Book on hepatitis from page 488 to 499

Book on hepatitis from page 488 to 499

488  Hepatology 2012
28. Alcoholic Hepatitis
Claus Niederau
Health and social problems due to alcohol
overconsumption
Mortality due to alcohol overconsumption is high, in particular among young men
(Mokdad 2000). Alcohol overconsumption not only increases the risk for liver
disease but is also responsible for malignancies, accidents, violence, and social
problems (Bellentani 1997, Vaillant 1995). Alcohol consumption in excess of 20-30
g for women and 40-60 g for men per day markedly increases the risk for liver
disease (Becker 1996, Lucey 2008). However, liver cirrhosis is seen only in a
minority of subjects with high alcohol consumption; less than 10% of subjects who
drink more than 120 g of alcohol daily have cirrhosis (Bellentani 1997). In addition
to the level of alcohol consumption, various other factors such as sex, other genetic
characteristics and comorbidities contribute to the risk for liver disease (Nishigushi
1991, Becker 1996, Bellentani 1997, McCollough 1998, de Alwis 2007, Lucey
2009).
Classification and natural history of alcoholic liver
disease
Alcohol overconsumption most often causes fat accumulation of hepatocytes, called
hepatic steatosis (Figure 1). Alcohol-induced steatosis is in general reversible after
alcohol abstinence. Continued alcohol overconsumption in the presence of steatosis
markedly increases the risk for development of hepatitis, fibrosis and cirrhosis (Teli
1995, Cubero 2009). Patients with alcohol-induced cirrhosis have a significantly
increased risk for hepatocellular carcinoma (McCollough 1998). Patients with only
fatty liver in the absence of inflammation and fibrosis have a much lower risk for
development of cirrhosis than those with fatty liver plus presence of inflammation
and fibrosis. The latter group of patients with alcoholic fatty liver, inflammation and
fibrosis is defined as alcoholic steatohepatitis (ASH). The liver histology of patients
with ASH is similar when compared to patients with non-alcoholic steatohepatitis
(NASH) that is often associated with obesity and diabetes (Ludwig 1980, Brunt
1999).
Alcoholic Hepatitis  489
The diagnosis of ASH by liver biopsy thus helps to define the risk for
development of cirrhosis. The histological diagnosis of ASH however should not be
confused with the term “alcoholic hepatitis” that is also called “acute alcoholic
hepatitis” although its course can be a rather chronic one (Lucey 2009). This
overview article concentrates on “alcoholic hepatitis” which is a clinical diagnosis
of a rather acute development of jaundice and liver failure associated with a high
short-term mortality.
It is not exactly known which factor(s) set off the development of severe alcohol
hepatitis. In general, pathogenesis and individual predisposition are governed by
gene-environment interactions in all types of alcoholic liver disease (Figure 1).
Based on the “second hit” or “multiple hits” hypothesis, patients are predisposed to
progressive alcoholic liver disease when a specific combination of gene and
environmental interaction exists (Tsukamoto 2009). A loss or gain of function
genetic model has become a popular experimental approach to test the role of a gene
as a second hit. Significant interactions for progressive development of alcoholic
liver disease have been proven in particular for female gender, obesity, various
drugs, iron overload, and hepatitis B and C viral infections (Mueller 2009, Machado
2009, Cubero 2009). These factors may also interact in the development of
hepatocellular carcinoma (HCC).
Figure 1. Effects of alcohol overconsumption on the liver.
A liver biopsy in someone with “alcoholic hepatitis” is often similar to a
histological feature  of ASH. Most patients with histological features of ASH
however will not develop “alcoholic hepatitis”. Alcohol overconsumption leads to a
severe form of hepatitis and liver failure associated with a high short-term mortality
only in some subjects. Such alcoholic hepatitis may be seen with or without
preexisting cirrhosis.
490  Hepatology 2012
Clinical features and diagnosis of alcoholic
hepatitis
Alcoholic hepatitis is a clinical diagnosis characterized by the rapid development of
jaundice and liver failure most often due to long-term alcohol overconsumption
(Naveau 1997, McCollough 1998, Lucey 2009). Further characteristics include
fever, ascites, and in some patients hepatic encephalopathy as well. Usually the liver
is enlarged and tender. Women have a higher risk for alcoholic hepatitis than men
assuming that both genders drink the same amount of alcohol. The type of alcohol is
not associated with the risk. Prevalence was 20% in a cohort of 1604 patients who
had a history of heavy alcohol consumption and underwent a liver biopsy (Naveau
1997).
