Objective:Comparative efficacy and safety profiles of tenofovir disoproxil fumarate (TDF) versus tenofovir alafenamide fumarate (TAF) in chronic hepatitis B (CHB) patients with high viral load.Methods:CHB patients with high viral load (hepatitis B virus (HBV) DNA>7 lg IU/mL) receiving TDF ( n=155) or TAF ( n=157) monotherapy were included between December 1st, 2022 and December 1st, 2023, to compare the rates of undetectable HBV DNA (<20 IU/mL), the alanine transaminase (ALT) normalization rate, hepatitis B e antigen (HBeAg) seroconversion rate, renal function and lipid profiles at 48 weeks of treatment. The statistical analysis was performed by the two independent samples t test, Mann-Whitney U test, chi-square test or Fisher′s exact probability test. Results:At week 48 of treatment, the TDF group achieved significantly higher HBV DNA undetectability rates (49.03%(76/155) vs 29.30%(46/157)) and greater mean reduction ((6.05±0.81) lg IU/mL vs (5.57±1.02) lg IU/mL) than the TAF group ( χ2=12.75, t=-4.65, both P<0.001). The ALT normalization rate of patients in the TDF and TAF groups were 77.19%(88/114) and 72.50%(87/120), and HBeAg seroconversion rates were 4.00%(6/150) and 2.67%(4/150), respectively, with no statistically significant differences between the two groups (all P>0.05). The differences in the changes and abnormal rates of serum creatinine and estimated glomerular filtration rate between the two groups of patients were not statistically significant (both P>0.05). The magnitude of decrease (0.15(-0.02, 0.38) mmol/L vs 0.06(-0.06, 0.20) mmol/L) and abnormal rate (27.73%(33/119) vs 16.39%(20/122)) in high-density lipoprotein cholesterol in the TDF group was higher than those in the TAF group, while the increases in low-density lipoprotein cholesterol (-0.09(-0.39, 0.25) mmol/L vs 0.05(-0.31, 0.42) mmol/L), total cholesterol (TC) (-0.23(-0.75, 0.23) mmol/L vs 0.08(-0.35, 0.57) mmol/L), and triglyceride (-0.12(-0.40, 0.06) mmol/L vs 0.00(-0.19, 0.24) mmol/L), as well as the abnormal rate of TC (4.31%(5/116) vs 15.75%(20/127)), in the TDF group were lower than those in the TAF group ( Z=-3.19, χ2=4.51, Z=2.17, Z=4.09, Z=3.71, χ2=8.59, all P<0.05). Conclusion:TDF demonstrated superior efficacy and better safety profiles compared to TAF in high viral load CHB patients.

Objective:To compare the anticoagulant efficacy and safety between argatroban and heparin in patients with liver failure complicated with hepatic encephalopathy undergoing artificial liver treatment.Methods:A total of 207 patients with liver failure complicated with hepatic encephalopathy who received artificial liver treatment in the intensive care unit (ICU) of Wuxi No.5 People′s Hospital from January 2021 to October 2024 were enrolled, including 105 cases in the argatroban group and 102 cases in the heparin group. Changes in coagulation function, hemoglobin (Hb), platelet (PLT) count, and model for end-stage liver disease (MELD) score before and after artificial liver treatment were compared between the two groups. The formation of deep vein thrombosis in the lower extremities, coagulation in the extracorporeal circulation circuit and plasma separator, bleeding at the deep venous catheter site were compared between the two groups. The 28-day survival outcome of the patient were recorded. Two independent sample t-test, rank sum test, and chi-square test were used for statistical comparisons, and the Kaplan-Meier method and log-rank test were used to analyze the survival rate of patients. Results:There were no statistically significant differences in activated partial thromboplastin (APTT), international normalized ratio (INR), Hb and PLT count before and after artificial liver treatment in the argatroban group ( Z=-1.74, -1.80, -1.26 and -0.52, respectively, all P>0.05), while the MELD score after treatment was lower than that before treatment and the difference was statistically significant ( t=6.49, P<0.001). After artificial liver treatment, the APTT in the argatroban group was 47.10(42.65, 51.90) s, which was shorter than that in the heparin group (56.05(50.02, 63.00) s). The INR, Hb, and PLT count in the argatroban group were 2.00(1.65, 2.54), 98.00(88.00, 112.00) g/L, and 92.00(75.50, 106.00)×10 9/L, respectively, which were all higher than those in the heparin group, which were 1.56(1.22, 1.93) g/L, 90.50(80.00, 104.75) g/L, and 74.00(64.75, 99.50)×10 9/L, respectively. The differences were all statistically significant ( Z=-7.16, -5.28, -3.05 and -3.32, respectively, all P<0.05). There was no statistically significant difference in MELD scores between the two groups ( P=0.250). The incidence of coagulation in the extracorporeal circulation circuit and plasma separator and bleeding at the deep venous catheter site in the argatroban group was 5.71%(6/105) and 1.90%(2/105), respectively, which were both lower than those in the heparin group (14.71%(15/102) and 9.80%(10/102), respectively). The differences were both statistically significant ( χ2=4.59 and 5.91, respectively, both P<0.05). At