The pathogenesis of IS involves multiple mechanisms, with neuroinflammatory responses playing a significant role in both the onset and prognosis of the disease. Following an IS event, the deprivation of oxygen and glucose—essential nutrients in the blood—results in the cessation of their supply to tissues. This leads to the inactivation of ion pumps and related channels, the release of excitatory neurotransmitters, and the initiation of oxidative stress reactions. Concurrently, extensive necrosis of cells and tissues occurs in the ischemic area. The necrotic cells and tissues, along with persistent oxidative stress reactions, induce the release of a substantial amount of DAMPs (Petrovic-Djergovic et al., 2016). High-mobility group box 1 (HMGB1) is classified as a damage-associated DAMPs, and its various oxidized forms are rapidly released from necrotic cells located in the ischemic core, thereby activating immune cells and initiating inflammatory responses. Additionally, HMGB1 can undergo oxidative modifications in the circulation, which further activates peripheral immune cells (Singh et al., 2016b). The release of substantial amounts of DAMPs triggers innate immune responses, including the activation of microglia and astrocytes, leading to cascading intravascular inflammatory reactions. While innate immune cells in the brain work to clear necrotic tissue, they concurrently release a significant quantity of inflammatory factors. Furthermore, the increased permeability of the blood-brain barrier, combined with the activity of peripheral immune cells, results in excessive immune activation at the lesion site, thereby exacerbating brain injury.

Neuroinflammation is initiated not only by innate immune cells but also by immune cells that infiltrate from the peripheral immune system. Following IS, neutrophils and monocytes are the first responders, entering the brain within 12 hours, followed by T cells and B cells, which infiltrate the brain hours to days after IS (DeLong et al., 2022). Within hours of the onset of IS, the peripheral neutrophil count increases exponentially, whereas the lymphocyte count exhibits a contrasting trend (Gill et al., 2018). In the field of acute ischemic stroke (AIS), the neutrophil-to-lymphocyte ratio has been recognized as a significant biomarker for patient prognosis. It predicts early neurological improvement following thrombolysis in patients with AIS and is associated with early neurological deterioration (Gong et al., 2021; Sharma et al., 2021). Following a stroke, monocytes migrate from peripheral tissues to the site of injury, guided by a gradient of secreted cytokines and chemokines, such as CCL2/CCR2, CX3CL1/CX3CR1, and MMP-2/9. At the injury site, these monocytes differentiate into macrophages, contributing to the inflammatory response and tissue repair (Kim and Cho, 2016). Furthermore, both the monocyte-to-high-density lipoprotein cholesterol ratio and the monocyte-to-lymphocyte ratio serve as predictive markers for IS risk. The combined application of these ratios provides more accurate diagnostic results (Liu et al., 2020a). T cells are recruited to the brain by inflammatory mediators, particularly the CD4+ Th1 and Th17 subtypes, which exacerbate the severity and outcomes of stroke. In contrast, regulatory T cells (Tregs) are beneficial for stroke recovery and are associated with improved outcomes following severe strokes (Endres et al., 2022). B cells play a critical role in modulating the inflammatory response during the acute phase of IS and are involved in the recruitment of immune cells (Wu et al., 2024). As the disease progresses into the subacute and chronic phases of IS, the function of B cells shifts from regulating inflammation to participating in tissue repair and the long-term modulation of the immune response (Malone et al., 2023). Following a stroke, both the innate and adaptive immune systems contribute to the cascade of brain injury, engaging in complex and interdependent roles that enhance the inflammatory response, clear necrotic tissue, and protect brain cells.

The microbiota represents a complex and diverse ecosystem within the human body. The primary sites of microbial colonization include the oral cavity, skin, respiratory tract, urogenital tract, eyes, and gastrointestinal tract (Cryan et al., 2019). Among these sites, the gut harbors the largest microbial population (Yuan et al., 2021). The gut microbiota comprises yeasts, archaea, parasites, viruses, and protozoa, with bacterial populations being the most predominant (Honarpisheh et al., 2022). These microorganisms play a crucial role in regulating host inflammatory immunity (Levy et al., 2017) and promoting intestinal barrier function (Nie et al., 2021), thereby serving as vital contributors to the maintenance of human health.

