Predictive Parameters After Molecular Absorbent Recirculating System Treatment Integrated With Model for End Stage Liver Disease Model in Patients With Acute-on-Chronic Liver Failure

Aim: The aim of study was to highlight parameters that in association with Model for End-stage Liver Disease (MELD) provide predictive criteria for long-term survival after treatment with the Molecular Adsorbent Recirculating System (MARS). Two homogenous groups were studied: one treated with standard medical therapy (SMT) and the other, with MARS. Materials and methods: Twenty acute-on-chronic liver failure patients on the waiting list for liver transplantation and affected by alcoholic cirrhosis with similar MELD scores (20-29) were evaluated for 7 days from inclusion and for 6-month survival. Ten patients (seven males and three females) were treated with MARS. Their mean age was 48.5 years (range = 35-61). The number of MARS applications was six for 6 consecutive days, and the length of the applications was 8 hours. Ten other patients (seven males and three females) were treated with SMT, including prophylaxis against bacterial infections and judicious use of diuretics. The precipitating factors were also treated appropriately. The mean age of the patients was 51 years (range = 37-64). All the variables that were significant upon univariate analysis were enrolled in a receiver operating characteristic analysis, with the intention to detect predictive parameters for patient death at 6 months. We considered a significant area under curve (AUC) value to be greater than 0.5. Results: Among 11 patients who died within 6 months there were in the MARS group and eight in the SMT group: the 3- and 6-month patient survival rates were 90% and 70% versus 30% and 20% in the two groups, respectively. Nine measures resulted in an AUC > 0.5: DeltaMELD; interleukin (IL)-8; IL-6; tumor necrosis factor- alpha, MELD score; creatinine, bilirubin international normalized ratio (INR) and cardiac index. DeltaMELD and postoperative IL-8 concentrations showed better results (AUC = 0.899), followed by postoperative creatinine (AUC = 0.879), postoperative cardiac index (AUC = 0.833), and postoperative INR (AUC = 0.818). Postoperative creatinine showed the best sensitivity (100%), while IL-8, the best specificity (88.9%). Conclusion: A combination of biochemical and clinical variables probably represent the best way to predict the survival of patients, allowing physicians to select the best therapies for each patient. © 2010 Elsevier Inc. All rights reserved.