Abstract
Background:
Perioperative neurocognitive disorders (PNDs) are the most common neurological complications in elderly patients undergoing surgery. Patients with PNDs have significantly greater incidences of postoperative disability and mortality. Currently, there are no specific treatments for PNDs.
Methods:
This review integrates the latest evidence examining the role of structural and functional changes in the blood–brain barrier (BBB) in the pathological mechanisms of PNDs, with the aim of identifying innovative preventive strategies and promising therapeutic targets for PNDs.
Results:
Researchers have proposed various hypotheses to shed light on the pathogenesis of PNDs, including neuroinflammation, neurotransmitter or receptor abnormalities, beta-amyloid (Aβ) deposition and tau protein phosphorylation, oxidative stress, iron homeostasis imbalance, circadian rhythm disruption, and changes in the gut microbiota. Damage to the BBB plays a critical role in the pathogenesis of PNDs.
Conclusion:
This article summarizes the role of BBB structural and functional changes in the pathogenesis of PNDs reported in recent studies, with the goal of providing new ideas for preventing and treating PNDs.
1 Introduction
Perioperative neurocognitive disorders (PNDs) refer to neuropsychological changes occurring during the perioperative period, including preoperative cognitive impairment, postoperative delirium, delayed neurocognitive recovery, and postoperative cognitive dysfunction (POCD) (Evered et al., 2018). PNDs are commonly observed in elderly individuals, with 25.8% of patients experiencing some degree of cognitive impairment within 1 week postoperatively, and 9.9% of elderly patients having cognitive impairment lasting longer than 3 months (Moller et al., 1998; Evered et al., 2022). PNDs can prolong hospital stays and are closely associated with increased risks of dementia and long-term mortality, imposing a significant burden on patients’ families and society (Evered et al., 2022). As no established preventive or treatment methods for PNDs are currently available, effective therapeutic strategies targeting their pathophysiological mechanisms must be urgently explored.
Disruption and dysfunction of the blood–brain barrier (BBB) are considered critical to the onset and progression of PNDs (Ni et al., 2018; Yuan et al., 2023). The BBB comprises endothelial cells, basement membranes, pericytes, tight junctions (TJs), and astrocyte peduncles and forms a selective barrier that maintains central nervous system homeostasis and prevents peripheral toxins and pathogens from entering (Kadry et al., 2020). Pathological mechanisms linked to PNDs, such as neuroinflammation, neurotransmitter or receptor abnormalities, β-amyloid (Aβ) deposition, tau protein phosphorylation, oxidative stress, iron homeostasis imbalance, circadian rhythm disruption, and changes in the gut microbiota, are believed to be associated with BBB damage. Specifically, during the development of PNDs, neuroinflammation is connected with a decrease in BBB pericyte coverage (Yuan et al., 2023), circadian rhythm disruption can influence astrocyte activation (Hou et al., 2019), gut microbiota dysbiosis can alter the expression of TJs (Wen et al., 2020), and Aβ can also damage brain microvascular endothelial cells (Zhao et al., 2025). Elevated serum levels of the inflammatory markers high-mobility group box 1 (HMGB1) and interleukin (IL)-6 in patients after major gastrointestinal surgery are associated with damage to the BBB and the development of PNDs (Lin et al., 2014). This evidence suggests that BBB damage and dysfunction are critical components of the multiple pathological mechanisms underlying PNDs. Reviews on PNDs over the past 2 years have predominantly focused on individual mechanisms, such as neuroinflammation and mitochondrial dysfunction (Liu et al., 2023, 2025), or centered on specific cell types, like microglia and neutrophils (Liu and Zhang, 2025; Cheng et al., 2025). However, the BBB is generally addressed only as a critical component of neuroinflammation, overlooking its multifaceted roles in PNDs. Therefore, this paper adopts a BBB-centric approach to synthesizing evidence that links to multiple PND mechanisms and BBB impairment. We propose that BBB impairment acts as a central hub where these pathological pathways converge. This framework not only provides a more integrated understanding of PND pathogenesis but also highlights novel therapeutic targets and preventive strategies aimed at preserving BBB function (Figure 1).
FIGURE 1
Diagram for the pathogenesis of perioperative neurocognitive disorders (PNDs), including neuroinflammation, neurotransmitter or receptor abnormalities, amyloid-β (Aβ) deposition, oxidative stress, iron homeostasis imbalance, circadian rhythm disruption, and gut flora dysbiosis.
2 Neuroinflammation and the BBB
Perioperative systemic inflammation disrupts the BBB, allowing peripheral inflammatory signals to enter the central nervous system. This activates and amplifies inflammatory responses within the multicellular networks of the neurovascular unit (NVU), ultimately damaging synapses and neural networks, manifesting as PNDs. BBB dysfunction facilitates the entry of peripheral inflammatory factors into the central nervous system while simultaneously promoting the maintenance and amplification of central inflammation.
Surgery and anesthesia can trigger systemic inflammation, disrupting the BBB. Peripheral inflammatory factors cross this barrier into the central nervous system, activating signaling pathways such as the nuclear factor-kappa B (NF-κB) pathway, inducing microglial polarization toward the pro-inflammatory M1 phenotype. Inflammation also activates the NOD-like receptor pyrin domain-containing 3 (NLRP3) inflammasome/caspase-1 axis and non-canonical caspase-4/5/11 pathways, contributing to the development of PNDs through inflammatory programmed cell death (pyroptosis) (Zhang and Yin, 2023; Cheng et al., 2022). Meanwhile, the mitogen-activated protein kinase (MAPK) pathway induces the secretion of active matrix metalloproteinases (MMPs), resulting in vascular endothelial cell apoptosis and ultrastructural disruption (Lee et al., 2024). Inflammation also alters the expression and localization of TJs. TJs are composed of membrane proteins (e.g., occludin and claudin) and cytoplasmic accessory proteins (e.g., zonula occluden [ZO]-1 and ZO-2) (Feldman et al., 2005). MMPs, pro-inflammatory cytokines (e.g., IL-1β, IL-6, tumor necrosis factor-α [TNF-α], IL-12, and IL-23), chemokines (e.g., C-C motif chemokine ligand [CCL] 2, CCL4, and CCL3), complement components (e.g., Component [C] 1q and C3), and adhesion molecules (e.g., intercellular adhesion molecule-1 [ICAM-1], P-selectin, and E-selectin) activate brain microvascular endothelial cells (BMECs) through intrinsic signaling pathways and synergistically drive the migration of leukocytes, microglia, and perivascular macrophages into the perivascular space, contributing to remodeling of TJs (Andjelkovic et al., 2023). Among these, specific cytokines can affect the tight junction complex of brain microvascular endothelium through multiple pathways, leading to increased barrier permeability. For example, IL-6 activates signal transducer and activator of transcription 3 (STAT3) pathway, subsequently downregulating TJs at both the transcriptional and protein levels (occludin, ZO-1, claudin-5), accompanied by decreased transendothelial electrical resistance (TEER) and increased macromolecular permeability (Digiovanni et al., 2024). TNFα, for instance, inhibits claudin-5 promoter activity in brain endothelial cells (Aslam et al., 2012). TNFα results in a rapid and transient upward band-shift of occludin, downregulation of occludin expression, morphological changes in BMECs, and increased BBB permeability (Ni et al., 2017). Studies indicate that normal human brain microvascular endothelial cells exhibit an irregular stellate morphology. In contrast, TNFα-treated cells exhibit a spindle-shaped appearance, narrower and more elongated, with significantly reduced TEER measurements (Ni et al., 2017). The BBB serves as a critical gateway for peripheral-to-central inflammatory transmission. Its disruption—the breach of this gateway—allows inflammatory signals to enter the central nervous system.
