Professional Documents
Culture Documents
NASH in Children
NASH in Children
Melania Manco, MD, PhD, GianFranco Bottazzo, MD, Rita DeVito, MD, Matilde Marcellini, MD,
Geltrude Mingrone, MD, PhD, FACN, and Valerio Nobili, MD
Scientific Directorate (M.M., GF.B), Dept. of Pathology (R.D.V.) and Hepato-Gastroenterology and Nutrition (V.N., M.M.),
“Bambino Gesu” Children’s Hospital and Research Institute, Dept. of Internal Medicine (G.M.), Catholic University,
Rome, ITALY
Key words: diet, insulin resistance (IR), Metabolic Syndrome (MetS), Non-Alcoholic Fatty Liver Disease (NAFLD),
Non-Alcoholic Steatohepatitis (NASH), pediatric obesity
In view of the epidemic obesity in childhood, facing the disease and its associated morbidities early at this
age becomes crucial for public health researchers and care givers. The present review focuses on pediatric Non
Alcoholic Fatty Liver Disease (NAFLD) among co-morbidities, being the disease yet under diagnosed and under
treated despite a prevalence growing exponentially.
Evidences suggest that the environmental background for the development of NAFLD may be established in
early life, and that the duration of the disease affects probably the likelihood of progression to more severe
disease (necro-inflammation or Non Alcoholic SteatoHepatitis, also termed, NASH; fibrosis and cirrhosis).
NAFLD associates with abdominal obesity, insulin resistance and features of metabolic syndrome. In
genetically prone individuals, malnutrition (i.e., excessive consumption of saturated fats and refined sugars)
leads to the derangement of the adipose tissue architecture and homeostasis, the peripheral and hepatic resistance
to insulin-stimulated glucose uptake, thus favoring a condition of chronic low-grade inflammation. Excessive
nutrients cannot be stored in the adipose tissue and overflow elsewhere, mainly to the muscle tissue and liver.
Fat deposition in both sites enhances insulin resistance and further deposition of fats in a vicious manner.
What is of special interest comparing NAFLD in children and adults is that the histological appearance of
the disease differs significantly, likely representing a yet physiological response to environmental stressors in
children and a long-term adaptation in adults.
In this article, we review the current concepts about paediatric NAFLD, its pathogenesis, diagnosis and
treatment, with particular regard to lifestyle and foods habits.
CURRENT SCENARIO IN PEDIATRIC reported from North America, Europe, Australia and Asia [1].
NAFLD. PREVALENCE OF THE It is almost certainly the most common cause of liver disease at
DISEASE this age, with cases described in children as old as 3 years [2].
The increase in its prevalence parallels the epidemic obesity. A
Non Alcoholic Fatty Liver Disease (NAFLD) may be recent survey on adolescent participants in the NHANES
deemed as a worldwide problem in childhood, with case series 1999 –2004 observes that elevated ALT level, a surrogate
Address correspondence to: Melania Manco, MD Ph.D, Scientific Directorate, Bambino Gesù Children’s Hospital, S. Onofrio 4 Square, 00165 Rome, ITALY. E-mail:
melaniamanco@tiscali.it
Journal of the American College of Nutrition, Vol. 27, No. 6, 667–676 (2008)
Published by the American College of Nutrition
667
Pandemic NAFLD in Children
marker of NAFLD in absence of other causes of liver disease, during the gestational life. According to the hypothesis of a
is present in 8.0% of US adolescents aged 12–19 years [3]. In fetal programming, undernutrition in mid and late gestation
a case series of autopsies performed on children 2–19 years of results in maintaining the brain at the expense of the trunk
age at time of death, the prevalence of histological proven growth, including visceral organs such as the liver [15]. This
NAFLD is 9.6% [4]. Prevalence of the disease increases when causes permanent alterations in the liver function [16]. Con-
samples of subjects are selected for obesity. NAFLD is reported trasting our and other’s thesis, the recent survey from the
to affect 20% of obese children and adolescents from US [5], NHANES 1999 –2004 founds no association of birth weight or
44% from Italy [6] and 74% from China [7]. The disease maternal smoking during pregnancy with elevated ALT level
associates significantly with type 2 diabetes mellitus [8] and all [3].
