hypothesized that environmentally induced hepatic gene specific DNA methylation changes contribute to BA development. Materials/Methods: Genome wide DNA methylation studies on liver samples from 6 non-syndromic early (less than 4 months of age) BA patients (to limit fibrosis related non-specific DNA methylation changes as much as possible) and 12 non-BA controls were performed with Illumina 450K Infinium Methylation BeadChip Kits. Validation of the arrays was done by bisulfite pyrosequencing at select loci. Results: Bisulfite pyrosequencing at select loci validated the arrays with significant correlation (r=0.94, p<0.001). A highly conservative selection resulted in the delineation of 45 genes where DNA methylation significantly associated with BA. Ontology analysis of this select group revealed an over-representation (p<0.001) of genes involved in response to biotic stimulus, positive regulation of developmental processes, regulation of immune effector processes, and cartilage development in connection with BA. Conclusions: Prenatal and/or perinatal hepatic DNA methylation changes may play a role in the developmental origins of BA.


Peter J. Lewindon -Advisory Committees or Review Panels: Janssen; Speaking and Teaching: Janssen, Abbott

The following people have nothing to disclose: Richard Kellermayer, Dorottya Nagy-Szakal, Ronald A. Harris, Tamara N. Perreira, Grant A. Ramm, Ross Shepherd


fibroblast growth factor 19 was aberrantly expressed in cholestatic hepatocytes, and its signal pathways were inappropriately down-regulated in biliary atresia children

Yasuhiro Hasegawa1, Hiroki Kondou1, Masanobu Kawai2, Take-hisa Ueno3, Yuki Miyahara1, Akiko Konishi1, Kie Nakao1, Takeshi Kimura1, Kayo Ikeda1, Makiko Tachibana1, Kazuhiko Bessho1, Yoko Miyoshi1, Toshimi Michigami2, Keiichi Ozono1;

1Department of Pediatrics, Graduate School of Medicine, Osaka University, Suita, Japan; 2Department of Bone and Mineral Research, Research Institute, Osaka Medical Center for Maternal and Child Health, Izumi, Japan; 3Department of Pediatric Surgery, Graduate School of Medicine, Osaka University, Suita, Japan

Backgrounds: Bile acid synthesis is regulated by negative feedback mechanism, in which Fibroblast growth factor (FGF) 19 induced by Farnesoid X receptor (FXR) in intestinal epithelium is transported through portal circulation to liver and binds to FGF receptor 4 (FGFR4) and its co-receptor β-klotho (KLB) on hepa-tocyte membrane, that results in suppression of cholesterol 7a-hydroxylase (CYP7A1) transcription via ERK and JNK pathway. However, it is not clear how these signaling pathways function in chronic cholestasis. Here we study the FXR/FGF1 9 signaling pathway in chronic cholestasis using serum and liver tissue from biliary atresia (BA) patients. Methods: 9 BA and 4 control (hepatoblastoma, hemanigoma, urea cycle disorder) patients were studied. Serum and tissue content of FGF19 were measured by ELISA. Quantification of FXR, FGF1 9, FGFR4, KLB, and CYP7A1 mRNA were achieved by q-RT PCR using total RNA extracted from hepatocytes, which were obtained using laser micro-dissection system. Localization of FGF19 mRNA was analyzed by in situ hybridization (ISH). Phosphorylation of ERK and JNK was studied by Western blotting (WB). Permission to perform this study was given by the Ethics Committee in our institute, and written informed consent was obtained from parents of all of the patients. Results: Serum concentration of bile acids was significantly higher in BA patients than in control. The protein level of FGF19 in both serum and liver tissue were significantly higher in BA patients than in control. Expression of FGF19 mRNA was significantly higher in BA hepatocytes. Moreover, ISH revealed that FGF19 were aberrantly synthesized in hepatocytes in BA liver. However, CYP7A1 mRNA, a representative target gene of FGF19, was not suppressed despite the high concentration of bile acids and FGF19 in BA hepatocytes. Next we examined the signaling molecules upstream of CYP7A1. FXR mRNA was significantly up-regulated in hepatocytes from BA patients, and so were FGFR4 and KLB. Furthermore, phosphorylation of FGFR4 in BA hepatocytes was increased above control. On the other hand, phosphorylation of both ERK and JNK was decreased in BA hepatocytes. Conclusion: FGF19 was aberrantly expressed in hepatocytes during chronic cholestasis, while FGF19 expression was not detected in normal hepatocytes. Although FGF19 production in ileum was not examined, FGF19 which activates FGFR4/KLB complex on hepatocytes was at least in part might derive from hepatocytes in BA liver. However, this signal was not fully activated to suppress CYP7A1 via ERK and JNK pathway. This might be one of the mechanisms which disturbs the regulation of bile acid synthesis in BA patients.


The following people have nothing to disclose: Yasuhiro Hasegawa, Hiroki Kondou, Masanobu Kawai, Takehisa Ueno, Yuki Miyahara, Akiko Konishi, Kie Nakao, Takeshi Kimura, Kayo Ikeda, Makiko Tachibana, Kazuhiko Bessho, Yoko Miyoshi, Toshimi Michigami, Keiichi Ozono


The expansion of PROMININ-1 -positive epithelial-mes-enchymal cells within periportal fibrosis of rotavirus-induced biliary atresia

Nirmala Mavila1, David James1, Pranavkumar Shivakumar2, Sarah Utley1, Katrina Mak1, Allison Wu1, Megan Groff1, Kasper S. Wang1;

1Children's Hospital Los angeles, Los angeles, CA; 2Division of Pediatric Gastroenterology, Hepatology and Nutrition, Cincinnati Children's Hospital Medical Center, Cincinnati, OH

Introduction: Most infants with biliary atresia (BA) experience progression toward cirrhosis and end-stage liver failure even after successful surgical drainage. In BA, ductular reactions comprised of hepatic progenitor cells (Stamp et al, Stem Cells 2012) expand in proximity to evolving intrahepatic fibrosis. Rountree et al (Stem Cells 2007) demonstrated proliferation of cells expressing putative stem/progenitor marker PROMININ-1 (PROM1) in a murine model of cholestasis/fibrosis with ductular reaction. However, the role of these progenitor cells remains unknown. Hypothesis: Expansion and mesenchymal differentiation of PROM1 positive (pos) cells are associated with periportal fibrosis in BA. Methods: Newborn BALB/c mouse pups were injected intraperitoneally with 1.5 x 106 FFU Rhesus rotavirus (RRV) or saline. Livers were collected at 7d and 14d post-injection (n=4–6). Human liver samples were collected with institutional IRB approval. Liver samples were subsequently analyzed by Fluorescence Activated Cell sorting (FACS), immuno-fluorescence (IF), Western blot, quantitative real time PCR (qPCR), RNA microarray, and Sirius Red staining. Results: FACS analysis demonstrated a 3-fold increase in PROM1 pos cells in 2-week RRV livers compared to saline controls (p<0.005). IF analysis demonstrated the emergence of PROM1 pos cells within the regions of periportal fibrosis in RRV livers. PROM1 pos cells are non-hepatocytes, subsets of which co-express epithelial (E-CADHERIN and CYTOKERATIN-19 (CK19)) and mesenchymal (VIMENTIN and α-SMOOTH MUSCLE ACTIN (αSMA)) markers. PROM1 pos cells also co-express COLLAGEN-1α. A subset of PROM1pos cells manifested Fibroblast Growth Factor signaling activation via AKT-depend-ent β-CATENIN activation in association with cell proliferation. Expression of mesenchymal markers (N-CADHERIN, αSMA, VIMENTIN and SLUG were significantly upregulated in RRV livers compared to control (p<0.05) while expression of biliary marker CK1 9 was not, suggesting a potential mesenchymal differentiation of PROM1pos cells. Activation of Transforming Growth Factor-β (TGFβ) signaling, a known profibrogenic pathway, was observed in subset of PROM1 pos cells with co-expression of phospho-SMAD3. Human BA livers exhibited similar periportal expansion of PROM1pos cells with activated SMAD3 as well as significant increased expression of PROM1 and progenitor cell gene CD49F compared to normal livers by qPCR and microarray analysis (p<0.05 and p<0.005, respectively). Conclusion: RRV-induced experimental BA is associated with the expansion of PROM1 pos dual epithelial-mesenchymal cells with activated TGFβ signaling within regions of periportal fibrosis.


The following people have nothing to disclose: Nirmala Mavila, David James, Pranavkumar Shivakumar, Sarah Utley, Katrina Mak, Allison Wu, Megan Groff, Kasper S. Wang


Differences in CNS metabolism between pups and adults in an animal model of biliary cirrhosis: longitudinal, non-invasive in vivo study

Cristina Cudalbu1, Olivier Braissant2, Valerie A. McLin3;

1CIBM, EPFL, Lausanne, Switzerland; 2Service of Biomedicine, University Hospital of Lausanne, Lausanne, Switzerland; 3Swiss Center for Liver Disease in Children, Department of Pediatrics, University Hospitals Geneva, Geneva, Switzerland

Chronic liver disease (CLD) affects both adults and children and is often associated with some degree of hepatic encephalopa-thy (HE). In childhood, acute hyperammonemia (HA) is associated with brain edema and leads to irreversible damage of the developing central nervous system (CNS). Long-term deficits of such a magnitude have not been described in adults with HA. Although cognitive deficits exist in children with CLD, the underlying mechanism is unclear. How the developing brain responds to the metabolic changes of CLD, and how these mechanisms differ from those in adult patients are two unknowns. We hypothesized that blood-brain-barrier permeability and energy metabolism may be different in adults and pups. Wistar adult and pup (21 days) rats underwent bile duct ligation (BDL) and were scanned before BDL and weekly thereafter. In vivo localized spectroscopy was performed in the hippocampus (TE=2.8ms) on a 9.4T system. The ADC-apparent diffusion coefficient was derived from the tensor and measured in cortex, striatum and hippocampus. Immunohistochemistry on brain tissue was performed using astrocytic and water channel markers (GFAP, AQP4). Following BDL, brain metabolism differed in pups and adults (Table 1). Changes in pups were more significant for all parameters studied. First, pups displayed a more pronounced increase of brain Gln which was associated with marked edema in spite of ongoing osmoregulation (decreased Ins). Enhanced expression of AQP4 in brain micro-capillaries suggests that the underlying mechanism for higher ADC values in pups (brain edema) may be related to greater blood-brain barrier (BBB) permeability. Second, pups showed a greater decrease in brain neurotransmitters and antioxidants coupled with an increase in energy metabolites. The underlying mechanistic hypothesis here is that NH4+ may have a more pronounced effect on the mitochondrial permeability transition during brain development than in adults. We conclude that osmotic and metabolic changes are greater in pups than adults, possibly owing to increased BBB permeability. How these two processes are linked and contribute to edema and neuro-cog-nitive changes remains to be determined.


The following people have nothing to disclose: Cristina Cudalbu, Olivier Brais-sant, Valerie A. McLin


Growing up with biliary atresia without liver transplantation; a single centre experience

Vandana Jain2, Vinod Kolimarala2, Mark Davenport2, Nigel Heaton1, Marianne Samyn2;

1 Institute of Liver Studies, KCH, London, United Kingdom; 2Paediatric Hepatology, Gastroenterogy and Nutrition centre, KCH, London, United Kingdom

Aims: To evaluate outcomes in adulthood of biliary atresia not requiring liver transplantation in the paediatric age group. Methods: A single centre retrospective analysis, comprising 268 patients who underwent KP between 1 980–96. Results: After KP, 1 07/268 (40%) survived with their native liver for at least 1 6 years. Ninety-seven (43M:54F) are followed up in our centre, and provide our study group; 1 0 were referred to other adult liver centres. Out of the remaining (n=1 61); 123 (76%) underwent liver transplantation (LT) <16yrs, 1 8 (1 1%) died and 20 (1 3%) were lost to follow up. At a median age of 20.6 yrs (range 16–32) at last follow-up, 81 (84%) remain with their native liver (Group 1) whilst 16 had LT or are currently listed (Group 2). 2 patients died, one post LT. Characteristics of Group 1 are listed in Table 1. 7% of patients had at least 1 episode of cholangitis, and in 9%, varices were seen at endoscopy. Heterogeneity of liver parenchyma on US was seen in 78% and present in all with SBR > 20 umol/l. In this group one patient died during pregnancy. In Group 2, 14/16 patients underwent LT and 2 are listed for LT. Documentation for 15 patients was available. Median age at LT was 18.8 yrs (range 16.5–27.1) after median waiting time for LT of 11.5 mths (range 2–63). Median SBR and albumin at time of listing were 137 umol/l and 32 mg/dl respectively. Indications for LT were recurrent cholangitis with synthetic failure (n=7), synthetic failure (n=4), cholangitis (n=2), hepatocellular carcinoma (n=1) and portal hypertension (n=1). The listed patients have been on the waiting list for 27 and 35 mths respectively. Model for End Stage liver disease (MELD) median score at listing was 17 (range: 1 1–31) and worsened in 5 patients whilst on the waiting list. Two patients required re-transplantation for chronic rejection, of which one died following CMV infection. Overall, four patients had 6 successful pregnancies of whom 1 is currently listed for LT and 1 is pregnant. One patient, with SBR 29 umol/l and portal hypertension died during pregnancy from a variceal bleed. Conclusion: In our patient cohort, 84% are alive with their native liver, with 69% of patients demonstrating normal SBR levels and 26% showing no evidence of progressive liver disease. The most common indications for LT in adulthood, are recurrent cholangitis and/or synthetic dysfunction.

Native liver (n=81)SBR <20 umol/l (n=55)SBR >20 umol/l (n=26)
AST (<50 IU/1)4614
Platelet count (>100)468
Albumin (>35 mg/dl) and INR <l .25319
No splenomegaly on US242
All of the above211


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

The following people have nothing to disclose: Vandana Jain, Vinod Kolimarala, Mark Davenport, Marianne Samyn


Effects of Histone De-Acetylation Inhibitors on biliary formation in zebrafish

Vivian Tang;

Children's Hospital of Philadelphia, Philadelphia, PA

Background: Biliary atresia (BA) is a progressive fibro-inflam-matory cholangiopathy affecting the intrahepatic and extra-hepatic bile ducts of neonates. Our previous studies have shown that decreased DNA methylation leads to biliary defects in zebrafish and is associated with biliary atresia (BA). Histone deactylease 1 (HDAC) 1 regulates histone acetylation to coordinate gene transcription in zebrafish embryos and is involved in liver organogenesis. Histone deactylase inhibitors (HDACi) are novel treatments for patients with neurologic and oncologic disorders, as they induce cell growth arrest and induce apop-totic cell death. These compounds lead to hyperacetylation of histones and increased gene expression. Purpose: As there is a robust overlap between DNA methylation and histone acetylation (HDAC), we are examining the potential role of histone acetylation in biliary development and disease pathogenesis. Methods: To determine if treatment of HDACi leads to biliary defects, we performed dose dependent treatments on zebrafish embryos with specific HDACi (Trichostatin A (TSA), APHA, AN-9, SAHA, VAHA). We then examined their biliary function using the lipid reporter PED6, and biliary development using cytokeratin immunostaining. Using qPCR, transcription factors specific for biliary targets were tested. Results: We found a moderate effect of trichostatin A (TSA) and APHA, with mild to little effects of the other compounds, on biliary function and development. We found significant up-regulation of Hedgehog target genes such as gli2a, ptch1, foxl1, znf697, ccnd1 in TSA-treated fish. Of note, patients with BA demonstrate increased Hedgehog activity, and we have found increased Hedgehog activity in several other zebrafish models of abnormal biliary development. TSA and APHA both inhibit class I, and thus our results support an importance of class I HDACs in mediating biliary defects. Conclusions: HDACi such as TSA affect biliary development, possibly through increased hedgehog signaling.


