The Brain-Heart Axis: Neuroinflammatory Interactions in Cardiovascular Disease

Abstract

Purpose of review:

The role of neuroimmune modulation and inflammation in cardiovascular disease has been historically underappreciated. Physiological connections between the heart and brain, termed the heart-brain axis (HBA), are bidirectional, occur through a complex network of autonomic nerves/hormones and cytokines, and play important roles in common disorders.

Recent findings:

At the molecular level, advances in the past two decades reveal complex crosstalk mediated by the sympathetic and parasympathetic nervous systems, the renin-angiotensin aldosterone and hypothalamus-pituitary axes, microRNA, and cytokines. Afferent pathways amplify proinflammatory signals via the hypothalamus and brainstem to the periphery, promoting neurogenic inflammation. At the organ level, while stress-mediated cardiomyopathy is the prototypical disorder of the HBA, cardiac dysfunction can result from a myriad of neurologic insults including stroke and spinal injury. Atrial fibrillation is not necessarily a causative factor for cardioembolic stroke, but a manifestation of an abnormal atrial substrate, which can lead to the development of stroke independent of AF.

Summary:

Central and peripheral neurogenic pro-inflammatory factors have major roles in the HBA, manifesting as complex bi-directional relationships in common conditions such as stroke, arrhythmia and cardiomyopathy.

Introduction

The role of neuroimmune modulation and inflammation in cardiovascular disease has been historically underappreciated. The expansion of multidisciplinary efforts in biomedical research has increased our awareness of the connectivity of central and peripheral neuronal mediation of key presentations that impact the course and therapy of CVD. Given that sensory neurons release transmitters that interact with the endothelium, neutrophils, macrophages, and other immune cells near axonal terminals and vascular targets, the role of brain modulation and generation of inflammatory factors in cardiovascular disease is significant and relevant [1].

The heart-brain axis (HBA) comprises a constellation of physiologic interactions between the cardiovascular and nervous systems. Bidirectional communication between the brain and the immune system is carried out through a complex network of autonomic nerves, hormones, and cytokines, influencing the pathogenesis of conditions such as hypertension, congestive heart failure (CHF), atrial fibrillation, and metabolic syndrome. For example, pro-inflammatory cytokines in the central nervous system impact sympathetic outflow, arterial pressure and cardiac remodeling in hypertension and CHF. Importantly, neurologic dysregulation can impair cardiac function, as seen in stress/emotion-mediated cardiac dysfunction (Takotsubo-cardiomyopathy) and stroke-related cardiac syndrome (stroke-heart syndrome). Conversely, cardiac dysregulation can impair neurologic function, as seen in arrhythmia-related neurologic complications, myocardial ischemia-related neurologic complications, and global microvascular compromise. While the existence of co-incident cardiac and neurologic disorders has been known for decades, emerging pathophysiologic evidence in the past decade has revealed insights into how these disorders develop in unison. There are instances where new paradigms have replaced older causative paradigms of disease etiology. We will begin with a review of the general anatomy of and physiology of the HBA.

Anatomy of the Heart Brain Axis

Neurons are distributed in the central nervous system (brain) and the peripheral nervous system (organs, skin, and vasculature). The efferent (“from the brain to periphery”) pathway begins with the medial prefrontal and insular cortex and culminates in cardiac myocytes and coronary epithelial cells. From top to bottom, the “levels” of the brain are the cortex, thalamus, hypothalamus, cerebellum, and brainstem, which in turn consists of the midbrain, pons, and medulla (Figure 1). The cortical level is responsible for high-level functions. Among the cortical areas, the prefrontal, insula, somatosensory cortices, and limbic system are particularly significant in the HBA. As part of the limbic system, the amygdala and hippocampus regulate emotion, memory and context with functional connections with the prefrontal cortex. The paraventricular nucleus of the hypothalamus (PVN) of the hypothalamus is the main effector of the hormonal stress response and can activate the hypothalamic-pituitary-adrenal axis by secreting corticotropin. Within the medulla, the rostral ventrolateral medulla (RVLM) is the final common efferent pathway for acute BP and blood flow increases, which derive from cardiac inotropic and chronotropic responses and peripheral arterial and venous constriction. In contrast, the Dorsal Vagal Nucleus (DVN) can receive input from the nucleus tractus solitarii (NTS) and promote parasympathetic signaling to the heart. [2]

