2. laboratory measures including serum levels of glucose, insulin, triglycerides, cholesterol, low and high density lipids, vitamin D; 3. demographic/clinical information including age, gender, ethnicity, income, smoking history, and frequency of exercise; diagnoses: hyperten- sion (HTN), diabetes mellitus (DM). Statistical analyses were conducted with SUDAAN 10.0. RESULTS: A total of 9,217 adult NHANES III participants were included of whom 2,325 had NAFLD and 6,892 were non-NAFLD controls. NAFLD was independently associated with older age, male gender, Hispanic ethnicity, BMI, DM, hypertension and dyslipidemia. Anthropometric measurements correlated strongly with each other and were also independ- ently associated with risk of NAFLD (P<0.01). Only metabolic conditions were found to be associated with overall mortality in NAFLD. Furthermore, the diagnosis of NAFLD and body fat (%) were independently associated with liver-related mortality (P=0.02). Body fat percentage was also found to be associated with higher risk of liver-related mortality among women (aHR=1.06, 95% CI: 1.03-1.09). CONCLUSIONS: Clinical measures of obesity including %BF are independently associated with the diagnosis of NAFLD and with liver- specific related mortality among individuals with NAFLD. Tu1044 Soluble CD14 are a Useful Hallmark of Liver Inflammation and Predictor for Progression of Fibrosis in NASH Yuji Ogawa, Kento Imajo, Masato Yoneda, Yoshiyasu Shinohara, Shingo Kato, Hironori Mawatari, Yuichi Nozaki, Koji Fujita, Wataru Shibata, Hiroyuki Kirikoshi, Koichiro Wada, Atsushi Nakajima Introduction: Presence of liver inflammation seems to be one of risk factors for progression of fibrosis in nonalcoholic steatohepatitis (NASH) (Argo CK, et al., J. Hepatol. 2009). We think that detecting the grade of liver inflammation may lead to predict the progression of NASH in future. At present, liver biopsy is the only method for diagnosis of liver inflammation, emphasizing the clinical importance of developing the biomarker for liver inflammation in NASH. We recently showed that overexpression of CD14, which serves as receptors for lipopolysaccharide (LPS) and regulates responsivity to LPS, in Kupffer cells is possible to trigger the progression of NASH via liver inflammation induced by increased responsivity to low-dose LPS (AASLD 2011). CD14 presents as a soluble protein in blood (sCD14) and its blood concentration is correlated with expression levels of membrane CD14 and reflects inflammation in liver. The aim of this study was to investigate, in humans, the relationship of sCD14 with histological findings in NASH. Methods: Our cohort consisted of 54 patients with nonalcoholic steatohepatitis (NASH), 30 patients with simple steatosis (SS) proven by liver biopsy and 15 age-matched healthy controls. Serum sCD14 levels in patients with NAFLD and healthy controls were measured using Enzyme-linked immunosorbent assay (ELIZA) kit. Results: Serum sCD14 levels were significantly increased in patients with NASH compared with both patients with SS and healthy controls. An association between serum sCD14 levels and the grade of liver inflammation or fibrosis stage was reflected in the significantly positive correlation present between these two variables in the patients with NAFLD (r=0.355, P<0.01, r=0.277, P<0.05, respectively), whereas no association between serum sCD14 levels and the grade of steatosis or ballooning degeneration. We also found that serum sCD14 levels in patients with NAFLD were associated with the development of fibrosis in NASH during 2 years. Conclusion: Our study indicated that serum sCD14 levels were significantly increased according to the development of liver inflammation. Serum sCD14 is a promising useful tool as the predictor of eventual histological progression and potentially a therapeutic target in NASH. Tu1045 Acoustic Radiation Force Impulse Imaging Elastography