Syncope in older adults

Approximately 3% of all visits to the emergency department are due to syncope.[1] Because rates of syncope increase with age, older adults are especially vulnerable to syncope-related injury.[1] Falls due to fainting, which often lead to hip fractures, hospital admissions, and institutionalization, have a direct cost to the Canadian health care system of $60 million a year.[2,3]

Syncope is commonly defined as “a transient loss of consciousness.”[4] Both medical school and residency training teach that syncope is diagnosed primarily by symptoms preceding the loss of consciousness (e.g., giddiness, lightheadedness, tunnel vision, nausea, spots in the vision). Unfortunately, approximately 30% of cognitively normal older patients who experience syncope under controlled laboratory conditions (i.e., during tilt table testing) will not recall the event.[5] This lack of recall is even worse in cognitively impaired patients. The fact that most falls in older adults are unwitnessed compounds the challenge presented by patients unable to recall presyncopal complaints.[6] Syncope should thus be considered a possible contributing factor in any unexplained fall in an older adult.

Causes of syncope
Often, the cause of a syncopal spell can be impossible to determine. A prospective study of a complete clinical evaluation for syncope, which included tilt table testing, Holter monitoring, and implantable loop recorder (ILR) monitoring, was unable find a cause for the episode one-quarter of the time.[7] This diagnostic uncertainty can lead to a multiplicity of expensive but very low-yield investigations.[8] In the emergency department, risk stratification scales have been developed and internally validated,[9] but consequent external validation has demonstrated poor specificity.[10] Neuroautonomic syncope, cardiac syncope, and syndromes with syncope-like symptoms (Table 1) are common causes of syncopal events. Regardless of the cause, all patients presenting with syncope should have their bone health assessed and be treated for osteoporosis if appropriate and prescribed hip protectors. The local jurisdiction’s laws respecting syncope and the operation of motor vehicles or aircraft should also be considered.

Neuroautonomic syncope
Neuroautonomic syncope is defined as “age-related changes in blood pressure and heart rate behavior, predominately resulting in hypotension and bradyarrhythmia.”[11] Neuroautonomic syncope encompasses four distinct syndromes: orthostatic hypotension (OH), vasovagal syncope (VVS), carotid sinus syndrome (CSS), and postprandial hypotension.[11]

When an adult assumes the standing position, approximately 300 to 800 mL of blood pools in the lower limb vasculature.[12] Under normal conditions, a neuroautomomic tachycardic response follows orthosis in order to maintain adequate cardiac output and peripheral vasoconstriction. Loss of consciousness occurs when the neuroautonomic response is impaired and cerebral blood flow is not maintained. Current guidelines define orthostatic hypotension as a drop in systolic blood pressure of 20 mm Hg or a drop in diastolic blood pressure of 10 mm Hg after standing upright for 3 minutes.[13] Identifying orthostatic hypotension at the bedside can be difficult as blood pressure responses are not always reproducible and repeat assessments are often needed. A significant decline in blood pressure that occurs immediately after standing but resolves within 3 minutes is not usually of clinical concern; however, failure to resolve has been linked to an increased risk of falling even in asymptomatic individuals.[14,15] During tilt table testing and measurement of orthostatic vital signs, OH is associated with an appropriate but ultimately ineffective tachycardic response as opposed to the cardioinhibition seen in VSS and CSS. Medication (especially cardiac and vasodilatory medications) and hypovolemia are the usual causes of OH, but there are many others (Table 2).

Once the underlying cause of OH has been addressed, first-line therapy for mild cases consists of nonpharmacological interventions. These include increased salt consumption, slow assumption of the upright position, and elevation of the head of the bed to 30 degrees at night. Bed elevation works by preventing the loss of intravascular volume due to nocturnal diuresis. Other factors that can worsen OH by exacerbating vasodilatation and causing venous pooling are hot ambient temperatures, physical exercise, and alcohol consumption. Syncopal spells can be prevented with isometric exercise maneuvers (e.g., squeezing crossed legs together), which have been shown to elevate blood pressure by approximately 20 mm Hg. OH that is refractory to nonpharmocological interventions should be treated initially with a mineralocorticoid agonist such as fludrocortisone to expand the intravascular volume. Contraindications to fludrocortisone consist mainly of congestive heart failure or other reasons to avoid worsening volume overload.[12] An alpha-1 agonist such as midodrine (a vasoconstrictive agent that has been validated in a randomized controlled trial)[16] should be used if mineralocorticoid agents are contraindicated or ineffective. Midodrine has the potential to cause urinary obstruction in patients with an enlarged prostate, so care should be taken when prescribing this agent.

