Hepatitis B virus (HBV) infection poses a persistent global health burden, driving the progression of chronic liver disease, cirrhosis, and hepatocellular carcinoma (HCC) (Hsu et al., 2023). In HBV-related cirrhosis, sustained liver enzyme abnormalities—even after achieving undetectable HBV-DNA through antiviral therapy—signal ongoing hepatic injury, accelerating fibrosis progression and increasing risks of decompensation and HCC (Boursier et al., 2016). These abnormalities may further reflect impaired hepatic metabolic functions, including disrupted protein synthesis and detoxification pathways, thereby compounding clinical morbidity (Albillos et al., 2020).

Accumulating evidence has characterized gut microbiome (GM) dysbiosis in HBV-related liver disease, with microbial alterations directly linked to disease severity and clinical outcomes. Chronic HBV infection exhibits reduced microbial α-diversity and enriched pro-inflammatory taxa (e.g., Enterobacteriaceae), correlating with hepatic inflammation and fibrosis progression (Shen et al., 2023). Longitudinal analyses further reveal dynamic GM shifts during disease evolution, where Streptococcus- and Veillonella-dominated profiles predict cirrhosis development (Chen et al., 2020). In decompensated cirrhosis, GM signatures (e.g., depletion of Lachnospiraceae) serve as prognostic markers for acute-on-chronic liver failure (Wang et al., 2021), while metabolomic perturbations—particularly disrupted bile acid and aromatic amino acid metabolism—differentiate disease stages (Shi et al., 2025). Collectively, these findings establish the GM as a key regulator of HBV pathogenesis, though the microbial-metabolite drivers of persistent inflammation in virologically suppressed patients remain mechanistically undefined.

Emerging evidence highlights the gut-liver axis as a pivotal mediator of liver disease pathogenesis. This bidirectional network integrates metabolic, immune, and neuroendocrine crosstalk via portal circulation and biliary excretion (Tripathi et al., 2018). A functionally intact intestinal barrier—comprising chemical (e.g., antimicrobial peptides), physical (e.g., epithelial tight junctions), and immunological (e.g., gut-associated lymphoid tissue) components—prevents microbial translocation. However, cirrhosis-induced dysbiosis disrupts this equilibrium, favoring pathobionts (e.g., Enterobacteriaceae) over commensals (e.g., Bifidobacterium, Lactobacillus), increasing intestinal permeability, and triggering hepatic inflammation through pathogen-associated molecular pattern (PAMP) release (Odenwald and Turner, 2017; Shu et al., 2022).

While 16S rRNA sequencing has delineated taxonomic shifts in cirrhosis, functional insights into host-microbiota interplay require metabolomic integration. Microbial-derived metabolites—including bile acids, short-chain fatty acids, and trimethylamine—directly modulate hepatic inflammation, fibrosis, and metabolic reprogramming (Sun et al., 2021). Fecal metabolomics thus bridges the gap between microbial composition and disease mechanisms, offering translational opportunities for therapeutic targeting.

Notably, a subset of HBV-related cirrhosis patients with virological suppression (undetectable HBV-DNA) exhibits persistent liver enzyme elevations, suggesting unresolved gut-liver axis dysregulation. However, the microbial and metabolic signatures underlying this phenotype remain uncharacterized. Here, we employed 16S rRNA gene sequencing and untargeted liquid chromatography-mass spectrometry (LC–MS) to profile gut microbiota and metabolites in this high-risk population, aiming to uncover mechanistic drivers and potential therapeutic targets.

Contrary to classical dysbiosis paradigms, virologically suppressed HBV-cirrhosis patients with persistent liver enzyme abnormalities exhibit elevated alpha diversity (Chao1, Shannon, and Simpson indices). We propose this reflects a metabolically driven maladaptive state characterized by: Antiviral therapy-induced niche remodeling, promoting taxonomic diversification without functional restoration; Depletion of hepatoprotective bile acids (UCA/TDC), reducing antimicrobial selection pressure and enabling pathobiont co-existence; Oxidative stress (evidenced by elevated 13(S)-hydroperoxylinolenic acid), favoring resilient but pro-inflammatory taxa (e.g., Ruminococcus). Despite higher diversity, functional impoverishment is evident—reduced SCFA producers (Blautia) and enriched TLR4 agonists (Escherichia-Shigella)—while beta-diversity confirms distinct microbial clustering (ANOSIM R = 0.177, p = 0.0049). These findings suggest alpha diversity may serve as a context-dependent biomarker in cirrhosis: elevated values during virological suppression could indicate compensatory but unstable microbial reorganization rather than restored health. Therapeutic strategies should prioritize functional restoration (e.g., bile acid supplementation, antioxidant therapy) over mere diversity modulation. Our observations contrast with classical cirrhosis-associated dysbiosis patterns (Bajaj et al., 2014) but align with recent reports of similar diversity paradoxes in metabolic dysfunction-associated steatotic liver disease (Zhang et al., 2021). This implies a conserved microbial stress response to hepatic metabolic disturbances, potentially independent of disease etiology. The co-enrichment of commensal (Faecalibacterium) and pathogenic (Ruminococcus) taxa in our cohort mirrors post-antiviral therapy studies (Chen et al., 2016), supporting the notion of incomplete microbial recovery despite virological control. In patients with normal liver enzymes, the milder dysbiosis (e.g., Bacteroides dominance, higher Blautia abundance) may eflect: Compensatory metabolic adaptation via SCFA-mediated immunomodulation (Wang et al., 2024); Antiviral therapy off-target effects on specific taxa (Benítez-Páez et al., 2020); Diagnostic latency, where microbial shifts precede biochemical abnormalities (Boursier et al., 2016). This dissociation between microbial (Benítez-Páez et al., 2020) and biochemical profiles underscores the need for multi-parameter assessments in cirrhosis management.

