Chapter 16. Syncope

• Sudden, unexpected, and transient loss of consciousness.
• Spontaneous and full recovery.

Syncope can be defined as a sudden, transient loss of consciousness and postural tone that fully resolves spontaneously without specific intervention (eg, cardiopulmonary resuscitation [CPR], electrical or chemical cardioversion). The common pathophysiologic mechanism responsible for most syncopal spells is a transient reduction in cerebral blood flow and cerebral hypoperfusion. Reduced cerebral blood flow from cardiovascular and neurocardiogenic causes accounts for most cases in which a diagnosis can be made. Even when cerebral blood flow is normal, a reduced delivery of such essential cerebral nutrients as oxygen and sugar can occasionally cause altered consciousness.

Syncope is a common condition experienced by up to 50% of adults in a lifetime. It is responsible for about 3% of hospital admissions and 4% of emergency department visits. Physicians are frequently consulted to evaluate this symptom and—more commonly—presyncope, dizziness, or light-headedness, which may have a similar pathogenesis.

Syncope has many causes (Table 16–1), most of which have a benign prognosis. Because cardiac causes are associated with greater morbidity and mortality, early recognition of structural heart disease or other cardiogenic causes is important in order to prevent sudden death or injury.

Cardiac Causes

Obstruction to Blood Flow

Any obstructive structural lesion of the left or right side of the heart can critically reduce the cerebral blood flow. Exertional symptoms are common with obstructive lesions because cardiac output does not rise normally with exercise and cerebral perfusion is not maintained. Obstruction to left ventricular outflow occurs with aortic valve stenosis, mitral stenosis, left atrial myxoma, prosthetic aortic or mitral valve dysfunction, and hypertrophic cardiomyopathy. The ventricular arrhythmias that can occur with valvular heart disease may be responsible for both exertional and nonexertional syncope as well as sudden death.

Lesions that obstruct flow through the right side of the heart include right atrial myxoma, pulmonary stenosis, tricuspid stenosis, pulmonary hypertension, and pulmonary emboli. Limitations to right ventricular outflow diminish the cardiac output and the ability to increase the output with exertion. Exertional syncope is common with severe pulmonary hypertension and severe pulmonic stenosis.

Nonexertional syncope can be the result of pulmonary emboli (hypoxia and obstruction of right ventricular outflow) and aortic dissection with pericardial tamponade, which impedes right ventricular filling and decreases cardiac output.

Electrical Disturbances

Bradyarrhythmias and Atrioventricular Block

Both bradyarrhythmias and tachyarrhythmias can cause transient decreased cardiac output with resultant cerebral hypoperfusion. Bradycardias result in symptoms when the rate is so slow that the compensatory increase in stroke volume is inadequate to maintain blood pressure. Periods of ventricular asystole as short as 5 seconds can cause syncope (from the ensuing cerebral hypoperfusion). Mechanisms of symptomatic ventricular bradycardia and asystole include sinus node disease (sinus exit block, sick sinus syndrome, marked sinus bradycardia, or sinus arrest) and second- or third-degree atrioventricular (AV) block. Syncope from Mobitz II second-degree AV block with paroxysms of several consecutive P waves that fail to conduct to the ventricle is called an Adams-Stokes attack. Medications can also cause syncope, and any patient with syncope from bradycardia must be thoroughly questioned regarding medications. Calcium channel blockers, digoxin, β-blockers (including optical formulations), sympatholytics, primary antiarrhythmics, and other medications can all decrease heart rate and increase AV block—enough to cause symptoms in susceptible patients.

Pacemaker malfunction is another possible cause. If a pacemaker was previously implanted for sinus node disease or high-degree AV block and the patient has a recurrence of syncope, there may be a pacemaker system problem, such as battery failure or lead fracture.

Tachyarrhythmias

Transient decreased cardiac output during ventricular or supraventricular tachycardias results when the ventricular rate is fast enough to decrease diastole significantly and thus decrease ventricular filling (preload). Concomitant peripheral vasodepression with resultant hypotension may also play a role in the pathophysiology of syncope with supraventricular arrhythmias.

Ventricular tachycardia occurs most frequently in patients with organic heart disease, particularly coronary artery disease with previous myocardial infarction. Ventricular tachycardia is an important and ominous cause of syncope because ventricular tachycardia usually precedes ventricular fibrillation. Symptoms and prognosis are related to the degree of underlying myocardial dysfunction and the rate and duration of the arrhythmia.

Supraventricular arrhythmias are more likely to cause palpitations and presyncope than true syncope. Although they occur often in young patients with structurally normal hearts, they are also prevalent in structurally abnormal hearts. As with ventricular arrhythmias, the severity of the symptoms is related to the ventricular rate of the arrhythmia and the degree of underlying myocardial dysfunction. Mechanisms associated with syncope include atrial arrhythmias (fibrillation, flutter, tachycardia) with rapid ventricular response, AV nodal reentrant tachycardia, and supraventricular arrhythmias associated with accessory pathways.