The following people have nothing to disclose: Vivian Tang


Determinants of cost in biliary atresia

Douglas Mogul, Mo Zhou, Paul Intihar, Kathleen B. Schwarz, Kevin Frick;

Johns Hopkins, Baltimore, MD

Background: Biliary atresia (BA) is the most common cause of neonatal cholestasis and the most common indication for pedi-atric liver transplantation in the U.S. However, despite the significant public health resources devoted to care of these patients, little is known regarding its financial cost. The purpose of this study is to evaluate the costs associated with BA in a large university setting and to determine which factors influence these costs. Methods: Patients diagnosed with BA at the Johns Hopkins Hospital (JHH) from January 1 993 to September 2012, or who transferred their care to JHH within four months of diagnosis, were included in the analysis. Patient demographic, BA diagnosis, and medical charge information were obtained from the JHH DataMart. Total charges at 1, 3, and 5 years after diagnosis were estimated separately to control for bias in follow-up time. The impacts of liver transplant and the type of BA (syndromic versus non-syndromic) on charges were examined using a t-test. Linear regression was used to estimate the correlation between age of diagnosis and charges. Charges were adjusted to 2012 U.S. dollars using annual medical consumer price index. Results: Seventy-three individuals with a minimum of 1 -year follow-up were included in the analysis; 3-year follow up data were available for 51 patients, and 37 patients were followed for 5 years. The 1-year, 3-year and 5-year treatment costs after BA diagnosis were $120,822, $254,148, and $289,376, respectively, and >93% of total costs came from inpatient services. The average total (inpatient + outpatient) charges for patients with a liver transplant (n = 42) was $55,673 per year as compared to $ 1 8,985 per year for non-transplanted individuals (n = 31) and this difference was statistically significant (P < 0.05). The costs of liver failure, represented by the 15-month transplant-related charges (3 months prior to transplant and 12 months after transplant), averaged $307,776, and were significantly more than the $4,031 in charges accrued over a comparable interval in age-matched, non-transplanted controls (P = 0.0006). There was no difference in charges between syndromic (19%) and non-syndromic BA (P > 0.05). Age of diagnosis was also significantly correlated with inpatient charges. For every day earlier that BA was diagnosed, there was a $1,390 reduction in 1-year inpatient charges (P = 0.01 1) and a $2,007 reduction in 3-year inpatient charges (P = 0.031). Conclusion: Liver transplantation and delayed diagnosis are significantly associated with increased costs in BA. Attempts to minimize liver transplantation and/or identify cases earlier may correlate with cost savings.


Kathleen B. Schwarz - Consulting: Novartis, Novartis; Grant/Research Support: Bristol-Myers Squibb, Gilead, Roche/Genentech, Bristol-Myers Squibb, Vertex, Roche

The following people have nothing to disclose: Douglas Mogul, Mo Zhou, Paul Intihar, Kevin Frick


Biliary diversion in children with intractable pruritus - A single centre experience

Palaniswamy Karthikeyan1, Mark Davenport2, A. S. Knisely3, Richard J. Thompson3, Sanjay Bansal1;

1Paediatric Hepatology, Gastroenterology and Nutrition, King's College Hospital, London, United Kingdom; 2Paediatric Surgery, King's College Hospital, London, United Kingdom; 3Institue of Liver Studies, King's College Hospital, London, United Kingdom

Management of children with intractable pruritus is a major challenge. Pharmacological therapy does not always achieve an adequate response. For this reason biliary diversion and, if unsuccessful, liver transplantation (LT) have been used. Aim: To assess the outcome of biliary diversion surgery in children with intractable pruritus due to cholestatic liver disease. Methods: Retrospective analysis of patients with intractable pruritus who had biliary diversion in a single centre. Results: 13 children (M:F 8:5) had biliary diversion (12 partial external, 1 internal). Median age at surgery was 2 yrs (range 0.25–10 yrs) with median follow-up of 1.87 yrs (range 0.58–1 0 yrs). Children fell into Group A (4 with improvement in pruritus) or Group B (9 with no improvement). Median ages at surgery were 1.88 yrs (range 1.08–10 yrs) and 2 yrs (range 0.25–3.7 yrs) respectively. Median total serum bile acid levels before and 6 months after surgery were respectively 90 μmol/L (range 78–158) and 20 μmol/L (5–153) in Group A. They were 215 μmol/L (54–390) and 169 μmol/L (47–252) for Group B. 5/9 in Group B had LT (1 died shortly thereafter) after a median period of 1.6 yrs(0.5–6.25 yrs). Three others were listed for LT, of whom 1 died whilst waiting. The remaining patient underwent surgery only 9 months ago, but exhibits little improvement. No FIC1 deficiency or Alagille syndrome patient had bridging fibrosis, or worse, at diversion. However, in Group A patients with BSEP deficiency or “other cholestatic diseases” fibrosis was minimal or absent. By contrast, all 5 Group B patients with these disorders had bridging fibrosis or cirrhosis at diversion. Conclusion: Biliary diversion can be effective in the management of intractable pruritus in some patients. In BSEP deficiency and “other” cholestasis, bridging fibrosis presaged poor outcome. These findings support the impression that with biliary diversion in intrahepatic cholestasis, the earlier performed the better.

GroupFIC 1 deficiencyBSEP deficiencyAlagille syndromeOther
  1. * Further management of Group B

  2. ** 1 death post LT and 1 death while on list

* LT/Listed/Undecided2**/l**/03/0/00/1/00/1/1


The following people have nothing to disclose: Palaniswamy Karthikeyan, Mark Davenport, A. S. Knisely, Richard J. Thompson, Sanjay Bansal


Sustained immunosuppression-free remission in pediatric autoimmune hepatitis: a population-based study

Mark Deneau, Linda Book, Stephen Guthery, M. K. Jensen;

Pedi-atric Gastroenterology, University of Utah, Salt Lake City, UT

Introduction: Limited data exist on outcomes in pediatric autoimmune hepatitis (AIH), particularly sustained immunosuppression-free remission (SIFR) after withdrawal of all immunosuppressive medications. Methods: We retrospectively reviewed all pediatric AIH patients in the region from 1986–201 1 using population-based methodology and followed them until biochemical remission, and/or SIFR. We defined SIFR as the presence of all of the following: 1. biochemical remission on immunosuppres-sion for at least one year, and 2. liver biopsy demonstrating no active inflammation, then 3. withdrawal of all immunosuppressive medications, then 4. at least one year with normal liver inflammatory markers and no clinical symptoms, and 5. no relapse of any kind through the end of follow-up. Results: We identified 56 AIH-patients [62.5% female, mean age 10.1 years (range:0.7–1 7.6)] followed for a median of 5.6 years [IQR: 2.8–8.6]. Cirrhosis was present in 14.0% and primary sclerosing cholangitis in 21.4%. Coexisting non-hepatic immune-mediated diseases occurred in 37.5%, with inflammatory bowel disease (19.6%), celiac disease (14.3%), and autoimmune thyroid disease (13.5%) most commonly seen. Screening of AIH patients for celiac and thyroid disease occurred in 66 and 50% of all cases, respectively. Biochemical remission on immunosuppression was achieved in 76.4% of all AIH patients after a median of 1.2 years [IQR:0.4–3.6]; 76.9% of these patients sustained biochemical remission through the end of follow-up. Elevated INR, positive anti-neutrophil cyto-plasmic antibody (ANCA) titer and/or cirrhosis on initial biopsy were predictors of failure to achieve biochemical remission on immunosuppression. Complete immunosuppression withdrawal was attempted in 16 patients and successful in 87.5% (14/16) and unsuccessful in 12.5% (2/16). Among all AIH patients in the cohort, the probability of achieving SIFR within five years after AIH diagnosis was 41.6% (95%CI:25.3–62.9). The presence of a non-hepatic autoimmune diagnosis and/or cirrhosis on initial biopsy were predictors of not achieving SIFR. Conclusions: Elevated INR, positive ANCA titer, cirrhosis on initial biopsy, or the presence of a non-hepatic autoimmune disease are patient characteristics associated with worse outcomes. The prevalence of celiac and thyroid diseases in pediatric AIH patients is high, but screening appears subop-timal. Universal screening of AIH patients for celiac and thyroid diseases may improve patient care. SIFR is an achievable goal for carefully selected AIH patients, particularly those in stable biochemical remission on immunosuppression.


The following people have nothing to disclose: Mark Deneau, Linda Book, Stephen Guthery, M. K. Jensen


Impact of IL28B polymorphism on Early virological response (EVR) to Pegylated interferon (peg IFN)/ribavirin therapy in Egyptian children with HCV genotype IV

Ahmed F. Abdalla1, Abeer Fathy1, Ahmed Megahed1, Tarik Barakat1, Mona A. Alsayed1, Wagdi Elkashef2, Khaled R. Zalata2;

1 Pediatrics, Hepatology abd GIT unit, Faculty of Medicine, Man-soura University, Mansoura, Egypt, Mansoura, Egypt; 2Pathology, Faculty of Medicine, Mansoura University, Mansoura, Egypt, Mansoura, Egypt

Background: IL28B polymorphism, reflection upon response to interferon therapy has been studied in adults and recently in genotype IV. For children, only few reports are available in literature. In current work we studied the reflection of IL28B polymorphism on EVR to Peg IFN plus ribavirin therapy in our ongoing treatement trial for children with chronic HCV infection. Methods: The study included naive 55 children with chronic HCV genotype IV, 33 male and 22 females with age ranged from 4 to 17years ( mean 1 1.20 ± 3.64). Liver biopsy was performed for all patients as a prerequisite. Basal HCV RNA viremia, and all liver function tests were done. In addition we evaluated IL 28B polymorphism at position - 3 1 76 C/T (rs 12979860) in all of them. HCV PCR was performed after initial 12 weeks of therapy as indicator for EVR. Results: Out of the studied 55 children, 37 (67.3%) had basal mild viremia, and 1 8 (32.7%) were of moderate viremia. Forty seven out of the 55 children were F0/F1; only two children had advanced fibrosis (F3/F4). Early virological response (EVR) was achieved in 40 children (72.7%) . Basal HCV RNA was significantly lower among children with EVR. Thirty one (77.5%) of responders were of mild viremia, on the other hand 6 (40%) of non-respon-ders were of mild viremia (p = 0.008). Regarding IL 28B polymorphism, 19 were CC, 23 CT and 13 TT. Eighteen (45%) of responders were CC, 15 (37.5%) CT and 7 (17.5%) TT. For non-responders, only one child (6.7%) was CC, 8 (53.3%) were CT and 6 (40%) TT. IL 28B polymorphism is significantly correlated with EVR (P= 0.022). Age, gender, fibrosis stage, or necroinflammation were of no significant differences among responders and none responders. Conclusion: Early virological response to peg IFN/Ribavirin therapy in children with HCV genotype IV is significantly correlated with basal HCV RNA viremia as well IL28B polymorphism. Testing for IL28B polymorphism may be considered as a prerequisite to peg IFN/Ribavirin therapy in children with HCV genotype IV.


The following people have nothing to disclose: Ahmed F. Abdalla, Abeer Fathy, Ahmed Megahed, Tarik Barakat, Mona A. Alsayed, Wagdi Elkashef, Khaled R. Zalata


Hepcidin and Iron Overload in HCV-infected Children with Acute Leukemia

Manal H. El-Sayed1, Randa M. Matter1, Amira N. Abdel-Gawad1, Manal M. Abdel-Aziz2, Gamal E. Esmat3, Galila M. Mokhtar1;

1 Department of Pediatrics, Ain Shams University, Cairo, Egypt; 2Department of Clinical Pathology, Ain Shams University, Cairo, Egypt; 3Department of Tropical Medicine, Cairo University, Cairo, Egypt

Background/Objective: Hepcidin plays a pivotal role in the pathogenesis of iron overload detected in children after therapy for acute lymphoblastic leukemia (ALL) and in patients with chronic HCV infection. This study evaluates iron overload and secondary hepatic siderosis in children treated for ALL with or without HCV infection and assesses whether serum hepcidin is useful in detecting iron load. Methods: Eighty patients with ALL and 20 healthy children and adolescents were enrolled in this study. Patients with relapsed ALL, decompensated liver disease or clinically proven acute infections were excluded from enrolment. Clinical examination and the number of hepatitis flares (transaminases>3XUNL) starting at diagnosis of ALL till sampling were recorded. Patients were divided into 4 groups (20 patients each): Children in the maintenance phase of chemotherapy without HCV (group A) or with HCV infection (group B) (mean age: 8.6±3.3 and 9.3±3.8 years respectively) and ALL survivors without HCV (group C) or with HCV infection (group D) (mean age: 11.0±3.0 and 9.0±2.6 years respectively). Iron status was assessed by serum iron, ferritin, total iron binding capacity (TIBC) and serum hepcidin. In subjects with iron overlaod, hepatic siderosis was evaluated byT2*MRI and liver histology. Liver fibrosis was also evaluated by fibroscan. Results: Serum hepcidin was higher among patients than controls (median: 38.5 ng/ml and 7.75 ng/ml respectively; p<0.01) and lower among HCV-infected patients compared to other groups, though not significant. Hepcidin correlated positively to serum ferritin (r=0.53, p>0.001), iron (r=0.45, p<0.001)and negatively to TIBC (r=-0.4, p<0.001). The stage of liver fibrosis as measured by fibroscan correlated to the frequency of hepatitis flares and serum ferritin (r=0.32, p=0.007; r=0.77, p<0.001 respectively). Ferritin was a predictor of hepatic iron content (correlated negatively to T2* MRI and positively to histological hepatic siderosis). Conclusion: Hepcidin and ferritin are important measurements for assessment of iron homeostasis in patients with ALL especially those with HCV infection. T2* MRI and fibroscan are reliable noninvasive methods for evaluation and follow-up of hepatic siderosis and fibrosis.


Gamal E. Esmat - Advisory Committees or Review Panels: MSD &BMS companies, MSD &BMS companies; Speaking and Teaching: Roche & GSK companies, Roche & GSK companies

The following people have nothing to disclose: Manal H. El-Sayed, Randa M. Matter, Amira N. Abdel-Gawad, Manal M. Abdel-Aziz, Galila M. Mokhtar


Outcome of hepatitis B and D (delta) virus infection in children undergoing chemotherapy for haematological malignancies: an eight-year single centre experience

Harshad Devarbhavi1, Keyur A. Sheth1, Karnam Ravikiran1, Mallikarjun Patil1, Venu H. Aradya1, T. R. Vijaykumar1, Adarsh Ck1, Sethumadhavan Muralidharan2, Ranjani Shamsundar2;

Department of Gastroenterology, St. John's Medical College Hospital, Bangalore, India; 2Department of Microbiology, St. John's Medical College Hospital, Bangalore, India

Aims: Hepatitis B reactivation has been studied in adults. The natural history, outcome and the optimal management of hepatitis B (HB) and hepatitis delta (HD) infection in children receiving chemotherapy is lacking. We report our experience with HB and HD in such patients. Material and methods: The study cohort consists of referred children with hepatitis that developed while receiving chemotherapy for haematological malignancies from 2005 to 2012. Investigations for acute hepatitis A, B, C, D and E by ELISA tests including HBeAg and anti-HBe and HBV DNA by PCR were carried out. The peak biochemical variables and other characteristics are depicted in Table. Patients received Lamivudine ± Adefovir. Results: 46 HBsAg negative children developed acute hepatitis B (n=15) and hepatitis D (n=31) during chemotherapy. The clinical, biochemical, virological characteristics are shown in Table 1. > 90% developed hepatitis during maintenance phase. Chemotherapy was temporarily stopped till resolution of jaundice or raised transam-inases. There were 3 deaths in the HDV group, two due to cholestatic hepatitis and another due super-added hepatitis A. Majority completed chemotherapy. Three developed leukemia recurrence while awaiting resolution of transaminase. Liver biopsy (N=4) in HD infection showed fibrosis. At follow up majority were stable with mild rise in transaminases. Surface antigen and e antigen seroconvertion was rare. Conclusion: HB and HD infection mostly manifests in maintenance phase of chemotherapy, is largely well tolerated, and results in treatment disruption. Chemotherapy can be resumed after normalization of liver tests under cover of oral antivirals in HBV and HDV. A risk of cholestasis, liver failure or decompensation is present in HDV but not in HBV. HBsAg and e Ag seroconvertion is rare in HBandHD.

