FATTY LIVER: it's affect in HIV & hepatitis
This report in the current issue of the journal Hepatology discusses the depositing and accumulation of fat in the liver in HIV negative individuals. The report discusses the causes and factors of fatty liver such as lipodystrophy, NRTIs, and insulin resistance. The article talks about the relationship between fatty liver and cirrhosis. I think HIV lipodystrophy researchers should pay attention to fatty liver syndrome and it's affect in HIV-infected individuals and in HCV and HBV coinfected individuals. NASH (non-alcoholic steatohepatitis) and NAFLD (non-alcoholic fatty liver disease) are the terms used to describe fatty liver.
"Nonalcoholic steatohepatitis: Summary of an AASLD Single Topic Conference'
Here are a few excerpted very cogent quotes from the article:
NAFLD and NASH are associated commonly with obesity, diabetes, and hyperlipidemia... lipodystrophy.. nucleoside analogues.. abnormal fat metabolism..mitochondrial injury or dysfunction... Liver biopsy is the only means of assessing the presence and extent of specific necroinflammatory changes and fibrosis... insulin resistance is at the top as an important antecedent metabolic abnormality in many patients.... Excess fat in the liver predisposes to hepatocellular injury... Hepatocellular injury may cause an inflammatory response with progressive fibrosis... The increased flux of free fatty acids through the liver in states of exuberant peripheral lipolysis may play a direct role in hepatocellular injury.. PPAR (peroxisomal proliferator activated receptor ) plays a central role in sensing excess free fatty acids and up-regulating the genetic program of fatty acid disposal... Congenital generalized lipodystrophy is characterized by nearly absent peripheral fat, severe hepatic steatosis, and a significant risk for cirrhosis..... Mutations of gene-encoding PPAR-, PPARG, are associated with partial lipodystrophies, although liver disease has not been explored in these patients... The absence of peripheral fat in the lipodystrophies also impairs leptin signaling because of a deficiency of adipocyte-derived leptin. A clinical trial of leptin administration to hypoleptinemic patients with partial lipodystrophy reduced liver volume and liver triglyceride content. It has not been established whether the response to leptin was due to improved leptin signaling in the liver or a central nervous system effect of leptin leading to diminished food intake...
.... Because adenosine triphosphate (ATP) is critical for maintaining cellular integrity, its depletion may predispose to hepatocellular injury. Studies as early as Dianzani's work in the 1950s have shown that hepatic ATP levels are depleted in experimental models of fatty liver.. Mitochondrial injury may be one cause of reduced hepatocellular ATP stores in NASH...
.... The response to chronic hepatocellular injury varies dramatically among individual patients with liver disease, possibly explaining why steatohepatitis is
relatively well tolerated in some yet associated with rapidly accumulating fibrosis in others. This variability may in part be explained by a variety of
polymorphisms of peptide mediators of the inflammatory cascade and their receptors..
.. Exercise and diet continue to be the cornerstones of therapy.140 Although typically recommended together, the concept of the fit fat individual (i.e., relatively well conditioned but obese) is relevant and suggests a benefit of exercise even in the absence of weight loss. Exercise alters substrate use in skeletal muscle and insulin sensitivity, although only about one third of patients achieve target levels of exercise and obese individuals may be resistant to these changes. A small number of studies of diet and exercise therapy have been reported in both adults and children. These typically reveal improved biochemical parameters but variable changes in histology. Histologic exacerbation has been observed when the rate of weight loss exceeded 1.6
g/wk. Higher-intensity exercise regimens are probably more effective in producing significant metabolic changes...
