Molecular adsorbent recirculating system for acute and acute-on-chronic liver failure: a meta-analysis.

Molecular adsorbent recirculating system (MARS) is an important option for patients with liver failure to give them additional time for recovery or to serve as a "bridge" to transplantation. However, its effect on survival for such patients is not well known. Our aim was to assess the treatment effects of MARS on patients with acute and acute-on-chronic liver failure. The outcomes measure evaluated was survival. We searched Medline (1966-2002) and EMBASE (1974-2002) using the terms liver failure, liver support systems, and MARS. Our search was extended to the Cochrane Controlled Trials Registry Database, published abstracts from 5 international conferences, Teraklin (the manufacturer of MARS), known contacts, and bibliographies from each full-published report. We included trials published in English and non-English languages. Eligible studies were randomized and nonrandomized controlled trials, which compared the treatment effects of MARS with standard medical treatment. Of the 206 articles screened, 4 randomized controlled trials including 67 patients were analyzed. Two nonrandomized trials with 61 patients were used for explorative analysis. The methodology, population, intervention, and outcomes of each selected trial were evaluated by duplicate independent review. Disagreements were resolved by consensus. In the primary meta-analysis, MARS treatment did not appear to reduce mortality significantly compared with standard medical treatment [relative risk (RR), 0.56; 95% confidence interval (CI), 0.28-1.14; P = .11]. Only 1 of the 4 randomized trials analyzed showed significant reduction in mortality. Sensitivity analysis of 3 peer-reviewed trials did not reduce mortality significantly with MARS treatment (RR, 0.72; 95% CI, 0.37-1.40; P = .33). Subgroup analysis of 2 trials for acute liver failure and another 2 trails for acute-on-chronic liver failure also did not reveal any benefit to survival with MARS treatment. In contrast, explorative analysis of 2 nonrandomized trials showed a significant survival benefit with MARS treatment (RR, 0.36; 95% CI, 0.17-0.76; P = .007). This was possibly related to bias in the selection of patients in the nonrandomized trials. In conclusion, MARS treatment had no significant survival benefit on patients with liver failure when compared with standard medical therapy. However, we found only a few trials with a small number of patients for the analysis, allowing for the possibility of false negative and erroneous conclusions. Well-conducted randomized trials are strongly recommended to define the role of MARS in the treatment of patients with liver failure.