Acute ischemic stroke (AIS) was a leading cause of death and chronic disability worldwide. Stroke survivors frequently had various complications, such as cognitive impairment and physical disability, which had a great impact on the quality of life (Duncan et al., 2021; Paul and Candelario-Jalil, 2021). Recent studies have shown that some risk factors including age, smoking and hyperlipidemia could affect the functional outcome after stroke (Meschia and Brott, 2018; Diener and Hankey, 2020). Hyperlipidemia could result in the neuroinflammation of brain and aggravated ischemic brain injury (Kim et al., 2014), and half of stroke patients were found to have hyperlipidemia (Rother et al., 2008). Hyperlipidemic stroke patients might suffer from functional deterioration after AIS. Kim et al. reported that the elevated plasma cholesterol levels were positively correlated with stroke severity in the hyperlipidemic mice (Kim et al., 2020). Elevated low-density lipoprotein cholesterol (LDL-C) was independently associated with severe stroke in patients with chronic kidney disease (Zhang et al., 2021). Currently, early detection of poor outcome of AIS with hyperlipidemia (POAH) was often challenging. Therefore, it is very urgent to find early biomarkers to evaluate the prognosis of hyperlipidemic stroke patients.

Recent studies have emphasized that the characteristic gut microbiota (GM) are associated with AIS. It was reported that stroke patients showed significant dysbiosis of bacteria with enriched short-chain fatty acids (SCFAs) (Li et al., 2019). Our previous studies showed that Proteobacteria was highly increased in the post-stroke cognitive impairment patients compared with the post-stroke noncognitive impairment patients (Ling et al., 2020). More and more evidence showed that GM have important influences on the occurrence, development and severity of stroke. Zhu et al. reported that GM directly impact cerebral infarct size and adverse outcomes following stroke through GM-derived metabolite trimethylamine-N-oxide (Zhu et al., 2021). GM have been increasingly recognized as vital determinants involved in the development of stroke and hyperlipidemia (Ling et al., 2022). The patients with hyperlipidemia showed abnormal GM composition (Gargari et al., 2018), which would aggravate dyslipidemia (Deng et al., 2019; Gu et al., 2020), while regulating GM could alleviate the abnormality of serum lipid in animal models (Yan et al., 2022). These findings demonstrated that GM might be an important regulator of the prognosis of hyperlipidemic stroke patients.

Recent evidences demonstrate that GM could be regarded as a diagnosis biomarker for many diseases. Our previous studies showed that patients with post-stroke comorbid cognitive impairment and depression exhibited an increased abundance of Proteobacteria, and a decreased abundance of several SCFAs-producing bacteria (Ling et al., 2020). It was reported that the abundance of Alcaligenaceae and Acinetobacter could remarkably distinguish autism spectrum disorders from the healthy group (Li et al., 2019). GM could distinguish stroke patients from healthy controls and the level of SCFAs appeared to effectively predict the severity and prognosis of stroke to some extent (Sun et al., 2021; Tan et al., 2021). The increased relative abundance of Finegoldia magna, Bifidobacterium dentium, and Clostridium clostridioforme could be used as a predictor of aging (Chen et al., 2022). Although the diagnostic application of GM has been well studied, the characteristic microbiota in POAH patients remains unclear.

Therefore, the present study was performed to investigate the characteristic GM of POAH patients. We further confirmed the correlation between characteristic GM and clinical parameters, as well as determined the gut microbial signature to predict POAH.

This study revealed that GM feature of POAH was that the abundance of Enterococcus increased while the abundance of bacteria producing SCFAs decreased, which was closely related to independent risk factors, such as cerebrovascular history, NIHSS score, and MAP. Moreover, the characteristic microbiota and microbiota plus with the three independent risk factors could establish a distinction for predicting POAH. These results indicated that GM might provide novel microbial biomarkers for predicting POAH.