Subsequently, the multicellular network within the NVU (including endothelial cells, microglia, astrocytes, pericytes, etc.) amplifies the central inflammatory response. Inflammation and complement activation induce microglia to transition into a pro-inflammatory state (M1 phenotype), releasing cytokines and chemokines that participate in multiple signaling pathways, including the C-C chemokine receptor type 5 (CCR5)-G-protein-coupled receptors (GPCRs)-Ras-MAPK pathway (Xu et al., 2023; Qi et al., 2024). Astrocytes also undergo morphological changes: their end-feet become swollen, connexin 43 in their gap junctions is damaged, and hemichannel activity increases, which can release harmful molecules into the extracellular space, thereby inducing neuroinflammation (Andjelkovic et al., 2023; Cho et al., 2021; Dong et al., 2022). Anesthesia and surgery induce MMP-9 activation, and MMP-9 degrades the extracellular matrix and cell adhesion molecules, detaching pericytes from the capillary wall, which leads to pericyte loss, increased transcytosis in brain endothelial cells, and exacerbated BBB leakage (Hu et al., 2024). Furthermore, interactions among microglia, astrocytes, and mast cells collectively amplify inflammation. For example, microglia release IL-1α, TNFα, and C1q to induce A1 neurotoxic reactive astrocytes, which in turn secrete neurotoxins that can induce the rapid death of neurons and oligodendrocytes. Additionally, there is potential crosstalk between astrocytes and microglia via C3/C3aR signaling, playing a key role in synapse elimination (Liddelow et al., 2017; Xiong et al., 2018). For instance, mast cells release histamine, tryptase, and various cytokines that can activate microglia. Microglia respond to mast cells through Toll-like receptor (TLR) 2/TLR4 signaling molecules, prompting mast cells to release cytokines that further activate microglia, thus forming a positive feedback loop (Skaper et al., 2014). As another example, mast cell proteases can induce the activation of astrocytes to release IL-33 via the p38/extracellular signal-regulated kinase (ERK)/NF-κB pathway. Astrocyte-derived IL-33, in turn, can prompt mast cells to release IL-6, IL-13, and CCL2 to modulate microglia (Huang et al., 2024; Skaper et al., 2014).
Perioperative BBB dysfunction and neuroinflammation impact proximal biological endpoints, including synapses and neural networks, through pathways such as the complement-microglia axis, receptor channels, and energy matching. Alterations in these endpoints are directly related to the cognitive phenotype of PNDs. For instance, the transformation of microglia to the M1 phenotype affects neuronal autophagy and mediates pathological synaptic pruning, leading to neuronal damage and cognitive decline (Qi et al., 2024; Xu et al., 2023). Furthermore, the disruption of interactions between neurons, astrocytes, and cerebral blood vessels (i.e., neurovascular coupling) impairs synaptic plasticity, affecting the release of neurotransmitters, neuromodulators, and vasoactive mediators, as well as normal brain function (Andjelkovic et al., 2023; Cho et al., 2021; Dong et al., 2022).
BBB dysfunction is a critical link in the influx and amplification of inflammatory signals during the pathogenesis of PNDs and should be a priority target for intervention. Current pharmacological strategies for PNDs focus on preserving BBB integrity and reducing neuroinflammation. Targeted agents include MMP inhibitors (e.g., GM6001), NF-κB antagonists, NLRP3/caspase-1 inhibitors, and mast cell stabilizers (e.g., cromoglycate) (Zhang et al., 2016; Cheng et al., 2022; Zhu et al., 2024). Additional anti-inflammatory agents, such as minocycline and non-steroidal anti-inflammatory drugs (e.g., ibuprofen), are also under investigation (Cheng et al., 2022). Non-pharmacological interventions, such as electroacupuncture preconditioning, may reduce oxidative stress and neuroinflammation by preserving the function of telomerase reverse transcriptase, potentially alleviating PNDs (Wang et al., 2024). While some animal studies suggest therapeutic efficacy, most approaches remain in preclinical stages. Further research is needed to confirm their clinical relevance in treating human PNDs.
3 Neurotransmitter or receptor abnormalities and the BBB
Deficits in the cholinergic system are key contributors to PNDs. Anesthesia and surgery reduce the number of cholinergic neurons and the expression of acetylcholine receptors like the α7 nicotinic acetylcholine receptor (α7-nAChR) in aged mice (Xu et al., 2019; Wei et al., 2022). α7-nAChR serves as a positive regulator of TJ expression in the BBB. As a positive regulator of TJ expression, selective activation of α7-nAChR increases claudin-5 and occludin levels, thereby strengthening the endothelial barrier (Kimura et al., 2019). Therefore, perioperative cholinergic deficits may directly increase BBB permeability.
Acetylcholine and its receptors also regulate the BBB via astrocytes. α7-nAChR activation in astrocytes reduces the secretion of inflammatory cytokines by inhibiting the NF-κB pathway (Patel et al., 2017). Strategies to enhance overall cholinergic signaling, such as using acetylcholinesterase inhibitors like tacrine, which reduce acetylcholine degradation in the synaptic cleft, have been shown to potentially play a significant role in the treatment of PNDs (Kotekar et al., 2018). In contrast, α7-nAChR inhibition has been reported to promote A2 astrocyte formation, enhance the release of neurotrophic factors, and increase glutamate uptake by inhibiting the NF-κB pathway (Chen et al., 2024). This apparent contradiction suggests that the function of α7-nAChR in PNDs may be highly dependent on the cell’s location, timing, and context (Letsinger et al., 2022). Consequently, future therapeutic strategies targeting this receptor require greater precision to modulate α7-nAChR function under specific temporal and signaling conditions.
Glutamate dysregulation also contributes to PNDs. Astrocytes normally maintain low extracellular glutamate levels, but their clearance capacity decreases perioperatively (Cao et al., 2022). Following peripheral surgery, reduced expression of glutamine synthetase (GS) and the glutamate-aspartate transporter (GLAST) in mouse hippocampal astrocytes impairs extracellular glutamate uptake and conversion (Femenía et al., 2018). Decreased expression of astrocyte glutamate transporter 1 (GLT-1) in hippocampal astrocytes further reduces glutamate clearance, increasing susceptibility to PNDs (Jiao et al., 2024).
Excess glutamate activates N-methyl-D-aspartate receptors (NMDARs), leading to endothelial damage and the suppression of claudin-5 and occludin via the Ras homolog (Rho)/Rho-associated kinase (ROCK) pathway, thus compromising BBB integrity (Yu et al., 2022). Elevated extracellular glutamate levels also activate astrocytes and promote the release of inflammatory factors, while deleting NMDAR subunit 2C in astrocytes alleviates neuroinflammation (Zuo et al., 2023; Gao et al., 2024). These findings suggest that NMDAR antagonists like amantadine may be potential therapeutic agents for PNDs (Alaçam et al., 2025).
4 Aβ deposition, tau protein phosphorylation and the BBB
Excessive Aβ deposition and tau protein phosphorylation initiate pathogenic cascades that damage surrounding neurons and synapses, contributing to cognitive dysfunction (Yang et al., 2022). In animal models of PNDs, increased central expression of Aβ and phosphorylated tau (p-tau) has been observed (Chen et al., 2022).
Aβ1-42, the most common Aβ isoform associated with PNDs, increases BMEC permeability and induces mitochondrial dysfunction, which is characterized by elevated reactive oxygen species (ROS) and reduced adenosine triphosphate (ATP) levels (Zhao et al., 2025). On the other hand, alterations in the structure and function of the BBB also affect the transport of Aβ. In aged PND mice induced by Sevoflurane, the receptor for advanced glycation end products (RAGE) is overexpressed in hippocampal endothelial cells (Liang et al., 2020). RAGE serves as the primary influx transporter for free Aβ from the circulation into the central nervous system. Peripheral HMGB1 crosses the compromised BBB and binds to TLR4 on endothelial cell membranes, thereby promoting RAGE transcription and translation, which leads to the excessive transport of plasma Aβ into the brain (Irzan et al., 2025; Li et al., 2025). BBB disruption also impairs multiple Aβ efflux transporters, such as low-density lipoprotein receptor-related protein 1 (LRP1) and P-glycoprotein (P-gp), through pathways including neuroinflammation (Versele et al., 2022; Huang et al., 2019). This ultimately forms a vicious cycle: Aβ damages the BBB, which in turn impairs Aβ transport, leading to further accumulation. In ApoE4 transgenic knock-in mice, cognitive impairment is more pronounced after surgery and is accompanied by greater Aβ accumulation in the hippocampal CA3 region (Kim et al., 2021). ApoE4 is a high-risk subtype of cerebral lipid transporters that not only affects cholesterol homeostasis but also specifically modulates Aβ-receptor interactions (Chen et al., 2025). ApoE4 binds to Aβ to form Aβ-ApoE4 complexes, which inhibit LRP1-mediated Aβ efflux transport in endothelial cells and further exacerbate Aβ deposition in the feedback loop (Zhou et al., 2023; Zhang et al., 2022). As a potential therapeutic, graphene oxide-based nanomaterials have been shown to inhibit the β-cleavage of amyloid precursor protein and improve the delivery of endosomal Aβ to lysosomes, thereby reducing Aβ generation and enhancing its degradation, which significantly lowers hippocampal Aβ levels and improves cognitive function in postoperative mice (Zhang et al., 2020).
