Acute Emotional Stress and Cardiac Arrhythmias

A 78-year-old man with an ischemic cardiomyopathy, left bundle-branch block, and a left ventricular ejection fraction of 20% to 25% had a single-chamber implantable cardioverter-defibrillator (ICD) placed several years earlier for primary prevention of sudden death. Since the time of a major myocardial infarction many years ago, the patient's wife of 56 years had taken over most of his former household chores. One day, she mowed the lawn and placed the mower in its usual spot when she was finished. While his wife went to get the weed trimmer to finish the edges, the patient absentmindedly moved the lawn mower from where his wife left it. As his wife slowly ambled backward using the weed trimmer to finish the edges, she fell over the mower and broke her arm. She immediately got up and screamed at her husband, “Hey, let me tell you something. If I have to go to the hospital, you’d better not be here when I come back!” When asked to recall how she felt later, the patient's wife noted, “Honey, I was crazy! I yelled at him and I went after him with my good arm and my brother took him down the street.”

The patient's wife phoned emergency medical services because of the injury to her arm, but by the time an ambulance arrived several minutes later, the patient's defibrillator had fired. The ambulance took the patient to the hospital, leaving his wife behind to call another ambulance for transport. Interrogation of the patient's ICD showed that he developed supraventricular tachycardia shortly after his wife began yelling. The supraventricular tachycardia subsequently degenerated to atrial fibrillation with a rapid ventricular response that was inappropriately interpreted as a ventricular arrhythmia by the ICD. The response by the ICD occurred because the ventricular rate was within the ventricular tachycardia detection zone, resulting in the delivery of an ICD shock. The patient underwent electrophysiology testing during which atrial flutter was easily induced and then degenerated to atrial fibrillation, requiring electrical cardioversion. Linear radiofrequency lesions were delivered in the area from which flutter originated and neither atrial flutter nor supraventricular tachycardia could be reinduced after ablation. A similarly stressful interaction has not occurred again and the patient has since remained free of atrial tachyarrhythmias for almost 4 years.

Emotional stress is a common triggering factor identified by patients with atrial fibrillation.¹ Episodes of atrial fibrillation are often initiated and maintained by ectopic impulses originating in myocardial cells in the pulmonary veins at or near their connection with the left atrium. Studies in dogs have shown that stimulation of autonomic ganglia^{2 - 3} and combined parasympathetic and sympathetic nerve stimulation⁴ may lead to spontaneous activity of these cells. Emotional stress may directly stimulate or alter the balance of autonomic input in these areas and lead to the initiation of atrial fibrillation.

Bettoni and Zimmermann⁵ examined 24-hour ambulatory electrocardiogram (Holter monitor) tapes from patients with episodes of atrial fibrillation and documented the low-frequency component (0.04 Hz-0.15 Hz) of heart rate variability, a measure of sympathetic stimulation of the heart. They also documented the high-frequency domain (0.15 Hz-0.4 Hz), which provides an assessment of the parasympathetic effects on the heart. They found that the low- to high-frequency ratio, an index of sympathovagal balance, demonstrated a linear increase until 10 minutes before the onset of atrial fibrillation, followed by a sharp decrease soon before atrial fibrillation began. Similar findings have been documented before episodes of atrial flutter.⁶

A 53-year-old man with diabetes with no known cardiac disease was at home with his wife when he collapsed in the bathroom. His wife phoned emergency medical services and an ambulance arrived soon afterward. The paramedics identified the patient's heart rhythm as ventricular fibrillation and initiated advanced cardiac life support. One of the paramedics asked the patient's wife to show her husband's medications. She retrieved the medications and returned to the room where her husband was lying unconscious, with paramedics performing cardiopulmonary resuscitation. She recalls thinking, “ . . . this is really happening . . . this is really happening . . . this is what I always feared.” She subsequently collapsed at the scene, was immediately examined by one of the paramedics who diagnosed ventricular fibrillation, and successfully defibrillated her heart to normal sinus rhythm. Unfortunately, the attempts to resuscitate her husband were unsuccessful.

After the heart was successfully defibrillated to normal sinus rhythm, the woman presented in this case was admitted to an intensive care unit with evidence of cardiogenic shock. Global left ventricular systolic dysfunction was noted, but there was no electrocardiographic evidence of acute myocardial infarction and the QT interval was not prolonged. She required administration of multiple vasoactive agents and spent several weeks in a coma using a ventilator. Although she was previously a healthy 53-year-old woman with no prior cardiovascular symptoms, she had a family history of premature coronary artery disease and was subsequently found to have multivessel coronary artery disease by coronary angiography. She was transferred to a rehabilitation facility after a long hospitalization and then underwent successful coronary artery bypass graft surgery.

