Neurocrit Care

https://doi.org/10.1007/s12028-018-0636-3

REVIEW

Cardiac Dysfunction in Neurocritical Care:

An Autonomic Perspective
Mohammad S. Ibrahim, Bennson Samuel, Wazim Mohamed and Kushak Suchdev*

© 2018 Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society

Abstract
A number of neurologic disorders can cause cardiac dysfunction by involving the conductive system and contractile
apparatus of the heart. This is especially prominent in the neurocritical care setting where the spectrum of cardiac
dysfunction due to acute neurologic injury ranges from trivial and isolated electrocardiographic changes to malig-
nant arrhythmias and sudden death (Table 1). The mechanism of these cardiac complications is complex and not
fully understood. An understanding of the neuroanatomical structures and pathways is of immense importance
to comprehend the underlying pathophysiology that culminates as cardiac damage and dysregulation. Once the
process is initiated, it can complicate and adversely affect the outcome of primary neurologic conditions commonly
seen in the neurocritical care setting. Not only are these cardiac disorders under-recognized, there is a paucity of data
to formulate evidence-based guidelines regarding early detection, acute management, and preventive strategies.
However, certain details of clinical features and their course combined with location of primary neurologic lesion on
neuroimaging and data obtained from laboratory investigations can be of great value to develop a strategy to appro-
priately manage these patients and to prevent adverse outcome from these cardiac complications. In this review, we
highlight the mechanisms of cardiac dysfunction due to catastrophic neurologic conditions or due to stress of critical
illness. We also address various clinical syndromes of cardiac dysfunction that occur as a result of the neurologic illness
and in turn may complicate the course of the primary neurologic condition.
Keywords: Autonomic nervous system, Sudden death, Neurocardiology, Cardiac dysfunction in critical care

Background and Historical Landmarks In the late nineteenth century, there were reports of an
The concept of psychological stress causing sudden car- increasing number of intraoperative deaths in England
diac death can be found in anthropological literature. The and Wales. Initially, these deaths were attributed to car-
first detailed scientific account of these ‘psychosomatic dio-inhibitory effect of chloroform overdose. However,
deaths’ was by Cannon et al. in his famous 1942 publi- the sequence of events leading to these deaths described
cation “Voodoo Death.” The mechanism proposed by in case reports in 1890s revealed that deaths occurred
Cannon et al. to explain these ‘cardiac’ deaths was neu- during the recovery phase from chloroform anesthe-
rogenic, specifically, a severe and persistent sympatho- sia rather than deep sedation. Interestingly, the deaths
adrenal response [1]. Half a century before Cannon’s ensued when a strong painful stimulus was applied to
work, the association of sympathetic stimulation and the patient during lighter phase of sedation [2]. Armed
sudden death already had begun to unfold, but in a dif- with this knowledge, Levy et al. performed a series of
ferent context. experiments on cats and demonstrated that sympathetic
activating maneuvers, such as intravenous adrenaline
administration, electrical stimulation of sympathetic
cardiac nerve or emotional stress, all of which triggered
*Correspondence: ksuchdev@med.wayne.edu
Department of Neurology, Division of Neurocritical care, Wayne State ventricular fibrillation. More importantly, these effects
University School of Medicine, Detroit, MI, USA were seen exclusively during light chloroform sedation,
suggesting that a neural mechanism was responsible for culminates in release of excessive catecholamine at car-
these arrhythmias and not chloroform overdose [3, 4]. diac post-synaptic adrenergic receptors.
A decade later, Neuberger et al. reported a case series of In order to understand the mechanisms by which
five patients who died during seizures and were found to neurologic disorders change the autonomic balance, we
have myocardial lesions with normal coronary vessels [5]. briefly discuss the neuroanatomical pathways between
Additionally, in animal models, myocardial lesions with- the nervous system and heart.
out involvement of coronary vessels could be produced
by imparting stress to metabolically deranged rats [6, 7]. Neuroanatomical Perspective
These lesions were pathologically different from liquefac- Introduction
tive necrosis of acute myocardial infarction, and adre- The autonomic nervous system is a widespread network
nalectomy did not prevent development of the lesions of nerves that connects central nervous system to the vis-
[7]. Raab et al. also demonstrated that reserpine, a drug ceral organs. Sympathetic and parasympathetic fibers are
that blocks catecholamine release from presynaptic adr- the peripheral efferents of this neural network that inner-
energic nerve terminals, prevented these stress induced vate and modulate their targets organs, such as receptors
lesions. However, drugs which blocked circulating cat- on the heart and vascular walls and smooth muscles of
echolamine did not have the same effect [6]. Addition- visceral organ. However, the origin and subsequent gov-
ally, in a clinical case series of patients who died acutely ernance of these peripheral efferents occur in cell bod-
from an aneurysmal subarachnoid hemorrhage, Green- ies clustered in nuclei, distributed at various locations in
hoot et al. found discrete necrotic lesions with interstitial central nervous system (Fig. 1). This forms a central neu-
hemorrhage in myocardium. Similar myocardial lesions ronal network that controls vital functions.
were reproduced in adult cats with stimulation of reticu-
lar formation in midbrain. Interestingly, these necrotic Sensory Nervous System of Heart (Fig. 1)
lesions were clustered around nerve endings of the heart Sensory receptors in the heart are distributed throughout
[8]. myocardial, pericardial and epicardial tissue and at the
All these findings in animals and clinical studies point junction of atria and great veins. These receptors are acti-
toward sympathetic overactivity, either due to stress or vated by pressure in the heart chambers, and impulses
severe neurologic damage, which initiates a process that are transmitted, predominantly via vagal afferent to