In patients older than 60 years, vasovagal syncope has been shown to cause approximately one-third of syncopal events. Susceptibility to VVS is increased by dehydration, either through the administration of medications such as diuretics or through poor oral intake. There are few effective pharmacological treatments for VVS, mainly because the mechanisms behind vasovagal syncope remain poorly understood.[17] Previously, the Bezold-Jarisch reflex (a reflex involving cardiac ventricular receptor stimulation) has been invoked to explain VVS, but this has been disproved in animal studies,[18] while some recent research suggests that an exaggeration of the arterial baroreflex may contribute to the development of VVS.[19] The standard presentation of VVS usually consists of a syncopal spell that occurs after a long time spent in the upright posture, usually in the setting of hypovolemia (such as during gastrointestinal illness or diuretic administration). An incorrect understanding of the mechanisms behind VVS previously led to the use of beta-blockers to reduce ventricular mechanoreceptor stimulation,[20] but randomized controlled trials have now shown that these medications actually worsen VVS.[21] Strong evidence supports using pacemaker insertion to treat the cardioinhibitory component of VVS[22] in troublesome cases that do not resolve with conservative care. Although VVS is usually a self-limiting condition, even a single fall can result in debilitating fractures in older adults.

Carotid sinus syndrome is characterized by an arterial baroreflex that is overly sensitive to oscillations in blood pressure levels, resulting in transient asystole and vasodilatation.[23,24] One-quarter of older adults who present to the emergency department after a fall have been shown to have CSS,[25] and CSS has been found to be an independent risk factor for hip fractures.[26,27] Occasionally, these episodes include a history of neck stimulation (e.g., from wearing tight collars or turning the head), but usually there is no identifiable trigger.[23] The only way to diagnose CSS definitively is by employing carotid sinus massage while the patient is on a cardiac monitor in the upright position.[28] Patients who present with a predominant asystolic response as opposed to vasodilation have been shown to respond to pacemaker insertion.[24,29-31] Pacemaker insertion in older adults with purely cardioinhibitory CSS has also been shown to reduce falling by 66% and injurious events by 70%.[24] A more recent trial, however, did not show any improvement in the fall rate,[32] suggesting that the role of pacemakers in CSS requires clarification.

An uncommon cause of neuroautonomic syncope is postprandial hypotension, a condition where increases in splanchnic blood flow in older adults result in hypotension.[33,34] A patient complaining of frequent lightheadedness or loss of consciousness 30 to 90 minutes after eating should be given a trial of diet modification (e.g., smaller more frequent meals, reduction in the simple carbohydrate portion of any meal).[11]

Cardiac syncope
Cardiac syncope, although less common then other forms, is associated with a significantly higher 1-year mortality.[35,36] The main causes for cardiac syncope can be divided into organic heart disease (usually aortic stenosis in older adults) and arrhythmias (usually sinus node disease or heart blocks). Cardiac syncope has a significant associated mortality, mainly due to injury.[35,36] Organic heart disease is detected commonly through cardiac auscultation assisted by echocardiography if necessary. Arrhythmias are detected commonly through electrocardiography and the targeted use of Holter monitoring and implantable loop recorder monitoring. Referral to a cardiologist for appropriate medical, surgical, and pacemaker therapy is required for all cardiac syncope cases.[31] Cholinesterase inhibitors have been shown to increase the risk of syncope and hip fractures in older adults with dementia,[37] suggesting that adverse events due to cognition-enhancing agents also need to be considered.

Syndromes with syncope-like symptoms
Hypoglycemia, seizures, transient ischemic attacks, and psychiatric conditions are also conditions that can alter consciousness. These syndromes should be considered along with other causes when a patient presents with syncope.

Assessing syncope
A step-by-step approach to managing syncope in the primary care setting begins with measuring orthostatic vital signs, obtaining an electrocardiogram, and determining bone density (Table 3). Information from a collateral historian (family member, care provider) can be useful, especially when a seizure of some kind may be involved.

