Approach

The initial assessment of patients with suspected syncope usually occurs in the accident and emergency (A&E) department or acute-care clinic. After determining as far as possible that the patient has experienced true syncope, the initial assessment should then focus on the following key points:

  • A detailed description of symptomatic events

  • Ascertaining whether clinically important structural heart disease is present (with an ECG and often an echocardiogram)

  • Clinical features of the history that suggest a diagnosis.

In experienced hands, the cause of syncope can be established by a comprehensive medical history and physical examination alone in approximately 60% of patients.[23][24] However, up to one third of patients with syncope have an undetermined cause after initial evaluation.[25] Further investigations are individualised so that the work-up is cost effective and appropriate.[1] Ultimately, the goal is to rule out more serious conditions and efficiently establish the cause(s) of syncope with sufficient confidence.[19]

History and physical examination

The medical history is the most valuable tool in the initial syncope assessment. In some cases, the history alone is diagnostic and no further testing is needed. The most common example is when the history is indicative of a classic vasovagal faint or a situational neurally mediated reflex faint. However, in other cases, the history may be unreliable, and patients (especially older adults) or witnesses may not be able to remember details. Patients often have multiple comorbidities, and as a consequence they may have several equally probable causes of fainting. If the history is suggestive but inconclusive, subsequent steps in the assessment can be focused, efficient, and cost effective.

Spend time eliciting all details of the event as best as possible. Look for patterns that may help narrow the diagnosis: document the preceding circumstances, premonitory symptoms, and subsequent outcome, and make careful note of comorbidities (e.g., diabetic neuropathy, autonomic dysfunction). A pre-prepared patient questionnaire may help save time while acquiring the needed detail.

Although the focus of this topic is NMRS, always consider the possibility of other causes of syncope (e.g., orthostatic, cardiovascular). The American College of Cardiology (ACC)/American Heart Association (AHA)/Heart Rhythm Society (HRS) recommend that the physical examination should include orthostatic blood pressure (BP) and heart rate changes in lying and sitting positions, on immediate standing, and after 3 minutes of upright posture.[1] Perform a basic neurological examination; sensory, motor, speech, or vision deficits suggest an underlying neurological problem requiring further investigation or referral.[1]

The diagnostic flow pathway illustrated (modified after that devised by the European Society of Cardiology Syncope Guidelines Task Force) may prove helpful.[2]

[Figure caption and citation for the preceding image starts]: Initial assessment and further management strategy of syncope based on European Society of Cardiology Task Force Guidelines. AECG, ambulatory ECG; CSM, carotid sinus massage; EP, electrophysiological; ILR, implantable loop recorder; SMU, syncope management unitFrom personal collection of Dr David Benditt; used with permission [Citation ends].com.bmj.content.model.Caption@1edbe425

Rule out structural heart disease

Underlying structural heart disease is among the most important factors to identify in a patient with suspected syncope. Obtain a 12-lead ECG in all patients to rule out atrioventricular (AV) block, bradycardia, asystole, long QT, or a bundle branch block. An echocardiogram can also be useful. The presence of heart disease is an independent predictor of a cardiac cause for syncope (i.e., a primary arrhythmic cause or a cause based on a structural cardiac abnormality leading to a transient haemodynamic disturbance), with a sensitivity of 95% and a specificity of 45%; by contrast, the absence of heart disease excludes a cardiac cause of syncope in 97% of the patients.[23]

Patients with structural abnormalities of the heart (such as left ventricular dysfunction) or ECG abnormalities (such as a prolonged QT or QRS interval) should be referred for immediate further evaluation.

Syncope in conjunction with exertion raises special concerns. In particular, if the faint occurs in 'full flight', consider the possibility of structural and/or dynamic heart lesions that produce a relatively fixed cardiac output in the setting of vascular dilation (e.g., severe aortic or mitral valvular stenosis, hypertrophic cardiomyopathy). However, syncope during or soon after exercise (even moderate exertion such as climbing stairs) can also occur in patients with severe autonomic dysfunction (e.g., pure autonomic failure) in whom vascular control is unable to maintain adequate cerebral perfusion pressure. On rare occasions, syncope accompanying exertion may occur as a consequence of a neurally mediated reflex faint (i.e., post-exertional variant of vasovagal faint); however, in these cases the faint typically occurs shortly after completion of, not during, the exercise.