Fig. 1 Autonomic-cardiac network: a afferent connections from sensory receptors of heart travels via IX and X cranial nerves to nucleus solitary
tract in medulla. Periventricular nucleus of hypothalamus is an integrating center as it has reciprocal connections (bidirectional arrows) with auto-
nomic nuclei from brain stem, other hypothalamic nuclei and from cortex. b Efferent autonomic pathway to the heart. Again shown are reciprocal
connections of periventricular hypothalamus with other central autonomic nuclei in cortex, hypothalamus, brain stem, and spinal cord. Sympa-
thetic efferents from higher centers descend to intermediolateral columns of upper thoracic spinal cord, whereas parasympathetic signals travel via
vagus nerve
nucleus of solitary tract (NTS) in the medulla. NTS also Higher Centers and Descending Pathways (Fig. 1)
receives afferents from baroreceptors, chemoreceptors As discussed above, efferent vagal nerve fibers from the
and skeletal muscle receptors [9]. nucleus ambiguous and dorsal vagal nucleus transmit
parasympathetic signals to the heart [11]. Sympathetic
Intrinsic Nervous System of Heart input to the heart comes from neurons in the interme-
Centrally derived autonomic nerves, as discussed in diolateral (IML) column at T1 to T5 level of spinal cord.
subsequent sections, are considered the dominant regu- This IML column is modulated by descending neurons
lator of cardiac function. However, there is evidence sug- from rostral ventrolateral medulla and paraventricular
gesting presence of an intrinsic network of cell bodies nucleus (PVN) of hypothalamus [11].
with interconnecting fibers, collectively known as car- Parvocellular neurons in PVN of hypothalamus, also
diac ganglionic plexus (CGP). Dominantly cholinergic, known as pre-autonomic neurons, are the only auto-
the CGP has been considered as relay station to para- nomic neurons which are recipients of autonomic affer-
sympathetic efferents [10]. However, due to a number ents and send both sympathetic and parasympathetic
of factors, they have been increasingly recognized as a efferents to other autonomic centers. Thus, PVN is con-
nervous system intrinsic to the heart. Firstly, they con- sidered to be the control center, modulating and opti-
tain other neuromodulators such as nitric oxide (NO), mizing balance between two opposite responses [12].
vasoactive intestinal polypeptide, neuropeptide Y and Other hypothalamic nuclei that play an important
intermedin in addition to traditionally known auto- role in cardiovascular regulation are anterior and lat-
nomic neurotransmitters. Secondly, specific receptors eral hypothalamic nuclei. In animal studies, stimula-
for each of these neuromodulators are present in the tion of anterior hypothalamus produced bradycardia,
heart (Table 4). whereas an opposite response was seen with stimula-
Parasympathetic stimulation interacts with sym- tion of lateral hypothalamus. This suggests that ante-
pathetic stimulation to counter various aspects of rior hypothalamus has parasympathetic whereas lateral
arrhythmogenesis. Intrinsic CGP along with these neu- hypothalamus has a sympathetic influence over the
romodulators plays a role in this interaction as detailed heart [13].
in Table 4. These actions of CGP are mediated by both In higher cortical centers, especially the insula, con-
modulation of acetylcholine effects and effects of these trol of sympathetic and parasympathetic outflow is
neuromodulators on their specific receptors. lateralized in a complex manner. Studies of unilateral
hemispheric inactivation with ipsilateral intra-carotid
Extrinsic Nerve Supply of Heart amobarbital injection [14–16], intraoperative stimulation
Table 4 depicts neurotransmitters and autonomic recep- of insular cortex [17], and positron emission tomography
tors of heart and effects of their stimulation via sym- scans during physical and mental stress all revealed that
pathetic and parasympathetic nerve endings. Beta2 sympathetic response was lateralized to the right insula
receptors, located predominantly on the basal segments and parasympathetic activity to the left [18]. In studies
and less commonly on the apical segments of the heart, of patients with stroke, results are conflicting. Involve-
are the most dominant recipients of sympathetic nerve ment of insula on either side was associated with exces-
supply. This distribution of sympathetic receptors plays a sive sympathetic activity, but was more pronounced in
pivotal role in etiopathogenesis of neurologic and neuro- patients with right insular stroke [19, 20]. Additionally,
genic cardiac dysfunctions as described in more detail in in animal studies, sympathetic chronotropic activity
later sections. The sympathetic nerve supply to the heart was found to be represented at rostral posterior insula,
arises from cardiac plexus, a bundle of nerves at arch of whereas parasympathetic chronotropy localized to cau-
aorta. The nerves in cardiac plexus are post-ganglionic dal posterior insula [21].
projections arising from paravertebral sympathetic cer-
vical and upper thoracic ganglions. The paravertebral
sympathetic ganglions in turn receive descending input Pathogenesis and Pathophysiology
from higher centers which exit central nervous system Several different mechanisms have been proposed to
(CNS) via upper five thoracic spinal cord segments as explain neurogenic myocardial dysfunction. The most
preganglionic fibers (Fig. 1). Parasympathetic nerve sup- widely accepted of these is focal damage or dysfunc-
ply to the heart has its preganglionic neurons located tion of CNS structures controlling autonomic outflow
dominantly in nucleus ambiguus and to lesser extent in (Fig. 2) leading to sympathetic over-dominance which in
dorsal medial nucleus of vagus. Descending vagal path- turn affects myocardial rhythm and integrity. Myocar-
ways converge directly on CGP as described in the previ- dial autonomic innervation and its impact on myocardial
ous section. pathophysiology depend on multiple factors.
 Fig. 2 Pathophysiology of autonomic cardiac dysfunction

Physiology of Myocardial Contraction Heterogeneity of Ventricular Sympathetic Response

Diastolic interval (DI) is the resting state of myocyte A number of studies have shown that there is a hetero-
during which membrane potential inside the cell mem- geneity of sympathetic supply among various walls and
brane is more negative relative to the outside. This segments of myocardium. Base of the left ventricle is
polarized state is achieved and transiently maintained, richly innervated by sympathetic nerves [23]. Ng et al.
predominantly due to the outward flow of potassium demonstrated that during sympathetic stimulation, there
­(K+) through rectifier ­K+ channels (Fig. 3a). was shortening of APD that was significantly more pro-
DI ends when electrical impulses lead to the onset of nounced at the base. Furthermore, expression of delayed
action potential (AP) by rapid opening of fast inward rectifier potassium channels through which catechola-
sodium ­(Na+) channels and consequent rapid N ­ a+ mine produces shortening of APD was highly upregu-
influx into the cell. This initiates a new state, the sys- lated in basal area during sympathetic stimulation [24].
tole, which lasts for a duration corresponding to the
AP duration (APD). Electrographically, this dura- Effect of Sympathetic Stimulation (Fig. 3b)
tion is represented by QT interval, i.e., summation of Sympathetic stimulation via its effect on delayed rectifier
durations of QRS complex, ST segment, and T waves. potassium channels directly affects AP and reduces its
APD is maintained transiently as calcium ­(Ca++) flows ­ + chan-
duration [24]. In canine models, slow rectifier K
in through L type C ­ a++ channels. At the end of APD, nels do not contribute to change in APD during the rest-
delayed rectifier potassium ­(K+) channels slowly open ing state, but cause significant APD shortening during
­ + efflux, thus ending APD by restoration of
leading to K beta adrenergic stimulation [25]. Heart rate (HR), which
polarized state [22]. is the sum of DI and APD, tends to increase as both of its
 Fig. 3 a Top panel: diastolic interval (DI) is the resting (polarized) state of myocytes due to outward flow of K via slow rectifier channels. This period
ends with the onset of action potential leading to depolarization (systole) initiated by rapid influx of Na. This state is maintained by Ca influx and
ends by K efflux through slow rectifier channels. Action potential duration (APD) corresponds to QT interval shown in bottom panel. b Myocardial
action potential during ongoing sympathetic stimulation. Shown is a relatively shorter first beat (first AP cycle with minimal shortening of DI and
APD). For maintenance of constant HR, next cycle has relatively longer DI and APD, which in turn giving rise to much shorter DI and APD (AP restitu-
tion described in text). This sequence of long and short cycles (AP alternans) if continued due to ongoing sympathetic stimulation will ultimately
lead to an AP duration longer than refractory period thus setting forth wave break and potentially, malignant arrhythmias