The diversity of microbial species increases from childhood to adulthood and subsequently declines in old age (Hou et al., 2022). In healthy adults, the fecal microbiota is predominantly composed of Proteobacteria, Bacillota, Actinobacteria, and Bacteroidota, which collectively account for over 90% of the gut microbiome, while the abundance of Fusobacteria and Verrucomicrobia remains relatively low. The reduction in anaerobic bacteria and Bifidobacteria is associated with the worsening of inflammatory conditions, as these bacteria play a crucial role in stimulating the immune system (Rinninella et al., 2019). With the increase in the size of cerebral infarction, Bacteroidota tend to overproliferate, whereas Bacillota and Actinobacteria show a decreasing trend. Certain members of Bacteroidota, such as the genus Bacteroides, may exacerbate neuroinflammation by promoting the polarization of pro-inflammatory Th17 cells. Bacillota include various genera that produce SCFAs, such as Lactobacillus, while Bifidobacterium within Actinobacteria exhibits immunomodulatory functions. A reduction in their populations may lead to weakened anti-inflammatory signals and compromised integrity of the intestinal barrier (Singh et al., 2016a). Disruption of gut microbiota diversity can adversely affect the immune system and increase intestinal permeability, resulting in enhanced infiltration of peripheral immune cells, including T cells and monocytes, into the brain following a stroke. This infiltration further amplifies neuroinflammatory responses and exacerbates brain tissue damage (Ma et al., 2024a).

TMAO is a metabolite produced by gut microbiota, primarily derived from dietary choline and L-carnitine through the action of intestinal microorganisms. It is transported to the liver via the portal vein, where it is oxidized into TMAO by flavin monooxygenase 3. The impact of TMAO on neuroinflammation occurs through multiple pathways. On one hand, TMAO may exacerbate neuroinflammation by influencing the activation state of microglia and the phenotypic transformation of monocytes and astrocytes, rendering them more susceptible to a pro-inflammatory phenotype, which leads to the secretion of elevated levels of inflammatory mediators such as IL-6 and MMP-9 (Haghikia et al., 2018; Su et al., 2021; Tu and Xia, 2024). On the other hand, TMAO may exacerbate neuroinflammation by increasing oxidative stress. TMAO inhibits mitochondrial function and activates oxidase systems, which leads to increased electron leakage and the production of superoxide anions. This process further promotes the generation of reactive oxygen species (ROS) (Ke et al., 2018; Chen et al., 2019). These ROS can damage intracellular macromolecules, including lipids, proteins, and DNA, thereby triggering cellular inflammatory responses. Furthermore, TMAO may influence the activity of antioxidant enzymes, such as superoxide dismutase, catalase, and glutathione peroxidase (Li et al., 2017; Brunt et al., 2020). Such influence can weaken the cellular antioxidant defense capacity, leading to an additional increase in ROS levels and exacerbating oxidative stress and neuroinflammation. Additionally, TMAO levels are positively correlated with the expression of various pro-inflammatory cytokines, such as IL-6 and TNF-α, which play a crucial role in post-stroke neuroinflammation. Their increase can further activate immune cells, amplify the inflammatory response, and result in aggravated brain tissue damage (Zhu et al., 2021).

Lipopolysaccharide (LPS), a component of the outer membrane of Gram-negative bacteria, is commonly referred to as an endotoxin. In healthy individuals, the gut microbiota serves as the primary source of LPS in the human body. Following a stroke, LPS can enter the bloodstream by disrupting intestinal barrier function, subsequently activating the body’s immune system. It possesses the ability to cross the blood-brain barrier (BBB) and bind to Toll-like receptor 4 (TLR4) on the surfaces of microglia and astrocytes in the brain. This interaction triggers a cascade of signaling pathways, including the nuclear factor-κB (NF-κB) pathway, which leads to the production of pro-inflammatory cytokines, activation of the innate immune response, and further exacerbation of the inflammatory response within the central nervous system (Fenton and Golenbock, 1998; Chidambaram et al., 2022). Nuclear factor erythroid 2-related factor 2 (Nrf2) is a pivotal transcription factor that serves as the guardian of redox homeostasis and is considered a promising therapeutic target for the treatment of stroke and inflammation-related diseases (Zhang et al., 2017). LPS can diminish the nuclear accumulation of Nrf2, leading to a decrease in its DNA-binding activity and a subsequent reduction in the expression of antioxidant enzymes, such as heme oxygenase-1 and NAD(P)H: quinone oxidoreductase 1. This decline results in a diminished antioxidant capacity in cells, which further exacerbates LPS-induced oxidative stress and inflammatory responses (Liao et al., 2020; Yang and Chen, 2021). Furthermore, LPS induces the production of ROS in microglia. ROS not only directly causes cellular damage but also triggers the activation of both brain-resident (microglia) and peripheral (leukocyte) immune pathways, leading to the release of additional inflammatory mediators and effector molecules. This inflammatory response may impair the reuptake of the neurotransmitter glutamate, resulting in elevated concentrations of glutamate in the synaptic cleft. Such elevations can lead to excitotoxicity, which damages neuronal structure and function, thereby creating a vicious cycle (Liao et al., 2020; Xin et al., 2023).