Figure 1:

Neural pathways involved in the brain-heart axis [92]. Sagittal view of the brain, showing the cerebral cortex, the thalamus, the hypothalamus, and the brainstem, which includes the midbrain, pons, and medulla oblongata (bottom left). The afferent (“from bottom to top”) pathway begins with input from chemoreceptors and baroreceptors and culminates in the insular cortex (bottom right). The efferent (“from top to bottom”) pathway begins with the medial prefrontal cortex and insular cortex and culminates in cardiac myocytes and coronary epithelial cells. Abbreviations: ACTH: adrenocorticotropic hormone; Hypothal: hypothalamic. Original figure created with BioRender.com.

Conversely, the afferent (“from the body to the brain”) pathway begins with input from chemoreceptors and baroreceptors and culminates in the insular cortex. Chemical and pressure signals from chemoreceptors and baroreceptors arrive at the thalamus, the brain’s center for relaying sensory signals to the cortex. The thalamus relays peripheral physiologic responsive signals to the insular cortex, a key integrator of hemodynamic regulation. [2]

Central and Peripheral Neurogenic Pro-Inflammatory Considerations in the Heart-Brain Interaction

It is well known that cytokines have effects on neuronal action in neurologic diseases such as epilepsy and cerebral ischemia [3] [4]. In cardiovascular disease, cytokines play a role in the co-morbid presentations of hypertension, CHF, metabolic syndrome, and stroke [5] [6] [7]. Several clues led to the recognition of the role of neurogenic pro-inflammatory processes in cardiovascular disease. There is evidence from various clinical reports that patients diagnosed with cardiovascular disease have higher levels of pro-inflammatory cytokines in their bloodstream, such as interleukin (IL)-1α, IL-6, TNF-α, and CRP [8] [9]. The administration of pro-inflammatory cytokines or anti-inflammatory cytokines into the central nervous system significantly impacts sympathetic outflow [10], arterial pressure and cardiac remodeling in models of hypertension and heart failure. Animal models have shown a correlation between hypertension and levels of proinflammatory cytokines. Both pro and anti-inflammatory cytokines interact with the brain's renin-angiotensin system (RAS) components to regulate blood pressure [11]. In the sections that follow, we describe the role of central and peripheral neurogenic properties in hypertension, CHF, and metabolic syndrome.

Hypertension

Glial cells and neurons produce pro-inflammatory cytokines and contribute the development of hypertension via vascular inflammation and activation of the renin-angiotensin-aldosterone system [7, 12-15]. Chronic overstimulation of the sympathetic nervous system affects the heart, peripheral vasculature, and kidneys, leading to increased blood pressure, cardiac output, vascular resistance, stiffness, and fluid retention [16-18]. These sympathetic efferent transmissions from the brain are affected by the PVN of the hypothalamus, the rostral ventral lateral medulla, and projections from the insular cortex to the nucleus of the solitary tract to the peripheral tissues [19, 20]. Given that vascular resistance and arterial stiffness are increased in hypertension, it is notable that these CNS regions demonstrate high expression of Ang-II type 1 receptors [7]. End organ damage in hypertension studies reveals that inflammatory factors such as T cells and macrophages accumulate in the kidneys and peripheral vasculature [21].

Conversely, these inflammatory cytokines originally generated in the brain can return from the periphery and disrupt the BBB [22] [23] [24] [25], enabling Ang-II to access the hypothalamus, medulla, and insular cortex/NTS [26]. Ang-II leads to elevation in BP via activation of nuclear factor kappa B (NFκB) in the hypothalamus/PVN [27], in conjunction with tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β), which heighten adrenocorticotropic hormone release (ACTH) and augment sympathetic outflow [28] [29] [30].

Congestive heart failure

The progression of CHF involves sustained inflammatory signaling and neurohormonal activation, with dysregulation of inflammatory cytokines such as TNF-α, IL-1, IL-6, IL-8, IL-10, MPO, iNOS, and CRP [30]. Sustained sympathetic nervous activity and RAAS activity have been demonstrated across the span of CHF from chronic preserved and reduced ejection fraction to cardiogenic shock [31] [6]. Hypothalamic nuclei such as the PVN and medullary nuclei like the RVLM and caudal ventrolateral medulla (CVLM) are key interfaces that promote the generation and interchange of inflammatory cytokines between the brain and the heart in CHF [32]. In fact, levels of circulating cytokine correlate with the severity of heart failure and prognosis [33]. Pro-inflammatory cytokines, such as tumour necrosis factor alpha (TNF-α), interleukin-1β, interleukin-2 and interleukin-4, can induce pulmonary edema, ventricular contractility abnormalities and perturbations in cardiac metabolism which may result in reduced cardiac function [34]. Injecting pro-inflammatory cytokines directly into the medullary nuclei can replicate the effects of circulating cytokines on sympathetic nerve activity [10].

