Definition and causes

Liver disease related to alcohol consumption fits into one of three categories—fatty liver, alcoholic hepatitis, or cirrhosis ( Table 1 ). Fatty liver, which occurs after acute alcohol ingestion, is generally reversible with abstinence and is not believed to predispose to any chronic form of liver disease if abstinence or moderation is maintained. Alcoholic hepatitis is an acute form of alcohol-induced liver injury that occurs with the consumption of a large quantity of alcohol over a prolonged period of time; it encompasses a spectrum of severity ranging from asymptomatic derangement of biochemistries to fulminant liver failure and death. Cirrhosis involves replacement of the normal hepatic parenchyma with extensive thick bands of fibrous tissue and regenerative nodules, which results in the clinical manifestations of portal hypertension and liver failure.

Back to Top

Prevalence and risk factors

The prevalence of alcoholic liver disease is influenced by many factors, including genetic factors (e.g., predilection to alcohol abuse, gender) and environmental factors (e.g., availability of alcohol, social acceptability of alcohol use, concomitant hepatotoxic insults), and is therefore difficult to define. In general, however, the risk of liver disease increases with the quantity and duration of alcohol intake. Although necessary, excessive alcohol use is not sufficient to promote alcoholic liver disease. Only one in five heavy drinkers will develop alcoholic hepatitis, and one in four will develop cirrhosis.

Different alcoholic beverages contain varying quantities of alcohol ( Table 2 ). Although fatty liver is a universal finding among heavy drinkers, up to 40% of those with modest alcohol intake (up to 10 g per day) will also exhibit fatty changes. Based on an autopsy series of men, a threshold daily alcohol intake of 40 g was necessary to produce pathologic changes of alcoholic hepatitis. Consumption of more than 80 g per day was associated with an increase in the severity of alcoholic hepatitis, but not in the overall prevalence. There is a clear dose-dependent relation between alcohol intake and the incidence of alcoholic cirrhosis. A daily intake of more than 60 g of alcohol in men and 20 g in women significantly increases the risk of cirrhosis. In addition, steady daily drinking, as compared with binge drinking, appears to be more harmful.

Back to Top

Pathophysiology and natural history

The liver and, to a lesser extent, the gastrointestinal tract, are the main sites of alcohol metabolism. Within the liver, there are two main pathways of alcohol metabolism, alcohol dehydrogenase and cytochrome P-450 2E1. Alcohol dehydrogenase is a hepatocyte cytosolic enzyme that converts alcohol to acetaldehyde. Acetaldehyde subsequently is metabolized to acetate via the mitochondrial enzyme, acetaldehyde dehydrogenase. Cytochrome P-450 2E1 also converts alcohol to acetaldehyde.

Liver damage occurs through several interrelated pathways. Alcohol dehydrogenase and acetaldehyde dehydrogenase cause the reduction of nicotinamide adenine dinucleotide (NAD) to NADH (reduced form of NAD). The altered ratio of NAD/NADH promotes fatty liver through the inhibition of gluconeogenesis and fatty acid oxidation. Cytochrome P-450 2E1, which is upregulated in chronic alcohol use, generates free radicals through the oxidation of nicotinamide adenine dinucleotide phosphate (reduced; NADPH) to NADP. Chronic alcohol exposure also activates hepatic macrophages, which then produce tumor necrosis factor a (TNF-α). TNF-α induces mitochondria to increase the production of reactive oxygen species. This oxidative stress promotes hepatocyte necrosis and apoptosis, which is exaggerated in the alcoholic individual who is deficient in antioxidants such as glutathione and vitamin E. Free radicals initiate lipid peroxidation, which causes inflammation and fibrosis. Inflammation is also incited by acetaldehyde that, when bound covalently to cellular proteins, forms adducts that are antigenic. Histologically, the earliest changes in alcoholic hepatitis are located predominantly around the central vein. Alcohol is known to cause an exaggerated gradient of hypoxia from the portal vein to the central vein, suggesting that the hypoxia induced by chronic alcohol use may contribute to hepatic damage.

With abstinence, morphologic changes of the fatty liver usually revert to normal. Although the short-term prognosis in patients with alcoholic steatosis is excellent, with longer follow-up it has been found that cirrhosis develops more commonly in alcohol abusers with fatty liver changes than in those with normal liver histology. Morphologic features predictive of progression to fibrosis, cirrhosis, or both include severe steatosis, giant mitochondria, and the presence of mixed macrovesicular-microvesicular steatosis.

