2021 年 7 巻 p. 5-14
The prevalence of dementia has increased with the rising aging population worldwide, and aging and dementia are known risk factors for delirium, a condition that is closely related to inflammation, such as coronavirus disease 2019 (COVID-19). COVID-19 is caused by severe acute respiratory syndrome coronavirus 2. It originated in Wuhan city, China in early December 2019 and has spread rapidly worldwide ever since. The elderly population are at a higher risk of contracting the virus than other age groups. Evidence has shown that many patients with COVID-19 exhibit symptoms of delirium, disproportionally affecting the elderly, and this is concerning due to the close association between dementia and delirium. Impairment of cerebrovascular functions, especially the blood-brain barrier, plays a key role in delirium and subsequent dementia caused by inflammation from infectious diseases, such as COVID-19.
With the increasing aging population worldwide, cognitive impairment has become an unavoidable problem affecting both the medical and the healthcare system. The number of people suffering from dementia, a chronic neurodegenerative disease, increases every year with an estimate of up to 43 million worldwide and predicted to double within the next 20 years1). Delirium, an acute condition with similar symptoms of dementia, has shown to affect 50% of elderly people (aged 65 years and older)2).
Risk factors for delirium can be divided into three categories: predisposing factors, facilitating factors, and precipitating factors. Predisposing factors include old age (60 years and older), heavy alcohol consumption, depression, and history of ischemia or stroke and brain damage, such as dementia. Predisposing factors are important in screening high-risk patients. Facilitating factors do not directly contribute to causing delirium, and these include stress, sleep impairment, and changes in the physical environment or physical restraint. Precipitating factors can directly cause delirium, and these include the use of anesthetics, hypoxia, infection, and invasive surgery. However, with sufficient support from the medical staff, facilitating and precipitating factors can be improved to a certain degree, ultimately decreasing the risk of developing delirium.
Dementia and delirium are distinct health conditions that are closely related. Study has shown that a single episode of delirium could increase the chance of developing dementia later in life2) and accelerates the rate of neurodegeneration. A study including 560 people aged 70 years or more without dementia found that those who were affected with delirium had significantly greater cognitive decline than those who were not3). Another cohort study including 309 acutely ill patients showed that 32% of patients experiencing delirium progressed towards dementia, while only 16% of those who were not delirious progressed to dementia4). A meta-analysis of 23 studies showed that patients who developed delirium during a hospital stay had a 2.3 times greater risk of developing dementia than those who did not5). Moreover, a study has shown that the duration of delirium could also be an important factor in developing dementia later in life, wherein patients with a longer duration of delirium experiencing worse global cognition and executive function at 3 and 12 months6). The reverse is also true, with delirium being more prevalent in patients with dementia. A study showed that 37% of patients with dementia experienced delirious attacks during their hospital stay.7) Furthermore, vascular dementia (VaD) is often associated with delirium8), and the pathological changes in patients with VaD are associated with cholinergic deficits9), which is also a risk factor of delirium. Studies have shown that 40% to 50% of patients with VaD experience delirium, the prevalence of which exceeds that of patients with early-onset Alzheimer’s disease (AD) or frontotemporal dementia7,10).
While the relationship between dementia and delirium remains unclear, several mechanisms have been proposed. For example, certain insults to the brain such as metabolic derangements, certain drugs (e.g., anticholinergics), ischemia, and immunological stressors could alter neurotransmitter concentrations causing acetylcholine deficiency or dopamine excess, thus resulting in neuronal dysfunctions11,12). In addition, hypoxia and cerebral ischemia may directly cause cerebral dysfunction through impaired cerebral blood flow and metabolism13). Evidence also showed that certain anesthetics can directly contribute to the accumulation of beta-amyloids, which is also a fundamental part of AD14). Another theory known as the threshold hypothesis has also been proposed. As the brains of patients with dementia have fewer neuronal connections, it becomes more difficult to deal with inflammation and infection, not only resulting in delirium but also advancing the disease. Inflammation in response to an infection or other stress (i.e., surgery or acute illness) could also cause neuronal dysfunction. Cases such as this can occur in various mechanisms, such as apoptosis, activation of microglia and astrocytes, and altered neurotransmission15). All these studies provide valuable insights into the close relationship between dementia and delirium. In this review, we focus on the blood-brain barrier (BBB) changes associated with systemic inflammation, and how these changes link dementia and delirium.
