approach to the patient with syncope: c history and...

By definition, syncope is asymptom, not a disease.Finding the cause of thissymptom, which can presentin myriad conditions, is a

challenge to the clinician. Syncope refers toa transient loss of consciousness and of pos-tural tone at the same time. Both events occurtogether when the brain fails to receive suffi-cient perfusion and, hence, insufficient oxy-gen and glucose. The constellation of othersymptoms that may or may not accompanysyncope is fundamental in guiding the clini-cian as to its underlying etiology. Althoughsyncope is usually brief in duration withspontaneous recovery, patients and their fam-ilies may be legitimately anxious about itsoccurrence because it may herald a conditionthat can be associated with serious morbidityor sudden death. Even if it occurs as a result

of a benign process, unpredictable loss of con-sciousness can be the cause of other trauma asa result of falls, motor vehicle accidents, andsimilar events. In addition, the search foranswers as to what precipitated the syncopalevent may require lengthy, costly, and some-times invasive diagnostic workups that fail tolead to a definitive diagnosis in 30% to 40%of patients.1,2

EPIDEMIOLOGY AND PATHOPHYSIOLOGY

EPIDEMIOLOGY: A PERVASIVE PROBLEM

Data from the Framingham study (n =5209 community-dwelling men andwomen) indicate that at least 3% of thepopulation will have a syncopal event with-in a 26-year period, and that 30% will havemore than one event.3 Prognosis is not nec-essarily malign or benign if syncope recurs.As individuals age, prevalence increases; the

Approach to the Patient with Syncope: History and Physical Examination Are KeyNora Goldschlager, MD

Dr Goldschlager is Professor of Clinical Medicine, University of California at San Francisco, and Director, CoronaryCare Unit, Electrocardiography, and Pacemaker Labs, San Francisco General Hospital, San Francisco, California.Dr Goldschlager reports she receives honoraria from Guidant Corporation; Medtronic, Inc; SiemensPacesetter, Inc; Pfizer, Inc; and Wyeth.Off-Label Product Discussion: The author of this article does not include information on off-label use of products.Correspondence to: Nora Goldschlager, MD, San Francisco General Hospital, 1001 Potrero Avenue, 5G1,San Francisco, CA 94110.

Syncope, the transient loss of consciousness that occurs due toinadequate cerebral perfusion, has many underlying etiologies, both cardiac andnoncardiac. Common conditions, such as vasovagal syncope, carotid sinus syn-drome, orthostatic hypotension, and primary autonomic nervous system failure,are discussed. The cardiac causes, however, are responsible for the significantmorbidity and mortality observed in all age groups and populations. Syncopeaccounts for 3% to 5% of emergency department visits and up to 6% of hospitalvisits annually in the United States. A detailed history and physical examinationare key in determining the cause and planning the subsequent management ofsyncope. It is critical to distinguish between true syncope and other conditions thatmay be misdiagnosed as syncope, such as seizures and other neuropsychiatricepisodes. Differential diagnosis via selection of appropriate diagnostic tests andmanagement of the common causes of syncope, including those in special popu-lations, such as athletes and the elderly, are the focus of this review.(Adv Stud Med. 2003;3(5):265-276)

ABSTRACT

CA

RD

IOLO

GY

Advanced Studies in Medicine 265

266 Vol. 3, No. 5 ■ May 2003

CARDIOLOGY

annual incidence of syncope in the elderly isapproximately 6%.4 Furthermore, in the UnitedStates, 3% to 5% of emergency department visitsand up to 6% of hospital admissions annually arefor patients who receive a diagnosis of syncope.1 Intotal, approximately 1 million patients are evaluatedfor syncope each year in the United States.5 Thesestatistics probably underestimate the actual inci-dence because many individuals who experience asolitary episode of syncope do not seek medicalattention. It is safe to say that syncope is very com-mon, crosses all age groups, and has varying signifi-cance depending on the age of the patient.

PHYSIOLOGY AND PATHOPHYSIOLOGY

Neurological and cardiovascular functions main-tain consciousness and the ability to stand upright.Upright posture causes pooling of blood in the lowerextremities due to gravity. With upright stance, out-put to the brain decreases and muscle stretchdecreases. The body detects these changes to bloodpressure and compensates via the baroreceptors,which are sensitive to stretch. Baroreceptors arelocated in the carotid sinuses, aortic arch, greatveins, right atrium, and both ventricles of the heart.Depending on the degree of stretch sensed by thebody’s receptors, sensory impulses are conducted tothe vasomotor center in the medulla and then to thehypothalamus. The sympathetic and parasympa-

thetic branches of the autonomic nervous systemmediate these responses back to the heart and bloodvessels. (Arterial baroreceptors respond to increasesin blood pressure; cardiopulmonary baroreceptorsmediate changes in blood volume.)

If blood pressure or volume is low, the sympa-thetic nervous system responds by increasing cardiacoutput (increased heart rate and stroke volume) aswell as by causing selective vasoconstriction at thearteriolar level. Blood pressure rises due to these con-ditions of increased peripheral resistance, and even-tually, a negative feedback loop returns the body tobaseline status.6 Chemoreceptors that regulate cere-bral vasodilatation (depending on pH, and on con-centrations of oxygen and carbon dioxide) and renaland endocrine system functions also play a role inmaintaining adequate circulation and, hence, cere-bral perfusion.

Syncope results when blood flow to the brainis inadequate to sustain consciousness. Cessationof perfusion for as few as 6 or 8 seconds has beenshown to cause loss of consciousness.7 Inadequateperfusion may be a result of decreased cardiac out-put (due to hypovolemia, structural defects in theheart, such as valvular heart disease, or arrhyth-mias), cerebrovascular atherosclerosis, or a neural-ly mediated decrease in heart rate, blood pressure,or both. Vasodilation decreases total peripheralvascular resistance, another major mechanism infainting that may occur in reflex syncope, hyper-thermia, and with certain medications.

Syncope may be caused by malfunctioning ofthe body’s normal protective mechanisms, auto-nomic nervous system responses, cardiovasculardisease, or outside influences (such as medicationsor blood loss from trauma). However, a variety ofconditions that range from epilepsy to metabolicstates, such as hypoglycemia, may at first appear tobe syncope. These conditions must be ruled out inthe diagnostic workup to distinguish true syncope.