Tau proteins can also disrupt BBB integrity independently of Aβ. Inhibiting tau expression with doxycycline restores BBB integrity (Blair et al., 2015). Varenicline mitigates tau protein mislocalization and improves cognitive outcomes (Wang J. et al., 2022).
However, the precise contributions of Aβ and tau to PNDs remain uncertain. Among community-dwelling older adults carrying the ApoE4 allele, preoperative functional connectivity in Alzheimer’s disease–vulnerable brain regions is elevated, followed by a steep postoperative decline. Nevertheless, cerebrospinal fluid levels of Aβ, tau, or p-tau, and cognitive trajectories, do not significantly differ between ApoE4 carriers and non-carriers (Browndyke et al., 2021). Furthermore, in cognitively intact older adults undergoing non-cardiac, non-neurological surgery, no consistent associations have been found between neurocognitive outcomes and postoperative Aβ or tau fluctuations (Berger et al., 2022). These findings indicate that Aβ and tau protein phosphorylation may contribute to the pathogenesis of PNDs in some populations. However, their role requires further investigation.
5 Oxidative stress and the BBB
Oxidative stress increases with age, surgery, and anesthesia, disrupting both the structure and function of the BBB through multiple pathways implicated in the development of PNDs. The detrimental effect of oxidative stress is first evident in the compromised integrity of TJs. Under hypoxia/reoxygenation conditions, ROS downregulate TJs, including claudin-5, occludin, and ZO-1, in BMECs, thereby increasing BBB permeability (Won et al., 2015). Free radical scavengers like secoisolariciresinol diglucoside (SDG) can attenuate hypoxia-induced BBB dysfunction and maintain tight junction integrity, highlighting the importance of ROS scavenging in preserving the barrier (Nzou et al., 2020).
Activated MMPs play a crucial role in degrading the proteins of the BBB basement membranes and tight junctions. Activated MMPs degrade TJs and the basement membranes, while ROS may regulate BBB permeability in aged mice via the cyclophilin A (CyPA)/MMP-9 pathway (Liu et al., 2022). ROS also induce cytoskeletal rearrangement via the ROS/Ras homolog family member A (RhoA)/phosphoinositide 3-kinase (PI3K)/protein kinase B (PKB) signaling cascade, further weakening barrier performance (Gu et al., 2013). Excessive ROS can also directly interfere with the key substance transport functions of BMECs. In vitro experiments have demonstrated that ROS accumulation within endothelial cells activates the intracellular ERK1/2 signaling pathway, significantly downregulating the mRNA expression of the ATP-binding cassette transporter ABCG2 and hindering its mediated efflux of neurotoxic substances, while antioxidants (e.g., N-acetylcysteine) or ERK1/2 inhibitors (e.g., U0126) can effectively reverse this downregulation (Newman et al., 2023). Excess ROS also compromise mitochondrial respiration and morphology in brain capillaries, impairing energy metabolism and survival of vascular endothelial cells (Velmurugan et al., 2024). Mitochondrial morphological homeostasis depends on the dynamic balance between fission and fusion, and oxidative stress is a key factor that drives excessive mitochondrial fission. Research has found that ROS can induce the translocation of coactivator-associated arginine methyltransferase 1 (CARM1) to the cytoplasm, where CARM1 methylates dynamin-related protein 1 (DRP1), triggering excessive mitochondrial fission. This, in turn, forms a positive feedback loop of ROS generation (i.e., the CARM1-DRP1-ROS axis), continuously reinforcing the mitochondrial morphological imbalance and thereby accelerating dysfunction and cellular senescence (Cho and Kim, 2024). Additionally, studies have found that mitochondria-targeted therapies (e.g., Mitoquinone) can effectively activate the AMP-activated protein kinase (AMPK) signaling pathway, upregulating the expression of key transcription factors Sirt1 and Pgc1α. This significantly improves mitochondrial respiratory function, thereby restoring cellular metabolic homeostasis and effectively reversing ROS-induced damage (Cai et al., 2025). Moreover, ROS can further activate NLRP3 through the binding of thioredoxin-interacting protein (TXNIP) to the NLRP3 inflammasome, inducing the production of IL-1β, which exacerbates degradation of TJs and barrier dysfunction (Cao et al., 2016).
In summary, oxidative stress is a key pathological basis for PNDs. ROS employ multi-target mechanisms that lead to both structural leakage and functional impairment of the BBB. Currently, antioxidant therapies have shown neuroprotective potential (Wang W. et al., 2023; Wang Y. et al., 2023), and future research should focus on two main directions. First, targeted mechanistic studies should be conducted to systematically analyze the key signaling axes of ROS-mediated BBB damage (e.g., the CARM1–DRP1–ROS pathway). Second, based on clearly identified pathological targets, emphasis should be placed on developing mitochondria-targeted antioxidants with BBB-penetrating capabilities (e.g., Mitoquinone) and combining anti-inflammatory and antioxidant interventions based on these mechanistic insights.
6 Iron dysregulation and the BBB
Iron is essential for normal brain function, and disruptions in its uptake, transport, storage, or utilization can lead to excessive intracellular accumulation of ferrous iron (Fe2+). Such excessive iron accumulation is common in neurological disorders such as Alzheimer’s disease, Parkinson’s disease, and other disorders that manifest neurodegeneration and associated brain iron accumulation (Reichert et al., 2020). In a laparotomy model of mice, brain iron overload and altered expression of iron metabolism-related proteins were observed for up to 14 days after surgery (Li et al., 2016). Another mouse study reported that sevoflurane induces apoptosis through cross-dysregulation of iron and glucose metabolism and induction of energy stress, implicating iron–glucose imbalance in sevoflurane-induced cognitive impairment (Ge et al., 2021). These metabolic changes occurred concurrently with cognitive decline. Increased iron deposition in the brain (especially the hippocampus) and oxidative stress are related to the pathogenesis of PNDs.
Iron dysregulation disrupts the BBB integrity by affecting multiple components of the NVU. Iron-induced ROS downregulate TJs and induce BMEC apoptosis. Mitochondrial ferritin (FtMt) protects against TJ loss and apoptosis by inhibiting iron dysregulation and ROS accumulation in BMECs (Wang P. et al., 2022). Vascular endothelial (VE)-cadherin is a key regulator of endothelial permeability. Additionally, an increase in the labile iron pool (LIP) reduces the expression of VE-cadherin on endothelial cell surfaces through hypoxia-inducible factor 2α (HIF2α), while the iron chelator Desferal (DFO) preserves BBB integrity via HIF2α regulation (Rand et al., 2021). Astrocytes respond to elevated intracellular iron by secreting hepcidin, which reduces iron influx by decreasing ferroportin 1 (FPN1) expression on endothelial cells, a mechanism that may mitigate BBB dysfunction in neurodegenerative diseases (You et al., 2022). DFO significantly reduces hippocampal iron concentration and ferritin levels in rats with PNDs, and reverses surgery-induced abnormal changes in iron transport-related proteins, suggesting that iron chelators may be a candidate strategy for the intervention of PNDs (Pan et al., 2016).
