Background Conventional diagnostic procedures and therapy of acute liver failure (ALF) and acute-on-chronic liver failure (ACLF) focus on to identify triggering events of the acute deterioration of the liver function and to avoid them. Further objectives are to prevent the development respectively the progression of secondary organ dysfunctions or organ failure. Most of the times the endocrinological function of the liver can to a wide extent be compensated, but the removal of toxins can only marginally be substituted by conventional conservative therapy. To improve this component of the liver function is the main objective of extracorporal liver support systems. The following principles of liver support systems can be differentiated: Artificial systems, bioartifical systems and extracorporal liver perfusion systems. This HTA report focuses on artificial systems (e.g. BioLogic-DT/-DTPF, MARS, Prometheus), because only these approaches currently are relevant in the German health care system. In 2004 a category "Extracorporal liver assist device" was introduced in the list of "additional payments" in the German DRG-system, which makes reimbursement for hospitals using the technology in inpatient care possible, based on an hospital's individual contract with statutory sickness funds. Objectives To report the present evidence and future research need on medical efficacy and economic effectiveness of extracorporal liver support devices for treatment of patients with ALF or ACLF based on published literature data. Are artificial liver support systems efficient and effective in the treatment of ALF or ACLF? Methods An extensive, systematic literature search in medical, economic, and HTA literature data bases was performed. Relevant data were extracted and synthesised. Results Relevant controlled trials were detected for BioLogic-DT and MARS. No randomised controlled trial on Prometheus was found. None of the included studies on BioLogic-DT showed advantages of the technology compared with standard conventional therapy concerning survival, clinical scores or clinical surrogate parameter like laboratory tests of liver function. Some studies reported complications and side effects of BioLogic-DT. All studies were methodologically insufficient. Concerning the use of MARS overall five studies - three of them randomised - were identified. Two studies reported a significant higher 30d-survival after MARS compared to controls, one study showed a non-significant trend to a better survival probability after one year. The studies showed statistically significant advantages in severity of hepatic encephalopathy, routine lab tests and hemodynamic parameter of the MARS group. None of the studies reported relevant complications or side effects. Although the methodological quality of the studies is seen as slightly better than in the studies on BioLogic-DT, there are methodological limitations: The largest sample size of the randomised trials was twelve patients per group and the study population was highly selected. Because of the methodological limitations the results can hardly be generalised. Only two economic publications presenting analyses of MARS could be de-tected. One publication shows major methodological mistakes which make a further interpretation of the results impossible. The other publication presents an incremental cost-effectiveness of MARS of 29,719 EUR per life year gained after one year from a payer's perspective (German statutory sickness fund, neglecting the intervention costs because of lacking reimbursement at this time), respectively 79,075 EUR per life year gained from a societal perspective. Including health related quality of life aspects the incremental costs per QALY (Quality adjusted life years) gained were calculated to be 44,784 EUR from a payer's perspective respectively 119,162 EUR from a societal perspective. The authors state that prolonging the time horizon of the calculations would improve cost-effectiveness ratios. The limitations of the study design also limit the scientific evidence of the results. Conclusion The results of the detected publications do not give any evidence for a positive medical efficacy of BioLogic-DT. Concerning MARS there is some evidence for positive effects on 30d-survival, clinical parameter, and some lab tests, although the evidence is limited by the small number of studies and their methodological weakness. The currently strongly limited evidence shows a trend to an acceptable cost-effectiveness of MARS, although the results are based on only one non-randomised trial. To give valid recommendations concerning the medical efficacy as well as the cost-effectiveness of artificial liver support systems further studies are necessary.