Interaction between gut microbiota and anesthesia: mechanism exploration and translation challenges focusing on the gut-brain-liver axis

Anesthetics can either directly or indirectly (via changes in factors such as pH, oxygen partial pressure, and mucus secretion) disrupt the gut microbiota, thus leading to a decrease in microbiota diversity, the proliferation of proinflammatory microbiota, and the development of metabolic disorders, which can correspondingly affect central nervous system function via immune, neurological, and endocrine pathways. Animal experiments have demonstrated that propofol exerts a certain antibacterial effect, and the specific mechanism may involve the inhibition of microbial growth by disrupting the integrity of the cell membrane. Moreover, the results of these experiments demonstrated that, at three hours after rats were injected with propofol, the abundances of Prevotella and Lactobacillus at the genus level decreased, with these bacteria exhibiting a recovery trend of increased abundances on the 14th day (Minerbi and Shen, 2022). The impact of inhaled anesthetics on the gut microbiota is relatively significant and may directly or indirectly cause alterations in the composition of the gut microbiota, such as by reducing the abundance of Lactobacilli, thereby affecting vagal nerve transmission signals (Liufu et al., 2020). This difference suggests that clinical interventions need to consider the effects of anesthesia duration, drug type, and host factors (such as age and underlying diseases) on the dynamics of microbiota recovery. Previous studies have demonstrated that after the administrations of multiple sevoflurane anesthetics, the abundance and alpha diversity (including measurements of the Shannon index) of the gut microbiota significantly decreased. Specifically, the abundance of beneficial bacteria like Akkermansia was significantly reduced, while the relative abundance of certain potentially pathogenic bacteria, such as Streptococcus, increased (Han et al., 2021; Han et al., 2024). This change also showed a recovery trend at 14 days (Han et al., 2021). Age also affects the sensitivity of the microbiota to sevoflurane. For example, repeated exposure to sevoflurane anesthesia during the neonatal period can significantly alter the gut microbiota of mice (including abnormal abundances of Streptococci and Spirochetes, among other bacteria) and is associated with cognitive impairment (Liu et al., 2021). The gut microbiota of the elderly population is inherently fragile, and anesthesia and surgery further exacerbate the imbalance of the microbiota, leading to an increased risk of postoperative delirium (Zhang Y. et al., 2023). Mechanistically, sevoflurane alters gut microbiota function by reducing the synthesis of secondary bile acids, such as deoxycholic acid (Zhao et al., 2024). Moreover, exposure to sevoflurane may lead to an increase in the microbial derivative trimethylamine oxide (TMAO), which in turn activates NLRP3 in microglia, stimulates the release of proinflammatory cytokines (such as IL-1β, IL-6, and TNF-α), and induces neuroinflammation (Zhao et al., 2019; Huang et al., 2021; Han et al., 2024). Isoflurane, as another commonly used inhaled anesthetic, can lead to a decrease in gut microbiota diversity. After 4 hours of exposure to isoflurane, the abundance of Firmicutes, Clostridia, Clostridiales, and Lachnospiraceae in the fecal microbiome of rats significantly increased, while the abundance of Bacteroidetes, Actinobacteria, Bacteroides, and Bacteroidaceae significantly decreased (Wang et al., 2019). Isoflurane also contributes to the accumulation of inflammatory cytokines (such as IL-1β and IL-6) in the central nervous system, which is closely associated with postoperative delirium symptoms, including behavioral and cognitive impairments (Jiang et al., 2019).

Opioid receptors (such as μ receptors) are widely expressed in intestinal epithelial cells and immune cells. Therefore, opioid drugs such as morphine can directly inhibit the secretion of antimicrobial peptides, alter intestinal pH and redox status, and selectively promote the proliferation of pathogenic bacteria while inhibiting beneficial bacteria (Thomas et al., 2022; Kolli et al., 2023). Numerous studies on rodents have shown that opioid drugs can cause rapid and significant dysbiosis of the microbiota. After discontinuation of medication, some microbial parameters may recover within several weeks or even longer, but complete recovery to baseline status is rare (Thomas et al., 2022). Mechanistically, opioid drugs impair intestinal barrier integrity by inhibiting intestinal peristalsis, reducing mucus secretion, and disrupting tight junction proteins (such as ZO-1 and occludin), thereby promoting the translocation of bacteria and their metabolites (including the entry of LPS into the bloodstream) (Zhang and Roy, 2021). Moreover, opioid drugs inhibit the production of SCFAs (such as butyrate and propionate) and suppress the signal of the FXR, thereby leading to the accumulation of primary bile acids and a decrease in secondary bile acids (Gicquelais et al., 2020). The accumulation of these harmful metabolites further exacerbates neuroinflammation, forming a vicious cycle. As a result, morphine and fentanyl can significantly reduce the alpha diversity of the gut microbiota (including reductions in the Shannon index and observed operational taxonomic units), which is a phenomenon observed in both human patients and animal models (Gicquelais et al., 2020). In terms of changes in the abundance of specific bacterial genera, Roseburia, which produces butyrate, decreased, and Bilophila, which participates in bile acid metabolism, decreased. And Bacteroides may become the dominant bacterial group, associated with various inflammatory responses (Gicquelais et al., 2020) (Table 1).