ARRIVING AT A DIAGNOSIS: WHAT SYNCOPE IS AND WHAT IT IS NOT

The European Society of Cardiology’s TaskForce on Syncope has classified transient uncon-scious states (Figure 1). In addition, eliciting agood history from patients is essential to deter-mine whether true loss of consciousness occurredor the patient experienced near-syncope (presyn-cope), dizziness, vertigo, or a drop attack (a fallwithout loss of consciousness).8

Several studies have compared various investi-gations that aimed at diagnosing syncope.Theclear outcome of these meta-analyses has been thatpatient history and physical examination are key to

Figure 1. Real or Apparent Transient Loss of Consciousness

Reprinted with permission from Task Force on Syncope, European Society of Cardiology.Eur Heart J. 2001;22:1256-1306. Copyright © 2001 with permission from Elsevier.


SYNCOPE

determining whether syncope did occur and deter-mining the cause of documented syncopal events.1,2

The Eastbourne Syncope Assessment Study per-formed a retrospective analysis of 1334 cases codedas syncope in a hospital in the United Kingdom dur-ing 1998.1 Of these patients, it was established that660 patients (mean age, 64 years) had suffered anepisode of syncope, and 40% of those had experi-enced recurrent syncope. In 61% of cases, the diag-nosis could be made on the basis of a detailedhistory and physical examination alone. In theEastbourne protocol, physical examination includedlying and standing blood pressures, 12-lead electro-cardiogram (ECG), complete blood count, plasmaglucose, urea, and electrolyte studies. If this protocolfailed to determine a cause, investigators found that7-day external cardiac event recorders activated bythe patient and tilt testing were 5 times more likelyto diagnose syncope compared with Holter moni-toring; they were 50 times more likely to diagnosethe problem compared with echocardiography andcomputed axial tomography (CT) scan.Electroencephalography (EEG), chest radiography,electrophysiologic studies, and coronary angiogra-phy yielded extremely little information in patientswithout focal neurologic or cardiac signs followingthe taking of history and a physical examination.

Kapoor arrived at similar findings in a US pop-ulation of 433 patients with syncope.2 The resultsshowed that history, physical examination, andECG most frequently determined the cause of syn-cope, when a cause could be determined (cause wasundetermined 41% of the time). When these stepsalone did not make the cause of syncope evident,they frequently led clinicians toward a probablediagnosis and selection of more extensive noninva-sive and/or invasive diagnostic testing. For example,an abnormal ECG might direct a clinician towardevent monitoring, cardiac catheterization, radionu-clide studies, stress testing, and/or electrophysiolog-ic testing. Signs and symptoms suggestive of seizureactivity might indicate EEG or CT scanning, where-as a vasovagal presentation might result in head-upright tilt table testing.

In 1997, Linzer et al reviewed the literature thatcovered diagnostic testing for syncope.9 The investi-gators included studies conducted in at least 10patients with syncope or near-syncope beginning in1980. Again, this meta-analysis confirmed history,physical examination, and ECG to have the highestyield and to be the most cost effective.9

CLUES FROM THE HISTORY

Asking the right questions is crucial in evaluatingpatients with syncope (Table 1). First, the patient and

any witnesses should be asked to report what hap-pened. Sudden and complete loss of consciousnesswithout any warning or finding the patient in thesupine position suggests a cardiac cause, whereas thepresence of a prodrome may incline the cliniciantoward either a seizure (not syncope) or a vasovagalevent, depending on the accompanying prodromalsymptoms. If the patient reports feeling flushed,diaphoretic, nauseated, and/or weak and dizzy priorto loss of consciousness, or if the environment wassomehow emotionally or physically uncomfortable, avasovagal episode is the likely cause. In contrast, if theprodrome included an aura, the clinician should pro-ceed with a workup for seizure.

The patient should be asked questions regard-ing any precipitating factors. For example, a sud-den change of posture (from lying to standing)may cause syncope in persons with orthostatichypotension, whereas exercise-induced syncopemay occur in the presence of certain cardiac con-ditions. In elderly patients with carotid sinus syn-

Questions about circumstances immediately prior to attack– Position (supine, sitting, or standing)– Activity (rest, change in posture, during or after exercise, during or immedi-

ately after urination, defecation, cough, or swallowing)– Predisposing factors (eg, crowded or warm places, prolonged standing, postpran-

dial period) and precipitating events (eg, fear, intense pain, neck movements)

Questions about onset of attack– Nausea, vomiting, abdominal discomfort, feeling of cold, sweating, aura, pain

in neck or shoulders, blurred vision

Questions about attack (eyewitness)– Way of falling (slumping or keeling over), skin color (pallor, cyanosis, flush-

ing), duration of loss of consciousness, breathing pattern (snoring), move-ments (tonic, clonic, tonic-clonic or minimal myoclonus, automatism) andtheir duration, onset of movement in relation to fall, tongue biting

Questions about end of attack– Nausea, vomiting, sweating, feeling of cold, confusion, muscle aches, skin

color, injury, chest pain, palpitations, urinary or fecal incontinence

Questions about background– Family history of sudden death, congenital arrhythmogenic heart disease

or fainting– Previous cardiac disease– Neurological history (parkinsonism, epilepsy, narcolepsy)– Metabolic disorders (diabetes, etc)– Medication (antihypertensives, antianginal agents, antidepressants, antiar-

rhythmic agents, diuretics, and QT-interval–prolonging agents)– (In case of recurrent syncope) Information on recurrences, such as time

from first syncopal episode and the number of episodes

Table 1. Clues to the Etiology of Syncope from the PhysicalExamination

Reprinted with permission from Task Force on Syncope, European Society of Cardiology. Eur Heart J. 2001;22:1256-1306. Copyright © 2001 with permission from Elsevier.

268 Vol. 3, No. 5 ■ May 2003

drome, turning the head or performing a Valsalvamaneuver may cause syncope.

Next, a history of medication use should beobtained. Table 2 lists some drugs that may causesyncope and the mechanisms by which theyact.10,11 The patient should be asked about othermedical conditions he or she may have, such asdiabetes, epilepsy, or narcolepsy. Family history ofsudden death should also be determined, becausecertain genetically based cardiac anomalies maypresent with syncope or sudden death.