Risk stratification

Assesses whether further hospital evaluation and treatment is necessary, establishes urgency, and estimates prognosis.[2][26]

The ACC/AHA/HRS guideline recommends evaluating short-term risk related to syncope (necessary for immediate decision making in the acute setting) as well as long-term adverse events (most likely to be associated with underlying medical conditions, many of which are cardiac).[1] The use of risk stratification scores may be reasonable.[1] However, the key determinant for further hospital evaluation and treatment is the presence of a serious medical condition potentially relevant to the cause of syncope.[1]

Serious medical conditions that may require further hospital evaluation and treatment include:

  • Cardiac arrhythmic conditions

  • Cardiac or vascular non-arrhythmic conditions

  • Non-cardiac conditions (e.g., gastrointestinal bleeding or severe anaemia).

Patients with documented severe bradyarrhythmias, ventricular tachycardia, acute aortic dissection, abrupt haemorrhage, or pulmonary embolism should be admitted. Similarly, it is prudent to admit patients with moderate to severe left ventricular dysfunction (e.g., ejection fraction ≤35%), valvular disease of clinical concern, or hypertrophic cardiomyopathy.

One study found a high prevalence of pulmonary embolism among patients hospitalised for their first episode of syncope (approximately 1 in 6 patients); this may be important to consider in patients without an alternative explanation, but further research is required.[27]

Where a serious medical condition is absent, it is reasonable to manage presumptive reflex-mediated syncope in the outpatient setting.[1]

Rules and criteria that may help to risk stratify patients presenting to the emergency department include: the San Francisco Syncope Rule (SFSR), the Canadian Risk Syncope Score (CSRS), the Osservatorio Epidemiologico sulla Sincope nel Lazio (OESIL) score, the Evaluation of Guidelines in Syncope Study (EGSYS) score, and the Risk Stratification of Syncope in the Emergency Department (ROSE) study.[28][29][30][31][32]​​​​ The two most commonly used criteria are the SFSR and the ROSE study. 

The SFSR has been implemented and validated in many accident and emergency departments and can be remembered using the mnemonic 'CHESS':[28]​​​​ [ San Francisco Syncope Rule TreeCalc Opens in new window ]

  • Congestive heart failure (CHF)

  • Haematocrit <30%

  • ECG abnormal

  • Shortness of breath

  • Systolic BP <90 mm Hg.

The newer ROSE study includes B-type natriuretic peptide (BNP) as part of the decision criteria and can be remembered using the mnemonic 'BRACES':[29]​​​​ [ ROSE score: risk stratification of syncope in the emergency department Opens in new window ]

  • BNP level ≥300 picograms/mL

  • Bradycardia ≤50 in emergency department or pre-hospital

  • Rectal exam positive for faecal occult blood

  • Anaemia (Hb ≤90 g/L)

  • Chest pain associated with syncope

  • ECG showing Q-waves (but not in lead III)

  • Saturation ≤94% on room air.

Other initial investigations

Consider targeted blood tests to rule out other causes of syncope, including:

  • Cortisol (to evaluate adrenal insufficiency)

  • Urea and creatinine (to evaluate for dehydration)

  • Haemoglobin (to rule out anaemia)

  • Blood glucose (to rule out hypoglycaemia)

  • Beta-hCG (human chorionic gonadotrophin; to rule out pregnancy)

  • Cardiac enzymes (to rule out myocardial infarction)

  • D-dimer (to rule out pulmonary embolism) and/or CT pulmonary angiogram.

Guidelines recommend against broad testing; only order investigations relevant to the clinical situation.[26][33]

Patients with recurrent syncope, or those who have experienced trauma due to syncope, warrant further evaluation with an ambulatory ECG monitor (i.e., Holter monitor), loop, or event monitor, or possibly tilt table testing. Non-cardiovascular causes can be examined through appropriate blood tests (e.g., glucose, cortisol), electroencephalogram (EEG), magnetic resonance imaging (MRI) of the brain, or autonomic testing.

