Alcoholic Liver Disease

Alcoholic liver disease.


The incidence of alcoholic liver disease (ALD) follows the trend of per capita alcohol consumption, with the spectrum of hepatic injury ranging from the preliminary stage of fatty liver to alcoholic hepatitis and/or cirrhosis. It is not clear how alcohol causes liver disease, but likely mechanisms include (1) oxidative stress and acetaldehyde produced during ethanol metabolism, and (2) innate and adaptive immune responses. Factors determining the variable susceptibility to liver disease in heavy drinkers include a variety of host and environmental factors which affect these mechanisms.

Clinical manifestations can be extremely variable, ranging from completely asymptomatic to severe hepatic failure. Although patients can present with a life-threatening complication, most often they do so with symptoms unrelated to the liver, typically nonspecific digestive symptoms or vague psychiatric complaints. The key to the early recognition of alcohol-related disease is having a high index of suspicion, with confirmation by

(1) direct questioning for alcohol history and alcohol-related symptoms;

(2) clinical examination for signs of chronic liver disease;

(3) supportive investigations, including elevation of serum γ-glutamyl transferase (γGT) and aspartate transaminase (AST); and

(4) liver biopsy, which is often required for accurate prognostication, revealing alcoholic fatty liver, alcoholic hepatitis, or cirrhosis.

Management is governed by the stage and severity of the liver disease, but always includes abstinence and adequate nutritional support. Corticosteroids and pentoxifylline can reduce mortality in selected patients with severe acute alcoholic hepatitis. Transplantation remains the only effective treatment for alcoholic cirrhosis, although this remains controversial, principally due to concerns over the risk of post-transplant recidivism.

Alcoholic liver disease in detail - technical


It is likely that humans have been consuming alcohol in excess since the Stone Age, and the ancient Greeks made the link between alcohol and liver damage. In spite of this association, the consumption of potentially hazardous quantities of alcohol persists and alcoholic liver disease (ALD) is currently the most common cause of liver disease in the Western world. In some countries consumption is increasing. The magnitude and range of the health and socioeconomic problems attributable to alcohol abuse are enormous, but this article will primarily focus on pathogenesis, epidemiology, and treatment.


There are several pathogenetic mechanisms through which ethanol can cause liver injury. These include the direct effect of ethanol metabolism on liver biochemistry and hepatocyte functioning, the release of cytokines as a result of portal endotoxinemia, and liver-directed adaptive immune responses generated towards new antigens formed through these first two mechanisms.

Ethanol metabolism and pathogenesis

Ethanol metabolism

Over 90% of circulating alcohol is oxidatively metabolized, primarily in the liver, and excreted as CO2 and water.

Alcohol oxidation in the liver takes place via three steps. First alcohol is oxidized, principally within the cytosol, to acetaldehyde. Then acetaldehyde is further oxidized to acetate, primarily within the mitochondria, and finally, acetate is released into the blood and oxidized to CO2 and water in peripheral tissues. At least three enzyme systems with the capacity to oxidize alcohol to acetaldehyde are present within the liver, although in normal individuals only the alcohol dehydrogenase enzymes are important.

ADHs catalyse the oxidation of a variety of alcohols to aldehydes and ketones. This includes catalysing the oxidation of ethanol to acetaldehyde and transferring hydrogen to the cofactor nicotinamide adenine dinucleotide (NAD) which is converted to its reduced form, NADH. In addition to ADH, alcohol is metabolized by the microsomal ethanol oxidizing system (MEOS), an accessory pathway that principally involves a specific alcohol-inducible form of cytochrome P450 designated CYP2E1. Its K m for alcohol is in the order of 50 to 80 mg/100 ml, so in view of its inducibility it appears to play an important role at high blood alcohol levels or following chronic alcohol abuse.

The third pathway for alcohol oxidation is catalysed by the enzyme catalase. This enzyme is located in the peroxisomes of most tissues and the pathway accounts for less than 2% of overall in vivo alcohol oxidation.