Finally, how an attack ends may be just as sig-nificant as how it begins. The patient or witnessesshould be questioned as to how the patientbehaved when regaining consciousness. Prolonged

confusion, sleepiness, or incontinence mandates aneurologic evaluation for seizures. Table 3 lists var-ious cardiac and noncardiac causes of syncope,along with common signs and symptoms.7,8

CLUES FROM THE PHYSICAL EXAMINATION

AND LABORATORY EVALUATION

A thorough cardiac examination (including 12-lead ECG) and neurological and psychiatric exami-nations add probably the most vital pieces to thesyncope puzzle that physical examination can pro-vide (Table 4). Detection of a heart murmur or gal-lop, abnormal rhythm, carotid bruit, focalneurologic sign, or an evaluation that mental status isimpaired will separate those patients requiring moreextensive cardiovascular or neurological testing. Inthese cases, echocardiography, stress testing, cardiaccatheterization, external or implantable devices forevent monitoring, and electrophysiologic testing arethe cardiac tests that may be required to rule out astructural anomaly, coronary artery disease (CAD),or arrhythmia. Conventional Holter monitoring isnot recommended because of the extremely low yieldand questionable accuracy obtained from such short-term recording.7 CT scan, magnetic resonance imag-ing (MRI) and/or EEG should be used only whenseizure disorder or stroke is suspected, because it isextremely rare for seizures or cerebrovascular attacksto present as syncope. Although carotid Dopplerultrasonography has been suggested, no studies con-firm its value, and carotid artery transient ischemicattacks (TIAs), like cerebrovascular attacks, are notassociated with loss of consciousness.7

Carotid sinus syndrome may best be evaluatedby performing carotid sinus massage in the appro-priate candidates. Head-upright tilt table testingwith or without administering medication to aug-ment the effects of gravity may achieve repro-ducible vasovagal syncope. Finally, standard bloodtests should be employed when a metabolic causeis suspected (eg, by obtaining plasma glucose lev-els in the case of possible diabetes, electrolytes fordehydration, or a complete blood count for hem-orrhage). Abnormal metabolic states more oftencause confusion, dizziness, or lightheadednessrather than true loss of consciousness.

CAUSES OF SYNCOPE

When the syncopal episode is transient or froma reversible cause, no further workup is necessary.However, once it has been established that anepisode is truly syncope, further evaluation is need-ed to differentiate the underlying etiology (Figure2), because more significant findings mandate refer-ral to a cardiac electrophysiologist (Table 5).

CARDIOLOGY

Agents that cause hypotension• Antihypertensive agents• Diuretics• Nitrates

Agents that influence cardiac output• Beta blocking agents• Digitalis

Agents that prolong QT interval• Antianginal agents

(Bepridil)• Antiarrhythmic agents

(Amiodarone, disopyramide, dofetilide, ibutilide, procainamide, quinidine, sotalol)

• Anesthetics/antiasthmatics(Adrenaline, epinephrine)

• Antibiotics(Azithromycin, clarithromycin, erythromycin, fluconazole, gatifloxacin, itraconazole, ketoconazole, moxifloxacin, sparfloxacin, trimethoprim)

• Antihistamines(Astemizole, clemastine fumarate, diphenhydramine hydrochloride)

• Antihyperlipidemics(Cholestyramine, colestipol hydrochloride)

• Anti-infectives(Pentamidine)

• Calcium channel blocking agents• Central nervous system drugs (eg, antidepressants

and antipsychotics)(Chlorpromazine, haloperidol, mesoridazine, tricyclic antidepressants, thioridazine)

• Toxins (eg, arsenic, organophosphate insecticides)• Miscellaneous

(Anabolic steroids, cocaine, ipecac, liquid protein diets, sumatriptan)

• Miscellaneous agents known to cause cardiac arrest/arrhythmias(Chinese herbs [eg, glycyrrhiza glabra, aconite alkaloids, ephedra alkaloids])

• Agents associated with altered sensorium(Alcohol, cocaine)

Table 2. Categories of Drugs that May Cause Syncope

Data from Goldschlager et al10 and Morag et al.11


The principle distinction that will guide man-agement of syncope is whether the cause of the syn-cope is cardiac or noncardiac. Although cardiacsyncope is more serious, because it includes organicheart disease (eg, aortic stenosis, pulmonary hyper-tension, pulmonary embolism, or myocardial infarc-tion [MI]) and arrhythmias, it is less commoncompared with noncardiac syncope. The cause ofsyncope was unknown in 34% of cases studied byLinzer et al; the “cause unknown” category was larg-er than the cardiac or noncardiac cause categories.9

NONCARDIAC CAUSES

Linzer et al reported that neurally mediatedsyncope (including vasovagal syncope, situationalsyncope, and carotid sinus syncope) is the mostcommon etiology.9 Other noncardiac causes ofsyncope include orthostatic hypotension, neuro-logical disease, medication-induced syncope, andpsychiatric disorders.

As with Linzer et al, Kapoor’s study of 433hospitalized patients failed to determine a cause inthe majority of cases (41%); noncardiac causes (ie,vasovagal, carotid sinus syndrome, situational, drug-related, orthostatic, and psychiatric) were the causein 33%, and the remaining 25% had cardiac condi-tions.2 The following sections discuss the most com-mon categories of syncope, including specificindications of any specialized testing, as well as man-agement for both noncardiac and cardiac causes ofsyncope. In general, EEGs, MRIs, and CT scansshould be reserved for patients with focal neurolog-ic signs and symptoms as opposed to those in whomsyncope is the more likely scenario.

NEURALLY MEDIATED SYNCOPE

Vasovagal Syncope. Vasovagal or neurocardio-genic syncope (the common fainting spell) is byfar the most common type of syncope, diagnosedin 40% of outpatient cases.5 It is also the mostenigmatic. Difficult to diagnose, vasovagal syn-cope often costs thousands of dollars per patientfor unnecessary diagnostic tests without the physi-cian being able to fully elucidate the specificunderlying mechanism or mechanisms. Excessivevenous pooling, which decreases venous return tothe right side of the heart, is often thought to ini-tiate the Bezold-Jarisch reflex (Figure 3), one pro-posed trigger for syncope. Instead of the normalphysiologic compensation that vasoconstriction andtachycardia would accomplish, there is a paradoxicalparasympathetic response and a withdrawal of sym-pathetic tone that lead to vasodilatation, bradycar-dia, hypotension, and loss of consciousness.5 Othertriggers may include serotonin, adenosine, opioids,

beta-endorphins, and emotional stress that elicitresponses from the higher central nervous system.5

Whatever the mechanism, vasovagal syncopegenerally does not confer great risk of mortality,12

although it may sometimes confer risk of injury fromfirst fainting and then falling. When vasovagal syn-cope occurs as a 1-time or rare event and doesn’tresult in damage, treatment is generally not necessary

SYNCOPE

Symptom or Finding Diagnosis to Consider

Episodes occur after sudden unexpected pain, Vasovagal attackfear, or unpleasant sight, sound, or smell

Episodes occur after prolonged standing at Vasovagal attackattention

Episodes occur in a well-trained athlete Vasovagal attackwithout heart disease after exertion

Episodes occur during or immediately after Situational syncopemicturition, cough, swallowing, or defecation

Syncope is accompanied by throat or facial Neurally mediated syncopepain (glossopharyngeal or trigeminal neuralgia) with neuralgia

Episodes occur with head rotation or pressure Carotid-sinus syncopeon carotid sinus (eg, due to tumors, shaving,or tight collars)

