Effect of Transjugular Intrahepatic Portosystemic Shunt on Intestinal Flora in Liver Cirrhosis: A Systematic Review

Abstract

Background

Liver cirrhosis disturbs the gut–liver axis, causing microbial imbalance and portal hypertension. Transjugular intrahepatic portosystemic shunt (TIPS) is a common method of managing portal hypertension, but it may predispose to hepatic encephalopathy (HE). The relationship between TIPS, intestinal microbiota, and HE is an active field of study, and there is increasing evidence to indicate that portal hypertension itself can lead to changes in microbes.

Purpose

To comprehensively assess the impact of TIPS on the gut microbiota of cirrhotic patients and to determine its correlation with clinical outcomes, especially HE.

Methods

A systematic search of prospective and cross-sectional studies was conducted. Studies included evaluated gut microbial composition pre- and post-TIPS with 16S rRNA sequencing and metabolomic profiling.

Results

Transjugular intrahepatic portosystemic shunt is not linked to a consistent pattern of microbes but is linked to fluctuating changes. The direction of these changes seems to be dependent on clinical outcomes. Increased production of short-chain fatty acids (SCFAs) by bacteria, such as Coprococcus and Roseburia, is associated with a reduced incidence of HE, which is reported in some cohorts at around 13%. Conversely, the increase in potentially pathogenic genera such as Haemophilus and Morganella, along with the loss of beneficial taxa, is associated with a 68% increase in HE rates. The metabolomic data also reveal an inverse correlation between some circulating conjugated bile acids and the severity of HE. There is also comparative evidence that portal hypertension and not cirrhosis per se may be central to the development of dysbiosis.

Conclusion

Transjugular intrahepatic portosystemic shunt has different impacts on the gut microbiome in different patients and affects the risk of HE. Maintaining positive microbial communities could provide a foundation for risk assessment and intervention.

Introduction

Cirrhosis of the liver is an indicator of the final stage of chronic hepatic injury. Its characteristics include the unremitting deposition of fibrous tissue, the formation of regenerative nodules that alter normal structure, and the drastic reorganization of the vascular network of the organ. These pathological processes act in concert to increase resistance to blood flow in the liver, a mechanical obstacle that triggers portal hypertension—the main driver of disease and mortality in this group. Statistics based on the Global Burden of Disease 2021 evaluation highlight the sheer magnitude of this problem, estimating that approximately 58.4 million people worldwide have the condition, and that over 1.4 million die each year. This number is expected to increase, as the cases of fatty liver disease that are not related to alcohol consumption are on the rise, as well as injuries caused by alcohol abuse and chronic viral liver infections.¹ Portal hypertension causes a cascade of serious events, such as bleeding due to enlarged veins, uncontrolled retention of abdominal fluid, and cognitive dysfunction due to liver dysfunction when it reaches a level of clinical importance. The emergence of such events indicates the onset of a decompensated stage of the disease and a much more pessimistic prognosis.²

The gut–liver axis is a concept that summarizes a two-way communication maintained by the portal venous system between the intestinal tract and the hepatic organ. A diverse, healthy gut microbial community is essential to maintain gut wall integrity, regulate bile acid metabolism, and tune immune surveillance of the body. The cirrhotic process is a severe impairment of this fragile balance, which introduces a dysbiosis characterized by contraction of the total microbial diversity, in addition to the proliferation of bacterial lineages with pathogenic tendencies, in particular, representatives of the Enterobacteriaceae family and the Bacilli class.³ These changes undermine the protective functions of the intestinal barrier, resulting in increased gut leakiness and enabling the migration of bacteria or their components into the circulation—a process known as bacterial translocation. The following increase in the number of endotoxins in the blood contributes to a systemic inflammatory response that increases the rate of liver dysfunction and increases susceptibility to infections and changes in mental status.^3–6

