BACKGROUND There are almost no prospective, controlled and randomized clinical trials comparing different approaches towards hepatic assist. In order to create a basis for comparing the value of the existing different hepatic assist methods this article offers a systematic review of the world experience with allogeneic or xenogeneic extracorporeal liver perfusion (ECLP). METHODS An Internet-assisted search was conducted in the international literature published from 1964 to 2000. Only articles with a clear description of methodology and outcome of patients were included. For multivariate analysis of variance the general linear method (GLM) procedure was used. Differences within the groups were analyzed by chi-square test. Data of 198 patients were included into the statistical analysis for systematic review. RESULTS The long-term survival rate (SVR) of these patients was 26%, thus not exceeding published data concerning SVR under standard intensive care. Age below 40 years (P<0.029), coma stage lower than III-IV (P<0.003), total perfusion time over 10 hours (P<0.024), hepatitis B as cause for acute liver failure (ALF) (P<0.05) as well as use of baboon and human livers (P<0.02) were identified as independent positive prognostic markers for improved survival. ECLP as bridging therapy to liver transplantation was successful in 12 of 14 patients. CONCLUSIONS ECLP using pig livers did not surpass the success of conventional intensive care treatment. An additional effect of transgenic expression of human regulators of complement regulation in porcine livers has not yet been proven. ECLP with human livers not suitable for liver transplantation might prove effective and practicable for temporary hepatic support. Bridging to liver transplantation by long-term ECLP using porcine and human livers appears to have comparable efficacy as bioartificial support methods.