Intensive care unit (ICU) patients often experience dysbiosis due to long-term sedation therapy (such as continuous infusion of midazolam), which may exacerbate increased intestinal permeability and systemic inflammatory response. A prospective cohort study of 61 ICU patients found that the abundance of Bifidobacterium genus was significantly higher in surviving patients compared to those who died during hospitalization (Kang et al., 2017). Another prospective study included 577 premature infants admitted to the ICU, collecting gut microbiota and conducting cluster analysis through 16S ribosomal RNA gene sequencing. The results showed that Cluster 4 (driven by Enterococcus) and Cluster 5 (driven by Staphylococcus) were associated with lower average gestational age, while Cluster 3 (driven by Escherichia/Shigella) was associated with higher average gestational age (Rozé et al., 2020). These findings suggest that certain cluster analysis of microbial communities may be noninvasive biomarkers. Additionally, long-term sedation may lead to decreased gastrointestinal motility and insufficient intake of enteral nutrition. Low enteral nutrition is significantly correlated with high-risk microbial clustering (such as Cluster 5), thereby indicating that inadequate nutritional support may exacerbate microbial dysbiosis (Rozé et al., 2020). In addition, ICU patients commonly use broad-spectrum antibiotics and experience prolonged courses of treatment, which further exacerbates microbiota disorders (Iapichino et al., 2008). In summary, critically ill ICU patients exhibit significant dysbiosis of the gut microbiota. However, the potential role of microbial-targeted therapy measures has not yet been determined.

The relationship between the gut microbiota and pain has been a research hotspot in the biomedical field in recent years and involves the investigation of multiple types of pain (such as visceral pain, neuropathic pain, inflammatory pain, headache, and cancer pain, among other types) and complex regulatory mechanisms (Minerbi and Shen, 2022). With respect to postoperative pain, the gut microbiota can participate in the occurrence of pain by affecting signaling of the vagus nerve and the release of inflammatory factors (Guo et al., 2019; Lee et al., 2023). In addition, preoperative antibiotic-induced dysbiosis can increase visceral pain sensitivity, whereas the restoration of microbial balance can reverse pain (Lucarini et al., 2022). A previous study revealed that preoperative microbiota characteristics can predict the risk of chronic postoperative pain. The decrease in the abundance of certain bacteria in breast cancer patients before surgery may be related to the increase in pain sensitivity (Yao et al., 2022). Following fecal microbiota transplantation, recipient mice with a "chronic postoperative pain microbiota" showed a significant increase in mechanical hyperalgesia and a decrease in the expression of Ppar-γ and arginase-1 in the spinal cord (Yao et al., 2022). Drugs such as somatostatin may reduce the risk of pancreatic fistula related pain by regulating postoperative microbiota composition (such as via reductions in Enterobacteriaceae) (Li et al., 2020). In summary, the causal mechanism of the relationship between the microbiota and postoperative pain, as well as individualized intervention plans, still need to be further explored, especially regarding the validation of different surgical types and patient populations (Guo et al., 2019; Morreale et al., 2022).

SCFAs: The first potential application of SCFAs involves anesthesia assistance, as certain SCFAs (or their metabolites) may produce anesthetic-like effects by enhancing the inhibitory neurotransmitter system. For example, acetic acid can enhance the anesthetic effect of sevoflurane and reduce its required concentration, which may be mediated by adenosine A1 receptors (Carmichael et al., 1991). Moreover, isovaleric acid exhibits direct anesthetic activity in tadpole models. Although propionic acid and methyl malonic acid do not exhibit separate anesthetic effects, they can significantly reduce the half effective concentration of isoflurane, thereby suggesting that they may be used as anesthetic adjuvants (Weng et al., 2009). Additionally, β-hydroxybutyric acid (a ketone body possessing a similar structure to the structures of SCFAs) produces anesthetic effects by enhancing GABAA receptor function, which is similar to the state of consciousness inhibition observed in ketotic patient (Yang et al., 2007). The second potential application involves assistance in postoperative recovery. Research has shown that supplementation with SCFAs (such as the regulation of the microbiota via the diet or electroacupuncture) may reduce postanesthesia neuroinflammation and cognitive impairment, especially in elderly or high-risk patients (Xu et al., 2021; Ke et al., 2022).