7 Disrupted circadian rhythm and the BBB
Circadian rhythm disruption refers to a misalignment between an organism’s internal biological clock and external environmental cues. Aging, a major risk factor for PNDs, is associated with dysregulation of core circadian genes, including the brain and muscle ARNT-like protein 1 (BMAL1) gene and the circadian locomotor output cycle kaput (CLOCK) gene (Wyse and Coogan, 2010). Disruption of circadian rhythms activates astrocytes and impairs pericyte and endothelial cell function, a systemic cellular dysregulation directly destabilizing the BBB. The role of abnormal clock gene expression in astrocyte activation and central nervous system degeneration remains controversial. For example, specific BMAL1 deletion can activate astrocytes and increase the expression of the autophagy chaperone Bcl2-associated athanogene 3 (Bag3), enhancing the phagocytosis of alpha-synuclein (αSyn) and tau proteins (Sheehan et al., 2023). Conversely, aberrant clock gene expression may exacerbate astrocyte activation and neuroinflammation, while BMAL1 activation can attenuate inflammation and alleviate PND symptoms (Hou et al., 2019; Niu et al., 2022). Thus, clock gene dysregulation appears to affect the inflammatory/phagocytic balance in astrocytes. Circadian disruption also directly impairs BBB structural integrity. Altered expression of BMAL1 gene and CLOCK gene can reduce the expression of claudin-5 and occludin at both the mRNA and protein levels (Tiburcio et al., 2024). BMAL1 deficiency also induces pericyte dysfunction by decreasing platelet-derived growth factor receptor β (PDGFRβ), a key regulator of BBB integrity, thereby contributing to age-dependent BBB breakdown, reduced cerebral blood flow, and increased accumulation of blood-derived neurotoxins (Nakazato et al., 2017). Therefore, the disruption of endothelial tight junctions and impaired pericyte function, driven by the dysregulation of clock genes, collectively lead to the breakdown and dysfunction of the physical BBB. Notably, circadian rhythms may also affect the BBB indirectly via the gut–brain axis. Disruption of gut microbiota homeostasis impairs gut barrier integrity and promotes BBB dysfunction (Li et al., 2021).
Preoperative melatonin administration mitigates the impact of surgery and anesthesia on the expression of melatonin receptor and clock gene, such as BMAL1, which improves postoperative neurobehavioral outcomes in aged mice (Jia et al., 2024). Additionally, nobiletin, an enhancer of the clock protein retinoic acid receptor-related orphan receptors (RORs), may protect against PNDs by supporting the expression of BMAL1 and RORs (Sun et al., 2022). A deeper understanding of circadian regulation of the BBB may therefore inform novel therapeutic strategies for PNDs.
8 The microbial–gut–brain axis and the BBB
The gut microbiota is implicated in neurodegenerative disorders through the microbiota–gut–brain (MGB) axis. Surgery and anesthesia can induce gut dysbiosis, contributing to PNDs primarily through two mechanisms: reduced production of beneficial metabolites and increased production of harmful compounds.
First, dysbiosis of the intestinal flora reduces beneficial metabolites of short-chain fatty acids (SCFAs), such as acetic acid, propionic acid, and butyrate. Reduced butyrate levels downregulate TJs and increase BBB permeability, contributing to PNDs. Butyrate supplementation has been shown to improve postoperative cognitive function (Wen et al., 2020). Second, dysbiosis may elevate harmful metabolites, including lipopolysaccharides, palmitamide, and trimethylamine N-oxide (TMAO), which damage the BBB and promote cognitive decline (Pan et al., 2023; Xiong et al., 2024). For example, TMAO is produced in the liver via the oxidation of trimethylamine, which is produced by the gut microbiota through the breakdown of dietary choline and carnitine. Gut TMAO levels are positively correlated with the severity of cognitive decline; TMAO is highly susceptible to crossing the BBB to reach the brain and may be involved in the development of PNDs by modulating microglial and astrocyte activation (Brunt et al., 2021; Xiong et al., 2024). Additionally, gut dysbiosis can activate immune pathways that influence PNDs. For example, it activates gut-derived T helper 17 (Th17) cells and increases IL-17 secretion. IL-17A upregulates its receptors on brain endothelial cells, elevates MMP levels, and reduces TJ expression, compromising BBB integrity and contributing to cognitive dysfunction (Ni et al., 2018; Wen et al., 2022). In summary, the microbiota-gut-brain axis, through its metabolic and immune-mediated pathways, acts on the BBB and disrupts its integrity. This positions the BBB as a key hub connecting perioperative dysbiosis to the central neuroinflammation that drives PNDs.
Preclinical studies have shown that cognitive function in patients with PNDs can be improved by gut microbiota modulation (e.g., probiotics), supplementation with beneficial metabolites (e.g., propionic acid pretreatment), dietary restriction, and intestinal barrier protection. These findings suggest that targeting the MGB axis could be a promising therapeutic strategy for PNDs. Identifying specific gut flora biomarkers may aid in screening at-risk populations and guiding interventions. However, current studies are limited and require further clinical validation and standardized implementation (Sugita et al., 2023; Dai et al., 2024; Ren et al., 2024).
9 Discussion
The pathogenesis of PNDs does not stem from a single pathological mechanism but rather a complex regulatory network centered on the dysfunction of the BBB. The various mechanisms systematically described in this paper demonstrate significant crosstalk and synergistic amplification effects. First, systemic inflammation induced by perioperative stress is a key initiating factor. By activating signaling pathways such as NF-κB and MAPK, it damages the tight junctions between cerebral microvascular endothelial cells, leading to increased BBB permeability (Cheng et al., 2022; Lee et al., 2024). The infiltration of peripheral inflammatory mediators activates microglia and astrocytes, creating a vicious cycle in which inflammation compromises BBB integrity, which in turn exacerbates the inflammatory response. Building on this core feedback loop, other pathological processes also contribute synergistically to BBB damage through mutual crosstalk. For instance, Aβ metabolic imbalance has a mutually reinforcing relationship with neuroinflammation. On the one hand, abnormal Aβ deposition can activate glial cells, leading to inflammatory responses (Hansen et al., 2018). On the other hand, the inflammatory microenvironment impedes Aβ efflux by downregulating Aβ clearance transporters such as P-gp, leading to endothelial cell damage and BBB leakage (Zhang et al., 2022; Zhao et al., 2025). Meanwhile, glutamate excitotoxicity and cholinergic deficits affect barrier integrity by altering the endothelial cytoskeleton and inflammatory factor secretion through the NMDAR-Rho/ROCK and α7-nAChR pathways, respectively (Yu et al., 2022; Patel et al., 2017). Iron-mediated ROS can induce the downregulation of TJs and the apoptosis of BMECs; furthermore, an increase in LIP can lead to a reduction in HIF2α-mediated VE-cadherin expression on the endothelial cell surface, collectively leading to BBB dysfunction (Wang P. et al., 2022; Rand et al., 2021). ROS can regulate BBB permeability through multiple inflammatory factor pathways. For example, ROS can further activate NLRP3 by binding TXNIP to the NLRP3 inflammasome, inducing the production of IL-1β which promotes degradation of TJs (Cao et al., 2016). Dysregulated expression of the core clock gene BMAL1 due to circadian rhythm disruption reduces the activity of efflux transporters, lowering the clearance efficiency of neurotoxic substances like Aβ, while also exacerbating astrocyte activation and increasing central inflammation levels, amplifying the damage to the BBB from upstream inflammatory signals (Niu et al., 2022; Hou et al., 2019; Zhang et al., 2021). Regarding the gut-brain axis, Th17-IL-17A signaling downregulates TJs (Ni et al., 2018; Wen et al., 2022), and gut dysbiosis leads to increased TMAO production which promotes the activation of microglia and astrocytes (Xiong et al., 2024; Brunt et al., 2021), all of which directly or indirectly lead to BBB damage. In summary, the pathophysiological network of PNDs is characterized by multiple mechanisms converging on BBB damage through various pathways. Whether it is systemic inflammation, oxidative stress, central protein or ion metabolism disorders, distal gut dysbiosis, or circadian rhythm dysregulation, their pathogenic signals ultimately all point to the disruption of BBB integrity. This highlights the pivotal role of the BBB as a key hub linking multi-dimensional pathophysiological changes with the final cognitive outcome.