Acetaldehyde is further oxidized in the liver to acetate by aldehyde dehydrogenases (ALDHs). ALDH, like ADH, uses NAD as a cofactor and further increases the NADH/NAD ratio. ALDH2 is responsible for the majority of acetaldehyde oxidation and exists in at least two allelic forms, ALDH2*1 and ALDH2*2, with the gene product of the ALDH2*2 allele (found in 50% of east Asians) having little or no catalytic activity. Homozygotes for the ALDH2*2 allele develop an unpleasant ‘flushing’ reaction after alcohol that is thought to be due to acetaldehyde. ALDH inhibitors such as disulfiram (Antabuse) use this mechanism in the treatment of alcoholism.

Ethanol metabolism may result in the generation of fatty liver (steatosis), oxidative stress/lipid peroxidation, and acetaldehyde, all of which are thought to be important in disease pathogenesis.

Pathogenesis of fatty liver

Accumulation of triacylglycerol is the consequence of increased substrate (free fatty acids and glycerol 3-phosphate) supply, increased esterification, and decreased export from the liver. Alcohol intake increases availability of free fatty acids through the lipolysis of adipose tissue, impairment of gluconeogenesis, increased fatty acid synthesis, and reduced oxidation. The altered NADH/NAD ratio generated during ethanol metabolism also increases the production of glycerol 3-phosphate, thereby again promoting triacylglycerol synthesis. The decrease in export of triacylglycerol from the liver appears to be primarily due to ethanol-induced down-regulation of microsomal triglyceride transfer protein (MTP), the principal enzyme responsible for packaging triacylglycerol and apolipoprotein B (apoB) into very low-density lipoprotein (VLDL) particles.

Oxidative stress and lipid peroxidation

An accumulating body of evidence now supports a role for oxidative stress and lipid peroxidation in the pathogenesis of ethanol-induced liver injury, though considerable controversy remains regarding the most important source of reactive oxygen species in ALD. The most likely candidates are microsomal CYP2E1 (the only source of hydroxyethyl radicals), the mitochondrial electron transport chain, inducible nitric oxide synthase, and Kupffer cells. Any liver injury due to oxidative stress is compounded by the depletion in antioxidant defences found after chronic alcohol consumption.


It has been known for some time that acetaldehyde can form Schiff bases with the valine, lysine, and tyrosine residues on cellular proteins, resulting in both stable and unstable adduct formation. This may disrupt protein function and can result in the production of immunodominant antigenic determinants. In addition, acetaldehyde mediates the ethanol-induced susceptibility of hepatocytes to cytotoxicity induced by tumour necrosis factor-α (TNFα).

Innate immune system

Endotoxin, which refers collectively to the lipopolysaccharide components of the cell wall of all Gram-negative bacteria, appears to play a central role in the development of ALD. Ethanol ingestion increases the translocation of endotoxin from the gut lumen to the portal circulation where it induces macrophages (Kupffer cells) to produce cytokines and Reactive oxygen species. These Kupffer cells are the primary intrahepatic source of TNFα, a cytokine believed to be central to disease pathogenesis.

Adaptive immune system

Several clinical features imply that adaptive immune mechanisms may have a role in disease pathogenesis. Patients often display hypergammaglobulinemia, and lymphocyte infiltration is a well-recognized histological feature of advanced disease. Studies to date have revealed that there are specific antibody and cellular responses to proteins formed during ethanol metabolism and in some patients there is also autoantibody formation. Further work is required to determine whether these specific immune responses are an important part of disease progression or an epiphenomenon.


It is still not clear why only around one-third of heavy drinkers develop alcoholic hepatitis and fewer progress to cirrhosis. The risk of developing ALD appears to begin at around 30 g/day of ethanol, but only around 5% of the individuals drinking this much show signs of liver disease. There is evidence to suggest that risk is higher in women and among individuals over 50 if alcohol is drunk outside mealtimes or consumed in a variety of different beverages. The highest risk is seen in drinkers who do not include wine in their drinking repertoire. No dose of alcohol confers a guarantee of developing cirrhosis, regardless of the length of time over which it is consumed, and relatively low doses can cause problems.