Our results showed that the composition and structure of microbiota were different between POAH and GOAH. Previous studies revealed that gut microbial communities in the group with adverse prognosis after stroke were distinct from those in the group with good prognosis, accompanied by an increase in the abundance of Bacteroidota, and Actinobacteriota, and the decreased abundance of Proteobacteria and the Bacteroidetes to Firmicutes ratio (B/F) (Benakis et al., 2016; Singh et al., 2016; Shimizu et al., 2019; Guo et al., 2021). The diversity of GM was affected by many factors, such as lipid homeostasis (Schoeler and Caesar, 2019). The decreased B/F induced dyslipidemia, leading to more severe outcomes, such as obesity and liver steatosis (Hussain et al., 2020). Our results showed that the abundance of Enterococcus in POAH was enriched, and positively related to the mRS score, indicating that the abundance of Enterococcus might be related to the risk of POAH. It was reported that Enterococcus was an opportunistic pathogen in the gastrointestinal tract, and the risen level of Enterococcus was relevant to many neurological and metabolic diseases, such as Parkinson’s disease, Alzheimer’s disease and diabetes (Underly et al., 2015; Li et al., 2017). Enterococcus appeared in subjects of the adverse outcome group, manifested as the post-stroke cognitive impairment (PSCI) and post-stroke affective disorder (Huang et al., 2021), which was consistent with our studies. Enterococcus could induce the secretion of proinflammatory cytokines, such as IL-6 (García-Solache and Rice, 2019), and further contribute to systemic inflammation (Stanley et al., 2016; Chen et al., 2019), which led to POAH (Suda et al., 2018). Evidence showed that Enterococcus faecalis disturbed the lipid metabolism (Huang et al., 2018; Zhu et al., 2021). Hu X et al. revealed that a higher abundance of Enterococcus had a closely related to poor prognosis of hypertriglyceridemia-related acute pancreatitis leading to poor prognosis in hypertriglyceridemia patients (Hu et al., 2021), suggesting that Enterococcus might be involved in the prognosis of hyperlipidemic stroke patients.

In this study, there was a significantly lower relative abundance of SCFAs-producing bacteria in POAH group, such as Lachnospiraceae, Faecalibacterium, Rothia and Butyricicoccus. Moreover, Lachnospiraceae, Faecalibacterium, and Butyricicoccus were associated with lower mRS score. Lachnospiraceae, a primary producer of butyrate, was related to the functional prognosis of diseases (Sorbara et al., 2020). Many studies showed that the abundance of Lachnospiraceae was significantly decreased in stroke patients and animal models (Zeng et al., 2019; Lin et al., 2021). The abundance of Lachnospiraceae in patients with post stroke cognitive impairment (Ling et al., 2020) and patients with nervous neurocritical illness (Xu et al., 2019) was less. In addition, lower blood lipid could increase the abundance of Lachnospiraceae and levels of SCFAs in hyperlipidemia model animals (Gui et al., 2019; Liu et al., 2021). In addition, our results showed that the relative abundance of Faecalibacterium in POAH group was significantly lower. Faecalibacterium is a butyrate-producing bacteria, belonging to Lachnospiraceae family. Previous studies showed that the relative abundance of Faecalibacterium had a lower relative abundance in patients with stroke (Silveira-Nunes et al., 2020), transient ischemic attack (Yin et al., 2015) and PSCI (Huang et al., 2021) was lower. Lee et al. reported that Faecalibacterium prausnitzii ameliorated post-stroke neurological deficits and elevated concentrations of intestinal SCFAs in aged mice with stroke (Lee et al., 2020). Faecalibacterium prausnitzii was decreased in fecal samples of hyperlipidemia adolescents (Gargari et al., 2018), and the abundance of Faecalibacterium prausnitzii in patients with mild hypercholesterolemia was significantly negatively correlated with TC and LDL (Xu et al., 2021). Furthermore, Faecalibacterium was observably elevated in the hyperlipidemia rats after probiotic intake, which could prevent the progression of hyperlipidemia (Shao et al., 2017). Enriched Faecalibacterium could reverse the increase of plasma TG level (Tong et al., 2018), and was positively correlated with plasma concentrations of butyric acid (Khan et al., 2018). Butyricicoccus, a butyrate-producing clostridial cluster genus, was related to reduced incidence of hyperlipidemia or hypercholesteremia in patients with colorectal cancer (Han et al., 2019). The abundance of Butyricicoccus was negatively correlated with the serum levels of LDL, TG and TC of obese patients, which could be used as a biomarker to predict obesity related lipid metabolism abnormalities (Zeng et al., 2019). Recent multiple studies have shown that SCFAs were closely linked to stroke and dyslipidemia. AIS patients, especially those with more severe stroke (Ling et al., 2020), showed a lack of SCFAs-producing bacteria and decreased levels of fecal SCFAs levels, which led to increased risks of post-stroke infection (Haak et al., 2021) and poor functional outcomes (Tan et al., 2021). Furthermore, the feces of young rats transplantation could effectively increase the concentration of SCFAs, and attenuate the neurological deficit and inflammation after stroke in elderly stroke mice (Lee et al., 2020) and in middle cerebral artery occlusion (MCAO) model rats (Chen et al., 2019). In addition, compared with control, subjects with hypercholesterolemia had a lower level of butyrate, which was negatively correlated with LDL (Granado-Serrano et al., 2019). SCFAs played an important role in reducing the risk of cholesterol and coronary heart disease, and valeric acid was negatively correlated with HDL-C in patients with mild hypercholesterolemia (Xu et al., 2021). These results indicated that decreased SCFAs-producing bacteria, such as Lachnospiraceae, Faecalibacterium, Rothia and Butyricicoccus and their metabolites SCFAs might participate in the occurrence of POAH.

