High Hepatic Iron and Low Serum Betaine Levels Are Associated With. Hepatic Steatosis Among Males With Chronic Hepatitis B. Asad Javaid, Allison B.
68.3% and 40.4%, respectively. The model for predicting NASH included age, BMI, waistto-hip ratio, ALT, AST, total bilirubin, albumin, HbA1c, HOMA-IR, HDL cholesterol, hematocrit, INR, and platelet count with a cross-validated AUROC of 0.81 (95% CI; 0.77 - 0.85). The specificity, sensitivity, negative predictive value (NPV), and positive predictive value (PPV) were 90.4%, 46.5%, 44.4%, and 91.1%, respectively, and the model correctly classified 60.6% of patients as having NASH. The model for predicting advanced fibrosis included age, BMI, waist-to-hip ratio, hypertension, ALT, AST, alkaline phosphatase, GGT, globulin, albumin, serum insulin, hematocrit, INR, and platelet count with a cross-validated AUROC of 0.82 (95% CI; 0.78 - 0.85). The specificity, sensitivity, NPV, and PPV were 90.2%, 56.3%, 75.2%, and 79.7%, respectively, and the model correctly classified 76.5% of patients as having advanced fibrosis. Conclusions: Routinely available clinical variables can be utilized to determine the likelihood of NASH or advanced fibrosis in diabetic patients with NAFLD. These data can guide clinicians when to refer the diabetic patients to a liver specialist. Sa1055 Use of Insulin and Sulfonylurea Is Associated With Hepatic Fibrosis in Diabetic Patients With Non-Alcoholic Fatty Liver Disease George Boon-Bee Goh, Jaividhya Dasarathy, Mangesh R. Pagadala, Aynur Unalp, Carol Hawkins, Ruth Sargent, Achuthan Sourianarayanane, Rish Pai, Lisa Yerian, Amer Khiyami, Srinivasan Dasarathy, Arthur J. McCullough Background: Type 2 diabetes mellitus (DM) is an established risk factor for advanced fibrosis and the development of NASH in Non-alcoholic fatty liver disease (NAFLD). However, not all diabetic patients with NASH develop advanced fibrosis. We hypothesized that medications taken by these patients may have influenced the development of advanced fibrosis in the setting of NAFLD. Aims: We explored the association between commonly used medications and advanced fibrosis in diabetic patients with NAFLD We also sought to confirm the close relationship between DM and advanced fibrosis in a cohort of biopsy proven NAFLD. Methods: The study cohort included 459 patients with biopsy proven NAFLD being followed up in the liver clinics at the Cleveland Clinic and MetroHealth Medical Centre in Cleveland, OH. Clinical information including demographics, anthropometry, medical history, concomitant medication history, biochemical and histological variables were ascertained. Risk factors for advanced fibrosis were explored, with advanced fibrosis defined as stage 3 or 4 fibrosis. We also compared the baseline characteristics of patients with and without DM. Risk factors for advanced fibrosis among the DM patients, specifically commonly used medications that were taken concomitantly were evaluated. Results: Age, presence of DM, total cholesterol, low density lipoprotein cholesterol and ballooning were independent risk factors for advanced fibrosis in NAFLD patients. Compared to non-DM patients, DM patients were older (52.16 vs. 46.03 years; p=0.000), with greater proportion of female gender (69.1 vs. 52.9%; p= 0.000), higher BMI (37.01 vs. 35.04 kg/m2; p=0.014) and higher prevalence of concomitant hypertension (70.9 vs. 43.1%; p=0.000) DM patients also had more lobular inflammation, ballooning and advanced fibrosis compared to non-DM patients. Amongst the DM patients, age (OR 1.091; 95%CI 1.039-1.146, p=0.000) and grade of ballooning (OR 5.586; 95%CI 2.687-11.611, p=0.000) had a positive relationship with advanced fibrosis. In addition, concomitant use of insulin (OR 4.950; 95%CI 1.647-14.880, p=0.004) and sulfonylurea (OR 5.070; 95%CI 1.870-13.745, p=0.001) were positively associated while statin use (OR 0.305; 95%CI 0.120-0.776, p=0.013) was negatively associated with advanced fibrosis. Conclusion: Presence of DM is a significant risk factor for advanced fibrosis in NAFLD. Amongst DM patients with NAFLD, the prevalence of advanced fibrosis was higher in patients treated with insulin and sulfonylurea, and was lower in patients treated with statins. These findings may provide support for statin use in diabetic patients with NAFLD while limiting the use of insulin and sulfonylurea as preferred hypoglycaemic agents in these patients.