components are shortened. Shortening of DI has another that was much less for a given increase in HR. As a result,
very crucial consequence (Fig. 3b). Through a feedback QT interval corrected for HR was prolonged [31].
mechanism attempting to keep HR under control, this AP alternans, described above is a precursor of malig-
shortened DI will lead to APD prolongation of next cycle nant arrhythmias and predictor of sudden death [32]. The
[26, 27]. This phenomenon of dependency of APD of a heterogeneity of sympathetic distribution among differ-
given cardiac cycle on DI of its preceding cycle is known ent walls and segments of ventricles can get exaggerated
as restitution of APD [28]. which promotes occurrence of AP alternans and conse-
At rest, the curve of this relationship of DI to APD quently malignant arrhythmias [33].
of the subsequent beat is relatively flat, i.e., magnitude Besides steepening of AP restitution curve via its effect
of change in APD is not significantly greater than the on delayed rectifier ­K+ channels, sympathetic stimulation
change in DI of previous cycle. With sympathetic stim- also significantly alters ­Ca++ levels by affecting L type
ulation, the slope of this curve steepens; a small change ­Ca++ channels. This leads to triggered activity as well as
in DI causes a marked change in AP duration of subse- after depolarizations, both of which are known precur-
quent cycle. If sympathetic stimulation is prolonged, this sors of ventricular arrhythmias.
phenomenon initiates a rhythm disturbance known as
AP alternans, i.e., a series of cardiac cycles with shorter Self‑Perpetuation of Sympathetic Effects
and longer durations alternating with each other (Fig. 3b) Increased sympathetic activity in the heart leads to a
[29]. phenomena, known as sympathetic neural remodeling
During this course of sympathetic stimulation, AP (SNR). This involves nerve sprouting and further hyper-
duration can be prolonged to the extent that a new stim- innervation of sympathetic nerves which has shown to
ulus may arise while myocardial cells are still in refrac- cause malignant arrhythmias and sudden death [34].
tory period of AP, leading to malignant arrhythmia by In rats that are subjected acute myocardial ischemia,
setting off a wave break or reentry [24, 29, 30]. Electro- the oxidative stress has shown to increase sympathetic
graphically, this process corresponds to prolonged QT activity and further stimulate SNR by promoting expres-
interval. Moreover, it also further affirms the clinical sion of nerve growth factor gene [35]. This mechanism
importance of correcting QT interval according to HR. can have potential therapeutic implications. Lithium,
Magnano et al. demonstrated that in 25 healthy subjects, for example, by activating antioxidant gene regulators,
isoproterenol infusion resulted in QT interval shortening has shown to inhibit NGF expression and thus protects
against ventricular arrhythmias. Intermedin, a member Clinical Features and Syndromes: Conductive
of calcitonin gene-related peptides that act via calcitonin Disturbances
receptors-like receptors and receptor activity-modifying Impaired Electrical Activity of Heart
protein (table), has also shown to inhibit oxidative stress Electrocardigraphic (EKG) changes are common in
and SNR in rats with heart failure due to acute MI. These patients with neurologic catastrophes. In patients with
effects in turn lead to reduced occurrences of ventricular acute stroke, repolarization abnormalities particularly
arrhythmias and heart failure [36]. QTc prolongation were most common. However, due
to similar vascular risk factors, a preexisting coronary
artery disease rather than neurogenic mechanism may
Degeneration and Necrosis of Myocytes be responsible for these EKG changes in a number of
Excessive catecholamine exposure results in increased patients. Yet, QTc prolongation and to some extent other
activity of cAMP in myocytes causing excessive C ­ a++ repolarization abnormalities, if present during acute
influx. Actin–myosin interaction can be prolonged phase of stroke, may be suggestive of a neurogenic mech-
enough to overwhelm their physical integrity. Severe and anism. Goldstein et al., by comparing EKGs of patients
prolonged catecholamine stimulation is well known to obtained during acute phase of stroke with their previous
produce myofibrillar degeneration and contraction band EKGs, found new QTc prolongation to be present in 32%
necrosis [37]. Characteristically located in sub-endocar- of these patients compared to 2% of controls [43].
dium, these band necrotic lesions are known to produce Another striking EKG feature in patients with stroke
malignant arrhythmias by involving the conducting tis- is the presence of abnormal T waves, which are seen in
sue [38]. Additionally, dysfunctions of protein synthesis almost 30% of patients with stroke [44]. In a study of
by endoplasmic reticulum causing apoptosis have also 150 patients with stroke and age-matched controls, new
been implicated in myocardial damage in rats who were T wave inversion and U waves were present in 15% of
subjected to traumatic stress [39]. patients with acute stroke but not in any patient from
the control group [43]. These T waves are sometimes also
referred to as “cerebral T waves” and defined as a T wave
Interaction of Parasympathetic and Sympathetic responses inversion ≥ 5 mm in depth in ≥ 4 contiguous precordial
Muscarinic receptors, supplied by a network of cholin- leads. A retrospective analysis of 800 patients with stroke
ergic post-ganglionic fibers, are distributed through- showed presence of T waves in 17 patients (2.1%). Even
out the ventricles [40]. In addition to directly producing more, approximately 20% of patients with T wave inver-
negative inotropic and chronotropic effects, this network sion had transient wall motion abnormalities on the
is a potent inhibitor of adrenergic response. The later echocardiogram consistent with stress-induced cardio-
effect is mediated by effect of acetylcholine on M3 sub- myopathy [45]. Table 3 highlights the major EKG abnor-
type muscarinic receptors, located at presynaptic sym- malities in ischemic and hemorrhagic strokes.
pathetic nerve terminals. Experimental studies involving Remarkably, QTc prolongation is not only a potential
enhancement of parasympathetic activity via vagal nerve indicator of underlying neurogenic mediated cardiac
stimulation (VNS) found that there was flattening of APD dysfunction but also may be a strong predictor of devas-
restitution slope, increase in effective refractory period tating cardiac and neurologic complications. In an outpa-
and hence increase in ventricular fibrillation threshold. tient study of stroke survivors of all types, prolonged QTc
Not surprisingly, VNS was found to be protective against in lead V6 was associated with a threefold increase in
ventricular arrhythmias [41, 42]. However, antiarrhyth- mortality with specificity approaching 94% for QTc inter-
mic effect of vagal stimulation was seen only during val ≥ 4.8 ms [46]. Also, QT prolongation along with tach-
simultaneous sympathetic stimulation and this protec- ycardia at baseline EKG was found to be independently
tion was lost with left stellate ganglionic ablation [42]. associated with development of cerebrovascular spasm
This ‘counter sympathetic’ effect also decreases sympa- [47]. Another study from the Czech Republic showed
thetic induced exacerbation of heterogeneity between that QTc prolongation after 48 h of stroke was associated
basal and apical segment of myocardium, thus preventing with mortality. Almost 40% of patients who died during
ventricular fibrillation. hospitalization after an acute stroke were found to have
In addition to all these cholinergic actions mediated via prolonged QTc after 48 h of stroke onset [48].
muscarinic receptors, intrinsic cardiac ganglionic plexus
as discussed in previous section, along with a number of Cardiac Arrest and Sudden Death
neuromodulators (Table 4), further facilitate this interac- Neurologic diseases are the leading cause of sudden
tion of two opposing mechanisms in order to maintain death of non-cardiac origin, especially in the younger
homeostasis. population [49].
 Sudden and unexpected deaths due to neurologic victims had any internal organ injury to explain the
catastrophes are best exemplified by acute aneurysmal deaths [61].
sub-arachnoid hemorrhage (aSAH) patients who die However, this ‘neurogenic’ myocardial damage is not
before reaching the hospital. These sudden deaths occur always immediately fatal as identified a decade later in
in 11–17% of patients with aSAH [50–53]. In Rochester, Japan by Sato and colleagues. They reported a series of
Minnesota, over a period of three decades, 113 patients cases of potentially reversible cardiomyopathy mim-
with ruptured intracranial aneurysm were studied. Of icking acute anterior wall myocardial infarction. These
these 113 patients, 13 (12%) died suddenly before reach- patients presented with acute onset of chest pain that
ing the hospital. Posterior circulation aneurysm was was temporally related to emotional stress. EKG revealed
more frequent (38%) in patients with sudden death. In ST segment elevation in precordial leads, but emergent
comparison, posterior circulation aneurysm was present coronary angiography was negative for atherosclerotic
only in 14% of patients who survived the initial insult. narrowing of coronary vessels. Ventriculography showed
Interestingly, intraventricular hemorrhage was present in apical and diaphragmatic akinesia along with hyperkine-
12 of these 13 patients with sudden death. Autopsy find- sia of basal segment resulting in apical ballooning. This
ings of these patients did not reveal any structural cardiac gave a peculiar shape of myocardium that was similar
abnormalities [51]. to ‘Takotsubu’—a Japanese fishermen’s octopus pot.
Data regarding sudden death are not as extensive in Therefore, this cardiomyopathy, that was transient and
other subtypes of acute stroke. However, sudden and resolved over time, was termed Takotsubu cardiomyo-
unexpected deaths seem to occur in the subacute or pathy (TCM) also known as ‘apical ballooning syndrome’
delayed phase of stroke. A 3-month follow-up of patients or’ broken heart syndrome’ [62].
who were enrolled in placebo arm of randomized trial In addition to emotional stress such as death of close
of trilizad in acute stroke (RANTTAS) showed that family or friend, financial loss or natural disasters, this
unexplained cardiac arrest occurred in 2% of these cardiomyopathy is also known to be precipitated by
279 patients [54]. In another postmortem study of 125 stress related to acute critical illness such as sepsis, acute
patients who died within one month after supra-tentorial respiratory failure and catecholamine infusion [63–66].
ischemic stroke, about 8% died suddenly and unexpect-
edly without any apparent cause at autopsy [55]. A pub- Neurogenic Stunned Myocardium
lished review by Soros and Hackinski gives a detailed Neurogenic stunned myocardium (NSM) is described as
account of sudden death in stroke patients [56]. a syndrome of acute left ventricular dysfunction triggered
Other than stroke, seizures are also associated with by an acute neurologic condition. Despite similarities
sudden death [57]. A prospective study of 4, 578 patients with TCM (Table 1), NSM is a distinct subtype of stress
with epilepsy showed occurrence of Sudden Unex- cardiomyopathy. This is due to a different triggering
pected Death in Epilepsy (SUDEP) at an incidence rate mechanism of NSM, in which there is a direct involve-
of 1.2/1000 patients per year. SUDEP was responsible ment of neural cardiovascular regulatory areas precipi-
for 18% of all deaths in epilepsy patients [58]. A 4-year tated by the acute neurologic condition rather than their
follow-up of patients in Western China found that 14.7% secondary involvement as a result of stress which is seen
of all deaths in epilepsy were sudden and unexpected in TCM. The clinical, biochemical and electrographic
[59]. Furthermore, in patients with epilepsy, retrospective manifestations of NSM are described in detail in Table 2.
measurement of QT interval before and after seizures Echocardiographic studies in patients with aSAH have
during simultaneous continuous EEG and EKG recording shown that myocardium can be affected either: (a) glob-
revealed abnormal prolongation of QT interval during 21 ally with severe reduction of left ventricular ejection
seizures (9 patients) out of 156 seizures recorded in 39 fraction (LVEF) or (b) with regional wall motion abnor-
patients [60]. malities (RWMA) with normal LVEF [67–69]. Global LV
dysfunction with LVEF of < 50% can be found in 15% of
patients after aSAH, whereas RWMAs are seen in 13–
Clinical Features and Syndromes: Contactile 27% of patients after rupture of intracranial aneurysm
Dysfunctions [68, 70]. The incidence of RWMA can be even higher in
Stress Cardiomyopathies patients with more severe neurologic injury due aSAH
In 1980, Cebelin et al. used the term ‘stress cardiomyo- [69, 71]. More than one-third of patients with high Hunt
pathy’ to describe myofibrillar necrotic lesions of heart and Hess score have RWMA on echocardiography [71].
found at autopsy of physical assault victims. These deaths The pattern of RWMA does not correlate with coronary
were presumably due to excessive sympathetic stimula- distribution. Basal and mid-ventricular segments of LV
tion triggered by severe stress of assault as none of these are most commonly affected and apex tends to be spared,
Table 1 Overview of neurogenic cardiac dysfunction
Autonomic mediated abnormality Proposed pathophysiologic Clinical spectrum and disorders Common inciting conditions
mechanism