In CHF, the activation of the renin–angiotensin system serves as a modifier of vascular and cardiac remodeling of the heart, fluid and electrolyte imbalance [31]. Angiotensin is local generation in the heart and brain in addition to its systemic production. Additionally, peripherally produced angiotensin II acting on CNS brainstem nuclei that lack a blood-brain barrier (e.g., the subfornical organ (SFO)) can elicit centrally mediated actions [32]. Thus, angiotensin II can cause an increase in sympathetic nerve activity, affecting both central and peripheral areas.

Despite the promise suggested by preclinical trials that targeting pro-inflammatory cytokines may be clinically useful in treating heart failure, clinical trials attempting to inhibit the actions of these cytokines have not met this expectation [35]. Trials targeting TNF-α using etanercept or infliximab in heart failure were stopped early due to lack of benefits or even worse patient outcomes. However, the CANTOS trial found that using canakinumab to inhibit interleukin-1β function reduced hospitalizations and mortality due to heart failure [36]. There is a strong connection between inflammation and heart failure; they are synergistic. Once this connection is sustained, therapeutic intervention is more challenging. It is likely important to control the inflammatory process early to prevent chronic inflammation and heart failure.

Metabolic Syndrome

Metabolic syndrome has been linked to the development of neurodegenerative diseases and decreased brain plasticity, as metabolic syndrome can cause systemic inflammation and increased permeability of the blood-brain barrier (BBB), with a corresponding increase in toxins, immune cells, and pathogens entering the brain. [37] [38] [39] This can disrupt neuronal function and cause hormonal dysregulation, increased immune sensitivity, and cognitive impairment depending on the affected brain region. Brain-derived neurotrophic factor (BDNF) plays a key role in maintaining energy balance and regulating cardiovascular function [40]. Animal studies suggest that hyperphagia and aggressiveness can be altered by altering BDNF levels in the brain and the peripheral circulation [41]. The control of food intake, glucose regulation and metabolism are dependent upon a fine balance between central regulatory inputs (primarily orchestrated by the hypothalamus) and a multitude of peripheral signals, such as insulin, adipokines (e.g., leptin, adiponectin, resistin), and gut hormones (e.g. cholecystokinin and ghrelin). The crosstalk between these pathways subserves the control of energy balance, stress responses and cardiovascular function.

Until recently, studies investigating the hypothalamus as a critical brain region for regulating energy homeostasis were primarily focused on the neuronal component of the hypothalamus. Recent studies have uncovered the vital role of glial cells as an additional player in energy balance regulation. It has recently been shown that inflammatory activation of glial cells in the hypothalamus under metabolic stress conditions contributes to various metabolic diseases. Infiltration and recruitment of monocytes and macrophages in the hypothalamus aids and maintains such inflammation and exacerbates the chromic manifestation of the disease. The hypothalamus is also responsible for the causation of systemic aging under metabolic stress [42]. Chronic, low-grade inflammation impairs energy balance and contributes to altered feeding habits, thermogenesis, and insulin and leptin signaling, eventually leading to metabolic disorders, such as diabetes, obesity, and hypertension. A better understanding of the multiple factors contributing to hypothalamic inflammation, the role of the different hypothalamic cells, and their crosstalk may help identify new therapeutic targets.

Effect of Neurologic Injury and Neurologic Risk Factors on Cardiac Function

The brain can exert a variety of effects that influence cardiac function through signaling via the SNS, PNS, RAAS, and HPA systems described above. The most well-known of the brain-heart relationships is that of stress-mediated cardiomyopathy, or Takotsubo cardiomyopathy. Studies conducted on Takotsubo have shown that certain pro-inflammatory cytokines, including tumor-necrosis factor-α and interleukin-6, can cause a lengthening of the action potential duration and QT prolongation [43] [44]. However, cardiomyopathy can also occur from stroke, seizure, and use of neuromodulating drugs. It is important to consider these as syndromes rather than diseases of the brain and heart in isolation, to improve diagnosis and management, especially as they often share underlying risk factors. We review some of these neuro-triggered cardiovascular syndromes in more detail below.

