The study is aimed to evaluate the additional role of ciprofloxacin therapy in severe alcoholic hepatitis combined to prednisolone therapy in an open-label placebo controlled manner.

Introduction: Alcoholic hepatitis (AH) is an inflammatory liver injury associated with longstanding excess alcohol consumption. Disease spectrum varies from asymptomatic transaminase elevations to fulminate liver failure. In hospital mortality in patients with severe alcoholic hepatitis not responding to corticosteroids is over 30 %.

Alcohol increases gut permeability and promotes the translocation of lipopolysaccharide (LPS) from the gut lumen the portal vein, and further to Kupffer cells, where LPS binds to CD14, ultimately activating multiple cytokine genes.

Diagnosis of AH is based on history of heavy alcohol use, symptoms like jaundice and on typical laboratory findings, and in uncertain cases on liver biopsy.

Determination of the severity of alcoholic hepatitis is essential for assessment of the disease prognosis and selection therapy. Cessation of alcohol consumption is mandatory for further therapy. Several scoring systems are available to assess the severity and the prognosis of alcohol hepatitis. Maddrey discrimination function (DF) is most widely used and enables to identify patients with severe alcohol hepatitis responding to corticosteroid therapy.

The first line therapy in severe alcoholic hepatitis (DF≥32) is prednisolone. However, those not responding to steroids have 77 % 6 months mortality.

New treatment options for severe AH are desperately needed. Although increased bacterial and LPS translocation are considered to have central role in the pathogenesis of AH no controlled studies of antibiotics in alcoholic hepatitis has been published. In Finland 600 AH requiring hospitalization are diagnosed annually.

Study objective: To evaluate to additional role of ciprofloxacin therapy in severe alcoholic hepatitis combined to prednisolone therapy.

Moreover, we try to find new and better predictors for liver injury and treatment response.

Patients: 150 AH patients, with Maddrey DF >32.

Randomization: Patients with severe AH are randomized at hospitalization 1:1 to receive:

1. Prednisolone 40 mg/day for 1 month, with decreasing by 5 mg/week + ciprofloxacin 1000 mg/ day for 120 days or
2. Prednisolone 40 mg/day for 1 month, with decreasing by 5 mg/week + placebo/ day for 120 days Measurement of response Early change in bilirubin levels (ECBL= S-Bil(Day 0)-S-Bil(Day7 )>0 Lille Score >0.45 day 7. Change in serum sterol levels as surrogate markers of cholesterol synthesis (reflecting liver function and severity of cholestasis) Primary end point Mortality at day 28, at 6 months and at 12 months Secondary end points: Proportion of patients with early change in bilirubin levels (ECBL= S-Bil(Day 0)-S-Bil(Day7 )>0 Proportion of patients with Lille Score >0.45 day 7 Recovery of liver function parameters in 1 and 3 months

Inclusion Criteria:

Severe alcoholic hepatitis; Maddrey above 300

Exclusion Criteria:

Viral hepatitis Remarkable bleeding in the gastrointestinal tract Serious bacterial infection Hepatorenal syndrome Earlier participation in this study Malignant disease not in remission Other liver disease affecting remarkably the outcome of alcoholic liver disease Mental retardation

No IPD will be shared with other researchers during the study or during the analysis of the results.