It is now clear that obesity and associated hyperglycaemia increase the incidence of all stages of ALD in heavy drinkers. Animal studies coupled with epidemiological data point to the fact that dietary fats, particularly polyunsaturated fats, may have a role in disease progression.

Evidence for the importance of genetic factors comes from a study showing that concordance rates for alcoholic cirrhosis are three times higher in monozygotic than in dizygotic twin pairs. This is not entirely explained by the difference in concordance rates for alcoholism per se and associations have been found between disease and polymorphisms in genes influencing the severity of steatosis (microsomal triglyceride transfer protein), ethanol metabolism (ADH2 and ALDH2), and cytokine production (TNFα and IL-10 promoters).


During a period of wine rationing in France during the Second World War, cirrhosis deaths reduced by 80%. A similar effect was observed during the Prohibition era in the United States of America (1919–1933). These and other studies have highlighted the close correlation between deaths from cirrhosis and per capita alcohol consumption. The worldwide increase in mortality from cirrhosis observed during the 1950s and 1960s was associated with a similar rise in alcohol consumption, attributed largely to the falling price of alcohol relative to income. This reduction in real price, the increased availability of alcohol, and heavy promotion by the drinks industry have lead to a doubling of alcohol consumption in the United Kingdom since 1960. This increase in consumption has been followed by a dramatic increase in cirrhosis mortality in the United Kingdom between the late 1980s and 2000. This contrasts sharply with a reduction in mortality in most other European countries and the United States of America. These latest data closely link cirrhosis mortality with a rise in alcohol consumption.

Clinical features


Alcohol-related liver damage is a spectrum, with the various lesions (fatty liver (steatosis), alcoholic hepatitis, and cirrhosis) occurring more commonly in combination than in isolation. The clinical manifestations can be extremely variable, ranging from completely asymptomatic to severe hepatic failure.

Initially, patients most commonly present with symptoms unrelated to the liver, typically nonspecific digestive symptoms or vague psychiatric complaints. The key to the early recognition of patients with alcohol-related disease is a high index of suspicion. Once the diagnosis is suspected it is usually easy to confirm by direct questioning for alcohol history and alcohol-related symptoms, careful clinical examination, and supportive laboratory investigations.

Important features to note on examination are the signs of chronic liver disease including hepatomegaly and signs indicative of alcohol-related pathology in other organs such as hypertension, atrial fibrillation, and a cushingoid appearance. Many of the classical signs of chronic liver disease, including spider nevi, Dupuytren’s contractures, palmar erythema, and parotid swelling, can occur in alcoholics in the absence of cirrhosis. Clinical signs and history cannot be relied upon to distinguish the various histological subtypes of ALD, since patients with cirrhosis can be asymptomatic while patients with hepatocellular failure may have only severe fatty change.

Biochemical and haematological tests can suggest the presence of alcohol abuse and indicate the presence of liver damage, but are not useful in determining the severity of the histological lesion. Elevation of γ-glutamyl transferase (γGT) has been reported in up to 90% of patients abusing alcohol but is not specific for alcohol abuse. Its main clinical use is probably in monitoring a period of supposed abstinence, since it falls within a week of cessation of drinking. A raised mean corpuscular volume (MCV) occurs in 80 to 100% of alcoholics with and without liver disease and is due to a direct toxic effect of alcohol on the marrow.

With regard to biochemical markers of alcohol-related liver damage, a rise in serum aspartate transaminase activity (AST) of up to five times normal is common in patients abusing alcohol and reflects the presence, but not the severity, of liver damage. Alanine transaminase (ALT) activity is raised less often than AST, and the AST/ALT ratio is usually greater than 1.

Liver biopsy is useful to accurately stage alcohol-related liver disease. Although many clinicians withhold biopsy, it is often required for accurate prognostication.

Alcoholic fatty liver

Alcoholic fatty liver is indistinguishable histologically from nonalcoholic fatty liver (NAFLD) associated with the metabolic syndrome. Patients with fatty liver are usually asymptomatic or present with nonspecific digestive symptoms. Rarely fatty liver may be associated with hyperlipidaemia, haemolytic anaemia and jaundice (Zieve’s syndrome, discussed below) or hepatic failure. Smooth, nontender hepatomegaly is usually the only clinical finding and the γGT, AST, and MCV are often mildly raised.