Torsades de pointes is a rapid polymorphic ventricular arrhythmia classically known to cause syncope in association with class I antiarrhythmic drugs. Because this arrhythmia can also cause sudden cardiac death, it is very important to understand the reversible conditions that can precipitate it. Most cases of torsades de pointes are associated with acquired long QT syndromes that can occur with several drugs (class Ia and III antiarrhythmics, erythromycin, certain antihistamines, and tricyclic antidepressants), electrolyte abnormalities (hypomagnesemia, hypokalemia, and hypocalcemia), myocardial ischemia, and central nervous system disorders. Rarely, long QT syndrome occurs congenitally, in hereditary forms of prolongation of the QT interval (with or without associated deafness).

Neurocardiogenic Causes

Vasovagal Syncope

This is the most common mechanism of syncope in young, otherwise healthy individuals, constituting approximately 60% of cases. The underlying mechanism is a paradoxical autonomic reflex resulting in profound hypotension (vasodepression) that is often associated with bradycardia (vasovagal). The underlying pathophysiology of vasovagal syncope has been elucidated with the aid of tilt-table testing (see the section on Diagnostic Studies). This technique provides a controlled means of reproducing vasovagal syncope and serves as a means to guide medical therapy. Before the development of tilt-table testing, the diagnosis of vasovagal syncope was largely a matter of exclusion.

Despite extensive study by numerous investigators, several controversies still remain regarding the underlying pathophysiology of vasovagal syncope. When one stands, a significant amount of venous pooling occurs, which can shift from 300 to 800 mL of blood away from the central circulation. To counteract this gravitational effect, baroreceptors in the carotid sinus, aortic arch, and left ventricle sense decreased stretch (low pressure) and send afferent neural messages to the vasomotor center in the brainstem's medulla, which in turn sends efferent stimuli to increase peripheral sympathetic and cardiac tone (vasoconstriction and increased heart rate and contractility) and decrease vagal tone. This is the normal neural reflex arc that maintains cerebral arterial pressure and prevents hypotension or syncope with orthostatic changes.

It is believed that this reflex arc goes awry in susceptible individuals, resulting in a mixed response of vasodepression (hypotension) and cardiac inhibition (bradycardia). This symptomatic reflex arc, commonly known as a vasovagal episode, can occur as pure vasodepression or, rarely, pure cardiac inhibition; the classic mixed response is most common, with hypotension often preceding the bradycardia.

Vasovagal syncope is often, but not always, preceded by predisposing factors such as fear, pain, injury, fatigue, prolonged standing, or such medical procedures as venipuncture. During the initial phase, it is believed that blood pressure and heart rate increase in response to circulating catecholamine increase, sympathetic nerve stimulation, and vagal withdrawal. This is abruptly followed by hypotension, bradycardia, and syncope caused by inhibition of sympathetic vasoconstrictor and cardiac stimulatory activity and increases in vagal tone.

Although the exact pathophysiology is not understood, two conditions are thought to bring on the paradoxical reflex in susceptible individuals: reduced left ventricular filling (an empty ventricle) from decreased venous return and increased left ventricular pressure from increased contractility caused by cardiac stimulation from circulating catecholamines. A vigorously contracting empty ventricle is thought to paradoxically increase left ventricular pressure (mimicking hypertension), which is sensed by left ventricular stretch receptors (C fibers). This afferent neural arc to the nucleus tractus solitarius in the medulla results in an efferent arc that paradoxically decreases sympathetic and increases vagal tone, causing hypotension and bradycardia, respectively. Because both the sinus node and AV node are richly innervated with autonomic nerves, bradycardia can be caused by both vagal depression of sinus node automaticity and decreased AV nodal conduction with AV block. Assuming a supine posture provides spontaneous resolution because the gravitational effects on cerebral blood flow are immediately neutralized. Figure 16–1 shows a two-lead ambulatory recording of a classic vasovagal episode in a patient with previously undiagnosed syncope.

Vasovagal syncope is usually considered a benign form that can often be treated with patient education and avoidance of the precipitant. Some individuals, however, have frequent severe spells that result in injuries and, rarely, sudden death. Several therapies (discussed later) are available for patients who require intervention.

Carotid Sinus Hypersensitivity

The carotid sinus has baroreceptors located in the internal carotid artery just above the bifurcation of the common carotid artery. These are sensitive to stretch and pressure and give rise to afferent impulses to the vasomotor center in the medulla. In susceptible individuals, particularly the elderly, activation of this reflex results in vagal efferent stimulation, which causes marked bradycardia or AV block (the cardioinhibitory response) with vasodepression (from efferent sympathetic inhibition). In contrast to vasovagal syncope, the cardioinhibitory response is usually dominant. Syncope from carotid sinus hypersensitivity usually is suggested by historical factors: precipitation with turning the head, shaving the neck, or wearing a tight shirt collar. Spontaneous attacks can also occur. Full consciousness is usually regained in less than a minute after attaining a supine position. When syncope from carotid sinus hypersensitivity is suspected, carotid sinus massage (CSM) can be used to diagnose this condition; hypersensitivity is generally defined as cardiac asystole of 3 seconds or more or a decline in systolic blood pressure of at least 50 mm Hg.

