Among non-communicable disorders (NCDs), stroke remains a paramount global health challenge. According to the most recent Global Burden of Disease (GBD) 2021 stroke burden report (Feigin et al., 2024), stroke is the second leading cause of death (accounting for 7.3 million deaths) and the third leading cause of death and disability combined worldwide. The World Stroke Organization’s Global Stroke Fact Sheet 2025 further highlights the escalating burden, noting a 70.0% increase in incident strokes, 44.0% of deaths attributable to stroke, and an 86.0% rise in prevalent strokes. This report underscores that metabolic risks contribute to 69.0% of all strokes, while environmental and behavioral risks constitute 37.0 and 35.0%, respectively (Feigin et al., 2025).

In light of these staggering figures, research has increasingly focused on identifying nontraditional risk factors influencing stroke prognosis, including gastrointestinal (GI) disorders. Notably, GI disorders—particularly those of a functional, inflammatory, or infectious gastrointestinal disorders—have been significantly associated with an elevated risk of ischemic stroke (Roth et al., 2020).

Gastrointestinal bleeding (GIB) is a major complication in patients with acute ischemic stroke (AIS), strongly linked to increased mortality and reduced independence (Roth et al., 2020; Du et al., 2020; Zhou et al., 2019). Evidence demonstrates that GIB is independently elevates the risk of stroke recurrence. A large-scale study of 12,415 ischemic stroke patients revealed that GIB was independently associated with higher recurrence risks at 3, 6, and 12 months post-stroke (Du et al., 2020). GIB is classified into overt GIB (e.g., hematemesis, melena) and occult GIB (detected by fecal occult blood test or iron deficiency anemia). While prior research has predominantly focused on overt GIB and its impact on stroke outcomes and recurrence (Du et al., 2020; Fu, 2019; Aziz et al., 2024), identifying risk factors such as infection, male sex, advanced age, and higher NIHSS scores, proton pump inhibitors (PPIs) have proven effective in reducing the incidence and mortality of overt GIB in AIS patients (Taha et al., 2015).

In contrast, the risks and consequences of occult GIB in stroke outcomes remain poorly understood. Occult GIB is particularly challenging to detect due to its asymptomatic nature and may originate from regions of the GI tract less susceptible to acid suppression (Taha et al., 2015; Raju et al., 2007). The mechanisms linking AIS to occult GIB are likely multifactorial, involving elements such as antiplatelet therapy and stress-induced mucosal injury (Fu, 2019).

Crucially, emerging research implicates gut dysbiosis—a well-documented consequence of ischemic stroke (Yin et al., 2015)—as a potential key player in compromising gastrointestinal integrity. Stroke-induced dysbiosis, characterized by an expansion of pro-inflammatory pathobionts, can increase intestinal permeability, potentially predispose the mucosa to micro-bleeding. Conversely, the presence of intraluminal blood might exacerbate dysbiosis, creating a vicious cycle (Freedberg et al., 2018; Xia et al., 2021; Yin et al., 2015). Ischemic stroke has been shown to rapidly induce pathological alterations in the gut microbiome, as evidenced in both animal models and clinical studies (Yin et al., 2015; Xu et al., 2021). This dysbiosis can, in turn, exacerbate brain infarction through systemic inflammatory pathways, forming a deleterious feedback loop via the brain-gut axis (Xu et al., 2021). Furthermore, gut colonization with specific pathogens, such as Enterococcus, has been independently associated with adverse outcomes (including mortality) in critically ill populations (Freedberg et al., 2018; Stein-Thoeringer et al., 2025), and our prior work identified Enterococcus as a risk factor for stroke-associated pneumonia (Xia et al., 2021). A recent 2025 study reinforced the role of brain-gut axis dysregulation in post-stroke complications by demonstrating significant alterations in gut microbiota structure and metabolic profiles following acute stroke (Chen et al., 2024).

