Background
The blood brain barrier (BBB) is a highly selective permeable barrier that separates the blood and the central nervous system. Reference Yang, Gu and Mandeville1,Reference Acharya, Goldwaser and Forsberg2 The normal functioning of the BBB is required to control the extracellular environment of the central nervous system. Reference Netto, Iliff and Stanimirovic3,Reference Simöes Da Gama and Morin-Brureau4 The neurovascular unit (NVU) is an important and closely related structure, composed in part of neural cells, glia, extracellular matrix, and endothelial cells with tight junctions contributing to the BBB. Reference Netto, Iliff and Stanimirovic3,Reference Simöes Da Gama and Morin-Brureau4 The NVU plays a critical role in regulating cerebral blood flow, along with the regional delivery of nutrients and oxygen. Reference Netto, Iliff and Stanimirovic3,Reference Simöes Da Gama and Morin-Brureau4 Dysfunction of the anatomic components or physiologic processes within the NVU may contribute to BBB disruption. Reference Netto, Iliff and Stanimirovic3,Reference Simöes Da Gama and Morin-Brureau4 BBB disruption has been described to occur through various mechanisms affecting both its structure, such as decreased production of its components and ultrastructural alterations, and its function, as observed through various markers of increased permeability. Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8 BBB breakdown has been implicated in the pathophysiology of several neurologic diseases, including stroke and multiple sclerosis, among others, through a variety of purported pathophysiologic processes, including neuroinflammation, edema, and ion dysregulation. Reference Profaci, Munji, Pulido and Daneman9
Anesthesia is an integral part of perioperative management of patients with a neurological condition. The impact of anesthesia on the integrity of the BBB and its structure and function is an evolving area of investigation. Given the potential implications for important clinical outcomes such as postoperative cognitive function (POCD), Reference Ghoneim and Block10 a comprehensive understanding of the influence of anesthetic exposure on the BBB is required, including any potential modifiable variables such as choice of anesthetic agent. The present systematic review aims to synthesize all literature pertaining to the effects of anesthetics on BBB.
Methods
Search Strategy and Selection Criteria
The literature search was conducted on the electronic databases MedlineALL, Embase, and Cochrane to identify relevant articles from 1978 to 2022. The database search strategies were generated with assistance from M.E., librarian. The search involved MESH and keyword searches, titles, and abstracts of the following words: anesthetics, anesthesia, general anesthesia, volatile anesthetics, intravenous anesthetics with blood brain barrier, BBB, with disruption, damage, and permeability. The search was limited to include only human and animal studies published in the English language. No sex restrictions were placed. Reference lists of relevant articles were also searched manually.
Inclusion and Exclusion Criteria
This systematic review was designed utilizing the PRISMA-P methodology flowchart. Two independent reviewers (AAR and EP) evaluated the search results to identify eligible studies that met the predefined inclusion and exclusion criteria; a third author (TC) was involved to make the final decision in the event of a discrepancy. Primary laboratory animal research, case series, case reports, randomized control trials, and letters were considered for inclusion. Studies were included if the sample comprised previously healthy adult human subjects and/or animal models, and the primary aim of the study was to measure changes in BBB integrity (i.e. structure or function), following the administration of any anesthetic agent (general anesthesia, spinal anesthesia, conscious sedation, or combination) without the context of a surgical procedure. Articles that not only analyzed BBB permeability but also discussed secondary findings (e.g. cognitive impairment) were included. Studies that employed an experimental model that produced BBB disruption (e.g. iatrogenic, osmotic, medications, surgery) were excluded so as to examine only the pure effect of anesthetics on BBB instead of a synergistic effect. Articles for which the full text was not accessible were excluded.
Data Extraction and Quality Assessment
Several key indices were obtained from all included studies: first author, year of publication, study design, sample size, sample characteristics (species, age, gender), anesthetic agent (type, molecule, concentration, dose), measures of BBB integrity, and results. Quality assessment was performed using the Newcastle–Ottawa Scale. Reference Murad, Sultan, Haffar and Bazerbachi11
Results
Our database-advanced search yielded a total of 1050 articles (MedlineALL, Embase, and Cochrane) (Figure 1) and search development strategies are described in detail in the supplementary material. A total of 121 articles were screened and out of those, 14 articles were finally included in the study (Table 1). Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8,Reference Sun, Satomoto, Adachi and Makita12–Reference Saija, Princi, De Pasquale and Costa20 Nine reports were of level II evidence and five were of level III evidence. Reference Pruka21 The articles included nine randomized control studies (64.3%) and five quasi-experimental studies (35.7%). Out of 14 articles, 3 studies utilized mice as their study models (n = 206 mice), and 11 articles used rat models (n = 265 rats).