Situational

The pathophysiologic mechanisms of syncope associated with micturition, defecation, coughing, and swallowing are not precisely elucidated. It is suggested that the vagal-sympathetic autonomic reflex (hypotension and bradycardia) is involved, with afferent neural stimulation coming from the involved viscera. The Valsalva maneuver (used during micturition, defecation, and coughing) is thought to contribute to hypotension by decreasing venous return, which decreases cardiac output.

In micturition syncope, which classically occurs on waking, a combination of physiologic changes that occur during sleep (eg, sudden decompression of the bladder, decreased heart rate and blood pressure) may contribute to syncope. Tussive syncope classically occurs in middle-aged men who drink alcohol, smoke, have chronic lung disease, and experience paroxysms of severe cough; it is thought to be largely due to decreased venous return caused by the Valsalva-like action of coughing. Deglutition syncope is believed to be entirely due to dysfunction of the afferent-efferent reflex arc and is associated with structural abnormalities of the esophagus (eg, diverticula, diffuse esophageal spasm, achalasia, stricture) and heart (eg, acute rheumatic carditis, myocardial infarction).

Orthostatic Hypotension

Volume depletion, medications, and autonomic dysfunction (primary and secondary) can all lead to cerebral hypoperfusion and syncope in the standing position. Primary chronic autonomic dysfunction is associated with Shy-Drager syndrome, Parkinson disease, and other rare dysautonomias. Secondary autonomic dysfunction with syncope has been described with diabetes mellitus, anemia, amyloidosis, multiple sclerosis, and human immunodeficiency virus (HIV) infection. Medications that commonly cause orthostatic hypotension include diuretics, antihypertensives (eg, angiotensin-converting enzyme [ACE] inhibitors, β-blockers, calcium channel blockers), and ethanol.

Essential to the diagnosis of orthostatic hypotension is evaluation of postural changes in blood pressure. Normal reflex mechanisms maintain blood pressure on standing (see discussion of neurocardiogenic syncope). The inability to maintain cerebral flow when standing or sitting upright, with resultant hypotension, dizziness, or syncope, is called orthostatic hypotension. Quantitatively, it has been defined as a 20 mm Hg or greater decline in systolic pressure or a 10 mm Hg or greater decline in diastolic pressure on assuming an upright position. Because this finding is frequent and asymptomatic in the elderly, the clinical diagnosis of syncope from orthostatic hypotension requires the presence of symptoms with blood pressure changes.

History & Physical Examination

According to several prospective studies, the initial clinical evaluation establishes the cause of syncope or suggests the necessary diagnostic test in up to 85% of the patients in whom a diagnosis will eventually be made. A detailed history and physical examination are therefore essential parts of the initial clinical evaluation. Relevant historical information includes all details leading up to the event, precipitating factors (micturition, cough, exertion), premonitory symptoms (aura), onset (sudden or slow), associated symptoms (palpitations, chest pain, headache), activity (at rest or with exercise), position (standing, sitting, changing position), and details about the episode (injury, incontinence, rapid recovery versus postictal state) and the frequency and severity of the events. The history should include factors suggestive of cardiac or other systemic illnesses, such as a family history of cardiac illness, arrhythmias, syncope, sudden death, or pacemaker implantation. Because about 10% of syncope is caused by the use of medications and recreational drugs, information regarding their use, prescribed dosages, and the amount actually taken is very important. All witnesses to the event (family, friends, bystanders) should be thoroughly interviewed because they can often supply details that the patient cannot. These witnesses may be able to provide information about the patient's complaints just prior to the event and observations during both the event (eg, pulse rate and rhythm, color, presence of spontaneous breathing, seizures, or seizure-like activity) and recovery. Electrocardiographic (ECG) recordings made by paramedics at the scene or recorded en route to the hospital and in the emergency department may provide important clues.

All patients need a complete cardiac, peripheral vascular, and neurologic examination, including an assessment of positional blood pressure and heart rate. Cardiac murmurs may be suggestive of structural heart disease, such as valvular stenosis. Differential blood pressure and pulse intensity may suggest aortic dissection or subclavian steal syndrome. Focal neurologic findings may point to seizure, stroke, or transient ischemic attack. The development of similar symptoms with upright posture and a corresponding decrease in blood pressure implicate orthostatic hypotension.