Our results showed that the characteristic bacteria in POAH patients were closely related to independent risk factors, such as increased, decreased MAP, and history of cerebrovascular disease. The higher NIHSS scores, the greater the risk of disability, the more serious the neurological impairment, and the larger the area of ischemic lesions (Cucchiara et al., 2019; Cucchiara et al., 2020; Wang et al., 2021). A study showed that stroke patients with a history of hyperlipidemia were associated with a higher NIHSS score on day 7 and were less likely to have neurological improvements (Restrepo et al., 2009). Higher MAP could maintain cerebral perfusion and cerebral blood flow velocity in stroke patients. MAP was found to be positively associated with adverse functional outcomes and recurrence risk in stroke patients. It was reported that there was a positive correlation between MAP and the adverse functional outcome and recurrence risk of stroke patients (Ma et al., 2019). Moreover, GM also had a close connection to the clinical parameters. Our results showed that the decrease of unclassified_f_Lachnospiraceae was associated with the increase of NIHSS score, and MAP was positively correlated with the abundance of Faecalibacterium, unclassified_f_Lachnospiraceae, and Butyricicoccus, while negatively correlated with Enterococcus. LEfSe was used to support the construction of POAH diagnostic model based on five characteristic genera. In addition, the prediction model based on the combination of five characteristics and three independent risk factors could predict the occurrence of POAH. Therefore, these finding revealed the close relationship between POAH and GM, and the characteristic GM could be used as a biomarker for early prediction of POAH.

However, several limitations of this study should be mentioned. First, this was a small sample observational study conducted in a single center. Meanwhile, we collected fecal sample of patients at a single time point, so we could not observe the dynamic changes of the interaction between GM and these parameters. In addition, the information on the concentration of microbial metabolites, such as SCFAs, was lacked, which was difficult to find out the causal relationship between GM and POAH. Despite these limitations, our study firstly described the Characteristic GM of POAH, which was helpful to understand the role of microbial biomarkers in predicting POAH.

In conclusion, these findings revealed the microbial characteristics of POAH, which were closely related to clinical parameters. The characteristic GM might facilitate the diagnosis of POAH, which highlighted the potential prediction of GM on POAH.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI, PRJNA894329.

The protocol of the study was reviewed and approved by The Ethics Committee of the Second Affiliated Hospital of Wenzhou Medical University (LCKY2020-207). The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

JL, JS and SC designed the experiments. JC, BC, JM, JZ, QG, HX, YK and SY performed the experiments and conducted the statistical analyses. All authors contributed to the article and approved the submitted version.

This work was supported by Clinical Medical Research Project of Zhejiang Medical Association (2022ZYC-D10).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.