VariablesHBV (N= 15) MeanHDV (N=31) Mean
Age (years)9.29.0
Sex (Males)926
Serum Proteins (g/dl)6.57.3
Serum Albumin (g/dl)3.84.1
Total Bilirubin (mg/dl)6.34.1
AST (U/L)1866.21504.3
ALT (U/L)1129.91418.5
ALP (U/L)680.4388.3
GGT (U/L)130.3103
Hemoglobin (g/dl)11.311.9
White Cell Count (103)6.710.5
Platelets (cells/dl)198000183958
HBV DNA (107IU/ml)1.63.3
Chemo - HBV/HDV detection (months)7.812.1
Follow up (months)33.630.9


The following people have nothing to disclose: Harshad Devarbhavi, Keyur A. Sheth, Karnam Ravikiran, Mallikarjun Patil, Venu H. Aradya, T. R. Vijaykumar, Adarsh Ck, Sethumadhavan Muralidharan, Ranjani Shamsundar


Pediatric NAFLD patients have a high prevalence of Hepatitis B non-immunity

Minesh Mehta2, Crystal Slaughter1, Stavra A. Xanthakos1, Rohit Kohli1;

1Department of Gastroenterology, Cincinnati Children's Hospital Medical Center, Mason, OH; 2Internal Medicine, University of Cincinnati, Cincinnati, OH

Most patients affected by nonalcoholic fatty liver disease (NAFLD) are obese. Hepatitis B vaccination is part of the universal vaccination schedule for children but obesity has been identified as a predictor of poor serologic antibody development after Hepatitis B vaccination. Therefore, the objective of this study was to determine the prevalence of immunity against hepatitis B in pediatric NAFLD patients. We hypothesized that hepatitis B surface antibody (HBs Ab) sero-prevalence would be low in children with NAFLD, despite universal immunization practices in place for hepatitis B. Methods: Clinical data was retrospectively analyzed from pediatric subjects with NAFLD, ages 6–1 8 years, who had been prospectively enrolled in an IRB-approved pediatric NAFLD registry. To be enrolled in the NAFLD Registry, all subjects had to have documentation of chronically elevated liver enzymes and negative exclusionary testing for other causes of steatohepatitis. The database was retrospectively reviewed for laboratory testing for the presence of HBs Ab; a marker for immunity and hepatitis B surface antigen (HBsAg); a marker for infection. Individuals with absence of HBs Ab in the setting of negative HBsAg were qualified as Hepatitis B non-immune. Results: 96 out of 200 subjects had no documented HBs Ab serology. Of the 104 subjects with documented HBs Ab status, only 29 were found to have positive HBs Ab serology. The remaining 75 subjects (72.1%) were Hepatitis B non-immune. No significant clinical differences were found between subjects with documented positive vs. negative HBs Ab (Table with Mean±SD). Conclusions: Our study identified a very high prevalence (72%) of hepatitis B non-immunity in pediatric patients with NAFLD, despite existing guidelines for universal pediatric vaccination for hepatitis B. This may be either from incomplete vaccination series or a diminished immunogenic response to hepatitis B vaccination in the setting of NAFLD. We propose that pediatric NAFLD patients should undergo comprehensive screening for hepatitis B immuno-genicity, in addition to screening for infection, and catch up or booster vaccinations should be administered to the non-immunized patients.

 Immune (29)Non-immune (75)p-value
Age (yr)13.34±3.913.37±3.4ns
BMI (kg/m2)34.71 ±7.135.31±6.5ns
AST (IU/L)58.83±28.755.05±33.4ns
ALT (IU/L)94.83±32.177155.9ns
Male % (n)51.7(15)62.6 (47)ns


Rohit Kohli - Grant/Research Support: Johnson and Johnson, Johnson and Johnson

The following people have nothing to disclose: Minesh Mehta, Crystal Slaughter, Stavra A. Xanthakos


Assessment and Promotion of Knowledge in Pediatric Liver Transplant Recipients using a Novel Transition Tool

Alexandra Tsouka1,2, Eric I. Benchimol3, Christiane Sokollik1,2, Yaron Avitzur1,2, Miriam Kaufman2,4, Vicky L. Ng1,2;

1 Division of Paediatric Gastroenterology, Hepatology and Nutrition, SickKids Transplant Center, The Hospital for Sick Children, University of Toronto, Toronto, ON, Canada; 2SickKids Transplant Centre, The Hospital for Sick Children, University of Toronto, Toronto, ON, Canada; 3Division of Paediatric Gastroenterology, Hepatology and Nutrition, Children's Hospital of Eastern Ontario, Ottawa, ON, Canada; 4Division of Adolescent Medicine, The Hospital for Sick Children, University of Toronto, Toronto, ON, Canada

Background: In the transition from paediatric to adult care, patients are expected to increase their level of self-care and knowledge. This study evaluated the knowledge of pediatric liver transplant (LTx) recipients in 4 major domains: primary disease, complications, medications and health services resources. Patients and Methods: Patient inclusion criteria were time since LTx of >6 mos AND patient age 8–1 8 yrs. Patient exclusion criteria included developmental delay precluding accurate completion of questionnaires. Patients and their parents completed a questionnaire called MyHealth Passport for Liver Transplantation as a cross-sectional assessment of knowledge. Responses were evaluated for accuracy by reviewing medical records. The proportion of correct responses between patients and parents was compared using Fisher's Exact test. Following correction of responses, each patient received his/her MyHealth Passport for use in emergency situations. Results: A total of 31 patients (median [range] age of 12.9 [8–18.3] yrs) and 29 parents participated. Median (interquartile range) time since LTx and patient age at time of LTx were 8.3 (5.8–13.1)yrsand 1.8 (0.7–8.3) yrs respectively. In all, 51.6% of patients and 89.7% of parents correctly named the primary liver condition. Most patients and parents accurately named the daily immunosuppression medication (87.1% vs. 93.1%, P=0.26), although the correct dose and frequency were known by only 51.6% and 41.9% of patients respectively. Parents tended to know patients' blood group type (62.1% vs 25.8% of pts, P=0.06). Regarding allograft re ection, compared to patients, parents tended to better recall if episodes had ever happened (89.7% vs 41.9% of pts, NS), total number of episodes (50% vs 8.3% of pts, NS) and date of the last episode (37.5% vs 8.3%, NS). Patients tended to be less aware than their parents of precautions following Varicella exposure (12.9% vs 51.7%, P=0.6). In all, 32.3% patients and 62.1% parents reported that they should not receive live vaccines following LTx. Details of health service resources such as names of insurance provider, family physician and pharmacy were known by 38.7%, 48.4%, and 67.7% patients. Conclusions: Patients tended to be less aware of important health issues compared with their parents. Significant gaps exist in the self-care and knowledge of pediatric LTx recipients. Future targeted educational interventions should address areas of weakness in adolescent knowledge, particularly immunosuppression medication details and health services resources.


The following people have nothing to disclose: Alexandra Tsouka, Eric I. Benchimol, Christiane Sokollik, Yaron Avitzur, Miriam Kaufman, Vicky L. Ng


Health Utilities in Pediatric Liver Transplant Recipients

Saeed Mohammad, Katie Neighbors, Estella M. Alonso; Pediatrics, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL

Health Utilities are preference based numeric measures of health states that are essential for assessing the comparative effectiveness of therapeutic strategies. We measured health utilities and health related quality of life (HRQOL) in pediatric liver transplant (LT) recipients. Methods: Patient-parent dyads completed surveys from which, indirect self-assessed and parent proxy health utilities were calculated. These included the Health Utility Index 2/3(HUI) for patients ≥12 yrs and the Child Health Utility Index 9D (CHU9D) for >7 yrs. Twenty-one adolescents ≥12 yrs assessed health utility directly using Standard Gamble and Time Trade Off techniques. All patients reported HRQOL using the PedsQL™ Generic Core Scale. Univariate analyses were performed to measure the association between demographic and medical variables and health utilities. Results: Sixty-six patient-parent dyads participated. The patient population was 58% female and 77% Caucasian. The mean age at survey was 13.1 ±3.3 yrs and mean time since transplant was 8.9±4.8 yrs. Forty-seven percent had biliary atresia and 56% had private insurance. In the last three years, 17 patients had late acute rejection, 6 were diagnosed with chronic rejection, 8 were treated for biliary obstruction, 3 were re-transplanted, and 1 was on transplant waiting list. Indirect (HUI, CHU9D) health utilities were significantly lower than published healthy controls (p<0.05). Significant correlation was noted between indirect health utilities and self-reported PedsQL™ Total scores (p<0.05). There was a highly significant correlation between Standard Gamble and Time Trade Off utilities (p<0.005), however these did not correlate with the indirect measures. Biliary obstruction was significantly associated with lower health utility on the self-assessed HUI2 (p<0.001) and HUI3 (p<0.005). Primary diagnosis, insurance status, acute rejection, age, gender, race and ethnicity did not affect health utility. Summary: In the first report of health utilities in pediatric LT recipients, we found health utilities are significantly lower than healthy peers. Standard Gamble and Time Trade Off utilities strongly correlate with each other indicating the feasibility of this assessment method in this population. The data support the prospect of using health utilities for decision analysis and comparative effectiveness trials in pediatric LT recipients.

Comparison to Normative Population

Health utility measureTransplant recipientsNormative populationp-value


The following people have nothing to disclose: Saeed Mohammad, Katie Neighbors, Estella M. Alonso


Long term outcome after liver transplant - a 15 year retrospective review

Nicola D. Ruth1,2, Maria Legarda1, Monica Smith1,3, Philipa J. Lewis1,3, Carla Lloyd1, Susan Paris3, Deirdre A. Kelly1,2;

1Hepatol-ogy, Birmingham Children's Hospital NHS Foundation Trust, Birmingham, United Kingdom; 2Institute of Biomedical Sciences, University of Birmingham, Birmingham, United Kingdom; 3Hepa-tology, Queen Elizabeth Hospital Birmingham, Birmingham, United Kingdom

Liver transplantation (LT) is a successful treatment for end stage liver disease. There are little data on the long term outcome of patients who underwent LT in childhood. Aim: To evaluate the long term survival and social adaptation of patients >15 years following LT. Methods: All patients who underwent LT >15 years ago were included. Medical notes were reviewed, patients interviewed and physicians contacted to complete data collection. Data included liver/renal function, hypertension, hyper-lipidemia, bone disease, recurrent disease, rejection, presence of graft/de Novo Autoimmune Hepatitis, psychological disorders, adherence, education/employment, relationships and socially accepted behaviours. Results: 181 patients underwent LT (1 985–1 995). 1 1 8 patients were alive >1 5 years post transplantation (median post-transplant survival 19.5 years, range 16–27 years). Median age at transplant was 25 months (range 1 5 days-1 6 years). Main indication was biliary atresia (51%). 23/116 patients were re-transplanted (the majority occurred in the first year post transplant). At >15 years post transplant, 89% had normal functioning grafts. 43% were on steroid immunosuppression and 54% on monotherapy (mycopheno-late/tacrolimus/cyclosporin). The main causes of graft dysfunction were chronic rejection (66%) and graft fibrosis (33%). Long term complications included renal dysfunction (22% had a calculated GFR <60ml/min/1 .73m2 and 3 patients required a renal transplant). 39% required antihypertensive therapy. 9% had had post transplant lymphoproliferative disorders/lym-phoma/leukaemia. 54% were compliant with treatment and appointments. 28% completed university; 43% went to college; 50% are currently employed, 29% are still at school. 21% are unemployed and not in education. 51% are married/in a relationship (1 3% have had children). 50% regularly drink alcohol, 23% smoke. Summary: The survival rate of LT patients >15 years post transplant is 64%, but most of the deaths were close to the transplant episode, reflecting early experience of transplantation. The majority of the survivors have normally functioning grafts, are in employment/education and are married/in a relationship. The main reasons for graft loss were chronic rejection or graft fibrosis. The complications of long term immunosuppression include hypertension, renal dysfunction and lymphoproliferative disorders. Non-adherence/compliance was present in 46% of the group. Conclusion: The long term outcome of liver transplantation is good, with patients having little or no graft dysfunction, allowing patients to have a normal social adaptation and functioning.


Deirdre A. Kelly - Consulting: Sanofi Pasteur; Grant/Research Support: BMS, Astellas, Acitllion, MSD, Roche

The following people have nothing to disclose: Nicola D. Ruth, Maria Legarda, Monica Smith, Philipa J. Lewis, Carla Lloyd, Susan Paris


Pediatric liver transplant for urea cycle disorders (UCD) and organic acidemias (OA): United Network for Organ Sharing (UNOS) data 2002–2012

Emily R. Perito1, Sue Rhee1, John P. Roberts2, Philip Rosenthal1,2;

1 Pediatric Hepatology/Gastroenterology, University of California San Francisco, San Francisco, CA; 2Surgery, University of California, San Francisco, San Francisco, CA

Background: Decision-making about liver transplant is unique in children with UCD & OA because they receive immediate high priority on the waiting list, not related to severity of their disease. There is limited national outcomes data on which to base recommendations about liver transplant for UCD/OA. Methods: Retrospective analysis of UNOS data on liver recipients <18yrs, transplanted 2002–12. Repeat & multi-organ transplants were excluded. Chi-squared & Kruskal-Wallis tests, Cox proportional-hazards, & logistic regression were used for analyses as appropriate. Results: 31 0 of 5751 (6%) of pediatric liver transplants were liver-only for UCD/OA. UCD/OA increased from 4.4% of transplants in 2002–05 to 7.4% in 2010–12 (p<0.001). 96% were deceased donor. In UCD/OA deceased donor recipients, 34% had transplant at age <1y & 26% at 1–2y. The proportion transplanted at <1y varied by region (p=0.03). Region 2 (21% at <1y) and Region 5 (46% at <1y) accounted for 48% of UCD/OA transplants but only 33% of all transplants. Median weight at transplant was 12.5kg (IQR 8.5–19.3kg, n=277) with less variation by region (p=0.058). 5-year graft survival was 78% for UCD/OA children <2y at transplant & 88% for those >2y (p=0.055). Vascular thrombosis caused 43% of graft losses; 70% occurred in children <2y at transplant. In multivariate analysis (n=260), age<2y, weight at transplant, split vs. whole liver, & cold ischemia time did not predict graft loss. Graft loss risk increased in females (HR 2.9, 1.4–6.0, p=0.003), Hispanics (HR 3.9, 1.8–8.6, p<0.001), Blacks (HR2.9, 1.1–8.3, p=0.04) &those hospitalized at transplant (HR 2.8, 1.1–7.1, p=0.03). 5-year patient survival was 88% in UCD/OA children <2y at transplant and 98% in those ≥2y (p=0.01). 75% of deaths occurred <6m post-transplant. In multivariate analysis (n=260), age<2y (HR 6.2, 1.3–30.8, p=0.02), split liver transplant (HR 3.5, 1.0–12.2, p=0.05) & Hispanic ethnicity (HR 3.9, 1.0–15.0, p=0.05) increased mortality risk. Data on cognitive & motor delay was only available from follow-up records. At last-follow up (4.2±2.8y), 43% of UCD/OA children transplanted at <2y & 41% at >2y had cognitive delay (p=0.8); Motor delay was reported in 35% transplanted at <2y & 24% at >2y (p=0.056). Cognitive & motor delay at last follow-up were not predicted by age or weight at transplant, gender, ethnicity, split vs. whole liver, or hospital-ization at transplant in uni- or multivariate analysis. Conclusion: Most liver transplants for UCD/OA occur in early childhood, but there is regional variation. Further research is needed on benefits of early transplant for UCD/OA, as younger age may increase post-transplant morbidity.