Laboratory tests show increases in serum aspartate aminotransferase (AST) to
approximately twice the upper limit of normal (ULN), while the increase in alanine
aminotransferase (ALT) is less pronounced. The ratio of AST to ALT is typically
>2 (Cohen 1979, Matloff 1980). Other laboratory abnormalities include increases in
peripheral leukocytes, serum bilirubin, and international normalized ratio (INR)
(Mathurin 2002, Orrego 1979). In the presence of an increase in serum creatinine
there  is a high risk for development of an often lethal hepatorenal syndrome
(Multimer 1993).
A liver biopsy usually shows big fat droplets and ballooning of hepatocytes that
may also include alcoholic hyaline (also called Mallory bodies); these changes are
accompanied by neutrophil infiltration and intrasinusoidal fibrosis (Figures 2 & 3)
(MacSween 1986).
Figures 2 & 3. Liver biopsies of alcoholic hepatitis.
The diagnosis of alcoholic steatohepatitis (ASH) requires the presence of fibrosis.
The role of liver biopsy in defining prognosis and treatment of alcoholic hepatitis in
the clinical setting remains unclear. Today, prognosis is usually not based on liver
biopsy but on clinical scoring systems (Lucey 2009).
Ultrasound is routinely done to look for hepatocellular carcinoma, biliary
obstruction, ascites, splenomegaly, portal vein thrombosis, and signs of portal
hypertension. Ascites should be checked for spontaneous bacterial peritonitis
routinely.
Differential diagnosis of alcoholic hepatitis includes severe non-alcoholic
steatohepatitis (NASH), acute or chronic viral hepatitis, drug-induced injury,
Alcoholic Hepatitis  491
autoimmune hepatitis, and Wilson’s disease. NASH shares the histological features
of ASH except for the rapid development of jaundice and liver failure.
After discontinuation of alcohol consumption the majority of patients will recover
from alcoholic hepatitis although jaundice, ascites and encephalopathy may persist
for weeks or months (Alexander 1971). Even so, a considerable percentage of
patients with alcoholic hepatitis still die today despite adequate treatment and
abstinence (Mathurin 2002, Orrego 1979).
Course and severity
Severe alcoholic hepatitis occurs in a small fraction of patients who overconsume
alcohol. The 28-day mortality is high and ranges from 30% to 50% in most cohorts
(Cohen 2009). Various scores have been used to predict the prognosis of alcoholic
hepatitis. Maddrey’s discriminant function (Maddrey 1978) and the Model for End-Stage Liver Disease (MELD; http://goo.gl/ksgu4) score may help to identify
patients who can benefit with corticosteriods. Most scores share some important
characteristics such as serum bilirubin and prothrombin time (Srikureja 2005).
Maddrey’s discriminant function is calculated as [4.6x (prothrombin time–control
prothrombin time, in seconds)]+serum bilirubin (mg/dL). A value of >32 indicates
severe alcoholic hepatitis and consequently calls for the use of corticosteroids
(Maddrey 1978). In two retrospective studies, the MELD score predicted short-term
mortality in alcoholic hepatitis as well as or even better than Maddrey’s
discriminant function (Dunn 2005, Srikureja 2005). A MELD score >21 was
associated with a 90-day mortality of 20%. The Lille score is based on pretreatment
data and on the response of serum bilirubin to a 7-day treatment with corticosteroids
and has been used to determine whether corticosteroids should be discontinued after
7 days because of treatment failure (Forrest 2005, Dunn 2005, Louvet 2007).
Patients with Maddrey’s discriminant function of <32 usually have mild disease
with a short-term survival of more than 90% and will not benefit from corticosteroid
treatment.
Investigators reported the results of a stepwise logistic-regression identifying
variables related to survival 1-4 months after hospital admission in patients with
alcoholic hepatitis (Forrest 2005); by using this data the Glasgow alcoholic hepatitis
score was developed (not to be confused with the Glasgow coma score). The score,
which includes age, peripheral leukocytes, urea nitrogen, bilirubin, and prothrombin
time, may help to identify high-risk patients who should receive corticosteroids.
