To assess and predict the effect and the clinical prognostic value of non-biotype artificial liver support system (NB-ALSS) in hepatitis B-related liver failure patients by model for end-stage liver disease (MELD) and Child-Turcotte-Pugh (CTP) score systems.

The randomly selected 210 cases of hepatitis B-related liver failure patients were divided into the artificial liver group(n = 115) and control group(n = 95), based on MELD and CTP scores. Each group were further divided into four subgroups: MELD< 20, 20 ≤ MELD< 30, 30 ≤ MELD< 40, and MELD> 40. The changes of MELD score between these two groups and of the laboratory parameters in the artificial liver group before and after the treatment were observed, and actual short-term mortality rate with expected mortality according to CTP score system were compared.

After the artificial liver treatment, prothrombin time international normalized ratio (INR) decreased significantly in four artificial liver subgroups [(1.3 ± 0.3) vs. (1.4 ± 0.3); (2.2 ± 0.8) vs. (2.6 ± 0.8); (4.1 ± 1.5) vs. (5.2 ± 1.7); (9.6 ± 2.8) vs. (12.2 ± 4.8); t = 4.303, 3.152, 3.545, 3.130; all P< 0.05], total bilirubin (TBIL) also showed significant improvement in these artificial liver groups after the treatment [(152 ± 74) μmol/L vs. (287 ± 118) μmol/L; (266 ± 160) μmol/L vs. (422 ± 114) μmol/L; (370 ± 144) μmol/L vs. (517 ± 126) μmol/L; (564 ± 180) μmol/L vs.(628 ± 121) μmol/L; t = 4.960, 5.951, 4.915, 2.577; all P< 0.05]. Except the MELD ≥ 40 subgroup, MELD scores in the other three artificial liver subgroups were markedly dropped after the treatment [(10.2 ± 3.4) vs. (16.6 ± 2.5); (18.2 ± 4.2) vs. (24.7 ± 2.6); (30.1 ± 7.5) vs. (36.2 ± 2.3); t = 7.036, 9.094, 5.476; all P< 0.05], MELD score decreased (△MELD) between the artificial liver groups and control groups were statistically significant in these three subgroups (t = 2.286, 2.906, 2.021; all P < 0.05). In the 20 ≤ MELD< 30 and 30 ≤ MELD< 40 subgroups, the mortality rates of the artificial liver groups were lower than expected mortality rates (40.8% vs. 76.0%, 51.4% vs. 83.0%; χ² = 12.119, 8.880; all P < 0.05), and also lower than that of the control groups (40.8% vs. 61.9%, 51.4% vs. 82.4%; χ² = 4.03, 4.71; all P < 0.05). Comparing the mortality rates between the artificial liver groups and control groups in the CTP C level: in the 20 ≤ MELD< 30 and 30 ≤ MELD< 40 subgroups, the mortality rates of two sets were significantly different(38.8% vs. 59.6%, 51.4% vs. 76.5%; χ² = 3.90, 4.40; all P < 0.05). Kaplan-Meier survival curves showed that in the 20 ≤ MELD< 30 and 30 ≤ MELD< 40 subgroups, the short-term survival rates in the artificial liver groups was higher than that in the control groups (χ² = 3.89, 5.70; all P < 0.05).

Combined MELD score and CTP classification system can assess the prognosis of NB-ALSS in treatment of hepatitis B-related liver failure patients.