Neurology · Medical Literature
Systematic review of gut microbiota structural characteristics in patients with ischemic stroke.
Abstract
Emerging evidence implicates the gut microbiota (GM) in the pathogenesis of ischemic stroke (IS). However, systematic evaluations synthesizing evidence on microbial shifts, functional alterations, and clinical correlations are lacking. This study aimed to comprehensively analyze GM characteristics in IS patients versus healthy controls (HC). A systematic search of PubMed, Web of Science, and Embase was conducted from inception to April 2025. Observational studies comparing GM in IS patients and HC were included. Data on alpha/beta diversity, relative microbial abundance (phylum to species), predicted functional pathways, and microbiota-clinical indicator correlations were extracted. A narrative synthesis summarized the direction and consistency of reported findings. Twenty-four studies involving 1,957 participants (1,159 IS, 798 HC) were included. The qualitative synthesis revealed no consistent differences in alpha diversity indices between IS and HC, while significant beta diversity separation was frequently reported (18/22 studies). At the phylum level, an increased relative abundance of Proteobacteria and a decreased abundance of Firmicutes were the most consistent findings in IS patients. Key genera frequently enriched in IS included Lactobacillus, Streptococcus, and Parabacteroides, while Faecalibacterium, Agathobacter, and Blautia were commonly depleted. LEfSe analysis confirmed these discriminative taxa. Predicted functional analysis suggested enrichment of pro-inflammatory pathways (e.g., lipopolysaccharide biosynthesis) and depletion of metabolic and neuroprotective pathways in IS. Clinically, Faecalibacterium abundance correlated negatively with NIHSS and mRS scores, while Lactobacillus and Streptococcus showed positive correlations with stroke severity and inflammatory markers. Ischemic stroke is associated with distinct gut microbial dysbiosis characterized by a pro-inflammatory, opportunistic pathogen-enriched, and butyrate-producer-depleted profile. These alterations correlate with disease severity and are linked to predicted disruptions in microbial metabolic functions. However, whether this dysbiosis is a cause, consequence, or exacerbating factor of IS remains to be determined. The gut microbiota may serve as a potential biomarker and therapeutic target in ischemic stroke. https://www.crd.york.ac.uk/PROSPERO/view/CRD420251117510, identifier CRD420251117510.
Clinical relevance
What are the consistent structural and functional characteristics of the gut microbiota in patients with ischemic stroke compared to healthy controls, and how do these microbial features correlate with clinical severity indicators? __SNAPSHOT__ {"studyDesign":"Systematic review with narrative (qualitative) synthesis; no meta-analysis performed","setting":"International; multi-database search of PubMed, Web of Science, and Embase from inception to April 2025","sampleSize":"n=1,957 across 24 included studies (IS: n=1,159; HC: n=798)","population":"Adult patients with confirmed ischemic stroke compared to healthy controls; specific inclusion criteria varied across the 24 primary observational studies","intervention":"Not applicable (observational exposure: presence of ischemic stroke)","comparator":"Healthy controls (HC) without ischemic stroke","primaryOutcome":"Gut microbiota structural characteristics, including alpha and beta diversity indices, relative microbial abundance from phylum to species level, predicted functional pathway enrichment/depletion (PICRUSt/KEGG-based), and correlation of microbial taxa with clinical indicators (e.g., NIHSS, mRS scores)","followUp":"Not reported (cross-sectional microbiome sampling in primary studies; timepoint relative to stroke onset varied across studies)","fundingSource":"Not reported in available full text"}
Key findings
The qualitative synthesis of 24 observational studies (n=1,957) identified the following directional patterns. Beta diversity: significant community-level compositional separation between IS and HC was reported in 18 of 22 studies (82%), indicating a reproducibly distinct microbial community structure in IS. Alpha diversity: no consistent difference in within-sample species richness or evenness was identified across studies — a notable discordance from the beta diversity signal. Phylum level: increased relative abundance of Proteobacteria (reported in 6 studies) and decreased abundance of Firmicutes (reported in 5 studies) were the most consistent phylum-level findings; changes in Bacteroidota and Actinobacteria were directionally inconsistent across studies. Genus level (vote-counting synthesis): genera significantly enriched in IS patients included Lactobacillus (6 studies), Streptococcus (4 studies), and Parabacteroides (4 studies); genera commonly depleted included Faecalibacterium, Agathobacter, and Blautia. LEfSe analysis across primary studies confirmed these discriminative taxa. Functional predictions (PICRUSt/KEGG-derived, indirect estimates only): IS gut microbiomes showed predicted enrichment of pro-inflammatory and pro-pathogenic pathways (lipopolysaccharide biosynthesis, flagellar assembly, pyrimidine metabolism) and depletion of host-nutritive and anti-inflammatory pathways (amino acid metabolism, cofactor/vitamin metabolism, butanoate/SCFA metabolism). One study reported paradoxical enrichment of butanoate metabolic genetic potential despite depletion of key SCFA-producing genera, illustrating the limitation of gene-content-based predictions. Clinical correlations: Faecalibacterium abundance correlated negatively with NIHSS and mRS scores and positively with cognitive function; Lactobacillus and Streptococcus showed positive correlations with stroke severity and inflammatory markers. No pooled effect sizes, confidence intervals, or p-values are reported in the narrative synthesis; individual study statistics were not aggregated. NNT/NNH: not calculable from this study design.
