Abstract
RATIONALE: Acute liver failure (ALF) is a life-threatening condition, characterised by serious damage to the liver cells and function. The aetiology of ALF determines the prognosis, and without immediate care or liver transplantation, people with ALF will develop life-threatening complications and die. Liver support systems are used as a bridge therapy to provide detoxification and restore liver function until either functional recovery occurs or a liver becomes available for transplantation. Liver support systems have continued to be used despite conflicting results of randomised clinical trials (RCTs). OBJECTIVES: To assess the benefits and harms of liver support systems plus standard medical care versus standard medical care for adults with acute liver failure. SEARCH METHODS: We identified trials through manual and electronic searches in The Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials, MEDLINE, Embase, LILACS, Science Citation Index Expanded, and Conference Proceedings Citation Index - Science. We also searched reference lists, trials registers, other sources, and contacted trial authors to identify trials for inclusion in the review. The latest search date was 10 July 2025. ELIGIBILITY CRITERIA: We included RCTs (published as abstracts or full-text articles) with a parallel-group design. We included the trials irrespective of the publication language and year, country of performance, or reported outcomes. The trials compared liver support systems plus standard medical care versus standard medical care. We included trials in adults ≥ 18 years old, of any sex, ethnicity, or race, diagnosed with ALF (including previous definitions) and irrespective of aetiology. We excluded trials with mixed populations where participants with ALF were < 10 % of the total number of participants; quasi-randomised trials and other observational studies. OUTCOMES: Critical outcomes were: all-cause mortality by day 28, all-cause mortality, serious adverse events (measured by the total number of people with one or more events), and the proportion of people with liver transplantation. Important outcomes included health-related quality of life, hepatic encephalopathy, and multi-organ failure (measured by the total number of events, including people who needed liver transplantation). Except for all-cause mortality by day 28, we planned to assess the remaining outcomes at their maximal follow-up. RISK OF BIAS: We assessed the risk of bias with the outcome-based RoB 2 tool. We judged the overall bias risk of each outcome result using the intention-to-treat principle. We resolved disagreements through discussion. SYNTHESIS METHODS: We followed the Methodological Expectations for Cochrane Intervention Reviews when conducting the review, and PRISMA 2020 for the reporting. We conducted meta-analyses of available data for each outcome at the prespecified time points. We analysed dichotomous data using the risk ratio (RR) and the random-effects model, and calculated the rate ratio from the total number of events in the separate groups of a trial, with a set-up of contrast-level data and the inverse variance method. For both types of analyses, we used a 95% confidence interval (CI). We assessed the certainty of evidence of seven prespecified outcome results using the five GRADE domains (risk of bias, heterogeneity, indirectness, imprecision using a minimally contextualised approach, and publication bias). INCLUDED STUDIES: We included 11 parallel-group RCTs with 681 randomised participants (342 to the experimental group and 339 to the control intervention groups). Two of the trials were only published as abstracts. In the trials reporting on age, there were a few participants (≤ 18 years old) with inseparable data (age range: 10 to 69 years), whom we included in the analyses. The trials compared artificial liver support systems (eight trials) or bioartificial liver support systems (three trials) plus standard treatment and care versus standard treatment and care alone in people with ALF. They were conducted in Europe (Austria, Germany, Denmark, Finland, France, the UK), India, and the USA. The trial participants were treated in intensive care units. The maximal follow-up for all-cause mortality and hepatic encephalopathy was 90 days; for serious adverse events, liver transplantation, and multi-organ failure, it was 360 days. SYNTHESIS OF RESULTS: We downgraded the certainty of evidence to very low due to risk of bias and imprecision in all outcome results, and heterogeneity in the two mortality outcome results. Liver support systems may have little to no effect on all-cause mortality by day 28 (RR 0.83, 95% CI 0.58 to 1.19; I² = 66%; 9 studies, 539 participants; very low-certainty evidence), all-cause mortality at maximal follow-up (RR 0.82, 95% CI 0.63 to 1.07; I² = 52%; 11 studies, 681 participants; very low-certainty evidence), on the number of serious adverse events (Rate ratio 0.93, 95% CI 0.81 to 1.07; I² = 0%; 11 studies, 681 participants; very low-certainty evidence), and on the proportion of people with liver transplantation (RR 1.03, 95% CI 0.87 to 1.21; I² = 0%; 5 studies, 519 participants; very low-certainty evidence). No trial included health-related quality of life as an outcome. The evidence is very uncertain about the effect of liver support systems on hepatic encephalopathy (RR 0.40, 95% CI 0.16 to 1.04; I² = 0%; 2 studies, 34 participants; very low-certainty evidence). Liver support systems may have little to no effect on the number of multi-organ failures (Rate ratio 0.92, 95% CI 0.72 to 1.18; I² = 0%; 6 studies, 501 participants; very low-certainty evidence). Five trials had no for-profit funding, four trials declared full or partial for-profit funding and two trials did not declare funding. One study is awaiting classification. We found no ongoing randomised clinical trials. AUTHORS' CONCLUSIONS: The trials assessed experimental liver support systems in adults with ALF. Based on very low-certainty evidence, we are very uncertain about the effects of liver support systems plus standard treatment and care versus standard treatment and care on all-cause mortality by day 28 and at maximal follow-up, serious adverse events, liver transplantation, hepatic encephalopathy, and multi-organ failure, including liver failure, due to insufficient data. We lack data on health-related quality of life. FUNDING: This Cochrane review had no dedicated funding. REGISTRATION: The protocol for this review is available via DOI: 10.1002/14651858.CD015059.