Several bedside maneuvers can be used to provoke syncope or presyncopal symptoms in appropriate patients. Arm flexion and extension may be used to provoke symptoms of subclavian steal syndrome; neck flexion and extension may elicit the symptoms of vertebrobasilar insufficiency. Open-mouthed hyperventilation for 1–3 minutes may elicit symptoms described in the history. CSM (when not contraindicated by presence of bruits or known carotid arteriosclerotic disease) can be performed at the bedside with ECG and blood pressure monitoring. Although the technique is not standardized, carotid massage is usually performed with the patient in the supine position, with 5 seconds of firm massage of each carotid body. Note that simultaneous bilateral massage is never done. A positive test is defined as at least 3 seconds of asystole with hypotension or reproduction of symptoms. Because of the high false-positive rate for CSM in the elderly, the diagnosis of carotid sinus hypersensitivity should only be considered if clinical events are associated with activities that press or stretch the carotid sinus.

Toxicology screens and medication levels may provide useful information and should be ordered if suggested by the patient's or the witness's history. Figure 16–2 outlines a suggested diagnostic approach.

Noninvasive Diagnostic Studies

Electrocardiography

A 12-lead ECG should be part of the routine clinical evaluation of syncope. A normal 12-lead ECG generally portends a good prognosis, with a very low incidence of either cardiogenic causes or diagnostic findings during invasive electrophysiologic testing. Unfortunately, the initial ECG rarely establishes arrhythmia as the cause unless complete heart block, ventricular tachycardia, or another abnormality is present at the time of the tracing. Given the unpredictably episodic nature of syncopal spells and the fact that most patients recover completely from them prior to the ECG, it is rare that the 12-lead ECG is diagnostic. More often, certain findings will suggest the possibility of a specific diagnosis. For example, Q waves suggestive of previous myocardial infarction correlate with abnormal electrophysiologic studies and the presence of inducible ventricular arrhythmias. Delta waves suggestive of Wolff-Parkinson-White syndrome make supraventricular tachycardia a very likely diagnosis. A long QT interval suggests torsades de pointes, and bifascicular block correlates with abnormal electrophysiologic findings, including high-degree AV block and inducible ventricular arrhythmias.

The most common abnormalities found in patients with syncope are bifascicular block, prior myocardial infarction, left ventricular hypertrophy, sinus bradycardia, and first-degree or Wenckebach AV block. Although these findings are all nonspecific, they may suggest a cardiac cause. The ECG or rhythm strips recorded by paramedics, the emergency department, or hospital ward lead to a specific diagnosis in about 10% of patients. The most common diagnoses include ventricular tachycardia and bradyarrhythmias caused by sinus node dysfunction or high-degree AV block.

Echocardiography

If the history, physical examination, and ECG do not suggest the diagnosis, or if underlying heart disease is suspected, an echocardiogram is a valuable diagnostic tool. Echocardiography may diagnose the likely cause of syncope (aortic stenosis, hypertrophic heart disease); but, more commonly, it is useful in directing further evaluation. Because morbidity and mortality with syncope are directly related to the presence and severity of structural heart disease, the echocardiogram can aid the physician in assessing the patient's prognosis and the necessity for further invasive evaluation. For example, the finding of a low left ventricular ejection fraction suggests ventricular arrhythmias and would be an indication for further electrophysiologic evaluation or an implantable defibrillator. Unsuspected findings on echocardiography are reported in 5–10% of unselected patients with syncope. Although this is similar to the ECG, the cost of the study limits it to persons with no obvious cause or suspected underlying heart disease.

Exercise Stress Testing

Exercise stress testing may be useful in patients with exertional symptoms. Exertional hypotension may occur as a result of underlying structural heart disease, chronotropic incompetence, or severe conduction disease resulting in AV block with increased atrial rates. Supraventricular and ventricular arrhythmias may be provoked with exercise. Hypotension and bradycardia at the termination of exercise can be diagnostic of reflexive vasomotor instability. Also, exercise or other forms of stress testing in those who cannot exercise may detect myocardial ischemia—a potential substrate for ventricular arrhythmias.

Ambulatory Monitoring

Prolonged ECG monitoring can be useful for documenting transient bradyarrhythmias or tachyarrhythmias. The most common method used is a Holter ambulatory ECG recording. Typically, two surface ECG leads are recorded for a period of 24–48 hours. Holter recording can be particularly helpful in diagnosing bradyarrhythmias, such as significant sinus pauses and transient high-degree AV block. When interpreting Holter recordings, clinicians must recognize that several abnormalities can occur in healthy, asymptomatic patients, and correlation with symptoms is often essential. Premature atrial and ventricular contractions, brief and paroxysmal atrial tachycardia, episodic AV Wenckebach block, sinus bradycardia, sinus pauses of up to 3 seconds especially when asleep, and nonsustained ventricular tachycardia can all be seen in normal, asymptomatic patients. Sinus pauses of longer than 3 seconds, Mobitz II AV block, complete heart block, and frequent nonsustained ventricular tachycardia are far less prevalent.