Other metabolic pathways: The associations between bile acid levels and anesthesia treatment mainly involve drug metabolism, enhanced treatment efficacy, and alternative treatment strategies. For patients with bile acid secretion disorders (such as progressive familial intrahepatic cholestasis) or liver injury, general anesthesia may exacerbate metabolic burdens, and the careful selection of anesthesia regimens is necessary (Clifton et al., 2011). Cholic acid and its derivatives have been found to enhance the local anesthetic effect of lidocaine (Posa et al., 2007). In addition, for high-risk patients who cannot tolerate anesthesia or surgery, piezoelectric lithotripsy combined with oral bile acids (such as ursodeoxycholic acid) is a safe and effective option. Research has shown that 69% of patients with single stones can achieve complete gallbladder clearance after lithotripsy without anesthesia (Pelletier et al., 1991). The supplementation of tryptophan metabolites can also represent a target for anesthesia intervention. For example, electroacupuncture treatment restores gut-brain axis function and reduces postoperative cognitive impairment and neuroinflammation by increasing the number of IPA-producing bacteria and their metabolites (Tang et al., 2024). Furthermore, the direct supplementation of IPA or the inhibition of key enzymes in the canine urinary ammonium pathway (such as IDO1) may become a new strategy for preventing and treating postoperative delirium (Zhou et al., 2023).

Antibiotics and probiotics are both ways that can directly alter the composition of the gut microbiota. Although antibiotics aim to prevent infections, they may increase the risk of complications. Prophylactic antibiotics during the perioperative period are one of the main factors that induce allergic reactions during anesthesia (Yasuda et al., 2021). Some antibiotics also have neurotoxic potential. Beta-lactam antibiotics (especially cephalosporins and carbapenems) are strongly associated with seizures and encephalopathy. Fluoroquinolones and macrolides can cause psychosis, insomnia, and neuropathy (Radkowski et al., 2025). This neurotoxicity, combined with the neuroprotective effect of anesthetic drugs themselves, may exacerbate the neurological damage in patients. Therefore, clinical decision-making needs to balance allergy history, surgical type, and anesthesia duration, and pay attention to the compatibility of medication timing with guidelines.

The preoperative or postoperative use of probiotics may alleviate anesthesia and surgery-related inflammatory reactions by regulating the balance of the gut microbiota. For example, in cardiac surgery, the combination of selective gastrointestinal decontamination, probiotics, and montmorillonite pretreatment can significantly reduce postoperative endotoxin and IL-6 levels and alleviate systemic inflammatory responses (Liu et al., 2018). In addition, probiotics may reduce the risks of postoperative infections and neurocognitive complications by maintaining the integrity of the intestinal barrier, thereby reducing the overgrowth of pathogenic bacteria and the release of metabolites such as endotoxins (Zanza et al., 2022). Adding specific beneficial bacteria (such as Lactobacilli and Bifidobacteria) has been found to improve cognitive function by regulating neuroinflammation and oxidative stress response (Liu et al., 2022; Tian et al., 2024). Currently, the auxiliary effect of probiotics on anesthesia is still controversial in human research. Different strains, doses, and timings of administration can also affect treatment efficacy. Therefore, it is necessary to combine individualized strategies (such as the combination of prebiotics and probiotics) to achieve optimal results.

FMT is mainly used to treat recurrent Clostridium difficile infections by transplanting functional microbiota from healthy donor feces into the patient's gut to restore gut microbiota balance (Khoruts et al., 2021). Currently, only preliminary studies have been performed on FMT-assisted anesthesia therapy. Animal studies have demonstrated that the correction of anesthesia-induced microbiota abnormalities via FMT may improve cognitive function. After the transplantation of healthy microbiota, cognitive impairment in young mice was observed to be alleviated, thereby suggesting that FMT may be a treatment for anesthesia-related neurotoxicity (Liu et al., 2021). The long-term use of antibiotics leading to gut microbiota dysbiosis (such as antibiotic-associated diarrhea) is a common problem observed in ICU patients. A prospective, multicenter, randomized controlled trial (NCT05430269) including 36 ICU patients with antibiotic-associated diarrhea revealed that FMT can accelerate diarrhea relief by restoring gut microbiota diversity (Cibulkova et al., 2024). However, patients under anesthesia or sedation should be given increased attention regarding operational safety when undergoing FMT. At the technical level, the technique involving the insertion of the middle section of the naso-intestinal tube has been proven to be safe and reliable. In addition, when ICU patients exhibit weakened immune function, FMT may cause bacterial translocation or systemic inflammatory reactions. Moreover, it is necessary to strictly screen donor feces and perform long-term cryopreservation (3–12 months) to reduce the risk of infection (Cibulkova et al., 2024). In conclusion, the efficacy and safety of FMT in anesthesia-exposed populations need to be systematically evaluated via more clinical trials, especially in terms of its preventive or therapeutic value for patients with cognitive impairments after the administration of multiple anesthetics. Moreover, the optimization of the implementation standards of FMT during the perioperative period, including donor selection, bacterial solution preparation methods, and administration timing, is necessary.