The patient also received grief counseling and emotional support from health care workers, family, and friends. While it was assumed that she would likely never again experience an emotional stress of this type, the importance of social support was emphasized as a means of potentially decreasing the effect of stressful life experiences on the cardiovascular system. The patient began taking a regimen of a β-blocker. She has been well for almost 2 years after the event. She has not had any recurrent cardiac events and her heart has normal left ventricular function.

In 1942, Cannon⁷ reported examples of sudden, dramatic, and unexplained deaths following a voodoo curse. He was among the first to propose a physiological basis for such occurrences, focusing on the effects of stimulation of the sympathetic nervous system as the underlying cause. Almost 4 decades ago, Engel,^{8 - 9} the physician who introduced the biopsychosocial model to medicine, published a collection of 170 cases of sudden death in the setting of stressful or disrupting life events. He noted that the types of emotionally stressful events in which death was reported to have occurred could be divided into 8 major categories. Five were related to deaths of others, 2 to personal danger or threats, and 1 occurred in the setting of relief, pleasure, or triumph. The most common reported setting in which women died suddenly was in response to the death of an individual with whom they were close. From his work, Engel concluded that sudden death in women more often occurred in response to death of another individual, whereas in men it more commonly occurred in response to danger.⁹

Significant adverse effects of acute emotional stress on the heart can be divided broadly into 3 areas: left ventricular contractile dysfunction, myocardial ischemia, or disturbances of cardiac rhythm ( Article ). Although these abnormalities are often only transient, their consequences can be gravely damaging and sometimes fatal.

Left Ventricular Dysfunction (Stress Cardiomyopathy or Takotsubo Cardiomyopathy)

Myocardial Ischemia

Ventricular Arrhythmias

Left ventricular dysfunction occurring as a result of acute emotional stress is well described. In most reported cases of what has been termed stress cardiomyopathy, takotsubo cardiomyopathy, or transient left ventricular apical ballooning, a patient (typically an older woman) develops sudden chest pain and shortness of breath soon after an emotionally stressful or traumatic experience.^{10 - 13} Investigation of such a patient typically demonstrates severe left ventricular dysfunction that most prominently affects the cardiac apex and produces transient apical ballooning.^{10 - 11} Although the pathogenesis of this phenomenon is debated, evidence suggests that the high levels of catecholamines resulting from the acutely stressful episode are the most likely cause of the acute left ventricular dysfunction. Catecholamines may stimulate brief periods of epicardial or microvascular coronary vasospasm,^{10 ,21} followed by more prolonged periods of myocardial stunning.¹¹ Myocardial stunning occurs when contractile dysfunction is observed even though there is no irreversible damage to the myocardium and even though coronary blood flow is normal. Transient changes in coronary flow velocity reserve, a measure of cardiac microvascular function, have been demonstrated in this condition.¹² This suggests that coronary microvascular dysfunction may produce brief periods of myocardial ischemia and lead to myocardial contractile dysfunction after emotional stress.

Wittstein et al¹³ reported 19 patients with this clinical syndrome, 18 of whom were women. Plasma catecholamine levels in affected patients were 2 to 3 times as high as among individuals with a Killip class III myocardial infarction (ie, a myocardial infarction with pulmonary edema on presentation). That catecholamines are involved in the pathogenesis of this syndrome is also suggested by the greater β-adrenergic receptor density and enhanced sensitivity to sympathetic stimulation of the cardiac apex compared with other areas of the heart,²² possibly leading to the unusual pattern of contractile dysfunction observed in this condition.

Myocardial ischemia is also well documented following episodes of acute emotional stress. In 1987, Verrier et al²³ developed a canine model that produced intense anger and showed that emotional stress could lead to reductions in coronary blood flow and to the development of myocardial ischemia. These authors measured blood flow in the left circumflex coronary artery of dogs after coronary luminal narrowing was achieved using an adjustable occluder placed distal to the flow probe. Within 2 to 4 minutes after inducing anger by having another dog challenge their access to food, significant reductions in coronary flow were observed, accompanied by evidence of myocardial ischemia shown on the electrocardiogram.

Ambulatory electrocardiographic monitoring was used to compare the frequency of emotional stress during the hour preceding ischemic ST-segment depression with the frequency of such stress during the remaining hours. This study by Gullette et al²⁴ used the case-crossover method and showed that feelings of tension, frustration, and sadness more than doubled the risk of myocardial ischemia in the subsequent hour. Rozanski et al²⁵ evaluated cardiac function in patients with coronary artery disease who were asked to perform a series of stress-producing tasks including arithmetic, the Stroop color task, and simulated public speaking. The majority of patients exhibited cardiac wall motion abnormalities shown on radionuclide ventriculography while performing these stressful tasks.