Historically, the 30-day mortality rate in patients with alcoholic hepatitis ranges from 0% to 50%. Clinical and laboratory features are powerful prognostic indicators for short-term mortality. Hepatic encephalopathy, derangement in renal function, hyperbilirubinemia, and prolonged prothrombin time are seen more frequently in patients who succumb to their illness than in those who survive. Both the discriminant function and the model for end-stage liver disease (MELD) score can be used to predict short-term mortality in patients with alcoholic hepatitis. The MELD score is calculated based on a patient's prothrombin time, serum creatinine, and bilirubin; a calculator can be found on the United Network for Organ Sharing website. Long-term survival in patients with alcoholic hepatitis who discontinue alcohol is significantly better than in those who continue to drink, although it remains considerably below that of an age-matched population. Three-year survival approaches 90% in abstainers, whereas it is less than 70% in active drinkers.

Cirrhosis has historically been considered to be an irreversible outcome of severe and prolonged liver damage. However, studies involving patients with liver disease from many distinct causes have shown convincingly that fibrosis and cirrhosis may actually have a component of reversibility. For those patients with decompensated alcoholic cirrhosis who undergo transplantation, survival is comparable with that of patients with other causes of liver disease (approximately a 70% 5-year patient survival rate).

Back to Top

Signs and symptoms

Patients with fatty liver are typically either asymptomatic or present with nonspecific symptoms that do not suggest acute liver disease. Supporting features on physical examination include an enlarged and smooth, but rarely tender, liver. In the absence of a super-imposed hepatic process, stigmata of chronic liver disease such as spider angiomas, ascites, or asterixis should be absent.

Alcoholic hepatitis is a syndrome with a spectrum of severity, and therefore manifesting symptoms vary. Symptoms may be nonspecific and mild and include anorexia and weight loss, abdominal pain and distention, or nausea and vomiting. Alternatively, more severe and specific symptoms may include encephalopathy and hepatic failure. Physical findings include hepatomegaly, jaundice, ascites, spider angiomas, fever, and encephalopathy. When alcoholic hepatitis is more severe, the differential diagnosis includes autoimmune hepatitis, acute viral hepatitis, acute Budd-Chiari syndrome, severe drug-induced liver injury, and an acute manifestation of Wilson's disease.

Established alcoholic cirrhosis may manifest with decompensation without a preceding history of fatty liver, alcoholic hepatitis, or any other alcohol-related diagnosis. Alternatively, alcoholic cirrhosis may be diagnosed concurrently with acute alcoholic hepatitis. The symptoms and signs of alcoholic cirrhosis do not help differentiate it from other causes of cirrhosis. Patients may present with jaundice, pruritus, abnormal laboratory findings (e.g., thrombocytopenia, hypoalbuminemia, coagulopathy), or complications of portal hypertension, such as variceal bleeding, ascites, or hepatic encephalopathy.

Back to Top

Diagnosis

Fatty liver is usually diagnosed in the asymptomatic patient who is undergoing evaluation for abnormal liver function; results include usually mildly elevated aminotransferase levels lower than twice the upper limit of normal. Laboratory tests are not diagnostic of fatty liver. Characteristic ultrasonographic findings include a hyperechoic liver with or without hepatomegaly. Liver biopsy is rarely needed to diagnose fatty liver in the appropriate clinical setting, but may be useful in excluding steatohepatitis or fibrosis. Typical histologic findings of fatty liver include fat accumulation in hepatocytes that is often macrovesicular, but may occasionally be microvesicular (Fig. 1). The centrilobular region of the hepatic acinus is most commonly affected. In severe fatty liver, however, fat is distributed throughout the acinus. Fatty liver is not specific to alcohol ingestion; it is associated with obesity, insulin resistance, hyperlipidemia, malnutrition, and various medications. Attribution of fatty liver to alcohol use therefore requires a detailed and accurate patient history.