Episodes occur immediately on standing Orthostatic hypotension

Patient takes medications that may lead to a Drug-induced syncopelong QT interval or orthostasis and bradycardia

Syncope is associated with headaches Migraines, seizures

Syncope is associated with vertigo, Transient ischemic attack,dysarthria, or diplopia subclavian steal, basilar

migraine

Episodes occur with arm exercise Subclavian steal

Patient is confused after episode, or loss Seizureof consciousness lasts longer than 5 min

Differences are found in blood pressure Subclavian steal or aorticor pulse between the 2 arms dissection

Syncope and murmur occur with changes Atrial myxoma or thrombusof position (from sitting to lying,bending, turning over in bed)

Syncope occurs with exertion Aortic stenosis, pulmonary hypertension, mitral stenosis,hypertrophic cardiomyopathy,coronary artery disease

Patient has a family history of sudden death Long QT syndrome, Brugada syndrome

Patient has a brief loss of consciousness with Arrhythmiasno prodrome and has heart disease

Patient has frequent syncope with somatic Psychiatric illnesssymptoms but no heart disease

Table 3. Clinical Features Suggestive of Specific Causes of Syncope

Reprinted with permission from Kapoor. N Engl J Med. 2000;343:1856-1862. Copyright © 2000 Massachusetts Medical Society. All rights reserved.

270 Vol. 3, No. 5 ■ May 2003

beyond reassuring the patient and educating him orher to avoid potential triggers or circumstances thatmay cause a recurrence (Sidebar, page 273).

Treatment with pharmacotherapy and/or animplantable pacemaker must be carefully individ-ualized and monitored. Although a variety ofmedications, including anticholinergic agents(scopolamine), centrally acting agents (serotoner-gic reuptake inhibitors), vasoconstrictors (alpha-adrenergic agonists), and negative inotropic agents(disopyramide) have been tried in the treatment ofsyncope, pharmacotherapy is often unnecessaryand may be ineffective.5,7 Furthermore, if carotidsinus syndrome is the cause rather than vasovagalsyncope, use of beta blocking agents may exacer-bate the problem by causing bradycardia. The useof the antiarrhythmic medication disopyramide also

may be dangerous in vasovagal syncope, because itmay precipitate life-threatening ventricular tach-yarrhythmias, urinary obstruction, or glaucoma.10

Cardiac pacing may be advisable if the patient hasrecurrent attacks or has experienced severe physicalinjury from an episode, and if head-upright tilt tabletesting has documented severe bradycardia or asys-tole.7 Permanent dual-chamber pacemakers designedto detect sudden falls in cardiac rate may then beimplanted. In patients with significant bradycardia asa cause of neurocardiogenic syncope, these devicesmay convert sudden syncope without prodrome intoa more gradual, symptomatic event for which thepatient can take anticipatory measures.10 It is impor-tant to rule out cardiac causes, especially in an olderpatient, but clinicians should also remember thatvasovagal syncope can present for the first time at an

CARDIOLOGY

Physical Examination

Finding Possible CausesOrthostatic hypotension Dehydration, medications (eg, antihypertensives, diuretics)

Ventricular asystole and/or hypotension during carotid Carotid sinus hypersensitivitysinus massage

Displaced or abnormal left ventricular impulse, S3 gallop, murmurs Ventricular tachycardiaof mitral regurgitation suggesting myocardial dysfunction

Sustained apical impulse, S4 gallop indicating left Need for atrioventricular synchrony to preserve optimal ventricular hypertrophy cardiac output

Ejection murmurs (aortic stenosis, hypertrophic Left ventricular outflow tract obstructionobstructive cardiomyopathy)

Right ventricular lift, loud P2, prominent A-wave in jugular Left ventricular underfilling, with reduced stroke volumevenous pulse, right ventricular S2 and S4 gallops suggesting pulmonary hypertension

12-Lead Electrocardiography

Finding Possible CausesLong QT interval Torsade de pointes ventricular tachycardia

Previous myocardial infarction Ventricular tachycardia

Atrioventricular conduction delay or block, Bradycardiabundle branch block, fascicle block

Ventricular hypertrophy Need for atrioventricular synchrony

Epsilon wave Arrhythmogenic right ventricular dysplasia

Right ventricular conduction delay with ST-segment Brugada syndromeelevation in precordial leads, typically V1 through V3

Atrial and ventricular ectopy Nonspecific; may indicate presence of a substratefor more sustained rhythm disturbance

Bradycardia Nonspecific; may indicate conduction system disease

Ventricular preexcitation Wolff-Parkinson-White syndrome

Table 4. Clues to the Etiology of Syncope from Physical Examination and 12-Lead Electrocardiography

Reprinted with permission from Goldschlager et al. Arch Intern Med. 2003;163:151-158.


older age. The scenario that precipitated the faintingspell may differ from that in a younger person, butthe diagnosis may be the same.

Tilt table testing now offers a fairly inexpensive,safe, and noninvasive method for diagnosing vaso-vagal syncope. Patients undergo constant ECG andblood pressure monitoring as the technician shiftsthem from the supine position (after 10 to 20 min-utes) to an angle of 80° that is maintained for a peri-od of approximately 30 to 45 minutes or untilsymptoms are reproduced. To enhance response,various challenging medications, such as isopro-terenol or sublingual nitroglycerin, may also be usedif the initial phase of the test is negative. Sensitivityand specificity are reported to range between 80% to90%, on average, for the reproduction of symptoms.A positive test is characterized by a sudden drop inblood pressure, pulse, and the occurrence of syn-cope.13 Tilt table testing should be performed todocument the response in individuals with recurrentsyncope that appears to be vasovagal, and in thosewho have experienced or may experience syncope ina high-risk situation. It should not be performed inpatients believed to have obstructive cardiac disease.5

Guidelines for the use of tilt table testing have beenissued by the American College of Cardiology andthe North American Society of Pacing andElectrophysiology.14

Carotid Sinus Syndrome. Hypersensitivity ofthe carotid sinus is another potential cause of neu-rally mediated syncope. It is responsible for 10%to 20% of presyncopal and syncopal spells, espe-cially in elderly patients with atherosclerosis.11

This syndrome may be diagnosed by performingcarotid sinus massage. The patient is attached to abeat-to-beat blood pressure device and cardiacmonitor. Massage is applied to the area of themaximal carotid pulse on one side of the neck at thelevel of the thyroid cartilage for 3 to 5 seconds. Theclinician is looking for either a 3-second interval ofasystole (cardioinhibitory response), a ≥50-mm Hgdecrease in blood pressure (vasodepressor response),both parameters, and/or syncope for a confirmeddiagnosis. The main risk associated with this proce-dure is the possibility of precipitating a TIA or cere-brovascular attack in individuals with carotid arterydisease. However, the complication rate for carotidsinus massage is extremely low (<1%). Carotid sinusmassage is generally avoided in patients with a pre-vious history of cerebrovascular attack, TIA, carotidendarterectomy, and/or in those with carotid bruits.