AASLD Abstracts
Pooled fibrosis progression rate in patients with nonalcoholic steatohepatitis, and baseline stage 0 fibrosis. Effect size represents the annual rate of progression of fibrosis stage. Sa1053 High Hepatic Iron and Low Serum Betaine Levels Are Associated With Hepatic Steatosis Among Males With Chronic Hepatitis B Asad Javaid, Allison B. Goldfine, Mary-Elizabeth Patti, Ping Ping Kuang, Peymei Wu, Teodoro Bottiglieri, Imad Nasser, Daryl Lau Nonalcoholic fatty liver disease (NAFLD) is an important cause of liver disease. The prevalence of hepatic steatosis in chronic hepatitis B (CHB) is not well understood. There are reports that low betaine level contributes to hepatic steatosis by increasing oxidative stress. Our aims are to determine the prevalence and impact of hepatic steatosis on CHB and the role of betaine in this disease state. In a single center, CHB patients with pretreatment liver biopsy between 1998 and 2013 were reviewed. The analysis excluded patients with alcoholism (>50g/day), HCV, HDV or HIV coinfection, other liver diseases and liver transplantation. To date, 231 patients (75% Asians) met the inclusion criteria. Serum betaine levels were measured in 40 randomly selected serum samples from patients with and without steatosis in both genders. Results: Total 127 (55%) pts had hepatic steatosis in biopsied tissue. Steatosis was more common in males (91/136, 67%) than females (36/95, 38%) [p=0.0001]. Among those with steatosis, males (mean 41.5 yrs) were younger than females (mean 48.5 yrs) [p=0.02]. For both genders, those with steatosis had higher rate of severe hepatic inflammation (M 63%, F 30%) compared to those without (M 5%, F 3%) [p=0.0001]. Advanced fibrosis (Stage 3+4) was more frequent in males (14%) than females (8%). Hepatic iron was present in 47 of 91 (52%) males with steatosis but in only 7 of 36 (19%) females with steatosis [p=0.0001]. Females with steatosis had higher rates of elevated LDL cholesterol (64% vs. 41%) [p=0.04], triglyceride (31% vs.13%) [p=0.02] and increased body weight (150lbs vs. 126lbs) [p=0.001] compared to those without. The lipid profiles and weight, however, were similar in males with and without steatosis. Interestingly, males with steatosis (N=10) had lower serum level of betaine compared those without steatosis (N=10) [28.3 vs. 43.3 umol/L; p=0.056]. In contrast, females with (N=10) or without steatosis (N=10) had similar betaine levels [42.9 vs. 39.3umol/L; p=0.3]. Viral factors such as serum HBV DNA levels, HBeAg status were not different between those with and without steatosis. Conclusions: High prevalence of hepatic steatosis is identified in this predominantly Asian cohort of patients with chronic hepatitis B. Steatosis is more common in males. For both genders, steatosis is associated with more severe hepatic inflammation. There appears to be a gender difference in predisposition to steatosis. In females, metabolic syndrome is a major contributing factor. In males, presence of hepatic iron and lower betaine levels correlate with steatosis and inflammation. It is possible that betaine attenuates iron-iduced oxidative stress in the liver. The gender differences in steatosis predisposition in CHB warrant further evaluation. Its understanding likely will contribute to better preventive and therapeutic strategies.