The diagnosis of alcoholic hepatitis is based on a thorough history, physical examination, and review of laboratory tests. Characteristically, the aspartate aminotransferase–to–alanine aminotransferase ratio (AST/ALT) is approximately 2:1, and the absolute value of the aminotransferase levels does not exceed 300 U/L unless a superimposed hepatic insult exists, such as acetaminophen toxicity. Other common and nonspecific laboratory abnormalities include anemia and leukocytosis. Liver biopsy is occasionally necessary to secure the diagnosis. The classic histologic features of alcoholic hepatitis include inflammation and necrosis, which are most prominent in the centrilobular region of the hepatic acinus (Fig. 2). Hepatocytes are classically ballooned, which causes compression of the sinusoid and reversible portal hypertension. The inflammatory cell infiltrate, located primarily in the sinusoids and close to necrotic hepatocytes, consists of polymorphonuclear cells and mononuclear cells. In addition to inflammation and necrosis, many patients with alcoholic hepatitis will have fatty infiltration and Mallory bodies, which are intracellular perinuclear aggregations of intermediate filaments that are eosinophilic on hematoxylin-eosin staining. Neither fatty infiltration nor Mallory bodies are specific for alcoholic hepatitis or necessary for the diagnosis.

The diagnosis of alcoholic cirrhosis rests on finding the classic signs and symptoms of end-stage liver disease in a patient with a history of significant alcohol intake. Patients tend to underreport their alcohol consumption, and discussions with family members and close friends may provide a more accurate estimation of alcohol intake. Patients may present with any or all complications of portal hypertension, including ascites, variceal bleeding, and hepatic encephalopathy. The histology of end-stage alcoholic cirrhosis, in the absence of acute alcoholic hepatitis, resembles that of advanced liver disease from many other causes, without any distinct pathologic findings (Fig. 3).

The overall clinical diagnosis of alcoholic liver disease, using a combination of physical findings, laboratory values, and clinical acumen, is very accurate (Box 1). However, liver biopsy can be justified in select cases, especially when the diagnosis is in question. In addition to confirming the diagnosis, liver biopsy is also useful for ruling out other unsuspected causes of liver disease, better characterizing the extent of the damage, providing prognosis, and guiding therapeutic decision making. As emphasized in the recent practice guidelines for alcoholic liver disease from the American College of Gastroenterology), liver biopsy is a relatively safe procedure, with morbidity lower than 0.6% and mortality lower than 0.03%. With a typical presentation, liver biopsy is not mandatory to diagnose alcoholic liver disease.

Summary

• All patients should be screened for alcoholic liver disease.
• Alcoholic liver disease is a heterogeneous disease.
• No physical examination finding or laboratory abnormality is specific for alcoholic liver disease.
• The diagnosis of alcoholic liver disease requires a detailed patient history, with supportive laboratory and imaging studies.
• Liver biopsy may be useful to confirm the diagnosis, rule out other diseases, and prognosticate.

Back to Top

Treatment

The foundation of therapy for alcoholic liver disease is abstinence. Patients are often unable to achieve complete and durable alcohol abstinence without assistance, and referral to a chemical dependency team is appropriate. Hospitalization is indicated to expedite a diagnostic evaluation of patients with jaundice, encephalopathy, or ascites of unknown cause. In addition, patients with known alcoholic liver disease who present with renal failure, fever, inadequate oral intake to maintain hydration, or rapidly deteriorating liver function, as demonstrated by progressive encephalopathy or coagulopathy, should be hospitalized.

Supportive care for all patients includes adequate nutrition. Almost all patients with alcoholic hepatitis have some degree of malnutrition, but estimating the severity of malnutrition remains a challenge because sensitive and specific clinical or laboratory parameters are lacking. The nutritionist plays a valuable role in assessing the degree of malnutrition and guiding nutritional supplementation in malnourished alcoholic patients. The degree of malnutrition correlates directly with short-term (1-month) and long-term (1-year) mortality. At 1 year from the time of diagnosis of alcoholic hepatitis, patients with mild malnutrition have a 14% mortality rate, compared with a 76% mortality rate in those with severe malnutrition. Attempts to correct protein-calorie malnutrition with supplemental oral or parenteral nutrition, or both, have met with mixed results. In general, enteral nutrition is preferable over parenteral supplementation, and protein should be supplied to provide positive nitrogen balance. Branched-chain amino acids are useful as a supplement to maintain positive nitrogen balance in patients who do not tolerate liberal protein intake because of the development of encephalopathy; however, their expense limits routine use in all alcoholic malnourished patients. Nutritional supplementation is generally associated with an improvement in liver test results, but only rarely with a mortality benefit. Refer to recent practice guidelines for a summary of recommendations for daily feeding in patients with alcoholic liver disease.