The exact molecular mechanism of the pathology of delirium remains unknown, which could be attributed to its multifactorial causation. However, evidence has shown that interaction between several biological factors might be the cause of the disruption of the large-scale neuronal network in the brain, leading to acute cognitive dysfunction2). These contributing factors of delirium include the dysregulation of neurotransmitters, inflammation, physiological stressors, metabolic derangements, electrolyte disorders, and genetic factors. Other factors can also directly interact with neurotransmission or cellular metabolism, including drugs, hypercortisolism, electrolyte disturbances, hypoxia, and impaired glucose oxidation. This review will focus on the role of inflammation in delirium to better understand the relationship between coronavirus disease 2019 (COVID-19) and delirium.
The dysregulation of cytokines is believed to be the main factor in the involvement of inflammation in delirium. Studies have shown that the use of interleukins in rat models alters the acetylcholine levels and activity16) and mediate exotoxic neurodegeneration17). A studyshowed that aging can alter the CNS and peripheral levels of certain cytokines (IL-1 beta and TNF alpha)18). On the contrary, other studyshowed that aging is not associated with cytokine level chages19). A study conducted on mice showed that changes in glial reactivity in the aging brain can exacerbate neuroinflammatory cytokine responses20). Symptoms associated with illness and delirium such as a decrease in cognition, depression of mood, and lethargy known as “sickness behavior” are now believed to have been caused by peripheral inflammatory cytokines21).
A study showed that out of 261 patients with dementia, 19.4% experienced delirium, in which 34.4% were diagnosed with VaD, dementia with the highest precentile22). VaD is well associated with cerebrovascular pathologies, such as disruption of the BBB. Interestingly, 31.8 % of patients were diagnosed with dementia with Lewy bodies (DLB). Recent evidence showed that BBB of synucleinopathy patients, which was initially believed to be intact, may be disrupted through a proinflammatory response23). Studies have also shown that alpha-synuclein can cross the BBB bidirectionally, with LRP1 being a potential efflux transporter for alpha-synuclein24). LRP1 is also involved in the efflux of beta-amyloids and is downregulated in AD25). If this downregulation is also present in synucleinopathies, it can result in impairment of alpha-synuclein clearance and the accumulation of alpha-synclein in the brain. Furthermore, it is well known that acetylcholine deficiency occurs in DLB26), a risk factor for delirium.
The BBB is composed of cerebrovascular endothelial cells between blood and the brain and selectively prevents substances in the blood from entering the CNS. In contrast to the leaky capillary endothelium in peripheral organs, the BBB is sealed by tight junctions and possesses various channels, receptors, and enzymes to allow substance transport. This allows transport across the BBB to be highly selective, which benefits the CNS in various ways. For example, pathogens, blood cells, and certain cytokines do not enter the CNS through the BBB. Antibodies and certain antibiotics are also prohibited from crossing the BBB, making drug and vaccine development difficult in some cases. Examples of molecules that are allowed to cross the BBB include insulin, leptin, TNF alpha, and epidermal growth factor, among others. The BBB is also important in supplying the brain with energy, as the brain lacks storage for carbohydrates. Disruption of the BBB can dramatically change the environment of the CNS. During systemic inflammation, the BBB can be changed both histologically and molecularly27), resulting in cytokine entry into the CNS. As stated, cytokine dysregulation can lead to sickness behavior and delirium. This phenomenon likely occurs through the dysregulation of neurotransmitters and the activation of glial cells in the presence of certain inflammatory cytokines.
Various conditions have also been associated with the disruption of BBB. Neuroinflammation following ischemic stroke is shown to disrupt the BBB. Following a stroke, the tight junction integrity in the BBB decreases, which can lead to vasogenic edema, hemorrhagic transformation, ultimately increasing mortality. The levels of inflammatory cytokines, such as IL-1 beta and TNF-1 alpha, are upregulated in the brain following cerebral ischemia, with evidence pointing that inflammatory cytokines can increase the permeability of the BBB28). Neurotransmitters released from hypoxic neurons during an ischemic stroke can also cause permanent neurotoxic damage to neural tissues, of which the BBB regulates29). Another condition associated with changes to the BBB is multiple sclerosis (MS). MS is an autoimmune disease characterized by the infiltration of lymphocytes and macrophages through the BBB and subsequent breakdown of the myelin sheath in neurons. A feature of MS involves the early breakdown of the BBB, allowing lymphocyte entry30).