Sa1056 Association of Nonalcoholic Fatty Liver Disease With Liver Malignancies Perla O. Schulz, Andrea Vieira, Maria de Fátima A. Nascimento, Fabio G. Ferreira, Mauricio A. Ribeiro, Luiz A. Szutan Background: The prevalence of nonalcoholic fatty liver disease (NAFLD) has been increasing worldwide. The relation among non-alcoholic steatohepatitis (NASH) cirrhosis and hepatocellular carcinoma (HCC) or intrahepatic cholangiocarcinoma (IHCC) has been reported, but it is still unknown if such associations exist in non-cirrhotic patients. Furthermore, it has been suggested that early stages of hepatic steatosis can already be considered a favorable microenvironment for the development of liver metastases of colorectal neoplasms (LMCN). Aims: To evaluate the association of NAFLD with primary and secondary malignancies of the liver and to compare the prevalence of this association in different tumor types. Methods: This was a retrospective study based on data from the anatomopathology records of 180 patients with hepatic neoplasms treated in a referral center of Brazil, from January 2007 to December 2011. Were excluded 60 patients due to previously known liver disease (viral hepatitis, alcohol abuse, autoimmune hepatitis, hemochromatosis and Wilson's disease), or absence of liver tissue free of tumor at histology material. We divided the remaining 120 cases in groups by type of cancer and a comparison was made among them, for each histological type. The histological samples were evaluated for the presence of hepatic steatosis, NASH and liver fibrosis. The presence of risk factors for NAFLD (glucose intolerance and/ or diabetes mellitus, dyslipidemia, arterial hypertension and overweight) was checked and the association of these predictors with the histological findings was quantified. Institution Ethic Committee approved the study and the statistical analysis was performed with the Statistical Package for Social Sciences (SPSS) version 13.0 and Epi Info version 3.4.3. The significance level for all tests was 5% (p < 0,05). Results: Both groups of each liver tumor were comparable regarding age and gender (Table 1). There was no difference in the association of hepatic steatosis with the general population (34.2% and 30% respectively, 95% CI: 25.8 to 43.4). The hepatic steatosis was more prevalent in patients with LMCN while the prevalence of liver fibrosis was higher in the HCC group (Table 2). No case of NASH in combination with any of the three tumor types was observed. Regarding the relationship of hepatic steatosis, NASH and liver fibrosis with risk factors, no association was found statistically significant in any of the tumors studied. Conclusion: NAFLD is
Sa1054 Clinical Model for NASH or Advanced Fibrosis in Patients With Diabetes and NAFLD Jessica Bazick, Michele Donithan, Brent A. Neuschwander-Tetri, David E. Kleiner, Elizabeth Brunt, Laura Wilson, Edward Doo, Joel E. Lavine, Rohit Loomba Background and aims: Approximately 18 million people in the United States have co-existing type 2 diabetes with NAFLD. It is not known who among these diabetic NAFLD patients has nonalcoholic steatohepatitis (NASH) or advanced fibrosis. Therefore, we aimed to determine factors that are associated with NASH or advanced fibrosis in diabetic patients with NAFLD in order to identify who should be prioritized for a liver biopsy or referred to a hepatologist for diagnostic evaluation and treatment. Methods: This prospective cohort study was derived from the NASH Clinical Research Network studies and included 1249 patients with biopsyproven NAFLD (including 441 with type 2 diabetes as defined by the American Diabetes Association Criteria). Two clinical models for presence of NASH and advanced fibrosis (stage ≥3) were developed with multiple logistic regression and cross-validated jack-knife methodology to obtain adjusted area under the receiver operator characteristic curves (AUROC) and their 95% confidence interval (CI). Results: The mean (±standard deviation) age and body-mass-index (BMI) of patients with diabetes and NAFLD was 52.4 ± 10.5 years and 35.8 ± 6.6 kg/m2, respectively. The prevalence of NASH and advanced fibrosis was
AASLD Abstracts
S-948
associated with liver metastases from colorectal cancer, but not with the other liver tumor types. Table 1 - Comparison of demographic group in accordance with the neoplasm found.