Patients with a Maddrey’s discriminant function >32 and a Glasgow alcoholic
hepatitis score of >9 who were treated with corticosteroids had an 84-day survival
of 59%, while untreated patients had a 38% survival (Forrest 2007). In one study the
Glasgow score indicated which subgroup of patients with a high score of Maddrey’s
discriminant function would benefit from corticosteroid therapy (Forrest 2007).
Child-Pugh (CP) and MELD scores have been widely used to predict survival in
cirrhotic patients. Recent studies have suggested that the addition of serum sodium
to MELD (MELD-Na score) may improve its prognostic accuracy. Another recent
study compared the performance of CP, MELD, and MELD-Na scores in predicting
6-month mortality in patients with alcoholic cirrhosis, and developed a new
prognostic score. In this study two French centres (Boursier 2009) enrolled 520
patients (mean age 56.4±10.2 years) with alcoholic cirrhosis randomly allocated
492  Hepatology 2012
into two groups. MELD, MELD-Na1, and MELD-Na2 were calculated according to
UNOS recommendations. Frequencies of CP classes were: A - 29.6%, B - 25.8%, C
- 44.6%. Of the 520 patients 93 died during the 6-month follow-up. In the whole
population, the values of CP, MELD, MELD-Na1, and MELD-Na2 for prediction of
6-month mortality were similar. Multivariate analysis identified age, bilirubin, urea,
prothrombin time, sodium, and alkaline phosphatase as independent predictors of 6-month mortality. The score combining these 6 variables was named the Prognostic
Score for Alcoholic Cirrhosis (PSAC) and compared to the 4 other scores. The
predictive value of PSAC was better than all other scores except for MELD-Na2.
By stepwise multivariate analysis, PSAC was identified as independently associated
with 6-month mortality at the first step, and CP at the second. The new PSAC score
may improve the prognostic accuracy to predict the 6-month outcome (Boursier
2009).
Another recent study analyzed the outcome of 79 patients who were admitted to
an Intensive Care Unit (ICU) because of alcoholic liver disease (Rye 2009). The
value of various scores was analyzed for predicting mortality including the Acute
Physiology, Age and Chronic Health Evaluation (APACHE II), Sequential Organ
Failure Assessment (SOFA), CP, and MELD scores. The major reason for
admission was sepsis (44%). The observed mortality in the ICU was 49% and
hospital mortality 68%. Compared to survivors, non-survivors had a significantly
higher serum bilirubin, creatinine and prothrombin time, and lower GCS and length
of ICU stay. Survival was affected by cardiac arrest pre-admission (mortality 75%)
and number of organs supported (mortality 8% with no organ support, 79% ≥2
organs, 100% ≥3 organs). Renal replacement therapy was associated with 100%
mortality. Mortality due to GI bleeding was only 33%. Thus, cirrhotics admitted to
the ICU with cardiac arrest pre-admission, need for renal replacement therapy, or
multiple organ support have a poor prognosis. The predictive accuracy of SOFA and
MELD scores were superior to APACHE II and Child-Pugh scores in cirrhotic
patients (Rye 2009).
A further study analyzed the mortality of 105 patients presenting with alcoholic
hepatitis (Hussain 2009). Patients were evaluated by the modified discriminant
function (mDF) for alcoholic liver disease, CP score, and Glasgow alcoholic
hepatitis score (GAHS). Mean survival for those alive at the end of the study (n=36)
was 34.6 ± 17.8 months. Mean survival for those who died (n=50) was 13.2 ± 14.4
months. The mDF, CP and GAHS scores were significant predictors of mortality in
this population. Prothrombin time was also a significant predictor of mortality
(Hussain 2009).
Mechanisms of alcohol-related liver injury
Alcoholic liver disease is initiated by different cell types in the liver and a number
of different factors including products derived from alcohol-induced inflammation,
ethanol metabolites, and indirect reactions from those metabolites, as well as genetic
predisposition (Colmenero 2007). Ethanol oxidation results in the production of
metabolites that have been shown to bind and form protein adducts, and to increase
inflammatory, fibrotic and cirrhotic responses. Lipopolysaccharide (LPS) has many
deleterious effects and plays a significant role in a number of disease processes by
increasing inflammatory cytokine release. In alcoholic liver disease, LPS is thought
Alcoholic Hepatitis  493
to be derived from a breakdown in the intestinal wall enabling LPS from resident
gut bacterial cell walls to leak into the blood stream. The ability of adducts and LPS
to independently stimulate various cells of the liver provides for a two-hit
mechanism by which various biological responses are induced and result in liver
injury.