A growing body of evidence suggests that, rather than being an epiphenomenon of excessive alcohol intake, steatosis may play a direct role in progression to more advanced disease. In several prospective studies of heavy drinkers the severity and pattern of steatosis on index biopsy predicts the subsequent risk of fibrosis and cirrhosis.

Alcoholic hepatitis

Alcoholic hepatitis consists of a constellation of histological abnormalities including ballooning degeneration of hepatocytes, Mallory bodies, and a neutrophil inflammatory cell infiltrate. In addition there is pericellular and perivenular fibrosis, producing a ‘chicken-wire’ appearance. Progression to cirrhosis is associated with the extent and degree of fibrosis, a panlobular distribution, and widespread Mallory body formation.

There is no good correlation between the severity of the histological lesion and the clinical presentation, although patients with severe histology usually present with symptoms specifically related to hepatocellular failure such as jaundice, ascites and encephalopathy, or variceal bleeding. Clinical decompensation may be precipitated by vomiting, diarrhoea, anorexia, increased alcohol intake, or intercurrent infection. The majority of patients have tender, smooth, hepatomegaly and signs of chronic liver disease may be present.

Blood abnormalities include decreased albumin and increased γGT, AST, bilirubin, alkaline phosphatase, and prothrombin time (PT). Blood urea and serum sodium and potassium are all low, unless hepatorenal syndrome supervenes, and hypoglycaemia may be present. Macrocytic anaemia, neutrophil leucocytosis, and thrombocytopenia are present in all but the mildest cases. Patients with severe alcoholic hepatitis often rapidly deteriorate in the days immediately following hospital admission. The pathophysiological basis of this is not clear.

Clinical and laboratory variables have been used to derive scores to predict short-term mortality in alcoholic hepatitis. The most widely used is the discriminant function which is based on PT and bilirubin only. This has been confirmed as a useful predictor of mortality prospectively. The recently derived Glasgow Alcoholic Hepatitis Score (GAHS) incorporates INR, bilirubin, age, creatinine, and white cell count. Although this scoring system needs validation in large numbers prospectively, initial work suggests that it is more sensitive and specific than the discriminant function at determining prognosis. If the patient survives to hospital discharge, then the long-term prognosis is determined by the initial histology, the progression to cirrhosis, and subsequent drinking behaviour.


With progressive injury, fibrous septa link hepatic and portal veins, and regenerative nodules eventually appear. This cirrhosis is usually micronodular and frequently reverts to a macronodular cirrhosis with abstention. Clinical presentation can range from asymptomatic hepatomegaly to hepatic failure and the complications of portal hypertension such as ascites or variceal bleeding. Presentation with severe hepatic decompensation usually implies the presence of continued drinking or the development of hepatocellular carcinoma or portal vein thrombosis. The clinical findings will depend on the presence of portal hypertension or encephalopathy and do not differ significantly from those observed in other forms of cirrhosis. Patients with compensated cirrhosis, particularly if abstinent from alcohol, can have completely normal laboratory investigations, while patients with continued intake will have a similar range of abnormal laboratory investigations to those seen in patients with alcoholic hepatitis. A raised α-fetoprotein suggests the presence of hepatocellular carcinoma and indicates the need for further investigations.

The survival of patients with alcoholic cirrhosis is determined by the clinical and histological severity of the disease at presentation and their subsequent drinking behaviour. In patients with alcoholic cirrhosis hepatocellular carcinoma occurs at a rate of 1 to 2%/year. It is most common in abstaining men.


Achieving abstinence

Measures aimed at establishing and maintaining abstinence are critical in the management of patients with ALD. This is best achieved by close liaison between liver physicians and addiction psychiatrists with support from specialist alcohol nurses and trained counsellors. Available treatments for alcohol-dependent patients can be divided into psychological and pharmacological (Table 1 below). ‘Brief intervention’ involves educating patients about the nature of their problem and providing then with advice on how to change their behaviour. In spite of the apparent simplicity of this form of management, brief interventions have been shown to significantly increase the chances of heavy drinkers moderating their drinking at 6 and 12 months in an outpatient setting. Cognitive-behavioural therapy and motivational enhancement therapy have also both been shown to reduce drinking in dependent patients in a randomized controlled trial.