Fatty liver disease that develops in the absence of alcohol abuse is recognized increasingly as a major health burden. This report summarizes the presentations and discussions at a Single Topic Conference held September 20-22, 2002, and sponsored by the American Association for the Study of Liver Diseases. The conference focused on fatty liver disorders. Estimates based on imaging and autopsy studies suggest that about 20% to 30% of adults in the United States and other Western countries have excess fat accumulation in the liver. About 10% of these individuals, or fully 2% to 3% of adults, are estimated to meet current diagnostic criteria for nonalcoholic steatohepatitis (NASH). Sustained liver injury leads to progressive fibrosis and cirrhosis in a fraction, possibly up to one third, of those with NASH, and NASH may be a cause of cryptogenic cirrhosis. NASH is now a significant health issue for obese children as well, leading to cirrhosis in some. The diagnostic criteria for NASH continue to evolve and rely on the histologic findings of steatosis, hepatocellular injury (ballooning, Mallory bodies), and the pattern of fibrosis. Generally recognized indications for biopsy include establishing the diagnosis and staging of the injury, but strict guidelines do not exist. Liver enzymes are insensitive and cannot be used reliably to confirm the diagnosis or stage the extent of fibrosis. Older age, obesity, and diabetes are predictive of fibrosis.
The pathogenesis of NASH is multifactorial. Insulin resistance may be an important factor in the accumulation of hepatocellular fat, whereas excess intracellular fatty acids, oxidant stress, adenosine triphosphate (ATP) depletion, and mitochondrial dysfunction may be important causes of
hepatocellular injury in the steatotic liver. Efforts are underway to refine the role of insulin resistance in NASH and determine whether improving insulin
sensitivity pharmacologically is an effective treatment. An altered lifestyle may be a more effective means of improving insulin sensitivity. The research
agenda for the future includes establishing the role of insulin resistance and abnormal lipoprotein metabolism in NASH, determining the pathogenesis of
cellular injury, defining predisposing genetic abnormalities, identifying better noninvasive predictors of disease, and defining effective therapy. (HEPATOLOGY 2003;37:1202-1219.)
Editorial note from Jules Levin: NRTIs used in HIV antiretroviral treatment may lead to mitochondrial toxicity to cells. But as far as I know we have not yet studied this effect in liver cells, particularly in persons with chronic hepatitis C and B. As well, HIV-infected individuals may suffer insulin resistance, diabetes, and increased lipids on HAART. Several small studies find persons coinfected with HIV and HCV or HBV are at increased risk compared to persons infected with HIV alone for body changes (lipodystrophy), metabolic abnormalities, and insulin resistance. These conditions may also have an effect on the liver for persons infected with hepatitis C and B. But the effect on the liver for persons with HCV and HBV has yet to be adequately explored.
The following is a summary of talks presented at the American Association for the Study of Liver Diseases (AASLD) Clinical Single Topic Conference on
Nonalcoholic Steatohepatitis (NASH) held in Atlanta, Georgia on September 20-22, 2002. The syllabus and lecture materials prepared by the conference's 25
presenters form the basis of this summary. As course organizers, the authors of this summary would like to acknowledge the presenters for their substantial
The objective of this particular conference was to bring together active clinical and basic science researchers in the field of fatty liver disease to critically analyze the latest developments and share ideas with respect to pathogenesis, diagnosis, and treatment. A focus of the meeting was the role of insulin resistance and hyperinsulinemia in the pathogenesis of nonalcoholic fatty liver disease (NAFLD) and NASH. Because of the focused nature of the conference, many key aspects of fatty liver disease could not be discussed in depth. The purpose of this report is to provide an overview of the conference and not summarize all aspects of NAFLD and NASH.
The term NASH, coined by Ludwig et al.1 in 1980 to describe the biopsy findings in patients with steatohepatitis in the absence of significant alcohol
consumption, has served the field well by bringing attention to this entity and promoting further research. However, this inclusive term has become problematic
because it requires a pathologist to make a clinical statement about alcohol consumption. As recently proposed,2,3 an alternative name—metabolic
steatohepatitis—was discussed but not uniformly accepted. Another alternative under consideration by pathologists is to further simplify the nomenclature by
reporting only steatohepatitis using specific criteria (see later) and leaving it up to the clinicians to assign causality and risk factors.
