The prevalence of alcoholic liver disease in any country depends largely on religious and other customs and on the relation between the cost of alcohol and earnings--the cheaper the alcohol the more lower socioeconomic groups are affected. World wide, alcohol consumption is increasing. However, in France there has been a decrease over the past 20 years, perhaps as a result of government propaganda, and likewise in the USA, where consumption, especially of spirits, has fallen, possibly as a result of changing lifestyles, with the need to conform in the face of unemployment and economic instability.

Alcohol excess can have various effects on the liver. 33% of heavy drinkers have no hepatic consequences; the reasons for this are not known. Others develop a fatty liver, which is a benign condition, and extensive zone 3 fibrosis, which is serious, preceding cirrhosis and, as a long-term sequel, hepatocellular carcinoma.
Risk factors for alcoholic liver damage Dose and duration

1 unit (10 g) of alcohol is contained in 28 mL (l fluid ounce) of whisky or similar spirits, 85 mL of wine, or 230 mL of beer. The nature of the beverage is not important; the key factor is the alcohol content.

Safe weekly limits are believed to be 21 units in men and 14 units in women. These figures may be too low; certainly little damage arises from 3 drinks a day, and there is some evidence that this level of consumption protects against coronary heart disease. 16 units (160 g) of alcohol daily for 5 years is probably the minimum associated with significant liver damage.
Sex

Women are more susceptible than men to alcoholic hepatic damage and are more likely to relapse after treatment. One underlying factor in women is the higher blood concentrations achieved for the same amount of alcohol; another is reduced alcohol metabolism by the stomach.
Genetics

Patterns of alcoholic drinking behaviour are inherited but no single genetic marker is clearly associated with susceptibility to alcoholic liver damage.[ 1] Nevertheless, twin studies show that there is a higher concordance rate for monozygotic than for dizygotic twins in the prevalence of alcoholism and cirrhosis. Genetic polymorphism in alcohol metabolising enzymes accounts for different rates of alcohol oxidation and generation of the toxic substance acetaldehyde,[ 2, 3] and there are probably differences in the genes controlling the transcription of collagen. The likelihood is that susceptibility is not caused by a single gene defect but by cumulative interaction of several genes and environmental factors. Alcoholism and alcoholic liver damage are polygenic disorders.
Nutrition

In patients with chronic stable alcoholic cirrhosis, body composition analysis shows protein depletion which is related to the severity of the liver disease.[ 4] The importance of malnutrition depends on the type of alcoholic; in patients of low socioeconomic status, protein-calorie malnutrition often precedes liver injury, whereas in the socially adequate with a good diet, liver damage seems little related to nutrition.

Animals show species variation.[5] Rats given alcohol develop liver damage only if the diet is deficient, whereas baboons develop cirrhosis while consuming a good diet. In rhesus monkeys alcoholic liver damage is prevented by increasing the dietary protein and choline. Patients with decompensated liver disease given a third of their calories as alcohol together with a nutritious diet improve steadily whereas liver function does not improve with alcohol abstinence if dietary protein remains low. Nutrition and hepatotoxicity may ac.t synergistically. Both alcohol and nutrition probably play a part in alcohol hepatotoxicity, alcohol being the more important. There may be a range of alcohol intake that is tolerated without liver damage under optimum dietary conditions. However, it is also likely that there is a threshold of alcohol toxicity beyond which no protection is afforded by dietary manipulation.
Hepatitis B and C infections

Coexistence of viral infections is believed to increase the severity of alcoholic liver disease. Nevertheless, features of viral disease may be dificult to distinguish from those of alcoholic damage.[6] The effect of abstinence on serum aminotransferases and serum gamma glutamyl transpeptidase (GGT), the identification of risk factors, and liver biopsy appearance can be helpful. Serological hepatitis B surface antigen may be absent and serum HBV DNA testing may be necessary to diagnose HBV infection.[7] Positive serum second-generation ELISA tests for hepatitis C antibodies usually correlate with a positive serum HCV RNA and allow diagnosis of complicating hepatitis C virus disease.[8]
Clinical recognition

About a third of physicians neglect to take a history of alcohol consumption from their patients. Such inquiries are essential; adequate time should be taken to determine the relevance of alcohol to the patient's lifestyle. Any suspicion that alcohol consumption is excessive should be followed by the CAGE questionnaire.

  C  Have you ever felt the need to cut down o your drinking?
  A  Have you ever felt annoyed by criticism of your drinking?
  G  Have you ever felt guilty about your dringking?
  E  Have you ever take a drink (eye opener) first thing in the
     morning?

Score 1 for each positive response. CAGE scores of 2 or more suggest the presence of alcohol-related problems.