the end of the 28-day follow-up, the mortality rates in the argatroban group and the heparin group were 22.9%(24/105) and 34.3%(35/102), respectively, and the difference was not statistically significant ( χ2=3.33, P=0.068). There was no statistically significant difference in the 28-day survival rate between the argatroban group and the heparin group ( χ2=2.09, P>0.05). Conclusions:Argatroban has a relatively minor impact on PLT count and Hb when it is used in artificial liver treatment for patients with liver failure complicated with hepatic encephalopathy. The incidence of coagulation in extracorporeal circulation circuits and plasma separators is low, and the risk of bleeding at the deep venous catheters is low. Argatroban is highly safe, which provides a new anticoagulation option for patients with a high risk of bleeding.

Objective:To explore the correlation and predictive value of the blood urea nitrogen to serum albumin ratio (BAR) in the development of sepsis among patients with acute-on-chronic liver failure (ACLF).Methods:A total of 410 patients diagnosed with ACLF who were admitted to Wuxi Fifth People′s Hospital between January 1st, 2020 and December 31st, 2024 were enrolled in this study. Demographic information, laboratory test indicators, and other clinical data were retrospectively analyzed. Participants were stratified into two groups using a 6∶4 allocation ratio, comprising a training set of 246 patients and a validation set of 164 patients, the clinical data of two groups were compared. Logistic regression was employed to evalute the influencing factors of sepsis during hospitalization in ACLF patients. Additionally, the predictive value of different factors for sepsis occurrence was evaluated using receiver-operating characteristic curve analysis. DeLong test was used to compare the area under the curve.Results:The comparison of baseline data between the training set and the validation set revealed no statistically significant differences (all P>0.05). A total of 197 sepsis cases were observed during the study period. Multivariate logistic regression analysis revealed that both BAR and the sequential organ failure assessment (SOFA) score were independent influencing factors for sepsis development in ACLF patients (odds ratio ( OR)=1.274, 95% confidence interval (95% CI) 1.075 to 1.510, P=0.005; OR=1.142, 95% CI 1.038 to 1.256, P=0.006). In the training set, the area under the curve (AUC) of BAR for predicting sepsis in ACLF patients was 0.802, which was superior to that of the SOFA score (AUC=0.706) ( Z=2.16, P=0.031). The validation set showed the predictive ability of BAR with an AUC of 0.726, which was superior to the SOFA score′s performance (AUC=0.606) ( Z=2.28, P=0.023). Conclusions:BAR could independently predict sepsis development in ACLF patients with significant prognostic value. BAR could be used as a clinically useful biomarker for sepsis risk stratification.

Objective:To explore the correlation and predictive value of the blood urea nitrogen to serum albumin ratio (BAR) in the development of sepsis among patients with acute-on-chronic liver failure (ACLF).Methods:A total of 410 patients diagnosed with ACLF who were admitted to Wuxi Fifth People′s Hospital between January 1st, 2020 and December 31st, 2024 were enrolled in this study. Demographic information, laboratory test indicators, and other clinical data were retrospectively analyzed. Participants were stratified into two groups using a 6∶4 allocation ratio, comprising a training set of 246 patients and a validation set of 164 patients, the clinical data of two groups were compared. Logistic regression was employed to evalute the influencing factors of sepsis during hospitalization in ACLF patients. Additionally, the predictive value of different factors for sepsis occurrence was evaluated using receiver-operating characteristic curve analysis. DeLong test was used to compare the area under the curve.Results:The comparison of baseline data between the training set and the validation set revealed no statistically significant differences (all P>0.05). A total of 197 sepsis cases were observed during the study period. Multivariate logistic regression analysis revealed that both BAR and the sequential organ failure assessment (SOFA) score were independent influencing factors for sepsis development in ACLF patients (odds ratio ( OR)=1.274, 95% confidence interval (95% CI) 1.075 to 1.510, P=0.005; OR=1.142, 95% CI 1.038 to 1.256, P=0.006). In the training set, the area under the curve (AUC) of BAR for predicting sepsis in ACLF patients was 0.802, which was superior to that of the SOFA score (AUC=0.706) ( Z=2.16, P=0.031). The validation set showed the predictive ability of BAR with an AUC of 0.726, which was superior to the SOFA score′s performance (AUC=0.606) ( Z=2.28, P=0.023). Conclusions:BAR could independently predict sepsis development in ACLF patients with significant prognostic value. BAR could be used as a clinically useful biomarker for sepsis risk stratification.

