Corticosteroids for alcoholic hepatitis--what's next?

Alcoholic hepatitis (AH) is observed in approximately 20% of heavy drinkers. The treatment of AH remains controversial and will, in the near future, constitute one of the main challenges to clinicians involved in the management of severe alcoholic liver disease [1]. Individual data from the last three randomized controlled trials (RCT), prospective studies and a recent RCT constantly showed that patients with Maddrey function (DF) R32 treated by corticosteroids had a 2-month survival of 80% [2–5]. Nevertheless, the efficacy of corticosteroids is still considered as a controversial issue for some authors. Alternative therapies are required to improve the prognosis of patients with a severe form of AH [6]. Progress in understanding the pathogenesis of AH is opening up an exciting new era and is lending impetus to future evaluation of new drugs targeting the TNF pathways [7–13]. However, the classical view that considers a treatment effective only when it improves survival needs to be modified. Indeed, the sample size of future studies comparing new drugs to corticosteroids will never be sufficient to detect a difference in short-term survival [14]. Clinicians need to promote new primary endpoints in future randomized controlled trials evaluating drugs in patients with severe AH (Table 1). We suggest that a treatment should be declared successful in patients demonstrating biological improvement and who remained alive 1 or 2 months.