The main priority in assessing syncope is to determine whether the spells are neuroautonomic or cardiac in origin. Not all patients presenting with loss of consciousness require an intensive workup for cardiac syncope. A better diagnostic yield results when patients have classic arrhythmia symptoms (syncope in the seated or supine position, exertional syncope, or palpitations), abnormal ECG results, or structural heart disease that has been confirmed by a physical exam or targeted echocardiography. Not all patients presenting with loss of consciousness require an intensive workup for neuroautonomic syncope either. Orthostatic hypotension can be diagnosed at the bedside and may require no further testing. More intensive investigations such as tilt table testing or carotid sinus massage for other neuroautonomic causes of syncope should only be performed in patients with recurrent syncopal spells, since these are usually treated by pacemaker insertion.

Echocardiography 
Echocardiography has quite a low diagnostic yield in patients who present to the emergency department with syncope, leaving its overall role in the workup for syncope somewhat uncertain.[7] The current consensus is that echocardiography should be used as an adjunct to the physical examination to detect structural heart disease or to check for low systolic function.

Holter monitoring 
Holter monitoring for 24 hours has a low overall diagnostic yield of 7%, but this yield has been found to double if the patient has electrocardiographic abnormalities.[7] Holter monitoring is not meant to be used as a general screening tool for older adults who have fallen; a study of older adults who fell frequently and older adults who did not found the two groups to have indistinguishable results on Holter monitoring.[38] Current best practice is to use targeted Holter monitoring based on the presence of ECG abnormalities, heart disease (including congestive heart disease), and highly suspicious symptoms such as exertional syncope or syncopal spells while seated or lying down. In the patient with a high pretest probability of cardiac syncope, targeted Holter monitoring consists of an initial 24-hour recording followed by a 48-hour recording if the first one was negative.

Implantable loop recorder monitoring 
Implantable loop recorders have become widely used in cases where arrhythmia-related syncope is suspected. An ILR surgically implanted in the subcutaneous tissue of the upper chest allows cardiac monitoring for arrhythmias over months instead of days. This technology is developing rapidly to incorporate wireless data transmission and home monitoring, and ILR use has been shown to increase diagnostic yield and reduce hospitalization rates in patients with recurrent syncope.[39] However, there is a paucity of results from randomized controlled trials, so we do not yet fully understand which patients require ILR monitoring.[40] In addition, most trials to date have been quite small and used the rate of ECG diagnosis as the main outcome rather than more clinically meaningful outcomes.[40] As well, the insertion of an ILR requires the involvement of a specialized cardiology centre and should be considered only if symptoms have become intolerable and repeated Holter monitoring has not proven diagnostically useful. A small study found pretest factors that increased the likelihood of ILR testing to find significant asystole or bradyarrhythmias include prolonged loss of consciousness, the absence of a prodrome, and convulsive movements during syncope.[37]

Tilt table testing
Tilt table testing is a way to reproduce a vasovagal response under controlled laboratory conditions. Tilt table testing can occasionally aid in the diagnosis of vasovagal syndrome, although the usefulness of this test in the primary care context is impeded by the lack of a gold standard for comparison,[41] poor reproducibility,[41] and the lack of availability in many centres.

Electroencephalography
The differential diagnosis for a syncopal spell should always include a seizure disorder. Symptoms such as a preceding aura, tongue biting, and tonic-clonic movements or incontinence increase the likelihood of seizure activity being a factor in syncopal spells and indicate an EEG is required. It should be noted, however, that EEG testing has a very low diagnostic yield, and the diagnostic yield does not improve when the test is ordered after neurological consultation.[42]

Summary
A transient loss of consciousness can result in falls that lead to hospital admissions and institutionalization. In older adults, neuroautonomic syncope, cardiac syncope, and syndromes with syncope-like symptoms all contribute to fall-related injuries, loss of independence, and mortality. Assessing syncope as a mechanism for falls can be quite challenging in the general population, and even more difficult in older adults with cognitive issues. A structured approach that includes obtaining information from collateral historians can help determine the cause of syncope to aid diagnosis and reduce the probability of future falls. 

Competing interests
The author receives operating grants from the Canadian Institutes of Health Research, the Heart and Stroke Foundation, and the Canadian Diabetes Association. There is no commercial association, source of funding, or other arrangement that poses a potential conflict of interest.

Acknowledgments
The work described in this article was supported by a grant from the Canadian Institutes of Health Research, the Canadian Diabetes Association, and the Heart and Stroke Foundation.

This article has been peer reviewed.

Dr Madden is an associate professor in the Gerontology and Diabetes Research Laboratory of the Division of Geriatric Medicine at the University of British Columbia. He is also a clinical scientist at the Centre for Hip Health and Mobility at UBC.