Stroke-Related Cardiac Dysfunction

Stroke, including both ischemic stroke and hemorrhagic stroke, involves compromise of the blood brain barrier and disrupts normal autoregulation of cerebral vascular tone, in a process called neurovascular uncoupling. After a stroke, patients experience a milieu of cardiac abnormalities, collectively known as stroke-heart syndrome, the incidence and severity of which peaks within the first 3 days post-stroke [45]. QT prolongation can occur in 20-65% of post-stroke patients, ST segment changes in 15-25%, and inverted “cerebral” T waves in 2-18% [46, 47]. Heart rate variability is reduced, and impairment in the baroreceptor reflex predisposes patients to acute hypertensive crises post-stroke [48]. While post-stroke patients experience these abnormalities, they are also at elevated risk of cardiac events. In fact, cardiovascular complications are the second leading cause of death in the post-stroke setting. After a subarachnoid hemorrhage, more than half of patients develop LV diastolic dysfunction even if the origin of the stroke is non-cardioembolic [49]. After an ischemic stroke, a third of patients may develop reduced ejection fraction [50].

Stroke-related cardiac dysfunction may occur by several mechanisms, including the HPA axis, inflammatory mediators, microRNA, and microvesicles. First, stroke activates the HPA axis, leading to increased catecholamine release into the circulation, including at myocardial nerve endings. In the immediate setting, the increased sympathetic tone can cause arrhythmias. In the longer term, the increased catecholamine exposure can lead to cardiomyocyte necrosis, hypertrophy, and fibrosis, because prolonged mitochondrial calcium overload triggers oxidative stress, osmotic swelling, and loss of ATP synthesis. Second, stroke promotes microgliosis, astrogliosis, endothelial cell activation, and secretion of cytokines and chemokines [51]. Some of these inflammatory cells, including neutrophils, can infiltrate myocardium post-stroke [52], suggesting that a non-specific inflammatory response may mediate cardiac manifestations post-stroke. Third, stroke alters the levels of cardiovascular microRNAs. MicroRNAs are short sequences of noncoding RNA which have the ability to regulate gene expression at the time of transcription and post-transcription. Deficiency of microRNA-126 has been shown to be associated with HF, AF, and CAD [53, 54]. Stroke decreases the level of endothelial cell-specific microRNA-126 for at least 24 weeks [55]. Mice with conditionally knocked out microRNA-126 developed increased cardiomyocyte hypertrophy, fibrosis, inflammatory factor expression, and cardiac dysfunction post-stroke compared to control mice, suggesting that decreased microRNA-126 expression after stroke may mediate cardiac dysfunction post-stroke [56]. Fourth, stroke increases the levels of circulating cellular microparticles, including microvesicles containing interleukin-6 (IL-6) [57]. Release of IL-6 induces vasospasm [58].

Stress-Related Cardiac Dysfunction

Cardiac dysfunction can also be precipitated by mental stress. In a famous study of the association between emotional stress and cardiac events in Germany, on days when the German national football team was playing at the World Cup, presentations for acute coronary syndrome or symptomatic arrhythmia had an increased incidence of 3.3 for men and 1.8 for women compared to control non-World-Cup periods [59]. The performance of mentally stressful tasks such as simulated public speaking can trigger transient asymptomatic wall motion abnormalities at lower heart rates than that in exercise-induced ischemia [60]. In patients with pre-existing epicardial CAD, mental stress-induced myocardial ischemia (MSIMI) has been shown to occur at lower levels of myocardial work (at a lower rate-pressure product) than that in exercise-induced ischemia [61].

Mental stress-related cardiac dysfunction has molecular correlates. Heightened cortisol response to mental stress is associated with detection of high-sensitivity troponin T even in healthy subjects independently of the presence of CAD [62]. In the World Cup example aforementioned, the relationship between emotional stress and cardiac events was mediated at the molecular level by increased levels of soluble CD40 ligand (sCD40L) and soluble vascular cell adhesion molecule (sVCAM-1), reflecting increased endothelial dysfunction, and of tumor necrosis factor-alpha (TNF-α) and monocyte chemoattractant protein (MCP)-1, molecules which have a central role in plaque rupture [63].