Philip Rosenthal - Advisory Committees or Review Panels: Ikaria, Gilead, Merck, General Electric; Consulting: Roche; Grant/Research Support: Roche, Bristol MyersSquibb, Gilead, Vertex

The following people have nothing to disclose: Emily R. Perito, Sue Rhee, John P. Roberts


Autoimmune hepatitis presenting as hyperacute liver failure with severe hepatic encephalopathy: rescue of the child by auxiliary liver transplantation and complete recovery of the native liver by combined immunosuppression

Giorgia Curia1, Enrico Gringeri2, Daniele Neri2, Giacomo Zanus2, Graziella Guariso1, Umberto Cillo2, Mara Cananzi1;

1 Unit of Gas-troenterology, Digestive Endoscopy, Hepatology and care of the Child with Liver Transplantation, University Hospital of Padova, Padova, Italy; 2Hepatobiliary and Liver Transplant Unit, University Hospital of Padova, Padova, Italy

Background. Acute liver failure (ALF) is a dramatic event. 5% of pediatric cases are determined by autoimmune hepatitis (AIH). When AIH does not respond to immunosuppressive treatment, liver transplantation remains the only therapeutic option. However, after transplantation AIH may recur in 20% of patients. Aim. To report the third pediatric case of ALF due to AIH successfully treated with auxiliary liver transplantation (ALT). Case Report and results. A 5-year-old girl was referred to our Center of Pediatric Hepatology for the sudden onset (<7 days) of jaundice and lethargy. Elevated liver enzymes (AST 999 U/L, and ALT 1583U/L), severe coagulopathy (international normalized ratio 3.3) and hyperammoniemia (158 umol/L) were present at admission. High titers of anti-nuclear and anti-smooth muscle antibodies were retrieved suggesting the presence of type 1 AIH. Despite a prompt steroid treatment, the girl showed a progressive neurological deterioration leading to hepatic coma in less than 24 hours after admission. Cerebral edema/hemorrhages were excluded by neurological imaging. 72 hours after admission the girl underwent a native left hepatectomy and received a left auxiliary graft implanted orthotopically. Histology of the explanted left lobe confirmed the diagnosis of AIH. Immunosuppression was based on tacrolimus, azathioprine and steroids. The girl showed a complete neurological recovery in the first week after ALT. Liver tests normalized 3 months after transplant. Liver biopsies and volumetric CT scans, performed at 1 and 6 months after ALT, respectively demonstrated a complete remission of AIH and a full volume recovery of the right native lobe. Steroid treatment was interrupted 6 months after ALT. Gradual withdrawal of tacrolimus was started 7 months after ALT. No significant complications have been observed at 1 0 months of follow up. Conclusions. ALT is an option for the treatment of ALF when the chances for the native liver to recovery are high but temporary support is critical for survival. Less than 50 cases of ALT have been reported in children with ALF so far; the main indications consisted of seronegative non-A-non-G hepatitis, drug toxicity and mushroom poisoning. We have described the third pediatric case of AIH treated by ALT with rapid restoration of liver function and reversal of hepatic encephalopathy. ALT should be always considered as a therapeutic option for AIH in the setting of a rapidly deteriorating liver function despite maximal medical treatment as, if the native liver recovers, it offers the advantage of avoiding lifelong immunosuppression.


Umberto Cillo - Grant/Research Support: Novartis, Bayer, Astellas

The following people have nothing to disclose: Giorgia Curia, Enrico Gringeri, Daniele Neri, Giacomo Zanus, Graziella Guariso, Mara Cananzi


Recurrence of Primary Sclerosing Cholangitis (rPSC) in Pediatric Liver Transplant (LTx) Recipients

Veena L. Venkat1, Sarangarajan Ranganathan2, George V. Mazariegos3, Rakesh Sindhi3;

1Division of Pediatric Gastroen-terology, Hepatology and Nutrition, Children's Hospital of Pittsburgh of UPMC, Pittsburgh, PA; 2Division of Pediatric Pathology, Children's Hospital of Pittsburgh of UPMC, Pittsburgh, PA; 3Hillman Center for Pediatric Transplantation, Children's Hospital of Pittsburgh of UPMC, Pittsburgh, PA

Purpose: There is little detailed clinical data on rPSC after LTx in children. Our purpose was to describe patient characteristics in children who experienced rPSC after LTx to identify potential risk factors for recurrence. Methods: After IRB approval, clinical information of pediatric patients transplanted for PSC between 1993 and 2012 was retrospectively reviewed, and variables related to the pre-transplant diagnosis of PSC and post-transplant variables were abstracted. Variables studied include CMV/EBV status, early/late rejection, induction regimen, immunosuppression in the first year, steroid-resistant rejection, diagnosis of inflammatory bowel disease (IBD), and HLA markers commonly associated with PSC. Pre and post-transplant histology from biopsies and explants was reviewed by a single pathologist who was unaware of the patients' clinical course. A diagnosis of rPSC was made based on radiographic features, histology or both. Patients were excluded from a diagnosis of rPSC if they had a history of hepatic artery thrombosis (HAT) or anastamotic stricture within 3 months after transplant. We report characteristics of patients who have undergone LTx for PSC and common factors observed in subjects with rPSC. Results: Twelve patients underwent LTx for PSC between 1993 and 2012. 8/12 patients were female. The age range at diagnosis of PSC was 3.9–15.3 yrs. The age range at transplantation was 13.0–19.0 yrs. Patient follow up ranged from 1.25–18 yrs. Patients received tacrolimus for maintenance immunosuppression after induction with steroids (n=6) or thymoglobulin (n=6). Three patients were diagnosed with rPSC at 44, 60 and 62 months after transplantation. A fourth patient has undergone re-transplantation for graft failure with features of both HAT and rPSC. This patient has distinct histologic features of rPSC in the second graft. After transplantation, 9/12 received either azathioprine or mycophenolate mofetil, 12/12 received ursodiol, 6/11 were treated for CMV, 8/11 had early rejection, 9/11 had IBD. 4/10 recipients and 2/10 donors were HLA-B8 (+), and 2/10 recipients were HLA-DR3,DQ2 (+) . Patient and graft survival was similar between the steroid and thymoglobulin groups. All four children with rPSC received a steroid-free, thymoglobulin induction. Conclusions: In this case series, thymoglobulin induction was the only common observed factor in children with rPSC. Steroid induction and tacrolimus maintenance is now our standard immunosuppression regimen for children undergoing LTx for PSC. Further study is required to determine whether thymoglobulin-induced loss of regulatory T-cells may be a contributing factor for rPSC.


The following people have nothing to disclose: Veena L. Venkat, Sarangarajan Ranganathan, George V. Mazariegos, Rakesh Sindhi


Outcome of Liver Transplantation in children under 10 Kilograms: experience at a Single Italian Tertiary Center (University of Padova, 1993–2012) and Review of the Literature

Mara Cananzi1, Enrico Gringeri2, Giorgia Curia1, Daniele Neri2, Giacomo Zanus2, Graziella Guariso1, Umberto Cillo2;

1Unit of Gastroenterology, Digestive Endoscopy, Hepatology and care of the Child with Liver Transplantation, University Hospital of Padova, Padova, Italy; 2Hepatobiliary and Liver Transplant Unit, University Hospital of Padova, Padova, Italy

Background. Infants (< 1 year) and small children (< 10 Kg) represent more than a third of pediatric liver transplant (LT) recipients. It is commonly assumed that young age and low body weight are associated with increased morbidity and mortality after LT. Objective. To evaluate the outcome of LT, in terms of patient and graft survival, in pediatric recipients < 10 kg. Methods. 1. All children with a body weight under 10kg transplanted at our Institution between January 1993 and May 2012 were retrospectively evaluated. 2. A systematic literature search was conducted through MEDLINE/PUBMED to retrieve all studies, published from January 2000 to September 2012, investigating LT outcomes in children under 1 0kg. Results. 1. 30 patients with a median weight of 8.4Kg (4.4–1 0) and a median age of 1 year (0.1–3.6) were included in the analysis. All children received cadaveric grafts, of which 66.6% were split livers. Median follow up time per patient was 5.7 years. Overall patient and graft survival were 96.7% and 84.5% respectively. Median re-transplantation and re-operation rates were 13% and 40%. 2. 17 articles, were retrieved. Overall patient survival ranged from 60% to 97%, while graft survival from 56% to 93%. Median re-transplantation and re-operation rates were respectively 7% and 35%. Conclusions. Although technically demanding, LT can be performed in children under 10kg with excellent patient and graft survival. Post-transplant complication rates are still high in this group of children; thus, further improvements are needed to reduce patient morbidity after transplant.


Umberto Cillo - Grant/Research Support: Novartis, Bayer, Astellas

The following people have nothing to disclose: Mara Cananzi, Enrico Gringeri, Giorgia Curia, Daniele Neri, Giacomo Zanus, Graziella Guariso


Incidence of mitochondrial disease in children presenting with acute liver failure under 2 years of age

Patrick J. McKiernan1, Sarah Ball2, Saikat Santra2, Girish Gupte1, Khalid Sharif1, Kate Hickman5, Robert McFarland5, Carl Fratter3, Joanna Poulton3, Shamima Rahman4, Robert W. Taylor5;

1 Liver unit, Birmingham Children's Hospital, Birmingham, United Kingdom; 2Department of inherited metabolic disease, Birmingham Children's Hospital, Birmingham, United Kingdom; 3NSCT Rare Mitochondrial Disease Services, Oxford, United Kingdom; 4NSCT Rare Mitochondrial Disease Services, London, United Kingdom; 5NSCT Rare Mitochondrial Disease Services, Newcastle, United Kingdom

Acute liver failure (ALF) in early life has a very poor outcome without liver transplantation (LT). ALF may be the initial presentation of a systemic mitochondrial disease where LT is con-traindicated. It is important to recognise mitochondrial disease as accurately and quickly as possible in order to avoid futile LT without denying this life-saving option to those who would benefit. The best approach to investigate mitochondrial disease in the presence of ALF in young children is unclear. Aim. To determine the incidence of recognised mitochondrial disease in a group of children presenting with ALF under two years. Methods. Consecutive children less than 2 presenting with ALF to a single centre between 2009 and 2011 were studied. Where possible, mitochondrial DNA (mtDNA) copy number was assessed in liver & sequencing of the POLG, MPV17, DGUOK and TRMU genes were undertaken. Results. 40 children (20 male) presented over the study period of whom 26 survived. 11 underwent LT with 8 early survivors. The underlying diagnosis was; infection (n=12), unexplained (n=11), mitochondrial disease (n=6), metabolic disorders (n=5), neonatal haemochro-matosis (NH, n=4) and other (n=2). Five of the children with mitochondrial disease had pathogenic mutations detected: DGUOK (n=2), POLG (n=2) and MPV17 (1). Four of these children died from progressive multisystemic disease while one recovered and remains well 4 years later. The remaining child with mitochondrial disease had equivocal mtDNA depletion in liver and muscle but no detected pathogenic mutations. She underwent LT but died later with evidence of progressive systemic disease. An additional 7 children had evidence of mtDNA depletion in liver tissue only (4 definite and 3 equivocal). One each of these was proven to have NH and enterovirus infection. The other 5 had unexplained liver failure. 3 underwent successful LT while the other 2 recovered. None of the surviving children have developed evidence of systemic disease during follow-up of up to 4 years. Summary. Mitochondrial disease is an important cause of ALF in children less than two years old. Where the genetic defect leading to mtDNA depletion can be characterised, inappropriate LT can be avoided. Evidence of mtDNA depletion in liver in the absence of a recognised genetic defect may be secondary to liver failure. Conclusion. Screening mtDNA maintenance gene mutations may be the most efficient way to exclude mitochondrial disease in ALF in the first two years of life.


Patrick J. McKiernan - Advisory Committees or Review Panels: Swedish Orphan Biovitrum AB

The following people have nothing to disclose: Sarah Ball, Saikat Santra, Girish Gupte, Khalid Sharif, Kate Hickman, Robert McFarland, Carl Fratter, Joanna Poulton, Shamima Rahman, Robert W. Taylor


Generation of Hepatocytes from Arthrogryposis-Renal dysfunction-Cholestasis (ARC) syndrome patient-derived induced Pluripotent Stem Cells

Maëlle Lorvellec1, Anna Straatman-Iwanowska4,5, Nicola D. Ruth1,3, Blerida Banushi4,5, Philip N. Newsome2, Jon Frampton1, Deirdre A. Kelly3, Paul Gissen4,5;

1School of Immunity and Infection. College of Medical and Dental Sciences, University of Birmingham, Birmingham, United Kingdom; 2National Institute for Health Research Biomedical Research Unit and Centre for Liver Research, University of Birmingham, Birmingham, United Kingdom; 3The Liver Unit, Birmingham Children's Hospital, Birmingham, United Kingdom; 4Medical Research Council Laboratory for Molecular Cell Biology, University College London, London, United Kingdom; 5Institute of Child Health, University College London, London, United Kingdom

ARC syndrome is an autosomal recessive multisystem disorder characterised by arthrogryposis, renal dysfunction and cholestasis with low gamma glutamyl transpeptidase activity and abnormal expression of hepatocyte canalicular membrane proteins such as Bile Salt Export Pump (BSEP) and Carcino Embryonic Antigen (CEA). Other features include severe failure to thrive, absence of platelet alpha granules and ichthyosis. Affected infants usually die in the first year of life although milder cases have recently been described. No specific medical treatment is currently available, hence the need to develop disease specific models. Induced Pluripotent Stem Cell (iPSC) technology can provide such patient-specific disease models since iPSCs have the potential to differentiate into any specialised cell of the body and constitute an almost unlimited source of cells for research purposes. Using a single excisable polycistronic lentiviral ‘stem cell cassette’ vector developed by Somers et al, efficient reprogramming of skin fibroblasts from ARC patients was achieved. Simultaneous transfer of the four reprogramming factors Oct4, Sox2, Klf4 and cMyc into human skin fibroblasts using this single vector allows derivation of human iPSCs containing a single excisable viral integration that upon removal generates cells virtually free of integrated transgenes. The iPSCs generated display embryonic stem cell-like morphology and express characteristic stem cell markers. The genetic mutation present in the patient was identified in the derived iPSCs by Sanger sequencing. Furthermore, ARC patient-derived iPSCs were differentiated into Hepatic Endoderm following a method developed by Sullivan et al, mimicking liver natural development. The iPSC-derived hepatocyte-like cells generated exhibited hepatic morphology. Expression of mature hepatocyte markers such as albumin together with fetal hepatocyte proteins such as alpha-fetoprotein, and a combination of cytokeratins 1 8 and 1 9 expression suggests that the cells generated have a phenotype closer to fetal hepatocytes. Unlike control cells, hepatocytes derived from ARC patients were not able to achieve polarised expression of CEA or BSEP, which correlates well with the patient phenotype. This work shows that iPSCs from fibroblasts of infants with rare metabolic diseases such as ARC syndrome can be generated and then differentiated into hepatocyte-like cells, thus providing an in vitro hepatic ARC model for further study of its specific cell phenotype and underlying pathophysiology.