Figure 1: Prisma flowchart.
Table 1: Study and demographic characteristics
N = Number; M = Male, F = Female, N/S = Not specified.
Demographic Characteristics
The studies included a total of 354 male (75.2%), 60 female animals (12.7%), and 57 unknown sex (12.1%). In addition, though many exact ages were not reported, mice ranged from 8 weeks in age to models described as "adults," n = 18 (8 weeks of age), not specified (n = 145).
Primary Findings (Table 2)
Diagnostic modalities utilized across studies included: transmission electron microscopy (n = 4 studies), western blots (n = 4 studies), immunohistochemistry (n = 4 studies), immunofluorescence (n = 3 studies), 14C-alpha-aminoisobutyric acid (n = 3 studies), 14C-iodoantipyrine (n = 2 studies), Evans Blue albumin extravasation (n = 2 studies), UV spectroscopy (n = 1 studies), 14C-glucose (n = 1 study), scanning electron microscopy (n = 1 study), and fluorimetry (n = 1 study).
Table 2: Main findings
Twelve studies used volatile anesthetics, of which three used sevoflurane, six used isoflurane, one used sevoflurane and isoflurane, one used nitrous oxide, and one used pentobarbital and halothane. One study used fentanyl intravenously. One study used pentobarbital or ketamine intraperitoneally.
Primary findings can be further categorized into BBB structure and BBB function leading to BBB disruption and permeability changes. BBB structural deficits following the administration of an anesthetic agent included ultrastructural deficits, decreases in tight junctions, and decreases in BBB components such as F-actin, occludin, claudin-3, claudin-5, and VE-cadherin. Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8,Reference Sun, Satomoto, Adachi and Makita12,Reference Cao, Li and Li17 Primary findings commenting on BBB function also suggest evidence of BBB permeability increases following exposure to volatile anesthetics, as characterized by IgG, NaF, fibrinogen, and Evans Blue markers. Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8,Reference Johansson and Linder14,Reference Cao, Li and Li17 However, one study found decreased BBB permeability as captured by [14C]alpha-aminoisobutyric acid after pentobarbital or ketamine exposure. Reference Saija, Princi, De Pasquale and Costa20 Other BBB function alterations reported within the literature include increased glucose transport across the BBB, Ki changes, and cerebral blood flow changes. Reference Chi, Anwar, Sinha, Wei, Klein and Weiss15,Reference Nemoto, Stezoski and MacMurdo16,Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19
Further, three studies explored the effect of anesthetic dose on the BBB. Hu et al demonstrated that exposure to 3.6% sevoflurane for 6 hours downregulated expression of BBB components and increased fibrinogen deposition at 24 hours after treatment, while animals exposed to lower doses (2.4% or 3.2% sevoflurane) did not have such changes. Reference Hu, Wang and Zheng7 Contrarily, Chi et al found that there was no significant difference in transport across the BBB in any brain region between groups exposed to 1 or 2% isoflurane, and groups exposed to low-dose fentanyl (25 µg/kg) and high-dose fentanyl (100 µg/kg). Reference Chi, Anwar, Sinha, Wei, Klein and Weiss15,Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19
Two studies investigated the effect of duration of anesthetic exposure. Cao et al identified that after 2 or 4 hours of isoflurane exposure, progressive changes occurred in the BBB ultrastructure morphology, but this did not occur after 30 minutes or 1 hour of isoflurane exposure. Reference Cao, Ni and Li5 The authors also reported that increased BBB permeability, as detected by NaF content and IgG immunoreactivity, was only observed after 4 hours of isoflurane exposure, and not after 30 minutes, 1 hour, or 2 hours of isoflurane exposure. Sun et al found that the BBB opened gradually with increasing duration of sevoflurane exposure. Reference Sun, Satomoto, Adachi and Makita12 The authors found that exposure to sevoflurane for shorter durations (30 minutes) did not induce any long-term effects.
