BPG is committed to discovery and dissemination of knowledge
Retrospective Study Open Access
Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
World J Hepatol. Jun 27, 2026; 18(6): 121259
Published online Jun 27, 2026. doi: 10.4254/wjh.121259
Liver stiffness-based composite indices for compensated and decompensated cirrhosis in chronic hepatitis B
Zhi-Zun Guo, Yan Wang, Xiao-Xuan Liu, Shan-Shan Yang, Fan Yu, Department of Infectious Diseases, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Third Hospital of Shanxi Medical University, Tongji Shanxi Hospital, Taiyuan 030032, Shanxi Province, China
ORCID number: Zhi-Zun Guo (0000-0003-1477-1191).
Author contributions: Guo ZZ participated in the conception and design of the study, wrote the manuscript, and was involved in the acquisition, analysis, and interpretation of data; Wang Y and Liu XX performed the statistical analysis; Yang SS and Yu F collected the data. All authors critically reviewed and provided final approval of the manuscript; and all authors were responsible for the decision to submit the manuscript for publication.
Institutional review board statement: The study was reviewed and approved by the Shanxi Bethune Hospital Institutional Review Board, Approval No. YXLL-2025-071.
Informed consent statement: The need for patient consent was waived due to the retrospective nature of the study.
Conflict-of-interest statement: There was no conflict of interest to be reported.
Data sharing statement: The anonymized dataset can be made available from the corresponding author at guozhizun@sxbqeh.com.cn.
Corresponding author: Zhi-Zun Guo, Department of Infectious Diseases, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Third Hospital of Shanxi Medical University, Tongji Shanxi Hospital, No. 99 Longcheng Street, Taiyuan 030032, Shanxi Province, China. guozhizun@sxbqeh.com.cn
Received: March 21, 2026
Revised: April 9, 2026
Accepted: April 24, 2026
Published online: June 27, 2026
Processing time: 99 Days and 2.2 Hours

Abstract
BACKGROUND

Chronic hepatitis B virus (HBV) infection remains a major global public health challenge. Accurate assessment of disease progression is essential for managing patients with chronic HBV infection. Despite the use of liver biopsy, imaging tests, and non-invasive models, their inherent limitations restrict widespread application. Thus, there is a pressing need to establish an effective non-invasive diagnostic model for assessing disease progression in patients with chronic HBV infection, particularly one that integrates imaging tests with demographic and hematologic parameters.

AIM

To develop novel non-invasive composite indices for evaluating the condition of chronic HBV-infected patients.

METHODS

This retrospective study enrolled 132 chronic HBV-infected patients who were admitted to the Department of Infectious Diseases, Shanxi Bethune Hospital, Taiyuan, China, between August 1, 2020 and June 30, 2024. Demographic variables, hematological parameters, and liver stiffness measurement (LSM) were recorded. Multivariable logistic regression was constructed to identify independent predictors of chronic hepatitis B (CHB), compensated and decompensated hepatitis B cirrhosis. The logistic regression diagnostic model was fitted with the selected predictors. Receiver operating characteristic (ROC) curves were generated to assess diagnostic performance of the model.

RESULTS

We developed three non-invasive models-A-index for CHB, MAPTAL for compensated hepatitis B cirrhosis, and APTAL for decompensated hepatitis B cirrhosis-which achieved area under the ROC curve of 0.948, 0.918, and 0.968, respectively.

CONCLUSION

The MAPTAL and APTAL indices, integrating LSM with routine demographic and hematologic variables, reliably predict progression to compensated and decompensated HBV-related cirrhosis, respectively.

Key Words: Liver cirrhosis; Hepatitis B virus; Diagnosis; Non-invasive; Transient elastography; Liver stiffness

Core Tip: This is a retrospective study aimed at developing novel non-invasive composite indices to evaluate the condition of chronic hepatitis B virus (HBV)-infected patients. Three models, named the A-index, MAPTAL index, and APTAL index, were constructed to predict disease progression stages. Importantly, the MAPTAL and APTAL indices, which incorporate readily available demographic, hematologic, and liver stiffness measurement parameters, can reliably predict compensated and decompensated HBV-related cirrhosis, respectively.



INTRODUCTION

Chronic hepatitis B virus (HBV) infection remains a major global public health challenge. The Polaris International Epidemiology Collaboration estimated that approximately 86 million individuals in China were chronically infected with HBV in 2016[1]. Furthermore, the World Health Organization reported that in 2019, an estimated 296 million individuals were chronically infected with HBV worldwide, leading to approximately 820000 deaths from related complications such as cirrhosis, liver failure, and hepatocellular carcinoma (HCC)[2].

The immune response elicited by HBV results in inflammation and necrosis of hepatocytes, while the persistence or recurrent occurrence of inflammatory necrosis facilitates the progression of chronic HBV infection towards liver fibrosis, cirrhosis, and potentially HCC[2]. Although liver biopsy is considered the gold standard for assessing the severity of liver inflammation, necrosis, and fibrosis, its clinical application is restricted due to several drawbacks, including sampling errors, complications, inter-observer variability, and substantial medical expenses[3].

