General measures for treatment of acute alcoholic hepatitis include abstinence from alcohol, nutritive support, relief of vitamin deficiencies and dietary adjustments if ascites or hepatic encephalopathy are present. Gastrointestinal bleeding and infections, particularly spontaneous bacterial peritonitis, are potential complications that can be specifically treated. Patients with mild alcoholic hepatitis may have marked improvement with abstinence and supportive care while those with deep jaundice, hepatic encephalopathy and marked abnormality in prothrombin have a 30-50% mortality and will almost certainly develop severe hepatic fibrosis or cirrhosis if they survive. A discriminant function developed by Maddrey and Colleagues (4.6 x prothrombin time-control prothrombin time + serum bilirubin in mg/dl) of >32 indicates a poor prognosis.

Since we have lacked knowledge of the pathogenesis of alcoholic liver injury, specific treatment of acute alcoholic hepatitis has been necessarily empirical. Potential objectives of treatment in severe cases have been suppression of hepatic inflammation, reduction of collagen formation, stimulation of hepatocyte regeneration and interruption of possible immunologically mediated or cytokine-induced hepatocyte damage.

A large number of randomized controlled trials (RCT) of a number of different treatments have been published. These include propylthiouracil, insulin and glucagon, anabolic-androgenic steroids, colchicine, penicillamine, and parenteral nutrition. None have shown unequivocal benefit. Much attention has been directed toward corticosteroid therapy, with 14 published RCTs. Most have shown no benefit, including three from our USC Liver Unit ( 2 published, 1 unpublished). However, in 3 of 14 trials there was a significantly better survival in a subgroup of patients with the most severe illness as indicated by the presence of spontaneous hepatic encephalopathy or a high Maddrey discriminant function. Three meta-analyses of these trials have been published. Two concluded that there was modest benefit from corticosteroid treatment in a subgroup of patients with spontaneous hepatic encephalopathy. The third and most recent meta-analysis concluded that there was no statistically significant benefit from corticosteroid, including the subgroup with encephalopathy.

In our Liver Unit we have recently completed a 4-year RCT of pentoxiphylline treatment designed to inhibit synthesis of tumor necrosis factor, a potentially harmful cytokine. This trial, not yet published, showed significant benefit from the treatment with short-term death rate of 23/52 in control patients vs. 12/49 in those treated with pentoxifylline. Currently we are considering a trial of monoclonal antibody to tumor necrosis factor.