Cardiac conduction disturbance Sympathetic over-dominance Changes in HR aSAH, also seen in acute ischemic
affecting myocardial ion channels Loss of HR variability and hemorrhagic stroke, seizures,
leading to alteration in action EKG repolarization abnormalities and TBI
potential and exaggeration of Elevation of cardiac enzymes
intrinsic heterogeneity Ventricular arrhythmias
Myocardial band necrosis, domi- Sudden death
nantly in sub-endocardium but
may involve conducting pathways
Myocardial inotropic dysfunction Increased catecholamine exposure Stress induced cardiomyopathy Psychological stress, critical illness,
to myocardium due to stress or natural disasters, medical and neu-
neurologic damage producing rologic disorders, catecholamine
enhanced cAMP production and infusion, spinal cord disorders
ER dysfunction, ­Ca++ influx and Neurogenic stunned myocardium All neurologic catastrophes but most
prolongation of actin–myosin often seen in high-grade aSAH
interaction. Apoptosis, inflamma-
tion and band necrosis are known
to occur
aSAH aneurysmal subarachnoid hemorrhage, cAMP cyclic adenosine monophosphate, EKG electrocardiographic, ER endoplasmic reticulam, HR heart rate, TBI
traumatic brain injury

Table 2 Clinical features of Takotsubu cardiomyopathy and neurogenic stunned myocardium

Clinical features Takotsubo CMP Neurogenic stunned myocardium

Presentation Chest pain, short of breath (pulmonary edema) Chest pain may be masked by coma or altered mentation.
Acute pulmonary edema is more common presentation
Age/gender Overwhelming majority of females of higher age group Majority are females of relatively younger age group. Also
affects significant number of males
Inciting event Stress (psychosocial, critical illness), no inciting event in Acute neurologic catastrophe especially high-grade aSAH
one-third of patients
EKG changes S–T segment elevation in precordial leads are most com- S–T elevation in precordial leads is less common. QT prolon-
mon gation and T wave inversions may be more frequent than
TCM
Wall motion abnormalities Apical wall motion abnormality with hypokinesia of basal Either of two patterns are more common (1) hypokinesia of
segments is more common. Mid-ventricular wall motion basal and mid-ventricular segments with apical sparing,
abnormality can also be seen (2) global hypokinesia of LV
Clinical course Significant improvement in LV function can occur in Depends on severity of primary neurologic injury. Known to
1–2 weeks adversely affect outcome from aSAH
Complications Hypoxemia/hypotension/SVT Hypotension and ventricular arrhythmias are more frequent
Ventricular arrhythmias are less common
aSAH aneurysmal subarachnoid hemorrhage, CMP cardiomyopathy, LV left ventricle, SVT supraventricular tachycardia, TCM Takotsubu cardiomyopathy (Refs. [60–67,
92])

a pattern which correlates with the distribution of sym- hemisphere, large vessel embolic stroke in posterior cir-
pathetic nerve endings [23, 67, 70]. culation or cryptogenic stroke involving basal ganglia
In addition to severity of neurologic injury and female (Table 3). Interestingly, 38% of patients with LV dysfunc-
sex, ST segment elevation in anterior precordial leads, tion had involvement of insula [75].
high cardiac troponin, history of cocaine and ampheta- NSM can complicate the course and outcome of pri-
mine use and posterior circulation aneurysms are known mary neurologic injury. Heart failure, pulmonary edema
risk factors of development of NSM in patients with and cardiac arrhythmias are the most common compli-
aSAH [71–74]. cations. Development of NSM in patients with aSAH
Despite paucity of data, LV dysfunction is known was associated with significantly higher risk of cerebral
to occur in almost all acute neurologic conditions. In vasospasms and delayed cerebral ischemia (DCI) [74].
patients with acute ischemic stroke, it typically affects These complications due to NSM adversely affect long-
older age women. LV dysfunction has been reported in term functional outcome and mortality [74, 76]. Addi-
association with cardio-embolic stroke affecting either tionally, low LVEF due to NSM poses a great challenge
Table 3 Clinical and electrographic cardiac events: a com- Severity of Primary Neurological Injury
parison between ischemic and hemorrhagic strokes Severity of primary neurological injury can also provide
Cardiac event Spontaneous ICH Ischemic valuable insight regarding risk stratification. Patients who
(%) stroke presented with a National Institutes of Health Stroke
(%) Scale (NIHSS) of > 10 had a higher rate of troponin eleva-
Incidences of cardiac events tion and ischemic changes in EKG compared to patients
Deaths during hospitalization 24 (1) 4.9–8 with NIHSS < 10 [81]. Similarly, patients with higher
Acute heart failure 3.8 4.4 Hunt and Hess score after aSAH have a much higher
Ventricular arrhythmias 0.3 1.1 incidence of left ventricular RWAs [71].
Elevated troponin 8.5–20 6–17
Post-stroke AMI 0–0.3 0.5–2.3
Location of Neurological Damage in Imaging
Common EKG changes
Location of neurologic damage on neuroimaging can also
QTc prolongation 34.3 9.8
provide guidance for risk stratification. Posterior circu-
ST segment depression 11.6 22
lation aneurysmal rupture was much more commonly
ST segment elevation 10.2 9.8
associated with sudden death as compare to anterior
U wave abnormality 9.6 9.8
circulation aneurysms. Intraventricular hemorrhage was
Pathologic Q wave 13.9 17.1
found in 12 of 13 patients with aSAH who died before
Abnormal T wave 39.9 43.9
reaching the hospital [51]. Another study of 118 intrac-
ranial hemorrhage patients in Japan revealed that insular
AMI acute myocardial infarction, EKG electrocardiographic, ICH intracranial
hemorrhage
involvement was an independent predictive factor of ST
depression, whereas insular involvement and the pres-
ence of intraventricular hemorrhage were independent
in treatment of DCI with hemodynamic augmentation predictive factors of QTc prolongation [82].
as there is a significant increase in systemic sympathetic
tone which increases LV afterload [77]. Consequently, EKG Changes
on the one hand there is further compromise of cer- Typical EKG changes described above are an important
ebral blood flow and worsening of cerebral vasospasms, marker of cardiac dysfunction. Serial EKG studies or con-
whereas on the other hand, use of vasopressors and vol- tinuous telemetry recording along with an echocardio-
ume infusion becomes more challenging and potentially gram are extremely useful in the identification of patients
counterproductive. Therefore, not only there is increased at risk and the diagnosis of NSM.
risk of development of cerebral vasospasm and delayed
cerebral ischemia, sympathetic mediated cardiovascular Biochemical Markers
dysfunction can significantly contribute to a vicious cycle Biochemical markers of myocardial damage, such as
that ultimately culminates in increased mortality and troponin T or I, are commonly elevated in patients with
poor functional outcome [74]. acute stroke even in patients without preexisting car-
diac history or renal dysfunction and are associated with
Risk Stratification and Diagnostic Workup increased mortality at 1-month follow-up [83, 84]. How-
Demographics and Historical Data ever, association of this increased mortality with sudden
In various studies, older age, severity of lesion, preexist- cardiac death has not been fully evaluated. In patients
ing coronary artery disease, arterial hypertension and with aSAH, elevation of troponin I is highly sensitive and
bradycardia have been shown to be risk factors for ven- specific marker of neurogenic myocardial damage [85]. In
tricular arrhythmias [78, 79]. Patients who were already comparison to severity of myocardial dysfunction with
taking angiotensin converting enzyme inhibitors (ACE- low ejection fraction (EF), troponin elevation is modest.
Is) or angiotensin receptor blockers (ARBs) were less This relatively modest elevation of troponin with severely
likely to develop ventricular arrhythmias than those who low EF on 2D Echo in aSAH is suggestive of a neurogenic
were not on this medications before aneurysmal rup- process rather than acute coronary syndrome [86].
ture [78]. Furthermore, female gender, which comprises
overwhelming majority of TCM cases, is more prone to Therapeutic Options and Future Directions
develop QTc prolongation after aSAH. In a multivariate Although cardiac dysfunction from neurogenic catastro-
analysis, female gender and hypokalemia were identified phes is associated with morbidity and mortality, there
as independent risk factors for QTc prolongation after are no clear guidelines for the acute management or for
SAH [80]. identification of high-risk patients. Management of these
Table 4 Overview of autonomic receptors and actions of cardiac neurotransmitters and neuromodulators
Neurotransmitter/modulator Receptors Actions Consequences