Table 1 Therapeutic options in patients with alcoholic liver disease


Target Treatment
Addiction Nonpharmacological
 Brief intervention
 Cognitive therapy
 Motivational enhancement therapy
Alcoholic hepatitis Probable benefit
 Pentoxifylline (one RCT only)
Possible benefit
Alcoholic cirrhosis Confirmed benefit
 Liver transplantation
Possible benefit

Both acamprosate and naltrexone aim to reduce alcohol craving and have been shown to reduce drinking days and increase abstinence rates. Disulfiram, an inhibitor of acetaldehyde dehydrogenase, has been used for many years with conflicting results. The only treatment to undergo assessment in drinkers with liver disease has been the GABAB-receptor antagonist baclofen, which has shown promise in small numbers with short follow-up.

Up to 50% of patients will either abstain completely or achieve a significant reduction in intake after being given simple advice by physicians during their initial presentation, with a significant improvement in survival compared to continued heavy drinkers.

Alcoholic hepatitis

Almost all treatment trials in patients with alcoholic hepatitis have been short-term (usually 1 month) and restricted to patients with a discriminant function greater than 32 and/or encephalopathy. Many treatment modalities have been tried in patients with alcoholic hepatitis, but none has achieved consensus status among practising hepatologists.


Steroids are aimed at suppressing the hepatic inflammatory response seen in liver biopsies from patients with severe alcoholic hepatitis. Concern over adverse effects coupled with a continued uncertainty over efficacy has contributed to the reluctance of many clinicians to prescribe steroids for patients with alcoholic hepatitis. This uncertainty may be due to the fact that initial studies were often poorly designed and included patients with a variety of disease severities. The most recent meta-analysis pooled the individual patient data from the three large randomized controlled trials, only including patients with encephalopathy and/or a discriminant function greater than 32. This study showed that steroids improved survival vs placebo (85% vs 65%), with increasing age and creatinine independent predictors of mortality on multivariate analysis. For these patients 40 mg prednisolone for 28 days with a 2-week taper is recommended. If the bilirubin has not fallen by the seventh day of steroid treatment the prognosis is very poor and alternative treatments should be considered. Steroid treatment is still relatively contraindicated in the large number of patients with concomitant infection and gastrointestinal bleeding.

Pentoxifylline is a nonselective phosphodiesterase inhibitor that is approved for use in claudication due to its effect on red blood cell deformability, but also has an anticytokine effect attributed to a reduction in TNFα gene transcription. In the only randomized controlled trial of pentoxifylline, the effective claudication dose (400 mg three times a day) led to a 40% reduction in 28-day mortality in patients with a discriminant function greater than 32. Importantly, almost all of the improvement in survival was due to a fall in mortality from hepatorenal syndrome, suggesting that pentoxifylline may have a specific beneficial effect in alcoholic hepatitis patients developing this ominous complication. Clearly, further trials are needed.

Protein–calorie malnutrition is seen in patients presenting with acute severe alcoholic hepatitis and there is a correlation between the severity of malnutrition and mortality. Enteral tube feeding with an energy-dense formula can reduce this mortality and is advised for all these patients.

In spite of the likely role of oxidative stress in disease pathogenesis there is evidence that antioxidants are ineffective in acute alcoholic hepatitis. Likewise, insulin and glucagon, anabolic steroids, and propylthiouracil have no proven role in treatment.

The anti-TNFα drugs infliximab and etanercept have been trialled in acute severe alcoholic hepatitis because of the putative role of TNFα in disease pathogenesis. Although the only large randomized trial using high-dose infliximab in combination with steroids was stopped as a result of a higher mortality in the treatment arm, these drugs may yet prove to have some role. Likewise, the molecular adsorbents recycling system (MARS) has shown promise in preliminary studies and further data is awaited.

A major improvement in survival in patient with severe disease has been seen with the use of glypressin and albumin for hepatorenal syndrome.

