Objective To establish a new model of indocyanine green (ICG) clearance test combined with total bilirubin actual resident rate (TBARR) for predicting the short-term prognosis of patients with hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF) treated with artificial liver support system (ALSS) therapy. Methods A retrospective analysis was performed for the clinical data of 136 patients with HBV-ACLF who underwent ALSS therapy in Department of Infectious Diseases, The Affiliated Hospital of Southwest Medical University, from June 2017 to July 2021, and according to the prognosis at 3-month follow-up, they were divided into survival group with 92 patients and death group with 44 patients. Related indicators were measured at the time of the confirmed diagnosis of ACLF, including biochemical parameters, coagulation, indocyanine green retention rate at 15 minutes (ICGR15), and effective hepatic blood flow (EHBF), and related indices were calculated, including Model for End-Stage Liver Disease (MELD) score, MELD difference (ΔMELD), Child-Turcotte-Pugh (CTP) score, total bilirubin clearance rate (TBCR), total bilirubin rebound rate (TBRR), and TBARR. The Mann-Whitney U test was used for comparison of continuous data with skewed distribution between two groups; the chi-square test was used for comparison of categorical data between groups. A binary logistic regression analysis was used to establish a combined predictive model for the prognosis of HBV-ACLF after ALSS therapy. The area under the ROC curve (AUC) was used to compare the accuracy of various models in judging the short-term prognosis of patients with HBV-ACLF after ALSS therapy, and the Z test was used for comparison of AUC. Results There were significant differences between the death group and the survival group in MELD score, ΔMELD, CTP score, ICGR15, EHBF, TBRR, TBARR, neutrophil count, percentage of neutrophils, lymphocyte count, platelet count, alkaline phosphatase, gamma-glutamyl transpeptidase, total bilirubin, albumin, prothrombin time, international normalized ratio, prothrombin time activity, prealbumin, fibrinogen, serum sodium, age, and the incidence rate of hepatic encephalopathy (all P < 0.05). The multivariate logistic regression analysis showed that age (odds ratio [OR]=1.096, 95% confidence interval [CI]: 1.056-1.137, P < 0.001), neutrophil count (OR=1.214, 95%CI: 1.044-1.411, P=0.012), TBRR (OR=0.989, 95%CI: 0.982-0.996, P=0.001), TBARR (OR=1.073, 95%CI: 1.049-1.098, P < 0.001), ΔMELD (OR=1.480, 95%CI: 1.288-1.701, P < 0.001), CTP score (OR=2.081, 95%CI: 1.585-2.732, P < 0.001), and ICGR15(OR=1.116, 95%CI: 1.067-1.168, P < 0.001) were independent influencing factors for short-term mortality in patients with HBV-ACLF after ALSS therapy. The binary logistic regression analysis was used to establish four combined predictive models for predicting the prognosis of HBV-ACLF after ALSS therapy, i.e., TBRR-ICGR15, TBARR-ICGR15, TBARR-ICGR15-ΔMELD, and TBARR-ICGR15-ΔMELD-age, with an AUC of 0.830, 0.867, 0.900, and 0.917, respectively, and the combined predictive models had a larger AUC than each index alone (age, neutrophil count, TBRR, TBARR, ΔMELD, MELD score, CTP score, and ICGR15), among which the TBARR-ICGR15-ΔMELD-age model had the largest AUC. The combined models TBARR-ICGR15-ΔMELD and TBARR-ICGR15-ΔMELD-age had sensitivities and specificities of > 80%. Conclusion The combined predictive model established by ICGR15 and TBARR has a good value for in predicting the short-term prognosis of patients with HBV-ACLF after ALSS therapy, and the combined predictive model has a better accuracy than the single model in judging prognosis. [ABSTRACT FROM AUTHOR]