4.3 g/dl respectively; p=0 0.015). Uric acid level was higher in group 2 and 3 patients compared to patients in group 1 and 2 (p= 0.02). The difference in uric acid level between groups 2 and 3 was not statistically significant but was significant between stage 1 versus stages 2-4 (p<001). Conclusion: Uric acid levels are higher in NAFLD patients with any degree of fibrosis compared to patients with no fibrosis suggesting a possible role of uric acid in development of fibrosis that needs further evaluation. Sa1566 FECAL MICROBIOTA TRANSPLANTATION IN NONALCOHOLIC STEATOHEPATITIS: A CASE SERIES Manida Wungjiranirun, Yesenia Risech-Neyman, Charles Wang, David J. Grand, Colleen R. Kelly, Kittichai Promrat Introduction Nonalcoholic steatohepatitis (NASH) is the most rapidly growing etiology of chronic liver disease and indication for liver transplantation in the United States. Treatment options for NASH are limited. Based on data supporting the role of intestinal dysbiosis in NASH, one treatment approach is to manipulate the gut microbiome through fecal microbiota transplantation. The aim of this pilot study is to evaluate the safety, tolerability and clinical effects of fecal microbiota transplantation from lean donors (FMT-L) in patients with biopsy proven NASH. Methods This is a pilot open labelled study with target enrollment of 5 participants. The primary outcome measure was the change in hepatic fat from baseline to week 12 as quantified by magnetic resonance imaging-estimated proton density fat fraction (MRI-PDFF). Participants were included if they were over the age of 18 with biopsy- proven NASH. Frozen donor stool was obtained from lean healthy unrelated donors without metabolic syndrome (body mass index (BMI)<25 kg/m 2 , hip/waist ratio<0.9 for men and<0.85 for women). FMT-L was administered through upper endoscopy into the distal part of the duodenum/proximal jejunum. Magnetic resonance imaging (MRI) to assess liver fat was obtained at baseline and at least 12 weeks after FMT-L. Blood samples were collected at visits at 0, 2, 4, 8, 12 and 24 weeks to assess liver enzymes, lipid panel, free fatty acids, and markers of insulin sensitivity. Stool samples were collected at each visit and adverse events (AEs) were assessed. Results Five participants were enrolled, one was lost to follow up prior to FMT-L. Four participants received FMT-L, one did not undergo a post FMT MRI due to loss to follow up. Mean age of study participants was 46.6 (range 28-62) years, and mean BMI was 34.4 (range 24.8 to 45.9) kg/m 2 . All had biopsy-proven NASH without cirrhosis within 6 months prior to FMT-L. Among the 3 subjects who underwent FMT-L and completed pre and post MRI-PDFF studies, there was no significant reduction in hepatic steatosis (subject#1 had a pre FMT PDFF of 19.5%, and a post of 20.75 %. Subject#3 had a pre FMT PDFF of 20.25% and a post of 23.0 %. Subject#4 had a pre FMT PDFF of 26.75% and a post of 23.0%). All subjects tolerated the procedure well and there were no serious adverse events observed during the study. There were 2 incidences of moderate AEs that were unrelated to the study intervention. Discussion In this small pilot study, FMT-L appears to be safe and well tolerated in patient with NASH. However, there was no significant improvement in the degree of hepatic steatosis as measured by MRI-PDFF at least 12 weeks after intervention. Liver chemistries and markers of insulin sensitivity did not improve significantly during the follow up period of this study. Microbiome analyses of stool samples are pending. Sa1567 FIBROSCAN PREDICTS FIVE YEAR LIVER OUTCOMES Paul R. Armstrong, Lisa Coffey, Jennifer Russell, Stewart Stephen Introduction Fibroscan (FS) is a non invasive tool to assess liver stiffness (LSM measured in kPa) which correlates with hepatic fibrosis. The clinically relevant cut-offs are>12kPa, which implies a 90% risk of advanced fibrosis and>25kPa, which has a 90% positive predictive value for clinically significant portal hypertension. Aims/Background We performed a retrospective study across liver disease aetiologies to assess whether baseline Fibroscan score associated with liver decompensation and liver related mortality. Method We reviewed all Fibroscans performed in the Liver Centre of the Mater Hospital in 2012 and selected all patients with a score of>35kPa. We then randomly selected similar numbers of patients with FS scores of<21kPa and 21-35kPa to achieve a total number of 100. We then determined liver outcomes at five years post Fibroscan. Results Patients with a baseline LSM>35kPa had a five year decompensation rate of 43% (8.6%/year) and liver related mortality of 20%. This compared to 19% (3.8%/year) and 12% in those