Alcoholic cirrhosis

At present the management of patients with advanced fibrotic liver disease is directed at preventing and treating the complications of portal hypertension, liver failure and hepatocellular carcinoma and deciding if and when to consider patients for orthotopic liver transplantation.

No medical therapy for alcoholic cirrhosis has yet reached clinical practice, but some agents have shown initial promise. These will need further trials before general use. Cirrhotic patients who continue to drink moderately may benefit from propylthiouracil, an antithyroid drug that targets the hypermetabolic state seen in ALD. A second agent, S-adenosylmethionine, acts both as an antioxidant by replenishing glutathione and as a methyl donor maintaining cell membrane fluidity. In one trial, a significant beneficial effect of treatment has been found in patients with Child’s A and B cirrhosis. Clearly further trials with this agent are awaited with interest. After promising results in baboons, a large, long-term study of phosphatidylcholine in humans with cirrhosis was negative, possibly due to the number of drinkers who dramatically reduced their consumption while under observation.

Liver transplantation

Transplantation for ALD remains controversial, principally due to concerns over the risk of post-transplant recidivism. In spite of initial concerns with regard to comorbidities, patients transplanted for cirrhosis have 5- and 10-year survival rates somewhere between those of patients transplanted for cholestatic and viral hepatitis-related liver disease. In general, transplantation should be restricted to patients with Child’s C cirrhosis or a MELD score of greater than 14. The potential effect of abstinence on the severity of disease has lead most units to adopt a policy of offering transplantation only to those patients who continue to have severe liver disease after a period of abstinence.

Around 10% of patients transplanted for ALD will return to problem drinking. This has no impact on early mortality, but mortality in recidivists is significantly higher than in abstainers 10 years post transplant. Efforts to minimize the risk of post-transplant recidivism are important, to avoid the likely adverse effect on the organ-donating public.

Many units have previously insisted on a 6-month period of abstinence. Although this can allow recovery, there is conflicting evidence about whether it can predict recidivism. The chance of recovery in patients with decompensated liver disease can be predicted as early as 3 months.

Transplantation for acute alcoholic hepatitis is very controversial. Although many of these patients undoubtedly have a poor prognosis, and there have been isolated reports of survival following transplantation, most clinical centres do not consider this therapeutic alternative.


Alcohol use and abuse is widespread and endstage liver disease is the result of prolonged heavy alcohol intake in only a small proportion of users. Nevertheless, these patients make up a significant proportion of the workload of most liver units in the Western world. The complex pathogenesis makes it an intriguing disease to study and the overlap of physical and psychological problems make this patient population a fascinating group. Unfortunately, improvements in management of alcoholic hepatitis are slow considering the high mortality of the condition. Liver transplantation remains the mainstay of treatment for advanced cirrhosis. Further research to understand the basics of hepatocyte injury is required to inform further clinical trials and improve mortality in this widespread disease.