The transjugular intrahepatic portosystemic shunt (TIPS) is an established treatment option to deal with the desperate effects of portal hypertension, namely bleeding varices that are not responsive to other treatment and fluid buildup that cannot be addressed by conventional diuretic treatment. Transjugular intrahepatic portosystemic shunt bypasses hepatic circulation; it does not restore normal physiology in the abdominal viscera and in the whole circulatory system by creating an artificial opening between the hepatic vein and a branch of the portal vein, which mechanically decreases the high portal pressure and reestablishes the normal patterns of blood flow. Sufficiently deep decrease in the portal pressure gradient is known to be a powerful predictor of prolonged survival, and an inappropriate hemodynamic reaction has its consequences of poorer disease progression.⁷ However, the same mechanism that decongests portal pressure, the diversion of venous blood of the hepatic sinusoids, also limits the liver's ability to process and counteract toxic substances that start in the gut. This physiological bypass has a direct role in the development of a common and problematic outcome: Post-TIPS hepatic encephalopathy (HE). The probability of this complication is closely related to the amount of blood that passes through the shunt, the extent of pressure drop that is attained, and the systemic concentration of ammonia and other neurotoxic substances.^4,8–10

An emerging body of modern literature suggests that the implantation of a TIPS device results in quantifiable changes in the composition as well as the functional output of the resident microbial community of the gut. Part of the literature suggests a postprocedural increase in the size of bacterial populations with an established tendency to produce ammonia, and the changes in the distribution of circulating metabolites are associated with an increased risk of encephalopathy.⁵ On the contrary, other studies have reported a partial recovery of beneficial microbial constituents and some metabolic recovery upon removal of the portal pressure load.^6,11,12 Even though the therapeutic approaches that attempt to control the microbiome, including antimicrobial medications and live beneficial cultures, as well as transplantation of whole fecal communities, have theoretical promise as a means of controlling HE, the evidence on their effectiveness is resolutely mixed.^8,11 The main questions remain unanswered: Whether portal hypertension is the major architect of the dysbiosis in such patients or whether cirrhosis exerts a more direct effect. Does TIPS intervention eventually drive the gut ecosystem to a state of long-term stability and normalization, or does it rather cause a new set of clinically significant disruptions?^9,10 The systematic review aims to synthesize and critically examine the existing literature on the implications of TIPS on the intestinal microbiota and its implications on patient care. The primary objective is to outline the particular alterations in the diversity of gut microbial communities and the proportion of species that take place in patients with liver cirrhosis directly due to the TIPS creation. Ancillary objectives include exploring the relationship between post-TIPS changes in microbial community and the development of HE; examining the evidence that can be used to establish the connection between TIPS-related changes in microbial community and indicators of bacterial migration and systemic immune response; and identifying gaps in knowledge and the methodological limitations of microbiome studies in this specific clinical scenario.

Methods

Study Design

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 framework.

Eligibility Criteria

Inclusion Criteria

Studies were included if they met the following requirements. Participants had to be adults aged 18 years or older with a confirmed diagnosis of liver cirrhosis, irrespective of etiology. The study needed to examine the elective placement of a TIPS as the primary intervention. Assessment of the gut microbiota had to be based on stool samples or direct intestinal specimens, analyzed using established methods such as culture techniques, 16S rRNA sequencing, shotgun metagenomics, or validated metabolite-based markers. Eligible studies reported changes in microbial composition, diversity, or associations with clinical outcomes, including HE or inflammatory markers. Both randomized trials and observational designs, including prospective or retrospective cohorts and case–control studies, were considered, provided they were published in English.

Exclusion Criteria

Studies were excluded if they involved animal experiments or in vitro models, or if the population included children or adolescents. Conference abstracts without full methodological detail were not considered. Research focusing on cirrhosis or TIPS without direct evaluation of gut microbiota was excluded, as were studies examining microbial communities outside the intestinal tract, such as oral or skin microbiota.

Systematic Review Protocol

Before starting the review, a comprehensive protocol was developed defining the aims of the study and the methodology to be used.

