Comparative Effectiveness of Pharmacological Interventions for Severe Alcoholic Hepatitis: A Systematic Review and Network Meta-analysis

BACKGROUND & AIMS: Severe alcoholic hepatitis (AH) has high mortality. We assessed the comparative effectiveness of pharmacological interventions for severe AH, through a network meta-analysis combining direct and indirect treatment comparisons. METHODS: We conducted a systematic literature review, through February 2015, for randomized controlled trials of adults with severe AH (discriminant function >= 32 and/or hepatic encephalopathy) that compared the efficacy of active pharmacologic interventions (corticosteroids, pentoxifylline, and N-acetylcysteine [NAC], alone or in combination) with each other or placebo, in reducing short-term mortality (primary outcome) and medium-term mortality, acute kidney injury, and/or infections (secondary outcomes). We performed direct and Bayesian network meta-analysis for all treatments, and used Grading of Recommendations Assessment, Development and Evaluation criteria to appraise quality of evidence. RESULTS: We included 22 randomized controlled trials (2621 patients) comparing 5 different interventions. In a direct met-aanalysis, only corticosteroids decreased risk of short-term mortality. In a network meta-analysis, moderate quality evidence supported the use of corticosteroids alone (relative risk [RR], 0.54; 95% credible interval [CrI], 0.39-0.73) or in combination with pentoxifylline (RR, 0.53; 95% CrI, 0.36-0.78) or NAC (RR, 0.15; 95% CI, 0.05-0.39), to reduce short-term mortality; low quality evidence showed that pentoxifylline also decreased short-term mortality (RR, 0.70; 95% CrI, 0.50-0.97). The addition of NAC, but not pentoxifylline, to corticosteroids may be superior to corticosteroids alone for reducing short-term mortality. No treatment was effective in reducing medium-term mortality. Imprecise estimates and the small number of direct trials lowered the confidence in several comparisons. CONCLUSIONS: In patients with severe AH, pentoxifylline and corticosteroids (alone and in combination with pentoxifylline or NAC) can reduce short-term mortality. No treatment decreases risk of medium-term mortality.