704 Development and Validation of a New Histological Score for Pediatric Nonalcoholic Fatty Liver Disease (NAFLD)

706 Figure 2: Cumulative probability of graft failure within 5 years in pediatric recipients of pediatric donor livers 704 Development and Validation of a New Histological Score for Pediatric Nonalcoholic Fatty Liver Disease (NAFLD) Naim Alkhouri, Rita De Vito, Anna Alisi, Rocio Lopez, Ariel E. Feldstein, Valerio Nobili

Background: Prolonged total parenteral nutrition (PN) may lead to cholestasis and liver disease (PNALD). The soybean oil-based lipid emulsion (Intralipid) and its constituent phytosterols have been implicated in PNALD. Phytosterols may induce cholestasis by antagonism of the nuclear bile-acid receptor, FXR, leading to increased synthesis and suppressed hepatocyte export of bile acids into bile. Clinical studies in PNALD patients show that the fish oil-based lipid emulsion (Omegaven), which is essentially devoid of phytosterols, can reverse cholestatic markers. We investigated the serum and hepatic tissue bile acid, phytosterol levels, and FXR target genes involved in bile acid homeostasis in TPN-fed preterm piglets given three different lipid emulsions. Methods: Preterm piglets were assigned to receive 14 days of either, 1) Enteral diet (EN); 2) TPN + Intralipid (100% soybean oil)(IL); 3) TPN + Omegaven (100% fish oil)(OV); or 4) TPN + SMOF (30% soybean, 30% coconut, 25% olive, and 15% fish oil). Results: Serum total bilirubin, bile acids, and GGT were markedly higher (P<0.05) in IL vs. EN pigs, but were lower in OV and SMOF vs. IL. Liver tissue bile acid levels were higher (P<0.05) in IL and OV pigs vs. EN and SMOF pigs. However, IL pigs tended to have a higher liver tissue bile acid level compared to OV pigs. Serum total phytosterol (beta-sitosterol, campesterol, stigmasterol) levels were significantly higher in IL (65 uM) vs. SMOF (21 uM), but barely detectable in OV (1.7 uM) pigs, reflecting the respective composition in the emulsions. These findings support the hypothesis that phytosterolemia leads to disruption of bile acid homeostasis and cholestasis, but this effect may be dose-dependent. FXR, CYP7A1, and BSEP expression were lower (P<0.05) and CYP27A1 expression tended to be lower in the three TPN groups vs. the EN group. However, among the TPN groups, CYP7A1 expression was also lowest (P<0.05) in OV pigs vs. the IL and SMOF pigs. SHP expression was not different among the groups. Importantly, OSTα expression was 6-fold higher (P<0.05) in the IL but marginally increased in OV and SMOF vs. EN pigs. In contrast, NTCP expression tended to be lower in all TPN groups vs. the EN group. These findings suggest that TPN decreases FXR-dependent synthesis and export of hepatocyte bile acids into bile. The net result of TPN is greater accumulation of hepatocyte bile acids in IL vs. OV and SMOF, which triggers a compensatory up regulation of bile acid export into the systemic circulation and biliary injury. Conclusions: Phytosterolemia induced by lipid emulsions produces a dose-dependent cholestatic phenotype, which is most severe with Intralipid. FXR and its target genes are suppressed by all three parenteral lipid emulsions, suggesting that the beneficial effects fish oil may occur via post-translational mechanisms or FXR-independent pathways.