Vasovagal syncope

Establishing the diagnosis of vasovagal faint relies heavily on obtaining a detailed medical history, including eyewitness accounts.

Classic features that suggest vasovagal syncope include:

  • Triggering factors:

    • Unpleasant sights

    • Pain from physical injury

    • Fatigue

    • Palpitations or bradycardia (consistent with but not diagnostic of vasovagal syncope)

    • Emotional stress (especially in warm, crowded environments)

    • Prolonged standing

    • Dehydration

    • Vigorous exercise (if syncope occurs post exercise)

  • Warning (prodromal) symptoms:

    • Feeling hot or cold

    • Sweating

    • Tachycardia (consistent with but not diagnostic of vasovagal syncope)

    • Shortness of breath

    • Loss of hearing or vision

    • Nausea

    • Change in breathing pattern

  • Physical signs reported by witnesses:

    • Marked pallor (white as a ghost, death-like)

    • Clammy skin

    • Confusion.

Recovery is typically rapid after a vasovagal faint if the patient remains recumbent, but a subsequent period of fatigue is common.

However, the classic features of vasovagal faint are often absent, only incompletely present, or not recollected (especially in older people). Patients may also experience symptoms (e.g., erratic muscle movements while unconscious, urinary or rarely bowel incontinence) that can result in confusion with seizure disorders and necessitate a neurological consultation. Family history of sudden death should trigger concern about other diagnoses (e.g., structural heart disease, long QT syndrome, Brugada syndrome).

Perform additional tests when the history alone is unclear. Head-up tilt-table testing is the most common confirmatory test for vasovagal syncope.[34][35][36][37][38]

Other NMRSs

These include carotid sinus syncope, situational faint, and glossopharyngeal and trigeminal neuralgia.

In carotid sinus syndrome (CSS), there is a history of accidental manipulation of the neck that results in external pressure on the carotid sinus baroreceptors. There are three types of CSS: bradycardic CSS, dominant vasodepressor CSS (predominant sign is a relative hypotension), or mixed (bradycardia and hypotension). Deliberate diagnostic carotid sinus massage (CSM) applied by a suitably experienced physician may reproduce symptoms and confirm the diagnosis in patients with CSS.[22][39] ECG in these cases will show CSS bradycardia.

Situational syncope can be triggered by acute haemorrhage, a cough, a sneeze, gastrointestinal stimulation (swallow, defecation, visceral pain), micturition, post exercise, or other situations including brass instrument playing, weightlifting, or eating a meal (post-prandial). Diagnosing situational syncope relies on the history; it is not easily assessed in the laboratory. Cough syncope may be an exception, but diagnostic criteria for haemodynamic response to induced cough have yet to be determined.

Glossopharyngeal neuralgia is an unusual pain syndrome associated with syncope. The hallmarks of this syndrome are episodic sensations of usually left-sided pharyngeal pain that frequently result in syncope. The pathogenesis relating the pain and syncope is unclear, but it is thought to be caused by connections between the glossopharyngeal and vagus nerves.[40]

Trigeminal neuralgia is a somewhat more common facial pain syndrome presenting with pain at one of the divisions of the trigeminal nerve. However, syncope resulting from trigeminal neuralgia is unusual and the pathogenesis is unclear. Trigeminal neuralgia has been associated with syncope, seizures, and even cardiac arrest.[41]

Laboratory studies: basic autonomic assessment

1. Head-up tilt-table testing (HUT)

The HUT is a widely available and well studied diagnostic tool that can unmask susceptibility to vasovagal syncope.[4][22][34][35][36][37][38] It is indicated when vasovagal syncope is suspected but the clinical history is equivocal. If the history is classic, HUT is not necessary. The most common test protocols are: HUT with glyceryl trinitrate; HUT with isoprenaline; or drug-free HUT. The drugs may be used to increase susceptibility to the test.[42]