Because the correlation of symptoms with an arrhythmia in patients with syncope provides the most valuable information, it is essential for the patient to keep an accurate diary of symptoms and activity during the recorded interval. Ambulatory monitoring is often useful in excluding arrhythmia mechanisms of syncope when patients experience syncope or presyncope without associated arrhythmias. Conversely, ambulatory monitoring may reveal significant conduction abnormalities or arrhythmias when the patient is asymptomatic. Frequent sinus pause of ≥ 3 seconds with associated atrial fibrillation indicates significant sinoatrial dysfunction. Asymptomatic Mobitz II AV block suggests distal conduction disease, with the possibility of prolonged AV block resulting in Adams-Stokes attacks. Nonsustained ventricular tachycardia in patients with significant left ventricular dysfunction (ejection fraction < 40%) correlates with a risk of sudden death from sustained ventricular arrhythmias and warrants further electrophysiologic evaluation (see section on Invasive Electrophysiology Studies).

A significant limitation of Holter ambulatory ECG recording is how infrequently patients experience either syncope or associated symptoms during the test. The results of studies suggest that Holter monitoring captures the heart rhythm during spontaneous syncope in only 4–10% of patients. For this reason, external ambulatory loop recorders have been used extensively for the evaluation of syncope. These devices record a single ECG lead continuously and can be worn by a patient for weeks. When activated by the patient or an observer, a rhythm strip is saved that includes several minutes surrounding the syncopal or presyncopal episode (Figure 16–3). The rhythm strip is sent to the laboratory using trans-telephonic equipment for added convenience for the patient and physician. Loop recorders are useful for evaluating patients with episodic syncope, presyncope, or dizziness, with or without associated palpitations.

For patients with infrequent symptoms, small loop recorders can be implanted under the skin and have a battery life of 1 year. They can be activated by the patient or a companion remotely, but newer models automatically record abnormal rhythms. This feature is especially valuable for patients who are rapidly incapacitated by their symptoms and cannot reliably activate a device. Although implantable loop recorders were once reserved as a last-resort diagnostic strategy, some physicians recommend earlier implantation because studies have shown diagnostic yields up to 87%.

Head-Up Tilt-Table Testing

Gravitational shifts in blood volume have long been recognized as a stimulus to neurocardiogenic syncope (also known as vasomotor, vasovagal, neurally mediated, or neurocardiogenic syncope). The use of the tilt table as a provocative maneuver in the diagnosis of unexplained syncope is decreasing because of concerns about the sensitivity, diagnostic yield, and reproducibility of the test. In patients with a normal cardiac evaluation, the pretest probability of neurocardiogenic syncope is high, so tilt-table testing is unlikely to be of value. The technique starts with monitoring the fasting patient for 10 minutes in the horizontal position, using noninvasive brachial or finger blood pressure, oxygen saturation, continuous ECG, and if somatization disorder is suspected, electroencephalogram (EEG) (true syncope can be differentiated from malingering). The patient is then tilted 60–80 degrees for up to 45 minutes; in children and adolescents, positive tests tend to occur sooner, and a period of 30 minutes is sufficient. The tilt produces a shift in blood volume distribution away from the central circulation and thorax to dependent peripheral vessels. This causes a decrease in central venous pressure, ventricular filling, stroke volume, and mean arterial blood pressure. Normally, activation of the baroreceptor-vasomotor reflex (described earlier) and renin-angiotensin system and release of catecholamines result in maintenance of blood pressure through increased heart rate and vasoconstriction. Monitoring of a normal passive tilt would show a small decrease in systolic blood pressure, with an increase in diastolic blood pressure, mean arterial pressure, and heart rate. All of these are considered normal adjustments to gravitational stress.

The test is considered positive if syncope or presyncope occurs with hypotension, with or without bradycardia. The patient is then quickly placed in the horizontal position, where normal compensatory mechanisms restore blood pressure and consciousness. The test is considered negative if symptoms and hemodynamic abnormalities fail to occur by 45 minutes of tilt. The tilt can be repeated with a provocative agent, such as the β-agonist isoproterenol, in selected patients (discussed later).

As previously discussed, various degrees of hypotension and bradycardia secondary to a paradoxical reflex that increases vagal tone and decreases sympathetic tone develop in susceptible patients. Both hypotension and bradycardia are present (a mixed response) in most patients. Hypotension usually predominates, however, and tends to occur before bradycardia (as shown in Figure 16–1). Hypotension that occurs alone, without bradycardia, is considered a purely vasodepressor response. Rarely, a patient may have a purely cardioinhibitory response (asystole).

The test has been shown to be 80–90% reproducible and specific, with a low false-positive rate (< 10%) in asymptomatic individuals. However, sensitivity ranges from 25% to 80% depending on the population studied. The 2006 American Heart Association (AHA)/American College of Cardiology Foundation (ACCF) scientific statement of the evaluation of syncope no longer recommends routine tilt-table testing for undiagnosed syncope. However, there are some patients in whom the diagnosis is unknown after a cardiac evaluation and the history is not helpful. Some of these patients would benefit from a tilt-table test, whereas in others a presumptive diagnosis of neurocardiogenic syncope is adequate.