Defining NAFLD and NASH, and the meaning of "nonalcoholic"
NAFLD is defined currently as fat accumulation in the liver exceeding 5% to 10% by weight, but it is estimated practically as the percentage of fat-laden
hepatocytes observed by light microscopy.4 Whether NAFLD with the minimum amount of fat is truly a disease, hence the D in the acronym, or simply a benign
condition, was debated. Progression to cirrhosis is rare in mild NAFLD, yet progression has been observed and any amount of fat may sensitize the liver to injury from other causes. Clearly, there is wide agreement on the need for a consensus regarding the precise criteria for classifying, grading, and staging histologic injury in these disorders collectively known as NAFLD (see later).
Inherent to defining NAFLD and NASH is the threshold at which steatohepatitis becomes alcohol related. This is not a sharply demarcated distinction. Many
centers accept up to 14 to 28 units of ethanol per week (up to 20-40 g/d in men and 20 g/d in women) whereas other investigators have used a cut-off level of 7 units/wk (10 g/d) or less. One report has suggested that limited alcohol intake is protective against NASH (as well as diabetes). Given the health benefits derived from modest ingestion of ethanol, this problem is unlikely to be resolved easily. As Dr. Oliver James has suggested, a reasonable compromise is to accept a working figure of 14 units/wk (20 g/d or roughly the equal of 2 glasses of wine per day) with acknowledgment that there will be uncertainty in the gray areas of this limit. This cut-off level is well below the traditional threshold for alcohol-induced liver disease.
The spectrum of histologic abnormalities defined by NAFLD includes simple steatosis (steatosis without other injury) and NASH as its more extreme forms. How NASH is best defined remains unsettled because there is significant diversity of opinion among expert pathologists regarding the necessity and character of specific findings. These include the amount and types of fat (macrovesicular and microvesicular), lobular inflammation (acute and/or chronic), and fibrosis (zone 3 and portal). In addition, the histologic findings in the pediatric population may differ from those in adults. One of the key histopathologic features of NASH in adults is the presence of perisinusoidal fibrosis, whereas in children portal fibrosis may be more characteristic.
A NAFLD classification system also has been proposed that correlates certain histologic features with the long-term prognosis8 (these groups are identified
variably as class or type). Class 1 constitutes simple steatosis, class 2 is steatosis with lobular inflammation, class 3 requires the additional presence of
ballooned hepatocytes, and class 4 requires the presence of either Mallory's hyaline or fibrosis. Within this system, class 3 and 4 NAFLD are similar and might be considered as a single group constituting NASH. Class 2 NAFLD is more controversial; it may be benign and includes relatively more men, often with a normal body mass index.
Class 3-4 NAFLD or NASH is described further by using 4 stages of fibrosis. A separate group of adult patients with primarily periportal fibrosis has been described, but this variant is not yet established as a distinct entity (V. Ratziu, unpublished data). Stage 4 NASH has been suggested to include NASH with cirrhosis, cirrhosis with features of NASH, and cryptogenic cirrhosis. It is now accepted that cryptogenic cirrhosis may represent a late phase of NASH that has lost the typical necroinflammatory and steatotic features in up to 80% of patients.
Prevalence and prognosis
NAFLD is perhaps the most common of all liver disorders. Wanless and Lentz found steatosis in 70% of obese and 35% of lean patients and NASH in
18.5% of obese and 2.7% of lean patients in a consecutive autopsy study. Among obese patients, the prevalence of class 1 NAFLD (simple steatosis) is about
60%, whereas NASH is found in 20% to 25% and 2% to 3% have cirrhosis. Among type 2 diabetic patients, it is estimated that 75% have some form of
A number of reports have addressed clinical predictors of more advanced histology on the initial diagnostic biopsy. Among these, age greater than 40 to 50 years, and the severity of obesity, diabetes, or hyperlipidemia (especially hypertriglyceridemia) are among the most reliable.5,14,26-29,36,37 The role of female gender has been more variable in reported series, but the relatively increased prevalence of women in patients with more advanced disease supports female gender as a risk for progression. Other reported predictors of advanced disease include elevation of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) and an AST:ALT ratio greater than 1. However, it is well known that significant liver disease may exist with liver enzymes
in the normal range among NAFLD patients. This could represent upward drift in the normal range, but treatment with antidiabetic medications also may
produce normalization of the aminotransferase levels despite pre-existing liver disease. Elevated serum immunoglobulin A level is under study as a potential
predictor of disease activity. It is important to note that all of these factors have not been studied adequately as predictors of progression over time but rather indicate the likelihood of finding more advanced disease on the initial biopsy. It is, however, likely that they also carry long-term prognostic significance.