Detailed discussion of tilt-table testing protocols can be found in the ACC expert consensus report and guideline documents from the European Society of Cardiology and the ACC/AHA/ HRS.[1][2][34] In brief, the first step is passive head-up tilt at 60° to 70°, during which the patient is supported by a footplate and gently applied body straps, for 20-45 minutes (protocols vary). If needed, tilt testing is repeated with a drug challenge. Until recently, the most frequently used provocative drug was isoprenaline, usually given in escalating doses. However, intravenous or sublingual glyceryl trinitrate has gained favour, in part because it expedites the procedure without adversely affecting diagnostic use.[43] The so-called Italian protocol, in which a 20-minute passive tilt is followed if necessary by sublingual glyceryl trinitrate, has become the most popular protocol.

An important clinical attribute of HUT is that it provides the opportunity to precipitate a typical vasovagal attack in the presence of the physician. This enhances the patient's confidence in the diagnosis. A positive test can also help patients learn to recognise symptoms and prevent an impending faint.

Tilt-table testing without pharmacological provocation exhibits a sensitivity of 37% and specificity of 99%; with pharmacological provocation (glyceryl trinitrate administration), the sensitivity is 66% and the specificity is 89%.[43] Nonetheless, HUT is imperfect and there is controversy regarding its optimal use. The false-positive rate is approximately 10%.[2] Acute reproducibility, in terms of whether syncope is induced, is 80% to 90%. Longer-term reproducibility (i.e., >1 year) is about 60%.[44]​ Experience of the practitioner can be a major limitation.

Another major weakness of HUT is that it does not always result in the same haemodynamic picture when repeated in the same patient, and thus is not appropriate for directing treatment strategy. A cardioinhibitory picture (i.e., bradycardia) may predominate on one occasion, whereas vasodepressor responses (vasodilation being principally responsible for the hypotension) may occur at other times.[7][45][46] In the International Study of Syncope of Unknown Etiology (ISSUE) study, even when tilt testing indicated a prominent vasodilatory component to the faint, the subsequent recording of spontaneous faints by an implantable loop recorder often revealed bradycardic events.[7] Further study of this phenomenon with devices that can monitor systemic BP, or a surrogate of BP, as well as heart rate is needed.

A positive tilt-table test (especially if it reproduces the patient's spontaneous symptoms) can be considered diagnostic for patients without severe structural heart disease, and no further tests are needed. However, for patients with significant structural heart disease, cardiac arrhythmias and iatrogenic factors (e.g., drug effects) should be excluded before relying on a positive tilt-test result.[43]

2. Carotid sinus massage (CSM)

Pressure at the site where the common carotid artery bifurcates produces a reflex slowing in heart rate and decrease in BP. This observation is the basis of the technique of CSM.

CSM is used to diagnose CSS in patients with a history of syncope. However, CSM can be positive in some people with no history of syncope (up to 40% of older adults); this is known as carotid sinus hypersensitivity.[2] Requiring reproduction of symptoms increases the specificity of CSM: in a study of 1855 patients with unexplained syncope, 24% demonstrated carotid sinus hypersensitivity, but only about one third of these (8.8%) had CSS when symptoms were required to make the diagnosis.[47]

Consensus opinion is that CSS may be diagnosed when CSM reproduces symptoms as a result of ≥1 of the following being induced: >3 seconds of asystole, paroxysmal atrioventricular (AV) block, a marked decrease in systemic arterial pressure (usually a drop in systolic BP ≥50 mmHg), or a combination of these.[1][2][48] CSM should be performed in both upright and supine positions, with pressure applied for 5 seconds in each position.[1] The European Society of Cardiology recommends applying pressure for 10 seconds to allow time for symptoms to develop.[48]

The most convincing results from CSM are obtained when it is undertaken with the patient in the upright position, as the impact of gravity exaggerates the haemodynamic effects of induced bradycardia or vasodilation. The diagnosis may be missed in about one third of cases if only supine CSM is performed. Arterial pressure and ECG should be recorded throughout. For measurement of arterial pressure, a non-invasive BP measurement device may be suitable, but satisfactory recordings are often difficult to obtain. A conventional sphygmomanometer is not appropriate.