with a baseline score of 21-35kPa and 6% (1.2%/year) and 0% in those with a baseline score of<21kPa. Aetiology did not impact on outcome. We found that a FS Score<35kPa has a Negative Predictive Value of 88% regarding risk of decompensation over 5 years, and that a FS Score>35kPa has a Relative Risk of 4 for decompensation over 5 years. Conclusions This retrospective study corroborates our prospective study showing that LSM can be used to determine risk of decompensation and liver mortality in liver disease regardless of aetiology. This finding can be used to prognosticate, ration therapies, and select patients for experimental therapeu- tic studies. S-1237 AASLD Abstracts Sa1568 PRETREATMENT WITH THE PHOSPHODIESTERASE-3 INHIBITOR MILRENONE REDUCES HEPATIC ISCHEMIA–REPERFUSION INJURY, MINIMIZING ZONE 3-BASED LIVER PAREHCHYMAL AND SMALL INTESTINAL INJURY IN RATS Shinichi Nakanuma, Tomoharu Miyashita, Hiroyuki Takamura, Rosuke Gabata, Mitsuyoshi Okazaki, Yoshinao Ohbatake, Isamu Makino, Hironori Hayashi, Koichi Okamoto, Jun Kinoshita, Keishi Nakamura, Hidehiro Tajima, Itasu Ninomiya, Sachio Fushida, Tetsuo Ohta Introduction: We previously reported the case of a liver transplant recipient who experienced liver allograft dysfunction caused by a severe zone 3 (pericentral)-based liver parenchymal injury, sinusoidal obstruction syndrome (SOS), which resulted in graft loss. Hepatic ischemia– reperfusion (I/R) injury is a main cause of graft parenchyma injury after liver transplantation (LT). We evaluated the association between zone 3-based liver parenchymal injury and hepatic I/R injury. The phosphodiesterase-3 inhibitor, Milrinone, had an inhibiting effect on hepatic I/R injury because of a vasodilatory effect and tolerance to endothelial cell damage. We also examined how Milrinone reduced hepatic I/R injury based on the zone 3-based liver parenchymal injury. We also assessed the effects of Milrinone on small intestinal injury after hepatic ischemia. Methods: Rats were randomly divided into three groups: sham, I/ R, and I/R+Milrinone (I/R+M). In the I/R+M group, Milrinone (30 mg/kg) was injected by intraportal administration 5 min before total hepatic ischemia. The hepatoduodenal ligament was clamped. After 30 min of total hepatic ischemia, the clamp was removed to initiate hepatic reperfusion. At 5 h after reperfusion, blood, liver and small intestine specimens were collected. Suzuki ` s criteria and the Park score were used in the histological analysis. Results: Serum aspartate aminotransferase and total bilirubin levels in the I/R+M group were significantly lower compared with the I/R group. Liver tissue assessment indicated that the congestion, cytoplasmic vacuolization, and total scores in zone 3 in the I/R group were significantly elevated compared with that of zone 1 (periportal). The congestion, necrosis, and total score at zone 3 in the I/R+M group significantly improved compared with those at the same zone in the I/R group. Assessment of the small intestine indicated that the Park and congestion scores in the villi of the I/R+M group were significantly decreased compared with those of the I/R group. All rats survived 30 min of total hepatic ischemia. With 45 min of total hepatic ischemia, the 7-day survival rate in the I/R group was 53.8%, while that in the I/R+M group was significantly improved to 92.3%. Conclusion: A hepatic I/R injury model with total hepatic ischemia caused zone 3-based liver parenchymal and small intestinal injuries, which were attenuated by Milrinone. We hypothesized that the vasodilatory effect of Milrinone reduced the congestion in sinusoidal vessels leading to improved blood flow in the liver parenchyma. Subsequently, small intestine congestion was also improved. Hepatocellular injury markers were decreased and the survival rate was increased in rats treated with Milrinone. Hepatic I/R injury may be partially associated with SOS after LT and Milrinone may be effective at preventing this refractory complication. Sa1569 LIVER TRANSPLANT OUTCOMES IN RECIPIENTS WITH ELDERLY DONORS, A SYSTEMATIC REVIEW AND META-ANALYSIS. Faisal Kamal, Muhammad Ali Khan, Arslan Talat, Hafiz Muhammad Sharjeel Arshad, Mahmoud Bayoumi, Raseen Tariq, Ashwani K. Singal, Sanjaya K. Satapathy Background: In this era of decreasing donor availability, elderly donors can possibly expand the donor pool. Several studies assessing transplant outcomes in transplant recipients with elderly donors have reported conflicting results. Aim: To compare transplant outcomes in liver transplant recipients receiving grafts from ≥70 and ≥80 years old donors with recipients receiving grafts from<70 and<80 years old donors respectively. Methods: We searched AASLD Abstracts