Our study delineates distinct gut microbiota and metabolite profiles in HBV-related cirrhosis patients achieving virological suppression (undetectable HBV-DNA), stratified by the presence or absence of persistent liver enzyme abnormalities. Key findings reveal a dichotomy between the two cohorts: The normal liver enzyme group exhibited dominance of Bacteroides and Blautia—taxa associated with SCFA production and immunomodulation—alongside enrichment of Escherichia-Shigella, a potential indicator of subclinical inflammation. In contrast, the abnormal liver enzyme group demonstrated marked enrichment of Faecalibacterium, Christensenellaceae_R-7_group, and Ruminococcus, suggesting a dysbiotic state that paradoxically combines commensal resilience with pro-inflammatory potential. The identification of 52 differentially abundant metabolites—including bile acids (e.g., UCA/TDCA), sterols, and oxidative stress markers—highlights two critical pathways: Bile acid disruption: Depletion of hepatoprotective bile acids correlates with impaired microbial conversion capacity, exacerbating cholestatic injury. Oxidative stress: Elevated lipid peroxidation products align with the enrichment of oxidative stress-resistant taxa (e.g., Prevotella_9), creating a feed-forward loop of inflammation. These findings challenge the paradigm that virological suppression universally resolves cirrhosis-associated dysbiosis. Instead, we propose: Persistent enzyme abnormalities reflect functional microbiota disruption beyond taxonomic shifts, evidenced by metabolite-microbe networks (e.g., Ruminococcus-bile acid associations). Therapeutic strategies should target metabolite restitution (e.g., bile acid supplementation) and oxidative stress mitigation to break the inflammation-dysbiosis cycle. The coexistence of beneficial (Faecalibacterium) and pathogenic (Ruminococcus) taxa in the abnormal group mirrors findings in metabolic dysfunction-associated steatotic liver disease, suggesting shared mechanisms of metabolic dysregulation post-viral suppression. Notably, the Bacteroides-dominant profile in the normal group resembles microbial recovery patterns observed in NAFLD remission, hinting at potential diagnostic or prognostic utility. While our metabolomics data robustly link specific microbes to functional outcomes, longitudinal studies are needed to establish causality. Additionally, the therapeutic potential of targeting identified metabolites (e.g., FXR agonists for bile acid restoration) warrants clinical validation.

In patients with undetectable HBV-DNA levels and normal liver enzymes, Bacteroides—a major producer of propionate, a short-chain fatty acid (SCFA) with anti-inflammatory, anti-lipogenic, and immunomodulatory properties—may contribute to maintaining metabolic and immune homeostasis (Zhang et al., 2025). Propionate-mediated signaling plays a critical role in regulating mucosal immunity, appetite, and cholesterol metabolism, and its deficiency is linked to chronic inflammatory and metabolic disorders, including type 2 diabetes, obesity, and colorectal cancer. Dietary interventions aimed at enhancing colonic propionate production represent a promising strategy for restoring immune and metabolic balance (Wang et al., 2024). The paradoxical elevation of Blautia, another SCFA-producing genus, may reflect compensatory mechanisms to counteract endotoxin-induced damage, as its abundance inversely correlates with portal hypertension severity (Bajaj et al., 2014). While some studies associate Blautia with metabolic disorders such as obesity and type 2 diabetes (Slyepchenko et al., 2017), others suggest its beneficial role in reducing inflammation, as evidenced by its negative correlation with fecal TNF-α levels in normal-weight children. This discrepancy highlights the context-dependent nature of Blautia’s role in host physiology (Benítez-Páez et al., 2020; Slyepchenko et al., 2017; Kashtanova et al., 2018). Conversely, the increased presence of Escherichia-Shigella, a well-established driver of TLR4-mediated pro-inflammatory responses, may indicate underlying gut dysbiosis, exacerbating liver injury in chronic hepatitis B (Duan et al., 2019), Similarly, Prevotella_9, enriched in patients with progressive hepatocellular carcinoma (HCC), is associated with tumor advancement and poorer clinical outcomes. Although Faecalibacterium is generally regarded as anti-inflammatory, its increased abundance in certain chronic liver diseases may paradoxically correlate with disease progression, potentially reflecting the host’s attempt to modulate immune responses in the face of ongoing liver injury (Lee et al., 2022). Although Faecalibacterium is generally regarded as an anti-inflammatory bacterium, its increased abundance in certain chronic liver diseases may paradoxically correlate with disease progression, potentially reflecting the host’s attempt to modulate immune responses in the face of ongoing liver injury.

