• acute-on-chronic liver failure (ACLF);
  • hepatitis B;
  • lamivudine;
  • model for end-stage liver disease (MELD);
  • plasma exchange (PE)


Background and Aim:  We used the model for end-stage liver disease (MELD) scoring system to predict the 3-month prognosis of patients with acute-on-chronic liver failure (ACLF) after plasma exchange (PE) and lamivudine treatment, and studied the predictive factors on the prognosis of patients.

Methods:  A total of 280 patients treated with lamivudine were randomly divided into PE and control groups. The relationship between mortality and influential factors of patients was studied by univariate and multivariate analysis.

Results:  The mortality (49.4%) of patients in the PE group with a MELD score from 30 to 40 was lower than that (86.1%) of the control group (χ2 = 24.546, P < 0.01). The total bilirubin (TBIL) rebound rate of the dead group was significantly higher than that of the survival group (P < 0.01). Univariate analysis showed that mortality was significantly related to age (P = 0.003), treatment method (P = 0.000), TBIL (P = 0.010), MELD score (P = 0.001), international normalised ratio (P = 0.014), pretreatment HBV-DNA load (P = 0.000), decline of hepatitis B virus (HBV)–DNA load during therapy (P = 0.013), encephalopathy (P = 0.019), and hepatorenal syndrome (P = 0.026). In multivariate analysis, MELD scores of 30–40, treatment method (P = 0.003), pretreatment HBV-DNA load (P = 0.009), decline of HBV-DNA load during therapy (P = 0.016), and encephalopathy (P = 0.015) were independent predictors of mortality; for MELD scores above 40, only the MELD score (P = 0.012) was an independent predictive.

Conclusions:  PE significantly decreased the mortality of patients with a MELD score of 30–40. For ACLF patients with a MELD score of 30–40, a low viral load pretreatment and quick decline of HBV-DNA load are good predictors for the survival with PE and lamivudine treatment.