Currently, there exists a notable deficiency in specific hematological diagnostic indicators for liver fibrosis, indicating that combined detection may enhance diagnostic efficacy[4]. Although non-invasive diagnostic models, such as the Aspartate-to-Platelet Ratio Index (APRI), Fibrosis-4 (Fib-4) index, albumin-bilirubin (ALBI) grade, and S index, have been proposed to evaluate the severity of liver fibrosis, each model possesses inherent limitations, rendering them unsuitable for all chronic HBV-infected patients. Furthermore, not all models incorporate demographic indicators, such as age and gender, thus failing to reflect the impact of demographic factors on disease progression. The APRI and Fib-4 index were developed using data from patients with chronic hepatitis C virus (HCV) infection to evaluate the extent of liver fibrosis[2]. Both models are characterized by their simplicity and practicality, however, the APRI exhibits reduced accuracy in evaluating HBV-related liver fibrosis, and the Fib-4 index inadequately reflects the reversal of liver fibrosis following antiviral therapy in chronic hepatitis B (CHB)[2,5,6]. The ALBI grade was formulated to assess liver function disorders in patients with HCC and has been evaluated in chronic liver disease (CLD) without HCC[7]. Wang et al[8] reported that the ALBI score could effectively predict the severity and long-term prognosis of CHB-related cirrhosis. Subsequently, Fujita et al[9] and Fujita et al[10] demonstrated that the ALBI score was capable of indicating liver fibrosis staging among Japanese patients with CHB or chronic hepatitis C (CHC), with diagnostic accuracy comparable to that of the APRI and Fib-4 index for distinguishing cirrhosis from non-cirrhotic stages. However, the ALBI model does not incorporate demographic variables, and its computation requires logarithmic transformation, limiting its clinical convenience. The S index was developed by Zhou et al[11] to predict significant liver fibrosis and cirrhosis in antiviral treatment-naive patients with chronic HBV infection. Its applicability in antiviral treatment-experienced patients requires further validation.

Transient elastography (TE) and magnetic resonance elastography (MRE) are validated non-invasive tools for staging liver fibrosis. Owing to higher costs and limited availability of specialized personnel, MRE is not routinely adopted in clinical settings[2]. Coincidentally, liver stiffness measurements (LSM) are influenced by various factors, including liver inflammation, necrosis, cholestasis, and severe steatosis, necessitating the interpretation of results in conjunction with serum alanine aminotransferase (ALT) and bilirubin levels[2]. The Chinese Expert Consensus on Clinical Application of TE Detecting Liver Fibrosis (2018 Update) therefore recommends integrating hematological parameters when TE yields indeterminate results[12]. The LSM-to-Platelet Ratio Index (LPRI) combines TE with hematological parameters to evaluate the degree of non-alcoholic fatty liver disease (NAFLD) as well as HBV/HCV-related liver fibrosis and cirrhosis[13]. It is equally critical to ascertain whether the LPRI is applicable to antiviral-experienced CHB patients. Consequently, establishing a comprehensive, non-invasive, and stage-specific diagnostic model that combines LSM with readily available clinical parameters remains urgently needed.

This study aims to develop novel non-invasive composite indices that integrate LSM with routine demographic and hematologic variables to specifically predict progression to CHB, compensated and decompensated cirrhosis in chronic HBV-infected patients.

MATERIALS AND METHODS
Study population

In this study, we retrospectively enrolled patients with chronic HBV infection who were admitted to the Department of Infectious Diseases, Shanxi Bethune Hospital, Taiyuan, China, between August 1, 2020 and June 30, 2024. Patients were included if they had chronic HBV infection diagnosed according to the Chinese Guidelines for the Prevention and Treatment of CHB (2022 edition)[2], including individuals with chronic HBV infection and normal ALT levels, those with CHB, and patients with compensated or decompensated hepatitis B-related cirrhosis. Exclusion criteria were as follows: (1) Co-infection with hepatitis A, C, D, or E virus, Epstein–Barr virus, cytomegalovirus, or other hepatotropic viruses; (2) Presence of alcoholic liver disease, metabolic dysfunction-associated fatty liver disease, autoimmune liver disease, inherited metabolic liver disorders, or drug-induced liver injury; (3) History of liver failure, HCC, or liver transplantation; (4) Concurrent malignancies or end-stage diseases requiring immunosuppressive or targeted molecular therapies; and (5) Pregnant women or individuals under the age of 18.

Ethical approval for the study was granted by the Ethics Committee of Shanxi Bethune Hospital (Ethics Approval Notice No. YXLL-2025-071).

Data collection

Demographic variables [sex, age, height, weight, body mass index (BMI)] were recorded for eligible patients. Before treatment initiation, the following hematological parameters were documented: ALT, aspartate aminotransferase (AST), alkaline phosphatase (ALP), γ-glutamyl transpeptidase (GGT), total bile acid (TBA), total bilirubin (T-Bil), direct bilirubin (D-Bil), indirect bilirubin (I-Bil), total cholesterol (TC), triglycerides (TG), albumin (ALB), platelet count (PLT), prothrombin time (PT) as well as prothrombin activity (PTA), and LSM together with controlled attenuation parameter by TE. All TE examinations were performed by certified operators with experience of ≥ 500 prior studies, following a standardized protocol that included: Fasting for ≥ 3 hours, appropriate probe selection, intercostal approach, acquisition of ≥ 10 valid measurements, and an inter-quartile range (IQR)/median ratio of ≤ 30%.

The data were accessed for research purposes from March 19, 2025 to March 30, 2025. The collected data were fully anonymized to safeguard the sensitive information of the patients. The dataset was complete for all included variables; no missing values were present, and therefore no imputation was required.

Existing non-invasive diagnostic model formulas

ALBI = 0.66 × log10 [T-Bil (µmol/L)] - 0.085 × ALB (g/L). APRI = AST (IU/L)/AST upper limit of normal (IU/L) × 100/PLT (109/L). LPRI = LSM (KPa) × 100/PLT (109/L). Fib-4 index = age (years) × AST (IU/L)/[PLT (109/L) × ALT (IU/L)1/2]. S index = 1000 × GGT (IU/L)/[PLT (109/L) × ALB (g/L)2].