The neural pathways illustrate how the brain influences intestinal function through both direct and indirect routes, with the autonomic nervous system (ANS) serving as the primary conduit for the complex bidirectional communication of the MGBA, encompassing both positive and negative feedback mechanisms. The ANS is divided into two components: the sympathetic nervous system (SNS) and the parasympathetic nervous system (PNS). The SNS plays a crucial role in regulating the integrity, permeability, and immune response of the intestinal barrier by modulating the secretion of intestinal mucus, the expression of tight junction proteins in epithelial cells, and the recruitment and activation of immune cells (Tsukamoto et al., 2007; Mallesh et al., 2022; Ten Hove et al., 2023).

The vagus nerve (VN) is a major component of the PNS, regulating intestinal functions, the composition and activity of the microbiota, and facilitating bidirectional communication between the gut and the brain. It plays a central role in the neural pathways of the gut-brain axis (Fülling et al., 2019). The VN transmits information between the brain and the gut through various mechanisms. On one hand, it conveys signals from the brain to the gut via efferent nerve fibers, regulating intestinal motility, secretion, and sensation. On the other hand, it detects metabolites produced by gut microbiota and hormones released by enteroendocrine cells through afferent nerve fibers, transmitting these signals to the central nervous system, thereby influencing autonomic neural integration and adaptive responses in the brain. Following a stroke, damage or dysfunction of the vagus nerve may disrupt the bidirectional communication network between the gut and the brain. The experiment demonstrated that common carotid artery occlusion and vagotomy alter intestinal morphology and microbiome composition, resulting in an increased abundance of Bifidobacterium while decreasing the abundance of Akkermansia and Ruminococcus. Concurrently, these interventions were associated with increased hippocampal neuronal cell death and a higher severity score of neurological functions, alongside changes in hippocampal lipid profiles, elevated lipid peroxidation, and altered expression of inflammatory cytokines. These findings suggest that the vagus nerve plays a crucial role in modulating neuroinflammation, neuronal damage, and the dynamic equilibrium of the gut microbiome following a stroke (Zhang et al., 2024). Electroacupuncture stimulation of the left vagus nerve in rats with cerebral infarction reperfusion significantly increased both the total movement distance and average speed in the open field test. Furthermore, it reduced Evans blue leakage and the fluorescence intensity of FITC-dextran, decreased microglial polarization, and improved the infiltration of inflammatory cells while preserving crypts in colon tissue, resulting in more intact tight junctions between colon cells (Wang et al., 2024). This study indicates that vagus nerve stimulation can influence pathological processes such as neuroinflammation, blood-brain barrier disruption, and intestinal barrier damage in the early stages following a stroke. Consequently, it provides an opportunity for early intervention that may mitigate acute injury from stroke, reduce disability rates, and lower mortality. Current clinical studies have confirmed the enhancing effect of VN on functional recovery following a stroke. This is evidenced by a randomized controlled trial that examined the efficacy of VNS combined with rehabilitation for upper limb motor function in patients with ischemic stroke (Dawson et al., 2021). The integration of vagus nerve stimulation with existing stroke treatment modalities, such as pharmacotherapy and rehabilitation training, results in a synergistic effect that promotes neurological recovery and enhances the quality of life for patients.

