Transient Ischemic Attack (TIA)

Overview and Recommendations

Background

● TIA is a transient episode of neurologic dysfunction caused by focal ischemia of the brain, spinal cord,
or retina without evidence of acute infarction on neuroimaging.

● TIAs may be caused by atheroemboli from carotid or vertebral artery stenosis, cardiogenic emboli in
patients with valvular stenosis or arrhythmias such as atrial brillation, paradoxical emboli through
intracardiac defects, aortic arch emboli, and in patients with hypercoagulable, prothrombotic, or
hyperviscous states such as sickle cell disease, thrombocytosis, or leukemia.

● Risk factors for TIA include:

⚬ metabolic conditions such as metabolic syndrome, diabetes mellitus type 2, and chronic kidney
disease
⚬ modi able risk factors such as obesity, smoking, hyperlipidemia, heavy alcohol use, and substance
abuse
⚬ cardiovascular conditions such as hypertension, atrial brillation, carotid stenosis, or history of
prior myocardial infarction, nonischemic cardiomyopathies, TIA, or stroke
⚬ hormone replacement therapy and oral contraceptives

● Most TIAs resolve within 1-2 hours, but the duration of symptoms does not predict the subsequent
risk of stroke.

Evaluation

● Neurologic symptoms consistent with cerebral ischemia from the anterior circulation include:

⚬ motor or sensory dysfunction of the contralateral face and/or contralateral extremities

⚬ ipsilateral vision loss or homonymous hemianopia
⚬ speech disturbances such as aphasia or dysarthria

● Neurologic symptoms consistent with a posterior circulation event (vertebrobasilar circulation) may
include:
⚬ motor or sensory dysfunction of the ipsilateral face and/or contralateral extremities (crossed
pattern)
⚬ visual loss in 1 or both homonymous visual elds including cortical blindness and Anton syndrome
(lack of awareness of blindness)
⚬ brainstem and cerebellar ndings such as ataxia, vertigo, diplopia, dysphagia, and dysarthria

● In patients referred for a suspected TIA or minor stroke, the most frequent noncerebrovascular
diagnoses include migraine, syncope, seizure, psychiatric conditions, vertigo, and peripheral nerve
conditions.

● Patients with a suspected TIA should be evaluated as soon as possible after the event (Strong
recommendation).

● Use validated tools such as Face Arm Speech Test (FAST) or Recognition of Stroke in the Emergency
Room (ROSIER) test to screen and diagnose patients for stroke or TIA; exclude hypoglycemia as cause
 of symptoms.

● Initial blood tests for the evaluation of patients with a suspected TIA may include (Weak
recommendation):
⚬ complete blood count with platelet count evaluating for in ammation, thrombocytosis, and
myelodysplastic disease
⚬ chemistry panel to assess for electrolyte imbalances and kidney dysfunction
⚬ prothrombin time (INR) and activated partial thromboplastin time (aPTT) to rule out coagulopathy
⚬ fasting lipid panel to assess for risk of atherosclerotic vascular disease
⚬ fasting glucose and consider obtaining a HbA1c if diabetes is suspected
⚬ serial cardiac biomarkers (troponin) if acute coronary syndrome suspected

● Consider a hypercoagulable workup in younger patients with a TIA, especially in those without
vascular risk factors or underlying identi able causes.

● Neuroimaging with magnetic resonance imaging (MRI), including di usion-weighted imaging (DWI), or
computed tomography (CT) (if MRI is not possible) should be obtained within 24 hours of symptom
onset (Strong recommendation).

● An electrocardiogram (ECG) should be obtained to assess for atrial brillation, ventricular

hypertrophy, or signs of cardiac ischemia (Strong recommendation).

● Consider prolonged cardiac monitoring (inpatient telemetry or Holter monitoring) in patients with an
etiology not identi ed after initial neuroimaging and ECG.

● Consider transthoracic and, depending on the clinical circumstance, a transesophageal

echocardiogram to assess for valvular or cardiogenic sources of emboli (Weak recommendation).