Of the studies that identified BBB changes, 10 employed inhalational (i.e. volatile) agents and 2 employed intravenous/intraperitoneal agents (Table 2). Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8,Reference Sun, Satomoto, Adachi and Makita12,Reference Johansson and Linder14–Reference Cao, Li and Li17,Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19,Reference Saija, Princi, De Pasquale and Costa20 Each of the 10 studies examining inhalational agents identified BBB disruption and/or increased BBB permeability, Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8,Reference Sun, Satomoto, Adachi and Makita12,Reference Johansson and Linder14–Reference Cao, Li and Li17 while the 2 studies examining intravenous/intraperitoneal agents, Chi et al and Saija et al, indicated decreased BBB permeability using fentanyl and pentobarbital/ketamine, respectively. Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19,Reference Saija, Princi, De Pasquale and Costa20
Three studies reported upon reversibility of BBB changes following anesthetic exposure. Cao et al found that the BBB ultrastructure morphological changes that occurred after 2 and 4 hours of isoflurane exposure gradually resolved at 24, 48, and 72 hours after exposure. Reference Cao, Ni and Li5 Also, the expression of occludin, a BBB component, which was decreased after 4 hours of isoflurane exposure, recovered to normal levels within 72 hours. Moreover, the authors found that BBB permeability returned to normal by 48 hours after isoflurane exposure as assessed by NaF content, and by 24 hours after isoflurane exposure as assessed by IgG immunoreactivity. Acharya et al found that increased BBB permeability, as assessed by IgG immunoreactivity, was not completely resolved at 24 hours after 3 hours of sevoflurane or isoflurane exposure. Reference Acharya, Goldwaser and Forsberg2 The authors found that the extent of reversibility after 24 hours following exposure may have been influenced by age, as morphological changes as assessed by scanning electron microscopy were more comparable to untreated controls in young and middle-aged animals, than in older animals. Sun et al reported that 48 hours after sevoflurane exposure, the percentage of capillaries that were destroyed decreased to 29%, as compared to 47% at 24 hours after exposure. Reference Sun, Satomoto, Adachi and Makita12
Discussion
Our systematic review highlights findings pertaining to the impact of anesthetic agents influencing BBB integrity in previously healthy models (Figure 2). Overall, our study showed that volatile anesthetics increase BBB permeability in previously healthy animal models. One study showed an increase in BBB permeability in older animals treated with sevoflurane but not isoflurane. Reference Acharya, Goldwaser and Forsberg2 Electron microscopy showed marked flattening of the luminal surfaces of brain vascular endothelial cells leading to such a permeability effect. Reference Acharya, Goldwaser and Forsberg2 Similarly, even with neonate models, 2% sevoflurane caused BBB disruptions in their hippocampus that was increased with exposure time. Reference Sun, Satomoto, Adachi and Makita12 This study not only showed the potential vulnerability of the developing brain to sevoflurane exposure but also highlighted the importance of duration of anesthesia and concentration. Reference Sun, Satomoto, Adachi and Makita12 Another study looking at nitrous oxide anesthetic explains that protein extravasation takes place in most rats when MAP exceeds 170 mmHg, suggesting a potential correlation to blood pressure changes in the intraoperative setting. Reference Johansson and Linder14 This is in line with another study that showed extreme hypertension and extreme hypotension, in combination with vasodilators may alter BBB permeability. Reference Chi, Anwar, Sinha, Wei, Klein and Weiss15 Interestingly, this study found that isoflurane significantly decreased the transfer of small hydrophilic molecules across the BBB. Reference Chi, Anwar, Sinha, Wei, Klein and Weiss15 This may be explained by a reduction in the perfused capillary surface area or as suggested in the paper, a potential direct effect of BBB disruption by isoflurane. Reference Chi, Anwar, Sinha, Wei, Klein and Weiss15
Figure 2: Plausible mechanisms for impact of anesthetic agents on blood brain barrier.
Our study findings were equivocal with respect to the impact of anesthetic dose on the BBB. While one study reported that downregulation of BBB components occurred only after exposure to a high dose of sevoflurane and not with lower doses, Reference Hu, Guo and Wang22 two other studies did not observe any difference in transport across the BBB with exposure to higher or lower doses of isoflurane and fentanyl, respectively. Reference Chi, Anwar, Sinha, Wei, Klein and Weiss15,Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19 By contrast, our study is more supportive of the notion that the impact of anesthetics on the BBB is related to the duration of exposure. Both studies investigating this question demonstrated that longer anesthetic exposure resulted in more significant BBB disruption, and that shorter durations of anesthetic exposure (e.g. 30 minutes) did not lead to significant BBB changes. Reference Cao, Ni and Li5,Reference Sun, Satomoto, Adachi and Makita12 These observations are hypothesis generating, and the influence of dose and duration of anesthetic exposure on the BBB is a clinically relevant topic that warrants further investigation.
