Dynamic changes of prognostic scores and related clinical indicators in hepatitis B virus-related acute-on-chronic liver failure patients without underlying liver cirrhosis and their relationship with clinical outcomes
- VernacularTitle:非肝硬化背景的HBV相关慢加急性肝衰竭预后评分和相关临床指标的动态变化及其与临床结局的关系
- Author:
Wenling WANG
1
;
Manman XU
1
;
Yu WU
1
;
Jiateng ZHANG
1
;
Huaibin ZOU
1
;
Yu CHEN
1
Author Information
- Publication Type:Journal Article
- Keywords: Hepatitis B Virus; Acute-On-Chronic Liver Failure; Liver Cirrhosis; Prognosis
- From: Journal of Clinical Hepatology 2025;41(9):1771-1778
- CountryChina
- Language:Chinese
- Abstract: ObjectiveTo investigate the dynamic trajectories of prognostic scores and key clinical indicators in hepatitis B virus-related acute-on-chronic liver failure （HBV-ACLF） patients without liver cirrhosis， to clarify their association with outcomes， and to provide new evidence for individualized prognostic assessment. MethodsA prospective study was conducted for the data of 154 non-cirrhotic HBV-ACLF patients who attended Beijing YouAn Hospital of Capital Medical University from January 2016 to December 2023， including prognostic scores and key biochemical indicators on Days 3， 7， 14， 21， and 28 of the disease course. According to the outcome of patients at 1 year， they were divided into death/liver transplantation group with 43 patients， liver cirrhosis group with 23 patients， and non-liver cirrhosis group with 88 patients， and the trajectory heterogeneity of different outcome subgroups was analyzed. A one-way analysis of variance was used for comparison of normally distributed continuous data among the three groups； the Kruskal-Wallis H test was used for comparison of non-normally distributed continuous data among the three groups； the Wilcoxon test was used between two groups. the chi-square test was used for comparison of categorical data between groups. The mean and its 95% confidence interval （CI） were calculated for each indicator at difference time points； the linear interpolation method was used to connect the means at adjacent time points and construct the group-specific longitudinal trend curve； the 95%CI was visualized using the semi-transparent ribbon area， with the transparency parameter （α=0.2） optimized to enhance the visual discrimination of overlapping intervals across multiple groups. A linear mixed-effects model was used to compare the longitudinal changing trend of each indicator between the patients with different outcomes； likelihood ratio was used to evaluate the significance of the interaction effect between time and group， and in case of the significant interaction effect， the slope based on the estimated marginal mean was used for comparison between two groups. ResultsThere were significant differences between the three groups in the incidence rates of ascites and grade Ⅲ — Ⅳ hepatic encephalopathy， MELD score， MELD-Na score， CLIF-C ACLF score， COSSH-ACLF Ⅱ score， total bilirubin （TBil）， international normalized ratio （INR）， alpha-fetoprotein， blood sodium， alanine aminotransferase， and procalcitonin at the baseline（all P0.05）. The analysis of dynamic trajectories showed that the death/liver transplantation group had high levels of prognostic scores and the biochemical parameters of TBil and INR （TBil400 μmol/L， INR2.5）， as well as a low level of platelet count （PLT） （100×10⁹/L）. The non-liver cirrhosis group had rapid improvements in indicators， with TBil200 μmol/L， INR1.5， and PLT100×10⁹/L by day 28， while the liver cirrhosis group showed a trend of recovery， with TBil200 μmol/L， INR2.0， and PLT 100×10⁹/L on day 28， with significant global heterogeneity in the temporal trends of the above indicators across the three groups （all P0.01）. ConclusionDynamic monitoring of prognostic scores and key clinical indicators can effectively stratify the 1-year outcomes of non-cirrhotic patients with HBV-ACLF. Patients with poor prognosis were typically characterized by INR 2.5 and TBil 400 μmol/L. Among those who survived beyond 1 year， individuals who subsequently progressed to cirrhosis were frequently identified by the presence of INR 1.5， TBil 200 μmol/L， and PLT 100×10⁹/L at day 28.