The structural and functional integrity of the BBB is essential for maintaining brain homeostasis. Multiple perioperative factors can impair BBB function, contributing to the onset and progression of PNDs. Numerous studies have explored therapeutic strategies aimed at preserving BBB function to prevent or treat PNDs. For example, the perioperative use of melatonin, which has been shown to reduce the incidence of PNDs in elderly patients (Fan et al., 2017); early monitoring of iron homeostasis and iron chelator, which may offer a novel treatment option for PNDs (Li et al., 2016; Pan et al., 2016); and the application of inflammation-related enzyme inhibitors and cytokine antagonists, are effective in animal models (Cheng et al., 2022; Zhu et al., 2024). Acetylcholinesterase inhibitors and NMDAR antagonists are also considered potential treatment options for PNDs due to their effects on neurotransmitter pathways (Kotekar et al., 2018; Alaçam et al., 2025). Moreover, strategies targeting Aβ deposition and tau protein phosphorylation represent another promising avenue (Zhang et al., 2020; Wang J. et al., 2022). Additionally, antioxidant therapies and gut microbiota remodeling have likewise shown neuroprotective effects in preclinical models (Wang W. et al., 2023; Wang Y. et al., 2023; Sugita et al., 2023; Dai et al., 2024; Ren et al., 2024).
This review also has certain limitations. First, it is a narrative review, not a systematic one, so the literature selection may be subject to a certain bias and may not have covered all relevant research. Second, most of these approaches are supported only by animal studies, and their efficacy and safety in humans remain uncertain. For example, differences between human and animal gut microbiota limit the generalizability of findings on gut microbiota remodeling. Finally, while this review elaborates on the central role of the BBB in the multiple mechanisms of PNDs, it does not explore with the same depth the precise initial signaling pathways through which perioperative stimuli, such as surgery and anesthesia, initiate the first step of BBB disruption. These limitations indicate the need for more systematic reviews in the future.
Although the role of BBB dysfunction in PNDs is evident, few studies have directly investigated how BBB damage mediates the development of PNDs, a topic worth further research. While the mechanisms underlying PNDs, particularly neuroinflammation, are increasingly understood, large-scale clinical evidence evaluating relevant interventions is lacking. Key areas such as drug selection, dosing regimens, and perioperative implementation strategies remain underexplored. Clarifying these aspects is essential for bridging the gap between experimental findings and clinical application. Despite these gaps, the current evidence suggests that multiple mechanistic pathways hold promise for understanding and preventing PNDs. Continued research will be instrumental in advancing our mechanistic understanding and the development of effective clinical interventions.
Statements
Author contributions
YS: Writing – original draft, Conceptualization, Visualization. YC: Writing – original draft, Conceptualization, Visualization. BZ: Writing – original draft. YH: Writing – original draft. ZL: Writing – original draft. XZ: Writing – review & editing. FG: Supervision, Writing – review & editing.
Funding
The author(s) declare that financial support was received for the research and/or publication of this article. This study was supported by the China National University Student Innovation & Entrepreneurship Development Program (202210392010).
Conflict of interest
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.
Generative AI statement
The authors declare that no Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Correction note
This article has been corrected with minor changes. These changes do not impact the scientific content of the article.
Publisher’s note
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.
References
1
Alaçam B. Başarır ŞN. Tuna A. T. Palabıyık O. Çakıroğlu H. (2025). Effects of amantadine on recovery, postoperative cognitive dysfunction, and pain after propofol anesthesia in mice.Behav. Brain Res.477:115290. 10.1016/j.bbr.2024.115290
2
Andjelkovic A. V. Situ M. Citalan-Madrid A. F. Stamatovic S. M. Xiang J. Keep R. F. (2023). Blood-brain barrier dysfunction in normal aging and neurodegeneration: Mechanisms, impact, and treatments.Stroke54661–672. 10.1161/strokeaha.122.040578
3
Aslam M. Ahmad N. Srivastava R. Hemmer B. (2012). TNF-alpha induced NFκB signaling and p65 (RelA) overexpression repress Cldn5 promoter in mouse brain endothelial cells.Cytokine57269–275. 10.1016/j.cyto.2011.10.016
4
Berger M. Browndyke J. N. Cooter Wright M. Nobuhara C. Reese M. Acker L. et al (2022). Postoperative changes in cognitive function and cerebrospinal fluid neurodegenerative disease biomarkers.Ann. Clin. Transl. Neurol.9155–170. 10.1002/acn3.51499
5
Blair L. J. Frauen H. D. Zhang B. Nordhues B. A. Bijan S. Lin Y. C. et al (2015). Tau depletion prevents progressive blood-brain barrier damage in a mouse model of tauopathy.Acta Neuropathol. Microbiol.3:8. 10.1186/s40478-015-0186-2
6
Browndyke J. N. Wright M. C. Yang R. Syed A. Park J. Hall A. et al (2021). Perioperative neurocognitive and functional neuroimaging trajectories in older APOE4 carriers compared with noncarriers: Secondary analysis of a prospective cohort study.Br. J. Anaesth.127917–928. 10.1016/j.bja.2021.08.012
7
Brunt V. E. LaRocca T. J. Bazzoni A. E. Sapinsley Z. J. Miyamoto-Ditmon J. Gioscia-Ryan R. A. et al (2021). The gut microbiome-derived metabolite trimethylamine-N-oxide modulates neuroinflammation and cognitive function with aging.GeroScience43377–394. 10.1007/s11357-020-00257-2
8
Cai L. Swetha M. Raji A. Terry A. V. Jr. Raju R. P. (2025). Mitochondria-targeted therapy with metformin and MitoQ reduces oxidative stress, improves mitochondrial function, and restores metabolic homeostasis in a murine model of Gulf War Illness.Redox Biol.85:103714. 10.1016/j.redox.2025.103714
9
Cao G. Jiang N. Hu Y. Zhang Y. Wang G. Yin M. et al (2016). Ruscogenin attenuates cerebral ischemia-induced blood-brain barrier dysfunction by suppressing TXNIP/NLRP3 inflammasome activation and the MAPK pathway. Int. J. Mol. Sci. 17:1418. 10.3390/ijms17091418
10
Cao Y. Lin X. Liu X. Yu K. Miao H. Li T. (2022). The role of astrocytes in the mechanism of perioperative neurocognitive disorders.Brain Sci.12:1435. 10.3390/brainsci12111435
11
Chen J. Zou J. Huang P. Gao X. Lun J. Li Y. et al (2024). KYNA ameliorates glutamate toxicity of HAND by enhancing glutamate uptake in A2 astrocytes.Int. J. Mol. Sci.25:4286. 10.3390/ijms25084286
12
Chen X. Gao F. Lin C. Chen A. Deng J. Chen P. et al (2022). mTOR-mediated autophagy in the hippocampus is involved in perioperative neurocognitive disorders in diabetic rats.CNS. Neurosci. Ther.28540–553. 10.1111/cns.13762
13