Although significant progress has been made in recent years in the study of the mechanism of interaction between gut microbiota and anesthesia, and potential pathways for regulating anesthesia efficacy and postoperative complications through the gut-brain-liver axis, as well as auxiliary strategies based on microbiota intervention, there is still a significant gap between current research and clinical application. Firstly, the depth of clinical evidence is insufficient. Most of the evidence for core signaling mechanisms comes from rodent models. There are differences in human physiology, immune system, and microbiota composition compared to animal models, and the exact pathways, strength of effects, and targets of these mechanisms in the human body still need to be directly validated in clinical populations. Existing human studies are mostly observational studies, making it difficult to establish whether changes in microbiota are a cause or a result of anesthesia/surgical complications. In addition, the significant impact of host factors (age, genetic background, dietary habits, underlying diseases, and concomitant medications) on microbiota anesthesia interactions has not been systematically quantified and integrated in clinical studies. This makes it difficult to predict an individual's response to anesthesia and sensitivity to microbiota intervention. Secondly, the breadth of clinical evidence is insufficient. At present, the number of clinical studies and the sample size is small, the observation period is short, and they are mostly concentrated in specific populations (such as ICU patients and specific surgical types), lacking universal conclusions. The lack of standardization in intervention plans makes it difficult to compare and generalize research results. In addition, we know very little about the potential long-term effects of microbiota regulation on anesthesia, such as its impact on immune homeostasis and metabolism.

Therefore, in order to overcome these clinical translation bottlenecks, possible directions include: 1) designing early clinical trials to evaluate the pharmacokinetics, safety, and preliminary efficacy of key microbial metabolites (such as specific SCFAs, IPA) in healthy volunteers or specific patient populations (such as reducing dosage as an anesthetic adjuvant). Combined with biological sample analysis, verify whether it reaches the target tissue and activates predetermined pathways (such as detecting receptor occupancy and downstream inflammatory marker changes). 2) Conduct a large-scale, prospective perioperative cohort study, systematically collect preoperative fecal microbiota, blood metabolome, and host genome data, and record anesthesia regimens, surgical types, and postoperative complications in detail. Integrate multidimensional data (including age, BMI, preoperative microbiota alpha diversity, abundance of key bacterial genera, baseline serum SCFAs, and host drug metabolism gene polymorphism such as CYP2B6) through machine learning (such as integrating XGBoost and random forest algorithm), screen microbiota characteristic spectra highly correlated with anesthesia sensitivity and postoperative cognitive impairment, and establish a clinically deployable predictive model (Lu et al., 2025). 3) Conduct multicenter, large sample, placebo-controlled randomized controlled studies targeting specific high-risk populations, such as elderly patients and patients undergoing major surgery. Clarify the optimal strain combination, dosage, timing of administration, and course of treatment. 4) Developing pro-anesthetics that can accurately determine their location in the intestine or across the blood-brain barrier, and promoting their early clinical evaluation, with the aim of improving intervention efficiency and reducing systemic side effects. For instance, construct a chitosan sodium alginate nanocarrier system loaded with butyric acid prodrugs (such as triglycerides) to achieve targeted regulation of the gut-brain axis; alternatively, CRISPR-Cas9 engineered bacteriophages can be designed to precisely inhibit specific metabolic pathways (such as butyryl CoA dehydrogenase gene) of perioperative pathogenic bacteria and reduce the production of liver toxicity intermediates related to propofol metabolism (Chen et al., 2023; Fan et al., 2024). 5) Build a "microbiota-anesthetic drug interaction database". 6) Based on accumulated evidence, develop consensus or preliminary guidelines on the use of probiotics, prebiotics, or specific dietary interventions during the perioperative period (Figure 2). Through this step-by-step effort, there is hope for gut microbiota and anesthesia to move from descriptive association studies to mechanistic causal verification, thereby assisting clinical treatment decisions.