The 5- and 10-year survival in NASH has been estimated at 67% and 59%, respectively, although death often may be from comorbid conditions. In Japan, the ratio of observed versus expected deaths was actually higher for cirrhosis than for heart disease in people with diabetes (2.67 vs. 1.81). Longitudinal
studies of NASH are few in number. Compiling figures from several reports (n = 32 patients), the risk in class 3 and 4 NAFLD for developing increased fibrosis
over approximately 5 years is 25% and for developing cirrhosis is 15%. Preliminary studies suggest a more benign course for class 2 NAFLD, although even simple steatosis also has been shown to progress occasionally to cirrhosis. Raw survival figures fail to reveal the morbidity among obese diabetic patients who develop cirrhosis and recent studies have indicated marked risk in such patients for complications of portal hypertension and a significant
risk for hepatocellular cancer in the subset who progress to cirrhosis.
NAFLD is reported increasingly in pediatric patients. Sixty percent of adolescents with elevated liver enzyme levels are obese or overweight and it is estimated that greater than 1% to 2% of adolescents have NAFL. The spectrum of histologic injury in this group clearly includes cirrhosis.6 Lavine presented
data showing the predictive power of the degree of insulin resistance and elevation of the aminotransferase levels. Unfortunately, no longitudinal studies yet exist.
Gender, ethnic, and familial considerations
It is now suspected that there is an even distribution of NASH among men and women although there may be gender variation among the specific classes.
Series of patients with more advanced disease have generally had more women, suggesting a more aggressive course. Surveys have suggested ethnic
variation with relative paucity among African Americans compared with European and Hispanic Americans. This may represent variation in referral patterns or genetic differences in body fat distribution or metabolic thermogenesis. Clustering within kindreds also has been described, further suggesting that genetic factors predispose to the development of NASH.
A history of obesity, diabetes, or hyperlipidemia is common but not invariable. An increasing number of patients have been described with normal body mass
index, although these individuals may have central adiposity and occult insulin resistance. Clinical findings include other features of metabolic syndrome
such as hypertension, hyperuricemia, and polycystic ovarian syndrome (hirsutism, oligomenorrhea, or amenorrhea). The importance of eliciting a previous history of features of the metabolic syndrome has been emphasized because changes in body composition due to aging and cirrhosis may mask prior severe and long-standing obesity.
Because of their associations with metabolic syndrome, NAFLD and NASH are associated commonly with obesity, diabetes, and hyperlipidemia, as well as
hypertension, hyperuricemia, and polycystic ovarian disease. Other conditions associated with these primary problems such as sleep apnea in obesity may be
observed. In addition, an association with lipodystrophy has been observed although the exact mechanism is not clear. Other noted associations include peroxisomal diseases, mitochondrialopathies, Weber-Christian disease, Mauriac Syndrome, Madelung's lipomatosis, Wilson's Disease, industrial solvent exposure, medications (amiodarone, tamoxifen, nucleoside analogues, and methotrexate), celiac disease, and abetalipoproteinemia. Many of these disorders have in common either abnormal fat metabolism and/or mitochondrial injury or dysfunction.
Symptoms, signs, laboratory, and imaging
Referral is often precipitated by abnormal liver enzyme levels detected at routine evaluation or during antihyperlipidemic drug therapy (Tables 3 and 4).
