Pharmacological interventions for acute hepatitis B infection: An attempted network meta-analysis

Background: Infection with hepatitis B virus (HBV) can be symptomatic or asymptomatic. Apart from chronic HBV infection, the complications related to acute HBV infection are severe acute viral hepatitis and fulminant hepatitis characterised by liver failure. The optimal pharmacological treatment of acute HBV infection remains controversial. Objectives: To assess the benefits and harms of pharmacological interventions in the treatment of acute HBV infection through a network meta-analysis and to generate rankings of the available treatments according to their safety and efficacy. As it was not possible to assess whether the potential effect modifiers were similar across different comparisons, we did not perform the network meta-analysis, and instead, assessed the benefits and harms of different interventions using standard Cochrane methodological procedures. Search methods: We searched CENTRAL, MEDLINE, Embase, Science Citation Index Expanded, WHO International Clinical Trials Registry Platform, and randomised clinical trials (RCTs) registers to August 2016 to identify RCTs on pharmacological interventions for acute HBV infection. Selection criteria: RCTs, irrespective of language, blinding, or publication status in participants with acute HBV infection. We excluded trials if participants had previously undergone liver transplantation and had other coexisting viral diseases such as hepatitis C virus and HIV. We considered any of the various pharmacological interventions compared with each other or with placebo, or no intervention. Data collection and analysis: We calculated the odds ratio (OR) and rate ratio with 95% confidence intervals (CI) using both fixed-effect and random-effects models based on available-participant analysis with Review Manager 5. We assessed risk of bias, controlled risk of random errors with Trial Sequential Analysis, and assessed the quality of the evidence using GRADE. Main results: Seven trials (597 participants) met our review inclusion criteria. All trials provided information for one or more outcomes; however, five participants were excluded from analysis by study authors. All the trials were at high risk of bias. Overall, all the evidence was low or very low quality evidence because of risk of bias (downgraded one level for risk of bias), small sample size (downgraded one level for imprecision), and wide CIs (downgraded one more level for imprecision in some comparisons). Of the seven trials, six were two-armed trials, while one trial was a three-armed trial. The comparisons included hepatitis B immunoglobulin (HBIG) versus placebo (one trial; 55 participants); interferon versus placebo (two trials; 200 participants); lamivudine versus placebo or no intervention (four trials; 316 participants); lamivudine versus entecavir (one trial; 90 participants); and entecavir versus no intervention (one trial; 131 participants). One trial included only people with acute HBV with hepatic encephalopathy (i.e. people with fulminant liver failure); one trial included only people with severe acute HBV, but it did not state whether any of the people also had fulminant HBV infection; three trials excluded fulminant HBV infection; and two trials did not report the severity of acute HBV infection. The mean or median follow-up period in the trials ranged from three to 12 months in the trials that provided this information. There was no evidence of any differences in short-term mortality (less than one year) in any of the comparisons: HBIG versus placebo (OR 1.13, 95% CI 0.36 to 3.54; participants = 55; 1 trial), lamivudine versus placebo or no intervention (OR 1.29, 95% CI 0.33 to 4.99; participants = 250; 2 trials); lamivudine versus entecavir (OR 1.23, 95% CI 0.13 to 11.65; participants = 90; 1 trial), or entecavir ver