Further reading  

Akriviadis E, et al. (2000). Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: a double-blind, placebo-controlled trial. Gastroenterology, 119, 1637–48. [The first trial of pentoxifylline in alcoholic hepatitis.][CrossRef] [Web of Science] [Medline] 
Becker U, et al. (2002). Lower risk for alcohol-induced cirrhosis in wine drinkers. Hepatology, 35, 868–75. [A very large epidemiological study into pattern of drinking and susceptibility.][CrossRef] [Web of Science] [Medline] 
Bellentani S, et al. (1997). Drinking habits as cofactors of risk for alcohol induced liver damage. The Dionysos Study Group. Gut, 41, 845–50. [A landmark study of liver disease in a given population.][Abstract/Full Text]
Cabre E, et al. (2000) Short- and long-term outcome of severe alcohol-induced hepatitis treated with steroids or enteral nutrition: a multicenter randomized trial. Hepatology, 32, 36–42. [RCT of enteral nutrition vs steroids.][CrossRef] [Web of Science] [Medline] 
Carithers RL, et al. (1989) Methylprednisolone therapy in patients with severe alcoholic hepatitis. A randomized multicenter trial. Ann Intern Med, 110, 685–90. [RCT of steroids in alcoholic hepatitis noted for the introduction of the discriminant function.]
Day CP (2006). Genes or environment to determine alcoholic liver disease and non-alcoholic fatty liver disease. Liver Int, 26, 1021–8. [Review of susceptibility to alcoholic liver disease.][CrossRef] [Web of Science] [Medline] 
Day CP, James OF (1998). Hepatic steatosis: innocent bystander or guilty party? Hepatology, 27, 1463–6. [Review of the role of fat in alcoholic liver disease.][CrossRef] [Web of Science] [Medline] 
Hrubec Z, Omenn GS (1981). Evidence of genetic predisposition to alcoholic cirrhosis and psychosis: twin concordances for alcoholism and its biological end points by zygosity among male veterans. Alcohol Clin Exp Res, 5, 207–15. [Only large twin study defining host factors that determine susceptibility.] [Web of Science] [Medline] 
Leon DA, McCambridge J (2006). Liver cirrhosis mortality rates in Britain from 1950 to 2002: an analysis of routine data. Lancet, 367, 52–6. [Study of cirrhosis mortality rates showing recent rises in UK.][CrossRef] [Web of Science] [Medline] 
Lelbach WK (1975). Cirrhosis in the alcoholic and its relation to the volume of alcohol abuse. Ann N Y Acad Sci, 252, 85–105. [Classic study of alcohol consumption and risk of cirrhosis.] [Web of Science] [Medline] 
Lieber CS (2004). The discovery of the microsomal ethanol oxidizing system and its physiologic and pathologic role. Drug Metab Rev, 36, 511–29. [Review of the characteristics of this important pathogenic pathway.][CrossRef] [Web of Science] [Medline] 
Mathurin P, et al. (2002) Corticosteroids improve short-term survival in patients with severe alcoholic hepatitis (AH): individual data analysis of the last three randomized placebo controlled double blind trials of corticosteroids in severe AH. J Hepatol, 36, 480–7. [Landmark ‘super’ meta-analysis showing benefit of corticosteroids in well selected patients.][CrossRef] [Web of Science] [Medline] 
Mathurin P, et al. (2003) Early change in bilirubin levels is an important prognostic factor in severe alcoholic hepatitis treated with prednisolone. Hepatology, 38, 1363–9. [Development of a criterion that allows early detection of patients not responding to steroids.] [Web of Science] [Medline] 
Mato JM, et al. (1999). S-Adenosylmethionine in alcoholic liver cirrhosis: a randomized, placebo-controlled, double-blind, multicenter clinical trial. J Hepatol, 30, 1081–9. [A rare positive RCT in alcoholic cirrhosis.][CrossRef] [Web of Science] [Medline] 
Neuberger J, et al. (2002). Transplantation for alcoholic liver disease. J Hepatol, 36, 130–7. [Thorough review of transplantation.] [Web of Science] [Medline] 
Sorensen TI, et al. (1984) Prospective evaluation of alcohol abuse and alcoholic liver injury in men as predictors of development of cirrhosis. Lancet, ii, 241–4. [Biopsy predictors of progression.]
Stewart SF, Day CP (2003). The management of alcoholic liver disease. J Hepatol, 38 Suppl 1, S2–13. [Review of current treatment for alcoholic liver disease.] [Web of Science] [Medline] 
Stewart SF, Day CP (2006). Alcoholic liver disease. In: Boyer T, Manns M, Wright T (eds) Zakim and Boyer’s hepatology, 5th edition, Chapter 30. Elsevier, Edinburgh.
Stewart SF, Day CP (2007). Ethanol metabolism and pathogenesis of alcoholic liver injury. In: Benhamou JP, et al. (eds) Textbook of hepatology; from basic science to clinical practice, 3rd edition, Chapter 12.2. Blackwell Publishing, Oxford.
Wilfred de Alwis NM, Day CP (2007). Genetics of alcoholic liver disease and nonalcoholic fatty liver disease. Semin Liver Dis, 27, 44–54. [Thorough review of the genetics of alcoholic liver disease.][CrossRef] [Web of Science] [Medline]