Statistical analysis

All statistical procedures were performed with IBM SPSS Statistics for Windows, version 27.0.1.0 (IBM Corp., Armonk, NY, United States). Data distribution was examined by the Kolmogorov-Smirnov test. Homogeneity of variances was evaluated with Levene’s test. Continuous variables were presented as mean ± SD or median (IQR) according to normality. One-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test was used to compare three or more independent groups meeting parametric assumptions, whereas the Kruskal-Wallis H test was employed for non-parametric data. Categorical variables were expressed as constituent ratios (%) and were analyzed with the χ2 test. Multivariable logistic regression was constructed to identify independent predictors of CHB, compensated hepatitis B cirrhosis, and decompensated hepatitis B cirrhosis, with results reported as odds ratio (OR) with 95% confidence interval (CI). The final models were derived through stepwise regression based on statistical significance, with results interpreted in the context of clinical relevance. Multicollinearity among candidate variables was evaluated prior to model construction by analyzing the stability of coefficient estimates and the presence of implausible standard errors. The logistic regression diagnostic model was fitted with the selected predictors. Diagnostic performance of the model was assessed by the area under the curve (AUC). Stability was evaluated by 1000 internal bootstrap resamples of the predicted probabilities. Two-tailed P value < 0.05 was considered statistically significant.

RESULTS
General information

In this study, a total of 132 patients met the inclusion and exclusion criteria, including 33 patients in the chronic HBV infection group, 49 patients in the CHB group, 27 patients in the compensated liver cirrhosis group, and 23 patients in the decompensated liver cirrhosis group. The general information of these patients, including age, height, weight, BMI, and gender composition ratios, is shown in Table 1.

Table 1 Baseline characteristics of chronic hepatitis B virus-infected patients by disease stage.
Indicator
Chronic HBV infection group (n = 33)
CHB group (n = 49)
Compensated liver cirrhosis group (n = 27)
Decompensated liver cirrhosis group (n = 23)
P value
Age (years)Median (IQR)46 (22.50)37 (19.50)52 (23.00)56 (8.00)< 0.001
Height (m)Median (IQR)1.67 (0.11)1.70 (0.14)1.70 (0.13)1.65 (0.14)0.054
Weight (kg)mean ± SD66.06 ± 10.8569.62 ± 14.7271.76 ± 14.8667.24 ± 14.210.438
BMI (kg/m2)mean ± SD24.12 ± 3.8824.39 ± 4.2424.86 ± 4.1024.72 ± 3.690.718
ALT (IU/L)Median (IQR)20.50 (12.00)126.30 (263.90)36.90 (31.70)26.80 (16.10)< 0.001
AST (IU/L)Median (IQR)22.50 (7.15)80.00 (99.15)35.30 (27.10)35.30 (37.30)< 0.001
ALP (IU/L)Median (IQR)76.50 (34.40)97.50 (53.50)89.00 (31.40)78.30 (57.40)0.009
GGT (IU/L)Median (IQR)18.20 (13.35)95.60 (124.75)35.40 (48.90)24.40 (58.70)< 0.001
TBA (µmol/L)Median (IQR)3.70 (5.60)5.50 (11.70)10.30 (22.20)12.30 (42.90)0.001
T-Bil (µmol/L)Median (IQR)12.60 (7.60)18.60 (13.65)20.60 (15.40)21.70 (20.00)< 0.001
D-Bil (µmol/L)Median (IQR)2.80 (1.45)4.40 (4.25)5.10 (3.30)5.10 (3.80)< 0.001
I-Bil (µmol/L)Median (IQR)9.70 (6.25)14.60 (10.20)15.30 (12.30)16.10 (16.50)0.001
TC (mmol/L)Median (IQR)4.39 (1.28)4.14 (1.41)3.74 (1.27)3.13 (1.44)0.002
TG (mmol/L)Median (IQR)0.93 (0.71)1.27 (0.99)0.93 (0.34)0.77 (0.50)0.001
ALB (g/L)Median (IQR)43.30 (5.75)41.90 (5.65)38.60 (8.00)34.30 (9.50)< 0.001
PLT (109/L)mean ± SD193.52 ± 61.26170.22 ± 61.06116.37 ± 55.5480.00 ± 48.25< 0.001
PT (seconds)Median (IQR)11.30 (1.00)11.60 (1.60)12.30 (2.50)13.20 (2.20)< 0.001
PTA (%)mean ± SD95.61 ± 10.2392.24 ± 16.0580.22 ± 15.7273.83 ± 15.24< 0.001
LSM (KPa)Median (IQR)5.00 (1.70)8.40 (8.15)13.30 (15.90)16.60 (19.00)< 0.001
CAP (dB/m)Median (IQR)216.00 (51.50)222.00 (76.50)222.00 (78.00)208.00 (52.00)0.466
GenderMale, n (%)16 (48.48)35 (71.43)20 (74.07)11 (47.83)0.045
Female, n (%)17 (51.52)14 (28.57)7 (25.93)12 (52.17)
Comparison of baseline characteristics among groups

Comparative analysis of observation indicators demonstrated that the differences in sex, age, ALT, AST, ALP, GGT, TBA, T-Bil, D-Bil, I-Bil, TC, TG, ALB, PLT, PT, PTA and LSM were statistically significant among the chronic HBV infection group, CHB group, compensated liver cirrhosis group, and decompensated liver cirrhosis group (all P value < 0.05, Table 1).