Other Testing

Routine laboratory tests (blood counts and chemistries) rarely reveal useful diagnostic information. Unless specifically suggested by the history and physical examination, it is not recommended in the evaluation of syncope. Likewise, EEG, computed tomography, or magnetic resonance imaging studies are of little use in patients whose history is not suggestive of a neurologic cause. These tests should be considered only in those patients whose neurologic history and physical examination are suggestive of seizure or other neurologic condition. Although routinely performed, the usefulness of transcranial Doppler ultrasonography in patients with syncope is unclear. Transient ischemic attacks rarely result in syncope without other associated symptoms. In patients with bruits on physical examination, ultrasonography is reasonable and may be useful in leading to a diagnosis.

Invasive Electrophysiology Studies

An invasive electrophysiology study (EPS) uses multielectrode catheters inserted percutaneously and guided under fluoroscopy or by magnetic sensors to specific cardiac locations. Electrode recording and pacing protocols are then performed to assess the patient's conduction system, including sinoatrial and AV nodal function and the distal conduction system (His-Purkinje). In addition, supraventricular and ventricular arrhythmias may be induced and their mechanisms determined in susceptible patients. In selected patients, ablation of the arrhythmic substrate with radiofrequency energy can be performed, often curing the patient of the condition.

Because patients are generally supine and sedated during EPS, one significant limitation is the inability to definitely correlate induction of arrhythmias with syncope; however, a rapid tachycardia associated with a significant decrease in systolic blood pressure generally supports the clinical importance of inducible arrhythmias. Induction of certain arrhythmias, such as atrial fibrillation, atrial flutter, polymorphic ventricular tachycardia, and ventricular fibrillation, can be a nonspecific finding. Asymptomatic patients with normal hearts can have all these induced with aggressive stimulation protocols.

The AHA/ACCF scientific statement on the evaluation of syncope does not recommend routine EPS in patients without underlying heart disease. It does recommend EPS in patients with cardiac disease and syncope that remains unexplained after appropriate evaluation, especially if the ejection fraction is reduced. In addition, EPS is reasonable in patients with recurrent unexplained syncope without structural heart disease and a negative head-up tilt test. In patients with no structural heart disease, normal ECG, normal ambulatory monitoring, and recurrent syncope not associated with injury, EPS is usually not diagnostic. EPS is not indicated in patients with a known cause of syncope for whom treatment will not be guided by electrophysiology testing.

The most significant finding at EPS is the induction of ventricular tachyarrhythmias. In addition, the induction of supraventricular tachycardias with associated hypotension is considered a positive study. Other significant findings include a prolonged corrected sinus node recovery time longer than 1000 ms, prolongation of the His-to-ventricle activation interval greater than 100 ms (normally 35–55 ms), or induction of infra-Hisian conduction block during rapid atrial pacing. These three findings are associated with significant bradyarrhythmias. Unfortunately, the diagnostic sensitivity of EPS for bradyarrhythmias is low, and further ambulatory monitoring is often necessary after a nondiagnostic EPS.

Seizure

True seizures are rarely established as the cause of a syncopal event, although they are frequently considered. A history of loss of bowel or bladder tone, witnessed convulsions, a preceding aura, and slow recovery with prolonged postictal state strongly suggest seizure. Syncope occurs frequently without a preceding aura or loss of sphincter tone and characteristically has a rapid spontaneous recovery without postictal state.

Some factors blur the distinction between these two entities: convulsions may occur with syncope (from cerebral anoxia or ischemia), the episode may be unwitnessed, the patient may not be able to recall associated symptoms, and the patient may have temporal lobe epilepsy, in which loss of consciousness is not associated with tonic-clonic movements. Temporal lobe epilepsy can also cause arrhythmias. Electroencephalography is indicated when the history cannot provide a clear distinction between syncope and seizure.

Metabolic Disorders and Hypoxia

Any condition that starves the cerebrum of essential nutrients (electrolytes, oxygen, glucose) or markedly changes pH can cause somnolence or coma. These conditions are not usually associated with spontaneous recovery (ie, without intervention); they tend to be longer lasting, and treatment is directed at the underlying abnormality. Hypoglycemia in outpatient diabetics, resulting from excessive insulin injection, can cause somnolence with a normal pulse and blood pressure; it is rapidly correctable with glucose.

Cerebral Vascular Insufficiency and Extracranial Vascular Disease

Although altered states of consciousness are known to occur with cerebral vascular events, true syncope is an uncommon presentation. Diseases of the intracranial and extracranial vessels can cause strokes and transient ischemic attacks, rarely syncope. Focal neurologic defects found during the physical examination are clues to the presence of cerebral vascular insufficiency, and differential peripheral arterial pressures or bruits suggest extracranial arterial disease. The extracranial arteries most commonly involved are the vertebrobasilar (occlusion), carotids (occlusion, emboli), aortic arch (dissection), and subclavian artery (stenosis), which may produce the subclavian steal syndrome.