as a Predictor of Significant Fibrosis in Patients With Chronic Liver Disease Pranav K. Mandal, Usha Goenka, Ashish K. Jha, Enam M. Khan, Sanjeev Kumar, Mahesh Goenka Introduction: Fibroscan is emerging as a useful non invasive technique to asses liver histology. Acoustic Radiation Force Impulse Imaging (ARFI) elastography is one of the newer modalities of performing liver elastography, which studies the localized mechanical properties of the liver tissue by measuring tissue displacement secondary to shear wave pulses. Aim of the study was to compare ARFI elastography with stage of fibrosis in patients with chronic liver disease (CLD) Methods: In this prospective study, 110 patients (n = 110) of CLD with varied etiology were recruited. Sixty of these patients had decompensated CLD, while 50 without any features of decompensation underwent liver biopsy. Liver histology was evaluated for stage of fibrosis using Metavir scoring system. Those patients who had decompensated CLD were assumed to be having F4 fibrosis. Results: There were 29 patients with chronic hepatitis B, 20 with ethanol related CLD, 18 with NAFLD, 5 with chronic hepatitis C, 4 with autoimmune hepatitis, 2 with drug induced CLD, while 32 patients were of unknown etiology. Of the total 110 patients; F0, F1, F2, F3 and F4 fibrosis was diagnosed in 18, 10, 7, 6 and 69 patients respectively. Mean ARFI value in F0, F1, F2, F3 and F4 fibrosis was 1.18 m/s, 1.34 m/s, 1.92 m/s, 2.12 m/s and 2.77 m/s respectively. Thus, there was a trend towards increasing ARFI value with increasing stage of fibrosis. Conclusion: ARFI elastography appears to be a promising non invasive alternative to liver biopsy to assess the stage of fibrosis in patients with CLD. Tu1046 Non-Alcoholic Fatty Liver Disease (NAFLD) With Type 2 Diabetes (T2D) Have the Lowest Level of Physical Activity Alita Mishra, Munkhzul Otgonsuren, Carey Escheik, Maria Stepanova, Lynn Gerber, Zobair M. Younossi BACKGROUND AND AIM: Type 2 Diabetes (T2D) is significantly associated with NAFLD. High intensity exercise improves metabolic status and potentially, NAFLD. This study com- pares the quantity and intensity of physical activity using accelerometer data, for patients with NAFLD with and without T2D. METHODS: Data were obtained from two continuous S-1019 AASLD Abstracts cycles of National Health and Nutrition Examination Survey (NHANES 2003-2004 and 2005- 2006). Extensive clinico-demographic, laboratory and activity counts from accelerometer readings (ActiGraph, Fort Walton Beach, FL) were available. For each participant, activity counts from 4 or more consecutive days were used to assess overall activity levels (counts/ min/d) and time spent in different levels of activity (sedentary, <100 cnt/min; light, 100- 2019 cnt/min; moderate, 2020-5999 cnt/min, and vigorous, 6000+ cnt/min). NAFLD was defined as a fatty liver index (FLI)> 60 in the absence of all other causes of chronic liver disease (e.g. HCV antibody, HBV surface antigen negative, normal transferrin saturation and alcohol consumption<20 gram/day). NAFLD patients with T2D were compared to controls using stratum-specific chi-square and t-tests. Simple linear regression analyses (with Taylor series linearized variance estimation and weighting) were used to determine the association between physical activity levels and NAFLD participants and were adjusted for age, gender, race/ethnicity, smoking status and education. P-value<0.05 or less were considered signific- ant. Statistical analyses were conducted using SUDAAN version 10.1 and SAS version 9.1. RESULTS: The study cohort included 4,565 NHANES (2003-2006) participants. Of these, 23% (N=1,065) met the study criteria for NAFLD and 3,500 were considered as controls. NAFLD patients were older and more obese (P<0.05). About 30% (N=320) of the NAFLD participants had T2D or fasting blood glucose of >126 mg/dL. NAFLD subjects with T2D