Alcohol (ethanol) can be oxidized by various enzymatic and non-enzymatic
pathways (Figures 2 and 3). In hepatocytes the most important pathway is oxidation
of ethanol via alcohol dehydrogenase (ADH) to acetaldehyde (Figure 4). In
mitochondria, acetaldehyde is converted to acetate and in turn acetate is converted
to acetyl CoA, which leads the two-carbon molecule into the TCA (tricarboxylic
acid cycle).
Figure 4. Oxidation of ethanol to acetaldehyde by enzymatic pathways.
This oxidation generates reducing equivalents, primarily reduced nicotinamide
adenine dinucleotide (NAD), i.e., NADH. The changes in the NADH–NAD+
potential in the liver inhibit both fatty acid oxidation and the TAC and may thereby
increase lipogenesis (You 2004a). Ethanol has also been shown to increase lipid
metabolism by inhibiting peroxisome-proliferator–activated receptor α  (PPARα)
and AMP kinase as well as by stimulation of sterol regulatory element-binding
protein (Fischer 2003, You 2004b, Ji 2006). All these mechanisms lead to hepatic
steatosis. Further enzymatic pathways of ethanol oxidation include catalase and the
“Microsomal Ethanol Oxidizing System” (MEOS), a cytochrome P450 component.
Oxidation of ethanol to acetaldehyde may also be due to non-enzymatic free radical
pathways (Figure 5). These include strong oxidizing intermediates such as the
hydroxyl radical which can abstract a hydrogen atom from ethanol, preferentially
producing the 1-hydroxyethyl radical; hypervalent iron complexes may also catalyse
this reaction without involvement of •OH (Reinke 1994, Welch 2002, Qian 1999).
Hydroxyethyl radicals may then react with oxygen to form a peroxy radical
intermediate which can rearrange to release acetaldehyde and superoxide.
Hydroxyethyl radicals can also react with proteins to produce antigenic adducts or
induce mitochondrial permeability transition (Clot 1995, Sakurai 2000).
There are probably various other mechanisms by which ethanol may cause or
contribute to liver disease. Ethanol increases the translocation of lipopolysaccharide
494  Hepatology 2012
(LPS) from the small and large intestines to the portal vein and on to the liver. In
Kupffer cells LPS can bind to CD14, which combines with toll-like receptor 4
(TLR4) thereby activating multiple cytokine genes (Schaffert 2009). In addition,
NADPH oxidase may release reactive oxygen species (ROS) that activate cytokine
genes within Kupffer cells, hepatocytes, and hepatic stellate cells. These cytokines
including TNF-α may cause fever, anorexia, and weight loss. Interleukin-8 and
monocyte chemotactic protein 1 (MCP-1) have been shown to attract neutrophils
and macrophages. Platelet-derived growth factor (PDGF) and transforming growth
factor b (TGF-b) contribute to the activation, migration, and multiplication of
hepatic stellate cells, thereby promoting liver fibrosis.
Figure 5. Oxidation of ethanol to acetaldehyde by non-enzymatic free radical pathways.
In the hepatocyte, ethanol is converted to acetaldehyde by the cytosolic enzyme
alcohol dehydrogenase (ADH) and the microsomal enzyme cytochrome P450 2E1
(CYP2E1). Acetaldehyde is converted to acetate. These reactions produce NADH
and inhibit the oxidation of triglycerides and fatty acids. ROS released by CYP2E1
and mitochondria cause lipid peroxidation. Inhibition of proteosomes due to ethanol
disturbs protein catabolism and may be partly responsible for the formation of
Mallory bodies. Reduction in enzymes that convert homocysteine to methionine
may increase homocysteine, thereby injuring the endoplasmic reticulum. Decrease
in binding of peroxisome proliferator–activated receptor a (PPAR-α) to DNA
reduces the expression of genes involved in fatty acid oxidation.