Additionally, stroke stress activates the hypothalamic-pituitary-adrenal (HPA) axis, resulting in the release of corticotropin-releasing hormone and cortisol. This process alters intestinal permeability and the composition of gut microbiota. Conversely, gut microbiota metabolites, such as tryptophan derivatives, can feedback to regulate HPA axis activity (Zhou et al., 2023). Within the gut-brain axis, cortisol modulates the activity of serotonin synthase in intestinal neurons, thereby influencing serotonin levels in the gut. Serotonin, in turn, can feedback to regulate HPA axis activity, creating a complex network of interactions.

The immune pathway represents a crucial mechanism through which the MGBA connects the gut microbiota to the brain. This pathway primarily regulates the activation of gut immune cells, the migration of these cells across systems, and the production of microbial metabolites. Following a stroke, immune cells, including dendritic cells and macrophages, detect pathogens and damage signals within the gut, thereby activating local immune responses through the secretion of pro-inflammatory cytokines, such as IL-6, IL-1β, and TNF-α (Honarpisheh et al., 2022). Additionally, these cells can migrate to the mesenteric lymph nodes and other immune organs, where they activate T cells and B cells to elicit specific immune responses (Cheng et al., 2023). γδ T cells and neutrophils within the gut have the capacity to migrate to the brain, thereby exacerbating inflammatory damage in this organ (Brea et al., 2021). Furthermore, alterations in gut microbiota and their metabolites may influence the differentiation and functionality of immune cells. For instance, SCFAs can inhibit the activity of histone deacetylases, enhance the expression of anti-inflammatory cytokines, and suppress the production of pro-inflammatory cytokines, thereby exerting immunomodulatory effects (Rooks and Garrett, 2016).

Metabolic pathways play a central role in the MGBA by generating microbial metabolites and regulating host physiological functions. These pathways significantly impact neural signal transmission, immune regulation, and energy homeostasis. They primarily operate through the systemic circulation of bioactive metabolites, cerebrospinal fluid circulation, the lymphatic system, and the increased permeability of the BBB. In terms of the circulation of metabolic products in body fluids, metabolites produced by gut microbiota, such as SCFAs, bile acids, and tryptophan metabolites, can enter the brain via the bloodstream and exert direct effects on brain cells. For instance, SCFAs can bind to G protein-coupled receptors in the brain, inhibiting the production of pro-inflammatory cytokines (Kimura et al., 2020). Furthermore, cerebrospinal fluid can exchange substances with the intestines through the lymphatic system, transporting the metabolic products of gut microbiota and inflammatory signals into the brain. Following a stroke, this function becomes impaired, negatively impacting substance exchange and immune regulation between the gut and the brain (Mestre et al., 2020; Chen et al., 2021; Shlobin et al., 2024). Besides, following a stroke, the permeability of the BBB increases, facilitating the entry of microbial metabolites, cytokines, and immune cells from the bloodstream into the brain parenchyma. Concurrently, the disruption of the BBB impairs normal communication between the brain and the gut, thereby disturbing the balance of the brain-gut axis (Wang et al., 2024).

All experimental models of ischemic stroke studied to date, including the intraluminal suture middle cerebral artery occlusion model, photochemically induced model, thromboembolic model, endothelin-1 induced model, embolic microsphere model, and genetically engineered models, have consistently demonstrated alterations in gut microbiota (Kaneko et al., 1985; Miyake et al., 1993; Chen et al., 2015; Sommer, 2017; Kuo et al., 2021; Tuo et al., 2022). Significant differences in gut microbiota exist between young and aged mice, with the Bacillota to Bacteroidota ratio being approximately nine times higher in aged mice compared to their younger counterparts. Following MCAO in mice, the gut microbiota of young mice resembles that of uninjured aged mice, evidenced by a more than threefold increase in the Bacillota to Bacteroidota ratio, alongside a reduction in the expression of intestinal tight junction protein and elevated plasma endotoxin levels. In aged mice, the Bacillota to Bacteroidota ratio increases by approximately 40%, accompanied by a marked decrease in the abundance of Akkermansia and an increase in conditional pathogens, such as Proteobacteria (e.g., Enterobacteriaceae) (Spychala et al., 2018). This indicates that stroke itself exacerbates dysbiosis of the gut microbiota, irrespective of the host’s age. Following a stroke, the abundance of beneficial bacteria, such as Bifidobacterium and Lactobacillus, significantly decreases, while the quantity of harmful bacteria associated with inflammatory responses and immune dysregulation, including Proteobacteria and Enterobacteriaceae, markedly increases. After a stroke, intestinal barrier function is compromised, and the levels of SCFAs are significantly reduced, leading to the translocation of endotoxins (e.g., LPS) into the bloodstream. This process may exacerbate intestinal inflammation and subsequently trigger systemic inflammatory responses. The use of antibiotics significantly impacts gut microbiota following a stroke. The widespread administration of antibiotics markedly reduces the diversity of gut microbiota, leading to a decline in the abundance of beneficial bacteria, including Lactobacillus and Ruminococcus. In contrast, conditionally pathogenic bacteria, such as Bacteroides, are not effectively suppressed (Wang et al., 2022). Relevant studies indicate that antibiotic-induced dysbiosis may persist for several weeks, impairing the normal functioning of the MGBA and delaying recovery post-stroke (Spychala et al., 2018). Conversely, other studies have reported that antibiotic treatment, whether administered alone or in combination, can reduce the volume of cerebral infarction (Benakis et al., 2020), as detailed in section 4.4.