features of the metabolic syndrome (MetS) [2]. Children start to eat dense calorie food (“junk food” rich in
NAFLD encompasses a spectrum of disease from asymp- both high saturated or hydrogenated fats and high glycaemic
tomatic steatosis, with or without elevated aminotransferases, index carbohydrates; sweaty beverages and carbonate drinks),
to cirrhosis with complications of liver failure and hepatocel- whilst their physical activity is inadequate to burn excessive
lular carcinoma [9]. The histological appearance ranges from calories. Fats accumulate largely in adipose tissue, and, inap-
simple steatosis to hepatocellular damage coupled with inflam- propriately, in muscle and liver. The sequence of events leading
mation (i.e., Non Alcoholic SteatoHepatitis, NASH) and/or to the ectopic accumulation of triglicerydes which causes in-
fibrosis [10]. Diagnosis of NAFLD requires serial measure- sulin resistance to develop has been referred to as the “overflow
ments of aminotransferases and ultrasound evaluation of liver hypothesis” [18]. Insulin resistance represents a physiological
brightness, whilst diagnosis of NASH and fibrosis necessitates energy sparing mechanism which favours survival during lim-
liver biopsy. Therefore, prevalence of NAFLD can be exactly ited food availability or increased energy needs, promoting
estimated in obese children, who are referred to secondary care energy accumulation as fat and reducing energy expenditure.
centers for the treatment of obesity. On the contrary, no accu-
rate information can be provided on the prevalence of NASH.
On the largest two samples of biopsy proven NAFLD described INSULIN RESISTANCE AND
in the literature [2,11], NASH is diagnosed in 84% of NAFLD OXIDATIVE DAMAGE
children from the Rome survey [2] and in 68% from the San
Diego sample [11]. The role of IR in the progression from NAFLD to NASH is
still unclear. Rate of oxidative and non-oxidative glucose dis-
posal seem to be reduced significantly in patients who develop
PATHOGENESIS: BEYOND THE NASH compared with those who do not [18], being almost half
“TWO HIT HYPOTHESIS” of the normal value even in NAFLD patients with no impaired
glucose tolerance [19]. This means that a higher demand of
After the disease was firstly described in 1980, two decades insulin is needed to maintain euglycemia. Consequently, insu-
later a first hypothesis was formulated on its pathogenesis [12] lin secretion is enhanced to balance the decreased glucose
and revised shortly [13]. In brief, excessive dietary fats over- disposal [2,18]. The pathogenetic mechanisms sustaining the
flow to the liver, and fat deposition is the first hit in the model relation between IR and development of NAFLD/NASH are,
(simple steatosis or NAFLD). Second insults, yet unknown, however, far from the intent of this review. They have been
determine the progression from NAFLD to NASH. To date, extensively afforded elsewhere [20]. Therefore, in Fig. 1, we
several researchers still bet on the pathogenetic role of a num- schematise simply main pathways of this relationship. In brief,
ber of mediators as candidate hits. A complex of several agents, IR results from the inability of the adipose tissue to expand and
and not a single one, may interact driving NAFLD to NASH. to accommodate excess fats. This imbalance favours the exces-
The interaction between genes and environment begins during sive release of adipocytokines which further enhances IR in a
the intrauterine and the early years of life to promote NASH. vicious cycle. To mediate signalling, insulin receptor uses
To support this hypothesis is the evidence that the disease is docking proteins, such as insulin receptors substrates (IRS).
diagnosed in children as old as three years [2], and children Tyrosine phosphorylation of IRS is a crucial step which can
born small weight for gestational age (SGA) have greater activate three major pathways: i) PI3K-Akt pathway involved
chance to develop NASH [14]. The concept of a genetic- in glucose, lipid and protein metabolisms; ii) MAPK pathway
environment conspiracy as a major basis for the progression of which mediates cell growth and differentiation; iii) CAP/Cbl/
metabolic diseases (thrifty genotype) applies also to NASH, Tc10 pathway, which, in muscle, controls the membrane trans-
which associates almost constantly with obesity, insulin resis- location of glucose transporter 4 (GLUT4). In liver, the acti-
tance (IR) and all the features clustering in the metabolic vation of PI3K-Akt pathway results in abnormally suppressed
syndrome [2]. Apart from predisposing to the development of lipolysis, increased gluconeogenesis and de novo lipogenesis
components of the MetS in the adult life, intrauterine factors [due to the up-regulation of lipogenic transcription factors_i.e.
