Treatment with MARS was associated with a decrease in serum bilirubin, with a reduction of serum creatinine to less than 1.5 mg/dL, and with a marked improvement of hepatic encephalopathy (decrease of HE from grade II-IV to grade 0-I) in a high proportion of patients. These features occurred more frequently in patients treated by MARS plus standard medical therapy than in those only receiving standard medical

treatment alone (percent decrease in serum bilirubin: 26.4% versus 8.9%; P < 0.001; improvement of hepatic encephalopathy: OR: 0.37; 95% CI 0.12-1.09; P = 0.07; reversal of HRS OR: 0.40; 95% CI 0.15-1.07; P = 0.07). It is important Ku-0059436 research buy to point out that the more frequent decrease of serum creatinine below the 1.5 mg/dL threshold observed in patients treated with MARS occurred despite that patients in the control group received treatment with terlipressin plus albumin and a significant number of them also received artificial renal support. There was a clear relationship between the dose of MARS and its effects in the support of organ function since these were evident within the first 4 days of treatment, a period in which patients received one MARS session per day, but not during the subsequent weeks in which, according to the study protocol, a maximum of one session every 2 days was administered. Despite the beneficial effects of the use

of MARS in organ support, no improvement in short-term or mid-term transplant-free survival was observed. The 28- and 90-day probabilities GS-1101 order of survival were similar in patients treated with MARS plus standard medical therapy 上海皓元 and in those

receiving standard medical therapy alone. This lack of a survival improvement was also observed in three different predetermined subgroups (patients with hepatic encephalopathy, HRS, or progressive hyperbilirubinemia). Finally, a logistic model adjusted to potential confounding variables not completely balanced at baseline showed a slight reduction in 28-day transplant-free mortality in the MARS arm, but the difference was nonsignificant. Similar results have recently been reported in another large randomized controlled trial using the fractionated plasma separation and absorption device.26 The lack of improvement in survival with MARS despite observed improvement of HE and amelioration of renal function is a paradoxical finding that represents a major point of our study. One potential explanation is that the beneficial effects of MARS could have been counterbalanced by complications associated with the procedure. Another possibility is that although renal and/or cerebral function improved in many patients in the MARS arm, the difference between the effects obtained with MARS plus standard medical therapy versus those with standard medical therapy alone was not high enough to influence survival. Our results suggest that this possibility is the most likely explanation.