CLINICAL GUIDELINE: Diagnosing Syncope: Part 1: Value of History, Physical Examination, and Electrocardiography

Syncope is a transient loss of consciousness that is accompanied by loss of postural tone. It is common [1] and can be dangerous [2], disabling [3], and difficult to diagnose [4]. Thousands of dollars can be spent evaluating a patient with syncope, only to result in a series of negative test results and a patient who continues to faint. Because the range of prognoses in syncope is wide, the physician's principal initial task is to distinguish between benign and life-threatening causes of syncope. We intend primarily to help clinicians maximize the diagnostic yield in the workup of syncope. Our secondary purpose is to summarize the literature that will aid clinicians in assessing risk to enable them to target hospitalization and invasive testing for the patient with syncope who is at high risk for an adverse outcome. The questions addressed by this two-part study are 1) Which diagnostic techniques are the most valuable for patients with syncope? 2) How can the clinical history help focus the workup for patients with syncope? and 3) When should patients with syncope be hospitalized? Methods We used the MEDLINE database to identify articles related to syncope and diagnostic testing. References that evaluated the diagnostic test in near syncope and dizziness were included if they also used the test in patients with syncope. When a medical subject heading did not identify a sufficient number of references about a particular diagnostic test (such as neurovascular testing or carotid Doppler ultrasonography), keyword searches (using such terms as transcranial Doppler) were done. To be included in the review, articles had to be published in the English language between 1980 and 1995. The studies had to be randomized trials, observational studies, cohort studies, or case series of more than 10 patients (review articles and case reports were excluded); had to focus on or include patients with syncope; and had to examine only patients 18 years of age or older (except for tilt-table studies, which often included adult and pediatric cases in the same articles). Articles that were candidates for review were evaluated in detail by one of the authors. Articles that met the selection criteria were used to prepare summary tables or paragraphs. Comparisons between groups (for example, the proportion of patients with and without heart disease who had tachyarrhythmias diagnosed by electrophysiologic testing) were made using the Fisher exact test. Selected national experts in cardiology and neurology were asked to review the findings in their area of expertise. The opinions of these experts were incorporated into the recommendations. Limitations of the Literature on Syncope In syncope, there is no diagnostic gold standard against which other diagnostic tests may be measured; thus, sensitivity and specificity may not be easily calculated. Moreover, the presence of a disease, such as coronary disease, in a patient who has fainted does not prove that the disease caused the syncope. Syncope is, at its core, a symptom and not a disease. Therefore, this review is not organized around a technology or a disease entity but focuses on the physiologic states that lead to a sudden, transient loss of consciousness. The literature that discusses syncope predominantly comprises case series or cohort studies based on referrals to tertiary care centers. We classified studies into three types: population-based studies (including unselected patients from the general population who were hospitalized or seen in emergency departments and other outpatient settings), referral-based studies (including patients referred to specialized centers for syncope workups), and small case series. To our knowledge, no randomized trials of the diagnostic workup or management strategies for patients with syncope have been done. A summary of the types of studies conducted in patients with syncope (Table 1) shows that most have been referral studies or case series. Table 1. Sample Characteristics of Studies of Diagnostic Tests and Syncope Definitions Organic Heart Disease Whenever possible, our definition of organic heart disease included coronary artery disease, congestive heart failure, valvular heart disease, cardiomyopathy, and congenital heart disease. Because conduction system disease is a separate predictor of the need for special diagnostic testing, it was kept apart except where indicated. Patients who had a history and physical examination that were negative for cardiovascular symptoms or signs and a normal electrocardiogram were considered to have normal hearts; however, we recognize that some investigators think that echocardiography should be done before patients are declared free of organic heart disease. Diagnostic Yield For most tests, the diagnostic yield reflects the number of patients with positive diagnostic test results divided by the number of tested patients. For Holter and loop monitoring, the