Philip N. Newsome - Grant/Research Support: Novo Nordisk

Deirdre A. Kelly - Consulting: Sanofi Pasteur; Grant/Research Support: BMS, Astellas, Acitllion, MSD, Roche

Paul Gissen - Advisory Committees or Review Panels: Synageva; Speaking and Teaching: Swedish orphan, Actelion

The following people have nothing to disclose: Maëlle Lorvellec, Anna Straatman-Iwanowska, Nicola D. Ruth, Blerida Banushi, Jon Frampton


Cholestatic liver disease in children with Kabuki syndrome

Anastasios Grammatikopoulos1,2, A. S. Knisely2, Giorgina Mieli-Vergani2, Richard J. Thompson1,2;

1Paediatric Liver GI & Nutrition Centre, King's College London at King's College Hospital NHS Trust, London, United Kingdom; 2Institute of Liver Studies, King's College London at King's College Hospital NHS Trust, London, United Kingdom

Background: Kabuki syndrome (KS) is a rare disorder with multisystem anomalies and developmental delay. Mutations in the Myeloid Leukemia Lineage 2 gene (MLL2) underlie KS. The protein encoded by MLL2 is a histone methyltransferase that methy-lates the Lys-4 position of histone H3 and belongs to the ASCOM complex, a transcriptional regulator of the nuclear far-nesoid X receptor (FXR), which regulates several genes involved in bile acid homeostasis. Individual reports describe a variety of cholestatic liver diseases (CLD) in KS with no genetic confirmation. Aim and methods: To describe hepatobiliary disease in KS and to corroborate its genetic association with MLL2 mutations, we reviewed our centre's cohort of paediatric patients with CLD. We identified four individuals with KS. All had high serum gamma-glutamyltransferase (GGT) activity. Clinical, laboratory and demographic data were collected and MLL2 sequencing was undertaken. Results: Patients 1–3 were of European, Patient 4 of Indian origin; the parents of Patient 4 were consanguine. There was no family history of CLD or of KS in any. Each patient was heterozygous fora predictedly pathogenic MLL2 mutation. In Patient 1, disease manifested in late childhood as severe portal hypertension and idiopathic cholestasis. In Patients 2–4, cholangiopathy was present neonatally (see Table). Patient 4 had neonatal sclerosing cholangitis. Patients 2 and 3 had biliary atresia and underwent hepatic portoenterostomy, followed by liver transplantation at age 11 years in Patient 3. Patients 1, 2, and 4 have on-going CLD. Conclusion: This is the first report of CLD in genetically defined KS. One novel (truncating) and three known (1 splice site, 2 truncating) MLL2 mutations were identified in four patients with KS. Clinicopathologic features did not suggest FXR dysregulation. In previous reports, early-onset cholangiopathy also has predominated among CLD of KS. Factors conducing to CLD in KS remain obscure.

Patient number and (gender)Age at presentationPortal tract expansionInflammationBile duct proliferationCholestasisBilirubin (mmol/L)AST (IU/L)GGT (IU/L)MLL2 mutationAmino acid change
1(F)10 years-+-+1753148C.9961C>TR3312*
2(F)3 weeks++++132134614c.74I1C>TR2471*
3(M)7 weeks-+++172193469c.5868–8C>T 
4(F)3 weeks-+++110110159c.1306G>TE436*


The following people have nothing to disclose: Anastasios Grammatikopoulos, A. S. Knisely, Giorgina Mieli-Vergani, Richard J. Thompson


Post-Developmental Genetic Screening for Zebrafish Models of Inherited Liver Diseases

Seok-Hyung Kim1, Simon Wu2, Joshua T. Gamse2, Kevin Ess1;

1Neurology, Vanderbilt University School of Medicine, Nashville, TN; 2Biological Sciences, Vanderbilt University, Nashville, TN

Hepatomegaly and steatosis are often manifestations of underlying problems such as metabolic disorders and infections. To establish zebrafish models of inherited liver diseases due to functional problems rather than liver specification defects during development, we screened 250 ethyl-N-nitrosourea (ENU) mutagenized zebrafish lines for fatty liver and/or hepatomegaly phenotypes at 6 to 8 days post fertilization (dpf). We identified 21 novel mutants showing liver specific defects after completion of normal liver development, which have not been identified during the previous two decades of zebrafish mutant screening. Twelve of mutants (vul02, vul03, vul04, vul05, vul08, vul09, vul12, vul13, vul15, vul16, vul17, 7579) showed hepatomegaly and fatty liver, 5 of mutants (vul01, vul06, vul18, 7580, 7539) have a hepatomegaly phenotype without steatosis symptom, and 4 mutants (vul04, vul07, vul10, vul 14) have fatty liver phenotype without hepatomegaly (Figure 1). Remarkably, 8 of mutants (vul02, vul03, vul09, vul12, vul13, vul16, vul17, vul18) developed ballooning degeneration of hepatocytes resembles steatohepatitis symptom in human and two mutants with severe liver defect (vul03, vul16) showed acute liver necrosis phenotype at 7 to 8 dpf. We also identified that 4 mutants (vul02, vul03, vul09, vul10) have decreased number of intrahepatic bile ducts. We first identified a nonsense mutation in vul02 among mutants showing hepatomegaly and steatosis symptoms. The mutated gene encodes electron transfer flavoprotein alpha subunit (Etfa) which is an essential protein for mitochondrial beta oxidation and recapitulates most of symptoms observed in patients with Multiple Acyl-CoA Dehy-drogenase Deficiency (MADD). Furthermore, molecular identification of these novel liver mutants will also enhance our understanding of genetic variations which could increase risk for alcoholic/non-alcoholic fatty liver disease development in adulthood.

Figure 1.

Three different types of liver specific mutants.


The following people have nothing to disclose: Seok-Hyung Kim, Simon Wu, Joshua T. Gamse, Kevin Ess


Early Phenotype Characterization of a Porcine Model of Hereditary Tyrosinemia Type 1

Jaime Glorioso1, Shennen A. Mao1, Raymond D. Hickey1, Joseph Lillegard1, James E. Fisher1, Milton Finegold3, Ronald J. Marler4, Markus Grompe2, Scott L. Nyberg1;

1Mayo Clinic, Rochester, MN; 2Oregon Health & Science University, Portland, OR; 3Texas Children's Hospital, Houston, TX; 4Mayo Clinic, Scottsdale, AZ

Human hereditary tyrosinemia type 1 (HT1) is an autosomal recessive inborn error of metabolism caused by a deficiency in the enzyme fumarylacetoacetate hydrolase (FAH) usually resulting in hepatic failure early after birth or cirrhosis with proclivity to hepatocellular carcinoma unless treated promptly with the protective drug [2-(2-nitro-4-trifluoro-methylbenzyol)-1,3-cyclo-hexanedione] (NTBC). To model this clinical condition, FAH-null heterozygote pigs were first produced through adeno-associ-ated virus-mediated gene targeting and homologous recombination to disrupt the porcine FAH gene. Heterozygote pigs were bred and a herd of FAH-null homozygote pigs has been produced for phenotype characterization. At birth, genotype was confirmed by PCR. Homozygote piglets were treated with NTBC during the initial postnatal period, and then assigned to various NTBC dosing regimens (0, 0.2, 0.5, and 1 mg/kg/day). Pigs were monitored for clinical symptoms of liver failure. Serologic evaluation was performed monthly and urine was analyzed. Liver ultrasound was obtained at 1 month and subsequent 90 day intervals. To date, heterozygote sows have farrowed 9 litters resulting in 32 homozygote FAH-deficient piglets. Initial evidence indicates that among pigs receiving no NTBC after the postnatal period, there is rapid progression of liver failure with increases in alkaline phosphatase, aspartate aminotransferase, and ammonia in conjunction with coagu-lopathy and hypoglycemia. Blood and urine amino acid analysis reveal significant elevation in succinylacetone consistent with HT1. In addition, there is near complete cessation in weight gain, accumulation of ascites, and clinical evidence of hepatic encephalopathy. Death from acute liver failure occurred at approximately 40 days from withdrawal of NTBC. In comparison, animals that remained on NTBC continued to have serologic analysis within normal limits, expected weight gain, and lack mental status changes during the first three months of life.

In summary, the FAH-null homozygote pig represents the first large animal model of metabolic liver disease. Postnatal withdrawal of NTBC results in a phenotype similar to human HT1 including biochemical and histological evidence of hepatic failure, poor weight gain, and early mortality. The FAH-null homozygote pigs will serve as an effective large-animal model of metabolic liver disease and potentially cirrhosis.


Markus Grompe - Board Membership: Yecuris Corp.; Consulting: Yecuris Corp.; Stock Shareholder: Yecuris Corp.

The following people have nothing to disclose: Jaime Glorioso, Shennen A. Mao, Raymond D. Hickey, Joseph Lillegard, James E. Fisher, Milton Finegold, Ronald J. Marler, Scott L. Nyberg


Role of Hippo Kinase Pathway in Pathogenesis of Poly-cystic Liver Disease

Genea Edwards1, Prachi C. Borude1, ChadM. Walesky1, Darren P. Wallace2,3, Udayan Apte1;

1Pharmacology, Toxicology and Therapeutics, University of Kansas Medical Center, Kansas City, KS; 2Department of Medicine, University of Kansas Medical Center, Kansas City, KS; 3The Kidney Institute, University of Kansas Medical Center, Kansas City, KS

Polycystic liver diseases (PLD) are an inherited genetic disorder characterized by formation of large fluid filled cysts in the liver accompanied by fibrosis and inflammation. PLD can be observed with and without presence of polycystic kidney disease. The autosomal recessive polycystic kidney disease associated with congenital hepatic fibrosis (ARPKD/CHF) has significant hepatic manifestations and is a leading cause of liver transplantation in pediatric patients. The mechanisms of hepatic cystogenesis in ARPKD/CHF are not known and the treatment options are extremely limited. Here we report that deregulation of Hippo Kinase signaling pathway is involved in pathogenesis of liver cyst formation in ARPKD/CHF. We utilized the PCK rats, which carry mutations in the PKHD-1 gene similar to the ARPKD/CHF patients, and PLD patient samples for this study. A time course analysis over postnatal days (PND) 0 to 90 in PCK rats revealed extensive hepatic cystogenesis along with significant hepatomegaly, pericystic fibrosis and inflammation. Interestingly, PCK rats also demonstrated progressive increase in hepatic steatosis. Immunohistochemical analysis revealed significant increase in expression and nuclear localization of Yes Associated Protein (Yap), the downstream regulator of the Hippo Kinase Pathway, in the cyst wall epithelium cells in the PCK rat livers. PCNA analysis showed increased cell proliferation in the cyst wall epithelium. Western blot confirmed the increase in Yap expression and revealed significant decrease in total and active Large Tumor Suppresor-1 (Lats-1) Kinase, the primary regulator of Yap. Substantial increase in Yap target genes including Survivin and CTGF was observed in cyst wall epithelium. The increase in CTGF was also consistent with extensive hepatic fibrosis as demonstrated by Trichrome staining and collagen-I immunohistochemistry. Increase in nuclear Yap, increased proliferation of cyst wall epithelium, and higher CTGF expression was also observed in human PLD samples. Taken together, these data indicate that deregulation of Hippo Kinase signaling pathway leading to increased Yap expression may be involved in hepatic cystogenesis in PLD. These studies were supported by grants to U.A. (P50 DK057301–11)


The following people have nothing to disclose: Genea Edwards, Prachi C. Borude, Chad M. Walesky, Darren P. Wallace, Udayan Apte


Medium-term outcome of children with liver disease secondary to PiZ alpha-1-antitrypsin deficiency

Alemka Jaklin Kekez, Susan M. Chambers, Nedim Hadzic;

King's College Hospital, London, United Kingdom

OBJECTIVE: Homozygous PiZ alpha-1-antitrypsin deficiency (A1ATD) is the commonest genetic cause of liver disease in childhood. The aim of this study was to investigate the prognostic parameters and outcome until adolescence, including requirement for liver transplantation (LT), in a large cohort of children diagnosed at a tertiary referral centre. METHODS: We retrospectively analysed institutional A1ATD database of children diagnosed between 1979 and 2012. Data regarding gender, age and serum bilirubin at diagnosis and disease severity were collected. The children were classified into following clinical phenotypes: a) severe (platelet count <50 x 109/L, albumin <35 g/L, INR >1.2), b) moderate (platelet count between 50–150 x 109/L, albumin >35 g/L), c) mild (abnormal AST, platelet count > 150 x 109/L), d) normalisation of biochemical markers, and e) asymptomatic family members. The children from group a) were typically considered for LT. RESULTS: A total of 292 children (113, 38% boys) were diagnosed with homozygous PiZ phenotype over these 33 years. Median age at genetic testing was 3.7 years (range, 0.04–20.5). Four patients died during pre-LT era. Complete follow up data were available for 240 (82.2%) patients. Of those, 96 (40%) were followed for more than 15 years [median: 19.8 years (range, 15.2 to 31.3)]. The disease was categorised as severe in 21 (21.9%), moderate in 4 (4.2%), mild in 23 (23.9%), normalisation of biochemical markers in 42 (43.7%), while 6 (6.3%) individuals identified through the family screening remained asymptomatic. All 21 (21.9%) patients with severe liver disease were transplanted. There was no difference in gender distribution between non-transplanted and transplanted patients (male/female, 28/47 vs. 11/10, p=0.31). The median age at LT was 6.4 years (range, 0.8 to 15.7). The children who were transplanted had significantly higher levels of serum bilirubin at presentation compared to non-transplanted patients (mean 104 vs. 16.8 μmol/L, p<0.001). CONCLUSIONS: Almost a half of symptomatic children diagnosed with PiZ A1ATD normalise their biochemical markers by early adolescence. About one quarter, typically presenting with higher serum bilirubin levels, require LT in childhood.


The following people have nothing to disclose: Alemka Jaklin Kekez, Susan M. Chambers, Nedim Hadzic


Hepatic Glyoxalase 1 is Downregulated and Hyperacetylated in Pediatric Non-alcoholic Fatty Liver Disease

J Bernadette Moore1, Christos Spanos1, Ciaran P. Fisher1, Caroline Evans2, Bernard M. Corfe2, Alberto Quaglia3, Emer Fitzpatrick3, Anil Dhawan3;

1Nutritional Sciences, University of Surrey, Guild-ford, United Kingdom; 2University of Sheffield, Sheffield, United Kingdom; 3Paediatric Liver Centre, King's College Hospital, London, United Kingdom

Background: Although pediatric non-alcoholic fatty liver disease (NAFLD) is now a common finding in children worldwide, aspects of its molecular pathogenesis are unknown and clinical biomarkers for early diagnosis and accurate disease staging remain scarce. Using iTRAQ-based proteomic profiling we identified hepatic glyoxalase 1 (GLO1) as downregulated in response to high fat feeding in an animal model of NAFLD. The purpose of these experiments was to examine the expression and post-translational modification of GLO1 using fatty acid treated HepG2 cells, primary steatotic hepatocytes and liver biopsies from pediatric NAFLD patients. Methods: GLO1 expression was examined by immunohistochemical staining of tissue biopsies from pediatric NAFLD patients and by immunoblotting of extracts from lipid loaded HepG2 cells. GLO1 was immunoprecipitated from HepG2 and primary human hepatocytes (steatotic and nonsteatotic) and acetylation assessed by immunoblotting and mass spectrometry. HepG2 intracellular lipid, GLO1 expression and acetylation were also assessed by immunofluorescence and high resolution confocal microscopy. Results: Liver expression of GLO1 was qualitatively decreased in 6 of 7 pediatric biopsies examined. Lipid accumulation was pronounced in HepG2 cells treated with oleate and a marked 50% reduction in GLO1 expression (p=0.0017, n=4) was observed compared to vehicle-treated control cells. Equally, pronounced hyperacetylation of both the GLO1 monomer and homodimer was observed in response to lipid loading. Mass spectrometry analyses confirmed the identification of GLO1 and its hyperacetylation in response to increased intracellular lipid. Further immunoprecipitation analyses showed a shift in this post-translational modification of GLO1 in hepatocytes from NAFLD patients. Conclusions: We have identified GLO1 as downregulated and hyperacetylated in pediatric NAFLD patients. This critical enzyme serves to detoxify intracellular, reactive methylglyoxal, the major source of advanced glycated endproducts; although GLO1 deficiency is established as contributing to the pathogenesis of diabetic complications, we are the first to report it altered in NAFLD. Consequently, increased serum methylglyoxal levels represent a potential bio-marker of NAFLD and we are examining this in patients exhibiting varying stages of the NAFLD disease spectrum.