Chen Y. Jin H. Chen J. Li J. Gãman M. A. Zou Z. (2025). The multifaceted roles of apolipoprotein E4 in Alzheimer’s disease pathology and potential therapeutic strategies.Cell Death Discov.11:312. 10.1038/s41420-025-02600-y
14
Cheng C. Wan H. Cong P. Huang X. Wu T. He M. et al (2022). Targeting neuroinflammation as a preventive and therapeutic approach for perioperative neurocognitive disorders.J. Neuroinflam.19:297. 10.1186/s12974-022-02656-y
15
Cheng J. Gao J. Li J. Tian H. (2025). Neutrophils: A new target for postoperative cognitive dysfunction.Apoptosis301117–1132. 10.1007/s10495-025-02098-w
16
Cho I. Kim J. M. Kim E. J. Kim S. Y. Kam E. H. Cheong E. et al (2021). Orthopedic surgery-induced cognitive dysfunction is mediated by CX3CL1/R1 signaling.J. Neuroinflam.18:93. 10.1186/s12974-021-02150-x
17
Cho Y. Kim Y. K. (2024). ROS-mediated cytoplasmic localization of CARM1 induces mitochondrial fission through DRP1 methylation.Redox Biol.73:103212. 10.1016/j.redox.2024.103212
18
Dai H. Y. Zhang Z. X. Tan C. Xian X. Ji D. Yang J. et al (2024). Propionic acid ameliorates cognitive function through immunomodulatory effects on Th17 cells in perioperative neurocognitive disorders.Heliyon10:e28817. 10.1016/j.heliyon.2024.e28817
19
Digiovanni S. Lorenzati M. Bianciotto O. T. Godel M. Fontana S. Akman M. et al (2024). Blood-brain barrier permeability increases with the differentiation of glioblastoma cells in vitro.Fluids Barriers CNS21:89. 10.1186/s12987-024-00590-0
20
Dong R. Han Y. Jiang L. Liu S. Zhang F. Peng L. et al (2022). Connexin 43 gap junction-mediated astrocytic network reconstruction attenuates isoflurane-induced cognitive dysfunction in mice.J. Neuroinflam.19:64. 10.1186/s12974-022-02424-y
21
Evered L. Atkins K. Silbert B. Scott D. A. (2022). Acute perioperative neurocognitive disorders: A narrative review.Anaesthesia7734–42. 10.1111/anae.15613
22
Evered L. Silbert B. Knopman D. S. Scott D. A. DeKosky S. T. Rasmussen L. S. et al (2018). Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery.Br. J. Anaesth.1211005–1012. 10.1016/j.bja.2017.11.087
23
Fan Y. Yuan L. Ji M. Yang J. Gao D. (2017). The effect of melatonin on early postoperative cognitive decline in elderly patients undergoing hip arthroplasty: A randomized controlled trial.J. Clin. Anesth.3977–81. 10.1016/j.jclinane.2017.03.023
24
Feldman G. J. Mullin J. M. Ryan M. P. (2005). Occludin: Structure, function and regulation.Adv. Drug Deliv. Rev.57883–917. 10.1016/j.addr.2005.01.009
25
Femenía T. Giménez-Cassina A. Codeluppi S. Fernández-Zafra T. Katsu-Jiménez Y. Terrando N. et al (2018). Disrupted neuroglial metabolic coupling after peripheral surgery.J. Neurosci.38452–464. 10.1523/jneurosci.1797-17.2017
26
Gao R. Ali T. Liu Z. Li A. He K. Yang C. et al (2024). NMDAR (2C) deletion in astrocytes relieved LPS-induced neuroinflammation and depression.Int. Immunopharmacol.132:111964. 10.1016/j.intimp.2024.111964
27
Ge X. Zuo Y. Xie J. Li X. Li Y. Thirupathi A. et al (2021). A new mechanism of POCD caused by sevoflurane in mice: Cognitive impairment induced by cross-dysfunction of iron and glucose metabolism.Aging1322375–22389. 10.18632/aging.203544
28
Gu Y. T. Xue Y. X. Wang Y. F. Wang J. H. Chen X. ShangGuan Q. R. et al (2013). Minoxidil sulfate induces an increase in blood-brain tumor barrier permeability through the ROS/RhoA/PI3K/PKB signaling pathway.Neuropharmacology75407–415. 10.1016/j.neuropharm.2013.08.004
29
Hansen D. V. Hanson J. E. Sheng M. (2018). Microglia in Alzheimer’s disease.J. Cell Biol.217459–472. 10.1083/jcb.201709069
30
Hou J. Shen Q. Wan X. Zhao B. Wu Y. Xia Z. (2019). REM sleep deprivation-induced circadian clock gene abnormalities participate in hippocampal-dependent memory impairment by enhancing inflammation in rats undergoing sevoflurane inhalation.Behav. Brain Res.364167–176. 10.1016/j.bbr.2019.01.038
31
Hu Y. Hu X. D. He Z. Q. Liu Y. Gui Y. K. Zhu S. H. et al (2024). Anesthesia/surgery activate MMP9 leading to blood-brain barrier disruption, triggering neuroinflammation and POD-like behavior in aged mice.Int. Immunopharmacol.135:112290. 10.1016/j.intimp.2024.112290
32
Huang L. Li B. Li X. Liu G. Liu R. Guo J. et al (2019). Significance and mechanisms of P-glycoprotein in central nervous system diseases.Curr. Drug Targets201141–1155. 10.2174/1389450120666190308144448
33
Huang X. Lan Z. Hu Z. (2024). Role and mechanisms of mast cells in brain disorders.Front. Immunol.15:1445867. 10.3389/fimmu.2024.1445867
34
Irzan F. I. Retinasamy T. Wen W. R. Sheng E. T. M. Shaikh M. F. Arulsamy A. (2025). The role of HMGB1 in infection-related cognitive deficits.Front. Biosci.30:25544. 10.31083/FBL25544
35
Jia X. Song Y. Li Z. Yang N. Liu T. Han D. et al (2024). Melatonin regulates circadian rhythm to ameliorate postoperative sleep disorders and neurobehavioral abnormalities in aged mice.CNS. Neurosci. Ther.30:e14436. 10.1111/cns.14436
36
Jiao X. H. Wan J. Wu W. F. Ma L. H. Chen C. Dong W. et al (2024). GLT-1 downregulation in hippocampal astrocytes induced by type 2 diabetes contributes to postoperative cognitive dysfunction in adult mice.CNS. Neurosci. Ther.30:e70024. 10.1111/cns.70024
37
Kadry H. Noorani B. Cucullo L. (2020). Overview of blood-brain barrier structure, function, impairment, and biomarkers of integrity.Fluids Barriers CNS17:69. 10.1186/s12987-020-00230-3
38
Kim J. H. Jung H. Lee Y. Sohn J. H. (2021). Surgery performed under propofol anesthesia induces cognitive impairment and amyloid pathology in ApoE4 knock-in mouse model.Front. Aging Neurosci.13:658860. 10.3389/fnagi.2021.658860
39
Kimura I. Dohgu S. Takata F. Matsumoto J. Kawahara Y. Nishihira M. et al (2019). Activation of the α7 nicotinic acetylcholine receptor upregulates blood-brain barrier function through increased claudin-5 and occludin expression in rat brain endothelial cells.Neurosci. Lett.6949–13. 10.1016/j.neulet.2018.11.022
40
Kotekar N. Shenkar A. Nagaraj R. (2018). Postoperative cognitive dysfunction: Current preventive strategies.Clin. Interv. Aging132267–2273. 10.2147/cia.S133896
41
Lee T. H. Chen J. L. Chang C. H. Tsai M. M. Tseng H. C. Chang Y. C. et al (2024). Brain-protective sterols from soft coral inhibits lipopolysaccharide-induced matrix metalloproteinase-9-mediated astrocytic migration.Biomedicines12:226. 10.3390/biomedicines12010226
42
Letsinger A. C. Gu Z. Yakel J. L. (2022). α7 nicotinic acetylcholine receptors in the hippocampal circuit: Taming complexity.Trends Neurosci.45145–157. 10.1016/j.tins.2021.11.006
43
Li W. Chen Q. Peng C. Yang D. Liu S. Lv Y. et al (2025). Roles of the receptor for advanced glycation end products and its ligands in the pathogenesis of Alzheimer’s disease.Int. J. Mol. Sci.26:403. 10.3390/ijms26010403
44
Li Y. Pan K. Chen L. Ning J. L. Li X. Yang T. et al (2016). Deferoxamine regulates neuroinflammation and iron homeostasis in a mouse model of postoperative cognitive dysfunction.J. Neuroinflam.13:268. 10.1186/s12974-016-0740-2
45
Li Y. Shao L. Mou Y. Zhang Y. Ping Y. (2021). Sleep, circadian rhythm and gut microbiota: Alterations in Alzheimer’s disease and their potential links in pathogenesis.Gut Microbes13:1957407. 10.1080/19490976.2021.1957407