Norepinephrine and epinephrine NE: Act dominantly on bet 1 adeno-receptors, but SA node activation, increased AV node and Increased HR, increased cardiac output and perfu-
also act on beta2, alpha1 and alpha 2 (presyn- his purkinje conduction, increased inotropy sion to vital organs. Increased propensity to
aptic inhibition). EPI: released from adrenal increased heterogeneity during repolarization, ventricular arrhythmias and myocardial necrosis
medulla and reach via circulation increased slope of APD restitution, reduced VFT
Acetylcholine Muscarinic receptors (M1 to M5 subtypes) in Inhibition of SA and AV node, reduced inotropy, Reduced HR and inotropy. Reduced propensity to
sympathetic and parasympathetic ganglia reduced heterogeneity and increased VFT ventricular arrhythmias
Post-synaptic receptors (dominantly M2) in
conductive tissue and myocytes of atria and
ventricles
Presynaptic sympathetic nerve terminal (M3
subtype)
Nitric oxide Released in ventricles with vagal nerve stimula- Accentuation of vagal mediated flattening of Protection from ventricular arrhythmias
tion. Possible existence of intrinsic nitrergic APD restitution slope and increase in VFT
neurons in ventricles
Vasoactive intestinal polypeptide Released in response vagal stimulation, choliner- Coronary and peripheral vasodilation, LV afterload Improves coronary flow during reperfusion and
gic drugs, and coronary occlusion. VIP-immu- reduction, enhance myocardial isometric force free radical scavenger, afterload reduction
noreactive fibers are located around coronary
sinus and around SA and AV nodes
Neuropeptide Y Released during prolonged sympathetic stimula- Reduce Ach release from parasympathetic Increased propensity to sympathetic-mediated
tion and act via Y2 receptors located at vagal nerve terminals and reduce vagal bradycardic arrhythmias.
presynaptic terminals response
Intermedin Released from myocytes during ischemia and act Reduce sympathetic activity, increases VFT and Protection from sympathetic mediated ventricular
via calcitonin receptors oxidative stress arrhythmias and heart failure in patient with
acute MI
Ach Acetylcholine, APD action potential duration, AV atrioventricular node, EPI epinephrine, HR heart rate, IMD Intermedin, LV left ventricle, MI myocardial infarction, NE Norepinephrine, NO Nitric oxide, NPY Neuropeptide Y,
SA sino-atrial node, VFT ventricular fibrillation threshold, VIP Vasoactive intestinal polypeptide
complications is generally based on the same principle ever feasible, especially in patients at high risk of car-
as for a non-neurologic cause. However, based on the diac complications. Correction of hypokalemia and
unique pathophysiology of these conditions, some spe- hypomagnesemia is prudent, for the same reason.
cific therapeutic options can be suggested: (4) In the management of DCI after aSAH, inotropic
agents are preferable over vasopressors for hemo-
(1) Centrally acting sympathetic blockers are a reason- dynamic augmentation, especially for patients with
able choice for patients with high blood pressure NSM [94–97]. Dobutamine has been shown to
requiring antihypertensive treatment with underly- increase cerebral blood flow (CBF) by 50% in patients
ing risk factors for neurogenic cardiac dysfunction. with severe vasospasm. The improvement in CBF
Clonidine, a centrally acting alpha-2 antagonist, with dobutamine was not only comparable to phe-
has been shown to suppress cardiac arrhythmias in nylephrine infusion, but was achieved without after-
patients with congestive heart failure (CHF) [87]. load augmenting effects of the latter. Also, dobu-
However, there is paucity of data on its effects on tamine produced 20% reduction in systemic vascular
acute neurogenic cardiac dysfunction. Antihyper- resistance (SVR) [94]. In another study, neurologic
tensive agents that block both central apha-2 recep- deficit due to DCI was reversed in more than two-
tors and imidazoline receptors, such as rilmenidine, thirds of patients treated with dobutamine [95].
have been shown to inhibit adrenaline-induced ven- Milrinone, a phosphodiesterase III inhibitor which
tricular arrhythmias in halothane anesthetized dogs. increases intracellular concentration of cyclic adeno-
However, the presence of vagal tone was critical for sine mono-phosphate in myocardium and vascular
this action [88]. Safety of these medications as anti- smooth muscles, was found to be more effective than
hypertensives agents is well documented but further dobutamine in patients with NSM who had higher
clinical research is needed in the neurocritical care SVR, whereas dobutamine was more effective in
setting. ACE-Is and ARBs are good choices for long- patients with lower SVR and blood pressure [96].
term blood pressure control [78]. However, choice However, as with most of hemodynamic augmenting
of short- and long-term antihypertensive agents in agents, effects of dobutamine and milrinone on myo-
patients who are at risk of neurogenic cardiac dys- cardium are adrenergic. Limited data are available for
function is an avenue where further research is efficacy of alternative methods of augmenting car-
needed. diac output (CO) such as intra-aortic balloon pump
(2) Protective effect of VNS, as discussed above, has [97]. Clinical cases have been reported success of use
been studied in clinical trials of patients with CHF of of intravenous insulin infusion in augmenting CO
primarily cardiac origin. It is worthwhile to mention in a patient with NSM secondary aSAH and acute
that altered cardiac autonomic activity, specifically ischemic stroke [98, 99].
reduction in vagal tone in patients with CHF after (5) A number of novel treatments are being studied
myocardial infarction plays a vital role in progressive based on the interplay of sympathetic and parasym-
worsening of symptoms and increased risk of sudden pathetic nervous systems. NO has been investigated
death [89]. Two small studies have shown safety and in animal subjects during CPR after ventricular fibril-
feasibility of implantable VNS, with improvement in lation induced cardiac arrest and showed improved
clinical parameters of CHF of primary cardiac origin success of CPR, neurologic outcome and mortality
[90, 91]. However, INOVATE-HF, a large randomized rate [100]. A phase II, double blind trial compar-
study of effect of implantable VNS in patients with ing efficacy of inhaled NO with placebo is currently
CHF failed to show any mortality benefits or reduc- enrolling patients with out of hospital cardiac arrest.
tion in episodes of CHF exacerbation [92]. Feasibil- The primary end points of the study are effects on
ity and efficacy of VNS in patients with neurogenic mortality, cardiac and neurologic morbidity.
cardiac dysfunctions is largely unexplored although a Beside NO, intermedin, a calcitonin gene-related