Table 3. Symptoms, Signs, Biochemistry and Imaging in NASH
Symptoms and physical findings
-Fatigue (correlates poorly with histologic stage)
-Right upper quadrant pain (usually mild but may be mistaken for gallstone disease)
-Bowel dysmotility and small bowel bacterial overgrowth
-Constipation (especially in children [J.E. Lavine, personal communication])
-Anthropometric (waist circumference indicates central adiposity
-Acanthosis nigricans (especially in children)
-Lipoatrophy/lipodystrophy (may be underrecognized in its focal or partial form)
-Panniculitis (rare, feature of Weber-Christian)
-Neurologic deficits (ocular muscle palsy, also, deafness may be part of maternally inherited deafness and diabetes)
-Palmer erythema, spider angiomata, and splenomegaly (cirrhosis)
Subacute liver failure
-Mild elevation of aspartate AST and ALT levels; levels seldom exceed 10? the upper level of normal and more typically are <1.5? the upper normal
-ALT > AST; AST > ALT suggests significant fibrosis or cirrhosis (altered with antidiabetic therapy)
-glutamyltransferase and alkaline phosphatase level elevation
- Hyperglycemia (caused by the association with diabetes present in about one third of patients)
- Hyperlipidemia (usually triglycerides) in approximately 20% to 25%1,11,227
- IgA deposition has been described in histologic sections in NASH patients and serum IgA level is elevated in about 25%
-Antinuclear antibody in about one third of patients
-Abnormal iron indices (common but generally do not indicate genetic hemochromatosis)
-Ultrasound, computed tomography, or magnetic resonance imaging: all insensitive to degrees of steatosis less than 25% to 30%
-None of these modalities are able to reliably identify fibrosis and stage the disease
-Magnetic resonance spectroscopy (MRI): fat content and ATP levels in fatty liver
Table 4. Tests of Insulin Homeostasis
A simple approach to assess insulin resistance or latent diabetes. Fasting insulin per se provides an approximate measure of insulin sensitivity but is not felt to be very precise.
QUICKI and HOMA
The Quantitative Insulin Check Index (QUICKI) and Homeostasis Model Assessment of Insulin Resistance (HOMA) are mathematic models based on the
product of the fasting insulin and glucose levels that provide similar measures of insulin sensitivity. The HOMA is calculated as insulin (mU/L) ? glucose (Ķmol/L)/22.5; the QUICKI is calculated as 1/log(insulin [mU/L] ? glucose [mg/dL]).
Hyperinsulinemic euglycemic clamp
The clamp technique is the gold standard. The insulin infusion rate can be tailored to examine hepatic sensitivity to insulin and/or peripheral glucose use.
The index of insulin sensitivity (SIClamp) is defined as M/(G ? I) corrected for body weight (where M = steady-state glucose infusion rate (mg/min), G is the
steady-state blood glucose concentration (mg/dL), and I is the difference between basal and steady-state insulin concentrations (mU/L)).
Frequently Sampled Intravenous Glucose Tolerance Test (FSIGT)
A less labor-intensive estimation of insulin sensitivity that provides the minimal model index of insulin sensitivity (SIMM). It correlates with glucose clamp
measurements in normal and obese subjects. The minimal model can generate an inaccurate index of insulin sensitivity in patients with impaired insulin secretion (i.e., in overt diabetes).
Hepatic degradation of insulin can be assessed by simultaneous measurement of fasting plasma insulin and C-peptide. A reduced C-peptide-to-insulin molar ratio therefore indicates impaired hepatic degradation of insulin. Renal impairment alters this relationship.
Oral Glucose Tolerance Test (OGTT)
Fasting plasma glucose (FPG) ≥110 and <126 mg/dL indicates impaired fasting glucose and FPG ≥126 mg/dL constitutes a provisional diagnosis of diabetes. Using a 75-g oral glucose tolerance test, the corresponding categories are used: 2 hour postload glucose < 140 mg/dL is normal; ≥140 and < 200 mg/dL is impaired glucose tolerance; and ≥200 mg/dL is consistent with diabetes. A firm diagnosis of diabetes requires repeat testing on another day.