The use of corticosteroids as specific therapy for alcoholic hepatitis has received a great deal of interest. The rationale behind their use is the possible role of the immune system in the initiation and perpetuation of hepatic damage. Three randomized controlled trials that investigated the use of corticosteroids (40 mg prednisolone/day, or the equivalent methylprednisolone, 28 mg/day, for 28 days) for patients with severe acute alcoholic hepatitis have suggested a significant decrease in short-term (30-day) mortality in patients randomized to prednisolone, but only those with more severe liver dysfunction, as manifested by hepatic encephalopathy or a markedly abnormal discriminant function. The discriminant function predicts the risk of early mortality in acute alcoholic hepatitis, and is calculated using the following formula:

with the prothrombin and control times in seconds.

Results from other randomized controlled trials have been contradictory. Several meta-analyses have been conducted in an effort to overcome low statistical power in individual trials. Analyses by Mathurin and colleagues and Imperiale and McCullough have supported the use of corticosteroids in a select group of patients with severe alcoholic hepatitis, manifest by a discriminant function of more than 32, hepatic encephalopathy, or both, whereas Christensen and Gluud have found no effect of corticosteroids on mortality. Most recent trials of corticosteroids for alcoholic hepatitis have excluded patients with certain coexisting conditions, such as gastrointestinal bleeding, active infection, diabetes, viral hepatitis, or acute pancreatitis, and therefore the applicability of these study findings is limited. Practice guidelines have supported the use of corticosteroids in patients with severe alcoholic hepatitis in whom the diagnosis is certain.

The most recent practice guidelines have not addressed the use of pentoxifylline in the treatment of alcoholic hepatitis, because studies with pentoxifylline were published after the guidelines. Pentoxifylline is an inhibitor of TNF synthesis. Elevated TNF levels have been associated with higher mortality from alcoholic hepatitis. TNF has been found in increased levels in both sera and liver tissue of patients with alcoholic hepatitis. A randomized, double-blind, controlled trial has investigated the effects of treatment with pentoxifylline on short-term survival and progression to hepatorenal syndrome in patients with severe alcoholic hepatitis. Pentoxifylline-treated patients had a significant decrease in mortality (24% vs. 46%, P = .037). The survival advantage was primarily due to a decrease in the development of hepatorenal syndrome in pentoxifylline-treated patients (50% vs. 91.7%, P = .009). Although promising, these findings have yet to be validated by independent investigation.

The histologic changes of alcoholic hepatitis are most pronounced in a perivenular distribution, resembling that of ischemic injury. It is well established that alcohol increases hepatic oxygen consumption. Propylthiouracil, which decreases metabolism and oxygen consumption, has been investigated as a means to reverse the hypermetabolic state of alcoholic hepatitis. An early study has shown improvement in laboratory parameters in patients with mild to moderate alcoholic hepatitis who were treated with propylthiouracil, but a survival advantage was not noted. A subsequent study involving patients with more severe alcoholic hepatitis found no difference in survival over initial 8 weeks and no difference in the rate of recovery of liver function abnormalities. A systematic review of more than 700 patients from six published randomized, controlled trials has failed to show any effect of propylthiouracil on mortality, individual laboratory parameters, liver histology, or liver-related complications. At present, propylthiouracil cannot be recommended for routine clinical use in the treatment of alcoholic hepatitis.

TNF-α is a proinflammatory cytokine produced by inflammatory cells in response to acute alcohol ingestion, and is believed to contribute to liver inflammation and hepatocyte damage. Infliximab is a chimeric human-mouse monoclonal antibody that binds to TNF-α and neutralizes its effects. Four small studies have investigated the use of infliximab in patients with severe alcoholic hepatitis, either alone or in combination with corticosteroids. Although early pilot studies have suggested that infliximab could be administered safely to patients with severe alcoholic hepatitis, the only randomized controlled study to date has demonstrated a higher probability of death at 2 months in those patients randomized to steroids and infliximab. In addition, there were significantly more severe infections in the infliximab-steroid group without any benefit in liver function or Maddrey scores. Anti–TNF-α therapies were investigated as treatment for alcoholic hepatitis after publication of the practice guidelines; however, routine use of these therapies cannot be recommended at present, and their use should be restricted to clinical trials.