Independent predictors of clinical phase (chronic hepatitis, compensated cirrhosis, decompensated cirrhosis)

Multivariable logistic regression analysis revealed that among variables significantly different across the four groups, ALT was an independent predictor of CHB (OR = 1.106, 95%CI: 1.011-1.210). Independent predictors of compensated hepatitis B cirrhosis included male sex (OR = 0.065, 95%CI: 0.005-0.856), age (OR = 1.133, 95%CI: 1.033-1.243), ALT (OR = 1.102, 95%CI: 1.006-1.208), TC (OR = 6.129, 95%CI: 1.140-32.956), PLT (OR = 0.962, 95%CI: 0.934-0.990), and LSM (OR = 1.679, 95%CI: 1.171-2.407). Variables independently associated with decompensated hepatitis B cirrhosis were age (OR = 1.226, 95%CI: 1.075-1.398), TC (OR = 8.922, 95%CI: 1.471-54.111), ALB (OR = 0.833, 95%CI: 0.700-0.992), PLT (OR = 0.951, 95%CI: 0.918-0.985), and LSM (OR = 1.595, 95%CI: 1.113-2.287) (all P value < 0.05, Tables 2, 3 and 4).

Table 2 Logistic regression estimates for predicting chronic hepatitis B.
Variable
β
SE
χ2
P value
OR
95%CI
Low
Up
β031.517 32.736 0.927 0.336
Male0.131 1.340 0.009 0.922 1.140 0.082 15.754
Age0.004 0.051 0.006 0.939 1.004 0.909 1.109
ALT0.101 0.046 4.861 0.027 1.106 1.011 1.210
AST0.070 0.057 1.505 0.220 1.073 0.959 1.200
ALP0.000 0.022 0.000 0.997 1.000 0.957 1.045
GGT0.011 0.021 0.262 0.609 1.011 0.969 1.054
TBA-0.014 0.069 0.039 0.843 0.986 0.862 1.129
T-Bil0.120 0.110 1.184 0.277 1.127 0.908 1.399
TC1.514 0.863 3.080 0.079 4.546 0.838 24.662
TG0.071 0.676 0.011 0.917 1.073 0.286 4.035
ALB-0.045 0.100 0.208 0.648 0.956 0.786 1.161
PLT-0.019 0.014 1.751 0.186 0.981 0.954 1.009
PT-2.176 1.831 1.411 0.235 0.114 0.003 4.111
PTA-0.189 0.124 2.313 0.128 0.828 0.650 1.056
LSM0.223 0.183 1.483 0.223 1.250 0.873 1.790
Table 3 Logistic regression estimates for predicting compensated hepatitis B virus-related cirrhosis.
Variable
β
SE
χ2
P value
OR
95%CI
Low
Up
β023.846 32.121 0.551 0.458
Male-2.732 1.314 4.320 0.038 0.065 0.005 0.856
Age0.125 0.047 7.022 0.008 1.133 1.033 1.243
ALT0.098 0.047 4.329 0.037 1.102 1.006 1.208
AST0.072 0.058 1.506 0.220 1.074 0.958 1.204
ALP0.004 0.022 0.040 0.841 1.004 0.961 1.050
GGT-0.037 0.022 2.828 0.093 0.963 0.922 1.006
TBA-0.005 0.068 0.005 0.943 0.995 0.871 1.137
T-Bil0.084 0.110 0.587 0.444 1.088 0.877 1.349
TC1.813 0.858 4.463 0.035 6.129 1.140 32.956
TG-0.521 0.767 0.460 0.497 0.594 0.132 2.673
ALB0.009 0.101 0.008 0.928 1.009 0.828 1.230
PLT-0.039 0.015 7.066 0.008 0.962 0.934 0.990
PT-2.043 1.752 1.359 0.244 0.130 0.004 4.021
PTA-0.176 0.125 1.974 0.160 0.838 0.656 1.072
LSM0.518 0.184 7.953 0.005 1.679 1.171 2.407
Table 4 Logistic regression estimates for predicting decompensated hepatitis B virus-related cirrhosis.
Variable
β
SE
χ2
P value
OR
95%CI
Low
Up
β032.084 33.010 0.945 0.331
Male-1.826 1.404 1.690 0.194 0.161 0.010 2.525
Age0.204 0.067 9.180 0.002 1.226 1.075 1.398
ALT0.050 0.051 0.968 0.325 1.051 0.952 1.160
AST0.108 0.058 3.493 0.062 1.114 0.995 1.248
ALP-0.014 0.024 0.342 0.559 0.986 0.942 1.033
GGT-0.033 0.024 1.852 0.174 0.968 0.924 1.014
TBA-0.019 0.068 0.078 0.781 0.981 0.858 1.122
T-Bil0.170 0.113 2.259 0.133 1.185 0.950 1.479
TC2.188 0.920 5.662 0.017 8.922 1.471 54.111
TG-1.521 1.384 1.208 0.272 0.218 0.014 3.293
ALB-0.183 0.089 4.214 0.040 0.833 0.700 0.992
PLT-0.050 0.018 7.653 0.006 0.951 0.918 0.985
PT-2.217 1.778 1.555 0.212 0.109 0.003 3.552
PTA-0.200 0.139 2.065 0.151 0.818 0.623 1.076
LSM0.467 0.184 6.461 0.011 1.595 1.113 2.287
Discrimination and internal stability of the logistic regression diagnostic model

According to the results of multivariable regression analysis, three non-invasive diagnostic models were established, which were designated by acronyms representing their constituent variables as follows: (1) The CHB group, A-index = 0.101 × ALT; (2) The compensated liver cirrhosis group, MAPTAL index = -2.732 × male + 0.125 × age - 0.039 × PLT + 1.813 × TC + 0.098 × ALT + 0.518 × LSM; and (3) The decompensated liver cirrhosis group, APTAL index = 0.204 × age - 0.05 × PLT + 2.188 × TC - 0.183 × ALB + 0.467 × LSM.