Ischemia provoked by these stress-producing tasks occurred at lower heart rates than ischemia induced by exercise. This may be because systemic vascular resistance decreases during physical stress but increases during mental stress, although both stress types increase myocardial oxygen demand.¹⁴ The sympathetic hyperactivation produced by acute emotional or mental stress may produce regional wall motion abnormalities even in individuals who have no evidence of cardiovascular disease.²⁶ It has been hypothesized that this occurs because acute emotional or mental stress provokes a sympathetic response that increases left ventricular afterload but fails to produce a sufficient increase in contractility to offset this effect.²⁶ The unique hemodynamic effects of emotional stress may make patients with coronary disease particularly prone to developing ischemia during stress-provoking life events. Mental stress was shown to provoke myocardial ischemia among study participants (6 of 21 participants with coronary artery disease who had negative exercise or chemical nuclear stress test results) while they enacted a difficult interpersonal situation during role play. ¹⁵

Lethal or potentially lethal ventricular arrhythmias induced by emotional stress, of particular relevance to the present case, are also well described.^{27 - 28} It has been estimated that at least 20% of episodes of serious ventricular arrhythmias or sudden cardiac death are precipitated by intense or unusual emotional stress.^{16 - 18} However, much of the evidence on which this is based is circumstantial and relies on patient recall or on the reports of eyewitnesses.

Patients with ICDs, as with the patient described in the first case, provide the opportunity to more closely link emotional stress and ventricular arrhythmias in time, since they provide a record of arrhythmias and device therapies. By inspecting stored electrograms, the timing of ventricular arrhythmias can be precisely determined and these events can then be temporally related to discrete episodes of emotional stress. In a study performed by Lampert et al,²⁹ patients with ICDs were asked to record their activities and emotions during the 15-minute period preceding an ICD shock and during the 2-hour to 15-minute period preceding the shock. As a control, they were also asked to complete a second diary page 1 week later on the same day of the week and time of day they originally received the shock. The authors examined 107 shocked ventricular arrhythmias from 42 patients and found that anger was significantly more likely to be identified by patients in the time immediately preceding a shock than in the corresponding control period. Lampert et al¹⁹ also found that when ventricular tachycardia was induced during mental arithmetic and anger recall (either serially subtracting 7 rapidly from a 3-digit number with mistakes corrected harshly or discussing an annoying or frustrating event while being asked irritating questions), it was faster and more difficult to terminate than it was without exposure to this stressful experience.

Evidence has emerged during the past several decades that characterizes the anatomical substrate and physiological pathways by which emotional stress may trigger ventricular arrhythmias. The neural input to the heart consists of vagal efferents that innervate the sinoatrial node and the atrioventricular node. Atrial muscle is richly innervated by vagal efferents as well, but the ventricular myocardium is only sparsely innervated by the vagus nerve. By contrast, sympathetic stimulation is widespread throughout the heart. Sympathetic efferents are present throughout the atria, especially in the sinoatrial node, and throughout the ventricular myocardium and cardiac conduction system. It has been known for decades that central sympathetic outflow to the heart can trigger ventricular arrhythmias.³⁰

Not surprisingly, early research about the relationship between emotional stress and ventricular arrhythmias focused on the mechanism whereby sympathetic stimulation affected ventricular vulnerability. Lown et al³¹ showed that the threshold for inducing repetitive ventricular ectopic beats was lower in dogs that had undergone stress inducement for 3 previous days by delivery of small electric shocks than in dogs not subjected to repeated stress of this type. Later work by this group showed that this increased susceptibility to ventricular arrhythmias was due to the action of the sympathetic nervous system on the myocardium. Stimulation of the posterior hypothalamus, which increases sympathetic neural traffic to the heart, produced a 40% reduction in the threshold for inducing ventricular fibrillation in dogs.³² This effect was abolished by β-adrenergic blockade.

More recent evidence indicates that asymmetric brain activity is particularly important in making the heart more vulnerable to developing ventricular arrhythmias. Lateralization of cerebral activity during emotional stress may stimulate the heart asymmetrically and produce areas of inhomogeneous repolarization that create electrical instability and facilitate cardiac arrhythmias.³³ The situation that is thought to occur in humans during episodes of emotional stress can be produced in animals by unilateral stimulation of sympathetic nerves, a process that increases the susceptibility to ventricular fibrillation.³⁴ For example, ventricular arrhythmia is consistently and repeatedly elicited in cats after a 2-minute occlusion of the left descending coronary artery and a 30-second stimulation of the left stellate ganglion.^{35 - 36}

There is evidence that a similar phenomenon may occur in humans during periods of emotional stress. Critchley et al²⁰ performed functional positron emission tomography neuroimaging studies while measuring heart rate variability in patients who were simultaneously performing mental arithmetic (rapid, continuous serial subtractions of 7 during a 3-minute period). Mental stress led to a marked increase in the low- to high-frequency ratio and to enhanced tracer uptake in the area of the right lateral midbrain. Laterality shown on positron emission tomography neuroimaging was highly correlated with several electrocardiographic markers of increased inhomogeneity of ventricular repolarization (the total cosine of the QRS-T angle, which characterizes global repolarization heterogeneity and T-wave residua, which is linked to regional repolarization dispersion). This finding suggests a strong relationship between asymmetric brain activity during mental stress and susceptibility to ventricular arrhythmias.