UCA and TDC were significantly reduced in the abnormal group, highlighting impaired bile acid homeostasis as a key feature of persistent liver enzyme abnormalities in HBV-related cirrhosis. UCA, a tertiary bile acid with hepatoprotective and anti-inflammatory properties, plays a crucial role in reducing hepatocyte apoptosis, modulating bile acid metabolism, and alleviating endoplasmic reticulum stress (Beuers et al., 2015). Recent studies further suggest that UCA exerts anti-fibrotic effects by inhibiting hepatic stellate cell activation, positioning it as a potential therapeutic agent for liver fibrosis and cirrhosis. Similarly, TDC, a conjugated bile acid essential for lipid emulsification, contributes to maintaining bile acid balance and reducing hepatic inflammation. The depletion of these protective bile acids underscores the role of bile acid dysregulation in driving liver injury. Dietary supplementation with bile acids, such as UCA, has been shown to reduce experimentally induced hepatic carcinogenesis in animal models, suggesting its potential for clinical translation (Fiorucci et al., 2021). Elevated DCA-Gly and 13(S)-hydroperoxylinolenic acid in HBV-related cirrhosis patients with undetectable HBV-DNA yet persistent liver dysfunction suggest distinct pathophysiological contributions. DCA-Gly accumulation may reflect impaired bile acid detoxification and gut-liver axis disruption, exacerbating hepatocyte injury via mitochondrial dysfunction (Beuers et al., 2015). 13(S)-hydroperoxylinolenic acid, a lipid peroxidation marker, indicates oxidative stress-driven inflammation, potentially activating pro-fibrotic pathways (Kühn and O'Donnell, 2006). These findings underscore metabolic dysregulation as a key driver of liver injury and suggest therapeutic targets, such as bile acid supplementation and antioxidant therapy, to restore metabolic balance and mitigate liver damage.

Our study has several limitations that warrant consideration. The relatively small sample size in microbiome and metabolomic analyses may limit the generalizability of our findings, necessitating validation in larger, multi-center prospective cohorts. Additionally, the use of 16S rRNA sequencing, while informative, provides only genus-level resolution, which may lead to inconsistent taxonomic assignments, as exemplified by the ambiguous abundance patterns of Escherichia-Shigella. To achieve species-level precision and functional insights, shotgun metagenomic sequencing is recommended for future studies. Furthermore, although non-targeted metabolomics offers a comprehensive overview of metabolic perturbations, it lacks the ability to establish causal relationships between microbiome composition and liver disease progression. Functional validation through in vitro and in vivo models is essential to elucidate these mechanisms. As a cross-sectional study, our design precludes the assessment of temporal changes in the gut microbiome and fecal metabolome, which are critical for understanding the dynamic progression of cirrhosis. Longitudinal studies incorporating serial fecal sampling at different disease stages are needed to identify robust microbiome and metabolite biomarkers associated with disease progression. Addressing these limitations through advanced sequencing technologies, larger cohort studies, and mechanistic investigations will significantly enhance our understanding of the gut-liver axis and inform the development of targeted therapeutic strategies for HBV-related cirrhosis.

In summary, this study demonstrates that gut microbiota dysbiosis and metabolic perturbations significantly influence persistent liver enzyme abnormalities in cirrhotic patients achieving HBV-DNA undetectability. The marked depletion of anti-inflammatory genera (e.g., Blautia) and hepatoprotective bile acids (e.g., ursocholic acid, UCA) correlates positively with hepatic dysfunction. This bidirectional gut-liver crosstalk manifests as: Progressive liver injury exacerbating intestinal ecological imbalance, and gut-derived metabolites accelerating disease progression through systemic inflammation. Notably, we propose the gut-liver axis remains a central metabolic regulator even post-virological control, with its associated microbial signatures and metabolite profiles serving as non-invasive prognostic tools. These findings establish a mechanistic foundation for targeted interventions including microbiota modulation and metabolite supplementation. Future investigations should employ longitudinal cohorts with serial fecal sampling across cirrhosis stages to dynamically track progression-associated biomarkers. Integrated multi-omics approaches, expanded cohorts, and mechanistic validation will further elucidate gut-liver interactions, advancing precision therapeutics for HBV-related cirrhosis.