Variation in the reference ranges among obese patients may partially explain normal levels despite histologic disease. It is not uncommon for patients to
present with a complication of previously unrecognized cirrhosis despite long-standing medical care because these patients often lack the classic nutritional
changes of cirrhosis. Because of the association between NAFLD and insulin resistance, laboratory evaluation of insulin sensitivity may be reasonable during
the evaluation of patients with NAFLD.
Liver biopsy is the only means of assessing the presence and extent of specific necroinflammatory changes and fibrosis. However, firm recommendations of
when to perform a liver biopsy in the routine clinical setting have not yet been developed and care will continue to require individualization. A pragmatic
approach in younger patients without clinical evidence of more advanced disease is a trial period of increased exercise and improved dietary habits. The role of baseline or serial biopsy in patients with liver abnormalities while using statin drugs has yet to be established. The use of surrogate markers such as
aminotransferases and fibrosis markers may have a limited role in pilot studies but are not adequate end points for definitive investigations. The degree of
sampling error and the significance of occasional apoptotic bodies are 2 areas that have not been studied adequately. Likewise, the use of ubiquitin for the
detection of Mallory hyaline has yet to be explored fully. Megamitochondria with crystalline inclusions are seen commonly at electron microscopy and
form a part of the pleomorphic mitochondrial changes in NAFLD, which likely represent varied degrees of injury (free radical damage) and adaptation
The development of steatosis, steatohepatitis, progressive hepatic fibrosis, and cirrhosis is most likely the result of multiple metabolic abnormalities taking
place in the right (or wrong) genetic environment. The field has yet to develop a unifying framework that successfully organizes and reconciles the many
diverse observations made to date. In lieu of such a framework, this conference summary will work from the simple model shown in Fig. 2.
Fig. 2. A parsimonious model for the development of NAFLD
and NASH. This model places insulin resistance at the top as
an important antecedent metabolic abnormality in many
patients. In what fraction of patients this is true is yet to be
established. Other abnormalities, alone or with insulin
resistance, also may contribute to hepatic fat accumulation.
Excess fat in the liver predisposes to hepatocellular injury in
some individuals. This may be caused by the direct cellular
toxicity of excess free fatty acids, oxidant stress and lipid
peroxidation, or other mechanisms. Hepatocellular injury may
cause an inflammatory response with progressive fibrosis in a
subset of patients; the extent of this adverse outcome most
likely depends on a variety of environmental and genetic
influences. Portions of this conceptual model have been
referred to as the 2-hit hypothesis, accounting for the
accumulation of fat as the first hit and hepatocellular injury in
the fatty liver as the second hit. Some investigators also
have used a taxonomic distinction of secondary NASH, or that
attributable to readily identifiable drugs, toxins, or genetic
abnormalities, and primary NASH, or that which is not
secondary and is probably related to insulin resistance. The
value of these simplifications is uncertain.
The increased flux of free fatty acids through the liver in states of exuberant peripheral lipolysis may play a direct role in hepatocellular injury.
However, one of the difficulties in understanding the role of free fatty acids in hepatocellular injury is the lack of reliable methods to measure their intracellular levels. Interestingly, polyunsaturated fatty acids are highly facile substrates for lipid peroxidation in animal models of alcohol-induced liver injury whereas saturated fatty acids exert a protective effect.
If free fatty acids are agents of destruction in the pathogenesis of liver disease, abundant and overlapping protective mechanisms against this toxicity would be
expected. Indeed such mechanisms exist. Hepatocytes in particular are well endowed with mechanisms to bind, transform, catabolize, and export excess free fatty acids through the concerted actions of fatty acid binding proteins, triglyceride synthesis, and secretion as very low density lipoprotein (VLDL),
mitochondrial -oxidation, and enzymatic removal of lipid peroxidation products. The nuclear receptor PPAR (peroxisomal proliferator activated receptor ) plays a central role in sensing excess free fatty acids and up-regulating the genetic program of fatty acid disposal. The flip side of this protective mechanism, at least in rats, may be a predisposition to carcinogenesis.