Psychiatric Disorders with Hyperventilation and Pseudoseizure

Psychiatric illnesses are an important cause of syncope, especially in younger patients with multiple episodes and no organic heart disease. Depression, anxiety attacks, somatization disorder, panic disorder, and substance abuse (eg, alcohol, cocaine, sedatives) have all been associated with syncopal spells. In addition, pseudoseizure that may or may not be associated with true seizure disorder may be present.

Hyperventilation may cause syncope by producing hypocapnia and metabolic alkalosis, both of which stimulate cerebral arterial chemoreceptors to produce cerebral arterial vasoconstriction and decrease blood flow. Hyperventilation is frequently associated with such psychiatric illnesses as anxiety, depression, and panic attacks, but it can also occur sporadically in patients without psychiatric disease. A history of rapid breathing and perioral numbness is suggestive. A monitored hyperventilation maneuver that reproduces the patient's symptoms helps confirm the diagnosis.

Medical therapy is primarily targeted toward tachyarrhythmias, and pacemakers are generally used for bradyarrhythmias. In some situations, however, both are needed. For example, in sick sinus syndrome, a pacemaker is needed to prevent sinus bradycardia and pauses, and antiarrhythmic medications are needed to prevent the tachycardia (paroxysmal atrial fibrillation and flutter). Moreover, antiarrhythmic drugs can exacerbate bradycardia in some patients and necessitate implantation of a pacemaker to prevent medication-induced bradycardia and syncope.

Pharmacologic and Nonpharmacologic

Supraventricular arrhythmias can be treated with an assortment of antiarrhythmic medications (sodium channel blockers, β-blockers, potassium channel blockers, calcium channel blockers) with a wide range in therapeutic efficacy. Inherent in the medical treatment of arrhythmias, unfortunately, is a consistently high rate of side effects and the occurrence of proarrhythmia, worsening the arrhythmia or creating new arrhythmias such as torsades de pointes. Sustained ventricular arrhythmias can be suppressed with antiarrhythmic drugs; however, recent studies favor their use only in combination with an implantable cardioverter-defibrillator (ICD) in most patients.

Coronary artery, valvular, and other structural heart disease can usually be managed medically. If the disease has reached the point of causing syncope, however, it is usually severe enough to warrant surgical intervention or percutaneous revascularization.

Neurocardiogenic and situational syncope can be treated in a number of ways (Table 16–2). Education and avoidance of known precipitants are a first step. α-Agonists, such as ephedrine or midodrine, can be used to enhance vasoconstriction and counteract the vasodilation portion of the reflex arc. β-Blockers (eg, metoprolol, pindolol) have been used effectively; they are thought to work by decreasing inotropy and preventing the afferent loop of the paradoxical reflex arc. Propantheline and scopolamine have been used for their anticholinergic effects against the vagal efferent loop of the reflex. The sodium channel blocker disopyramide has been used successfully, although its mechanism of action against neurocardiogenic syncope is poorly understood. Theophylline, an adenosine antagonist, has also been used successfully (adenosine is thought to be a mediator of vasodepression in the reflex arc). Some success has been reported using the antidepressants (serotonin reuptake inhibitors) fluoxetine and sertraline. Although the mechanism of action is not fully understood, serotonin is believed to play a central role in the development of vasovagal syncope. Volume expansion with liberalization of salt and water intake, administration of the mineralocorticoid fludrocortisone, and the use of thigh-high elastic stockings are effective because they can augment venous return. Also, isometric exercise such as handgrip and leg tensing can raise systemic blood pressure and prevent syncope if applied as soon as premonitory symptoms are appreciated.

None of these therapies has been shown to have any particular therapeutic advantage over the others. The specific therapy chosen should be based on the patient's age, lifestyle, and medication side effects. Each therapy can be given an empiric trial until an effective agent is found, or a repeat tilt-table test response can be used as a guide to long-term therapy.

When drug-induced syncope is suspected, the offending agent should be discontinued and the patient monitored closely for recurrence. In the case of documented or suspected torsades de pointes, the underlying drug; electrolyte abnormality; or thyroid, neurologic, or cardiac cause must be sought and corrected. Intravenous magnesium, lidocaine, or isoproterenol or temporary ventricular pacing can be used to stabilize a patient with torsades de pointes while the underlying problem is corrected.

