Objective:To explore the predictive value of changes in prealbumin for the prognosis of patients with hepatitis B virus-associated acute-on-chronic liver failure (HBV-ACLF) after artificial liver treatment.Methods:From January 1, 2018 to December 31, 2021, the clinical data (including prealbumin, platelet count, lymphocyte count, alanine transaminase (ALT), etc.) of 87 patients with HBV-ACLF who received artificial liver treatment at the Department of Gastroenterology of the General Hospital of Western Theater Command PLA were retrospectively collected. The 90-day survival status of all the patients was followed up, and the patients were divided into the survival group and the mortality group according to the survival status. The clinical characteristics and the changes of prealbumin on day 1 to 3, day 3 to 7, and day 1 to 7 after artificial liver treatment were compared between the 2 groups. Multivariate logistic regression analysis was used to analyze the independent influencing factors of the 90-day prognosis of HBV-ACLF patients after artificial liver treatment, and the nomogram prediction model was established and the receiver operating characteristic curve (ROC) was drawn to assess the area under the curve (AUC). Hosmer-Lemeshow goodness-of-fit test, calibration curve and clinical decision curve were performed to evaluate the goodness of fit, consistency and clinical value of the prediction model. Paired t-test and Mann-Whitney U test were used for statistical analysis. Results:There were 69 cases enrolled into the survival group, and 18 cases enrolled into the mortality group. The levels of albumin, prealbumin, platelet count, lymphocyte count, and ALT before treatment, and the level of prealbumin at the 3rd day after treatment of the survival group were all higher than those of the mortality group (32.5 (30.6, 35.2) g/L vs. 29.4 (27.6, 32.3) g/L, 66.0 (52.5, 81.5) mg/L vs. 56.5 (39.2, 65.0) mg/L, 103.0 (72.5, 145.0)×10 9/L vs. 63.5 (40.0, 92.5)×10 9/L, 1.1 (0.8, 1.4)×10 9/L vs. 0.9 (0.5, 1.1)×10 9/L, (514.7±86.4) U/L vs. (328.2±93.4) U/L, 90.0 (69.5, 102.5) mg/L vs.68.5(60.0, 75.8) mg/L), and the age, the level of total bilirubin, international normalized ratio, and prothrombin time before treatment of the survival group were all lower than those of the mortality group (48.0 (42.0, 57.0) years old vs. 48.5 (47.0, 56.0) years old, 323.9 (261.2, 409.2) μmol/L vs. 452.2 (405.8, 510.8) μmol/L, 1.5 (1.3, 1.9) vs. 1.9 (1.4, 2.1), 17.3 (14.6, 20.8) s vs. 21.4 (16.6, 23.2) s), and the differences were statistically significant ( Z=-3.38, -2.87, -2.38 and -2.01, t=2.39, Z=-4.11, 3.00, 3.64, 2.18 and 2.37; all P<0.05). The change of prealbumin on day 1 to 3 after treatment in the mortality group was greater than that in the survival group (-0.182 (-0.321, -0.026) vs. -0.043 (-0.133, 0.093)), and the difference was statistically significant ( Z=-3.42, P=0.001). The results of multivariate logistic regression analysis showed that the age, total bilirubin before treatment, and the change of prealbumin on day 1 to 3 after treatment were independent influencing factors for the 90-day prognosis in HBV-ACLF patients after artificial liver treatment (all P<0.05), and the nomogram model was established based on the above 3 factors. The results of ROC analysis showed that the AUC of the prediction model was 0.933 (95% confidence interval: 0.866 to 1.000, P<0.001), with a sensitivity of 0.933 and a specificity of 0.825. The results of the Hosmer-Lemeshow goodness-of-fit test showed that the prediction model had a good fit( P=0.700). The results of calibration curve analysis indicated that the actual curve of the prediction model was close to the calibration curve, with an average absolute error of 0.034, the consistency between the predicted probability and the actual probability was good. The clinical decision curve analysis suggested that the prediction model had significant clinical benefits. Conclusions:The changes of prealbumin after artificial liver treatment in HBV-ACLF patients can reflect the recovery of liver function. The nomogram prediction model based on the change of prealbumin on day 1 to 3 after treatment, age, and total bilirubin before treatment can better predict the 90-day prognosis of HBV-ACLF patients after artificial liver treatment.