Practice implications
This systematic review establishes that ischemic stroke is reproducibly associated with a distinct gut dysbiotic pattern — characterized by enrichment of potentially pro-inflammatory genera (Lactobacillus, Streptococcus, Parabacteroides, Proteobacteria at the phylum level) and depletion of key butyrate-producing commensals (Faecalibacterium, Agathobacter, Blautia, Roseburia) — across a geographically and methodologically diverse body of literature. The highly consistent beta diversity separation (18/22 studies) provides the strongest evidence for a distinct IS-associated microbial community structure. However, the study does not — and by design cannot — establish whether gut dysbiosis precedes and contributes causally to stroke, whether it is a consequence of stroke physiology (e.g., gut hypoperfusion, immobility, post-stroke antibiotic use, dietary changes), or both. The clinical correlations between Faecalibacterium depletion and stroke severity (NIHSS, mRS) are hypothesis-generating but derived from associative data subject to unmeasured confounding. Functional pathway predictions are indirect (PICRUSt/KEGG applied to 16S rRNA gene data) and cannot confirm that metabolic shifts actually occur at the protein or metabolite level. The studied population is likely predominantly East Asian, given the geographic distribution of included studies, limiting generalizability to other ethnic and dietary contexts. The absence of a pooled quantitative meta-analysis means that effect sizes are not estimated; the narrative synthesis reflects vote-counting, which is prone to bias toward larger or more frequently replicated associations. Practice should not change based on this review; it provides a well-organized foundation for future hypothesis-driven prospective studies and therapeutic trials. Only abstract-level detail was available for a subset of primary studies, and the full dataset of individual study statistics was not accessed during this analysis. __LIMITATIONS__ Narrative synthesis only — no meta-analysis performed; vote-counting methodology used for genus-level findings is susceptible to false consensus and cannot quantify effect magnitude or heterogeneity.,Fundamental causal ambiguity: cross-sectional observational design of all included studies cannot distinguish whether dysbiosis precedes ischemic stroke (contributing to risk) or results from it (acute stroke physiology, immobility, dysphagia, antibiotic exposure, altered nutrition).,High methodological heterogeneity across included studies: variable 16S rRNA gene hypervariable regions sequenced, different bioinformatic pipelines, sequencing depths, and taxonomic reference databases preclude reliable cross-study quantitative comparison.,Functional pathway predictions are entirely indirect, derived from PICRUSt or Tax4Fun applied to 16S rRNA amplicon data — a method with known limitations in accuracy compared to shotgun metagenomics; actual metabolite production is not confirmed.,Likely strong geographic concentration of included studies (predominantly East Asian populations, particularly Chinese cohorts), severely limiting generalizability to Western, African, or South Asian patient populations with different dietary patterns and baseline microbiome compositions.,Critical uncontrolled confounders across primary studies: post-stroke antibiotic use, proton pump inhibitor or statin use, nasogastric feeding, swallowing impairment, bowel dysmotility, and pre-stroke dietary patterns — all of which independently alter gut microbiota — were not uniformly reported or adjusted for.,Single time-point microbiome sampling in most primary studies; it is unknown whether the dysbiosis pattern reflects the acute stroke period, subacute recovery, or a chronic pre-existing state.,Alpha diversity inconsistency across studies undermines a simple model of reduced microbial richness in IS; this finding is not adequately explained by the current synthesis.,Clinical correlations (e.g., Faecalibacterium vs. NIHSS) are associative and subject to unmeasured confounding; no causal inference methodology (e.g., Mendelian randomization) was applied.,Funding source not reported, precluding assessment of potential sponsor-related bias in study selection or interpretation. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ This systematic review of 24 studies (n=1,957) demonstrates that ischemic stroke is consistently associated with a distinct gut microbiome composition, characterized by enrichment of Proteobacteria, Lactobacillus, and Streptococcus and depletion of butyrate-producing commensals such as Faecalibacterium, with community-level (beta diversity) separation found in 82% of studies. However, causality cannot be established from this cross-sectional evidence base, functional pathway data are indirect estimates only, and substantial methodological heterogeneity precludes quantitative effect size estimation. Clinicians should regard these findings as hypothesis-generating: they do not support any change in current post-stroke management but should motivate large-scale, prospective, multi-timepoint cohort studies with rigorous confounder control and, ultimately, randomized trials of microbiome-targeted adjuvant therapies.