Furthermore, 12 out of 14 studies included in the current review tested inhalational anesthetics. A consistent directionality toward BBB disruption and/or increased BBB permeability was reported amongst studies that employed such agents. Reference Acharya, Goldwaser and Forsberg2,Reference Cao, Ni and Li5–Reference He, Wang and Le8,Reference Sun, Satomoto, Adachi and Makita12–Reference Zheng, Meng, Li, Lu, Wu and Chen18 This was contrasted by studies employing intravenous or intraperitoneal agents, such as fentanyl, pentobarbital, or ketamine. Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19,Reference Saija, Princi, De Pasquale and Costa20 Despite there being a paucity of such studies included in the current work, there appeared to be a trend toward decreased BBB permeability subsequent to intravenous or intraperitoneal anesthetic exposure. Reference Chi, Wei, Anwar, Sinha, Klein and Weiss19,Reference Saija, Princi, De Pasquale and Costa20
Our study also suggests that BBB changes following anesthetic exposure demonstrate some reversibility over the short-term. Two of the studies found that both structural and functional changes in the BBB following anesthetic exposure gradually resolved over time, within 24, 48, or 72 hours following exposure depending on the finding in question. Reference Cao, Ni and Li5,Reference Sun, Satomoto, Adachi and Makita12 However, it is possible that the extent of reversibility depends in part on factors such as the recovery time and patient age, as suggested by one study that found 24 hours to be insufficient for complete recovery of BBB changes, and that the extent of recovery was reduced in older animals. Reference Acharya, Goldwaser and Forsberg2
The impact of anesthesia on the BBB is particularly relevant to the phenomenon of POCD in older adults. Reference Ghoneim and Block10 Unfortunately, the underlying causes of POCD have yet to be determined, and no definitive descriptions regarding mechanisms associated with the BBB disruption have been made. Reference Wang, Li and Cao13,Reference Hu, Guo and Wang22 Studies in animal models have suggested that exposure to agents such as sevoflurane is associated with BBB compromise, in an age-dependent manner. Reference Acharya, Goldwaser and Forsberg2 One proposed mechanism is a decrease in expression of tight junction proteins such as occludin, causing increased leakage of molecules and mediating isoflurane-induced hippocampus BBB disruption. Reference Cao, Ni and Li5 Further, a study also proposed the involvement of specific danger-associated molecular patterns with a pivotal role in mediating acute damage response and causing BBB dysfunction during surgery. Reference He, Wang and Le8
To the best of our knowledge, this is the first systematic review aiming to explore the pure effect of anesthetics on BBB integrity using data from healthy subjects. Notably, the present work identified a clear gap within the literature in that each of the included studies was in animal subjects. While our adopted inclusion criteria may have been too restrictive to draw upon human literature examining the impact of anesthetic on the BBB (i.e. healthy human subjects receiving anesthetics without surgical confounders), our review highlights the paucity of available human work related to this research question. Moreover, implications drawn toward the field are predominantly based on findings from preclinical studies employing volatile anesthetics; thus, additional studies including human subjects and utilizing intravenous anesthetics that test comparable research questions are warranted going forward.
Recommendations
This systematic review elucidated some potential findings and mechanisms that should inspire future work to aid practitioners and healthcare teams potentially better care for patients. It appears from our findings that a significant number of models with previously healthy brains suffered disruptions in BBB integrity following administration of different anesthetic agents. Therefore, in such a context, the risks and benefits of using particular sedative/opioid agents should be incorporated into the discussion. While the model studied in this review involved previously healthy models, it is plausible that greater post-surgical care and vigilance may be required for patients with previous neurological focal deficits prone to BBB disruption.
Limitations
There are several limitations to our systematic review. Overall, there was a limited number of studies looking at reversible or modifiable variables such as anesthetic agents that influence BBB integrity in previously healthy models. As a result, the conclusions regarding the impact of such agents on BBB integrity are limited. Subsequently, the results of this review should be considered preliminary, supporting the need for future prospective studies to examine effects on previously pathologic brains. In addition, attrition bias could not be eliminated due to heterogeneity in the study data. However, to the best of our abilities, we attempted to organize and standardize the information presented by each study. In addition, our review illustrated similar outcomes despite this variability.
Conclusion
Our study elucidated interesting findings pertaining to potential mechanisms of BBB integrity changes in response to anesthetic agents. Although preliminary, we hope this study would inspire future work to explore this area of research further so we can collectively provide better healthcare to patients.
Supplementary Material
To view supplementary material for this article, please visit https://doi.org/10.1017/cjn.2022.319.
Acknowledgement
We thank Marina Englesakis, UHN librarian for providing some assistance to develop search strategies.
Funding
No funding was received for this research.
Conflicts of Interest
The authors have no conflicts of interest to declare.