As mentioned above, the most well-recognized disorder in the HBA is Takotsubo syndrome, or stress cardiomyopathy, in which patients with a mentally or physically stressful precipitating event develop sudden onset congestive heart failure, typically with apical ballooning of the left ventricle. In patients with stress cardiomyopathy, catecholamine levels exceed levels in MI [64]. Excess catecholamine stimulation of cardiomyocytes disrupts calcium homeostasis and leads to hypercontraction of sarcomeres with increased oxidative and metabolic stress and impairment in coronary microcirculation [43]. In the long term, elevated catecholamine release leads to histologically visible alterations in cardiomyocytes, including inflammatory cell infiltration, increased extracellular matrix protein levels, and contraction band necrosis, known as myocytolysis. [44] More generally, chronic congestive heart failure is recognized as a state of persistently activated sympathetic system leading to myocytolysis, LV hypertrophy, excitation-contraction coupling, and beta-adrenergic receptor desensitization, leading to myocyte necrosis and apoptosis [65]. For this reason, neurohormonal blockade is a mainstay of guideline-directed medical therapy for heart failure.

Cardiac Dysfunction from Other Neurologic Conditions

Apart from stroke-related cardiac dysfunction and stress-related cardiac dysfunction, cardiac dysfunction may also arise from seizures, spinal cord injury, baroreflex failure, Guillain-Barre syndrome, and use of neuromodulating drugs. Seizures that involve the medial parietal frontal cortex, insular cortex, or amygdala can cause tachyarrhythmias via sympathetic hyperreactivity [66]; whereas seizures involving the left temporal lobe can cause ictal bradycardia, syncope, and asystole [67]. Cervical or thoracic spinal cord injuries above the T5 level can cause autonomic dysreflexia that persists after recovery from acute spinal shock, manifesting as severe hypertension that may lead to hypertensive encephalopathy with intracranial or retinal hemorrhage or sudden cardiac death [68].

Effect of Cardiac Injury and Cardiovascular Risk Factors on Neurologic Function

The converse of neurologically triggered cardiovascular syndromes is also true: while the brain can influence cardiac function through the mechanisms described above, the cardiovascular system can also directly lead to neurovascular pathologies. The development of cognitive dysfunction in individuals with cardiovascular disease has important clinical implications for their longitudinal cardiovascular care, as comorbid dementia presents challenges for medication adherence and persistence, as well as invasive cardiovascular procedures which carry higher risks of in-hospital mortality and complications in those with underlying cognitive impairment [69] [69]. Perhaps the most well-known heart-brain relationships is that of atrial fibrillation (AF)-related thromboembolism resulting in ischemic stroke [70]. Thus, we will now discuss the pathophysiology of AF-related stroke, MI-related stroke, MI-related cognitive impairment and dementia, and MI-related delirium.

Arrhythmia-Related Stroke

AF is the most common cardiac arrhythmia, affecting two million people in the US [71] and 33 million people globally [70]. Traditionally, it has been understood that because atrial fibrillation causes atrio-ventricular dyssynchrony and atrial hypokinesis, blood stasis and activation of Virchow’s triad results in formation of thromboembolism in low-flow areas such as the left atrial appendage [72]. The contemporary model of cardiac mechanisms of stroke emphasizes two subtle distinctions (Figure 2). First, AF is considered to be a manifestation of abnormal atrial substrate, which is the causal etiology of both AF and stroke, as abnormal atrial substrate has been shown to lead to stroke regardless of the presence of AF [70]. Consistent with this model, increased left atrial size [73], ECG-defined left atrial abnormality [74, 75], paroxysmal supraventricular tachycardia [76], and premature ventricular contractions [77] have been associated with development of stroke independently of AF, and often precede the appearance of AF. Structural remodeling (e.g. fibrosis, atrial dilatation) is a process that would require multiple weeks of sustained AF [78], not just the six minutes of AF that signify an increased stroke risk [79].

Figure 2:

Relationship between AF and development of stroke. In the old paradigm, AF was considered a causative factor of stroke, with left atrial size being merely a surrogate marker for chronicity of AF. In the new paradigm, AF is considered to be a manifestation of abnormal atrial substrate, which can lead to development of stroke independently of AF. Cerebral vascular risk factors and central vascular risk factors predispose both to abnormal atrial substrate and development of stroke. Original figure created with BioRender.com.