number of animal studies have shown its beneficial peptide, is another novel agent. Released from myo-
effect in restoring autonomic balance [87, 93]. Anti- cytes in response to ischemia, it acts on calcitonin
arrhythmic mechanism of parasympathetic nervous receptors and calcitonin-like receptors (Table 4).
system and protective effect of VNS has been dis- Intermedin administration has been shown to
cussed comprehensively in a review [29]. improve cardiac function, reduce sympathetic medi-
(3) A number of pharmacologic agents frequently used ated neural remodeling, and increase ventricular
in neurocritical care settings such as anticonvulsants, fibrillation threshold in rats with ischemia induced
neuroleptics, or certain antimicrobial agents can CHF [36].
prolong QTc interval and should be avoided when-
 Despite the agents and strategies available, there is 15. Yoon BW, Morillo CA, Cechetto DF, Hachinski V. Cerebral hemispheric
lateralization in cardiac autonomic control. Arch Neurol. 1997. https​://
paucity of evidence-based data addressing cardiac doi.org/10.1001/archn​eur.1997.00550​18005​5012.
dysfunctions in neurologic catastrophes. Rand- 16. Hilz MJ, Dütsch M, Perrine K, et al. Hemispheric influence on autonomic
omized clinical studies are needed for development modulation and baroreflex sensitivity. Ann Neurol. 2001. https​://doi.
org/10.1002/ana.1006.
of parameters for risk stratification, approaches for 17. Oppenheimer SM, Gelb A, Girvin JP, Hachinski VC. Cardiovascular effects
acute intervention, and strategies for long-term pre- of human insular cortex stimulation. Neurology. 1992. https​://doi.
vention of cardiac complications. org/10.1212/WNL.42.9.1727.
18. Critchley HD, Corfield DR, Chandler MP, et al. Cerebral correlates of auto-
nomic cardiovascular arousal: a functional neuroimaging investigation
in humans. J Physiol. 2000. https​://doi.org/10.1111/j.1469-7793.2000.
Acknowledgments
t01-1-00259​.x.
The authors would like to acknowledge Dr. Robert P Lisak, MD for his help in
19. Meyer S, Strittmatter M, Fischer C, et al. Lateralization in autonomic dys-
editing the manuscript and his thoughtful comments.
function in ischemic stroke involving the insular cortex. NeuroReport.
2004;15:357–61.
Author Contribution
20. Barron SA, Rogovski Z, Hemli J. Autonomic consequences of cerebral
MI and KS were responsible for study conception, literature search, writing and
hemisphere infarction. Stroke. 1994. https​://doi.org/10.1161/01.
editing the manuscript. BS and WM were involved in editing the manuscript.
STR.25.1.113.
21. Oppenheimer SM, Cechetto DF. Cardiac chronotropic organization of
Source of support
the rat insular cortex. Brain Res. 1990. https​://doi.org/10.1016/0006-
No funding.
8993(90)91796​-J.
22. Guyton AC. Guyton and Hall textbook of medical physiology. Amster-
Compliance with ethical standards
dam: Elsevier; 2015. p. 110–2.
23. Winter J, Tanko AS, Brack KE, et al. Differential cardiac responses to uni-
Conflict of interest
lateral sympathetic nerve stimulation in the isolated innervated rabbit
The authors declare that they have no conflict of interest.
heart. Auton Neurosci Basic Clin. 2012. https​://doi.org/10.1016/j.autne​
u.2011.08.004.
24. Ng GA, Mantravadi RK, Walker WH, et al. Sympathetic nerve stimulation
produces spatial heterogeneities of action potential restitution. Hear
Rhythm. 2009. https​://doi.org/10.1016/j.hrthm​.2009.01.035.
25. Volders PG, Stengl M, van Opstal JM, et al. Probing the contribution
of IKs to canine ventricular repolarization: key role for beta-adrenergic
receptor stimulation. Circulation. 2003. https​://doi.org/10.1161/01.
References CIR.00000​68344​.54010​.B3.
1. Cannon WB. “Voodoo” death. American Anthropologist, 1942;44(new 26. Karagueuzian HS, Chen PS. Graded response and restitution hypoth-
series):169–181. Am J Public Health. 2002;44(new series):169–81. eses of ventricular vulnerability to fibrillation: insights into the mecha-
2. Wilson A. The mechanism of death under chloroform. Lancet. 1894. nism of initiation of fibrillation. J Electrocardiol. 1999;32:87–91.
https​://doi.org/10.1016/S0140​-6736(01)59812​-8. 27. Fox JJ, McHarg JL, Gilmour RF. Ionic mechanism of electrical alternans.
3. Levy AG. The exciting causes of ventricular fibrillation in animals under Am J Physiol Heart Circ Physiol. 2002. https​://doi.org/10.1152/ajphe​
chloroform anesthesia. Heart 1913;4:1912–1913. art.00612​.2001.
4. Levy AG. Chloroform anesthesia, chapter IV. Cambridge: Wood Publish- 28. Bass BG. Restitution of the action potential in cat papillary muscle. Am J
ers; 1922. p. 26–43. Physiol. 1975;228:1717–24, 2753–60.
5. Erickson TC. Cardiac activity during epileptic seizures. Arch Neuropsy- 29. Ng GA. Neuro-cardiac interaction in malignant ventricular arrhyth-
chol. 1939;41:511–8. mia and sudden cardiac death. Auton Neurosci. 2016. https​://doi.
6. Raab W, Stark E, Macmillan WH, Gigee WR. Sympathogenic origin and org/10.1016/j.autne​u.2016.07.001.
antiadrenergic prevention of stress-induced myocardial lesions. Am J 30. Karma A. Electrical alternans and spiral wave breakup in cardiac tissue.
Cardiol. 1961. https​://doi.org/10.1016/0002-9149(61)90207​-7. Chaos. 1994. https​://doi.org/10.1063/1.16602​4.
7. Seyle H. The role of electrolytes in pathogenesis of experimental cardio- 31. Magnano AR, Holleran S, Ramakrishnan R, et al. Autonomic nervous
myopathies without vascular involvement. Fundam Asp Physiol Pathol. system influences on QT interval in normal subjects. J Am Coll Cardiol.
1965;1:135–60. 2002. https​://doi.org/10.1016/S0735​-1097(02)01852​-1.
8. Greenhoot JH, Reichenbach DD. Cardiac injury and subarachnoid hem- 32. Rosenbaum DS, Jackson LE, Smith JM, et al. Electrical alternans and
orrhage. J Neurosurg. 1969. https​://doi.org/10.3171/jns.1969.30.5.0521. vulnerability to ventricular arrhythmias. N Engl J Med. 1994. https​://doi.
9. Shepherd JT. The heart as a sensory organ. J Am Coll Cardiol. 1985. org/10.1056/NEJM1​99401​27330​0402.
https​://doi.org/10.1016/S0735​-1097(85)80533​-7. 33. Pastore JM, Girouard SD, Laurita KR, et al. Mechanism linking T-wave
10. Wake E, Brack K. Characterization of the intrinsic cardiac nervous sys- alternans to the genesis of cardiac fibrillation. Circulation. 1999. https​://
tem. Auton Neurosci Basic Clin. 2016;199:3–16. doi.org/10.1161/01.CIR.99.10.1385.
11. Agarwal SK, Calaresu FR. Electrical stimulation of nucleus tractus soli- 34. Cao JM, Chen LS, KenKnight BH, et al. Nerve sprouting and sudden
tarius excites vagal preganglionic cardiomotor neurons of the nucleus cardiac death. Circ Res. 2000. https​://doi.org/10.1161/01.RES.86.7.816.
ambiguus in rats. Brain Res. 1992. https​://doi.org/10.1016/0006- 35. Lee TM, Lin SZ, Chang NC. Antiarrhythmic effect of lithium in rats after
8993(92)90833​-U. myocardial infarction by activation of Nrf2/HO-1 signaling. Free Radic