seem to be important for liver development and function yet the sterol regulatory binding protein-1c (SREBP-1c) and the
LIVER HISTOLOGY
Histological features of NASH differ in childhood respect
with adults [41]. Whilst the adult pattern (termed NASH type 1)
is characterised by the presence of steatosis (mainly macro-
vescicular) with ballooning degeneration and/or perisinusoidal
fibrosis in absence of portal features (Fig. 3A), the pediatric
NASH (NASH Type 2) is defined by the presence of steatosis
along with portal inflammation and/or fibrosis in absence of
ballooning degeneration and perisinusoidal fibrosis (Fig. 3B).
In the San Diego series, NASH type 2 is more common (51
patients out of 100); Type 1 is observed in 17 subjects, and an
overlap of type 1 and 2 is found in 32 cases. When the same
criteria proposed by Schwimmer et al. [41] are applied to the
Rome cohort [42], only 2.4% of 84 patients meet the criteria for
NASH type 1, and 29% the criteria for NASH type 2. The
combination of both types is observed in the largest part of
patients (52%) (Fig. 3C). Simple steatosis is found in the 17%
of the sample. Thus, in the Rome series NASH type 1 is rarely
seen as compared with San Diego survey, while the overlap of
type 1 and 2 pattern is the commonest histological report. In the
Rome series, fibrosis is noted in 58% patients; mostly of mild
severity (stage 1 in 51%), with only 5% patients showing septal Fig 3. Panel A: NASH type 1. Steatosis (S) with ballooning (B), no
fibrosis (stage 3). Fibrosis is significantly associated with older fibrosis and/or inflammation in the portal tract (EE 10X). Panel B:
age, increased body mass index [42], metabolic syndrome and NASH type 2. Steatosis (S) with inflammation (I) and fibrosis in the
its components, decreased insulin secretion and sensitivity [2]. portal tract (EE 10X). Panel C: NASH Overlap. Steatosis (S) and
In the San Diego survey, fibrosis is present in 60% of patients ballooning degeneration (B) are associated with inflammation and
fibrosis in the portal tract (EE 10X).
being moderate to severe in half of them. Cirrhosis is observed
in 3% of children. Fibrosis is positively related to the amount
of fat at the biopsy, and to the presence of lipogranulomas and
portal inflammation [41].
Difference in histological prevalence may resemble the dif- on fatty liver disease in adults with the intent to enable clini-
ferent ethnicity of the two samples: all Caucasians in the Rome cians to evaluate the most appropriate diet for NAFLD/NASH
series; mostly Hispanics and Asians in the San Diego sample. patients and make rational decisions based on this perspec-
As far as the different patterns of histology observed in children tive—in the absence of controlled trials—to help their patients
and adults are concerned, it has been speculated that the liver [47]. Basically, all the speculations made by the Authors apply
responds differently in the pediatric age to an insult which is to children, too and are currently in use at our Unit.
not yet inveterate as in the adults. In this context, periportal First of all, sudden or quick weight loss achieved through
damage may represent the earlier stage of the injury [43]. dietary modification may accelerate the progression from sim-
What is more significant is that there is still no consensus on ple fatty liver to NASH, therefore gradual weight loss is
how to evaluate and score histological features of liver involve- strongly recommended. Consumption of carbohydrate should
ment. The scoring system for the semi quantitative evaluation of be moderate and of low-glycemic index (GI) foods preferred. A
NASH, proposed by Elizabeth Brunt et al [44] assesses steatosis, higher carbohydrate intake is associated with a greater odd of
necro-inflammatory activity (grading) and architectural alterations inflammation than a higher fat intake [48]. A recent meta-
related to fibrosis (staging). The system has been developed for analysis [49] has highlighted all the beneficial effects of low in
NASH, but it does not seem sufficient for the assessment of GI dietary components in individuals with impaired insulin
NAFLD. Based on the initial system of Elisabeth Brunt, the response and dyslipidemia. According to these evidences, it
Pathology Committee of the NASH Clinical Research Network seems reasonable to conclude that the inclusion of carbohy-
has designed and validated a tailored scoring system that encom- drates that are high in indigestible and fermentable fiber and
passes the full spectrum of lesion of NAFLD. This system intro- low in GI can be helpful in maintaining glucose, insulin, and
duces the NAFLD activity score (NAS) including only features of FA concentrations in individuals with IR, such as NAFLD
active injury that are potentially reversible: steatosis, lobular in-
patients.
flammation and ballooning [45].