Receiver operating characteristic analysis yielded an optimal A-index cut-off of 4.60 (Youden index), with sensitivity 0.86, specificity 1.00, and AUC 0.948 (95%CI: 0.900-0.997); this was higher than ALBI grade (AUC 0.656; Figure 1A). Additionally, MAPTAL (cut-off 12.76) achieved AUC 0.918 (95%CI: 0.853-0.983), outperforming APRI (AUC 0.895), Fib-4 (AUC 0.838), and S index (AUC 0.854) but slightly inferior to LPRI (AUC 0.942; Figure 1B). APTAL (cut-off 11.39) reached AUC 0.968 (95%CI: 0.930-1.007), surpassing all conventional scores (Figure 1C). Detailed results are presented in Table 5.

Figure 1
Figure 1 Receiver operating characteristic curves of the proposed models and established indices for diagnosing liver disease in chronic hepatitis B patients. A: The A-index and albumin-bilirubin grade for diagnosing chronic hepatitis B; B: The MAPTAL index, Aspartate Platelet Ratio Index (APRI), liver stiffness measurement to Platelet Ratio Index (LPRI), and Fibrosis-4 (Fib-4) for diagnosing compensated cirrhosis; C: The APTAL index, APRI, LPRI, and Fib-4 for diagnosing decompensated cirrhosis. A-index = 0.101 × alanine aminotransferase (ALT); MAPTAL index = -2.732 × male + 0.125 × age - 0.039 × platelet count (PLT) + 1.813 × total cholesterol (TC) + 0.098 × ALT + 0.518 × liver stiffness measurement (LSM); APTAL index = 0.204 × age - 0.05 × PLT + 2.188 × TC - 0.183 × albumin + 0.467 × LSM. ALBI grade: Albumin-bilirubin grade; APRI: Aspartate Platelet Ratio Index; LPRI: LSM to Platelet Ratio Index; Fib-4: Fibrosis-4; AUC: Area under the curve.
Table 5 Diagnostic performance of the proposed models and established indices for chronic hepatitis B, compensated cirrhosis, and decompensated cirrhosis.
Disease
Model
Cut-off value
Sensitivity
Specificity
AUC
95%CI
CHBA-index> 4.600.861.000.9480.900-0.997
ALBI grade> -2.850.690.610.6560.534-0.778
Compensated liver cirrhosisMAPTAL index> 12.760.81 0.88 0.918 0.853-0.983
APRI> 0.560.70 0.94 0.895 0.818-0.972
LPRI> 4.320.93 0.85 0.942 0.887-0.997
Fib-4 index> 2.310.70 0.88 0.838 0.735-0.941
S index> 0.110.78 0.82 0.854 0.758-0.949
Decompensated liver cirrhosisAPTAL index> 11.390.870.940.9680.930-1.007
APRI> 0.450.910.880.9290.855-1.002
LPRI> 4.850.960.880.949 0.873-1.024
Fib-4 index> 3.760.740.970.9380.880-0.996
S index> 0.110.830.820.8770.780-0.974

Bootstrap analysis (1000 resamples) showed wide bias-corrected and accelerated 95%CI for all models’ predicted probabilities (A-index 0.213-0.864, MAPTAL 0.024-0.294, APTAL 0.004-0.296), extending below the 0.70 threshold and indicating insufficient internal stability.

DISCUSSION

Accurate assessment of disease progression is essential for managing patients with chronic HBV infection. Despite the use of liver biopsy, imaging, and non-invasive models, their inherent limitations restrict widespread application. Therefore, a comprehensive, non-invasive, and stage-specific diagnostic model that integrates readily available demographic, hematologic, and imaging parameters is urgently needed.

We retrospectively enrolled 132 hospitalized patients with chronic HBV infection, regardless of prior antiviral therapy, and stratified them into four stages: Chronic HBV infection with normal ALT, CHB, compensated HBV-related cirrhosis, and decompensated HBV-related cirrhosis. Demographic characteristics, hematologic indices, and LSM by TE were compared across groups. Three non-invasive models were then constructed: The A-index for CHB, the MAPTAL index for compensated cirrhosis, and the APTAL index for decompensated cirrhosis. These novel models-based on readily available demographic, hematologic, and LSM parameters-can reliably predict the disease progression stages of chronic HBV infection and facilitate early recognition of cirrhotic decompensation.

Zheng and Tian[14] used Global Burden of Disease data to show that the incidence of cirrhosis in China rose steadily with age from 1990 to 2019, peaked at 40-50 years, and was significantly higher in men than in women. Previous work has demonstrated a strong positive correlation between age and histological fibrosis stage in patients with CHB, indicating that advanced age accelerates progression to cirrhosis and HCC. Except for primary biliary cholangitis, men have a higher incidence of advanced fibrosis and cirrhosis than women, and fibrosis in women is usually delayed until after menopause; cirrhosis is rarely seen in women of reproductive age. In the present study, age and sex differed significantly across groups: Median age increased stepwise from the CHB to the compensated-cirrhosis to the decompensated-cirrhosis group, supporting the concept of age-dependent disease progression. Logistic regression identified age as an independent risk factor for both compensated and decompensated cirrhosis. Male sex was inversely associated with compensated cirrhosis in this sample; however, this observation is likely attributable to limited sample size and selection bias. While sex differences in CHB outcomes have been reported in larger cohorts, potentially through sex-specific gene expression and hormonal pathways[15,16], our data do not permit definitive conclusions regarding biological mechanisms. This finding should be interpreted with caution pending validation in prospective studies.