Objective:To explore the predictive value of changes in prealbumin for the prognosis of patients with hepatitis B virus-associated acute-on-chronic liver failure (HBV-ACLF) after artificial liver treatment.Methods:From January 1, 2018 to December 31, 2021, the clinical data (including prealbumin, platelet count, lymphocyte count, alanine transaminase (ALT), etc.) of 87 patients with HBV-ACLF who received artificial liver treatment at the Department of Gastroenterology of the General Hospital of Western Theater Command PLA were retrospectively collected. The 90-day survival status of all the patients was followed up, and the patients were divided into the survival group and the mortality group according to the survival status. The clinical characteristics and the changes of prealbumin on day 1 to 3, day 3 to 7, and day 1 to 7 after artificial liver treatment were compared between the 2 groups. Multivariate logistic regression analysis was used to analyze the independent influencing factors of the 90-day prognosis of HBV-ACLF patients after artificial liver treatment, and the nomogram prediction model was established and the receiver operating characteristic curve (ROC) was drawn to assess the area under the curve (AUC). Hosmer-Lemeshow goodness-of-fit test, calibration curve and clinical decision curve were performed to evaluate the goodness of fit, consistency and clinical value of the prediction model. Paired t-test and Mann-Whitney U test were used for statistical analysis. Results:There were 69 cases enrolled into the survival group, and 18 cases enrolled into the mortality group. The levels of albumin, prealbumin, platelet count, lymphocyte count, and ALT before treatment, and the level of prealbumin at the 3rd day after treatment of the survival group were all higher than those of the mortality group (32.5 (30.6, 35.2) g/L vs. 29.4 (27.6, 32.3) g/L, 66.0 (52.5, 81.5) mg/L vs. 56.5 (39.2, 65.0) mg/L, 103.0 (72.5, 145.0)×10 9/L vs. 63.5 (40.0, 92.5)×10 9/L, 1.1 (0.8, 1.4)×10 9/L vs. 0.9 (0.5, 1.1)×10 9/L, (514.7±86.4) U/L vs. (328.2±93.4) U/L, 90.0 (69.5, 102.5) mg/L vs.68.5(60.0, 75.8) mg/L), and the age, the level of total bilirubin, international normalized ratio, and prothrombin time before treatment of the survival group were all lower than those of the mortality group (48.0 (42.0, 57.0) years old vs. 48.5 (47.0, 56.0) years old, 323.9 (261.2, 409.2) μmol/L vs. 452.2 (405.8, 510.8) μmol/L, 1.5 (1.3, 1.9) vs. 1.9 (1.4, 2.1), 17.3 (14.6, 20.8) s vs. 21.4 (16.6, 23.2) s), and the differences were statistically significant ( Z=-3.38, -2.87, -2.38 and -2.01, t=2.39, Z=-4.11, 3.00, 3.64, 2.18 and 2.37; all P<0.05). The change of prealbumin on day 1 to 3 after treatment in the mortality group was greater than that in the survival group (-0.182 (-0.321, -0.026) vs. -0.043 (-0.133, 0.093)), and the difference was statistically significant ( Z=-3.42, P=0.001). The results of multivariate logistic regression analysis showed that the age, total bilirubin before treatment, and the change of prealbumin on day 1 to 3 after treatment were independent influencing factors for the 90-day prognosis in HBV-ACLF patients after artificial liver treatment (all P<0.05), and the nomogram model was established based on the above 3 factors. The results of ROC analysis showed that the AUC of the prediction model was 0.933 (95% confidence interval: 0.866 to 1.000, P<0.001), with a sensitivity of 0.933 and a specificity of 0.825. The results of the Hosmer-Lemeshow goodness-of-fit test showed that the prediction model had a good fit( P=0.700). The results of calibration curve analysis indicated that the actual curve of the prediction model was close to the calibration curve, with an average absolute error of 0.034, the consistency between the predicted probability and the actual probability was good. The clinical decision curve analysis suggested that the prediction model had significant clinical benefits. Conclusions:The changes of prealbumin after artificial liver treatment in HBV-ACLF patients can reflect the recovery of liver function. The nomogram prediction model based on the change of prealbumin on day 1 to 3 after treatment, age, and total bilirubin before treatment can better predict the 90-day prognosis of HBV-ACLF patients after artificial liver treatment.