Objective:Comparative efficacy and safety profiles of tenofovir disoproxil fumarate (TDF) versus tenofovir alafenamide fumarate (TAF) in chronic hepatitis B (CHB) patients with high viral load.Methods:CHB patients with high viral load (hepatitis B virus (HBV) DNA>7 lg IU/mL) receiving TDF ( n=155) or TAF ( n=157) monotherapy were included between December 1st, 2022 and December 1st, 2023, to compare the rates of undetectable HBV DNA (<20 IU/mL), the alanine transaminase (ALT) normalization rate, hepatitis B e antigen (HBeAg) seroconversion rate, renal function and lipid profiles at 48 weeks of treatment. The statistical analysis was performed by the two independent samples t test, Mann-Whitney U test, chi-square test or Fisher′s exact probability test. Results:At week 48 of treatment, the TDF group achieved significantly higher HBV DNA undetectability rates (49.03%(76/155) vs 29.30%(46/157)) and greater mean reduction ((6.05±0.81) lg IU/mL vs (5.57±1.02) lg IU/mL) than the TAF group ( χ2=12.75, t=-4.65, both P<0.001). The ALT normalization rate of patients in the TDF and TAF groups were 77.19%(88/114) and 72.50%(87/120), and HBeAg seroconversion rates were 4.00%(6/150) and 2.67%(4/150), respectively, with no statistically significant differences between the two groups (all P>0.05). The differences in the changes and abnormal rates of serum creatinine and estimated glomerular filtration rate between the two groups of patients were not statistically significant (both P>0.05). The magnitude of decrease (0.15(-0.02, 0.38) mmol/L vs 0.06(-0.06, 0.20) mmol/L) and abnormal rate (27.73%(33/119) vs 16.39%(20/122)) in high-density lipoprotein cholesterol in the TDF group was higher than those in the TAF group, while the increases in low-density lipoprotein cholesterol (-0.09(-0.39, 0.25) mmol/L vs 0.05(-0.31, 0.42) mmol/L), total cholesterol (TC) (-0.23(-0.75, 0.23) mmol/L vs 0.08(-0.35, 0.57) mmol/L), and triglyceride (-0.12(-0.40, 0.06) mmol/L vs 0.00(-0.19, 0.24) mmol/L), as well as the abnormal rate of TC (4.31%(5/116) vs 15.75%(20/127)), in the TDF group were lower than those in the TAF group ( Z=-3.19, χ2=4.51, Z=2.17, Z=4.09, Z=3.71, χ2=8.59, all P<0.05). Conclusion:TDF demonstrated superior efficacy and better safety profiles compared to TAF in high viral load CHB patients.

