Predictive value of the Chinese group on the study of severe hepatitis B-acute-on-chronic liver failure score in the short-term prognosis of patients with hepatitis B virus-related acute-on-chronic liver failure

• Main Authors: Jing-Jing Tong, Wei Zhao, Xiu-Ying Mu, Xiang Xu, Hai-Bin Su, Xiao-Yan Liu, Jing Chen, Xing-Ran Zhai, Yu Wang, Jin-Hua Hu, Ning-Ning Wang
• Format: Article
• Language: English
• Published: Wolters Kluwer 2019-07-01
• Series: Chinese Medical Journal
• Online Access: http://journals.lww.com/10.1097/CM9.0000000000000298
• ISSN: 0366-6999; 2542-5641

Abstract. Background:. As a large, prospective, multicenter study-based prognostic score for hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF), the Chinese group on the study of severe hepatitis B-acute-on-chronic liver failure score (COSSH-ACLFs), has been approved by some foreign scholars; however, its predictive value needs to be verified. This study investigated the predictive value of COSSH-ACLFs for short-term prognosis in Chinese patients with HBV-ACLF. Methods:. This retrospective cohort study included 751 patients with HBV-ACLF admitted to the Fifth Medical Center of Chinese PLA General Hospital between January 2011 and December 2014. Spearman method was used to assess the correlation of COSSH-ACLFs with classical scores. Different COX multivariate regression models were used to confirm the relationship between COSSH-ACLFs and short-term prognosis in patients with HBV-ACLF, and stratified analysis was used to further verify the stability of this relationship. We compared the predictive powers of COSSH-ACLFs and other classical scores using area under the receiver operating characteristic curve (AUROC) and Z-test. Results:. A total of 975 patients with HBV-ACLF were screened, and 751 were analyzed (623 male and 128 female). COSSH-ACLFs was the highest in patients with end-stage ACLF, followed by those with middle- and early-stage ACLF (H = 211.8, P < 0.001). In the fully adjusted model, COX multivariate regression analysis revealed that COSSH-ACLFs (as a continuous variable) was independently and positively correlated with mortality risk in patients with HBV-ACLF at 28 days (hazard ratio [HR]: 1.37 [1.22, 1.53], P < 0.001) and 90 days (HR: 1.43 [1.29, 1.58], P < 0.001). The same trend could be observed in the crude model and minimally adjusted model. The AUROCs of COSSH-ACLFs for 28-day and 90-day prognoses in patients with HBV-ACLF were 0.807 and 0.792, respectively, indicating a stronger predictive accuracy than those of classic models. Conclusions:. COSSH-ACLFs, with a superior predictive accuracy compared with other classical scores, can strongly predict short-term prognosis in Chinese patients with HBV-ACLF.