12. Swanson LW, Sawchenko PE. Paraventricular nucleus: a site for the inte- Biol Med. 2014. https​://doi.org/10.1016/j.freer​adbio​med.2014.08.022.
gration of neuroendocrine and autonomic mechanisms. Neuroendocri- 36. Xu B, Xu H, Cao H, et al. Intermedin improves cardiac function and
nology. 1980. https​://doi.org/10.1159/00012​3111. sympathetic neural remodeling in a rat model of post myocardial
13. Melville KI, Blum B, Shister HE, Silver MD. Cardiac ischemic changes and infarction heart failure. Mol Med Rep. 2017. https​://doi.org/10.3892/
arrhythmias induced by hypothalamic stimulation. Am J Cardiol. 1963. mmr.2017.6776.
https​://doi.org/10.1016/0002-9149(63)90281​-9. 37. Samuels MA. Neurogenic heart disease: a unifying hypothesis. Am J
14. Zamrini EY, Meador KJ, Loring DW, et al. Unilateral cerebral inactivation Cardiol. 1987. https​://doi.org/10.1016/0002-9149(87)90678​-3.
produces differential left/right heart rate responses. Neurology. 1990. 38. Samuels MA. The brain–heart connection. Circulation. 2007;116:77–84.
https​://doi.org/10.1212/WNL.40.9.1408.
39. Liu M, Xu F, Tao T, et al. Molecular mechanisms of stress-induced myo- 62. Sato H. Takotsubo-like left ventricular dysfunction due to multivessel
cardial injury in a rat model simulating posttraumatic stress disorder. coronary spasm. Tokyo: Kagakuhyoronsha Publishing Company; 1990.
Psychosom Med. 2016. https​://doi.org/10.1097/PSY.00000​00000​00035​3. p. 56–64.
40. Kawano H, Okada R, Yano K. Histological study on the distribution of 63. Bybee KA, Kara T, Prasad A, et al. Systematic review: transient left
autonomic nerves in the human heart. Heart Vessels. 2003. https​://doi. ventricular apical ballooning: a syndrome that mimics ST-segment
org/10.1007/s0038​00300​005. elevation myocardial infarction. Ann Intern Med. 2004;141:858–65.
41. Kolman BS, Verrier RL, Lown B. The effect of vagus nerve stimulation 64. Templin C, Ghadri JR, Diekmann J, et al. Clinical features and outcomes
upon vulnerability of the canine ventricle. Role of sympathetic para- of Takotsubo (stress) cardiomyopathy. N Engl J Med. 2015. https​://doi.
sympathetic interactions. Circulation. 1975. https​://doi.org/10.1161/01. org/10.1056/NEJMo​a1406​761.
CIR.52.4.578. 65. Watanabe H, Kodama M, Okura Y, et al. Impact of earthquakes on Takot-
42. Ng GA, Brack KE, Patel VH, Coote JH. Autonomic modulation of electri- subo cardiomyopathy. J Am Med Assoc. 2005;294:305–7.
cal restitution, alternans and ventricular fibrillation initiation in the 66. Y-Hassan S, Settergren M, Henareh L. Sepsis-induced myocardial
isolated heart. Cardiovasc Res. 2007. https​://doi.org/10.1016/j.cardi​ depression and Takotsubo syndrome. Acute Card Care. 2014. https​://
ores.2006.12.001. doi.org/10.3109/17482​941.2014.92008​9.
43. Goldstein DS. The electrocardiogram in stroke: relationship to patho- 67. Zaroff JG, Rordorf GA, Ogilvy CS, Picard MH. Regional patterns of
physiological type and comparison with prior tracings. Stroke. 1979. left ventricular systolic dysfunction after subarachnoid hemorrhage:
https​://doi.org/10.1161/01.STR.10.3.253. evidence for neurally mediated cardiac injury. J Am Soc Echocardiogr.
44. Putaala J, Lehto M, Meretoja A, et al. In-hospital cardiac complica- 2000. https​://doi.org/10.1067/mje.2000.10576​3.
tions after intracerebral hemorrhage. Int J Stroke. 2014. https​://doi. 68. Banki N, Kopelnik A, Tung P, et al. Prospective analysis of prevalence,
org/10.1111/ijs.12180​. distribution, and rate of recovery of left ventricular systolic dysfunction
45. Stone J, Mor-Avi V, Ardelt A, Lang RM. Frequency of Inverted electro- in patients with subarachnoid hemorrhage. J Neurosurg. 2006. https​://
cardiographic T waves (cerebral T waves) in patients with acute strokes doi.org/10.3171/jns.2006.105.1.15.
and their relation to left ventricular wall motion abnormalities. Am J 69. Sugimoto K, Watanabe E, Yamada A, et al. Prognostic implications of left
Cardiol. 2018. https​://doi.org/10.1016/j.amjca​rd.2017.09.025. ventricular wall motion abnormalities associated with subarachnoid
46. Wong KYK, Mac Walter RS, Douglas D, et al. Long QTc predicts future hemorrhage. Int Heart J. 2008. https​://doi.org/10.1536/ihj.49.75.
cardiac death in stroke survivors. Heart. 2003. https​://doi.org/10.1136/ 70. Khush K, Kopelnik A, Tung P, et al. Age and aneurysm position predict
heart​.89.4.377. patterns of left ventricular dysfunction after subarachnoid hemor-
47. Ibrahim GM, MacDonald RL. Electrocardiographic changes predict rhage. J Am Soc Echocardiogr. 2005. https​://doi.org/10.1016/j.
angiographic vasospasm after aneurysmal subarachnoid hemorrhage. echo.2004.08.045.
Stroke. 2012. https​://doi.org/10.1161/STROK​EAHA.112.65815​3. 71. Kothavale A, Banki NM, Kopelnik A, et al. Predictors of left ventricular
48. Hromádka M, Seidlerová J, Rohan V, et al. Prolonged corrected QT regional wall motion abnormalities after subarachnoid hemorrhage.
interval as a predictor of clinical outcome in acute ischemic stroke. J Neurocrit Care. 2006. https​://doi.org/10.1385/NCC:4:3:199.
Stroke Cerebrovasc Dis. 2016. https​://doi.org/10.1016/j.jstro​kecer​ebrov​ 72. Kono T, Morita H, Kuroiwa T, et al. Left ventricular wall motion
asdis​.2016.08.005. abnormalities in patients with subarachnoid hemorrhage: neuro-
49. Lim Z, Gibbs K, Potts JE, Sanatani S. A review of sudden unexpected genic stunned myocardium. J Am Coll Cardiol. 1994. https​://doi.
death in the young in British Columbia. Can J Cardiol. 2010;26:22–6. org/10.1016/0735-1097(94)90008​-6.
50. Fogelholm R. Subarachnoid hemorrhage in middle-Finland: incidence, 73. Kuroiwa T, Morita H, Tanabe H, Ohta T. Significance of ST segment
early prognosis and indications for neurosurgical treatment. Stroke. elevation in electrocardiograms in patients with ruptured cerebral
1981. https​://doi.org/10.1161/01.STR.12.3.296. aneurysms. Acta Neurochir (Wien). 1995. https​://doi.org/10.1007/BF014​
51. Schievink WI, Wijdicks EF, Parisi JE, et al. Sudden death from aneurysmal 20064​.
subarachnoid hemorrhage. Neurology. 1995;45:871–4. 74. Kilbourn KJ, Levy S, Staff I, et al. Clinical characteristics and outcomes
52. Sheikhazadi A, Gharehdaghi J. Survey of sudden death from aneurys- of neurogenic stress cardiomyopathy in aneurysmal subarachnoid
mal subarachnoid hemorrhage in cadavers referred to legal medicine hemorrhage. Clin Neurol Neurosurg. 2013. https​://doi.org/10.1016/j.
organization of Tehran, 2001–2005. Am J Forensic Med Pathol. 2009. cline​uro.2012.09.006.
https​://doi.org/10.1097/PAF.0b013​e3181​bfcd6​4. 75. Murthy SB, Shah S, Venkatasubba Rao CP, et al. Clinical characteristics of
53. Ljunggren B, Säveland H, Brandt L, Zygmunt S. Early operation and myocardial stunning in acute stroke. J Clin Neurosci. 2014;21:1279–82.