Objective:To compare the anticoagulant efficacy and safety between argatroban and heparin in patients with liver failure complicated with hepatic encephalopathy undergoing artificial liver treatment.Methods:A total of 207 patients with liver failure complicated with hepatic encephalopathy who received artificial liver treatment in the intensive care unit (ICU) of Wuxi No.5 People′s Hospital from January 2021 to October 2024 were enrolled, including 105 cases in the argatroban group and 102 cases in the heparin group. Changes in coagulation function, hemoglobin (Hb), platelet (PLT) count, and model for end-stage liver disease (MELD) score before and after artificial liver treatment were compared between the two groups. The formation of deep vein thrombosis in the lower extremities, coagulation in the extracorporeal circulation circuit and plasma separator, bleeding at the deep venous catheter site were compared between the two groups. The 28-day survival outcome of the patient were recorded. Two independent sample t-test, rank sum test, and chi-square test were used for statistical comparisons, and the Kaplan-Meier method and log-rank test were used to analyze the survival rate of patients. Results:There were no statistically significant differences in activated partial thromboplastin (APTT), international normalized ratio (INR), Hb and PLT count before and after artificial liver treatment in the argatroban group ( Z=-1.74, -1.80, -1.26 and -0.52, respectively, all P>0.05), while the MELD score after treatment was lower than that before treatment and the difference was statistically significant ( t=6.49, P<0.001). After artificial liver treatment, the APTT in the argatroban group was 47.10(42.65, 51.90) s, which was shorter than that in the heparin group (56.05(50.02, 63.00) s). The INR, Hb, and PLT count in the argatroban group were 2.00(1.65, 2.54), 98.00(88.00, 112.00) g/L, and 92.00(75.50, 106.00)×10 9/L, respectively, which were all higher than those in the heparin group, which were 1.56(1.22, 1.93) g/L, 90.50(80.00, 104.75) g/L, and 74.00(64.75, 99.50)×10 9/L, respectively. The differences were all statistically significant ( Z=-7.16, -5.28, -3.05 and -3.32, respectively, all P<0.05). There was no statistically significant difference in MELD scores between the two groups ( P=0.250). The incidence of coagulation in the extracorporeal circulation circuit and plasma separator and bleeding at the deep venous catheter site in the argatroban group was 5.71%(6/105) and 1.90%(2/105), respectively, which were both lower than those in the heparin group (14.71%(15/102) and 9.80%(10/102), respectively). The differences were both statistically significant ( χ2=4.59 and 5.91, respectively, both P<0.05). At the end of the 28-day follow-up, the mortality rates in the argatroban group and the heparin group were 22.9%(24/105) and 34.3%(35/102), respectively, and the difference was not statistically significant ( χ2=3.33, P=0.068). There was no statistically significant difference in the 28-day survival rate between the argatroban group and the heparin group ( χ2=2.09, P>0.05). Conclusions:Argatroban has a relatively minor impact on PLT count and Hb when it is used in artificial liver treatment for patients with liver failure complicated with hepatic encephalopathy. The incidence of coagulation in extracorporeal circulation circuits and plasma separators is low, and the risk of bleeding at the deep venous catheters is low. Argatroban is highly safe, which provides a new anticoagulation option for patients with a high risk of bleeding.

Objective:To establish a novel laryngopharyngeal reflux model in Bama minipigs excluding concurrent gastroesophageal reflux through endoscopic cricopharyngeal myotomy.Methods:Twelve 8-month-old male Bama minipigs were randomly assigned to three groups: Group 1 underwent cricopharyngeal myotomy alone, Group 2 underwent combined cricopharyngeal and lower esophageal sphincter myotomy, and Group 3 served as the control group. Following a one-week acclimatization period, the respective surgical procedures were performed. At 2 weeks postoperatively, laryngopharyngeal pH monitoring was conducted on all pigs. At 8 weeks, histopathological assessment using hematoxylin and eosin (HE) staining, transmission electron microscopy of the laryngopharyngeal mucosa, and quantification of pepsin in the laryngopharyngeal and distal esophageal mucosa were performed to analyze intergroup differences and to elucidate the occurrence and pathologic featuresof LPR.Results:Two weeks postoperatively, experimental groups exhibited laryngopharyngeal reflux episodes with pH<5.0, in contrast to the control group. HE staining at 8 weeks revealed inflammatory changes in the laryngopharyngeal mucosa of Groups 1 and 2, accompanied by increased intercellular spaces and decreased desmosome density under electron microscopy, indicating a pathogenic mechanism involving disruption of intercellular junctions by refluxate. Statistically significant differences in pepsin expression ofthe vocalcords mucosal were observed among groups( F=88.427, P<0.001).Group 2 exhibited elevated pepsin expression in the distal esophageal mucosa than Groups 1 and 3, suggesting concurrent GERD only occured in Group 2. Conclusion:Endoscopic cricopharyngeal myotomy induces LPRD in Bama minipigs without concurrent GERD by reducing upper esophageal sphincter pressure, thereby offering a model that closely resembles the clinical features of LPRD.