Glutathione transport from the cytosol into the mitochondria is reduced by
ethanol. Ethanol may also activate Fas and TNF receptor 1 (TNF-R1) thereby
activating caspase 8, causing mitochondrial injury and opening the mitochondrial
transition pore (MTP), releasing cytochrome c, and activating caspases; all these
processes contribute to apoptosis. Activation of TNF-R1 leads to nuclear factor
kappa B (NFkB) activation (Schaffert 2009).
Gut permeability and the circulating LPS endotoxin levels of the outer wall of
gram-negative bacteria are increased in patients with alcoholic liver injury (Uesugi
2002, Bjarnson 1984, Urbaschel 2001). In various animal studies alcohol exposure
promoted the transfer of LPS endotoxins from the intestine into portal blood (West
2005). Oral treatment with antibiotics reduced such increases in LPS endotoxins and
ameliorated alcoholic liver injury in animals (Uesugi 2001, Nanji 1994, Adachi
Alcoholic Hepatitis  495
1995). Activation of Kupffer cells by LPS endotoxins involves CD14, toll-like
receptor 4 (TLR4), and MD2 (Uesugi 2001, Akira 2001, Yin 2001). The
downstream pathways of TLR4 signalling include activation of early growth
response 1 (EGR1), NFkB, and the TLR4 adapter also called toll-interleukin-1
receptor domain-containing adapter inducing interferon-ß (TRIF) (McCuillen 2005,
Zhao 2008). TRIF-dependent signalling may contribute to alcohol-induced liver
damage mediated by TLR4 (Hritz 2008).
Many animal studies have shown that alcohol ingestion increases various markers
of oxidative stress (Meagher 1999, Wu2009). Studies in rats and mice suggest that
activated macrophages (Kupffer cells) and hepatocytes are the main sources of
alcohol-induced free radicals (Bailey 1998, Kamimura 1992). Oxidative stress may
mediate alcohol-induced liver injury, e.g., via cytochrome P450 2E1 (Mansuri 1999,
Lu 2008), leading to mitochondrial damage, activation of endoplasmic reticulum–
dependent apoptosis, and up-regulation of lipid synthesis (Ji 2003,  Yin 2001).
Activated Kupffer cells will also release TNF-α; this cytokine plays an important
role in the pathogenesis of alcoholic hepatitis. Circulating TNF-α concentrations are
higher in patients with alcoholic hepatitis than in heavy drinkers with inactive
cirrhosis, heavy drinkers who do not have liver disease and persons who do not
drink alcohol and who do not have liver disease (Adachi 1994, Bird 1990).
Circulating TNF-α concentrations are associated with high mortality (Bird 1990). In
animal studies, knockouts of the TNF receptor 1 and administration of the anti-TNF
agent thalidomide both ameliorated alcohol-induced liver injury (Yin 1999, Imuro
1997, Enomoto 2002). Ethanol was also shown to release mitochondrial cytochrome
c and to induce expression of the Fas ligand which may then cause apoptosis via the
caspase-3 activation pathway (Zhou 2001). Both TNF- and Fas-mediated signals
may increase the vulnerability of hepatocytes (Minagawa 2004).
Treatment
Abstinence from alcohol
After recovery from liver failure all patients with alcoholic hepatitis patients need to
have psychological and social support in order to assure continued abstinence (Saitz
2007).
Supportive therapy
There is still a lack of specific therapy for patients with alcoholic hepatitis although
prednisolone and pentoxifylline may have beneficial effects in severe disease. It is,
however, generally accepted that all complications and risks such as ascites,
encephalopathy, hepatorenal syndrome, and infections should be treated like other
decompensated liver diseases (Kosten 2003, Sanyal 2008, Lim 2008). The daily
protein intake should be at least 1.5 g/kg. Vitamin B1 and other vitamins should be
administered according to recommended references (Barr 2006).