numerator includes the sum of the true-positive test results (arrhythmias during fainting) plus the true-negative test results (normal rhythm during symptoms). This expanded definition reflects the prognostic importance of a negative result on electrocardiography during syncope. For certain tests, the absolute value of the diagnostic yield may not be as important as the ability of the test to exclude a serious diagnosis (for example, intracardiac electrophysiologic studies may be of considerable benefit when they exclude ventricular tachycardia in a patient in whom that diagnosis was strongly considered). Data Synthesis Differential Diagnosis The first category of syncope is neurally mediated syncope, which results from reflex mechanisms that are associated with inappropriate vasodilatation, bradycardia, or both (Table 2). This category includes vasovagal, vasodepressor, situational, and carotid sinus syncope. Neurocardiogenic mechanisms are also implicated in syncope associated with ventricular outflow obstruction (such as with aortic stenosis and pulmonary embolism) as well as supraventricular tachyarrhythmias [5-9]. The second category is orthostatic hypotension, which may result from age-related physiologic changes, volume depletion, medication, and autonomic insufficiency [10, 11]. Psychiatric disorders related to syncope (such as anxiety, depression, and conversion disorders) form a third category. The fourth category includes neurologic disorders, although these rarely cause syncope unless patients with seizures are included. Neurologic causes of syncope include transient ischemia (almost exclusively involving the vertebrobasilar territory), migraines (basilar artery), and seizures (atonic seizures, temporal lobe epilepsy, and unwitnessed grand mal seizures) [12]. Table 2. Causes of Syncope Cardiac causes of syncope include coronary disease, congenital and valvular heart disease, cardiomyopathy, arrhythmias, and conduction system disorders. Coronary disease, congestive heart failure, ventricular hypertrophy, and myocarditis may set the stage for arrhythmia and syncope. Exertional syncope results from heart disease characterized by a fixed cardiac output that does not increase with exercise. Exertional syncope may also reflect arrhythmic or neurocardiogenic disorders or an anomalous coronary artery. Syncope may be the presenting symptom in elderly patients with acute myocardial infarction [13]; it rarely occurs with coronary artery spasm and aortic dissection. We used five population-based studies of unselected patients to estimate the prevalence of various causes of syncope [14-18]; the summary of these studies is necessarily limited by the variability in diagnostic criteria. The most common causes of syncope were vasovagal episode, heart disease and arrhythmias, orthostatic hypotension, and seizures. The cause of syncope could not be determined in approximately 34% of patients. All of these studies were done several years ago, and the proportion of patients with unexplained syncope is probably lower now, given wider use of event monitoring, tilt testing, electrophysiologic studies, attention to psychiatric illnesses, and recognition that the cause of syncope in elderly patients may be multifactorial. Approach to Syncope The algorithm depicted in Figure 1 provides a diagnostic approach to syncope. It is intended to provide a framework for clinical judgment, not to replace it. Key points in the algorithm that will be discussed in the text include the following. Figure 1. Algorithm for diagnosing syncope. 1. History, physical examination, and electrocardiography are the core of the workup for patients with syncope. 2. Carotid sinus massage may be useful in elderly patients but should not be done by the generalist if bruits are present, if the patient has a history of ventricular tachycardia, or in the setting of a recent stroke or myocardial infarction. A false-positive test result should be suspected if carotid massage is positive but the history does not suggest carotid hypersensitivity. 3. Special issues for elderly patients include the multifactorial nature of syncope, polypharmacy, use of carotid sinus massage, and cardiac testing (exercise stress test and echocardiography) to exclude cardiac disease. 4. Nondiagnostic arrhythmias found on Holter monitor readings should not usually be treated. 5. Intracardiac electrophysiologic studies are most useful in patients who have organic heart disease and otherwise unexplained syncope. 6. In a patient with exertional syncope, echocardiography should precede exercise stress testing. 7. The assessment of patients with a normal heart who have frequent episodes of syncope should include a loop recorder and psychiatric evaluation. 8. The workup of patients with a normal heart who have infrequent episodes of syncope should include a tilt test and psychiatric evaluation. 9. Neurologic testing, including electroencephalography, computed tomography, and carotid and transcranial Doppler ultrasonography, should be rese