Antioxidant agents and iron reduction therapy
Among the promising agents are vitamin E, S-adenosyl-methionine, betaine,and N-acetylcysteine. Betaine (a methyl donor in an alternative pathway for remethylation of homocysteine to methionine) has shown encouraging results in adults as has vitamin E in a pediatric population. Interestingly, the use of vitamin E in patients with coronary artery disease is associated with blunted efficacy of statin drugs. Silymarin, a popular milk thistle extract, is used commonly by patients with liver disease but we are not aware of published studies in NAFLD. Variable results were seen in one study of combination antioxidants. Histamine, which possesses indirect antioxidant properties, and a group of substances known as lazaroids (21-aminosteroids), may warrant pilot work. One study has shown improvement in liver enzyme levels and insulin sensitivity in a group of HFE gene-negative patients treated with serial phlebotomy for iron reduction.
Insulin therapy, sometimes recommended early in the course of type 2 diabetes, and sulfonylureas, have not been addressed adequately. The
thiazolidinediones have shown promise. These agents activate the PPAR nuclear transcription factor, alter skeletal muscle glucose uptake (through
increased GLUT4 activity), decrease central adiposity, promote adipocyte differentiation, alter mitochondrial mass, and alter thermogenesis. The
efficacy of troglitazone in lipodystrophy suggests a primary effect on lipid metabolism. Metformin has undergone limited study in NAFLD. It
down-regulates hepatic gluconeogenesis and also appears to divert fatty acids from triglyceride production to mitochondrial beta oxidation. Other
candidate agents include acarbose (an -glucosidase inhibitor), acipimox (inhibits lipolysis), and d-chiro-inositol.
Fibrates alter lipoprotein metabolism through the PPAR receptor but had no benefit in early reports. However, bezafibrate showed benefit in
tamoxifen-associated steatohepatitis. Basaranoglu et al. showed improvement in liver enzyme levels but histology was not measured in a study of
gemfibrozil. A pilot study has showed improvement in biochemical and histologic parameters in a small sample of patients treated with the HMG-CoA
reductase inhibitor atorvastatin. However, a recent report showed no significant histologic differences between controls and patients using various statin drugs. Recent reports of subclinical skeletal muscle toxicity characterized by formation of ragged red fibers and mediated by mitochondrial injury are justifiable cause for concern for the use of these drugs in NAFLD. Other lipid-lowering agents (such as colesevelam or other resin-binding agents) have not been investigated. The potential role of lipid-lowering agents is questioned by observations of inherent defects of apoprotein metabolism in NASH and NAFLD.
Liver transplantation and disease recurrence
Many patients with advanced disease are poor candidates for transplantation due to comorbid conditions such as obesity and complications of diabetes. Both
recurrence of NASH in patients with previously established NASH215-218 and de novo occurrence of NASH after transplantation for cryptogenic cirrhosis12,16
can occur. Posttransplantation progression to cirrhosis may develop although predictive factors and treatment have not been well defined. Immunosuppression could play a role due to the promotion of fatty liver and diabetes with corticosteroid use and more direct effects such as the effect of cyclosporine on the mitochondrion.
The Future: a call to action
Many issues remain unresolved regarding the diagnosis and treatment of NASH. Large series of well-characterized patients will need to be followed-up to better establish the natural history of NAFLD. To achieve this ambitious goal, collaborative groups such as the National Institutes of Health-supported NASH Clinical Research Network will need to pool their data for collective reporting of outcomes. Shown in Table 7 is a partial list of key areas identified during the
conference that need immediate clarification to move this field ahead toward effective diagnosis, prevention, and treatment of NASH.
Table 7. Controversial Areas and Future Research Goals
Epidemiology and risk factors
-- What is the prevalence of NAFLD among specific populations, especially among individuals with diabetes or hyperlipidemia?
--What are the risk factors for progression and what is the best means of NAFLD classification that accounts for these risk factors?
-- Is there a non-insulin-resistant group of individuals who have primary disorders that produce a clinical picture similar to that found in the metabolic
--What is the role of occupational exposures such as hydrocarbon fumes?
--What are the effects of modest alcohol consumption?