Electrophysiologic Therapies

Implantable devices, such as the ICD and pacemakers, surgical procedures, and catheter ablation have all been used to treat syncope in appropriate patients. Besides causing syncope, ventricular tachycardia has the potential to degenerate to ventricular fibrillation. Many electrophysiologists consider syncope with ventricular tachycardia to be aborted sudden cardiac death. For this reason, patients with structural heart disease, documented ventricular tachycardia, and associated syncope are best treated with an ICD; moreover, patients with structural heart disease, syncope of unknown origin, and inducible ventricular arrhythmias during EPS are also most commonly treated with an ICD. The device continuously monitors the patient's heart rate and can apply a direct-current (DC) shock to convert ventricular fibrillation or tachycardia to sinus rhythm. ICDs have been found to reduce the likelihood of mortality in appropriate patients when compared with antiarrhythmic drugs. Although an ICD does not always prevent syncope, it effectively prevents sudden cardiac death. In addition, an ICD can terminate some ventricular tachycardias with rapid ventricular antitachycardia pacing (ATP). This therapy is painless and often unnoticed by the patient.

Ventricular tachycardia with associated ventricular aneurysm has also been successfully treated with open heart surgical ablation. With the development of nonthoracotomy ICD implantation, surgical ablation has become less common. In addition, nonsurgical catheter ablation techniques have been developed that allow for the treatment of some selected patients in the electrophysiology laboratory. In most patients, surgical or catheter ablation is used to reduce frequent appropriate ICD shocks and does not replace ICD therapy.

Single- and dual-chamber pacemakers are very effective in reducing symptomatic recurrences and preventing sudden death in patients with syncope caused by bradyarrhythmias from sinus node dysfunction, sick sinus syndrome with bradyarrhythmias and tachyarrhythmias, high-degree AV block, and carotid sinus hypersensitivity. Pacemaker therapy has been shown to have little role in treating neurocardiogenic syncope, largely because vasodepression is often the dominant component of the reflex and is not corrected by pacing. Pacemaker implantation may occasionally be appropriate for these patients if temporary pacing shows that the symptoms can be minimized (converting syncope to dizziness only) or if the episodes are purely cardioinhibitory. It may also be considered if multiple medications have been ineffective in severely diseased patients with mixed (vasodepressive and cardioinhibitory) episodes and the desired goal is to convert syncopal to presyncopal spells by abolishing the cardioinhibitory component.

In the past decade, electrophysiologic percutaneous catheter techniques using radiofrequency energy as an ablative source have been more than 95% effective in curing patients with supraventricular arrhythmias from AV nodal reentry or accessory pathways. Although successful ablation of atrial tachycardias, atrial flutter, some ventricular tachycardias, and even atrial fibrillation can be performed in selected patients, recurrence is not uncommon. In the case of atrial fibrillation with an uncontrollably rapid and symptomatic ventricular response, creation of complete heart block with catheter ablation and implantation of an activity-modulated ventricular pacemaker can effectively ameliorate symptoms.

The studies to date suggest that mortality rates in patients with syncope are strongly correlated with the presence or absence of underlying structural heart disease. Dividing syncope patients into three groups is helpful for assessing their prognosis. In patients with cardiac causes of syncope, mortality is high at 1 year, ranging from 18% to 33%. In patients with noncardiac causes of syncope, 1-year mortality is much lower and ranges from 0% to 12%. Patients in whom syncope remains of unknown origin (despite an appropriate directed workup) do fairly well with a 6% 1-year mortality rate. The incidence of sudden death in the cardiac patients is also much higher than in the other two groups. Recurrence rates in all three groups are similar: up to 15% per year. Recurrences do not predict an increase in mortality, although they are associated with increased morbidity (eg, fractures, soft tissue injury).

In the emergency department, the main goal of triage is to identify high-risk patients who would benefit from hospitalization or observation with monitoring for 24–48 hours and appropriate treatment. Presumably low-risk patients could be discharged with outpatient follow-up. To facilitate this effort, several rules and scores have been developed. The easiest to use are the rules that are lists of risk factors and if one or more are present, the patient is at high risk of an untoward event. The only externally validated rule is the San Francisco Syncope Rule, which states that if any of the following five conditions is present, the patient should be admitted: congestive heart failure; hematocrit < 30%; ECG abnormalities; shortness of breath; or systolic blood pressure < 90 mm Hg (acronym CHESS). The scores assign risk points to several variables, and the derived total score is related to risk. The scores tend to focus on cardiac adverse events and were not designed for use in initial triage.

Recurrent syncope of unknown origin remains a difficult management problem. Patients with multiple atraumatic episodes and structurally normal hearts are more likely to have psychiatric illness or neurocardiogenic syncope and less likely to have electrophysiologic abnormalities at EPS. Memory-loop event recorders (to rule out bradyarrhythmias and tachyarrhythmias), tilt-table testing, and neuropsychiatric evaluation are probably the best means of evaluating these patients further. Patients with syncope of unknown origin and no structural heart disease have an excellent prognosis. They need to be reassured, but close follow-up and episodic reevaluation are also recommended. Intensive follow-up and reevaluation are needed for the patient with underlying heart disease and syncope of unknown origin. Such patients are not rare, and therapy should be directed at likely causes (such as EPS-induced sustained ventricular tachycardia). Empiric therapy (such as antiarrhythmic medication to suppress ventricular ectopy) is not without significant risk, however, and cannot be recommended as a general approach.