The following people have nothing to disclose: J Bernadette Moore, Christos Spanos, Ciaran P. Fisher, Caroline Evans, Bernard M. Corfe, Alberto Quaglia, Emer Fitzpatrick, Anil Dhawan


Effects of docosahexaenoic acid on liver histology, hepatic progenitor cell activation and macrophage pool in pediatric nonalcoholic fatty liver disease

Valerio Nobili3, Guido Carpino1, Anna Alisi3, Rita De Vito3, Antonio Franchitto2, Paolo Onori2, Gianfranco Alpini4, Eugenio Gau-dio2;

1University of Rome “Foro Italico”, Rome, Italy; 2Sapienza University of Rome, Rome, Italy; 3Bambino Gesù Children's Hospital, Rome, Italy; 4Texas A&M Health Science Center College of Medicine, Temple, TX

Nonalcoholic fatty liver disease (NAFLD) is one of the most important causes of liver-related morbidity and mortality in children. NAFLD has been independently correlated with atherosclerosis and cardiovascular risk. Docosahexaenoic acid (DHA), the major dietary N-3 long-chain polyunsaturated fatty acids, has been proven to be a effective treatment for children with NAFLD. AIM. to investigate in pediatric NAFLD the mechanisms underlying the effects of DHA administration on histo-pathological aspects, hepatic progenitor cell (HPC) activation and macrophage (MΦ) pool. Methods: 11 children with untreated NAFLD were included. Children were treated with DHA for 18 months. Liver biopsies before and after the treatment were analyzed by immunohistochemistry or immunofluorescence. G protein-coupled receptor (GPR)120 expression, HPC activation (presence of Epithelial Cell Adhesion Molecule and Cytokeratin-7 positive HPCs), and macrophage pool were evaluated and correlated with clinical and histo-pathological parameters. Results: hepatocytes, liver MΦs and HPCs expressed GPR120, which is the main receptor for DHA. The treatment with DHA was able to determined: i) a significant improvement of liver steatosis, hepatocyte ballooning, and NAFLD Activity Score (NAS); ii) a significant reduction of HPC activation and ductular reaction extension which were correlated with histo-pathological parameters; iii) the reduction of the number of GPR120 positive (inflammatory) MΦs which was directly correlated with the decrease of pro-inflammatory cytokine serum levels; iv) the increase of GPR120 expression in hepatocytes; v) the reduction of serine-311-phosphorylated NF-kB nuclear translocation in hepatocytes and MΦ which was also in correlation with serum inflammatory cytokines. Conclusion: DHA treatment seems to modulate HPC activation, hepatocyte survival and MΦ polarization through the interaction with GPR120 and NF-kB repression. The modulation of GPR120 exploits a novel crucial role in the regulation of the cell-to-cell cross-talk that drives inflammatory response, HPC activation and hepatocyte survival.


The following people have nothing to disclose: Valerio Nobili, Guido Carpino, Anna Alisi, Rita De Vito, Antonio Franchitto, Paolo Onori, Gianfranco Alpini, Eugenio Gaudio


Colitis enhances liver injury and inflammation in a preterm porcine model of parental nutrition associated-liver disease

Emma M. Tillman1, Dennis D. Black1, Randal Buddington4, Pamela B. Sylvestre3, Peihong Guan1, Richard Helms2;

1Children's Foundation Research Center, The University of Tennessee, Memphis, TN; 2Clinical Pharmacy, The University of Tennessee, Memphis, TN; 3Methodist Le Bonheur Health Care, Memphis, TN; 4University of Memphis, Memphis, TN

Background: One of the most devastating complications affecting infants and children with intestinal failure is parenteral nutrition (PN)-associated liver disease (PNALD). The etiology of PNALD is not well understood, but prematurity, prolonged PN, and concurrent intestinal injury are considered significant risk factors in the development and severity of disease. The aim of this study was to evaluate liver injury after exposure to PN and colitis in a preterm porcine model. Methods: Preterm pigs were obtained via caesarian section on day 104 of 114 day porcine gestation. All animals received PN for the first 24 hours and were then randomized to enteral nutrition (EN), EN + colitis, PN, or PN + colitis. Colitis was induced with dextran sodium sulfate (DSS) given enterally every 12 hours for three doses starting 24 hours after delivery. Necropsy was performed on day 12. Serum, as well as snap-frozen and formalin-fixed liver, were collected for analysis. Liver RNA was extracted from frozen tissue to evaluate eighty-four key genes involved in inflammation using the Pig Inflammatory Cytokines and Receptors RT Profiler PCR Array system (SABiosciences, Frederick, MD) using quantitative RT-PCR. Liver histology was scored by a pathologist blinded to the experimental groups. Results: Eighteen pigs were initiated, nine pigs treated with DSS expired prior to the scheduled necropsy, and nine completed the day twelve-day protocol. There was no difference in baseline weight, serum transaminases, direct bilirubin (DB), or total bile acids (TBA) among the four groups. At necropsy, mean DB and TBA were higher in the PN and PN+DSS groups compared to control, but this did not reach statistical significance. Animals treated with PN developed mild steatosis; whereas, animals treated with PN+DSS had moderate steatosis. In addition, PN+DSS pigs had significant ballooning degeneration and presence of Mallory bodies, but this was not observed in the other groups. Analysis of liver tissue from animals treated with PN+DSS compared to control (EN) identified 26 genes that were up- or down-regulated using a boundary of 4 when the two groups were plotted using 2-ΔCT. Conclusion: Preterm pigs developed only mild hepatic biochemical changes after twelve days of PN, although histological changes were present. The degree of steatosis increased when a colitis insult was combined with PN treatment. Liver mRNA analysis revealed up and down-regulation of key inflammatory genes. These data suggest that gut inflammation has a significant impact on liver health in the preterm pig exposed to PN.


The following people have nothing to disclose: Emma M. Tillman, Dennis D. Black, Randal Buddington, Pamela B. Sylvestre, Peihong Guan, Richard Helms


Non invasive assessment of hepatic fibrosis in children: performance of liver stiffness measurement and aspar-tate transaminase to platelet ratio

Azza Youssef1, Tawhida Y. Abdel ghaffar1, Gamal Esmat2, Abdel Aziz Abdel Wanis1;

1 Pediatrics, Faculty of medicine, Ain Shams Univesity, Cairo, Egypt; 2Tropical medicine, Faculty of medicine, Cairo university, Cairo, Egypt

Background : Because of the limitations of liver biopsy, there is strong interest in developing less invasive methods to assess liver fibrosis specially in children. LSM and APRI score are another non invasive means of liver fibrosis assessment utilizing simple commonly performed tests. Objective: to evaluate LSM and APRI score as non invasive means of fibrosis assessment in children with chronic liver disease in general and Wilson disease and Autoimmune hepatitis in particular. Patients and methods : Liver biopsy was done for 52 children (25 boys & 27 girls with mean age of 11.5 years) suffering from various chronic liver diseases . The stage of fibrosis was assessed according to METAVIR system. APRI score was calculated for 42 of the children and LSM by transient elastography was performed to each of the children as well as to 18 healthy age and sex matched controls. The correlation between METAVIR fibrosis stage and each of the APRI score and LSM was studied. Cutoff points for differentiation of no or mild fibrosis from significant fibrosis in the group as a whole was calculated for APRI and LSM using ROC curves. Correlation studies and cutoff points of LSM were further calculated for patients with Wilson disease (n=20) and those with autoimmune hepatitis (n=16). Results : Twenty three children had no or mild fibrosis(<F2 Metavir )(G1) while 29 had significant fibrosis (>=F2 Metavir) (G 2). The mean APRI score of the patients was 0.71 +/- 0.48. It was significantly higher in G2 vs G1 patients (0.71 vs 0.3) and showed significant correlation with METAVIR staging (r=+0.53, P<0.001). At a cut off of 0.58 it had 63%, 73%, 70.4%, 66.7% sensitivity, specificity, PPV & NPV respectively for significant fibrosis. LSM was significantly higher in patients vs controls (38.8 vs 11.1 kPas &P<0.000) and in G2 vs G1 (15.4 vs 6.9 Kpas& P<0.000). It showed significant positive correlation with both METAVIR staging & APRI score (r=0.58, 0.69 respectively& P<0.001). A cut off of 8.1Kpas had a 78%, 73%, 77.8% &73.3% sensitivity, specificity, PPV & NPV respectively for significant fibrosis. In WD cut off for LSM was 8.5Kpas with 0.84, 69%, 71%, 81.8%, 55.6% AUR, sensitivity, specificity, PPV & NPV respectively . In AIH a cut off of 6.40 for LSM had 0.95, 90%, 83%, 90%, 83.3%AUR, sensitivity, specificity, PPV and NPV respectively for diagnosis of significant fibrosis. Conclusion: In children, APRI score and LSM perform reasonably well in fibrosis assessment. Diseases with different etiologies have different cutoff values for significant fibrosis.


The following people have nothing to disclose: Azza Youssef, Tawhida Y. Abdel ghaffar, Gamal Esmat, Abdel Aziz Abdel Wanis


The expression of IL-22 and clinical characters in Chinese pediatric patients with chronic hepatitis B

Yanwei Zhong, Shishu Zhu, Hongfei Zhang, Yi Dong, Zhiqiang Xu, Dawei Chen, Limin Wang, Gan Yu, Fuchuan Wang, Zhibin Li, Yuan Yang;

302hospital, Beijing, China

Background: Interleukin (IL)-22 induces acute phase reactants and favors anti-microbial defence and protects tissues from damage. IL-22 is important in chronic skin inflammation, however, its role in chronic hepatitis B (CHB) in pediatric patients remains unclear. We explored the expression of IL-22 and evaluated the association between IL-22 and clinical characters in pediatric patients with CHB. Methods: The level of IL-22 was measured using flow cytometry analyses from 45 pediatric patients with CHB and 9 healthy controls. The serum HBV DNA loads and alanine aminotransferase levels were measured according to international standard method. The degree of liver inflammation was graded using the histological activity index (HAI), and the fibrosis stage was staged according to a modified METAVIR scoring system. Results: Compared with healthy controls, the level of IL-22 in peripheral blood was significantly higher in pediatric patients with CHB. In CHB patients, the expression of IL-22 was a significant negative correlation with both liver HAI score and fibrosis stage,but expression of IL-22 had no correlation with serum HBV DNA loads and alanine aminotransferase levels. Conclusion: IL-22 may provide protection in liver inflammation/fibrosis in the pediatric patients during CHB infection.


The following people have nothing to disclose: Yanwei Zhong, Shishu Zhu, Hongfei Zhang, Yi Dong, Zhiqiang Xu, Dawei Chen, Limin Wang, Gan Yu, Fuchuan Wang, Zhibin Li, Yuan Yang


The Prevalence and Spectrum of Abnormal Liver Biochemistry in a Large Cohort of Children with Inflammatory Bowel Disease

Pamela L. Valentino1,2, Brian M. Feldman2,3, Thomas D. Walters1,2, Anne M. Griffiths1,2, Simon C. Ling1,2, Eleanor Pullenayegum4, Binita M. Kamath1,2;

1Division of Gastroenterology, Hepatology, and Nutrition, Hospital for Sick Children and the University of Toronto, Toronto, ON, Canada; 2Department of Paediatrics, University of Toronto, Toronto, ON, Canada; 3Division of Rheumatology, Hospital for Sick Children and the University of Toronto, Toronto, ON, Canada; 4Child Health Evaluative Sciences, Hospital for Sick Children and the University of Toronto, Toronto, ON, Canada

Background: There is scant literature regarding abnormal liver biochemistry (LB) in inflammatory bowel diseases (IBD). We sought to characterize the prevalence, spectrum and clinical associations of abnormal LB in children with IBD. Methods: A random sample of 300 children with IBD was retrospectively ascertained from the IBD database at the Hospital for Sick Children in Toronto. Serial LB measurements from the time of IBD diagnosis and onwards were evaluated to detect abnormalities and a Kaplan-Meier time to event analysis was performed. Cox-proportional hazards models identified clinical variables associated with abnormal LB. Results: IBD in this population was classified as Crohn disease in 163 (54%), ulcerative colitis in 100 (33%), and IBD-Undefined in 37 (12%), with a median duration of follow-up of 2.8years (interquartile range [IQR]: 1.5, 4.8years). 297/300 patients (99%) had LB measured at some point after their diagnosis of IBD. Overall, 160 children with IBD (53%) developed abnormal LB, with a median time to development of 30months following the IBD diagnosis (95% confidence interval [CI]: 21.7, 41.1 months). In 24% (72/300) the LB was abnormal at ≥2x the upper limit of normal (ULN). The abnormal LB persisted beyond 30 days in 16% at 1–2xULN, and 8% at ≥2xULN. 4.3% (13/300) children developed primary sclerosing cholangitis (PSC) and 1.7% (5), autoimmune sclerosing cholangitis (ASC). Variables associated with the increased development of abnormal LB included: steroid use (hazard ratio [HR] 1.8 [CI 1.1,2.9], p=0.012), antibiotic use (metronidazole or ciprofloxacin, HR 1.7[CI 1.1,2.6], p=0.016), and Exclusive Enteral Nutrition use (EEN, HR 2.4[CI 1.2,4.8], p=0.017). The association between abnormal LB and medication use was more pronounced at a biochemistry threshold of >2xULN with a HR of 3.7 for steroid use (CI 2.1,6.3, p<0.0001) and 2.4 for antibiotics use (CI 1.2,4.3, p=0.004). There were no significant associations between methotrexate use and abnormal LB, and infliximab had a protective effect (HR 0.56 [CI 0.32,0.98], p=0.043). Conclusions: Abnormal LB is a prevalent problem in pediatric IBD occurring in 53% of this cohort within the first 3 years after diagnosis. However, the degree of abnormal LB is typically mild, and not persistent, and the development of chronic liver disease is a relatively rare event. Substantial elevations in LB are associated with steroid, antibiotic, and EEN use, which may be markers for IBD exacerbations. These data suggest that conservative observation is appropriate for most children with IBD and abnormal LB, in whom costly and invasive investigations are unnecessary.