Patients with carotid sinus syndrome shouldavoid pressure to the carotid sinus, such as thatwhich can occur with head turning, tight collars,and Valsalva maneuvers. Medications that may

SYNCOPE

Figure 2. Approach to the Patient with Syncope

*Syncope after recent head trauma, seizures, headache, or new neurologic abnormality: referto neurologist. †Atrioventricular block, ventricular tachycardia, supraventricular tachyarrhyth-mias including atrial fibrillation and flutter, sinus bradycardia, myocardial infarction, ectopy, QTinterval 0.50 sec or longer (receiving/not receiving QT-prolonging drugs), arrhythmogenicright ventricular dysplasia, hypertrophic cardiomyopathy, Wolff-Parkinson-White syndrome,and Brugada syndrome.ECG = electrocardiogram; BBB = bundle branch block; LVEF = left ventricular ejection frac-tion; RX = treatment as appropriate.Adapted with permission from Goldschlager et al. Arch Intern Med. 2003;163:151-158.

1. Neurocardiogenic (including vasovagal) syncope, especially if refractory toavoidance of triggers and drug therapy or associated with prolonged pauses incardiac rhythm

2. Arrhythmia identified during evaluation, including ventricular tachycardia due toany cause; bradyarrhythmia, such as sinus bradycardia or atrioventricular block notclearly caused by drug therapy that can be withheld or changed; supraventriculartachycardia, especially in the presence of Wolff-Parkinson-White conduction

3. Congenital long QT syndrome diagnosed or suspected from the electrocardiogram

4. Brugada syndrome diagnosed or suspected from the electrocardiogram

5. Structural heart disease, such as dilated or hypertrophic cardiomyopathy; sus-pected arrhythmogenic right ventricular dysplasia; or coronary artery anomalyidentified during evaluation

6. Syncope in athletes

7. Syncope during exercise

8. Origin of syncope remains unknown and prolonged arrhythmia monitoring bymeans of an implantable loop recorder is being considered

Table 5. Indications to Refer to a Cardiac Electrophysiologist

Reprinted with permission from Goldschlager et al. Arch Intern Med. 2003;163:151-158.

272 Vol. 3, No. 5 ■ May 2003

precipitate bradycardia and/or hypotensionshould also be avoided.10,11 Treatment for this syn-drome is the implantation of a cardiac pacemakerto manage the bradycardia when a cardioinhibito-ry rather than a vasodepressor component is doc-umented to be primary.10,11

Kenny et al randomized 175 patients olderthan 50 years with cardioinhibitory carotid sinushypersensitivity who had experienced nonacci-dental falls (with or without syncope) to 2groups in the Syncope and Falls in the Elderly:Pacing and Carotid Sinus Evaluation (SAFEPACE) trial.15 One group had dual-chamberpacemakers implanted (n = 87), and the controlgroup (n = 88) received standard therapy. Pacedsubjects experienced significantly fewer falls andsyncopal events compared with control subjects(odds ratio, 0.42; 95% confidence interval [CI],0.23–0.75). Falls were reduced by two thirdsand injuries by 70%, and syncopal events werenearly one half as common (n = 28 in pacedpatients; n = 47 in controls).

Situational Syncope. Like vasovagal syncope, situ-ational syncope is vagally mediated. However, in sit-uational syncope, which is seen most often in theelderly, a specific situation consistently triggers a lossof consciousness. Common triggers include cough,swallowing, shaving, micturition, and defecation.Wherever possible, treatment is aimed at avoiding or

lessening exposure to the situation. For example,patients with chronic obstructive pulmonary diseaseor asthma may suffer from cough syncope. Coughsuppression and smoking cessation may be benefi-cial. Advising individuals to sit during and for a fewminutes after urinating and defecating may also helpto prevent loss of consciousness.

OTHER NONCARDIAC CAUSES

Orthostatic Hypotension. Orthostatic hypoten-sion is defined as a drop in systolic blood pressureof ≥20 mm Hg, or a decrease of systolic bloodpressure to <90 mm Hg, after the patient hasbeen standing for 1 to 3 minutes following aperiod of having been supine for 5 minutes. Itmay also be indicated by a decline in diastolicblood pressure of at least 10 mm Hg after 3 min-utes, or by a slower, steadier drop over a longerperiod (10 to 15 minutes).10

A variety of acute and chronic etiologies havebeen implicated in orthostatic hypotension. Volumedepletion, such as that which occurs with trauma,gastrointestinal bleeding, or dehydration, may causeorthostatic changes. Replacement of fluids and elec-trolytes will reverse orthostatic hypotension in thesesituations. In more chronic cases and especially inthe elderly, failure of the autonomic system may bedue to medications the patient is taking (Table 2).Diuretics, vasodilators, and alcohol are among themost common offenders. Whenever possible, thesesubstances should be discontinued. Compressionhosiery may also be of some use in patients who canwear them without discomfort.4,7 Patients withorthostatic hypotension should be counseled tochange positions slowly (eg, when getting out of bedfor the first time in the morning). These patientsshould also be cautioned to avoid environmentswith high temperatures, such as saunas or Jacuzzis,and to make lifestyle changes that include changingthe water temperature for baths and showers fromhot to warm and avoiding heavy meals, excessiveexertion, and alcohol. Restricting intake of salt andincreasing fluids may also be helpful in selectedpatients. For example, salt intake may be detrimen-tal in elderly patients, particularly those with hyper-tension or heart failure.

Autonomic Nervous System Failure. When thereis symptomatic evidence of a failure of the auto-nomic nervous system (ANS), it is sometimes nec-essary to consider a neurological evaluation ofsyncope. Orthostatic hypotension was discussedpreviously. Other situations result from seriousdegenerative diseases that cause primary failure ofthe ANS. The main etiologies are Parkinson’s diseaseor atrophy of multiple systems (formerly including

CARDIOLOGY

Figure 3. Autonomic Nervous System Regulation ofCardiovascular Hemodynamic Responses

Reprinted with permission from Fenton et al. Ann Intern Med. 2000;133:714-725.