Although direct evidence establishing a causal or independent correlative link between occult GIB and gut dysbiosis specifically in ischemic stroke patients is still evolving, a compelling body of recent research strongly suggests a plausible and mechanistically supported pathophysiological connection via the brain-gut axis (Xu et al., 2021; Meng et al., 2023; Zhang et al., 2025). While previous studies, including our own, have established stroke-induced gut dysbiosis and its association with systemic complications (Stanley et al., 2016), the specific link between the abundance of gut pathogens and the occurrence of occult GIB during the acute phase of stroke—a common yet underdiagnosed complication—remains markedly understudied. Therefore, to address this significant gap in knowledge, the present study aims to: (1) investigate the association between occult GIB and long-term prognosis in AIS patients, including 90-day dependency and one-year major adverse cardiovascular events (MACEs, including death and recurrent ischemic stroke); (2) explore the association between gut potential pathogens (and their metabolites), and occult GIB; and (3) examine the potential link between increased gut potential pathogens and long-term outcomes.

This study diverges from previous research predominantly focusing on overt gastrointestinal bleeding (GIB) post-stroke, by examining the impacts of occult GIB on long-term outcomes in acute ischemic stroke (AIS) patients and revealing its association with an enrichment of specific gut pathogens. Our findings revealed that occult GIB is highly prevalent in 13.9% of AIS patients, significantly higher than overt GIB (6.2%). Compared to non-GIB patients, those with occult GIB exhibited higher rates of bacterial infections and a heightened abundance of specific gut pathogens, including Enterococcus, Staphylococcus, and Pseudomonas, which were identified as independent risk factors for occult GIB. Elevated gut permeability biomarkers, (LBP and LPS) and a more severe inflammatory profile (higher IL-23, lower IL-10) were observed in occult GIB patients, correlating with worse outcomes, including higher 90-day dependency and 1-year major adverse cardiovascular events (MACEs).

Gastrointestinal bleeding is a significant contributor to increased mortality in stroke patients, with much attention historically focused on the treatment and prevention of overt GIB. In our study, the prevalence of overt GIB was 6.2%, aligning with prior studies where GIB incidence ranged from 1.24 to 8.1% (Du et al., 2020). Notably, our multi-center cohort study highlighted a rising incidence of occult GIB in AIS patients, revealing distinct clinical characteristics and long-term outcomes compare to both overt GIB patients and non-GIB patients. Lower admission hemoglobin levels were observed in both GIB groups, consistent with previous findings (Taha et al., 2015). Patients with overt GIB exhibited higher admission WBC levels and were more likely to experience dysphagia, infections, and 90-day dependency compared to those with occult GIB or without GIB. However, at the one-year follow-up, patients with occult GIB had similar mortality and MACEs rates to those with overt GIB. Occult GIB post-stroke was identified as an independent risk factor for 90-day unfavorable outcomes and one-year MACEs after adjusting for traditional vascular risk factors in AIS patients. These findings highlights the need for vigilant management of occult GIB post-stroke and long-term care of these patients. While this study cannot establish a cause-effect relationship, these results have important clinical implications.