Consumption of sweetened drinks causes excess intake of as
much as 150 –300 kcal/d [50]. Therefore intakes of refined
sugars and high-fructose or high-glucose foods and beverages
ROLE OF DIET AND should be reduced in NAFLD children. High intakes of both
PHARMACOTHERAPY sugars can stimulate de novo synthesis of fatty acids [51] and
lead to changes in long-term energy balance regulation at the
Diet level of the central nervous system, inducing mainly a reduced
post meal suppression of the orexigenic hormone ghrelin [52].
To date, therapeutic strategies for NAFLD/NASH have
Intake of saturates fats must be limited to ⬍10% of total
revolved around i) identification and treatment of associated
energy, since they promote endoplasmic reticulum stress as well as
metabolic conditions such as obesity, diabetes and dyslipide-
hepatocyte injury [53], insulin resistance [54] and development of
mia; ii) amelioration of insulin resistance by weight loss, ex-
all components of the metabolic syndrome. Consumption of trans
ercise or pharmacotherapy; iii) use of hepato-protective agents
such as antioxidants to protect liver from secondary insults. fatty acids must also be limited. Although the specific pathoge-
A program targeting gradual weight reduction and physical netic mechanisms of trans FAs are not yet clear, the recommen-
exercise continues to be the gold standard of treatment for dation to avoid the intake of those fatty acids deriving from
NAFLD in children as well as in adults, even though little to no hydrogenated oils seems well founded, since they may increase
scientific evidence is available on diet and NAFLD. Change in inflammatory markers [55], induce endothelial dysfunction [56],
lifestyle per sè represents the only effective therapeutic option. and unfavorably alter the blood lipid profile [57]. Conversely,
Lifestyle advice, including mainly low-fat and low-glycaemic some bacterially derived trans FAs typically found in dairy prod-
index diets and regular physical exercise, causes weight loss, ucts seem to not have adverse effects on lipoprotein profiles [58].
improvement of body composition, and amelioration up to Intake of monounsaturated fatty acids (MUFAs) has been largely
resolution of metabolic associated morbidities, including im- recommended. MUFAs are a key component of the Mediterranean
pairment of glucose metabolism, dyslipidemia and blood hy- diet, representing the olive oil the main source. The beneficial
pertension [42]. Ultrasound liver brightness and liver enzymes effects of MUFAs on cardiovascular disease risk and blood lipid
normalize or improve early after the beginning of the treatment profiles have been extensively addressed [59], as well as a neutral
[42]. This occurs in parallel with the amelioration or resolution in vivo effect on insulin resistance and secretion has been postu-
of steatosis and necro-inflammation, while no significant lated [54]. Contrasting, in part, the latter hypothesis, a recent study
change comes about grade of fibrosis after two years of a [60] evaluating expression of the phosphatase and tensin homo-
treatment based on a nutritional counsel alone [46]. A recent logue on chromosome 10 (PTEN) in the liver of rats and humans
review discusses systematically the effect of macronutrient having steatosis, found that oleic acid induces specifically down
content (carbohydrate, fat, and protein ratios) and specific food regulation of PTEN transcription. PTEN is a major regulator of the
components, such as soluble fiber, n-3 fatty acids, and fructose, insulin receptor substrate, and its down regulation may favor
MJ, Sterling RK, Luketic VA, Shiffman ML and Clore JN: Non 40. Nobili V, Vizzutti F, Arena U, Abraldes JG, Marra F, Pietrobattista
alcoholic steatohepatitis: association of insulin resistance and mi- A, Fruhwirth R, Marcellini M, Pinzani M: Accuracy and repro-
tochondrial abnormalities. Gastroenterology 120:1183–1192, ducibility of transient elastography for the diagnosis of fibrosis in
2001. pediatric nonalcoholic steatohepatitis. Hepatology 2008 Apr 18.