Search Strategy

The electronic databases were searched thoroughly up to April 2026. The databases were PubMed/MEDLINE, Embase, Scopus, and Web of Science Core Collection. Search terms were created with the help of a medical information specialist and included controlled vocabulary (e.g., Medical Subject Headings) as well as applicable free-text terms. It was a strategy that centered on three key concepts: TIPS, liver cirrhosis, and gut microbiota. The entire PubMed search strategy is shown in Supplementary Table 1. Only English-language studies were included in the results.

Table 1.

Simplified risk of bias assessment of included studies

Evidence	Author, year	Study design	Key limitations	Overall rating
11	Li et al., 2022	Prospective cohort	Short to mid-term follow-up only	Good
12	Gitto et al., 2023	Prospective cohort	Small sample, unclear participation rate	Fair
13	Zhao et al., 2024	Prospective cohort	Very short follow-up (1 month)	Fair
14	Swilling et al., 2024	Pilot observational	Poor reporting, unclear population, and measures	Poor
15	Heller et al., 2024	Controlled cohort	Residual confounding possible	Good
16	Machado et al., 2023	Cross-sectional	No temporal assessment, minimal HE cases	Fair
17	Chen et al., 2023	Observational cohort	Follow-up duration unclear	Fair
18	Gedgaudas et al., 2022	Cross-sectional	No longitudinal inference	Good
19	Bloom et al., 2021	Cross-sectional	Single time-point exposure and outcome	Fair

HE, hepatic encephalopathy

Data Sources

The search protocol was applied to the electronic repositories of PubMed/MEDLINE, Embase, Scopus, and Web of Science Core Collection.

Study Selection and Data Extraction

All citations found in the database searches were gathered, and duplicates were detected and removed using EndNote 20 reference management software (Clarivate, Philadelphia, PA, USA), along with the Rayyan web-based platform to screen systematic reviews. Titles and abstracts of the collected records were first evaluated by two reviewers working separately, and then the entire text of any potentially relevant articles was evaluated to confirm adherence to the previously established eligibility criteria. The two reviewers would have resolved any differences in judgment between them by mutual deliberation or by requesting the opinion of a third independent reviewer, which would be binding. A PRISMA flowchart (Fig. 1) visually illustrates the entire process of identifying the studies, screening them, and ultimately including them. Two reviewers used a standard, pilot-tested data collection form to retrieve information in the studies that met all the criteria. The content extracted was divided into the following categories: (1) basic study characteristics (authorship, year, geographical location, design); (2) baseline characteristics of the patient group (age, sex distribution, cause of liver disease, severity scores including MELD and Child–Pugh); and (3) technical characteristics of the TIPS procedure.

Fig. 1.

PRISMA flow diagram detailing the screening process

**Records were excluded after title/abstract screening because they did not meet the eligibility criteria

Data Analysis

Since the expected level of variability among studies in terms of patient groups, experimental designs, and definitions of outcomes was high, a quantitative statistical pooling of data (meta-analysis) was neither practical nor methodologically appropriate. The results are therefore synthesized through a descriptive narrative method and structured in the manner of the main and secondary research questions as stated above. The studies were clustered based on the type of outcomes that they reported, and a comparative summary of the main attributes and results is compiled in Table 1. The narrative synthesis incorporates a qualitative evaluation of the strength, reproducibility, and biological coherence of the evidence, as well as the limitations of the study designs used.

Risk of Bias Assessment

The reviewers evaluated the methodological quality and susceptibility to bias in each of the included studies separately using the National Institutes of Health (NIH) Quality Assessment Tool, which is designed to evaluate observational cohort and cross-sectional studies. Depending on the overall level of risks in a variety of areas, such as the sufficiency of the selection methods used, the quality and reliability of exposure and outcome measures, the management of participant dropout, and the statistical management of confounding factors, each study was given a final quality rating of good, fair, or poor.

Results

Flow Diagram of Studies

The PRISMA flow diagram (Fig. 1) details the screening process, showing the number of records identified, screened, and included.

Study Selection and Characteristics

Nine studies met the inclusion criteria, comprising prospective cohorts, pilot observational studies, and cross-sectional analyses.^11–19 Sample sizes ranged from 13 to 106 participants. Detailed study characteristics are summarized in Supplementary Table 2.