The underlying mechanism linking AIS and occult GIB remains unclear. Proposed mechanisms contributing to mucosal injury after ischemic stroke include antiplatelet use, stress, vagal hyperactivity (Fu, 2019) and disruption of the brain-gut axis, which may increase gastrointestinal mucosal injury risk. Some preventive strategy, including selective NSAIDs such as coxibs (Taha et al., 2015), or other antithrombotic drugs, appear less likely to cause mucosal damage in the lower gastrointestinal tract. However, managing occult GIB solely with acid inhibitors may be unrealistic, given the multiple etiological factors and the complex anatomical sites. These drugs are only effective for lesions in acid-susceptible areas such as the esophagus, stomach, or duodenum. Interestingly, researchers have found that the incidence of occult GIB was significantly correlated with the prescriptions of proton pump inhibitors (PPI) (Taha et al., 2015). Post-stroke alternations in the gut microbiome, particularly the enrichment of pathogenic species, may also contribute to occult GIB development. Recent research, including our previous studies (Yin et al., 2015; Xu et al., 2021; Xia et al., 2019), have demonstrated that ischemic stroke induces pathological alternations in the host gut microbiome, as evidenced in both animal and clinical studies. Increased pathogenic species and associated metabolites have been reported to increase the risk of ischemic stroke [e.g., atherosclerosis, obesity, diabetes mellitus, and hypertension (Boulangé et al., 2016; John and Mullin, 2016; Mazidi et al., 2016)] and post-stroke complications [e.g., pneumonia (Stanley et al., 2016; Stevens et al., 2022; Fu et al., 2024), and cognitive impairment (Olson et al., 2021; Pan et al., 2023; Wang et al., 2023)]. In our previous study, patients with stroke-associated pneumonia (SAP) exhibited increased gut microbiota dysbiosis in the acute stage, with Enterococcus identified as an independent risk factor of SAP (Xia et al., 2021). Increased abundance of gut pathogenic species, including Enterococcus, Pseudomonas, Staphylococcus, and Klebsiella, were positively associated with specific bacteria infection and even mortality, as confirmed by stool or swab culturesand 16S sequencing (Freedberg et al., 2018; Shimasaki et al., 2019). In this study, we observed that patients with occult GIB had a significantly higher risk of infection, with culture-proven bacteria events. We further found that certain pathogens, including Enterococcus, Pseudomonas, and Staphylococcus, were more prevalent in occult GIB patients during the early acute stage, both in the age-, dysphagia- and NIHSS- matched cohort, as well as in the entire AIS patient cohorts. Multivariate logistic regression analysis confirmed that these pathogens were independent risk factors for occult GIB in AIS patients. Our study identified specific pathogens (Enterococcus, Staphylococcus, Pseudomonas) that were independently associated with occult GIB. This association raises the possibility of a link between this specific pathogen enrichment and gastrointestinal mucosal integrity post-stroke, though the direction of this relationship requires further investigation. Furthermore, the co-occurrence of gut pathobionts and systemic infections suggests a shared underlying dysbiotic state or potential cross-talk, but our study design cannot confirm a direct gut origin for post-stroke infections. We also assessed the biomarkers of gut permeability and inflammation in patients. The systemic biomarker profile observed in the early acute phase—elevated LBP, LPS, and IL-23, and reduced IL-10—in occult GIB patients is indicative of a heightened systemic inflammatory state. While this pattern could be consistent with increased gut permeability and mucosal injury, it is crucial to note that these single early measurements may also reflect, and are likely confounded by, the acute systemic inflammatory response to the stroke itself. In terms of long-term outcomes, increased Enterococcus were independently associated with a higher risk of 90-day dependency while increased Staphylococcus was linked to one-year MACEs and recurrent stroke. However, the direct association between these biomarkers and occult GIB was not fully elucidated in this study. Targeting these enriched gut pathobionts and gut-derived metabolites may help to improve GIB management and enhance long-term outcomes.

Our study has limitations. First, we did not monitor longitudinal changes in gut pathogens and occult GIB over time, or explore the underling mechanisms linking gut pathogens to adverse outcomes, limiting our ability to established causality. Secondly, our analysis was focused on a pre-selected set of pathogens based on their clinical relevance. We did not perform α- and β-diversity analyses to characterize the overall gut microbial community structure. Therefore, our findings should be interpreted as demonstrating an enrichment of specific pathogens associated with occult GIB, rather than providing a complete picture of global gut dysbiosis. Future studies including comprehensive community-level analyses are warranted. Additionally, our sample size may limit the statistical power to detect certain associations. Furthermore, due to the cross-sectional nature of the microbiota assessment at admission, we cannot establish causality or rule out reverse causality—whether occult GIB (or its predisposing conditions) alters the gut environment promoting pathogen growth, or whether pre-existing or stroke-induced dysbiosis contributes to mucosal vulnerability leading to occult GIB. Further research should focus on elucidating these mechanisms and developing targeting interventions to manage gut pathogens and improve outcomes in AIS patients.

In this prospective multicenter study, occult GIB was prevalent in nearly 14% of AIS patients and was associated with significantly worse short- and long-term functional outcomes and survival. The enrichment of specific gut pathogens (Enterococcus, Staphylococcus, Pseudomonas) in the acute phase was independently associated with occult GIB. These findings highlight occult GIB as a clinically significant yet often overlooked complication of AIS. The association with this specific gut pathogen profile warrants further investigation to elucidate potential causal mechanisms and explore targeted interventions aimed at modulating the gut microbiome to improve stroke outcomes.