Patients may present with non-specific digestive symptoms such as anorexia, morning nausea with dry retching, diarrhoea, vague right upper quadrant abdominal pain, and tenderness or pyrexia. They may seek medical advice because of the effects of alcoholism such as social disruption, poor work performance, accidents, violent behaviour, fits, tremulousness, or depression. Physical signs include tender hepatomegaly, prominent vascular spider naevi, and associated features of alcoholism.

The combination of a raised mean corpuscular volume and serum GGT will identify 90% of the alcohol-dependent population. Serum aspartate aminotransferase (APT) may be modestly raised and is usually less than alanine aminotransferase (ALT). High ALT values suggest concomitant ingestion of paracetamol. Serum carbohydrate-deficient transferrin is a specific and sensitive test of alcoholism, irrespective of complicating liver disease, but is not readily available.[9] Blood alcohol concentrations may be useful diagnostically and to confirm abstention. The PGA index and serum procollagen III peptide may also be useful in prediction of non-benign disease.

Liver biopsy is necessary to confirm the extent of liver disease and for prognosis.[5] Perivenular (zone 3) sclerosis, which is much more serious than simple fatty change and a precursor of cirrhosis, can only be diagnosed histologically.
Management

Patients must abstain completely from alcohol. Less severely affected individuals can have "brief interventional counselling" from doctor, nurse, or similar person. The main ingredients are termed FRAMES.

  F  Feedback to patient about risks
  R  Personal responsibility for change
  A  Advice
  M  Menu of options for change
  E  Empathic interviewing
  S  Self-efficacy

Such counselling results in a 38% treatment benefit, albeit often temporary.[10] More severely affected patients should be referred to a psychiatrist. Unfortunately at least 25% of patients with alcoholic liver disease will continue to imbibe irrespective of therapy.

The development of a withdrawal syndrome (delirium tremens) should be anticipated by administration of a benzodiazepine. Stable middle-class alcoholics should take a mixed diet containing 2000 calories with 1 g of protein per kg body weight and a small vitamin supplement. Severe alcoholic hepatitis with jaundice is rare, and such patients should be admitted to hospital. Precipitating factors such as infection and bleeding must be sought and treated. Parenteral vitamins are given but the use of corticosteroids is controversial. Steroids are probably useful for encephalopathy without gastrointestinal bleeding, systemic infection, and renal failure. Corticosteroids given for 28 days reduce early mortality in patients with alcoholic hepatitis who are severely ill.[11] Long-term oxandrolone (an anabolic steroid) is useful in patients with moderately severe disease but ineffective in those with severe malnutrition and inadequate calorie intake.[12] The need to provide protein and improve nutrition overrides the fear of precipitating hepatic encephalopathy. Oral or intravenous aminoacid supplementation should be reserved for the very few jaundiced and obviously malnourished individuals.[13] Most patients can take adequate natural protein by mouth. Accelerated improvement follows the use of casein-based nasoduodenal tube feeding supplements (1.5 g protein per kg body weight).[14] However, there is only a trend towards increased survival. Colchicine has not improved short-term survival in patients with alcoholic hepatitis.

Alcohol induces a hypermetabolic state that potentiates centrizonal anoxic liver injury. Propylthiouracil reduces hypoxic liver injury in hypermetabolic animals. A controlled trial confirmed the benefit of propylthiouracil, particularly long-term, in those who continued to drink modestly.[15] Nevertheless, propylthiouracil has never gained general acceptance for the treatment of alcoholic liver disease.
Transplantation

Survival after liver transplantation for alcoholic cirrhosis is similar to or even better than for other end-stage liver diseases. Selection of alcoholics for operation is difficuk. The cirrhosis is self-inflicted; at least 20% of patients will return to alcohol; and compliance with immunosuppressive regimens may be poor. Should these patients compete with others when donor livers are in short supply? I believe those selected should have a stable psychiatric and socioeconomic background, a job to return to after the transplant, and no evidence of extrahepatic (eg, cerebral) alcoholic damage. They should have been abstinent for at least 6 months; such abstinence proves to be the most important predictor of post-transplant recidivism.[16]

Selection is crucial.[17] Patients deemed unsuitable because they are too well need continued surveillance because their survival can decline. Those excluded because they are too sick or psychiatrically unsuitable have a significantly lower survival than those transplanted. More worrying is the observation from Pittsburgh that 23 of 330 persons transplanted for alcoholic liver disease had definitely resumed alcohol abuse.[18] Within 6 months to 2 years, 22 of 23 patients studied had liver biopsy evidence of alcoholic hepatitis and 4 of these were cirrhotic. Hepatic transplantation for acute alcoholic hepatitis is even more difficult to justify than for end-stage alcoholic cirrhosis since a period of pre-transplant sobriety is unlikely. Transplants should not be done for acute alcoholic hepatitis until we have reliable techniques to predict recidivism and especially the potential to return to problem drinking.[19] Well-designed controlled trials must be carried out.