Table 2.

Synthesis of results addressing review objectives

Analytical domain	Key results and supporting evidence	Consistency and strength of evidence
Intestinal flora composition and diversity after TIPS	Postprocedure microbial patterns clustered into improved, stable, or deteriorated trajectories rather than a uniform shift.11  Reported taxonomic changes were heterogeneous, including increased Flavonifractor, depletion of Clostridiaceae, and enrichment of Haemophilus and Morganella in affected patients.12,13  After adjustment for antibiotics, no independent association with α-diversity loss was observed.15	Moderate: Alterations are consistently detected, but direction and taxa differ across studies, likely reflecting methodological heterogeneity and confounding (e.g., antibiotics, diet, PPI use).
Microbiota patterns and HE	Markedly different encephalopathy rates were observed across microbial trajectories (13.3 vs 68.2%).11  Encephalopathy was associated with depletion of SCFA-producing genera (Coprococcus, Roseburia, Anaerostipes, Butyricicoccus) and expansion of potentially pathogenic taxa (Haemophilus, Eggerthella, Morganella).11,13  Reduced fecal acetate, propionate, and butyrate levels were reported in affected patients.19	Strong: Prospective cohort and metabolomic data converge on a consistent association between dysbiotic progression and increased encephalopathy risk.
Bacterial translocation and systemic inflammation	Circulating microbial DNA in portal hypertension correlated with inflammatory markers (e.g., IL-8).18  Conjugated bile acid profiles in blood were inversely related to encephalopathy severity after TIPS.16  Loss of SCFA-producing bacteria provides a plausible mechanistic link to impaired barrier function.11,13,19	Limited but supportive: Evidence is indirect and largely cross-sectional, with few longitudinal assessments incorporating direct translocation or inflammatory markers.
Comparison with non-TIPS populations	No significant microbiota differences were observed between TIPS and non-TIPS cirrhotic patients after accounting for antibiotic exposure.15  Similar dysbiosis patterns in non-cirrhotic portal hypertension and cirrhosis implicate portal hypertension as a major driver of microbial alteration.9	Limited for TIPS-specific effects; informative for pathophysiology: Direct comparative data remain sparse.

HE, hepatic encephalopathy; SCFA, short-chain fatty acid; TIPS, transjugular intrahepatic portosystemic shunt

Synthesized Findings

Evidence suggests that TIPS placement is associated with measurable, heterogeneous changes in gut microbial composition and function. The key synthesized findings are presented in Table 2.

Microbiota Composition and Diversity after TIPS

Li et al. followed 106 cirrhotic patients with paired fecal sampling.¹¹ They identified distinct post-TIPS microbial trajectories (improved, stable, deteriorated). Patients whose microbiota shifted toward enrichment of autochthonous, short-chain fatty acid (SCFA)-producing taxa had significantly lower HE incidence (13.3 vs 68.2% in the deteriorated group). Gitto et al. reported a statistically significant increase in Flavonifractor and reduction in Clostridiaceae at 3 months post-TIPS, with no notable change in α-diversity.¹² In contrast, Zhao et al. found genus-level shifts associated with early HE onset, including enrichment of Morganella and depletion of butyrate-producing genera within 1 month.¹³ Heller et al. found that after controlling for antibiotic exposure, TIPS had no significant independent effect on gut microbiota, suggesting antibiotics are a major confounder.¹⁵

Hepatic Encephalopathy and Microbiota-linked Risk

Post-TIPS HE rates ranged from 21 to 33%.^11,13,17 Consistently, patients developing HE showed depletion of SCFA-producing taxa and expansion of pathogenic or urease-associated organisms.^11,13 For example, Li et al. found an inverse correlation between HE severity and genera like Coprococcus and Ruminococcus.¹¹ Bloom et al. provided functional corroboration, showing significantly lower fecal acetate, propionate, and butyrate in HE patients.¹⁹