had 80 ct/min/day less physical activity than controls (P<0.01). Furthermore, they spent less time participating in activity of any level (P<0.01) than controls. About 50 % of NAFLD with T2D had lower average physical activity and moderate-to-vigorous physical activity level than all other groups. A negative association determined by linear regression models was noted between physical activity levels and NAFLD subgroup (P<0.01), but the magnitude of the association was strongest for NAFLD with T2D -49.33[(95% CI:-63.02; -35.64), p<0.01)]. Furthermore, the lower intensity physical activity levels were associated with higher BMI, waist circumference, and sum of skinfolds; especially NAFLD patients with T2D. CONCLUSIONS: Diabetic NAFLD patients have the lowest level of physical activity. These data suggest the importance of developing activity programs for this group of patients. Tu1047 Predictive Value of ALT for Diagnosing Non-Alcoholic Steatohepatitis (NASH) and Advanced Fibrosis in Patients With Non-Alcoholic Fatty Liver Disease (NAFLD) Siddharth Verma, Prashant Sharma, John Hart, Smruti R. Mohanty INTRODUCTION: Traditionally, radiological imaging and elevated ALT have been utilized in the work-up and diagnosis of NAFLD. However, patients with normal ALT could also have progressive form of NAFLD, such as NASH or cirrhosis. AIM: To investigate the best cut off value of ALT from the area under receiver operating characteristics curve (AUROC) in predicting sensitivity and specificity for NASH and advanced fibrosis in our ethnically diverse patient population. METHODS: A retrospective cohort study of patients from Univer- sity of Chicago Medical Center pathology database (June 1, 1995 to June 30, 2005) with biopsy proven NAFLD were selected. Patients with other chronic liver diseases including liver transplant were excluded. Computerized medical records were utilized to obtain demo- graphic, clinical, and laboratory data, including ALT levels. Normal ALT was defined as < 35 IU/L while elevated ALT was > 35 IU/L. All liver biopsy were scored by a single pathologist. Using the NASH CRN scoring system, NAS score > 5 was considered NASH, and fibrosis score >2 was considered a marker of significant fibrosis. RESULTS: Two hundred twenty two patients met biopsy proven criteria for NAFLD and had documented ALT levels. Fifty six (23%) had normal ALT levels. Compared to elevated ALT group, NAS score > 5 was lower in normal ALT group (10.7% vs. 28.9 %; p< 0.01), while fibrosis score >2 was slightly higher in normal ALT group (26.8% v 18.1%; p=0.1). For NAS score > 5, sensitivity were 88.9%, 88.9-50%, and 50% for ALT < 35, >35-70, and >70, respectively, while specificity were 28.6%, 28.6-60.7%, and 60.7% for ALT <35, >35-70, and >70, respectively. For fibrosis score > 2, sensitivity were 68.9%, 68.9-40%, and 40% for ALT <35, >35-70, and >70, respectively, while specificity were 22.6%, 22.6-57.6%, and 57.6% for ALT <35, >35- 70, and >70, respectively. AUROC were 0.6248 for NAS score > 5 and 0.464 for fibrosis score >2. On multivariate analysis, ferritin per 100 ng/ml increase (P=0.01; OR 1.2; 95% CI 1.03-1.34), Mallory bodies (P=0.02; OR 3.17; 95% CI 1.16-8.64), and ballooning (P <0.01; OR 17.6; 95% CI 6.0-51.7) were independently associated with NASH, while Mallory bodies (P=0.04; OR 2.6; 95% CI 1.03-6.42), ballooning (P<0.01; OR 17.3; 95% CI 6.8- 43.8) and inflammation (P<0.01; OR 36.4; 95% CI 4.0-330.7) were significantly associated with fibrosis in all patients. CONCLUSION: Our ROC curve analysis does not show an optimal ALT threshold in predicting NASH or advanced fibrosis. Therefore, ALT levels should not be used alone in determining whether NAFLD patients need liver biopsy to confirm NASH and advanced fibrosis. Hence, metabolic risk factors for NAFLD in predicting NASH and advanced fibrosis may be utilized in deciding to pursue a liver biopsy. AASLD Abstracts