46
Liang R. Ou S. Han Y. Xu J. Zhou S. (2020). Plasma amyloid beta level changes in aged mice with cognitive dysfunction following sevoflurane exposure.Exp. Gerontol.129:110737. 10.1016/j.exger.2019.110737
47
Liddelow S. A. Guttenplan K. A. Clarke L. E. Bennett F. C. Bohlen C. J. Schirmer L. et al (2017). Neurotoxic reactive astrocytes are induced by activated microglia.Nature541481–487. 10.1038/nature21029
48
Lin G. X. Wang T. Chen M. H. Hu Z. H. Ouyang W. (2014). Serum high-mobility group box 1 protein correlates with cognitive decline after gastrointestinal surgery.Acta Anaesthesiol. Scand.58668–674. 10.1111/aas.12320
49
Liu F. Wu X. Wang Z. Li A. Luo Y. Cao J. (2025). Mitochondrial dysfunction in postoperative cognitive dysfunction: From preclinical mechanisms to multimodal diagnostics and precision intervention.Ageing Res. Rev.111:102845. 10.1016/j.arr.2025.102845
50
Liu L. F. Hu Y. Liu Y. N. Shi D. W. Liu C. Da X. et al (2022). Reactive oxygen species (ROS) contribute to delirium-like behavior by activating CypA/MMP9 signaling and inducing blood-brain barrier impairment in aged mice after anesthesia and surgery.Front. Aging Neurosci.14:1021129. 10.3389/fnagi.2022.1021129
51
Liu X. Zhang A. (2025). Microglial modulation as a therapeutic avenue for perioperative neurocognitive disorders: Unveiling pathophysiological mechanisms and clinical implications.CNS Neurosci. Ther.31:e70481. 10.1111/cns.70481
52
Liu Y. Yang W. Xue J. Chen J. Liu S. Zhang S. et al (2023). Neuroinflammation: The central enabler of postoperative cognitive dysfunction.Biomed. Pharmacother.167:115582. 10.1016/j.biopha.2023.115582
53
Moller J. T. Cluitmans P. Rasmussen L. S. Houx P. Rasmussen H. Canet J. et al (1998). Long-term postoperative cognitive dysfunction in the elderly ISPOCD1 study. ISPOCD investigators. International study on post-operative cognitive dysfunction.Lancet351857–861. 10.1016/s0140-6736(97)07382-0
54
Nakazato R. Kawabe K. Yamada D. Ikeno S. Mieda M. Shimba S. et al (2017). Disruption of Bmal1 impairs blood-brain barrier integrity via pericyte dysfunction.J. Neurosci.3710052–10062. 10.1523/jneurosci.3639-16.2017
55
Newman S. A. Short J. L. Nicolazzo J. A. (2023). Reduction in ABCG2 mRNA expression in human immortalized brain microvascular endothelial cells by ferric ammonium citrate is mediated by reactive oxygen species and activation of ERK1/2 signaling.Pharm. Res.40651–660. 10.1007/s11095-022-03458-w
56
Ni P. Dong H. Wang Y. Zhou Q. Xu M. Qian Y. et al (2018). IL-17A contributes to perioperative neurocognitive disorders through blood-brain barrier disruption in aged mice.J. Neuroinflam.15:332. 10.1186/s12974-018-1374-3
57
Ni Y. Teng T. Li R. Simonyi A. Sun G. Y. Lee J. C. (2017). TNFα alters occludin and cerebral endothelial permeability: Role of p38MAPK.PLoS One12:e0170346. 10.1371/journal.pone.0170346
58
Niu X. Zheng S. Li S. Liu H. (2022). Role of different doses of ketamine in postoperative neurocognitive function in aged mice undergoing partial hepatectomy by regulating the Bmal1/NMDA/NF-K b axis.Eur. Surg. Res.63182–195. 10.1159/000520210
59
Nzou G. Wicks R. T. VanOstrand N. R. Mekky G. A. Seale S. A. El-Taibany A. et al (2020). Multicellular 3D neurovascular unit model for assessing hypoxia- and neuroinflammation-induced blood-brain barrier dysfunction.Sci. Rep.10:9766. 10.1038/s41598-020-66487-8
60
Pan C. Zhang H. Zhang L. Chen L. Xu L. Xu N. et al (2023). Surgery-induced gut microbial dysbiosis promotes cognitive impairment via the regulation of intestinal function and palmitic amide metabolites.Microbiome11:248. 10.1186/s40168-023-01689-6
61
Pan K. Li X. Chen Y. Zhu D. Li Y. Tao G. et al (2016). Deferoxamine pre-treatment protects against postoperative cognitive dysfunction of aged rats by depressing microglial activation via ameliorating iron accumulation in hippocampus.Neuropharmacology111180–194. 10.1016/j.neuropharm.2016.09.004
62
Patel H. McIntire J. Ryan S. Dunah A. Loring R. (2017). Anti-inflammatory effects of astroglial α7 nicotinic acetylcholine receptors are mediated by inhibition of the NF-κB pathway and activation of the Nrf2 pathway.J. Neuroinflam.14:192. 10.1186/s12974-017-0967-6
63
Qi Z. Peng J. Wang H. Wang L. Su Y. Ding L. et al (2024). Modulation of neuroinflammation and cognitive function in postoperative cognitive dysfunction via CCR5-GPCRs-Ras-MAPK pathway targeting microglial EVs.CNS Neurosci. Ther.30:e14924. 10.1111/cns.14924
64
Rand D. Ravid O. Atrakchi D. Israelov H. Bresler Y. Shemesh C. et al (2021). Endothelial iron homeostasis regulates blood-brain barrier integrity via the HIF2α-Ve-Cadherin pathway.Pharmaceutics13:311. 10.3390/pharmaceutics13030311
65
Reichert C. O. de Freitas F. A. Sampaio-Silva J. Rokita-Rosa L. Barros P. L. Levy D. et al (2020). Mechanisms of ferroptosis in neurodegenerative diseases.Int. J. Mol. Sci.21:8765. 10.3390/ijms21228765
66
Ren L. Liang H. Zhu L. Yang X. Zhang H. Sun N. et al (2024). Dietary restriction improves perioperative neurocognitive disorders by inhibiting neuroinflammation and gut microbial dysbiosis.Neuroscience54048–67. 10.1016/j.neuroscience.2024.01.012
67
Sheehan P. W. Nadarajah C. J. Kanan M. F. Patterson J. N. Novotny B. Lawrence J. H. et al (2023). The astrocytic BMAL1-BAG3 axis protects against alpha-synuclein and tau pathology.Neuron1112383–2398.e7. 10.1016/j.neuron.2023.05.006
68
Skaper S. D. Facci L. Giusti P. (2014). Mast cells, glia and neuroinflammation: Partners in crime.Immunology141314–327. 10.1111/imm.12170
69
Sugita S. Tahir P. Kinjo S. (2023). The effects of microbiome-targeted therapy on cognitive impairment and postoperative cognitive dysfunction-A systematic review.PLoS One18:e0281049. 10.1371/journal.pone.0281049
70
Sun Z. Yang N. Jia X. Song Y. Han D. Wang X. et al (2022). Nobiletin attenuates anesthesia/surgery-induced neurocognitive decline by preserving the expression of clock genes in mice.Front. Neurosci.16:938874. 10.3389/fnins.2022.938874
71
Tiburcio D. Adolphe S. Akpu I. B. Teglas T. Naranjo O. George S. H. et al (2024). Abstract 829: Phthalate exposure results in circadian rhythm-induced BBB dysfunction increasing the risk for breast cancer brain metastasis.Cancer Res.84:829. 10.1158/1538-7445.am2024-829
72
Velmurugan G. V. Vekaria H. J. Hartz A. M. S. Bauer B. Hubbard W. B. (2024). Oxidative stress alters mitochondrial homeostasis in isolated brain capillaries.Fluids Barriers CNS21:81. 10.1186/s12987-024-00579-9
73
Versele R. Sevin E. Gosselet F. Fenart L. Candela P. (2022). TNF-α and IL-1β modulate blood-brain barrier permeability and decrease amyloid-β peptide efflux in a human blood-brain barrier model.Int. J. Mol. Sci.23:10235. 10.3390/ijms231810235
74
Wang J. Zhu S. Lu W. Li A. Zhou Y. Chen Y. et al (2022). Varenicline improved laparotomy-induced cognitive impairment by restoring mitophagy in aged mice.Eur. J. Pharmacol.916:174524. 10.1016/j.ejphar.2021.174524
75