overall outcome in aneurysmal subarachnoid hemorrhage. J Neuro- 76. Crago EA, Kerr ME, Kong Y, et al. The impact of cardiac complications on
surg. 1985. https​://doi.org/10.3171/jns.1985.62.4.0547. outcome in the SAH population. Acta Neurol Scand. 2004. https​://doi.
54. Johnston KC, Li JY, Lyden PD, et al. Medical and neurological complica- org/10.1111/j.1600-0404.2004.00311​.x.
tions of ischemic stroke: experience from the RANTTAS trial. Stroke. 77. Temes RE, Tessitore E, Schmidt JM, et al. Left ventricular dysfunction and
1998. https​://doi.org/10.1161/01.STR.29.2.447. cerebral infarction from vasospasm after subarachnoid hemorrhage.
55. Silver FL, Norris JW, Lewis AJ, Hachinski VC. Early mortality following Neurocrit Care. 2010. https​://doi.org/10.1007/s1202​8-010-9447-x.
stroke: a prospective review. Stroke. 1984. https​://doi.org/10.1161/01. 78. Michael Frangiskakis J, Hravnak M, Crago EA, et al. Ventricular arrhyth-
STR.15.3.492. mia risk after subarachnoid hemorrhage. Neurocrit Care. 2009. https​://
56. Sörös P, Hachinski V. Cardiovascular and neurological causes of sudden doi.org/10.1007/s1202​8-009-9188-x.
death after ischaemic stroke. Lancet Neurol. 2012;11:179–88. 79. Kallmünzer B, Breuer L, Kahl N, et al. Serious cardiac arrhythmias
57. Nashef L, So EL, Ryvlin P, Tomson T. Unifying the definitions of sudden after stroke: incidence, time course, and predictors—a systematic,
unexpected death in epilepsy. Epilepsia. 2012;53:227–33. prospective analysis. Stroke. 2012. https​://doi.org/10.1161/STROK​
58. Walczak TS, Leppik IE, D’Amelio M, et al. Incidence and risk factors in EAHA.112.66431​8.
sudden unexpected death in epilepsy: a prospective cohort study. 80. Fukui S, Otani N, Katoh H, et al. Female gender as a risk factor for
Neurology. 2001. https​://doi.org/10.1212/WNL.56.4.519. hypokalemia and QT prolongation after subarachnoid hemorrhage.
59. Mu J, Liu L, Zhang Q, et al. Causes of death among people with convul- Neurology. 2002;59:134–6.
sive epilepsy in rural West China: a prospective study. Neurology. 2011. 81. Wira CR, Rivers E, Martinez-Capolino C, et al. Cardiac complica-
https​://doi.org/10.1212/WNL.0b013​e3182​23c78​4. tions in acute ischemic stroke. West J Emerg Med. 2011. https​://doi.
60. Brotherstone R, Blackhall B, McLellan A. Lengthening of corrected org/10.5811/westj​em.2011.2.1765.
QT during epileptic seizures. Epilepsia. 2010. https​://doi.org/10.111 82. Takeuchi S, Nagatani K, Otani N, et al. Electrocardiograph abnormali-
1/j.1528-1167.2009.02281​.x. ties in intracerebral hemorrhage. J Clin Neurosci. 2015. https​://doi.
61. Cebelin MS, Hirsch CS. Human stress cardiomyopathy: myocardial org/10.1016/j.jocn.2015.04.028.
lesions in victims of homicidal assaults without internal injuries. Hum 83. Kerr G, Ray G, Wu O, et al. Elevated troponin after stroke: a systematic
Pathol. 1980. https​://doi.org/10.1016/S0046​-8177(80)80129​-8. review. Cerebrovasc Dis. 2009;28:220–6.
84. Song HS, Back JH, Jin DK, et al. Cardiac troponin T elevation after stroke: 93. Li M. Vagal nerve stimulation markedly improves long-term sur-
relationships between elevated serum troponin T, stroke location, and vival after chronic heart failure in rats. Circulation. 2003. https​://doi.
prognosis. J Clin Neurol. 2008. https​://doi.org/10.3988/jcn.2008.4.2.75. org/10.1161/01.CIR.00001​05721​.71640​.DA.
85. Parekh N, Venkatesh B, Cross D, et al. Cardiac troponin I predicts myo- 94. Kim DH, Joseph M, Ziadi S, et al. Increases in cardiac output can reverse
cardial dysfunction in aneurysmal subarachnoid hemorrhage. J Am Coll flow deficits from vasospasm independent of blood pressure: a study
Cardiol. 2000. https​://doi.org/10.1016/S0735​-1097(00)00857​-3. using xenon computed tomographic measurement of cerebral blood
86. Bulsara KR, McGirt MJ, Liao L, et al. Use of the peak troponin value to flow. Neurosurgery. 2003. https​://doi.org/10.1227/01.NEU.00000​88567​
differentiate myocardial infarction from reversible neurogenic left .59324​.78.
ventricular dysfunction associated with aneurysmal subarachnoid hem- 95. Levy ML, Rabb CH, Zelman V, Giannotta SL. Cardiac performance
orrhage. J Neurosurg. 2003. https​://doi.org/10.3171/jns.2003.98.3.0524. enhancement from dobutamine in patients refractory to hyperv-
87. Zhang YH, Zhu J, Song YC. Suppressing sympathetic activation with olemic therapy for cerebral vasospasm. J Neurosurg. 1993. https​://doi.
clonidine on ventricular arrhythmias in congestive heart failure. Int J org/10.3171/jns.1993.79.4.0494.
Cardiol. 1998. https​://doi.org/10.1016/S0167​-5273(98)00127​-2. 96. Naidech A, Du Y, Kreiter KT, et al. Dobutamine versus milrinone
88. Kagawa K, Hayashi Y, Itoh I, et al. Identification of the central imidazoline after subarachnoid hemorrhage. Neurosurgery. 2005. https​://doi.
receptor subtype involved in modulation of halothane-epinephrine org/10.1227/01.NEU.00001​44780​.97392​.D7.
arrhythmias in rats. Anesth Analg. 2005. https​://doi.org/10.1213/01. 97. Taccone FS, Lubicz B, Piagnerelli M, et al. Cardiogenic shock with
ANE.00001​84185​.69471​.F6. stunned myocardium during triple-h therapy treated with intra-aortic
89. La Rovere MT, Bigger JT, Marcus FI, et al. Baroreflex sensitivity and heart- balloon pump counterpulsation. Neurocrit Care. 2009. https​://doi.
rate variability in prediction of total cardiac mortality after myocardial org/10.1007/s1202​8-008-9135-2.
infarction. ATRAMI (Autonomic Tone and Reflexes After Myocardial 98. Vanderschuren A, Hantson P. Hyperinsulinemic euglycemia therapy
Infarction) Investigators. Lancet. 1998. https​://doi.org/10.1016/S0140​ for stunned myocardium following subarachnoid hemorrhage. J Neuro-
-6736(97)11144​-8. surg. 2009. https​://doi.org/10.3171/2008.4.17507​.
90. De Ferrari GM, Crijns HJGM, Borggrefe M, et al. Chronic vagus nerve 99. Devos J, Peeters A, Wittebole X, Hantson P. High-dose insulin therapy
stimulation: a new and promising therapeutic approach for chronic for neurogenic-stunned myocardium after stroke. BMJ Case Rep. 2012.
heart failure. Eur Heart J. 2011. https​://doi.org/10.1093/eurhe​artj/ehq39​ https​://doi.org/10.1136/bcr-2012-00662​0.
1. 100. Brücken A, Derwall M, Bleilevens C, et al. Brief inhalation of nitric oxide
91. Premchand RK, Sharma K, Mittal S, et al. Autonomic regulation therapy increases resuscitation success and improves 7-day-survival after
via left or right cervical vagus nerve stimulation in patients with chronic cardiac arrest in rats: a randomized controlled animal study. Crit Care.
heart failure: results of the ANTHEM-HF trial. J Card Fail. 2014. https​:// 2015. https​://doi.org/10.1186/s1305​4-015-1128-x.
doi.org/10.1016/j.cardf​ail.2014.08.009.
92. Gold MR, Van Veldhuisen DJ, Hauptman PJ, et al. Vagus nerve stimula-
tion for the treatment of heart failure: the INOVATE-HF trial. J Am Coll
Cardiol. 2016. https​://doi.org/10.1016/j.jacc.2016.03.525.