--Are there more sensitive serum markers for detecting NAFLD than the transaminases?
--What is the value of newer imaging modalities such as magnetic resonance spectroscopy that appear capable of sampling various areas of the liver and
measuring ATP homeostasis?
--What is the significance of immunologic markers such as antinuclear antibodies and IgA elevations, which are common in NAFLD?
-- What is the extent of sampling error on liver biopsy?
-- What is the spectrum of liver disease associated with insulin resistance?
-- What is the significance of predominantly portal-based inflammation in association with steatosis?
--What is the best measure of insulin sensitivity in patients with NAFLD?
-- Can we identify serum markers of NAFLD and hepatic fibrosis that reliably predict who has benign steatosis and who is at risk for hepatocellular injury and progressive fibrosis?
--Similarly, can we identify genetic markers that would predict who might be predisposed to either insulin resistance or to progressive liver disease?
--Is NAFLD a consequence of too much insulin signaling in the liver, too little, both, or neither?
--In states of insulin resistance associated with NAFLD, which of the insulin signaling pathways are impaired and in which tissues are these pathways relevant?
--Conversely, which of the insulin signaling pathways are overactivated by hyperinsulinemia and what are the metabolic consequences of this overzealous
--What are the relevant cytokines and other peptide mediators of insulin resistance in NAFLD?
--What is the role of increased free fatty acid levels in mediating cellular injury in NAFLD?
-- What is the role of increased free fatty acid levels in mediating insulin resistance in NAFLD?
--Is oxidant stress an important process in the pathogenesis of cellular injury and fibrosis in NAFLD or is it an epiphenomenonal consequence of cellular
--Why are the characteristic pathologic features of NASH often lost when it progresses to cirrhosis
--What are the most effective behavioral and pharmacologic approaches to insulin resistance in NAFLD?
--Are dietary polyunsaturated fats helpful or harmful to the liver?
--What are the effects on NAFLD of treating comorbid conditions such as diabetes and hyperlipidemia with sulfonylureas or statins?
The authors and course organizers are indebted to the following individuals who presented this material at the NASH Single Topic Conference:
Arun Sanyal, M.D., Medical College of Virginia; Joel Lavine, M.D., Ph.D., University of California at San Diego Medical Center, Joint Program Pediatric GI & Nutrition; Zobair M. Younossi, M.D., Inova Fairfax Hospital, Center for Liver Diseases; Elizabeth M. Brunt, M.D., Department of Pathology, St. Louis University; Gerald I. Shulman, M.D., Ph.D., Section of Endocrinology, Yale University; Nir Barzilai, M.D., Division of Endocrinology, Albert Einstein College of Medicine; Gökhan S. Hotamisligil, M.D., Ph.D., Harvard School of Public Health; Jerry Radziuk, M.D., Ph.D., Ottawa Hospital (Civic Campus); Abhimanyu Garg, M.D., Center for Human Nutrition, University of Texas Southwestern Medical Center; Nicholas O. Davidson, M.D., Division of Gastroenterology, Washington University; Nathan M. Bass, M.D., Ph.D., Division of Gastroenterology, University of California at San Francisco; Geoffrey C. Farrell, M.D., Storr Liver Unit, Westmead Millennium Institute; Christopher P. Day, M.D., Ph.D., Department of Medicine, University of Newcastle Medical School; Stephen H. Caldwell, M.D., Division of Gastroenterology and Hepatology, University of Virginia; Anna Mae Diehl, M.D., Johns Hopkins University; Brent A. Neuschwander-Tetri, M.D., Division of Gastroenterology and Hepatology, St. Louis University; Oliver F.W. James, F.R.C.P., University of Newcastle upon Tyne; Keith D. Lindor, M.D., Division of Gastroenterology and Hepatology, Mayo Clinic Foundation; Arthur L. Weltman, University of Virginia; Eve A. Roberts, M.D., F.R.C.P.C., University of Toronto, Hospital for Sick Children; and Ian R. Wanless, M.D., University of Toronto, Toronto General Hospital.