To further understand the differences in gut microbiota between stroke patients and healthy controls, we identified the 15 most recent clinical studies that compare the microbiota of stroke patients with that of healthy controls ( Table 1 ). The selection of these articles was guided by the PRISMA statement, which assesses the quality of the included literature. Detailed search strategies and flow diagrams used in each database are provided in the supplementary materials.

(1) Most studies utilized 16S rRNA gene sequencing, while a minority employed metagenomic sequencing, focusing exclusively on the bacterial components of the gut microbiota. (2) The research involved teams from various countries and regions, including China, Italy, Brazil, South Korea, and the Netherlands, reflecting a global interest in the relationship between gut microbiota and stroke. Dietary habits, environmental factors, and other variables across different regions may influence both the gut microbiota and the incidence of stroke.(3)Each study employed distinct grouping methods tailored to the research objectives and the specific conditions of the patients. Some studies categorized patients based on the type of stroke, such as large artery atherosclerosis or cardioembolic stroke, while others classified them according to the severity of the condition, including mild, moderate, and severe stroke. Additionally, some studies considered coagulation function, differentiating between hypercoagulable states and normal coagulation function, or the time of onset, distinguishing between the subacute phase and the recovery phase. Furthermore, certain studies concentrated on specific types of stroke, including cryptogenic stroke and lacunar infarction. These varied grouping approaches enable a more nuanced investigation of the role of gut microbiota across different types and stages of stroke. (4)Conversely, the differences in α-diversity between stroke patients and healthy controls exhibited variability across studies. This inconsistency may be attributable to several factors, including the characteristics of the study subjects, sample size, study design, and the type and severity of stroke.

Overall, Parabacteroides has been referenced in numerous studies, demonstrating a significant increase in abundance among patients with AIS and those suffering from large artery atherosclerosis (LAA) stroke. Similarly, Alistipes has also shown increased abundance in research involving AIS and IS patients. Streptococcus is more abundant in the gut microbiota of patients with IS and cryptogenic stroke (CS). Klebsiella exhibits an increasing trend in abundance among patients with LAA-type stroke and some patients with AIS. Lactobacillus shows increased abundance in stroke patients, particularly in those with specific types of stroke, such as lacunar infarction, or during particular stages, such as the recovery phase. The abundance of Escherichia-Shigella is elevated in patients with CS and some individuals with IS. Research has demonstrated that the prevalence of Enterobacteriaceae increases in patients diagnosed with CS. Conversely, the abundance of Prevotella significantly declines in patients with AIS, particularly those with LAA type stroke, as well as in individuals with IS. The abundance of Bacteroides is reduced in the gut microbiota of patients with AIS and CS. Faecalibacterium shows a declining trend in patients with CS and some individuals with IS. Moreover, Roseburia displays a reduced abundance in stroke patients, including those with AIS. Research has also indicated a decrease in the abundance of Clostridiales among patients with IS. Additionally, Lachnospiraceae exhibits decreased abundance in stroke patients, particularly among certain AIS patients. Finally, Akkermansia is found to be less abundant in specific stroke patients, including those with IS.