Midbrain MedullaNucleus tractus solitariusNucleus ambiguusVentrolateral medulla

Higher CenterHypothalamusAmygdalaParaventricular cortexInsular cortexAnterior cingulate cortex Mechanoreceptors and

BaroreceptorsCarotid sinusAortic archHeart Lung (cardiopulmonary)

Spinal cordIntermediolateral cell column

Parasympathetic efferents(vagus)

Sympatheticefferents

Vascular tone

Heart Rate

(Glossopharyngeal)

(Vagus)Afferents


Shy-Drager syndrome, striatonigral degeneration,and olivopontocerebellar degeneration). This is aprogressive disorder characterized by orthostatichypotension along with other symptoms of ANSfailure, such as ataxia and parkinsonism.

Failure of the ANS may also be secondary tocertain illnesses (eg, diabetes or liver or kidney fail-ure) or due to certain drugs (monoamine oxidaseinhibitors and tricyclic antidepressants, phenothi-azines, antihistamines, or levodopa) or alcoholuse. Although a neurological workup is not neces-sary in all cases of syncope, it is imperative in thesyncopal patient who is taking these medicationsor demonstrates signs and symptoms of autonom-ic failure (eg, difficulty urinating, impotence), dia-betes, or Parkinson’s disease.

Cerebrovascular Events. Vascular or subclaviansteal syndrome is a rare cerebrovascular cause of syn-cope that occurs when the circulation to the arm isblocked due to stenosis or occlusion of the subcla-vian artery. Blood flow is essentially “stolen” by thevertebral artery. If there are excessive demands onthe extremity (eg, during exercise), there may beinsufficient perfusion to the brain stem, resulting inloss of consciousness. The diagnosis, which is mostcommon in middle-aged men, is suggested by thecircumstances surrounding the symptoms, and canbe confirmed via Doppler ultrasonography. Onphysical examination, there is a discrepancy in bloodpressure between the upper and lower extremities.Treatment is the use of revascularization techniques.

Some types of basilar artery or complicatedmigraine headaches may have syncope as a compo-nent, although the patient’s loss of consciousnessmay last for too long to be considered syncope, andthe constellation of other symptoms may help theclinician differentiate migraine from syncope. Thesemigraines generally occur in young adults.

TIAs and cerebrovascular attacks rarely causesyncope as a primary symptom. Both are muchmore likely to present with complete or partialweakness or paralysis, slurred speech, abnormaleye movements or vision disturbances, alteredmental status, and severe headache. It is generallynot necessary to obtain expensive or invasive radi-ologic testing to rule out a TIA or cerebrovascularattack when syncope occurs.

Epilepsy. Syncope has sometimes been misdiag-nosed as epilepsy because both involve loss of con-sciousness. Because cardiovascular causes of syncopemay result in cerebral hypoperfusion and seizure-like movements, these syncopal events may be par-ticularly difficult to differentiate from true epilepsy.Zaidi et al investigated a population of 74 subjectswho had been given a presumptive diagnosis of

epilepsy, but who had presented with atypicalseizures or convulsions that were resistant to treat-ment.16 All of the study participants had experiencedmultiple episodes of seizures over a prolonged peri-od of time (median, 25 incidents over 60 months).Subjects underwent 80° head-upright tilt table test-ing and carotid sinus massage with simultaneousEEG, ECG, and blood pressure monitoring. Aselect group of 10 patients in whom a cardiac causewas strongly suspected also had loop recordersimplanted for long-term continuous ECG monitor-ing. Nineteen patients (25.7%) had positive head-upright tilt table tests, 7 patients (9.5%) hadsignificant pauses with carotid sinus massage, and 2patients had prolonged bradycardia (diagnosed vialoop recorder). Many of these patients had seizure-like activity during the course of their syncopalepisodes, but their EEGs were normal. Two subjectswere determined to have psychogenic seizures. Ofthe 74 subjects in this study, 31 (41.9%) were foundto have convulsive syncope from vasovagal, cardiac,or psychogenic causes rather than from epilepsy.

A careful history taken from the patient and eye-witnesses to the event is instrumental in arriving at aproper diagnosis. Clinicians should determinewhether the individual experienced aura, whether themovements were tonic-clonic in nature, and whetherthere was a postictal state marked by sleepiness orconfusion. Although movements may occur duringsyncope, they are generally not synchronous and theyoccur after the person falls down. Such movementscontrast with what occurs in epilepsy, in which the

SYNCOPE

Counseling Patients for Avoidance of Neurally Mediated Syncope

• Avoid environments that may trigger a fainting spell, such as hot, crowded roomsor the need for prolonged standing.

• Avoid stressful events that predispose some individuals to fainting, such asvenipuncture or other exposure to blood, pain, fear, or highly emotional settings.

• Avoid dehydration. Be certain that proper attire, ventilation, and adequate flu-ids and salt intake are available during exertion, as with sports participation.

• Be aware of and attempt to minimize situations that trigger syncope, such ascoughing or wearing tight collars.

• Discontinue medications, such as vasodilators, that increase susceptibility tofainting spells.

• Be aware of warning signs, such as feeling nauseated, sweaty, dizzy, or light-headed, and sit or lie down to attempt to prevent loss of consciousness.

• Your healthcare provider may prescribe salt tablets or medications, such as fludro-cortisone, midodrine, propranolol or other beta blocking agents, or anticholinergicagents for severe or recurrent episodes. He or she may also refer you to an elec-trophysiologist who may recommend implantation of a pacemaker in certain cir-cumstances after the appropriate diagnostic tests have been performed.

274 Vol. 3, No. 5 ■ May 2003

forceful and synchronous movements may precedecollapse and may continue while the person is on theground. Persons experiencing syncope will beupright prior to losing consciousness (except in thecase of a cardiac arrhythmia, which may occur inother positions), and they are limp in muscle tone.By contrast, a seizure may occur in any position, andmuscle tone is rigid. Incontinence and tongue bitingalso suggest seizure activity rather than syncope.However, because it is sometimes difficult for a layperson to differentiate a seizure from a syncopalepisode (Table 3), patients with seizures that do notseem to respond to adequate doses of pharmacother-apy should be evaluated to rule out syncope (partic-ularly cardiac syncope) as the true etiology.

Psychogenic Episodes. Certain psychiatric condi-tions, including anxiety, depression, and panicattacks, may cause syncope-like symptoms, althoughthere may not be definite loss of consciousness.Psychogenic seizures (pseudoepilepsy), hyperventila-tion syndrome, cataplexy (marked by sudden loss ofmuscle control [drop attacks] triggered by laughteror another emotion), narcolepsy, and somatizationdisorder may all need to be considered in differenti-ating syncope from other similar conditions. A psy-chiatric evaluation and the appropriate treatmentshould then be initiated when there are symptomssuggestive of one of these etiologies.

