Treatment of alcoholic hepatitis


  • Potential conflict of interest: Nothing to report.


Watch a video presentation of this article

Watch the interview with the author

Answer questions and earn CME

Three steps must be completed before starting treatment for alcoholic hepatitis (AH). First, the diagnosis of alcoholic hepatitis must be made based on history and clinical and/or histological data. Second, the severity of AH must be determined using Maddrey's discriminant function, the model for end-stage liver disease,1 the Glasgow alcoholic hepatitis score,2 or the ABIC (albumin-bilirubin-INR-creatinine) score.3 These scores determine whether a patient should be treated (except ABIC), and the short-term prognosis. Finally, patients need to be assessed for infection with chest X-ray, blood cultures, urinalysis (preferably culture), and paracentesis (cell count and preferably Gram stain and culture). Approximately 25% of patients are infected at the time of hospital admission, and an additional 25% become infected while receiving treatment.4


AASLD, American Association for the Study of Liver Diseases; ABIC, albumin-bilirubin-INR-creatinine; AH, alcoholic hepatitis; NAC, N-acetylcysteine; PTX, pentoxifylline.


Abstinence is critical to long-term survival. We recommend discussing abstinence with the patient and referral to an addiction specialist. We leave the decision about the use of drugs to reduce craving (e.g., baclofen, acamprosate, or naloxone) to the addiction specialist. However, use of these agents in patients with severe AH is not recommended due to risk of hepatotoxicity. Patients who return to drinking risk recurrence.


All patients with severe AH are malnourished.5 Although nutritional treatment has not been proven to improve survival,6 failure to consume sufficient calories is a marker of poor survival among patients with AH and severe malnutrition.7 We use the guidelines shown in Table 1 for nutritional support.8 Enteral feeding can be used in anorexic patients. Fifty milligrams of elemental zinc with meals may improve gut barrier function.9

Table 1. Recommended Nutritional Treatment for AH
  1. Calories and protein are given according to ideal body weight.

Calories35-40 cal/kg/day
Protein1.2-1.5 g/kg/day
Sodium2 g/day
Meals4 or 5 meals per day, including an evening snack
Multivitamins, mineralsThiamine, B2, B12, folate, magnesium, zinc, selenium

Drug Treatment


Prednisolone at 40 mg/day for 28 days with or without a 2-week taper is the recommended treatment by the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver10, 11 but was judged to be ineffective in improving overall or liver-related survival by the Cochran Reviews12 (Fig. 1). Patients suitable for treatment are those with a discriminant function >32 or with hepatic encephalopathy (Table 2). Exclusions include recent (i.e., past 2-4 days) upper gastrointestinal bleeding, uncontrolled diabetes, or significant renal insufficiency. Patients with infection can be treated after the infection has been controlled with antibiotics for several days. Patients with hepatitis C or hepatitis B (controlled with drugs) can receive prednisolone.

Figure 1.

Meta-analysis of survival among subjects receiving prednisolone or placebo grouped according to the inclusion criteria used in the clinical trial. Prednisolone treatment was associated with improved survival in trials that enrolled subjects with Maddrey's discriminant function >31 or hepatic encephalopathy, but not in clinical trials that used other inclusion criteria nor when all trials were combined. Reprinted with permission from Alimentary Pharmacology & Therapeutics.12 Wiley-Blackwell, 2008.

Table 2. Inclusion and Exclusion Criteria for Treatment of AH
Severity of liver diseaseInfection uncontrolled by antibiotics
Hepatic encephalopathyRecent upper gastrointestinal bleeding
 Hepatorenal syndrome
 Uncontrolled diabetes
 Uncontrolled hepatitis B
 Human immunodeficiency virus

We agree with the AASLD's recommendation of prednisolone treatment for 28 days11 among patients with severe AH. Many hepatologists assess response to prednisolone treatment using the Lille score (, which measures baseline prognostic markers as well as biochemical response to prednisolone after 7 days of therapy.13 Patients with a Lille score <0.45 should continue prednisolone treatment for a total of 28 days, because prednisolone treatment improves survival in this subset of patients. There is debate regarding the management of patients with a Lille score ≥0.45 (poor response to prednisolone). In their initial evaluation, Louvet and colleagues13 suggested that prednisolone be discontinued among patients with a Lille score ≥0.45, whereas their recent report, based on a larger sample and alternative cut points, suggests that patients with a Lille score ≥0.56 are unlikely to benefit from prednisolone.14 Our suggestion is to consider discontinuing prednisolone in patients with Lille score ≥0.56.


Orally administered pentoxifylline (PTX) at 400 mg thrice daily for 28 days is an alternative treatment for severe AH.11 There are no absolute contraindications to PTX treatment. In the initial study of 101 patients, PTX improved survival by decreasing the incidence of hepatorenal syndrome.15 Subsequent trials with PTX have been smaller but have confirmed improved outcomes with PTX treatment. The Cochran Review concluded that PTX reduced mortality by approximately 35%, with seven patients needed to be treated to save the life of one patient16 (Fig. 2).