Compared to the 19 control subjects (Table), cases had more prior exposure to methotrexate. PNPLA3 genotyping revealed mutations in 75% of the cases. Conclusion: We describe 8 patients who developed features of NAFLD during anti-TNF- α therapy, most of whom had not gained weight, and whose ALT levels normalized when anti-TNF- α treatment was stopped. Compared to patients on anti-TNF- α who did not develop liver problems, our cases had no difference in risk factors for NAFLD, although the cases had more PNPLA3 gene mutations and more prior exposure to methotrexate. The mechanism underlying this previously unrecognized and seemingly paradoxical effect of anti-TNF- α treatment is not clear.
AASLD Abstracts
LMNCN - liver metastases of non-colorrectal neoplasms; LMCN- liver metastases of colorrectal neoplasms; HCC - hepatocellular carcinoma; IHCC - intrahepatic cholangiocarcinoma; Others - lymphoproliferative tumors and sarcomas; ♂ - Male; N -number of cases. Table 2 - Prevalence of steatosis and hepatic fibrosis according to the study group.
Sa1057 Safety and Effect of Bariatric Surgery in Patients With Known Cirrhosis ± Portal Hypertension: A Single Center Experience Laura Pestana, Ross A. Dierkhising, Michael L. Kendrick, Kymberly D. Watt Background: Data on outcomes of bariatric surgery in cirrhotic patients is limited. Aim: To assess safety and metabolic or liver related outcomes after bariatric surgery in patients with known cirrhosis ± portal hypertension. Methods: 14 patients with known Child's A cirrhosis ± portal hypertension with 6 months to 2 years followup were included (11 sleeve gastrectomy, 3 gastric bypass). Results: Mean age 55.5 years, 10/14 female. 14/14 with NASH (2 with HCV and NASH). 3 patients had portal hypertension on EGD (2 portal gastropathy, 1 small varices). Mean weight decreased from 125 ± 18 kg pre-op to 94 ± 17kg and 80 ± 38kg by 1 (p30% steatosis). Bilirubin increased above 2mg/dL in 1 patient after 1 year post-op. One patient developed encephalopathy 3 years post-op. No patients developed post-operative bleeding or surgical complications. Conclusion: Bariatric surgery in patients with well compensated cirrhosis is well tolerated with beneficial effects of weight loss and metabolic syndrome improvement as well as reduced hepatic steatosis. No bleeding complications were noted in the few patients with mild endoscopic portal hypertension.
Sa1059 Low Rates of Weight Loss in Nonalcoholic Fatty Liver Disease: The Role of Physician Counseling Anwar Dudekula, Vikrant Rachakonda, Beebijan Shaik, Jaideep Behari Background: Nonalcoholic fatty liver disease (NALFD) is a leading cause of chronic liver disease worldwide. Although multiple conditions are associated with the development NAFLD, obesity and insulin resistance are the strongest risk factors for the disease. Weight loss improves aminotransferases and histologic features of NAFLD; however, maintenance of weight loss outside of investigational protocols is difficult to achieve. To date, no studies have assessed the effectiveness of weight loss recommendations for NAFLD outside of clinical trials. Aims: The primary goal of this study was to characterize patterns of long-term weight loss in an ambulatory cohort of patients with NAFLD. In addition, we determined clinical predictors of weight loss in NAFLD. Methods: We retrospectively reviewed clinical records of patients with NAFLD and without cirrhosis presenting to a large, tertiary liver center for outpatient care from May 1st 2007 to December 31st 2012. Demographics, body mass index, follow-up duration, and clinical factors associated with obesity were collected. Percent weight change between the first and last clinical visits was calculated. Logistic regression analysis was used to determine predictors of successful weight loss. Results: 924 patients met inclusion criteria. The mean baseline BMI was 33.3 ± 6.6 kg/m2, and the mean followup duration was 17.3 ± 17.6 months. The majority of patients with NAFLD were in either overweight (26.1%) or class I obesity (30.5%) categories at baseline, while the prevalence of underweight and class III obesity was significantly lower (0.2 % and 15.4%, respectively). Overall, there was no change in mean weight or BMI during the follow-up period. A majority (610 patients; 66%) experienced no weight change, while 183 patients (19.8%) lost at least 5% body weight during the follow up period. Factors associated with successful weight loss included number of clinic visits (OR = 3.38 for 3 visits) and baseline BMI (OR = 1.12). Conclusion: Mildly obese patients are at high risk for the development of NAFLD, and weight loss is largely unsuccessful in ambulatory patients with this condition. Frequent clinical encounters are associated with enhanced weight reduction, thus highlighting the role of physician interaction in this group. Future studies are required to define effective weight loss strategies in patients with NAFLD.