Changes to the BBB may also play an important role in AD via the accumulation of beta-amyloid through intake at the BBB and the decrease of the clearance of beta-amyloids. Evidence suggests that 85% of beta-amyloid clearance is through the BBB, via LRP-1, LRP-2, and APOJ31). Some studies have shown that vascular factors such as hypertension32), diabetes33), hyperlipidemia34), and cardiovascular disease35) can be risk factors of AD. Several AD animal models, which include models derived from mutations in APOE4, APP, and PSEN1 have shown to develop an early BBB breakdown36). Cerebral amyloid angiopathy (CAA) is also related to the BBB breakdown. CAA is the pathology that involves the accumulation of beta-amyloid on the walls of cerebral vessels. While CAA can be observed in physiological changes related to aging, it is known to be more prevalent in AD. In CAA, beta-amyloids are known to accumulate in the outer and middle layers of vascular smooth muscle cells. The main component of the accumulated beta-amyloid is amyloid-beta 40, which is in contrast to the main component of amyloid plaques or amyloid-beta 42. The accumulation of beta-amyloid on the walls of cerebral vessels can lead to the degeneration of tunica media and endothelium, resulting in the disruption of the BBB. Studies with animal models and postmortem brain of patients with AD suggest that CAA is not accompanied by the increase in production of APP, the precursor to beta-amyloid, but is caused by the dysfunction of beta-amyloid clearance from the brain37). Beta-amyloids in brain interstitial fluid are known to be excreted to the cerebral spinal fluid through the perivascular space. However, if this mechanism is obstructed, the flow of beta-amyloids can stagnate and accumulate on the walls of blood vessels. During severe systemic inflammation where vascular endothelial cells are potentially damaged by inflammatory cytokines, the BBB disruption caused by CAA can be possibly amplified. In addition, the accumulation of beta-amyloids can induce inflammation, i.e., CAA in itself can cause vascular inflammation. Albumin quotient, a common biofluid marker of BBB breakdown, is shown to elevate in AD38). This phenomenon is believed to be induced by vascular inflammation caused by CAA and BBB disruption. As beta-amyloids in CAA disrupt the BBB, peripheral inflammation can have a stronger influence on the CNS. This event can be the mechanism causing delirium accompanied by systemic inflammation and eventually resulting in the onset of dementia.
Coronavirus disease 2019 (COVID-19) was first identified in Wuhan city, China in early December 2019 and has spread rapidly worldwide, with over 100 million confirmed cases and over 2.5 million deaths to date39). It is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). COVID-19, delirium, and dementia all disproportionally affect the elderly. Age is the major risk factor for neurodegenerative diseases, and over 50% of elderly people in hospitals are affected by delirium2), while older patients with COVID-19 experience the most severe and prolonged course of the disease40). Data have shown that up to 67% of patients with COVID-19 exhibit signs of delirium (Table 1). Differences in the prevalence of delirium as reported in various studies could be attributed to the differences in the average age of the cohorts and the difficulty of diagnosing delirium. As shown, delirium can be listed as a feature of COVID-19. Signs of delirium could sometimes be present without the usual febrile responses such as cough or fever. With data suggesting that 40% of all cases have no radiographic abnormalities on presentation41), currently, there are proposals listing delirium as one of the diagnostic criteria for COVID-19, because the possibility of overlooking a potential COVID-19 case without considering delirium exists. While the exact cause of this phenomenon remains unknown, it could be attributed to the multifactorial nature of delirium. Many aforementioned risk factors of delirium can be associated with COVID-19, such as inflammation, ischemia, and anesthetics. A mass cohort study has shown that patients with COVID-19 have a higher risk of developing complications including dementia, mood disorder, insomnia, and ischemic stroke than influenza patients 6 months after the diagnosis42).
| Reference | Study size | Age | Prevalence of delirium in COVID-19 patients |
|---|---|---|---|
| Pun, Brenda T. et al., 202157 | 2088 | 64 (median) | 55% |
| Helms, J. et al., 202058 | 58 | — | 67% |
| Kennedy, M. et al., 202056 | 817 | 77 (mean) | 28% |
| Ticinesi, A. et al., 202059 | 852 | 73 (mean) | 11% |
| Mao, L. et al., 202060 | 214 | 52 (mean) | 8% |
| Khan, Sikandar H. et al., 202061 | 268 | 58 (mean) | 29% |
One of the main features of severe COVID-19 is systemic inflammation known as the “cytokine storm.” As previously stated, systemic inflammation can lead to undesirable changes to the BBB, which presents as sickness behavior or delirium. During some cases of systemic inflammation caused by sepsis, sepsis-associated encephalopathy43) and sepsis-associated delirium44) were observed. The sepsis-associated encephalopathy can often trigger severe cognitive impairments and exacerbates neurodegenerative pathology44). It has been suggested that the systemic inflammation triggered by some severe COVID-19 cases is similar in scale to that of sepsis45), thus similar cognitive deterioration can occur and will boost existing neurodegenerative pathologies. Since some neurodegenerative diseases are associated with BBB changes, it is likely that the BBB change caused by the systemic inflammation can directly contribute to the development of some neurodegenerative diseases.