Figure 2.

Meta-analysis of survival among subjects receiving PTX or placebo. Reprinted with permission from: The Cochrane Database of Systematic Reviews.16 Wiley, 2009.

Treatment of Nonresponders

There is uncertainty regarding appropriate treatment of patients who do not respond to prednisolone (Lille score ≥0.45 or ≥0.56) or PTX (vaguely defined). Analysis of retrospective databases suggests that switching poorly responding patients from prednisolone to PTX does not improve survival.17 A randomized controlled trial that was presented at the AASLD annual meeting in 2011 but is not yet published showed that combination treatment with prednisolone and PTX was not better than treatment with prednisolone alone: the 6-month survival (∼68%) and the incidence of hepatorenal syndrome (∼10%) was similar in both groups.18 Not only does this trial suggest that combination treatment is not better than prednisolone alone, it also suggests that PTX alone is not better than prednisolone alone. Thus, at the current time, there is no evidence that adding a second drug, or switching to the alternative treatment, is beneficial for patients who respond poorly to their initial treatment.

Are antioxidants the Answer?

There is widespread acceptance that AH is a condition of excessive oxidative stress. Three clinical trials published prior to 2011 using a variety of antioxidant cocktails, including N-acetylcysteine (NAC), reported no difference in survival when compared with enteral nutrition,19 worse survival when compared with prednisolone,20 and no improvement in survival when added to prednisolone.21 However, the results of a recent study by Nguyen-Khac et al.,22 in which NAC was given intravenously for 5 days in combination with prednisolone, suggest that NAC may be beneficial.22 In this study, 174 patients were randomized to prednisolone + placebo (intravenous saline) or prednisolone + NAC for 5 days, followed by prednisolone for the next 23 days. Mortality at 1 month was better in the NAC + prednisolone arm (8%) compared with the prednisolone alone arm (24%) (P < 0.006). The primary outcome of the trial, however, was 6-month mortality that was similar in the NAC + prednisolone arm (27%) and the prednisolone alone arm (38%) (P = 0.07) (Fig. 3). Although this trial suggests that NAC may improve survival, given the lack of efficacy with antioxidant therapy in prior clinical trials, our opinion is that additional clinical trials are needed before NAC can be recommended for treatment of AH.

Figure 3.

Six-month survival among patients with AH randomized to receive 28 days of prednisolone or 28 days of prednisolone and 5 days of intravenous NAC. Reprinted with permission from The New England Journal of Medicine.22 Boston, Massachusetts Medical Society, 2011.

Liver Transplantation, the Next Frontier

Patients with a Lille score >0.45 have a 70% mortality by 6 months,13 the usual amount of time a patient needs to be abstinent before considering liver transplantation. Consequently, Mathurin et al.23 conducted a clinical trial of liver transplantation in highly selected patients who failed prednisolone treatment. The enrolled patients had no prior history of alcoholic liver disease (first episode), had a supportive family, and were deemed acceptable candidates by their primary medical team (residents and attending physician), an addiction specialist, their hepatologists, and their surgical team (anesthesiologist and surgeon). A total of 26 patients were placed on the liver transplantation list an average of 13 days after hospital admission. Survival among patients receiving liver transplantation (77% at 6 months) was significantly higher than predicted survival (30%) (P < 0.001) and only slightly less than estimated survival among patients with a Lille score < 0.45 (85%) (P = 0.33) (Fig. 4). Approximately 2% of patients with AH at the participating hospitals qualified for liver transplantation, and 3% of liver transplantations performed at the hospitals were for patients with AH. Three patients returned to drinking, but only one to harmful levels. No patient needed retransplantation. These investigators continue to perform liver transplantation for highly selected patients as part of a clinical trial. Early liver transplantation for AH is not generally accepted among liver transplantation centers in the United States.

Figure 4.

Kaplan-Meier estimates of survival among 26 subjects with AH undergoing liver transplantation and randomly selected matched control subjects failing to respond to prednisolone (Lille score ≥0.45) and responding to prednisolone (Lille score <0.45). Reprinted with permission from The New England Journal of Medicine.23 Boston, Massachusetts Medical Society, 2011.

The Future

Clinical trials are underway in Europe comparing prednisolone with PTX and comparing prednisolone with prednisolone + NAC. Transplantation centers in the United States are beginning to discuss liver transplantation for AH. Recently, the National Institute on Alcohol Abuse and Alcoholism has funded two consortia to evaluate new treatments for AH. The clinical trials in these consortia, as well as the detailed study of pathophysiology of liver injury that is obtained in conjunction with these trials, should provide critical data for the identification of future treatments.