CARDIAC CAUSES

In 1990, Kapoor determined that, in his popu-lation of 433 patients with syncope, those with acardiac cause had a much poorer prognosis thanthose in whom a noncardiac cause was found. After1 year, patients with a cardiac cause had an 18% to33% mortality rate compared with 0% to 12% inthose with noncardiac causes and 6% in patientswith syncope of undetermined etiology. At 5 years,patients with a cardiac cause had a mortality rate of50.5% compared with 30% in patients with non-cardiac causes and 24.1% in people for whom acause of syncope could not be determined (P <.001). Cox regression analyses were performedbecause patients with a cardiac cause of syncopewere older and tended to have preexisting cardiacdisorders and risk factors. Despite these factors, theresults of these analyses supported the fact that acardiac cause of syncope was an independent pre-dictor of mortality, especially sudden death.2

Although this study is somewhat dated, and todaywe have more sophisticated tools at our disposal(such as electrophysiologic testing, event monitor-ing, implantable loop recorders, and more advancedechocardiography), it stresses the vital importance ofexcluding a cardiac cause of syncope.

A study published by Soteriades et al in 2002evaluated the incidence, specific causes, and prog-nosis of syncope among the 7814 men andwomen who participated in the FraminghamHeart Study between 1971 and 1998.12 The find-ings suggested that individuals found to have acardiac cause for their syncope were a high-riskgroup, predisposed to greater morbidity and mor-tality. After adjustment for other variables, thehazard ratios related to syncope from any causewere 1.31 (95% CI, 1.14 to 1.51) for death fromany cause, 1.27 (95% CI, 0.99 to 1.64) for MI ordeath from CAD, and 1.06 (95% CI, 0.77 to1.45) for stroke. Corresponding hazard ratios forcardiac syncope were 2.01 (95% CI, 1.48 to 2.73),2.66 (95% CI, 1.69 to 4.19) for MI/CAD, and2.01 (95% CI, 1.06 to 3.80) for stroke. Overallprobability of survival of participants with syn-cope from cardiac causes was lower than that ofparticipants without syncope (P < .001).12 Thesesignificant findings show that it behooves the clin-ician to carefully evaluate patients for cardiac caus-es of syncope, treat them, and monitor themcarefully to avert morbidity and mortality.

ARRHYTHMIAS

Arrhythmias are a primary cardiac cause ofsyncope. One potential mechanism (eg, in heartblock) is the slow transmission of impulsesthrough the heart. This eventually causes inade-quate cerebral perfusion, thus causing syncope orpresyncope.7 These abnormal cardiac rhythmsmay arise as a result of structural defects in theheart, impeding the proper conduction of impuls-es. Some of these arrhythmias include ventriculartachycardia, supraventricular tachycardia, sinusnode disease (manifested by sinus bradycardia orlong sinus pauses), and heart block (complete atrio-ventricular block or Mobitz II). Electrophysiologicstudies are generally the means of diagnosingarrhythmias in syncope of unclear etiology,although fortuitous timing via an ECG or Holtermonitor may reveal these arrhythmias. For personswho have abnormally slow heart rates, permanentpacing is indicated and effective for symptoms.7

Other arrhythmias, such as torsade de pointes(a polymorphic ventricular tachycardia associatedwith a prolonged QT interval or prominent Uwave on ECG), may be congenital, but are fre-quently the result of certain medications.17

Discontinuation of the suspect medication willcure the syncope. Ironically, some antiarrhythmicagents will cause torsade de pointes, as will certainpsychoactive medications, antimicrobial drugs,and nonsedating antihistamines (Table 2).7,10,17

CARDIOLOGY


Concomitant ingestion of grapefruit juice withthese medications may interfere with thecytochrome P450-3A4 enzyme system, causingelevated levels of the drugs and increasing the riskof prolonged QT interval. Chinese herbs (gly-cyrrhiza glabra), aconite and ephedra alkaloids,and anabolic steroids in large doses may also pre-cipitate ventricular arrhythmias and syncope.10

STRUCTURAL CARDIAC AND

CARDIOPULMONARY DISEASE

Syncope also may occur in the midst of cardiacarrest secondary to MI or CAD. Less commoncauses of cardiac syncope include aortic stenosis,aortic dissection, pericardial tamponade, hyper-trophic obstructive cardiomyopathy, atrial myxo-ma, mitral stenosis, pulmonary stenosis,pulmonary embolism, pulmonary hypertension,and permanent pacemaker malfunction.4 Syncopein these cases may be arrhythmia associated, medi-ated by neurocardiac reflexes, or due to inadequateblood flow from a mechanical obstruction.

SUMMARY AND TREATMENT

Patients who have cardiac as opposed to noncar-diac syncope generally have little prodrome.Whereas vasovagal syncope may be postural, andsyncope secondary to a mechanical defect in theheart may be exertional or postexertional, arrhyth-mia-induced syncope is usually neither postural norexertional, and comes on with very little warning.There is a sudden drop in cardiac output and cere-bral blood flow in both types of cardiac syncope —arrhythmia based and structurally based.Treatmentof the underlying arrhythmia, or of the structuralcardiac or cardiopulmonary disease, is the goal inpreventing recurrence of syncope in these patients.This may be accomplished via pharmacotherapy,surgery, or both, according to underlying etiology.Specific details of these treatments are beyond thescope of this article. The patient should be referredto a cardiologist for specific diagnostic testing andmanagement (Table 5). Hospitalization may also benecessary in some instances.

SPECIAL SITUATIONS: THE ATHLETE

One in 250 000 athletes, including 25 adoles-cents, die annually while engaging in exercise,including competitive sports. Syncope in athletesshould therefore be carefully evaluated. It may bebenign but may also signal serious — potentiallyfatal — cardiac pathology (Table 6). Risk variesaccording to age, sex, intensity of the activity, pre-vious syncope, and/or known cardiac disease.10

The nature and intensity of training can have

a physiologic effect. Sympathetic activity and thebaroreceptor reflex can be blunted in the athlete.Aerobic exercise increases oxygen uptake, heartrate, cardiac output, and vagal tone. Isometricexercise (eg, weight training) can increase bloodpressure to very elevated levels (>350/150 mmHg).10 The clinician should consider a potentiallylethal arrhythmia in any athlete who has anepisode of syncope that is both quick in onset (noprodrome) and in recovery. A structural anomaly ofthe heart (aortic stenosis) or of a coronary bloodvessel may be present. Clinicians should considerthese possibilities if an athlete collapses shortly afterbeginning physical activity or with prolonged activ-ity, although ischemia presenting as syncope inresponse to exercise is unusual. Because athletessometimes develop physiologic left ventricularhypertrophy, it may be necessary to restrict exercisefor several months in order to attempt to differenti-ate the “athletic heart,” which will regress, fromhypertrophic cardiomyopathy, which will not.10