Since the development of the original NAFLD activity score (NAS), it became evident that pediatric NAFLD has a distinct histopathological pattern including the presence of portal inflammation (PI). The aim of this study was to develop and validate a new grading score for Pediatric NAFLD that takes into account the presence of PI and the weight of each histological feature. Methods: Our training set consisted of 203 pediatric patients with biopsy-proven NAFLD. The diagnosis of nonalcoholic steatohepatitis (NASH) versus not NASH was based on Brunt criteria. The individual histological features of NAFLD were scored as follows: steatosis (0-3), lobular inflammation (0-3), ballooning (0-2), and portal inflammation (0-2; 0= no PI, 1= mild PI, 2= more than mild). Logistic regression analysis was performed to apply weight to each histological feature (to improve correlation with NASH). The new score was called the Pediatric NAFLD Histological Score or PNHS. A p value < 0.05 was considered statistically significant. The validation set consisted of 100 children with biopsy-proven NAFLD. Results: The mean age of the initial cohort was 12.4 ± 3.4 years, the median BMI percentile was 97% (94%-98%), and significant fibrosis (fibrosis stage ≥ 2) was present in 26 patients (12.8%). NASH was diagnosed in 135 patients with a mean NAS of 4.5 ± 1.4. The mean PNHS in the NASH group was 89 ± 20.5 compared to 21.9 ± 24.5 in not NASH group, p< 0.001. PNHS correlated with the presence of NASH according to pathologist diagnosis better than NAS, p= 0.011. The area under the ROC curve (AUC) for diagnosis of NASH was 0.96 for PNHS. Similar findings were noted in the validation set with AUC of 0.94. Conclusion: PNHS may be used for histological grading of pediatric NAFLD with better correlation with the presence of NASH than NAS. The utility of this score in other populations and in therapeutic trials should be further investigated. 705 Prevalence of Suspected NAFLD is Increasing Among US Adolescents Miriam B. Vos, Jean Welsh Introduction: Nonalcoholic fatty liver disease is the most common liver disease in children and screening for this disease is an important public health challenge. Recent studies have suggested that lower cutoffs should be utilized in assessing the prevalence of NAFLD. We applied established and newly recommended, lower, gender-specific cutoffs to examine trends in prevalence of NAFLD over the past 2 decades. Methods: We utilized nationally representative data in 12-18 year olds from the National Health and Examination Survey datasets spanning 1988 to 2008. Subjects were excluded for missing data and known liver disease. Suspected NAFLD was defined as overweight/obese (BMI≥85th percentile) and elevated ALT (ALT > 25.8 for boys or ALT > 22.1 for girls) and also as overweight/obese and ALT>30 to allow for comparisons with earlier studies. Results: In total, 10,359 adolescents were included in the analysis. Age, sex, race and % overweight did not differ over time. Percent of adolescents who were obese increased from 11.2 ± 1.1 to 20.6 ± 2.1% (p for trend <.0001). Among all adolescents, using the gender specific cutoffs, suspected NAFLD increased from 3.6%± 0.6 to 9.9%± 1.3 (p<.0001). In overweight adolescents, the prevalence of elevated ALT was 13.2% in 2007-08 and there was no significant linear increase in ALT levels over time. However, among obese adolescents, prevalence of elevated ALT increased 120% from 16.7% to 36.9% (p=.006; Figure 1). Using the established cutoffs of ALT>30, similar increases were observed. Conclusions: Suspected NAFLD in adolescents is increasing in prevalence and affects approximately 10% of all adolescents. The increase is primarily occurring in obese adolescents and is more rapid than the increase in obesity itself. These data strongly support recommendations to screen for NAFLD in obese adolescents.

707 MRI Identifies Unsuspected Liver Dysfunction in Patients After Fontan Procedure Pinar Bulut, Rene Romero, Michael McConnell, Kiery Braithwaite, William Mahle, Nitika A. Gupta, Miriam B. Vos, Adina Alazraki Background: Long term survivors after Fontan procedure are at increased risk of developing hepatic dysfunction/cirrhosis and its related complications. The degree of liver dysfunction is often underestimated by serum biochemical testing. At our center, pediatric cardiologists have begun to routinely refer post-Fontan patients for hepatologic assessment. We postulated whether MRI might detect abnormalities before clinical or biochemical perturbations in postFontan patients. Material & Methods: Children post- Fontan repair who were referred to the liver program for evaluation between 2010-2011 were retrospectively reviewed. Evaluation included history and physical examination, comprehensive metabolic panel, gamma-glutamyl transpeptidase, INR, complete blood count, hepatitis B and C serology, ceruloplasmin, alpha1 antitrypsin level and phenotype. All but 1 patient was scanned on a SiemensTrioTrim 3T magnet; 1 patient was scanned on a GE Twinspeed 1.5T magnet with an equivalent protocol due to orthodontics. A standardized departmental protocol was utilized. MRI findings were correlated with age at surgery and years since surgery. MRI images were reviewed independently by 2 pediatric radiologists. IRB approval was obtained. Results: 29 patients met inclusion criteria. Mean age at time of Fontan surgery was 2.69 years (1.48 - 12.5 years). Mean age at hepatology evaluation was 14.61 years (9.1- 19.2 years). Mean time elapsed between Fontan surgery and hepatologic evaluation was 10.97 years (4.7- 17.75 years). On physical examination 9/29 (31%) had hepatomegaly, 2/29 (6.9%) had isolated splenomegaly. Evaluation for coexisting chronic liver disease as outlined was negative in all patients. All patients had normal albumin and INR. 17/29 patients (58%) underwent abdominal MR imaging study, 4 patients had MRI incompatible hardware and 8 patients were not scanned due to insurance denial. MR images were evaluated for presence of fibrosis, congestion and any other hepatic abnormalities. Fibrosis was determined based on specific pattern of delayed reticular enhancement in combination with liver morphology. All patients demonstrated

S-925

AASLD Abstracts

AASLD Abstracts

Differential Molecular Regulation of Bile Acid Homeostasis by Soy Lipid Induced Phytosterolemia and FISH Oil Lipid Emulsions in TPN-Fed Preterm Pigs Kenneth Ng, Hester Vlaardingerbroek, Barbara Stoll, Hans van Goudoever, Leo Kluijtmans, Oluyinka Olutoye, Douglas Burrin