Thomas D. Walters - Advisory Committees or Review Panels: Janssen, AbbVie; Consulting: Janssen, AbbVie; Grant/Research Support: Janssen, AbbVie; Speaking and Teaching: Janssen, AbbVie, Ferring

Simon C. Ling - Consulting: Gilead, Isis Pharmaceuticals; Grant/Research Support: Bristol Myers Squibb

The following people have nothing to disclose: Pamela L. Valentino, Brian M. Feld-man, Anne M. Griffiths, Eleanor Pullenayegum, Binita M. Kamath


A Biomarker Study: Plasma Osteopontin Levels Reflect Hepatic Fibrogenesis-Related Gene Expression in Biliary Atresia

Akihiro Asai1,2, Samyukta Malladi2, Jonathan Misch2, Padmini Malladi2, Jorge A. Bezerra3, Peter F. Whitington1,2;

1Pediatrics, Northwestern University, Chicago, IL; 2Pediatrics, Ann & Robert H. Lurie Children's Hospital of Chicago Research Center, Chicago, IL; 3Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, OH

Background: Osteopontin (OPN) is a profibrogenic cytokine that is markedly over expressed in liver of Biliary Atresia (BA). Hedgehog (Hh) pathway is highly active and stimulates biliary dysmorphogenesis in BA. OPN is a known Hh response element. In BA, reactive ductules along the limiting plate are both Hh- responsive and produce OPN. We hypothesized that in BA liver gene expression of OPN correlates with expression of genes related to Hh activation and fibrogenesis, and because it is a secreted protein that plasma and/or urine OPN is a potential biomarker for fibrogenesis. Methods: We studied matched liver tissue, plasma and urine obtained at the time of Kasai procedure from 30 infants with BA acquired from the Childhood Liver Disease Research and Education Network. OPN concentrations were measured by MSD immunoassay; liver gene expression by microarray; and degree of fibrosis by

Sirius Red computerized quantitative morphometry and Ishak fibrosis score (0–6). Results: Plasma OPN in BA ranged from 185–827ng/ml (normal infant [N=5] 43–311). Urine OPN, whether corrected for Cr or not, did not significantly correlate with plasma OPN. Liver OPN mRNA expression ranged from 3–27 fold-normal and the fibrosis gene COL1A1 expression ranged from 5–20 fold-normal. OPN gene expression correlated with COL1A1 (p=0.002, r2=0.3), as did expression of Hh receptor patched1 (p=0.012, r2=0.2) and Hh-related EMT marker vimentin (p<0.001, r2=0.55). Plasma OPN levels correlated with liver OPN gene expression (p=0.029, r2=0.1 6). In 27 BA cases with adequate tissue the area density of fibrosis ranged from 3.5–40% (NL infant < 1%) and the fibrosis scores from 3–6: 3 (n=6), 4 (n=14), 5 (n=6), and 6 (n=1). Tissue COL1A1 and OPN gene expressions correlated with fibrosis area density (p=0.046, r2=0.15 and p=0.026, r2=0.18, respectively) as well as fibrosis score (ANOVA p=0.007 and p=0.01, respectively). Plasma OPN failed to significantly correlate with either fibrosis area density or score. Conclusions: The correlation of gene markers of Hh signaling activity, Hh-driven EMT and OPN with COL1A1 gene expression further suggest an important role for Hh activation and OPN in BA fibrogenesis. The correlation of plasma OPN concentration with liver OPN gene expression suggests potential as a non-invasive biomarker for this fibrogenic process. Urine OPN concentration does not accurately reflect plasma OPN and lacks biomarker potential. Further study is needed to determine if plasma OPN measurement might significantly contribute to a biomarker panel predictive of fibrosis related outcome (i.e. portal hypertension and need for transplant) in BA.


Jorge A. Bezerra - Grant/Research Support: Molecular Genetics Laboratory, CHMC

The following people have nothing to disclose: Akihiro Asai, Samyukta Malladi, Jonathan Misch, Padmini Malladi, Peter F. Whitington


Urinary Bisphenol-A is Associated with Non-Alcoholic Steatohepatitis in Adolescents: National Health and Nutrition Examination Survey (NHANES) Data 2003–2010

Sofia G. Verstraete, Emily R. Perito, Janet M. Wojcicki, Philip Rosen-thal;

Division of Pediatric Gastroenterology, Hepatology, and Nutrition, U.C.S.F., San Francisco, CA

Bisphenol-A (BPA) is a chemical used to coat polycarbonate bottles as well as food and beverage cans. BPA alters lipid metabolism and insulin resistance in hepatocytes and adipocytes in vitro and in animal models. Its exposure in humans has been associated with type 2 diabetes, cardiovascular disease, obesity, and increased serum GGT and alkaline phosphatase concentrations. Urinary BPA levels correlate with oxidative stress metabolite levels, which are thought to contribute to liver damage in non-alcoholic steatohepatitis (NASH). The relationship between BPA exposure and NASH has not been studied. Methods: We used 2003–2010 NHANES data to do a cross-sectional analysis of 2160 subjects aged 12–19 years without liver disease of other etiology who had been randomly selected for measurement of urinary BPA concentration. We used appropriately weighted multivariate logistic regression to investigate the association between urinary BPA and NASH, considering NHANES complex survey sampling design. We defined NASH as serum ALT >40U/L in overweight/obese adolescents with insulin resistance (HOMA IR [Homeostasis model assessment of insulin resistance] >3.16). Results: We created unweighted sample quartiles for urinary BPA (Table1). Prevalence of NASH was highest in the second quartile (2.3%, SE 0.6). In multivariate analyses, participants in the 2nd and 3rd quartiles of urinary BPA had > 4 times the odds of NASH than the lowest quartile, after adjusting for race, gender, age, BMI Z-score, country of birth, and metabolic syndrome. Other predictors associated with NASH in the multivariate model included Hispanic race (p=0.007), male gender (p=0.002), and BMI Z-Score (p<0.001). Conclusions: Urinary BPA levels are associated with an increased risk of NASH. The relationship of BPA exposure and NASH requires further investigation.

Table 1. Association Between NASH, Urinary BPA, and Race*
  Odds Ratio (95% Confidence Interval)p-value
  1. *Adjusted for Gender, Age, Race, Metabolic Syndrome, Body Mass Index Z-Score,Country of Birth, Urinary Creatinine.

Urinary Bisphenol-A Quartile(ng/ml)<1.3Reference 


Philip Rosenthal - Advisory Committees or Review Panels: Ikaria, Gilead, Merck, General Electric; Consulting: Roche; Grant/Research Support: Roche, Bristol MyersSquibb, Gilead, Vertex

The following people have nothing to disclose: Sofia G. Verstraete, Emily R. Per-ito, Janet M. Wojcicki


Dietary intake may influence essential fatty acid and long chain polyunsaturated fatty acid (LCPUFA) expression in children with nonalcoholic fatty liver disease (NAFLD)

Diana Mager1,2, Vera Mazurak1, Spencer Proctor1, Susan M. Gilmour2, Jason Yap2;

1Agricultural, Food & Nutritional Science, Univ of Alberta, Edmonton, AB, Canada; 2Pediatrics, University of Alberta, Edmonton, AB, Canada

Background: Alterations in hepatic fat metabolism induced by hyperinsulinemia and insulin resistance (IR) and diets high in saturated fat/simple sugars may contribute to an increased risk for nonalcoholic fatty liver disease (NAFLD) in obese children and adolescents. We hypothesized that children with NAFLD and hyperinsulinemia/IR who consume diets high in saturated fat/simple sugars (e.g fructose) would have altered essential fatty acid (EFA) and long chain polyunsaturated fatty acid (LCPUFA) expression when compared with healthy lean children. Methods: We prospectively studied 32 children (n=13 NAFLD with biopsy proven disease/n= 19 healthy lean children) between 7–18 years. Fasting blood for analysis of ALT, AST, total-cholesterol (TC), HDL-cholesterol, LDL-cholesterol, triglyc-erides (TG), glucose, insulin, nonesterified free fatty acids (NEFA), inflammatory markers (IL-6, IL-10, TNF-α, CRP) Apolipoproteins (Apo-B100, ApoB-48 and ApoC-III) and FA composition in red blood cell (RBC) fractions was collected. IR was assessed using the homeostasis model of IR (HOMA-IR>3 abnormal). RBC-FA composition in phosphatidylethanolamine (PE)/phosphatidylcholine (PC) fractions were analyzed using gas chromatography. Dietary intake was measured using three-day food intake records. Results: Children with NAFLD had a significantly higher percentage of dietary energy derived from fat (34.1 ± 7.3% vs 25.9 ± 7.8%; p=0.0008) and lower amounts from carbohydrate (49.9 ± 7.6% vs 57.1 ± 7.8%; p=0.0006) and monounsaturated fatty acid (MUFA) (7.5 ± 4.2 % vs 4.7 ± 2.2%; p=0.02) when compared to healthy controls. Significant elevations in ALT (80 ± 53 vs 16± 5 U/L), AST (48 ± 28 vs 23 ± 5 U/L), ApoB-100 (687 ± 317 vs 305 ± 129 μg/mL), Apo-CIII (20.6 ± 14.1 vs 9.9 ± 6.0 μg/dL), insulin (30 ± 16vs 10±5mU/L), HOMA-IR (6.8 ± 4.5 vs 2.0 ± 1.1), TG (1.4 ± 0.8 vs 0.7 ± 0.3 mmol/) and CRP (3.3 ± 0.4 vs 0.6 ± 0.2 mg/dL) were observed in children with NAFLD (p<0.05). Children with NAFLD also had lower proportions of C1 8:2n6, C18:3n3, C20:5n3, C22:6n3, ΣMUFA, Σ n-3 along with higher proportions of C18:0 and ΣSFA in RBC PE/PC fractions when compared to controls (p<0.05). Diets higher in MUFA, omega-6 FA, glycemic load, fructose and increased IR were associated with decreased proportions of C1 8:2n6, C1 8:3n6 and increased proportions of C20:4n6 in RBC-PE fractions (p<0.05), but showed no direct relationships to C22:5n4 and C22:6n3 composition. . Conclusions: Children and adolescents with NAFLD have reduced EFA and LCPUFA in PE fractions of RBC when compared to lean children, which may be due in part to dietary composition.


The following people have nothing to disclose: Diana Mager, Vera Mazurak, Spencer Proctor, Susan M. Gilmour, Jason Yap


The use of bone marrow aspiration as an investigative tool in pediatric liver disease

Cherry Alviani1, Emer Fitzpatrick1, Sue Height2, Nedim Hadzic1;

1Dept. Paediatric Hepatology, King's College Hospital, London, United Kingdom; 2Dept. Paediatric Haematology, King's College Hospital, London, United Kingdom

Introduction: Bone marrow aspiration (BMA) has traditionally been an important tool in the investigation of pediatric liver disease, particularly in the diagnosis of storage conditions or Haemophagocytic lymphohistiocytosis (HLH). It is an invasive measure however, and diagnostic yield is uncertain. We reviewed experience of BMA in a tertiary pediatric liver centre to determine its best application in practice. Methods: Pediatric liver patients who had a BMA for suspected storage conditions over a 10 year period were identified though a haematology database. Clinical, metabolic, haematological and histological data were reviewed. SPSS v 17.0 was used for analysis. Results: During the study period, 154 children (105 boys) underwent 182 BMAs for suspected storage conditions. Median age of presentation was 0.39 years (IQR 0.19, 1.63). Presenting features were conjugated hyperbilirubinemia in 75 (38 also splenomegaly (SM), 3 hepatomegaly (HM) and 14 hepatosplenomegaly (HSM)), acute liver failure in 29, isolated HM in 7, SM in 10 and HSM in 23. Final diagnosis was Nie-mann Pick C (NPC) in 13, NPA or B in 3 and Wolman disease/cholesterol ester storage disease in 2, Glycogen storage disease in 4, mitochondrial cytopathy in 6, HLH in 4. Aetiology was not identified in 70 (46%), 7 of whom have died and the remainder discharged or followed with stable disease. BMA was haemodilute/inadequate in 20(13%). The sensitivity of initial BMA for detection of storage conditions/HLH in the setting of ALF was 57% (86% on repeat) with a positive predictive value (PPV) of 100% and a negative predictive value (NPV) of 83%. In infants < 1year with conjugated hyperbilirubinemia /HSM, the sensitivity for detection of storage disorders was 71% with a PPV of 91% and a NPV of 76%. In children >1 year, the sensitivity was 83% with PPV of 100% and NPV of 95%. 60(40%) had white cell enzyme analysis, this was diagnostic in 11 (18%) with a PPV of 50% and NPV of 100%. In 6(10%) cases results were inconclusive. The assay took up to 4 weeks. Fibroblast culture with Filipin staining is the gold standard for diagnosis of NPC. Skin biopsy was done in 49. This was diagnostic in 18 with a PPV of 95% and a NPV of 96%. All of these children had a previously diagnostic BMA or WCE. Time from biopsy to results was a median 3 months. Discussion: The overall sensitivity PPV and NPV for BMA as an investigative tool for investigation of pediatric storage disorders were high, particularly in the setting of acute liver failure when a rapid diagnosis was necessary. Inadequate samples requiring repeat procedure are frequent however, and for non-urgent indications, WCE and skin biopsy may be preferred.


The following people have nothing to disclose: Cherry Alviani, Emer Fitzpatrick, Sue Height, Nedim Hadzic


Partial Splenic Embolization in the Management of Portal Hypertension in Children

Katherine A. Orellana1, Nadia Ovchinsky1, Mercedes Martinez1, Peter Shlossberg2, Jonathan Susman2, Vladimir Sheynzon2, Steven J. Lobritto1;

1Pediatrics, Columbia University, Morgan Stanley Children's Hospital of New York, New York, NY;2Radiology, Columbia University, New York, NY

Complications of portal hypertension such as hypersplenism and variceal hemorrhage can cause significant morbidity in children. Partial splenic artery embolization (PSE) has been performed in adults and children for the treatment of complications of portal hypertension. In this report, we retrospectively review our single center experience with PSE in pediatric patients with portal hypertension of various etiologies. Fourteen patients underwent PSE due to hypersplenism or recurrent variceal hemorrhage between March 2010 and March 2012. Five patients (33%) were female and the average patient age was 12 years (range 1.5 to 20 years). The primary diseases leading to portal hypertension included BA with a functioning Kasai portoenterostomy (3), portal sclerosis (1), primary sclerosing cholangitis (2), autoimmune hepatitis (1), portal vein thrombosis (4), nodular regenerative hyperplasia (1), cystic fibrosis (2), hepatocellular carcinoma (1), and idiopathic portal hypertension (1). PSE was accomplished using embosphere particles with catheter positioned in distal splenic artery beyond pancreatic branches. The embolized splenic volume targeted was 60–70%. Patients were admitted after PSE for pain management. Broad-spectrum antibiotics were administered preoperatively and continued for 5–7 days. Vaccinations for pneumococcus, meningococcus, and H. influenza were administered prior to embolization. The average hospital length of stay was 4.8 days. Four children were readmitted for fever with pain (3) and dehydration (1). The platelet count improved after PSE with values greater than 100,000 in eleven patients (78%). Children with prior esophageal varices showed improvement after PSE with only five children (35%) requiring further endo-scopic therapy. Two patients had PSE > 5 years after liver transplantation (LT) and one patient had PSE 23 months prior to eventual LT. The follow-up period after PSE ranged from 6 to 24 months. After PSE, patients experienced transient abdominal pain and distention, fever, and peri-splenic fluid collections. However, no serious complications such as splenic abscess, splenic rupture, bleeding, pancreatic infarction, opportunistic infection or death were observed. In conclusion, PSE is an effective and safe alternative in the management of pediatric portal hypertension in select situations. We have successfully performed PSE in both pre- and post-LT patients as well as cirrhotic and non-cirrhotic pediatric patients. PSE is considered a less invasive alternative to splenectomy and portal systemic shunting since it preserves functional spleen mass and avoids post-procedure accelerated liver disease or encephalopathy.