25. Leclerq IA, Da Silva Morais A, Schroyen B, Van Hul N, Geerts A: 41. Schwimmer JB, Behling C, Newbury R, Dcutsch R, Nievergelt C,
Insulin resistance in hepatocytes and sinusoidal liver cells: Mech- Schork NJ, Lavine JE: Histopathology of pediatric nonalcoholic
anisms and consequences. J Hepatol 47:142–156, 2007. fatty liver disease. Hepatology 42:641–649, 2005
26. Manco M, Marcellini M, Giannone G, Nobili V: Correlation of 42. Nobili V, Marcellini M, Devito R, Ciampalini P, Piemonte F,
serum TNF-alpha levels and histologic liver injury scores in pae- Comparcola D, Sartorelli MR, Angulo P: NAFLD in children: a
diatric non-alcholic fatty liver disease. Am J Clin Pathol 127:954– prospective clinical-pathological study and effect of lifestyle ad-
960, 2007. vice. Hepatology 44:458–465, 2006.
27. Nobili V, Manco M, Ciampalini P, Diciommo V, Devito R, Pi- 43. Roberts EA: Pediatric nonalcholic fatty liver disease (NAFLD): A
emonte F, Comparcola D, Guidi R, Marcellini M: Leptin, free “growing” problem? J Hepatol 46:1133–1142, 2007.
leptin index, insulin resistance and liver fibrosis in children with 44. Brunt EM: Non-alcoholic steatohepatitis: definition and pathology.
non-alcoholic fatty liver disease. Eur J Endocrinol 155:735–743, Semin Liver Dis 21: 3–16, 2001.
2006. 45. Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ,
28. Zou CC, Liang L, Hong F, Fu JF, Zhao ZY: Serum adiponectin, Cummings OW, Ferrell LD, Liu YC, Torbenson MS, Unalp-Arida
resistin levels and non-alcoholic fatty liver disease in obese chil- A, Yeh M, McCullough AJ, Sanyal AJ: Nonalcoholic Steatohepa-
dren. Endocr J 52:519–524, 2005. titis Clinical Research Network. Design and validation of a histo-
29. Manco M, Bedogni G, Marcellini M, Devito R, Ciampalini P, logical scoring system for nonalcoholic fatty liver disease. Hepa-
Sartorelli MR, Comparcola D, Piemonte F, Nobili V: Waist cir- tology 41:1313–1321, 2005.
cumference correlates with liver fibrosis in children with non 46. Nobili V, Manco M, Devito R, Di Ciommo V, Comparcola D,
alcoholic steatohepatitis. Gut 57:1283–1287, 2008. Sartorelli MR, Piemonte F, Marcellini M, Angulo P: Lifestyle
30. Keskin M, Kurtoglu S, Kendrici M, Atabek ME, Yazici C: Ho- intervention and antioxidants in children with non alcoholic fatty
meostasis model assessment is more reliable than the fasting glu- liver disease: a randomised controlled trial. Hepatology 48:119–
cose/insulin and quantitative insulin sensitivity index for assessing 128, 2008.
insulin resistance among obese children and adolescents. Pediatrics 47. Zivkovic AM, German JB, Sanyal AJ: Comparative review of diets
115:E500–E503, 2005. for the metabolic syndrome: implications for nonalcoholic fatty
31. Uwaifo GI, Tjahjana M, Freedman RJ, Lutchman G, Promrat K: liver disease. Am J Clin Nutr 86:285–300, 2007.
Acanthosis nigricans in patients with nonalcoholic steatohepatitis: 48. Solga S, Alkhuraishe AR, Clark JM, Torbenson M, Greenwald A,
an uncommon finding. Endocr Pract 12:371–379, 2006. Diehl AM, Magnuson T: Dietary composition and nonalcoholic
