POSTERS P1069 THE ASSOCIATION BETWEEN FATTY LIVER STATUS CHANGE AND INCIDENCE OF DIABETES MELLITUS IN JAPAN K. Nagai1 , T. Tsuboya2,3 , H. Yamazaki1 , A. Tomonari1 , K. Tsuji1 , H. Maguchi1 . 1 Centre of Gastroenterology, Teine Keijinkai Hospital, Sapporo, Japan; 2 Social and Behavioural Sciences, Harvard School of Public Health, Boston, United States; 3 International and Community Oral Health, Tohoku University Graduate School of Dentistry, Sendai, Japan E-mail:
[email protected] Background and Aims: An association between fatty liver (FL) and diabetes mellitus (DM) has been reported. However, it is still uncertain if a change in fatty liver status affects DM incidence. The aim of our study was to assess the risk of future DM incidence in people whom fatty liver status had changed. Methods: Eligible participants were those who had a health check in 2000, 2001, and at least once between 2002 and 2013. Those who had positive HBs antigen, positive HCV antibody, or DM in 2000 or 2001 were excluded, leading to the eligible 12,904 participants. The participants were divided into 4 groups; (1) “Improved FL” group (n = 234), which had FL in 2000 and no FL in 2001; (2) “Sustained FL” group (n = 1,146), which had FL in both 2000 and 2001; (3) “Newly developed FL” group (n = 760), which had no FL in 2000 but had FL in 2001; (4) “Non-FL group” (n = 10,764), which had no FL both in 2000 and 2001. Fatty liver was diagnosed using ultrasonography. DM was ascertained by fasting plasma glucose ≥126 mg/dL, HbA1c ≥6.5%, self-reported physician-diagnosed DM, or having any medication for DM. The outcome was cumulative incidence of DM during 2002–2013. Logistic regression model was used to estimate crude and multivariate odds ratios (ORs) and confidence interval (CI) for the association between DM incidence and the groups. Age, sex, family history of DM, overweight, hypertension, dyslipidaemia, frequency of exercise and smoking status in 2000 were adjusted in the multivariate model. Results: During the follow-up period, cumulative incidence of DM was as follows: 2/234 (0.85%) in “improved FL”, 35/1,146 (3.05%) in “sustained FL”, 22/760 (2.89%) in “newly developed FL”, and 50/10,764 (0.46%) in “non-FL”. The multivariate ORs were 1.507 (95% CI 0.368–6.351) in “improved FL”, 4.830 (95% CI 2.717– 8.588) in “newly developed FL”, and 5.024 (95% CI 2.977–8.479) in “sustained FL”, compared with “non-FL”. Comparing with “improved FL”, multivariate ORs were 2.571 (95% CI 0.569–11.629) in “newly developed FL”, and 3.492 (95% CI 0.825–14.787) in “sustained FL”. Conclusions: FL was a risk factor of future DM. The association between FL improvement and reduced incidence of DM appeared clinically significant, but was not statistically significant. P1070 THE MARKER OF MACROPHAGE ACTIVATION SOLUBLE CD163 IS ASSOCIATED WITH LIVER HISTOLOGY IN CHILDREN WITH NON-ALCOHOLIC FATTY LIVER DISEASE K. Kazankov1 , H.J. Møller2 , A. Alisi3 , R. De Vito4 , H. Vilstrup1 , V. Nobili3 , H. Grønbæk1 . 1 Department of Hepatology and Gastroenterology, 2 Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark; 3 Hepato-Metabolic Disease Unit and Liver Research Unit, 4 Unit of Pathology, “Bambino Ges` u” Children’s Hospital, IRCCS, Rome, Italy E-mail:
[email protected] Background and Aims: Macrophages are involved in inflammation and fibrosis in non-alcoholic fatty liver disease (NAFLD). Soluble (s)CD163 is a specific marker for macrophage activation and in a previous study, sCD163 was increased in obese children with indices of NAFLD. We aimed to measure sCD163 in children with biopsy-proven NAFLD and associate it with clinical, biochemical and histological parameters of NAFLD. S750
Methods: We investigated 155 children with biopsy-proven NAFLD. Demographic, clinical and biochemical data were recorded. Liver biopsies were scored according to the NAFLD Clinical Research Network criteria. sCD163 was measured by ELISA. Results: Soluble CD163 correlated with the body mass index standard deviation scores (r = 0.21, p = 0.05). sCD163 was associated with hepatocyte ballooning (rho=0.17, p = 0.03), but not steatosis or lobular inflammation. It showed a trend to association with the fibrosis score (rho=0.14, p = 0.09), and children with bridging fibrosis (F=3, n = 14) tended to have higher sCD163 compared to those with lower fibrosis stages [2.4 (IQR 1.7–2.9) vs. 1.8 (IQR 1.5–2.4) mg/L, p = 0.07]. The number of CD163 positive macrophages was associated with scores of steatosis, lobular inflammation and hepatocyte ballooning; however, the cell counts showed no association with sCD163 levels (r = 0.12, p = 0.24). Conclusions: sCD163 reflecting macrophage activation was associated with histological parameters of liver injury and fibrosis, suggesting a role for macrophage activation in pediatric NAFLD. However, there was no association with CD163 immunohistochemistry. P1071 THE ASSOCIATION BETWEEN SERUM LEVELS OF URIC ACID AND ALANINE AMINOTRANSFERASE IN A POPULATION-BASED COHORT L. Rabinowich1 , O. Ben-Assuli2 , M. Green3 , A. Goldstein4 , A. Magid3 , V. Shalev4 , O. Shibolet1 , G. Chodick4 , S. Zelber-Sagi1,3 . 1 The Liver unit, Gastroenterology Department, Tel-Aviv Sourasky Medical Center, Tel Aviv, 2 Faculty of Business administration, Ono Academic College, Kiryat Ono, 3 School of Public Health, University of Haifa, Haifa, 4 MACCABI Institute for Health Services Research/ Medical Division, Maccabi Healthcare Services, Tel Aviv, Israel E-mail:
[email protected] Background and Aims: Elevated serum uric acid is frequently observed in patients with the metabolic syndrome. A strong correlation exists between this syndrome and nonalcoholic fatty liver disease (NAFLD). Therefore, we aimed to test the association between uric acid and elevated alanine aminotransferase (ALT) in a large population-based cohort. Methods: A cross-sectional study using real-world data from a large public health organization in Israel (Maccabi Healthcare System). The population consisted of individuals aged 20–60 years old who underwent blood tests for ALT and uric acid for any reason during 1997–2012. Individuals with secondary liver disease, celiac and inflammatory bowl-disease were excluded. Subgroup analysis was performed in subjects who were diagnosed with fatty liver in the medical records (n = 2,628). This database includes medical history, diagnoses, patient consultations, prescription drug purchase, laboratory and imaging tests. Results: The study population included 82,608 people (32.5% men, mean age 43.91±10.15 years). Subjects in the upper quartiles (>5.6 mg/dL) of uric acid level were significantly (P < 0.001) more likely to have a poorer metabolic profile in terms of glucose, serum lipids, diabetes and hypertension. Categorizing serum uric acid into deciles demonstrated a significant positive dose-response association with the rate of elevated serum ALT (P for trend <0.001). In multivariate logistic regression analysis, controlling for potential confounders, the association between uric acid and elevated ALT persisted both in the total population (OR = 1.18, 95% CI 1.10–1.28; OR = 1.49, 1.38–1.62; OR = 2.10, 1.93–2.29, for the 2nd, 3rd and 4th quartiles respectively compared to the 1st) and in the subgroup of subjects who were diagnosed with fatty liver (OR = 1.77, 1.22–2.57, for the 4th quartile compared to the 1st). With stratification by gender or BMI categories, the association between uric acid and elevated ALT was maintained in all categories.
Journal of Hepatology 2015 vol. 62 | S263–S864
POSTERS Conclusions: Serum uric acid is independently associated with elevated ALT, as a surrogate for NAFLD, and thus may serve as a serum marker and should be further investigated as a risk factor for NAFLD. P1072 THE INTERACTION BETWEEN VISCERAL ADIPOSE AND HEPATIC TISSUE AFFECTS LIVER INJURY IN NAFL AND NASH PATIENTS M. Schlattjan1 , J.-P. Sowa1 , S. Sydor1 , G. Gerken1 , L.P. Bechmann1 , A. Canbay1 . 1 Department of Gastroenterology and Hepatology, University Hospital, University Duisburg-Essen, Essen, Germany E-mail:
[email protected] Background and Aims: Non-alcoholic fatty liver disease (NAFLD) and the more severe form non-alcoholic steatohepatitis (NASH) are the most common liver diseases in western countries. The interaction between liver and adipose tissue is fundamental for the development of NAFLD and NASH. Aim of this study was to investigate the interaction between adipose tissue and liver at the time of bariatric surgery. Methods: Blood, visceral adipose tissue, and liver tissue samples were obtained from 40 (median age 46±8.6 y; 29 w/11 m; BMI 51.3±8.1 kg/m2 ) morbidly obese patients undergoing bariatric surgery. Histopathological assessment of the biopsies was done according to the NAFLD activity score (NAS). Blood samples taken before surgery were analyzed for parameters of liver injury (M30, M65). Hepatic and adipose tissue mRNA levels of one gene for triacylglycerol regulation (CGI-58) and a central regulating gene for fatty acid storage and glucose metabolism (PPARg2) were assessed by qrtPCR. Results: CGI-58 mRNA expression was increased in adipose tissue of morbidly obese patients and showed a strong correlation to low density lipoprotein (LDL) in sera and with the NAS. In morbidly obese individuals mRNA of PPARg2 was significantly upregulated in hepatic and adipose tissue. Moreover, PPARg2 mRNA levels in adipose tissue were correlated significantly with hepatic PPARg2 mRNA levels. The ratio of markers for apoptosis and necrosis (M30, M65) in serum also correlated significantly with mRNA levels of metabolic regulators in adipose tissue. Conclusions: The presented data shows that adipose CGI-58 and PPARg2 mRNA expression is associated on the one hand with mRNA expression in the liver and on the other hand with the grade of liver injury. P1073 INTERLEUKIN 15 IN NONALCOHOLIC FATTY LIVER DISEASE AND OBESITY O. Kurinna1 , G. Fadieienko1 , O. Babak1 , T. Solomentseva1 , K. Syntyk1 . 1 GI National Institute of Therapy of the National academy of medical sciences of Ukraine, Kharkiv, Ukraine E-mail:
[email protected] Background and Aims: Background: Non-alcoholic fatty liver disease (NAFLD) and obesity represent widespread pathologies associated with low-grade inflammatory processes. Experimental data suggest crucial role of anabolic cytokine interleukin 15 (IL-15) in NAFLD development. Y. Cepero-Donates et al. showed that increased secretion of IL-15 promotes fat accumulation in the liver stimulating hepatic inflammatory response in mice. The aim was to investigate IL-15 concentration in patients with NAFLD associated with obesity depending on steatosis degree and anthropological parameters. The study included 32 patients with NAFLD associated with obesity, 31 normal weight patients with NAFLD and 26 normal weight volunteers without hepatic steatosis. Methods: NAFLD diagnosis and assessment were performed by abdominal ultrasound examination. Steatosis degree was evaluated using hepatorenal index (HRI). For all NAFLD patients other causes of hepatic steatosis were excluded. Obesity was measured
using body mass index (BMI) and waist circumference (WC). Concentration of IL-15 were measured using enzyme-linked immunosorbent assay kit. Results: The results showed that IL-15 concentration was significantly increased in patients with NAFLD comparing to control group (p < 0.05). Concentrations of IL-15 observed in NAFLD patients with concomitant obesity were significantly higher than those measured in NAFLD patients with normal weight (p < 0.05). In all NAFLD patients IL-15 correlated with HRI supporting its role in hepatic fat accumulation. Furthermore, in patients with NAFLD there was significant correlations of IL-15 concentration with BMI (p < 0.05) and WC (p < 0.05). Conclusions: Patients with NAFLD and concomitant obesity might have more significant proinflamatory status that could be caused by adipose tissue dysfunction and cytokine synthesis abnormalities. Increased synthesis of IL-15 observed in NAFLD obese patients supports its role in hepatic lipid accumulation found in experimental studies. Further investigations are needed to explore correlations of IL-15 with histological findings in NAFLD obese patients. P1074 NO EVIDENCE FOR PLATELET HYPERACTIVITY IN PATIENTS WITH NON-ALCOHOLIC FATTY LIVER DISEASE W. Potze1 , M.S. Siddiqui2 , S.L. Boyett2 , J. Adelmeijer1 , K. Daita2 , T. Lisman1 , A.J. Sanyal2 . 1 Surgical Research Laboratory, Department of Surgery, University Medical Center Groningen, Groningen, Netherlands; 2 Div. of Gastroenterology, Hepatology and Nutrition, Dept. of Internal Medicine, Virginia Commonwealth University School of Medicine, Richmond, VA, United States E-mail:
[email protected] Background and Aims: The mechanism of increased prevalence of cardiovascular disease in patients with non-alcoholic fatty liver disease (NAFLD) is unknown. Platelet activation has been implicated as a contributor of the increased risk of cardiovascular disease in the metabolic syndrome, but its role in NAFLD is unclear. We, therefore, assessed the platelet activation status in lean and obese individuals in comparison to patients with various histological severities of NAFLD. Methods: Blood was drawn from 68 patients with biopsy-proven NAFLD (simple steatosis n = 24, NASH n = 22, and NASH cirrhosis n = 22), 20 lean controls (BMI <25 kg/m2 ), 20 overweight controls (BMI >25 kg/m2 ), and 15 patients with alcoholic (ASH) cirrhosis. Subjects with congenital coagulation disorders, recent infection (<2 weeks), anticoagulant or anti-platelet therapy, and recent transfusion with blood products were excluded. We studied basal and agonist-induced platelet activation using flow cytometry. In addition, we studied plasma levels of von Willebrand factor (VWF), the VWF-cleaving protease ADAMTS13, and the platelet activation marker soluble P-selectin. Results: Basal platelet activation was comparable between lean and overweight controls, patients with NASH, and patients with NASH or ASH-related cirrhosis. There was only a slight increase in basal platelet activation in patients with simple steatosis compared to overweight controls. Agonist-induced platelet activation was decreased in patients with cirrhosis, most notably in patients with ASH-related cirrhosis, but similar between patients with non-cirrhotic NAFLD and controls. Plasma levels of VWF were increased in patients with NASH or ASH cirrhosis compared to all other groups; however levels were comparable between lean and overweight controls, patients with simple steatosis, and patients with NASH. ADAMTS13 levels were comparable between all patients and controls. Soluble P-selectin levels were mildly elevated in plasma from patients with NASH, NASH cirrhosis, or ASH cirrhosis compared to lean and overweight controls.
Journal of Hepatology 2015 vol. 62 | S263–S864
S751