Second, aging and systemic vascular risk factors cause mechanical changes in atrial and cerebral substrates that lead to stroke independently of AF. Curiously, a single, short episode of AF is associated with a two-fold increase in stroke risk in older adults [79], but not in young adults [80], suggesting that there are age-related biologic factors which modify the relationship between AF and stroke. In older adults, age related vascular and atrial abnormalities such as fibrosis, endothelial dysfunction [81] and chamber dilatation at left atrial appendage may lead to stroke incidence due to mural thrombus formation, independently of AF [82]. The above-mentioned changes can also cause AF due to structural alterations in atrial wall, which can directly result in stroke onset. So this can be a reason why geriatric population has an increased risk of post AF Stroke onset as compared to young individuals [82].

By acknowledging the role of aging, systemic vascular risk factors, and atrial substrate abnormalities as causes of thromboembolic stroke, we may better understand why a third of ischemic strokes have no obvious cause, and why only one third of patients with cryptogenic stroke after 3 years of continuous rhythm monitoring via loop recorder [83]. Treatment of cardioembolic stroke should include addressing systemic vascular risk factors and causes of atrial substrate abnormalities, such as metabolic syndrome and obesity. While clinicians might be inclined to discontinue anticoagulation during stretches of normal sinus rhythm, the contemporary model suggests that the decision should also take into consideration the patient’s systemic vascular risk factors and whether causes of atrial substrate abnormalities have been addressed [70].

Myocardial ischemia-Related Stroke

While stroke is a well-known complication of AF, stroke can also be a complication of MI. An analysis from the Global Registry of Acute Coronary Events (GRACE) registry showed that while post-MI stroke was rare, at 0.9%, it carried a high mortality rate of 32.6%. The incidence of stroke is higher with ST-Elevation MI (STEMI) compared with non-STEMI (NSTEMI) or Unstable Angina (UA), with an incidence of 1.3%, 0.9%, and 0.5% (p<0.001) for STEMI, NSTEMI, and UA respectively [84]. The incidence of stroke is significantly lower in the 30 days post-PCI (0.4%) than in the 30 days post-CABG 3 [85]. Traditionally, stroke has been considered a complication of MI if it occurs within 30 days of the stroke [86], but recent evidence suggests that patients with acute MI are at elevated risk for ischemic stroke compared to patients without acute MI up to 12 weeks after their acute MI [87]. Notably, the effect was present after accounting for periprocedural (PCI- and CABG-related) strokes, and after exclusion of patients with prior or concurrent diagnosis of AF [87], suggesting a pathophysiologic link between MI and stroke independent of AF and independent of procedures. The pathophysiologic mechanisms linking MI and stroke involve changes in cerebral blood flow, neuronal integrity, synaptic transmission, BBB permeability (Figure 3).

Figure 3:

Relationship between acute MI and development of stroke. In the old paradigm, acute MI predisposed to development of LV thrombus which led to stroke. In the new paradigm, it is recognized that stroke can develop post-MI regardless of whether an LV thrombus is present. This occurs through multiple mechanisms, including decreased cerebral blood flow from decreased cardiac output, neuronal injury from microRNA and neuroinflammation, altered synaptic transmission from microRNA and neuroinflammation, increased blood brain barrier permeability from upregulation of angiotensin II, DAMPs, and pro-inflammatory cytokines, and activation of astrocytes, microglia, choroid plexus, and endothelial cells. Original figure created in BioRender.com.

Myocardial ischemia occurs due to supply-demand mismatch of oxygen to myocardium, stemming from a heterogeneous spectrum of pathologic processes including but not limited to stable ischemic heart disease in the context of functional demand, spontaneous coronary artery dissection, coronary thromboembolism, and Acute Coronary Syndromes (ACS) due to acute atherosclerotic plaque rupture [88]. At a hemodynamic level, the decreased cardiac output resulting from acute MI results in decreased cerebral blood flow. At the molecular level, stressed and dying cardiomyocytes release damage-associated molecular patterns (DAMPs) that can be recognized by pattern recognition receptors (PRRs) on immune cells, leading to a robust systemic inflammatory response through upregulation of cytokines, chemokines, and interleukins in the brain within minutes [89]. The circulating pro-inflammatory cytokines, DAMPs, along with angiotensin II released from ischemic cardiomyocytes, reduce the expression of junctional proteins such as occluding, claudin-5, and ZO-1, compromising the integrity of the BBB [90]. The heightened neuroinflammatory state can persist up to 8 weeks after MI, even after the initial peripheral inflammation has attenuated [89].