Jonathan Susman - Board Membership: Angiodynamics; Consulting: Covidien

The following people have nothing to disclose: Katherine A. Orellana, Nadia Ovchinsky, Mercedes Martinez, Peter Shlossberg, Vladimir Sheynzon, Steven J. Lobritto


Outcomes of varices in children

Juan Cristóbal Gana3, Tassos Grammatikopoulos4, Yael Mozer-Glassberg6, Jason Y. Yap5, Veronique Morinville7, Yaron Avitzur1,2, Sofía Verdaguer3, Paula Troncoso3, Simon C. Ling1,2;

1Hospital for Sick Children, Toronto, ON, Canada; 2Department of Paediatrics, University of Toronto, Toronto, ON, Canada; 3Pontificia Universi-dad Catolica de Chile, Santiago, Chile; 4King's College Hospital, London, United Kingdom; 5Unversity of Alberta, Edmonton, AB, Canada; 6Schneider Children's Medical Centre, Petach Tikva, Israel; 7Montreal Children's Hospital, Montreal, QC, Canada

Background: There is little published data on the outcomes of children with endoscopically-proven varices. We aimed to describe 2-year outcomes in a group of children with esophageal varices characterized at endoscopy at multiple centres. Methods: 2-year follow-up data was collected on children previously identified to have esophageal varices during our previous multicentre study of non-invasive diagnosis of varices (1). Data collected included details of subsequent endoscopies and outcomes including bleeding episodes, transplant and death. We explored potential predictors of bleeding, including platelet count and our previously described clinical prediction rule (CPR) (1). Results: 44 patients were included in this follow-up study. At baseline 36% had grade 1 varices by the Paquet classification, 32% grade 2, 25% grade 3 and 2% grade 4 varices. Of 24 children (55%) who were re-scoped during the follow-up period, 42% had the same grade of varices, 21% had increased grade by one, 4% had increased grade by two, 25% had decreased grade by one, and 4% had decreased grade by two. Treatment to prevent bleeding was provided to 11 children (8 beta-blockers, 2 banding, 1 combined). Three patients (7%) had bleeding episodes, 1 with initially grade 2 varices and 2 with initially grade 3 varices (one of whom still had grade 3 varices when re-scoped). Platelet counts were 49, 95 and 72 x10^9/l in the bleeders, compared to a mean (SD) of 95 (50) x10^9/l in non-bleeders. The CPR was not available in one bleeder, and calculated as 129 and 93 in the other two. CPR mean (SD) in non-bleeders was 103 (19). Conclusions: In this cohort of children with varices, the incidence of variceal bleeding was low (7%) during a 2-year follow-up period. Variceal grade progresses slowly in a minority of patients. Reference: (1) Gana JC et al, Gastroenterology 2011; 141(6): 2009–16.


Simon C. Ling - Consulting: Gilead, Isis Pharmaceuticals; Grant/Research Support: Bristol Myers Squibb

The following people have nothing to disclose: Juan Cristóbal Gana, Tassos Grammatikopoulos, Yael Mozer-Glassberg, Jason Y. Yap, Veronique Morinville, Yaron Avitzur, Sofía Verdaguer, Paula Troncoso


Abernethy malformation: associations, complications, and outcomes - a single center experience

Frank DiPaola, Ashley Walther, Greg Tiao, Maria H. Alonso, Jorge A. Bezerra, Jaimie D. Nathan;

Cincinnati Children's Hospital Medical Center, Cincinnati, OH

Abernethy malformation is a rare congenital malformation wherein venous return from the intestines and spleen bypasses the liver. The portal vein may be absent (type I) or present (type II). Patients are at risk for complications including encephalopathy, intrahepatic masses, hepatopulmonary syndrome, and pulmonary hypertension. Complications and response to treatments, including shunt ligation and liver transplant, are poorly defined. To obtain insight into the clinical presentation and outcomes, we performed a retrospective chart review of all patients with Abernethy malformation who have received care at Cincinnati Children's Hospital Medical Center. Results: Ten cases of Abernethy malformation (7 females and 3 males ranging in age from neonate to 26 years) were characterized by interventional angiography. Nine had other anomalies including congenital heart disease, other vascular abnormalities, and gastrointestinal, genitourinary, and vertebral anomalies. Three had Turner syndrome: 2 patients with type I and 1 patient with type II. Three cases (ages 9, 10, and 17 years) of type II Abernethy malformation underwent interventional shunt occlusion and subsequent imaging confirmed increased portal flow: 2 patients had severe hyperammonemia and episodic encephalopathy which resolved after shunt occlusion. Notably, one of these three patients was initially misdiagnosed with a type I malformation by computed tomography prior to interventional angiography. Three cases had liver masses: focal nodular hyperplasia in a 10 year old male, hepatic adenoma (with β-catenin and HNF-1α mutations indicating malignant potential) in a 17 year old female, and hepatocellular carcinoma in an 11 year old male. The latter patient underwent segmental liver resection with shunt ligation, and remained well with no further treatment. One three month old child with type I Abernethy malformation presented with severe hypoxemia and pulmonary arteriovenous malformations consistent with hepatopulmonary syndrome. After orthotopic liver transplant, the hypoxemia resolved completely in two months. Conclusions: 1) serious complications of Abernethy malformation can present in early childhood, 2) interventional angiography should be performed to avoid mischaracterizing a type II Abernethy malformation as a type I, 3) shunt occlusion in cases of type II Abernethy malformation is effective in reversing encephalopathy, 4) a variety of mass lesions may occur and hepatic adenomas should be carefully evaluated for malignant potential, and 5) severe hepatopulmonary syndrome can occur early in infancy in patients with type I Abernethy malformation and may resolve following liver transplant.


Jorge A. Bezerra - Grant/Research Support: Molecular Genetics Laboratory, CHMC

The following people have nothing to disclose: Frank DiPaola, Ashley Walther, Greg Tiao, Maria H. Alonso, Jaimie D. Nathan


Correlation of liver stiffness, as determined by transient elastography, with the presence and severity of clinical liver disease in children and young adults with cystic fibrosis

Amal A. Aqul1, Sarah Harney1, Roshan Raza1, Gregory S. Saw-icki2, Paul D. Mitchell3, Maureen M. Jonas1, Rima Fawaz1;

1Pedi-atric Gastroenterology and Hepatology, Boston Children's Hospital/ Harvard Medical School, Boston, MA; 2Pediatric Pul-monology, Boston Children's Hospital/ Harvard Medical School, Boston, MA; 3Clinical Research Center, Boston Children's Hospital/ Harvard Medical School, Boston, MA

Background Liver disease (LD) is the third leading cause of death in cystic fibrosis (CF). CFLD includes a range of abnormalities with variable degrees of fibrosis. No test has been recognized as a reliable tool to identify patients with or at risk for advanced fibrosis. Liver biopsy is an imperfect indicator of fibrosis severity due to the focal nature of CFLD. Aim To evaluate whether or not liver stiffness, determined by transient elastography (TE), correlates with the presence and severity of LD in children and young adults with CF. Methods Subjects were divided into 3 groups based on clinical, biochemical, sonographic, and endoscopic findings; Group A: no CFLD, Group B: CFLD without portal hypertension (PHTN), and Group C: CFLD with PHTN. Patients underwent TE during routine CF visits. Median and inter-quartile range (IQR) for Liver Stiffness Measurement (LSM) was calculated for each group. LSM was compared to AST/platelet ratio index (APRI), an indicator of hepatic fibrosis. Results: 209 CF patients (52% male; mean age 13.2±6.9y; 9(4%) <2y and 54 (26%) 18–26y) underwent LSM. 187 (89%) subjects with complete data and a valid LSM resulted in 121 subjects (65%) in Group A, 59 (31%) in Group B, and 9 (5%) in Group C. Elevated LSM (>6 kPa) was found more often in patients >11y (N=45, 37%), irrespective of liver disease severity (P<0.0001). The median (IQR) LSM was different among the 3 groups: 4.6 (3.8, 5.4) kPa in Group A, 5.1 (4.4,6.3) kPa in Group B and 23.1 (11.8,24.8) kPa in Group C (P<0.0001) with all pairwise comparisons different from one another (P<0.05). Similarly, median (IQR) APRI was different among the 3 groups: Group A 0.22 (0.17, 0.28), Group B 0.26 (0.18, 0.33), Group C 0.54 (0.29, 0.95); P= 0.0002. Analysis of receiver operating characteristics for discriminating Group C from Groups A and B resulted in cut-points at 10.6 kPa for LSM and 0.4 for APRI. LSM was superior to APRI in discriminating CF with PHTN from other groups (P=0.05). Conclusions: Liver stiffness determined by TE correlates with the presence and severity of LD in children and young adults with CF. LSM increases with age and disease severity. Although APRI provided some information regarding degree of hepatic fibrosis, LSM performed better than APRI in children and adults with CFLD.


Maureen M. Jonas - Advisory Committees or Review Panels: Gilead Sciences; Consulting: Novartis; Grant/Research Support: Bristol Myers Squibb, Roche, Merck Schering Plough

The following people have nothing to disclose: Amal A. Aqul, Sarah Harney, Roshan Raza, Gregory S. Sawicki, Paul D. Mitchell, Rima Fawaz


Assessment of Liver Stiffness in Pediatric Post-Fontan Patients Using Transient Elastography

Becky Chen1, Richard A. Schreiber1, Derek Human2, James E. Potts2, Orlee R. Guttman1;

1Gastroenterology, Hepatology and Nutrition, British Columbia Children's Hospital, Vancouver, BC, Canada; 2Children's Heart Centre, British Columbia Children's Hospital, Vancouver, BC, Canada

Background: Hepatic abnormalities have been well documented among patients post- Fontan surgery for complex congenital heart disease. Hepatic fibrosis (HF) is a recognized complication, though its extent in pediatric Fontan patients is poorly understood as liver biopsy is not routine in these patients. Transient elastography (TE), a well-established noninvasive tool for the assessment of HF in adults, has recently been evaluated in several pediatric populations. There is a paucity of literature examining the role of TE in pediatric post-Fontan patients. Methods: Post-Fontan patients and controls were enrolled through the cardiology and gastroenterology clinics at BC Children's Hospital. Patients with known pre-existing liver disease (e.g., Alagille syndrome) were excluded. Demographic information, hepatic biochemistry, echocardiography and cardiac catheterization data were collected. Patients underwent TE measurements using age- and size-appropriate probes. Results: Nineteen post-Fontan patients (median age 12.2 y, 5.9–16.7) and 19 age-matched controls were studied. The median interval from Fontan to TE measurement was 9.1 y (1–13.5). Nine of the Fontan patients had hepatomegaly and 2 had splenomegaly; none were in cardiac failure. Liver stiffness values were significantly higher in Fontan patients compared with controls (19 kPa (8.1–42.2) vs. 4.9 kPa (2.4–6.1), p<0.0001). GGT was significantly higher in Fontan patients than controls (49 U/L (10–148) versus 11 U/L (8–20), p=0.0003). There was no association between TE values and patient age or time since Fontan. Conclusions: TE is a feasible non-invasive tool for the assessment of liver stiffness in children post-Fontan surgery. Pediatric Fontan patients have markedly elevated liver stiffness suggesting a high risk of advanced HF. TE has potentially important utility in the follow-up of pediatric Fontan patients.


The following people have nothing to disclose: Becky Chen, Richard A. Schreiber, Derek Human, James E. Potts, Orlee R. Guttman


Hepatic Involvement in Langerhans Cell Histiocytosis

Eirini Kyrana1, Juliana Puppi1, Penelope Brock2, Nedim Hadzic1;

1Paediatric Liver, GI and Nutrition Centre, King's College Hospital, London, United Kingdom; 2Paediatric Haematology and Oncology Department, Great Ormond Street Hospital, London, United Kingdom

INTRODUCTION: Langerhans cell histiocytosis (LCH) is a rare, non-malignant disease of unknown aetiology, characterised by extreme heterogeneity and an unpredictable course. Hepatic involvement in LCH is associated with worse prognosis. METHODS: A retrospective review of children with LCH referred to a tertiary paediatric hepatology centre from 1999 to 2012. RESULTS: Ten children (7 girls) with LCH were referred at a median age of 30 months (range 10 - 72). Median age at diagnosis was 17 months (range 8 - 27). Multisystem disease was documented in all but one, where initial information was not available. At presentation, skin rash was found in all and hepatic involvement was noted in 8, including 4 with hepatomegaly and 1 with hepatosplenomegaly. Diagnosis was confirmed by CD1a/S100 immunostaining on skin biopsies. All patients except one underwent a liver biopsy, suggestive of sclerosing cholangitis (SC) in 7 patients, with CD 1a-positive cell infiltrates in 3. Paucity of intrahepatic bile ducts and biliary changes compatible with graft-versus-host disease (GvHD) were the findings in the other 2 biopsies. Systemic chemotherapy was the initial treatment for 8 patients. Two children had a very aggressive LCH and liver disease and died at 23 and 31 months without liver transplant (LT). Six patients with severe SC and progressive portal hypertension (PHT) received LT at median age of 84.5 months (range 13 - 139). All except one were jaundiced (median bilirubin 193 μmol/L range 13 -732). Splenomegaly, thrombocytopenia and oesophageal varices were present in 5, 4 and 3 patients, respectively. Liver explants showed biliary cirrhosis in all. One patient died shortly after LT from fungal sepsis. He also had lung LCH involvement at the time of LT. Two girls had biopsy-proven LCH recurrence in the graft at 6 and 60 months after LT, respectively. One of them had been successfully treated for post-transplant lymphoproliferative disease. They received further chemotherapy which controlled the disease in one, but the other required re-LT for progressive disease 3 years later. Five of the transplanted patients are alive after a median follow up of 14 years (range 0.75 - 18) post LT. One of the patients had no preceding liver involvement and was diagnosed with post-haematopoietic stem cell transplant GvHD. Another patient normalised his liver function tests after chemotherapy. They are both asymptomatic 41 and 56 months after diagnosis, respectively. CONCLUSION: Hepatic involvement is almost universal in multisystem LCH. Progression of PHT often heralds a requirement for LT. This is a viable management option, if the LCH is in prolonged remission.


The following people have nothing to disclose: Eirini Kyrana, Juliana Puppi, Penelope Brock, Nedim Hadzic


Plant Sterol Levels Correlating with Markers of Hepatocyte Injury in Hospitalized Patients Receiving Parenteral Nutrition

Alisha M. Mavis2, T. Hang Nghiem-Rao3, Elizabeth M. Polzin4, Mary F. Firary4, Grzegorz W. Telega2, Shailendra B. Patel1,5;

1 Medical College of Wisconsin, Menomonee Falls, WI; 2Pediatric Gastroenterology, Hepatology, & Nutrition, Medical College of Wisconsin, Milwaukee, WI; 3Neonatology, Medical College of Wisconsin, Milwaukee, WI; 4Children's Hospital of Wisconsin, Milwaukee, WI; 5Clement J. Zabocki Veterans Medical Center, Milwaukee, WI

Background: Parenteral nutrition (PN) is crucial for the survival of children unable to tolerate enteral feeds and those with intestinal failure. PN-associated liver disease (PNALD) is a serious complication of long term PN. The impact of intravenous lipid emulsions on the development of PNALD has been investigated in recent years. Plant sterols in lipid emulsions, such as Intralipid (IL), have been shown to be a significant contributor to cholestasis and may cause direct hepatocyte injury. Very little is known about the sterol metabolism during IL infusions. Also, the relative importance of specific sterols in the development of PNALD is not known. Clarifying the associations between specific plant sterol levels and liver injury may offer opportunities for prevention of PNALD. Objectives: To determine correlations between specific plant sterol levels and markers of hepatocyte injury in hospitalized patients receiving PN. Methods: This is a prospective study that included children (0–18 years) receiving PN during hospitalization. We measured stigmasterol, campesterol, sitosterol, and cholesterol levels in plasma samples obtained in longitudinal fashion while on PN. In addition, we collected laboratory data to determine presence of cholestasis and hepatocellular injury. 80 patients were recruited, 56 patients received PN, 23 patients were controls on full enteral feeds, and 1 was excluded due to a genetic duplication. Results: We identified two distinct patters of sitosterol metabolism: fast metabolizers who maintained sitosterol levels below 0.5, slow metabolizes in whom sitosterol levels exceeded 0.5 and usually continued to increase during the study period. We found significant positive correlation between conjugated hyperbilirubinemia and increased sitosterol levels in infants and older children with congenital bowel disease and/or intestinal surgery. Conclusions: There are 2 distinct types of sitosterol metabolism. Intestinal surgery and/or congenital bowel disease is a risk factor for conjugated hyperbilirubinemia in patients who have decreased sitosterol metabolism.


T. Hang Nghiem-Rao - Grant/Research Support: The Gerber Foundation

The following people have nothing to disclose: Alisha M. Mavis, Elizabeth M. Polzin, Mary F. Firary, Grzegorz W. Telega, Shailendra B. Patel