Objective:To establish a novel laryngopharyngeal reflux model in Bama minipigs excluding concurrent gastroesophageal reflux through endoscopic cricopharyngeal myotomy.Methods:Twelve 8-month-old male Bama minipigs were randomly assigned to three groups: Group 1 underwent cricopharyngeal myotomy alone, Group 2 underwent combined cricopharyngeal and lower esophageal sphincter myotomy, and Group 3 served as the control group. Following a one-week acclimatization period, the respective surgical procedures were performed. At 2 weeks postoperatively, laryngopharyngeal pH monitoring was conducted on all pigs. At 8 weeks, histopathological assessment using hematoxylin and eosin (HE) staining, transmission electron microscopy of the laryngopharyngeal mucosa, and quantification of pepsin in the laryngopharyngeal and distal esophageal mucosa were performed to analyze intergroup differences and to elucidate the occurrence and pathologic featuresof LPR.Results:Two weeks postoperatively, experimental groups exhibited laryngopharyngeal reflux episodes with pH<5.0, in contrast to the control group. HE staining at 8 weeks revealed inflammatory changes in the laryngopharyngeal mucosa of Groups 1 and 2, accompanied by increased intercellular spaces and decreased desmosome density under electron microscopy, indicating a pathogenic mechanism involving disruption of intercellular junctions by refluxate. Statistically significant differences in pepsin expression ofthe vocalcords mucosal were observed among groups( F=88.427, P<0.001).Group 2 exhibited elevated pepsin expression in the distal esophageal mucosa than Groups 1 and 3, suggesting concurrent GERD only occured in Group 2. Conclusion:Endoscopic cricopharyngeal myotomy induces LPRD in Bama minipigs without concurrent GERD by reducing upper esophageal sphincter pressure, thereby offering a model that closely resembles the clinical features of LPRD.

Objective:To analyze the impact of baseline quantification of hepatitis B core antibody (qHBcAb) on prognosis of patients with hepatitis B virus (HBV) related acute-on-chronic liver failure (HBV-ACLF).Methods:A total of 91 HBV-ACLF patients (HBV-ACLF group), who admitted to Wuxi No.5 People′s Hospital from July 1, 2019 to December 30, 2021, were included in this study. Fifty chronic hepatitis B (CHB) patients (CHB group) and 50 chronic HBV carriers (HBV carrier group) were enrolled as controls. Baseline clinical data such as qHBcAb, blood routine examination biochemical, and coagulation indices, HBsAg, hepatitis B e antigen (HBeAg), HBV DNA levels were recorded and analyzed retrospectively. The HBV-ACLF, HBsAg and HBV-DNA data were converted logarithmically. Patients were followed-up for 90 days. Cox regression was used to analyze the correlation between HBV-ACLF and survival outcome; survival rate was estimated by the Kaplan-Meier method; receiver operating characteristic (ROC) curve was used to evaluate the predictive value of baseline qHBcAb for the prognosis in patients with HBV-ACLF.Results:The baseline qHBcAb level in HBV-ACLF patients was (4.83±0.42) IU/ml, which was significantly higher than that in the CHB group [(4.59±0.54) IU/ml] and chronic HBV carrier group [(3.86±0.74) IU/ml] (all P<0.05). At the end of 90 days follow-up, 46 patients (50.55%) survived, and 45 patients (49.45%) died in the HBV-ACLF group. The baseline qHBcAb level was significantly higher in the survival group [(4.93±0.22) IU/ml] than in the death group [(4.70±0.52) IU/ml, P<0.01]. Significant differences were also found in the alpha fetoprotein, international normalized ratio, prothrombin activity, antithrombin Ⅲ activity, platelet, end-stage liver disease model score and hepatic encephalopathy complication between the two groups ( P<0.05). Cox regression analysis showed that the baseline qHBcAb was an independent risk factor affecting the 90-day survival of HBV-ACLF patients [hazard ratio=0.027,95% confidence interval ( CI) 0.001-0.696, P<0.05]. The area under the ROC curve of baseline qHBcAb level for predicting the 90-day survival outcome of HBV-ACLF patients was 0.639 (95% CI 0.525-0.752, P<0.05), with a cut-off value of 4.89 IU/ml. The cumulative survival rate of patients with baseline qHBcAb≥4.89 IU/ml was higher than that of patients with baseline qHBcAb<4.89 IU/ml ( P<0.05). Conclusions:Higher baseline qHBcAb level is associated with favorable outcome of HBV-ACLF patients and baseline qHBcAb may be used as a new biomarker to predict the clinical outcome of HBV-ACLF patients. HBV-ACLF patients with serum qHBcAb lower than 4.89 IU/ml face increased risk of short-term death.