32. Kim HJ, Kim HJ, Lee KE, Kim DJ, Kim SK, Ahn CW, Lim SK, fatty liver disease. Dig Dis Sci 49:1578–1583, 2004.
Kim KR, Lee HC, Huh KB, Cha BS: Metabolic significance of 49. Anderson JW, Randles KM, Kendall CW, Jenkins DJ: Carbohy-
nonalcoholic fatty liver disease in nonobese, nondiabetic adults. drate and fiber recommendations for individuals with diabetes: a
Arch Intern Med 164:2169–2175, 2004. quantitative assessment and meta-analysis of the evidence. J Am
33. Matsuda M, De Fronzo RA: Insulin sensitivity indices obtained Coll Nutr 23:5–17, 2004.
from oral glucose tolerance testing: comparison with the euglyce- 50. Popkin BM, Armstrong LE, Bray GM, Caballero B, Frei B, Willett
mic insulin clamp. Diabetes Care 22:1462–1470, 1999. WC: A new proposed guidance system for beverage consumption
34. Mari A, Pacini G, Murphy E, Ludvik B, Nolan JJ: A model-based in the United States. Am J Clin Nutr 83:529–542, 2006.
method for assessing insulin sensitivity from the oral glucose 51. McDevitt RM, Bott SJ, Harding M, Coward WA, Bluck LJ,
tolerance test. Diabetes Care 24:539–548, 2001. Prentice AM. De novo lipogenesis during controlled overfeeding
35. DeFronzo RA, Tobin JD, Andres R: Glucose clamp technique: a with sucrose or glucose in lean and obese women. Am J Clin Nutr
method for quantifying insulin secretion and resistance. Am J 74:737–746, 2001.
Physiol 237:E214–E223, 1979. 52. Teff KL, Elliott SS, Tschop M, Kiefer TJ, Rader D, Heiman M,
36. Bergman RN: Lilly lecture 1989. Toward physiological under- Townsend RR, Keim NL, D’Alessio D, Havel PJ: Dietary fructose
standing of glucose tolerance. Minimal-model approach. Diabetes reduces circulating insulin and leptin, attenuates postprandial sup-
38:1512–1527, 1989. pression of ghrelin, and increases triglycerides in women. J Clin
37. American Diabetes Association: Diagnosis and classification of Endocrinol Metab 89:2963–2972, 2004.
Diabetes Mellitus. Diabetes Care 31:S55–S60, 2008. 53. Wang D, Wei Y, Pagliassotti MJ: Saturated fatty acids promote
38. Angulo P, Hui JM, Marchesini G, Bugianesi E, George J, Farrell endoplasmic reticulum stress and liver injury in rats with hepatic
GC, Enders F, Saksena S, Burt AD, Bida JP, Lindor K, Sanderson steatosis. Endocrinology 147:943–51, 2006.
SO, Lenzi M, Adams LA, Kench J, Therneau TM, Day CP: The 54. Manco M, Calvani M, Mingrone G: Effects of dietary fatty acids
NAFLD fibrosis score: a noninvasive system that identifies liver on insulin sensitivity and secretion. Diabetes Obes Metab 6:402–
fibrosis in patients with NAFLD. Hepatology 45:846–854, 2007. 413, 2004.
39. Rosenberg WM, Voelker M, Thiel R, Becka M, Burt A, Schuppan 55. Lopez-Garcia E, Schulze MB, Meigs JB, Manson JE, Rifai N,
D, Hubscher S, Roskams T, Pinzani M, Arthur MJ: European Liver Stampfer MJ, Willett WC, Hu FB; Consumption of trans fatty
Fibrosis Group. Serum markers detect the presence of liver fibro- acids is related to plasma biomarkers of inflammation and endo-
sis: a cohort study. Gastroenterology 127:1704–1713, 2004. thelial dysfunction. J Nutr 135:562–566, 2005.
56. Mozaffarian D: Trans fatty acids—effects on systemic inflamma- 65. Marchesini G, Brizi M, Bianchi G: Metformin in non-alcoholic
tion and endothelial function. Atheroscler Suppl 7:29–32, 2006. steatohepatitis. Lancet 358:893–894, 2001.
