As reported in the literature, the mortality rates for patients with Acute Hepatic Failure (AHF) approaches 80% in cases in which liver transplantation is not possible. Post-transplant mortality mostly depends on the severity of the neurological condition at the time of the operation (20% in I-II degree coma patients and 44% in III degree coma patients). The primary indications for liver transplantation in AHF are Fulminant Hepatitis (FH)(93%), Subfulminant Hepatitis (5%) and other indications (2%). Other causes of AHF are Primary Non-Function (PNF) and Delayed Function (DF), which occur in 7-10%. Therefore it becomes necessary to monitor the patients with a Liver Support Device to be able to improve the clinical condition of the patients before liver transplantation (LT). In our experience we used the Molecular Adsorbent Recirculating System (MARS) (MARS Monitor; Teraklin AG, Rostock Germany), which enables the selective removal of albumin-bound substances accumulating in liver failure by the use of albumin-enriched dialysate. The system is used as a bridging device to orthotopic liver transplantation (OLT) of patients with FHF. We studied 34 patients, including 16 males and 18 females: 9 were affected by Primary-Non-Function (PNF), nine by Fulminant Hepatitis (FH), six by Delayed-Non-Function (DNF), and ten by Acute on Chronic Hepatic Failure (AOCHF). The average age of the patients was 41.8 years and the average number of applications was 6.4; the median length of application was about eight hours. The parameters that we monitored, before and after each treatment, were neurological status (EEG, cerebral CT, Glasgow Coma Score), haemodynamic parameters, acid base equilibrium, and blood gas analysis. We also monitored hepatic and renal function. In addition, the clinical conditions of the patients were monitored using kidney and liver ultrasound/ultrasonography (US). Inclusion criteria were bilirubin > 15 mg/dL, ammonia > 160 micro g/dL and a Glasgow Coma Score between 6 and 11. The reduction of bilirubin and ammonia were very significant (P < 0.01), whereas the changes of International Normalized Ratio (INR) were not significant. Also the modifications of albumin, total protein, sodium, potassium and calcium were not significant. In conclusion, four out of nine patients with PNF are alive without a second transplantation and were discharged after about 48 days; four out of nine underwent OLT, while one out of nine died; five out of six patients with DF are alive without a second transplantation, and they were discharged after an average time of 55.5 days, one out of six died; six out of nine patients with fulminant hepatitis underwent OLT and four of these are alive, while two died due to sepsis; three patients are alive without OLT. Four patients with AOCHF underwent OLT and are alive, three patients are alive and on a waiting list, two died while on a waiting list and one patient who experienced reactivation of HBV infection during chemotherapy for non-Hodgkin's lymphoma is alive. In spite of the limited number of cases of our study, we believe that MARS can be applied with high tolerance for a very long period of time. In addition, its repeatability allows it to be used in patients with DNF and FH as a bridge to transplant. In patients with DNF, it is used while waiting for complete recovery of the transplanted organ.