Metabolomic and Circulating Microbiome Evidence

Machado et al. found that specific conjugated bile acids inversely correlated with HE grade post-TIPS.¹⁶ Chen et al. reported that several depleted portal serum metabolites showed predictive value for post-TIPS hepatic encephalopathy, with five metabolites having an area under the receiver operating characteristic curve (AUC) ≥ 0.75.¹⁷ Gedgaudas et al. demonstrated correlations between circulating Escherichia/Shigella abundance and IL-8 levels in portal hypertension, linking microbial translocation to inflammation.¹⁸

Assessment of Risk of Bias

As summarized in Tables 1 and 2, the risk of bias varied. Studies with larger samples, longitudinal design, and control for confounders like antibiotics were rated “Good.”^11,15 Studies were downgraded for small samples, short follow-up, or cross-sectional design, limiting causal inference.

Discussion

Summary of Main Findings

This systematic review demonstrates that TIPS placement leads to measurable, heterogeneous changes in the gut flora of cirrhotic patients. The microbial response is not uniform but follows distinct trajectories that are strongly correlated with clinical outcomes, particularly HE. Evidence from the largest prospective cohort indicates that patients whose microbiota shifts toward a more beneficial, SCFA-producing profile post-TIPS have a markedly lower risk of HE compared with those with a deteriorating microbial profile.¹¹ Metabolomic and circulating microbiome data support a mechanistic link through altered bile acid metabolism, bacterial translocation, and systemic inflammation.

Limitations

The evidence base has several limitations. Most included studies had small sample sizes and short follow-up durations, which restricts the generalizability of the findings. There was substantial methodological heterogeneity in sequencing platforms, sampling intervals, metabolomic methods, and bioinformatic pipelines. Important confounders, including antibiotic use, rifaximin, proton-pump inhibitors, dietary intake, and disease severity, were inconsistently reported or controlled across studies. In addition, the predominance of within-patient pre-post designs and the limited number of robust matched non-TIPS control groups restrict causal inference regarding TIPS-specific microbiome effects vs the natural history of cirrhosis and portal hypertension.

Conclusion

Transjugular intrahepatic portosystemic shunt placement alters the intestinal milieu, and the subsequent response of the gut microbiota appears to mediate clinical outcomes like HE. The risk of post-TIPS HE is not uniformly increased but is closely tied to microbiota trajectories. A shift toward a restorative profile with beneficial SCFA-producing bacteria is associated with better neurological outcomes. These findings underscore the gut-liver-brain axis in TIPS pathophysiology and suggest that microbiome assessment could aid in risk stratification. They also provide a rationale for interventional trials testing microbiota-directed therapies alongside TIPS to steer the microbiome toward a neuroprotective state and improve patient outcomes.

Clinical Significance

This systematic review highlights the potential clinical relevance of gut microbiota changes after TIPS placement. Microbial and metabolomic profiles may help identify patients at increased risk of post-TIPS hepatic encephalopathy and may support future risk stratification and microbiota-directed preventive strategies. These findings also provide a basis for future interventional studies evaluating microbiome-modulating therapies in patients undergoing TIPS.

Ethical Approval

Nil. Ethical approval was not required because this study is a systematic review of previously published literature and did not involve new human or animal participants.

Data Availability Declaration

No new datasets were generated or analyzed in this study. All data discussed in this review are available from the previously published articles cited in the manuscript.

Artificial Intelligence (AI) Disclosure

Artificial intelligence-assisted language support was used only during proof correction to help improve wording, formatting, and responses to publisher queries. The authors reviewed, verified, and approved all content and take full responsibility for the final manuscript.

Acknowledgments

Nil.

Authors' Contributions

Abdala Mohamed conceptualized the study, performed the literature search, screened studies, extracted data, and drafted the manuscript. Mohamed Bade contributed to study screening, data extraction, risk of bias assessment, and manuscript revision. Zhao Xiang-An supervised the study, reviewed the intellectual content, and critically revised the manuscript. All authors read and approved the final manuscript.

Supplementary Materials

All the supplementary materials are available on its journal website www.ejohg.com

Orcid

Abdala Mohamed https://orcid.org/0009-0008-6648-610X