Corticosteroids
Various studies and meta-analyses  show controversial results for the use of
corticosteroids in alcoholic hepatitis (Imperiale 1990, Christensen 1999, Imperiale
1999, Rambaldi 2008). In general, corticosteroids have not been shown to increase
survival, in particular during longer follow-up (Rambaldi 2008). However, there is
496  Hepatology 2012
evidence that corticosteroids do reduce mortality in a subgroup of patients with a
Maddrey’s discriminant function >32 or in those presenting with hepatic
encephalopathy (Rambaldi 2008). A meta-analysis of three studies corroborated that
corticosteroids given for 28 days increase 1-month survival by 20% in severe
alcoholic hepatitis (Maddrey’s discriminant function >32) (Mathurin 2002). In these
studies Maddrey’s discriminant function >32 resembled a MELD score of >21. In
most studies prednisolone was given at 40 mg a day for 28 days. In some studies
prednisolone was stopped completely at 28 days (Mathurin 2003), while the dose
was gradually reduced in other studies (Imperiale 1990). Corticoids should not be
given in the presence of sepsis, severe infection, hepatorenal syndrome, chronic
hepatitis B, or gastrointestinal bleeding (O’Shea 2006).
The mechanisms by which corticosteroids improve short-term survival in severe
alcoholic hepatitis are not fully understood. In general, corticosteroids inhibit
various inflammatory processes by acting on activator protein 1 and NFkB (Barnes
1997). In some studies in patients with alcoholic hepatitis, the administration of
corticosteroids was associated with a decrease in circulating levels of
proinflammatory cytokines such as interleukin-8, TNF-α and others (Taieb 2000,
Spahr 2001).
Recent reviews and recommendations conclude that corticosteroids should not be
given to patients with a Maddrey’s discriminant function <32 or a MELD score <21
until further data can identify patients with a high short-term risk (Lucey 2009).
Corticosteroids are thus ineffective in a large group of patients with alcoholic
hepatitis and probably do not affect long-term outcome. There is also evidence that
corticosteroids can be discontinued after 7 days if there is no obvious improvement
in clinical signs and symptoms and in serum bilirubin (Maddrey 1978, Dunn 2005,
Forrest 2005, Louvet 2007).
Pentoxifylline
Pentoxyfilline (400 mg TID for 28 days) reduced short-term mortality in severe
alcoholic hepatitis (Maddrey’s discriminant function >32) in a randomized,
controlled trial; mortality was 24% in the pentoxifylline group and 46% in the
placebo group (p<0.01) (Akrivadis 2000). This trial did not include a group on
corticosteroid treatment. Although the phosphodiesterase inhibitor pentoxifylline
has been suggested to act as an anti-TNF agent, TNF-α concentrations did not differ
significantly between the two groups. Thus, the mechanisms by which
pentoxifylline may improve the prognosis in alcoholic hepatitis remains unknown.
Interestingly, almost all deaths (22 of 24; 92%) in the placebo group were
associated with hepatorenal syndrome while hepatorenal syndrome was considered
the cause of death in only 6 of 12 patients (50%) in the pentoxifylline group. Thus,
one might speculate that pentoxifylline may exert its beneficial effects by
preventing the development of hepatorenal syndrome. A recent study (De BK 2009)
compared the efficacy of pentoxifylline and prednisolone in the treatment of severe
alcoholic hepatitis. 68 patients with severe alcoholic hepatitis (Maddrey score >32)
received pentoxifylline (400 mg TID for 28 days) (n=34) or prednisolone (40 mg
QD for 28 days) (n=34) for 28 days in a randomized double-blind controlled study,
and subsequently in an open-label study (with a tapering dose of prednisolone) for a
total of 3 months, and were followed over a period of 12 months. Twelve patients in
the corticosteroid group died by the end of month 3 in contrast to five patients in the
Alcoholic Hepatitis  497
pentoxifylline group (mortality 35.3% vs 14.7%, p=0.04). Six patients in the
corticosteroid group but none in the pentoxifylline group developed hepatorenal
syndrome. Pentoxifylline was associated with a significantly lower MELD score at
the end of 28 days of therapy when compared to corticosteroids (15.5 ± 3.6 vs 17.8
± 4.6, p=0.04). Reduced mortality, improved risk:benefit profile and renoprotective
effects of pentoxifylline compared with prednisolone suggest that pentoxifylline is
superior to prednisolone for treatment of severe alcoholic hepatitis. Interestingly,
another recent study showed that long-term pentoxifylline therapy effectively
achieved sustained biochemical improvement and even histological improvement in
non-alcoholic steatohepatitis (Satapathy 2007).