Bikers and runners in particular may suffer fromnoncardiac causes of syncope, including dehydra-tion, hypoglycemia, or a sudden fall in blood pres-sure at the end of their exercise session. Athletes mayhave vasovagal episodes from neurally mediatedhypotension and bradycardia. Hyperventilation orValsalva maneuvers induced in the process of exer-cising may also lead to syncope or presyncope.Supplements (eg, ephedra [mahuang], androgenicsteroids, creatine) may cause potentially fatalarrhythmias in athletes who use these substances,which may be obtained over the counter. Both crea-tine and ephedra have been implicated in suddendeaths in athletes.10 In light of more than 100 deathslinked to the use of ephedra, the US Food and DrugAdministration (FDA) has issued a warning regard-

SYNCOPE

• High catechol levels

• Exposure to excessive heat or cold

• Hormonal shifts

• Mental stress

• Dehydration

• Exercise type (aerobic vs isometric)

• Hyperventilation

• Valsalva maneuvers (weight lifting)

• Substance abuse (cocaine, amphetamines)

• Supplements (steroids, ephedra, creatine)

Table 6. Factors to Consider in the Evaluationof Syncope in the Athlete

276 Vol. 3, No. 5 ■ May 2003

ing its use as a weight-loss aid and as an enhancer ofsports performance. The FDA is also considering anoutright ban on ephedra or requiring a warninglabel on the package.18

The workup of an athlete with syncope is bestleft to the cardiologist who can diagnose and treatthe underlying etiology, including making a deter-mination of prognosis and whether restriction orcurtailing of athletics is indicated. Diagnostic test-ing generally includes ECG and echocardiography.Treadmill testing, Holter monitoring, and externalloop recorders have a low diagnostic yield, as doeselectrophysiologic testing in athletes without docu-mented heart disease. Tilt table testing, cardiaccatheterization, or myocardial biopsy may be indi-cated in specific instances determined by the cardi-ologist. Selection of the appropriate diagnostic testshinges on a complete and accurate record of the cir-cumstances that cause or caused the syncope.

As with all syncopal situations, appropriate treat-ment is determined by underlying etiology, and mayinvolve surgery or pharmacotherapy. Lifestyle mod-ifications may also be part of the treatment plan,such as curtailing use of dangerous supplements.Whether these lifestyle modifications include rec-ommendations for restricted participation in sportsdepends on the nature of the sport, the presence ofdocumented cardiac disease (and whether it can becontrolled or cured with treatment), as well as the preferences of the individual athlete.Arrhythmogenic right ventricular dysplasia, torsadede pointes, and hypertrophic obstructive cardiomy-opathy all mandate restriction or avoidance of exer-cise. Ablation or an implantable defibrillator may becorrective for other anomalies and/or arrhythmias.10

CONCLUSION

Although differentiating the cause of syncopein a patient challenges the clinician, it is essentialfor reducing morbidity and mortality. Because thescope of etiologies crosses many systems and disci-plines (eg, cardiac, neurological, psychiatric), thepatient may be subject to lengthy, costly, and inva-sive diagnostic workups and hospitalizations. Onthe other hand, the price of not diagnosing thecause of syncope may be unacceptably high, if thepatient has a serious consequence (eg, fall, motorvehicle accident, or even death) secondary tomissed or improper diagnosis. Fortunately, evi-dence-based studies confirm that data obtainedfrom a detailed history, physical examination, andcarefully selected diagnostic testing (basic thoughthese tests may be) are frequently successful indetermining etiology and appropriate treatment.

REFERENCES

1. Farwell D, Sulke N. How do we diagnose syn-cope? J Cardiovasc Electrophysiol. 2002;13(suppl):S9-S13.

2. Kapoor WN. Evaluation and outcome of patientswith syncope. Medicine. 1990;69:160-175.

3. Savage DD, Corwin L, McGee DL, Kannel WB, WolfPA. Epidemiologic features of isolated syncope: theFramingham Study. Stroke. 1985; 16:626-629.

4. Chang-Sing P, Peter CT. Syncope: evaluation andmanagement. Cardiol Clin. 1991;9:641-651.

5. Fenton AM, Hammill SC, Rea RF, Low PA, Shen W.Vasovagal syncope. Ann Intern Med. 2000;133:714-725.

6. Hansen M. Pathophysiology: Foundations ofDisease and Clinical Intervention. Philadelphia: WB Saunders Co; 1998:342-343.

7. Task Force on Syncope, European Society ofCardiology. Guidelines on management (diagnosisand treatment) of syncope. Eur Heart J. 2001;22:1256-1306.

8. Kapoor WN. Syncope. N Engl J Med. 2000;343:1856-1862.

9. Linzer M, Yang EH, Estes M, Wang P, VorperianVR, Kapoor WN. Diagnosing syncope. Part 1.Value of history, physical examination, and electrocar-diography. Ann Intern Med. 1997;126:989-996.

10. Goldschlager N, Epstein AE, Grubb BP, et al.Etiologic considerations in the patient with syncopeand an apparently normal heart. Arch Intern Med.2003;163:151-158.

11. Morag R, Brenner B. Syncope [eMedicine Web site].Available at: http://www.emedicine.com/emerg/topic876.htm#. Accessed February 9, 2003.

12. Soteriades ES, Evans JC, Larson MG, et al.Incidence and prognosis of syncope. N Engl JMed. 2002;347:878-885.

13. Grubb BP, Kimmel S. Head-upright tilt table testing.Postgrad Med. 1998;103:133-138.

14. Benditt D, Ferguson D, Blair P, et al. ACC expertconsensus document: tilt table testing for assessingsyncope. J Am Coll Cardiol. 1996;28:263-275.

15. Kenny RA, Richardson DA, Steen N, Bexton RS, ShawFE, Bond J. Carotid sinus syndrome: a modifiable riskfactor for nonaccidental falls in older adults (SAFEPACE). J Am Coll Cardiol. 2001;38:1491-1496.

16. Zaidi A, Clough P, Cooper P, Scheepers B,Fitzpatrick AP. Misdiagnosis of epilepsy: manyseizure-like attacks have a cardiovascular cause. J Am Coll Cardiol. 2000;36:181-184.

17. Center for Education and Research on Therapeutics.Drugs with risk of torsade de pointes [University ofArizona Center for Education on Research andTherapeutics Web site]. Available at:http://www.torsades.org/medical-pros/drug-lists/drug-lists.htm. Accessed April 17, 2003.

18. Smith S. FDA proposes warning label for ephedra.Boston Globe. March 1, 2003:A1.

CARDIOLOGY

approach to the patient with syncope: c history and...

Documents