PLT represents a well-established inverse predictor of liver fibrosis stage in patients with CLD[17]. In this cohort, we confirmed an inverse correlation between PLT and the condition of chronic HBV infection; thrombocytopenia emerged as an independent risk factor for both compensated and decompensated HBV-related cirrhosis. The pathogenesis of thrombocytopenia in CLD is multifactorial, encompassing impaired hepatic thrombopoietin production, bone-marrow suppression, splenic sequestration, and autoimmune-mediated platelet destruction[18].

ALT and AST are ubiquitous hepatic enzymes released from injured hepatocytes and are the most commonly used serum markers of hepatocellular injury[19]. In the present cohort, elevated ALT was an independent risk factor for both hepatitis and compensated cirrhosis, and its discriminative ability for CHB was comparable to that of the A-index. The value of the AST-to-ALT ratio (AAR) as a predictor of fibrosis or cirrhosis remains controversial. Guéchot et al[20] observed a progressive rise in AAR with increasing Metavir fibrosis stage in patients with CHC, yet they concluded that AAR was not a reliable discriminator of fibrosis. However, Lai et al[19] reported that a higher AAR predicted incident cirrhosis in patients with chronic HBV infection. Although logistic regression analysis in our study did not identify AST as an independent correlate of disease stage, AAR increased stepwise from the CHB to the compensated-cirrhosis to the decompensated-cirrhosis group, suggesting a positive association between AAR and disease severity. These data warrant external validation before AAR can be recommended as a biomarker of progression in chronic HBV infection.

ALB, synthesized exclusively by hepatocytes, exhibits antioxidant, immunomodulatory and detoxification properties[8]. Hypoalbuminaemia reflects hepatic synthetic failure and predicts adverse outcomes in cirrhotic patients[8]. Albumin-based indices-such as the ALBI score[9], neutrophil-to-albumin ratio[21], and age-bilirubin-albumin index[22]-are already used to stage liver fibrosis. In the present cohort, serum ALB levels declined stepwise with chronic HBV disease progression, and a reduced ALB level was identified as an independent risk factor for decompensated cirrhosis, corroborating previous data.

Cholesterol regulates diverse cellular processes, including membrane fluidity and gene transcription[23]. Both excess and deficiency are pathogenic. Although serum TC has been linked to fibrosis severity in NAFLD, evidence in chronic HBV infection is scarce. Janičko et al[24] observed that deceased patients with cirrhosis had higher serum TC levels than survivors and identified TC as an independent predictor of mortality. In the present HBV cohort, TC was additionally found to be a risk factor for both compensated and decompensated cirrhosis. These data underscore the need for mechanistic studies clarifying how hypercholesterolaemia may accelerate cirrhosis progression.

The discriminative performance of a model is considered excellent when the AUC is 0.90-1.00, good at 0.80-0.90, and fair at 0.70-0.80[9]. In the present study, all three developed models achieved excellent accuracy for staging disease progression of chronic HBV infection, especially for cirrhosis. At the optimal cut-off of 12.76, the MAPTAL index identified compensated cirrhosis with sensitivity 0.81, specificity 0.88, and an AUC of 0.918, outperforming the APRI, Fib-4, and S index, although remaining slightly below LPRI. Similarly, at a cut-off of 11.39, the APTAL index yielded sensitivity 0.87, specificity 0.94, and an AUC of 0.968 for decompensated cirrhosis, surpassing APRI, LPRI, Fib-4, and S index.

The primary strength of our study is the inclusion of patients with chronic HBV infection regardless of prior antiviral therapy, enhancing the generalizability of our findings to routine clinical practice. Furthermore, the established models use simple arithmetic and readily available laboratory and elastography measurements, making them cost-effective and easily implemented in resource-limited settings. Importantly, these indices address the limitations of prior models by providing stage-specific predictions using readily available clinical parameters, thereby enhancing clinical utility across HBV populations. To our knowledge, MAPTAL and APTAL are the first non-invasive indices to integrate demographic characteristics, hematologic indices, and LSM, thereby providing clinicians with a comprehensive tool for assessing chronic HBV disease progression.

This study has several inherent limitations. First, the retrospective, single-center design precluded full control of operator- and patient-related factors, such as obesity and ascites, which may have introduced selection and procedural bias. Second, lifestyle variables (alcohol intake, dietary patterns, and physical activity) and certain comorbidities were not systematically recorded, so residual confounding cannot be excluded. Third, the modest sample size limits the precision of our estimates, underscores the risk of over-fitting, and precludes formal calibration assessment; therefore, the generalizability of MAPTAL and APTAL to diverse HBV populations remains to be established, the model should be considered hypothesis-generating rather than definitive. To address these weaknesses, we are establishing a prospective, multicenter cohort that: (1) Pre-specifies BMI and ascites status as stratification variables; (2) Employs standardized questionnaires to record lifestyle factors; and (3) Uses ICD-10 coding for complete comorbidity capture. Building upon this expanded dataset, we will construct a support vector machine classifier through machine learning to facilitate intuitive risk stratification and enhance clinical applicability. We explicitly propose this multicenter data collection, calibration assessment, and model refinement as our immediate next step, and strongly recommend that any future clinical implementation be preceded by large-scale prospective validation across diverse centers and populations.