Asymmetry of brain activity during emotional or mental stress is exaggerated in patients with coronary artery disease. Soufer et al³⁷ used positron emission tomography to measure brain activation patterns in response to mental stress in 10 men with coronary disease and in 6 age-matched men without coronary disease (controls). During positron emission tomography neuroimaging, an echocardiographic probe was placed on the chest to assess cardiac wall motion and heart rate and blood pressure were simultaneously recorded. Mental and emotional stress was produced by having these participants perform simple serial subtraction, although they were frequently prompted to go faster while the base number from which they were subtracting was changed. In addition, errors were corrected harshly in an effort to increase emotional stress. Compared with healthy controls and compared with their own baseline conditions, individuals with coronary artery disease demonstrated brain hyperactivation during mental stress in several areas including the left parietal cortex, the left anterior cingulate, the right visual association cortex, the left fusiform gyrus, and the cerebellum. Mental stress also decreased regional cerebral blood flow in individuals with coronary disease compared with healthy controls in the right thalamus, the right superior frontal gyrus, and the right middle temporal gyrus.

The ability of social support to buffer the effects of emotional or mental stress on cardiovascular reactivity has been repeatedly demonstrated.^{38 - 40} β-Blockers have an important role in the treatment of all sudden death survivors. Kiès et al⁴¹ observed 300 sudden death survivors with ischemic heart disease and characteristics like the patient presented in this case and found that the use of β-blockers was associated with a 50% reduction in recurrent ventricular arrhythmias during a mean follow-up period of 2.5 years. β-Blockers have been shown to reduce subsequent arrhythmias in patients presenting with life-threatening ventricular arrhythmia in the setting of emotional stress.²⁷

Many other therapeutic options are available to help patients more successfully manage emotional or mental stress, and the effects of several types of available treatments on the cardiovascular system have been evaluated (Table). Patients who were taught relaxation therapy after having sustained a myocardial infarction were shown to experience fewer cardiac events in a 5-year period than patients who were not taught this technique in a study by van Dixhoorn and Duivenvoorden.⁵¹ Patients were randomly assigned to either physical exercise training only or to exercise training plus individualized instruction on relaxation therapy (focusing on relaxing imagery and initiating voluntary changes in posture, muscle tension, and breathing techniques). Although relaxation therapy was taught to these individuals for only 1 hour per week for 6 weeks, the effects were long lasting. A meta-analysis of 27 studies in which patients with ischemic heart disease were taught relaxation therapy found sufficient evidence that this form of treatment decreases resting heart rate and the incidence of angina, arrhythmias, cardiac events, and cardiac death.⁴⁴

Biofeedback, in which external cues are used to control processes usually considered involuntary (eg, heart rate or blood pressure), can be used to favorably influence sympathetic neural traffic to the heart. Patients can use biofeedback machines to provide information about physiological responses and to help them focus on abdominal breathing and alter the depth and frequency of their respirations in an effort to regulate these processes. Patients with coronary artery disease have successfully used biofeedback to increase heart rate variability,⁴⁵ which is indicative of reduced sympathetic stimulation of the heart. Hypnosis has been shown to reduce the ability of emotional stress to induce repolarization inhomogeneity in healthy participants. Taggart et al⁴⁶ examined participants who were prompted to experience a variety of emotions intended to elicit a wide range of autonomic responses. Increases in low- to high-frequency ratio during emotional stress were positively correlated with changes in ventricular repolarization. Hypnosis reduced the LF/HF ratio and significantly reduced the changes in ventricular repolarization in response to emotional stress. Evidence also exists for a role of several other treatments intended to reduce sympathetic activity in patients with cardiovascular disease or cardiac risk factors including controlled slow breathing,⁴⁷ slow breathing guided by heart rate variability biofeedback,⁴⁸ meditation,^{42 - 43} and yoga.^{49 - 50}

The 2 patients presented in these cases demonstrate some of the effects of acute emotional stress on the heart. Although avoidance of life situations likely to produce emotional or mental stress may be desirable for many individuals, this is seldom feasible. Instead, to promote healthy living in general, patients who have experienced cardiovascular events during stressful experiences are advised to focus on strategies to reduce cardiovascular reactivity to acutely stressful experiences. β-Blockers are appropriate for patients with ischemic heart disease who have survived an episode of sudden cardiac death, but nonpharmacological approaches to manage emotional stress in patients with and without coronary artery disease are also clearly appropriate and merit additional investigation in randomized trials.