Moreover, systemic inflammation induces a prothrombotic state from a variety of factors, including an imbalance between procoagulants and anti-coagulants that is caused by endothelial and cellular activation, Interleukin 1 beta, Tumor Necrosis Factor (TNF), and adhesion molecules such as P-Selectins, E-selectins, Intracellular adhesion molecule-1 (ICAM-1), Vascular Cell Adhesion Molecule-1 (VCAM-1) and interferons. For example, activated platelets also increase stroke risk by releasing P-Selectin that causes aggregation of platelets and the blood clot formation which can lead to cerebral arterial occlusion and stroke incidence [91]. Consistent with this, the platelet to lymphocyte ratio (PLR) can be used a measure of thrombotic risk after MI [92] with an increased PLR associated with an increase in the incidence of stroke. Increased Matrix Metelloprotein-9 (MMP-9) release by lymphocytes can cause degradation of the blood brain barrier, which enhances the risk of hemorrhagic stroke and also increases overall mortality [93]. While anterior MI is known to have the potential to involve to aneurysmal changes in the left ventricular wall resulting in hypokinesis, which increases the risk of left ventricular thrombus formation and, thus, embolization leading to stroke, stroke can occur in MI of any territory. By acknowledging the role of inflammatory cytokines and miRNAs changes in cerebral blood flow, neuronal integrity, synaptic transmission, BBB permeability, we may better understand why half of post-MI strokes occur in non-anterior Mis.

Myocardial Ischemia-Related Dementia and Cognitive Impairment

Even in the absence of stroke, meta-analysis shows that MI is prospectively associated with increased risk of developing cognitive impairment or dementia [94]. The pathophysiologic link between MI and cognitive impairment or dementia may be mediated by shared vascular risk factors, increased platelet activity, miRNA-mediated injury. First, ACS and cognitive impairment share common vascular factors such as hypercholesterolemia, diabetes, hypertension, smoking, and genetic factors such as mutations in the APOE-e4 allele [95]. The importance of vascular contributions to dementia and cognitive impairment was recently recognized in an AHA/ASA scientific statement [96].

Second, platelets, which are upregulated in ACS, contain P-selectin and GPIIbIIIa receptors which facilitate platelet adhesion not just at sites of coronary plaque rupture but also at sites of cerebrovascular lesions, which can trigger a cascade of perivascular inflammation and neurologic injury [97]. Activated platelets can also cause cerebral vasoconstriction that may lead to small brain infarcts in the hippocampus, accelerating progression to dementia [98]. Moreover, experiments in mice have shown that platelets are responsible for the deposition and accumulation of beta-amyloid in blood vessels after a thrombotic event [99]. Beta-amyloids are peptides resulting from proteolytic cleavage of amyloid precursor proteins (APPs) by beta- and gamma-secretase and can cross the blood brain barrier. The aggregation of beta-amyloid in the perivascular space of brain areas associated with memory like hippocampus and amygdala can cause senile plaques that are characteristic of Alzheimer’s dementia [100].

Third, microRNAs released by the heart in ACS have been found to affect neuronal function. MicroRNAs are small, non-coding RNA sequences that have the ability to inhibit protein synthesis by interacting with the 3’ untranslated region of messenger RNAs (mRNAs) that code for proteins [101]. miR-1, miR-133a/b, have been shown to be consistently decreased, and miR-208a/b, and miR-499 have been shown to be consistently increased in the serum of humans and animals after acute MI [102-104]. miR-1, miR-133a/b, miR-208a/b, and miR-499 have been shown to modulate neuronal apoptosis, microtubule dissolution, synaptic transmission, and demyelination [90]. Further studies are ongoing in an effort to link these molecular changes to clinical dementia in post-MI patients.