Objective:To assess the predictors of outcomes for different subtypes of liver failure, and the effectiveness of artificial liver support systems in the treatment of liver failure.Methods:The clinical data of 112 patients with hepatitis B virus (HBV)- and non-HBV-related liver failure admitted to the intensive care unit (ICU) of the Fifth People's Hospital of Wuxi were collected from January to December 2020. The relevant etiologies of acute, subacute, acute-on-chronic, subacute-on-chronic, chronic subtype liver failure were analyzed. The efficacies of artificial liver support systems in the treatment of various subtypes of liver failure were also compared. The correlation of various indicators was analyzed by Spearman correlation analysis, the risk factors affecting the prognosis of patients with liver failure were analyzed by multivariate Logistic regression equation, and receiver operator characteristic curve (ROC curve) of subjects was plotted to evaluate the predictive value of each risk factor for the prognosis of patients with liver failure.Results:Among the 112 liver failure patients, 63 were caused by hepatitis B and 49 were caused by non-hepatitis B. The liver failure caused by hepatitis B was 6 times higher than for men than for women, which was higher than that of non-HBV liver failure group (1.33 times). Antithrombin Ⅲ (AT Ⅲ) and total bilirubin (TBil) levels of subacute liver failure were higher than those of pre-liver failure in the HBV liver failure group [AT Ⅲ: (59.33±14.57)% vs. (35.66±20.72)%, TBil (μmol/L): 399.21±112.94 vs. 206.08±126.96, both P < 0.05]. The levels of AT Ⅲ in patients with pre-liver failure and chronic liver failure in the non-HBV liver failure group were significantly higher than those with acute liver failure [(58.33±15.28%), (44.00±19.10)% vs. (31.33±7.57)%, both P < 0.05], patients with acute liver failure had significantly lower level of TBil than pre-liver failure (μmol/L: 107.83±49.73 vs. 286.20±128.92, P < 0.05), the TBil levels in patients with subacute and acute-on-chronic liver failure were also significantly higher than that in pre-liver failure group (μmol/L: 417.27±118.60, 373.00±187.00 vs. 286.20±128.92, both P < 0.05). Patients with subacute liver failure, subacute-on-chronic liver failure and chronic liver failure in the non-HBV failure group were significantly longer than those in acute liver failure (days: 36.00±8.31, 27.52±11.71, 27.72±22.71 vs. 11.00±1.41, all P < 0.05). There was no statistically significant difference in the case fatality rate of using the artificial liver support system between the HBV failure group and the non-HBV failure group (55.6% vs. 50.0%, P < 0.05), the levels of AT Ⅲ in the two groups of surviving patients were significantly higher than that of the dead [HBV liver failure group: (36.20±6.26)% vs. (27.33±8.87)%, non-HBV liver failure group: (41.06±4.16)% vs. (28.71±12.35)%, both P < 0.01]. Correlation analysis showed that there was a clear positive correlation between AT Ⅲ and TBil in the dead patients of HBV liver failure group and the survival and death patients of non-HBV liver failure group ( r values were 0.069, 0.341, 0.064, and P values were 0.723, 1.196 and 0.761, respectively); there was a significant inverse correlation between AT Ⅲ and TBil in the HBV liver failure group ( r = -0.105, P = 0.745). Multivariate Logistic regression analysis showed that AT Ⅲ was an independent risk factor affecting the prognosis of patients with non-HBV liver failure [odd ratio ( OR) = 1.023, 95% confidence interval (95% CI) was -0.001 to 0.001, P = 0.007]. TBil was an independent risk factor affecting prognosis of patients with HBV liver failure ( OR = 1.005, 95% CI was -0.002 to -7.543, P = 0.033). The analysis of ROC curve showed that AT Ⅲ had a predictive value for the prognosis of patients with non-HBV liver failure, the area under the ROC curve (AUC) = 0.747, the 95% CI was 0.592-0.902, P = 0.009. When the optimal truncation value was 39.5%, its sensitivity and specificity were 83.33% and 56.25%, respectively. Conclusions:Artificial liver support system treatment of liver failure was difficult to effectively reduce the mortality of patients with end-stage liver failure. In addition to AT Ⅲ, TBil also could be used as an indicator to assess liver compensatency and predict prognosis in liver failure patients.