Other therapies that have been investigated in the treatment of alcoholic hepatitis but not found to be beneficial include insulin and glucagon, calcium channel blockers, and antioxidants, such as vitamin E.

Treatment of the patient with alcoholic cirrhosis mirrors the care of patients with any other type of cirrhosis; it includes prevention and management of ascites, spontaneous bacterial peritonitis, variceal bleeding, encephalopathy, malnutrition, and hepatocellular carcinoma. Once advanced cirrhosis has occurred with evidence of decompensation (i.e., ascites, spontaneous bacterial peritonitis, hepatic encephalopathy, variceal bleeding), patients should be referred to a transplantation center. Acute alcoholic hepatitis, no matter how severe, is an absolute contraindication for liver transplantation. For more than a decade, alcoholic cirrhosis has been the second leading indication for liver transplantation in the United States. Most transplantation centers currently require patients with a history of alcohol abuse to have documented abstinence of at least 6 months before undergoing transplantation. This requirement theoretically has a dual advantage of predicting long-term sobriety and allowing recovery of liver function from acute alcoholic hepatitis. This “6-month abstinence rule” may not have much prognostic significance in predicting recidivism, however. Alcohol use of any quantity after transplantation for alcohol-related liver disease approaches 50% during the first 5 years, and abuse occurs in up to 15% of patients. Table 3 summarizes investigated treatments for alcoholic liver disease.

Summary

• Abstinence is the cornerstone of treatment of alcoholic liver disease.
• Nutritional deficiencies should be sought and treated aggressively.
• Corticosteroids should be used in patients with a definite diagnosis of severe alcoholic hepatitis, as defined by a discriminant function of more than 32, hepatic encephalopathy, or both. Corticosteroids have not been evaluated in patients with renal failure, active infection, pancreatitis, or gastrointestinal bleeding.

Back to Top

Prevention and screening

As emphasized in the most recent national practice guidelines, health care providers must be attentive for signs of covert alcohol abuse. Many patients will not openly disclose an accurate history of alcohol use. In addition, no physical examination finding or laboratory abnormality is specific for alcoholic liver disease. All patients should therefore be screened for alcohol abuse or dependency. The CAGE questionnaire (cutting down on drinking, annoyance at others' concerns about drinking, feeling guilty about drinking, using alcohol as an eye opener in the morning) is the preferred screening tool, with two or more positive answers providing a sensitivity of 70% to 96% and a specificity of 91% to 99% for alcohol dependency.

Back to Top

Suggested Readings

• Pentoxifylline improves short-term survival in severe acute alcoholic hepatitis: A double-blind, placebo-controlled trial. Gastroenterology. 119: 2000; 1637-1648.
• Glucocorticoids are ineffective in alcoholic hepatitis: a meta-analysis adjusting for confounding variables. Gut. 37: 1995; 113-118.
• Epidemiology of alcoholic liver disease. Semin Liver Dis. 8: 1988; 12-25.
• Do corticosteroids reduce mortality from alcoholic hepatitis?. Ann Intern Med. 113: 1990; 299-307.
• Corticosteroids improve short-term survival in patients with severe alcoholic hepatitis (AH): Individual data anlysis of the last three randomized placebo controlled double blind trials of corticosteroids in severe AH. J Hepatology. 36: 2002; 480-487.
• Alcoholic liver disease: Proposed recommendations for the American College of Gastroenterology. Am J Gastroenterol. 93: 1998; 2022-2036.
• Alcohol consumption and alcoholic liver disease: Evidence of a threshold level of effects of ethanol. Alcohol Clin Exp Res. 17: 1993; 1112-1117.
• Nutritional therapy in alcoholic liver disease. Semin Liver Dis. 13: 1993; 196-209.
• Alcoholic liver disease: Clinical patterns and diagnosis. Acta Med Scand (Suppl). 703: 1985; 103-110.
• Alcoholic liver disease: New insights into mechanisms and preventative strategies. Trends Mol Med. 7: 2001; 408-413.
• Timing of liver transplantation in alcoholic cirrhosis. J Hepatol. 39: 2003; 302-307.
• Model for End-Stage Liver Disease (MELD) Calculator, 2007. Available at http://www.unos.org/resources/MeldPeldCalculator.asp?index=98