Delirium & the Heart

Patients with MI are also susceptible to developing delirium, a fluctuating, transient psychological state that affects cognition, behavior, speech, attention, consciousness, orientation, and social interactions. By definition, it is almost always triggered by an underlying disorder and most commonly impacts the older adults population [105]. Delirium has important clinical consequences, including increased hospital stay, health expenditures, and poor clinical outcomes with increased burden of morbidity and mortality [105, 106] Patients hospitalized for cardiovascular diseases are at significantly higher risk of developing delirium, which complicates their subsequent management and outcomes [107] In a prospective cohort study, seven percent of patients with non ST elevation MI (NSTE-MI) developed delirium during their hospitalization, with even higher rates among octogenarians [108]. In a study, 28.8% of the subjects (>65 age) undergoing primary PCI, suffer delirium post-procedure. Importantly, in-hospital delirium was identified as a key predictor of 6-month mortality [108].

The pathophysiologic link between ACS and delirium may involve decreased cardiac output and cerebral perfusion, procedure-related and anesthesia-related disorientation, and use of narcotics for pain control. At a hemodynamic level, in ACS, reduced cardiac output can lead to compromised cerebral perfusion, potentially damaging grey areas that include the left medial frontal cortex, left cingulate and precuneus, left and right parahippocampal gyri and right and left middle temporal gyri [109]. As described above, ACS mounts an inflammatory response with endotoxins, cytokines, and chemokines that can cause neuroinflammation and lead to neurological changes that manifest as delirium after the procedure [90]. At a clinical level, factors that influence the risk of developing delirium among ACS patients include older age, prolonged hospital stay, coronary intervention, medication side effects and polypharmacy [106]. For example, narcotics are frequently utilized for pain control in patients with ACS, especially those undergoing percutaneous coronary intervention (PCI), but increase the risk of delirium [110]. Nitroglycerin is used to manage angina pain and to reduce preload in patients with ACS, but also increase the risk of delirium, possibly by interfering with the cerebral blood flow, glucose, and lactate supply [111].

Delirium has important consequences for the post-MI patient. Delirium can prolong the procedural duration of PCI, while increasing procedural pain, peri-procedural risk and length of hospital stay, which in turn can increase the frequency and intensity of delirium itself [106]. In addition to prolonged procedural duration, an agitated patient poses a risk to both themselves and the clinicians caring for them. Delirium is associated with an increased hospital stay, poor outcomes in ACS patients, functional decline, and low Quality of Life (QOL), increased morbidity and mortality. For every passing 48 hours of in-hospital delirium, mortality is increased by 11% [112] [113]. The most effective treatment for delirium is the prevention of delirium onset. Controlling pain, fever, and other stressors are of prime importance. Sensory impairments should be addressed and access to newspapers, radios, and television should be encouraged. The patient should be frequently re-oriented to both time and place. Physical restraints and invasive procedures should be minimized in patients at high risk for delirium. Early mobilization is a key factor in improving cardiac outcomes as well as delirium [107].

Conclusion

In conclusion, the HBA is greater than the sum of its component parts. Advances in the past two decades have revealed complex crosstalk between the two organ systems, mediated by the SNS, PNS, RAAS, and HPA axes at the pathway level, and via miRNA, cytokines, and neurotransmitters at the molecular level. Central and peripheral neurogenic pro-inflammatory factors play a major role in the Heart-Brain interaction, manifesting as bi-directional relationships in cardiovascular disorders such as hypertension, metabolic syndrome, and congestive heart failure. While stress-mediated cardiomyopathy is the prototypical disorder of the brain-to-heart axis, cardiac dysfunction can result from a myriad of neurologic insults including stroke, seizures, spinal cord injury, baroreflex injury, and use of neuromodulating drugs as well. Meanwhile, in the heart-to-brain axis, it is now recognized that atrial fibrillation is not necessarily a causative factor for cardioembolic stroke, but a manifestation of abnormal atrial substrate, which can lead to the development of stroke independently of AF. Similarly, it is recognized that MI can lead to the development of stroke independently of LV thrombus formation, due to the effects of miRNA-mediated neuronal injury and cytokine-mediated neuroinflammation. Post-MI cognitive impairment and dementia is recognized to have a molecular and cellular basis and is no longer merely an epidemiologic curiosity. Future research should be aimed at investigating whether these pathophysiologic insights can be translated into differences in management. Ongoing experiments involving catecholamine receptor blockade, free radical scavengers, antioxidants, targeted anti-inflammatory agents and miRNA inhibitors may someday benefit patients with disorders in the HBA.