Systematic review and meta-analysis of survival following extracorporeal liver support

Background: Extracorporeal liver support (ELS) systems offer the potential to prolong survival in acute and acute-on-chronic liver failure. However, the literature has been unclear on their specific role and influence on mortality. This meta-analysis aimed to test the hypothesis that ELS improves survival in acute and acute-on-chronic liver failure. Methods: Clinical trials citing MeSH terms 'liver failure' and 'liver, artificial' were identified by searching MEDLINE, Embase and the Cochrane registry of randomized controlled trials (RCTs) between January 1995 and January 2010. Only RCTs comparing ELS with standard medical therapy in acute or acute-on-chronic liver failure were included. A predefined data collection pro forma was used and study quality assessed according to Consolidated Standards of Reporting Trials (CONSORT) criteria. Risk ratio was used as the effect size measure according to a random-effects model. Results: The search strategy revealed 74 clinical studies including 17 RCTs, five case-control studies and 52 cohort studies. Eight RCTs were suitable for inclusion, three addressing acute liver failure (198 participants) and five acute-on-chronic liver failure (157 participants). The mean CONSORT score was 14 (range 11-20). Overall ELS therapy significantly improved survival in acute liver failure (risk ratio 0.70; P = 0.05). The number needed to treat to prevent one death in acute liver failure was eight. No significant survival benefit was demonstrated in acute-on-chronic liver failure (risk ratio 0.87; P = 0.37). Conclusion: ELS systems appear to improve survival in acute liver failure. There is, however, no evidence that they improve survival in acute-on-chronic liver failure.