N-acetyl cysteine
A multicentre, randomised, controlled trial (Nguyen-Khac 2009) analysed treatment
of severe acute alcoholic hepatitis via corticoids plus N-acetyl cysteine (C+NAC)
versus corticoids (C) alone. The background to this approach was the hypothesis
that the glutathione precursor NAC may rebuild antioxidant stocks in the
hepatocyte. Deaths were significantly lower in the C+NAC group than in the C
group at month 1 (n=7/85 (8.2%) vs. 21/89 (23.6%), p=0.005) and at month 2
(13/85 (15.3%) vs. 29/89 (32.6%), p=0.007) but not at month 3 (19/85 (22.4%) vs.
30/89 (33.7%), p=0.095) or at month 6 (23/85 (27.1%) vs. 34/89 (38.2%). NAC
may improve short-term survival. This improvement, however, is lost by month 3.
Anti-TNF-α therapy
Some smaller studies have shown beneficial results using the TNF-α receptor
antagonists infliximab and etanercept in patients with acute alcoholic hepatitis
(Spahr 2007, Mookerjee 2003, Tilg 2003, Menin 2004). A larger randomized,
controlled clinical trial compared the effects of infliximab plus prednisolone vs
placebo plus prednisolone in patients with severe alcoholic hepatitis (Maddrey’s
discriminant function >32) (Naveau 2004). The trial was stopped early by the safety
monitoring board because of a significant increase in severe infections and a
(nonsignificant) increase in deaths in the infliximab group. Similarly, etanercept
reduced 6-month survival when compared with placebo in a randomized, placebo-controlled trial (Boetticher 2008). Thus, TNF-α receptor antagonists should not be
used for clinical therapy of alcoholic hepatitis (Lucey 2009).
Nutritional support
Many patients with alcoholic hepatitis have signs of malnutrition associated with
high mortality (Mendenhall 1984, Mendenhall 1986, Stickel 2003). Parenteral and
enteral nutrition have been shown to improve malnutrition in alcoholic hepatitis but
has not improved survival (Mendenhall 1984). A randomised, controlled clinical
trial looked at the effects of enteral nutrition of 2000 kcal/day via tube feeding
versus treatment with 40 mg/day prednisolone for 28 days in severe alcoholic
hepatitis. Survival in both groups was similar after one month and one year. It may
be concluded that nutritional support is as effective as corticosteroids in some
patients (Cabre 2000). However, corticoids in many studies failed to improve long-term survival.
498  Hepatology 2012
Other pharmacologic treatments
The anabolic steroid oxandrolone failed to improve survival in patients with
alcoholic hepatitis (Mendenhall 1984). Numerous studies have shown that alcoholic
hepatitis is accompanied by oxidative stress. So far, all studies with antioxidants
such as vitamin E, silymarin (milk thistle) and others have failed to improve
survival in alcoholic hepatitis (Pares 1998, Mezey 2004). Older studies did show
that colchicine, propylthiouracil, insulin and glucagon failed to improve survival in
alcoholic hepatitis (Lucey 2009).
Liver transplantation
In current guidelines for liver transplantation, the patient needs to have at least a 6-month period of alcohol abstinence before they can be evaluated for transplantation,
thus alcoholic hepatitis is usually a contraindication for liver transplantation (Lucey
1997, Everhardt 1997, Lucey 2007).
Figure 6. Treatment algorithm in alcoholic hepatitis.
Summary
Alcoholic hepatitis is a clinical diagnosis based on a history of heavy alcohol
consumption, jaundice, other signs of liver failure, and the absence of other causes
of hepatitis. A liver biopsy may be helpful but is not required either to determine the
diagnosis or prognosis. Abstinence from alcohol is the prerequisite for recovery.
Patients with signs of malnutrition should have adequate nutritional support.
Subjects with severe alcoholic hepatitis (Maddrey’s discriminant function >32 or
MELD score >21) who do not have sepsis or other corticosteroid contraindications
Alcoholic Hepatitis  499
may receive 40 mg prednisolone daily for 28 days (McCullough 1998, Lucey 2009).