CONCLUSION

In conclusion, we provide early proof-of-concept that two parsimonious, non-invasive models (MAPTAL and APTAL), constructed solely from routine demographics, hematologic indices and LSM, reliably predict disease progression stages in chronic HBV-infected patients and remain cost-effective for resource-limited settings. This retrospective, single-center analysis should be viewed as an initial scaffold that now demands rigorous, prospective, multicenter validation before consideration for routine clinical use.

ACKNOWLEDGEMENTS

The staffs in the Department of Infectious Diseases at Shanxi Bethune Hospital supported this work.

References
1.  Polaris Observatory Collaborators. Global prevalence, treatment, and prevention of hepatitis B virus infection in 2016: a modelling study. Lancet Gastroenterol Hepatol. 2018;3:383-403.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 1376]  [Cited by in RCA: 1282]  [Article Influence: 160.3]  [Reference Citation Analysis (4)]
2.  Chinese Society of Hepatology;  Chinese Medical Association; Chinese Society of Infectious Diseases, Chinese Medical Association. [Guidelines for the prevention and treatment of chronic hepatitis B (version 2022)]. Zhonghua Gan Zang Bing Za Zhi. 2022;30:1309-1331.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 48]  [Reference Citation Analysis (2)]
3.  Kozumi K, Kodama T, Murai H, Sakane S, Govaere O, Cockell S, Motooka D, Kakita N, Yamada Y, Kondo Y, Tahata Y, Yamada R, Hikita H, Sakamori R, Kamada Y, Daly AK, Anstee QM, Tatsumi T, Morii E, Takehara T. Transcriptomics Identify Thrombospondin-2 as a Biomarker for NASH and Advanced Liver Fibrosis. Hepatology. 2021;74:2452-2466.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 140]  [Cited by in RCA: 119]  [Article Influence: 23.8]  [Reference Citation Analysis (0)]
4.  Chinese Society of Hepatology Chinese Medical Association; Chinese Society of Gastroenterology Chinese Medical Association;  Chinese Society of Infectious Diseases, Chinese Medical Association. [Consensus on the diagnosis and therapy of hepatic fibrosis in]. Zhonghua Gan Zang Bing Za Zhi. 2019;27:657-667.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 11]  [Reference Citation Analysis (0)]
5.  Wai CT, Cheng CL, Wee A, Dan YY, Chan E, Chua W, Mak B, Oo AM, Lim SG. Non-invasive models for predicting histology in patients with chronic hepatitis B. Liver Int. 2006;26:666-672.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 105]  [Cited by in RCA: 108]  [Article Influence: 5.4]  [Reference Citation Analysis (0)]
6.  Dong XQ, Wu Z, Zhao H, Wang GQ; China HepB-Related Fibrosis Assessment Research Group. Evaluation and comparison of thirty noninvasive models for diagnosing liver fibrosis in chinese hepatitis B patients. J Viral Hepat. 2019;26:297-307.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 36]  [Cited by in RCA: 35]  [Article Influence: 5.0]  [Reference Citation Analysis (1)]
7.  Johnson PJ, Berhane S, Kagebayashi C, Satomura S, Teng M, Reeves HL, O'Beirne J, Fox R, Skowronska A, Palmer D, Yeo W, Mo F, Lai P, Iñarrairaegui M, Chan SL, Sangro B, Miksad R, Tada T, Kumada T, Toyoda H. Assessment of liver function in patients with hepatocellular carcinoma: a new evidence-based approach-the ALBI grade. J Clin Oncol. 2015;33:550-558.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 2296]  [Cited by in RCA: 2251]  [Article Influence: 204.6]  [Reference Citation Analysis (5)]
8.  Wang J, Zhang Z, Yan X, Li M, Xia J, Liu Y, Chen Y, Jia B, Zhu L, Zhu C, Huang R, Wu C. Albumin-Bilirubin (ALBI) as an accurate and simple prognostic score for chronic hepatitis B-related liver cirrhosis. Dig Liver Dis. 2019;51:1172-1178.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 69]  [Cited by in RCA: 58]  [Article Influence: 8.3]  [Reference Citation Analysis (0)]
9.  Fujita K, Oura K, Yoneyama H, Shi T, Takuma K, Nakahara M, Tadokoro T, Nomura T, Morishita A, Tsutsui K, Himoto T, Masaki T. Albumin-bilirubin score indicates liver fibrosis staging and prognosis in patients with chronic hepatitis C. Hepatol Res. 2019;49:731-742.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 58]  [Cited by in RCA: 48]  [Article Influence: 6.9]  [Reference Citation Analysis (2)]
10.  Fujita K, Nomura T, Morishita A, Oura K, Yoneyama H, Kobara H, Tsutsui K, Himoto T, Masaki T. Albumin-Bilirubin Score Differentiates Liver Fibrosis Stage and Hepatocellular Carcinoma Incidence in Chronic Hepatitis B Virus Infection: A Retrospective Cohort Study. Am J Trop Med Hyg. 2019;101:220-225.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 7]  [Cited by in RCA: 19]  [Article Influence: 3.2]  [Reference Citation Analysis (0)]