There is evidence showing that the SARS-Cov-2 can directly infect the brain, via various proposed pathways. The virus infects the cell by the binding of the receptor-binding domain of the spike protein to the receptor angiotensin-converting enzyme 2 (ACE2), after cleavage by furin46). Many neurons and neuroglia with high vascularization and permeable BBB express both ACE2 and furin, therefore prone to SARS-CoV-2 infections. Another pathway suggests that SARS-CoV-2 enters the brain through the nasal epithelium, which brings the virus into the olfactory bulb47). Studies also showed that neuropilin-1 may also be involved in the viral infection of SARS-CoV-2 into the brain48), and tests showed that monoclonal anti-neuropilin-1 antibody can significantly reduce viral infection49). However, in all cases, the SARS-CoV-2 infection to the brain is yet to be fully understood.
Another clinical symptom of COVID-19 is the decrease in blood oxygenation caused by pneumonia. In severe cases, widespread inflammation of the lung is associated with respiratory failure and severe hypoxia. This omnipresent hypoxia is likely to affect the brain and may have negative effects including respiratory alkalosis and energy deprivation. A decrease in arterial oxygen saturation below 75% can cause impairment of neuronal activity50). Hypoxia can also cause damage to neural cells due to the rapid increase in reactive oxygen species, against which the weak brain antioxidative defenses have few ways to defend.
Systemic inflammation associated with COVID-19 is known to increase blood levels of fibronectin, likely through the stimulation of its liver synthesis51). This phenomenon can facilitate blood clot formation, and it has been shown that approximately 20% to 50% of COVID-19 cases were complicated by thrombotic and thromboembolic manifestations52), with stroke reported in approximately 5% of hospitalized patients53). Stroke is associated with secondary neurodegeneration and increased risk of both VaD and AD. Therefore, the relationship can be drawn linking COVID-19 associated thrombosis and neurodegenerative diseases.
Furthermore, the indirect effects of COVID-19 can be seen when considering the nature of the pandemic. The risk of developing delirium can be reduced up to 40% with steps such as having a family member assisting the patient with self-orientation54), which is a difficult task when dealing with COVID-1955). The psychological stresses involved in coping with COVID-19 infection, the lack of physical contact, and the isolation from family are likely to trigger depression, which in itself is a well-known risk factor of dementia.
BBB is key in understanding CNS conditions, such as dementia and delirium. As BBB is highly important in maintaining the environment of the CNS, changes to the BBB may lead to drastic results. For example, the pathology of CAA involves the accumulation of beta-amyloid protein in cerebrovascular walls through the impairment of the clearance of beta-amyloids, most of which occurs across the BBB. Other conditions such as DLB, MS, and stroke also involve BBB changes, the mechanism which is likely to be necessary for understanding the said conditions.
Dementia and delirium are two closely linked conditions that are prevalent worldwide with the increasing aging population. Patients who have experienced a delirious attack are more likely to develop dementia than those who have not, thus making the understanding of the mechanism linking dementia and delirium an important task for both the medical and scientific communities. Studies have also shown that cerebrovascular pathologies are likely key interfaces in linking dementia and delirium. CAA pathologies seen in dementia, such as AD, can disrupt BBB, and this BBB disruption can cause delirium through the entry of various cytokines and dysregulation of neurotransmitters. Changes to the BBB independent of dementia may also cause delirium, weakening the brain and increasing the risk of developing dementia.
COVID-19 exhibits diverse clinical symptoms, the understating of which will be key for overcoming the pandemic. Recent evidence has shown that delirium presents as a clinical feature of COVID-1956). We proposed the following mechanism linking COVID-19, delirium, and the risk of developing dementia later in life. Infection caused by various pathogens, including SARS-CoV-2, can cause systemic inflammation, resulting in BBB disruption. As a result, the disruption allows inflammatory cytokines to enter the CNS, dysregulating neurotransmitters and activating glial cells, causing delirium. Furthermore, SARS-CoV-2 can directly infect the brain, and although not proven, it possible may be the cause of delirium observed in COVID-19. Through various neuropathologies, delirium can ultimately cause dementia, and pathologies present in dementia can cause additional BBB disruption, resulting in a vicious cycle (Figure 1).
Delirium and dementia are and may continue to be a challenge for the world that requires urgent intervention. With the global COVID-19 pandemic, both dementia and delirium will more likely become more prevalent in the future due to the direct and indirect interaction between COVID-19, delirium, and dementia. We believe this review will provide valuable insights to the existing literature to better understand the cerebrovascular mechanism involved in linking the three conditions.
This work was supported by JSPS KAKENHI Grant Number 17H05080 (grant-in-Aid for Young Scientists (A)) and the research grant from Cell Science Research Foundation.
Authors have no potential conflicts of interest to declare.