A treatment algorithm is shown in Figure 6. After 7 days of corticosteroid treatment,
patients without obvious clinical benefit, without significant improvement of
jaundice and with a Lille score >0.45 may have disease that will not respond to
continued treatment with corticosteroids or an early switch to pentoxifylline (Louvet
2008). In situations where administration of corticosteroids appears to be risky,
pentoxifylline may be tried (Lucey 2009); this drug may decrease the risk of
hepatorenal syndrome that is often lethal in alcoholic hepatitis. Patients with less
severe alcoholic hepatitis have a good short-term survival of >90% and should not
be treated with corticosteroids or pentoxyfilline (Mathurin 2002).
References
Agarwal K, Czaja AJ, Jones De, et al. Cytotoxic T lymphocyteantigen-4 (CTLA-4) gene
polymorphisms and susceptibility to type 1 autoimmune hepatitis. Hepatology
2000;31:49-53. (Abstract)
Ahmed M, Mutimer D, Hathaway M, et al. Liver transplantation for autoimmune hepatitis: a 12-year experience. Transplant. Proc. 1997;29:496. (Abstract)
Alvarez F, Berg PA, Bianchi FB, et al. International Autoimmune Hepatitis Group Report: review
of criteria for diagnosis of autoimmune hepatitis. J. Hepatology 1999;31:929-38.
(Abstract)
Alvarez F, Ciocca M, Canero-Velasco C, et al. Short-term cyclosporine induces a remission of
autoimmune hepatitis in children. J. Hepatol. 1999;30:222-7. (Abstract)
Angulo P, Smith C, Jorgensen RA. Budesonide in the treatment of patients with primary biliary
cirrhosis with suboptimal response to ursodeoxycholic acid. Hepatology 2000;20:471-90.
Bach N, Bodian C, Bodenheimer H, et al. Methotrexate therapy for primary biliary cirrhosis. Am
J Gastroenterol 2003;98:187-93. (Abstract)
Bandin O, Courvalin Jc, Poupon R, et al. Specificity and sensitivity of gp210 autoantibodies
detected using an enzyme-linked immunosorbent assay and a synthetic polypeptide
in the diagnosis of primary biliary cirrhosis. Hepatology 1996;23:1020-4. (Abstract)
Beaune Ph, Lecoeur S, Bourdi M, et al. Anti-cytochrome P450 autoantibodies in drug-induced
disease. Eur J Haematol Suppl 1996;60:89-92. (Abstract)
Berg PA, Doniach D, Roitt IM. Mitochondrial antibodies in primary biliary cirrhosis. I.
Localization of the antigen to mitochondrial membranes. J Exp Med 1967;126:277-90. (Abstract)
Berg PA, Klein R. Mitochondrial antigen/antibody systems in primary biliary cirrhosis: revisited.
Liver 1995;15:281-92. (Abstract)
Berg PA, Klein R. Mitochondrial antigens and autoantibodies: from anti-M1 to anti-M9. Klin
Wochenschr 1986;64:897-909. (Abstract)
Bergquist A, Broome U. Hepatobiliary and extra-hepatic malignancies in primary sclerosing
cholangitis. Best Pract Res Clin Gastroenterol 2001;15:643-56. (Abstract)
Bergquist A, Ekbom A, Olsson R, et al. Hepatic and extrahepatic malignancies in primary
sclerosing cholangitis. J Hepatol 2002;36:321-7. (Abstract)
Beuers U, Rust C. Overlap syndromes. Semin Liver Dis 2005;25:311-20. (Abstract)
Bloch DB, Chiche JD, Orth D, et al. Structural and functional heterogeneity of nuclear bodies.
Mol Cell Biol 1999;19:4423-30. (Abstract)
Bluthner M, Schafer C, Schneider C, et al. Identification of major linear epitopes on the sp100
nuclear PBC autoantigen by the gene-fragment phage-display technology.
Autoimmunity 1999;29:33-42. (Abstract)
Boberg KM, Aadland E, Jahnsen J, et al. Incidence and prevalence of primary biliary cirrhosis,
primary sclerosing cholangitis, and autoimmune hepatitis in a Norwegian population.
Scand. J. Gastroenterol. 1998;33:99-103. (Abstract)
Boberg KM, Bergquist A, Mitchell S, et al. Cholangiocarcinoma in primary sclerosing
cholangitis: risk factors and clinical presentation. Scand J Gastroenterol
2002;37:1205-11. (Abstract)