11.  Zhou K, Gao CF, Zhao YP, Liu HL, Zheng RD, Xian JC, Xu HT, Mao YM, Zeng MD, Lu LG. Simpler score of routine laboratory tests predicts liver fibrosis in patients with chronic hepatitis B. J Gastroenterol Hepatol. 2010;25:1569-1577.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 94]  [Cited by in RCA: 89]  [Article Influence: 5.6]  [Reference Citation Analysis (4)]
12.  Chinese Foundation for Hepatitis Prevention and Control; Chinese Society of Infectious Disease and Chinese Society of Hepatology, Chinese Medical Association;  Liver Disease Committee of Chinese Research Hospital Association. [Consensus on clinical application of transient elastography detecting liver fibrosis: a 2018 update]. Zhonghua Gan Zang Bing Za Zhi. 2019;27:182-191.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 29]  [Reference Citation Analysis (0)]
13.  Zhou JL, Wang BQ, Sun YM, Meng TT, Wu SS, Ma H, Ou XJ, You H, Jia JD, Wu XN. [Application value of liver stiffness measurement - to - platelet ratio index score in diagnosis of hepatitis B liver fibrosis and liver cirrhosis]. Linchuang Gandanbing Zazhi. 2022;38:1529-1533.  [PubMed]  [DOI]  [Full Text]
14.  Zheng WM, Tian YY. [Age-period-cohort analysis and prediction of incidence and mortality of cirrhosis in China, 1990-2019]. Jibing Jiance. 2024;39:1489-1494.  [PubMed]  [DOI]  [Full Text]
15.  Wu XN, Wang MZ, Zhang N, Zhang W, Dong J, Ke MY, Xiang JX, Ma F, Xue F, Hou JJ, Ma ZJ, Wang FM, Liu XM, Wu R, Pawlik TM, Ye K, Yu J, Zhang XF, Lyu Y. Sex-determining region Y gene promotes liver fibrosis and accounts for sexual dimorphism in its pathophysiology. J Hepatol. 2024;80:928-940.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 15]  [Cited by in RCA: 25]  [Article Influence: 12.5]  [Reference Citation Analysis (0)]
16.  Shimizu I, Ito S. Protection of estrogens against the progression of chronic liver disease. Hepatol Res. 2007;37:239-247.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 72]  [Cited by in RCA: 77]  [Article Influence: 4.1]  [Reference Citation Analysis (0)]
17.  Gotlieb N, Schwartz N, Zelber-Sagi S, Chodick G, Shalev V, Shibolet O. Longitudinal decrease in platelet counts as a surrogate marker of liver fibrosis. World J Gastroenterol. 2020;26:5849-5862.  [PubMed]  [DOI]  [Full Text]
18.  Giannini EG. Review article: thrombocytopenia in chronic liver disease and pharmacologic treatment options. Aliment Pharmacol Ther. 2006;23:1055-1065.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 151]  [Cited by in RCA: 136]  [Article Influence: 6.8]  [Reference Citation Analysis (0)]
19.  Lai X, Chen H, Dong X, Zhou G, Liang D, Xu F, Liu H, Luo Y, Liu H, Wan S. AST to ALT ratio as a prospective risk predictor for liver cirrhosis in patients with chronic HBV infection. Eur J Gastroenterol Hepatol. 2024;36:338-344.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 13]  [Cited by in RCA: 11]  [Article Influence: 5.5]  [Reference Citation Analysis (0)]
20.  Guéchot J, Boisson RC, Zarski JP, Sturm N, Calès P, Lasnier E; ANRS HCEP 23 Fibrostar Group. AST/ALT ratio is not an index of liver fibrosis in chronic hepatitis C when aminotransferase activities are determinate according to the international recommendations. Clin Res Hepatol Gastroenterol. 2013;37:467-472.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 11]  [Cited by in RCA: 18]  [Article Influence: 1.4]  [Reference Citation Analysis (0)]
21.  Bao B, Xu S, Sun P, Zheng L. Neutrophil to albumin ratio: a biomarker in non-alcoholic fatty liver disease and with liver fibrosis. Front Nutr. 2024;11:1368459.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 15]  [Reference Citation Analysis (0)]
22.  Yilmaz B, Kayadibi H, Yeniova AO, Koseoglu H, Simsek Z. The age, bilirubin and albumin (ABA) index: a novel noninvasive index for predicting liver fibrosis in patients with chronic hepatitis C infection. Eur J Gastroenterol Hepatol. 2021;33:e290-e296.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 3]  [Cited by in RCA: 4]  [Article Influence: 0.8]  [Reference Citation Analysis (0)]
23.  Schade DS, Shey L, Eaton RP. Cholesterol Review: A Metabolically Important Molecule. Endocr Pract. 2020;26:1514-1523.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 29]  [Cited by in RCA: 173]  [Article Influence: 34.6]  [Reference Citation Analysis (0)]
24.  Janičko M, Veselíny E, Leško D, Jarčuška P. Serum cholesterol is a significant and independent mortality predictor in liver cirrhosis patients. Ann Hepatol. 2013;12:581-587.  [PubMed]  [DOI]
Footnotes

Peer review: Externally peer reviewed.

Peer-review model: Single blind

Specialty type: Gastroenterology and hepatology

Country of origin: China

Peer-review report’s classification

Scientific quality: Grade A, Grade C

Novelty: Grade B, Grade C

Creativity or innovation: Grade B, Grade C

Scientific significance: Grade A, Grade C

P-Reviewer: Liu Q, PhD, China; Paudel D, MD, Chief Physician, Nepal S-Editor: Qu XL L-Editor: A P-Editor: Wang CH

Write to the Help Desk