CSM should not be performed in patients who have experienced transient ischaemic attack or stroke within the past 3 months or in patients with carotid bruits (unless carotid Doppler studies convincingly exclude significant carotid artery narrowing).[2]

On a rare occasion, CSM may elicit self-limited atrial fibrillation of little clinical significance. Finally, findings recorded in the period after an acute myocardial infarction may not be representative and should not be used to make a diagnosis of CSS.

3. Adenosine triphosphate (ATP) test

The value of bolus administration of ATP remains controversial.[2] The ATP test may be useful to identify a form of syncope associated with neurally mediated paroxysmal AV block in certain older people in whom other causes have been excluded, and may guide a decision to initiate permanent pacing by unmasking occult AV nodal disease in susceptible people.[1] The diagnostic and predictive value of the test remains to be confirmed by prospective studies. In the absence of sufficient data, the test may be considered at the end of the diagnostic work-up. 

The most thoroughly tested protocol involves injection of a 20-mg bolus of ATP into a brachial vein in a supine patient who is undergoing continuous ECG monitoring.[49] BP is monitored non-invasively. The outcome of the test (positive or negative) depends on the duration of the ATP-induced cardiac pause. A pause >10 seconds, even if interrupted by escape beats, is defined as abnormal. Some reports suggest that a pause >6 seconds is sufficient to declare the test abnormal.[50] For patients with abnormal responses, reproducibility is approximately 80% both in the short and the long term.

Because of the possibility of bronchospastic reactions, the ATP test is contraindicated in patients with known asthma.

4. Insertable loop recorder

An insertable loop recorder can be used to rule out brady- or tachyarrhythmia if events are relatively infrequent (<1 or 2 per month) and an arrhythmia is suspected.[26]

5. Electrophysiological study

Electrophysiological study is infrequently indicated.[1] It may be used if structural heart disease is present and a tachyarrhythmia is suspected, and in patients who may be experiencing intermittent high-grade conduction block. It is also warranted if there is suspicion of paroxysmal supraventricular tachycardia as the cause of symptoms, especially if the patient describes palpitations prior to syncope.

Miscellaneous autonomic system tests

1. Valsalva manoeuvre

The Valsalva manoeuvre is used to assess the integrity of the arterial baroreceptor reflex arc and may indicate the presence of autonomic failure that can cause syncope through orthostatic hypotension.[2] In itself, however, the Valsalva manoeuvre does not directly implicate a mechanism for syncope. The Valsalva response is typically tested under continuous ECG and BP monitoring in addition to monitoring the patient for clinical symptoms.

2. Active standing test

As its name implies, the active standing test assesses a patient's response to active movement from supine to upright posture. It can be considered to rule out orthostatic hypotension.[51]

Active muscle movement is expected to propel more blood towards the central circulation to provide the increase in cardiac output required for upright posture. However, active use of leg muscles may also aggravate peripheral vascular dilation and, as a result, induce greater hypotension. The balance between these effects determines the net effect on systemic BP.

A 3-minute active standing test is useful to assess initial orthostatic hypotension (first 30 seconds of upright posture).[2] The initial hypotension occurs on active standing but not on tilting. To detect the pressure decrease, beat-to-beat pressure monitoring (usually non-invasive) is essential. A ≥3-minute period of active standing may also demonstrate later diminution of BP indicative of classic orthostatic hypotension (delayed orthostatic hypotension).[1][48]

3. Cold pressor test

The cold pressor test is performed by immersing the hand into a container of ice-cold water, usually for 1 minute. Any changes in blood pressure and heart rate are recorded. As with the Valsalva manoeuvre, this test provides insight into the integrity of the autonomic reflex. It has not been used to identify a specific diagnosis.

4. Cough test

Induced cough may help assess susceptibility to cough (tussive) syncope, but few data are available.[52] As with CSM, it may be best undertaken with the patient in the upright posture. Diagnostic criteria for haemodynamic response to induced cough have yet to be determined.

Use of this content is subject to our disclaimer