Sa1058 Non-Alcoholic Fatty Liver Disease: A Potential Consequence of Anti-TNF-α Therapy Linda A. Feagins, Avegail Flores, Cristina Arriens, Christina Park, Andreas Reimold, Geri Brown Background: Non-alcoholic fatty liver disease (NAFLD) is strongly associated with metabolic syndrome. Studies suggest that inflammatory cytokines, including tumor necrosis factor (TNF)-α, contribute to the development of that syndrome, and TNF- α has been proposed as a potential therapeutic target for preventing metabolic syndrome and its consequences. However, we have observed a number of patients who developed abnormal ALT levels during treatment with anti-TNF-α agents, who had liver biopsies showing NAFLD, and whose ALTs normalized when anti-TNF-α therapy was stopped. Available reports of patients who developed abnormal liver tests during anti-TNF- α therapy have described liver biopsies showing features of autoimmune hepatitis or drug-induced hepatotoxicity. In this study, we investigated potential risk factors for the development of NAFLD during anti-TNF- αtherapy. Methods: In this case-control study, we reviewed medical records of patients in our inflammatory bowel disease and rheumatology clinics to identify those who developed abnormal ALTs during anti-TNF-α therapy that normalized when that therapy was stopped, and whose liver biopsies showed NAFLD [steatohepatitis (NASH) or steatosis (NAFL)]. We also identified 3 control groups: 1) patients on anti-TNF treatment, but with normal liver tests, matched to cases based on race, gender, age, and BMI, 2) patients with biopsy-proven NASH but no apparent inflammatory disease, and 3) patients without diabetes or liver disease. Data on demographics, lipid panels, and clinical features of metabolic syndrome were recorded for all patients. Genotyping was performed for I148M PNPLA3, a polymorphism known to predispose to NASH, on peripheral blood. Results: 8 cases (4 rheumatoid arthritis, 1 ankylosing spondylitis, 2 psoriatic arthritis, 1 Crohn's) had liver biopsies showing NASH in 5 and NAFL in 3; anti-TNF-α therapy included infliximab (3), adalimumab (2), and etanercept (3). Peak ALTs (mean 110 U/L) were observed as early as 1 month after initiation of antiTNF-α (mean 12 months). 5 of the 8 cases had not gained weight during anti-TNF- α therapy. When anti-TNF-α was stopped, liver tests normalized within 2 to 8 months.
Sa1060 Do Types of Fatty Liver Disease Influence the Susceptibility of Diabetes Mellitus, Hypertension, and Dyslipidemia? Nobuyuki Toshikuni, Atsushi Fukumura, Tomiyasu Arisawa, Mikihiro Tsutsumi [Background] Alcoholic liver disease (ALD) and nonalcoholic fatty liver disease (NAFLD) are two major types of fatty liver disease (FLD). The relationships between FLD types and diabetes mellitus (DM), hypertension (HT), and dyslipidemia (DL) have not been fully understood. Our objective was to clarify whether FLD types influence the susceptibility of these diseases. [Patients and Methods] Between 2009 and 2011, we conducted a nationwide survey of Japanese subjects to evaluate the status of FLD; Data were obtained from 1020
S-949
AASLD Abstracts