57. Tricon S, Burdge GC, Kew S, Banerjee T, Russell JJ, Jones EL, 66. Lin HZ, Yang SQ, Chuckaree C, Kuhajda F, Ronnet G, Diehl AM:
Grimble RF, Williams CM, Yaqoob P, Calder PC: Opposing Metformin reverses fatty liver disease in obese, leptin-deficient
effects of cis-9,trans-11 and trans-10,cis-12 conjugated linoleic mice. Nat Med 6:998–1003, 2000.
acid on blood lipids in healthy humans. Am J Clin Nutr 80:614– 67. Belfort R, Harrison SA, Brown K, Darland C, Finch J, Hardies J,
620, 2004. Balas B, Gastaldelli A, Tio F, Pulcini J, Berria R, Ma JZ, Dwivedi
58. Tricon S, Burdge GC, Jones EL, Russell JJ, El-Khazen S, Moretti S, Havranek R, Fincke C, DeFronzo R, Bannayan GA, Schenker S,
E, Hall WL, Gerry AB, Leake DS, Grimble RF, Williams CM, Cusi K: A placebo-controlled trial of pioglitazone in subjects with
Calder PC, Yaqoob P: Effects of dairy products naturally enriched nonalcoholic steatohepatitis. N Engl J Med 355:2297–2307, 2006.
with cis-9,trans-11 conjugated linoleic acid on the blood lipid 68. Lavine JE, Schwimmer JB: Clinical Research Network launches
profile in healthy middle-aged men. Am J Clin Nutr 83:744–753, TONIC: trial for treatment of non alcoholic fatty liver disease in
2006. children. J Pediatr Gastroenterol Nutr 42:129–130, 2006.
59. Lichtenstein AH. Thematic review series: patient-oriented re- 69. Schwimmer JB, Middleton MS, Deutsch R, Lavine JE: A phase 2
search. Dietary fat, carbohydrate, and protein: effects on plasma clinical trial of metformin as a treatment for non-diabetic paediatric
lipoprotein patterns. J Lipid Res 47:1661–1667, 2006. non-alcoholic steatohepatitis. Aliment Pharmacol Ther 21:871–
60. Vinciguerra M, Veyrat-Durebex C, Moukil MA, Rubbia-Brandt L, 879, 2005.
Rohner-Jeanrenaud F, Foti M: PTEN down-regulation by unsatur- 70. Lavine JE: Vitamin E treatment of nonalcoholic steatohepatitis in
ated fatty acids triggers hepatic steatosis via an NF kappaBp65/ children: a pilot study. J Pediatr 136:734–738, 2000.
mTOR-dependent mechanism. Gastroenterology. 134:268–280, 71. Vajro P, Mandato C, Franzese A, Ciccimarra E, Lucariello S,
2008. Savoia M, Capuano G, Migliaro F: Vitamin E treatment in pedi-
61. Goheen SC, Larkin EC, Rao GA: Severe fatty liver in rats fed a atric obesity-related liver disease: a randomized study. J Pediatr
fat-free ethanol diet, and its prevention by small amounts of dietary Gastroenterol Nutr. 38:48–55, 2004.
arachidonate. Lipids 18:285–290, 1983. 72. Nobili V, Manco M, DeVito R, Ciampalini P, Piemonte F, Mar-
62. Bouziane M, Prost J, Belleville J: Dietary protein deficiency af- cellini M: Effect of vitamin E on aminotransferase levels and
fects n-3 and n-6 polyunsaturated fatty acids hepatic storage and insulin resistance in NAFLD children. Aliment Pharmacol Ther
very low density lipoprotein transport in rats on different diets. 24:1553–1561, 2006.
Lipids 29:265–272, 1994. 73. Nobili V, Manco M, Ciampalini P, Alisi A, Devito R, Bugianesi E,
63. Clarke SD, Jump DB: Polyunsaturated fatty acid regulation of Marcellini M, Marchesini G: Metformin use in children with
hepatic gene transcription. Lipids 31:S7–S11, 1996. nonalcoholic fatty liver disease: An open-label, 24-month, obser-
64. Kim HJ, Takahashi M, Ezaki O: Fish oil feeding decreases mature vational pilot study. Clin Ther 30:1168–1176, 2008.
sterol regulatory element-binding protein 1 (SREBP-1) by down-
regulation of SREBP-1c mRNA in mouse liver. A possible mech-
anism for down-regulation of lipogenic enzyme mRNAs. J Biol
Chem 274:25892–25898, 1999. Received March 19, 2008; revision accepted July 22, 2008.