Wang P. Ren Q. Shi M. Liu Y. Bai H. Chang Y. Z. (2022). Overexpression of mitochondrial ferritin enhances blood-brain Barrier integrity following ischemic stroke in mice by maintaining iron homeostasis in endothelial cells.Antioxidants11:1257. 10.3390/antiox11071257
76
Wang W. Chen C. Wang Q. Ma J. G. Li Y. S. Guan Z. et al (2024). Electroacupuncture pretreatment preserves telomerase reverse transcriptase function and alleviates postoperative cognitive dysfunction by suppressing oxidative stress and neuroinflammation in aged mice.CNS Neurosci. Ther.30:e14373. 10.1111/cns.14373
77
Wang W. Gao W. Gong P. Song W. Bu X. Hou J. et al (2023). Neuronal-specific TNFAIP1 ablation attenuates postoperative cognitive dysfunction by targeting SNAP25 for K48-linked ubiquitination.Cell Commun. Signal.21:356. 10.1186/s12964-023-01390-z
78
Wang Y. Cai Z. Zhan G. Li X. Li S. Wang X. et al (2023). Caffeic acid phenethyl ester suppresses oxidative stress and regulates M1/M2 microglial polarization via the Sirt6/Nrf2 pathway to mitigate cognitive impairment in aged mice following anesthesia and surgery.Antioxidants12:714. 10.3390/antiox12030714
79
Wei P. Lyu W. Xu L. Feng H. Zhou H. Li J. (2022). α7 nicotinic acetylcholine receptor may be a pharmacological target for perioperative neurocognitive disorders.Front. Pharmacol.13:907713. 10.3389/fphar.2022.907713
80
Wen J. Ding Y. Wang L. Xiao Y. (2020). Gut microbiome improves postoperative cognitive function by decreasing the permeability of the blood-brain barrier in aged mice.Brain Res. Bull.164249–256. 10.1016/j.brainresbull.2020.08.017
81
Wen Y. Feng S. Dai H. Mao M. Zhou Z. Li B. et al (2022). Intestinal dysbacteriosis-propelled T helper 17 cells activation mediate the perioperative neurocognitive disorder induced by anesthesia/surgery in aged rats.Neurosci. Lett.783:136741. 10.1016/j.neulet.2022.136741
82
Won S. Sayeed I. Peterson B. L. Wali B. Kahn J. S. Stein D. G. (2015). Vitamin D prevents hypoxia/reoxygenation-induced blood-brain barrier disruption via vitamin D receptor-mediated NF-kB signaling pathways.PLoS One10:e0122821. 10.1371/journal.pone.0122821
83
Wyse C. A. Coogan A. N. (2010). Effects of aging on the diurnal expression patterns of CLOCK and BMAL1 in the mouse brain.Brain Res.133721–31. 10.1016/j.brainres.2010.03.113
84
Xiong C. Liu J. Lin D. Zhang J. Terrando N. Wu A. (2018). Complement activation contributes to perioperative neurocognitive disorders in mice.J. Neuroinflam.15:254. 10.1186/s12974-018-1292-4
85
Xiong Y. Pu Y. N. Li L. Y. Su Y. Niu J. Y. Xiao Z. Y. (2024). The gut microbiota-derived metabolite, trimethylamine-N-oxide, aggravates cognitive dysfunction induced by femoral fractures in mice.Kaohsiung J. Med. Sci.40732–743. 10.1002/kjm2.12873
86
Xu F. Han L. Wang Y. Deng D. Ding Y. Zhao S. et al (2023). Prolonged anesthesia induces neuroinflammation and complement-mediated microglial synaptic elimination involved in neurocognitive dysfunction and anxiety-like behaviors.BMC Med.21:7. 10.1186/s12916-022-02705-6
87
Xu H. Chen L. Zhang X. Jiang X. Tian W. Yu W. et al (2019). Central cholinergic neuronal degeneration promotes the development of postoperative cognitive dysfunction.Lab. Investig.991078–1088. 10.1038/s41374-018-0174-9
88
Yang Y. S. He S. L. Chen W. C. Wang C. M. Huang Q. M. Shi Y. C. et al (2022). Recent progress in understanding the role of noncoding RNA in postoperative cognitive dysfunction.Front. Cell Neurosci.16:1024475. 10.3389/fncel.2022.1024475
89
You L. Yu P. P. Dong T. Guo W. Chang S. Zheng B. et al (2022). Astrocyte-derived hepcidin controls iron trafficking through the blood-brain barrier by regulating ferroportin 1 in microvascular endothelial cells.Cell Death Dis.13:667. 10.1038/s41419-022-05043-w
90
Yu Y. Wu Y. Wei J. Huang F. Mao F. Nong W. et al (2022). NMDA mediates disruption of blood-brain barrier permeability via Rho/ROCK signaling pathway.Neurochem. Int.154:105278. 10.1016/j.neuint.2022.105278
91
Yuan H. Sun D. Ji Y. Meng B. Lu B. Liu R. et al (2023). Pericyte loss impairs the blood-brain barrier and cognitive function in aged mice after anesthesia/surgery.Brain Res. Bull.204:110799. 10.1016/j.brainresbull.2023.110799
92
Zhang J. Zhu S. Jin P. Huang Y. Dai Q. Zhu Q. et al (2020). Graphene oxide improves postoperative cognitive dysfunction by maximally alleviating amyloid beta burden in mice.Theranostics1011908–11920. 10.7150/thno.50616
93
Zhang M. Yin Y. (2023). Dual roles of anesthetics in postoperative cognitive dysfunction: Regulation of microglial activation through inflammatory signaling pathways.Front. Immunol.14:1102312. 10.3389/fimmu.2023.1102312
94
Zhang S. L. Lahens N. F. Yue Z. Arnold D. M. Pakstis P. P. Schwarz J. E. et al (2021). The circadian clock regulates efflux through the blood-brain barrier in mice and human cells.Nat. Commun.12:617. 10.1038/s41467-020-20795-9
95
Zhang S. Dong H. Zhang X. Li N. Sun J. Qian Y. (2016). Cerebral mast cells contribute to postoperative cognitive dysfunction by promoting blood-brain barrier disruption.Behav. Brain Res.298158–166. 10.1016/j.bbr.2015.11.003
96
Zhang Y. L. Wang J. Zhang Z. N. Su Q. Guo J. H. (2022). The relationship between amyloid-beta and brain capillary endothelial cells in Alzheimer’s disease.Neural Regen. Res.172355–2363. 10.4103/1673-5374.335829
97
Zhao H. Fan M. Zhang J. Gao Y. Chen L. Huang L. (2025). Amyloid beta-induced mitochondrial dysfunction and endothelial permeability in cerebral microvascular endothelial cells: The protective role of dexmedetomidine.Brain Res. Bull.220:111137. 10.1016/j.brainresbull.2024.111137
98
Zhou X. Shi Q. Zhang X. Gu L. Li J. Quan S. et al (2023). ApoE4-mediated blood-brain barrier damage in Alzheimer’s disease: Progress and prospects.Brain Res. Bull.199:110670. 10.1016/j.brainresbull.2023.110670
99
Zhu B. Cao A. Chen C. Zhou W. Luo W. Gui Y. et al (2024). MMP-9 inhibition alleviates postoperative cognitive dysfunction by improving glymphatic function by regulating AQP4 polarity.Int. Immunopharmacol.126:111215. 10.1016/j.intimp.2023.111215
100
Zuo T. Gong B. Gao Y. Yuan L. (2023). In vitro study of the stimulatory effects of extracellular Glu on astrocytes.Mol. Biol. Rep.506611–6617. 10.1007/s11033-023-08601-3
Summary
Keywords
perioperative neurocognitive disorders, blood–brain barrier, endothelial cells, neuroinflammation, pathogenesis
Citation
Su Y, Chen Y, Zheng B, Huang Y, Liao Z, Zheng X and Gao F (2026) Structure and function of the blood–brain barrier in perioperative neurocognitive disorders. Front. Neurosci. 19:1690354. doi: 10.3389/fnins.2025.1690354
Received
21 August 2025
Accepted
23 October 2025
Published
02 January 2026
Corrected
08 January 2026
Volume
19 - 2025
Edited by
Mohammad Badruzzaman Khan, Augusta University, United States
Reviewed by
Thomson Patrick Joseph, University of Regina, Canada
Fei Fei Xu, Air Force Medical University, China
Updates
Copyright
© 2026 Su, Chen, Zheng, Huang, Liao, Zheng and Gao.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Fei Gao, fjmugaofei@163.com
†These authors have contributed equally to this work and share first authorship
Disclaimer
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.