● Noninvasive vascular imaging is recommended for cervical and intracranial vessels with ultrasound,
computed tomographic angiography, or magnetic resonance angiography (Strong recommendation).

Acute Management

● For acute management of TIA, o er aspirin (unless contraindicated) to be started immediately.

● Assess the patient for risk of subsequent stroke as soon as possible to assist in determining
management.
⚬ Results of magnetic resonance imaging, vascular imaging, blood tests, echocardiogram, and
electrocardiography may contribute to assessing risk factors and subsequent treatment.
⚬ TIA patients with carotid stenosis and those with transient symptoms with infarction are at a
higher risk of both short- and long-term subsequent stroke than in patients with other etiologies.

● Determine the appropriate care setting depending on the timing and nature of symptoms and the risk
of stroke.
⚬ If presenting within 48 hours of the suspected TIA, the patient is considered to be:

– at the HIGHEST RISK for stroke if the symptoms include transient, uctuating, and/or persistent
unilateral weakness, or speech disturbance, and should be immediately sent to an emergency
department with the capacity for advanced stroke care
– at a HIGH RISK for stroke if there are transient, uctuating, or persistent symptoms without
motor weakness or speech disturbance, and should be sent for a same-day assessment at
 closest stroke prevention clinic or to an emergency department with the capacity for advanced
stroke care
⚬ Patients presenting within 2-14 days after the suspected TIA are considered to be at INCREASED
RISK for stroke and should have a comprehensive assessment by a healthcare professional with
stroke expertise as soon as possible and:
– no more than 24 hours after rst contact if the symptoms include transient, uctuating, and/or
persistent unilateral weakness, or speech disturbance
– no more than 14 days after rst contact If there are transient, uctuating, or persistent
symptoms without motor weakness or speech disturbance
⚬ Patients considered to be at a LOWER RISK for stroke include those:

– presenting more than 14 days after the suspected TIA who should be assessed by a neurologist
or stroke specialist as soon as possible (no more than 30 days after onset of the symptoms)
– with atypical sensory symptoms (such as patchy numbness and/or tingling) who should be
assessed by a stroke specialist as required
⚬ Also consider hospitalizing patients with a TIA if presenting within 72 hours of event if (Weak
recommendation):
– unclear if outpatient workup can be completed within 2 days
– other evidence indicates event caused by focal ischemia

Prevention of Stroke After TIA

● Implement measures to reduce the risk stroke.

● Antiplatelet agents should be used rather than anticoagulation for patients with noncardioembolic TIA
(Strong recommendation); options for most patients include
⚬ combination aspirin plus clopidogrel for up to 21 days, with single antiplatelet thereafter
⚬ aspirin 81-325 mg orally once daily
⚬ combination aspirin 25 mg plus extended-release dipyridamole 200 mg orally once daily
(Aggrenox)
⚬ clopidogrel 75 mg orally once daily
⚬ see Antiplatelet Therapy for Secondary Prevention of Stroke or Transient Ischemic Attack for a
complete discussion of strategies.

● Use oral anticoagulation in patients with atrial brillation (Strong recommendation). See
Thromboembolic Prophylaxis in Atrial Fibrillation for a complete discussion of strategies.

● For patients with carotid artery stenosis:

⚬ Perform carotid endarterectomy (CEA) for patients with a perioperative morbidity and mortality
risk estimated to be < 6% who have had a TIA or nondisabling ischemic stroke within the last 6
months and have either:
– ipsilateral severe (70%-99%) stenosis documented by noninvasive imaging (Strong
recommendation)
– ipsilateral moderate (50%-69%) stenosis documented by catheter-based imaging (Strong
recommendation)
⚬ Consider carotid artery stenting (CAS) as an alternative to CEA for symptomatic patients with an
average or low risk of complications associated with endovascular intervention plus stenosis > 70%
by noninvasive imaging or > 50% by catheter angiography (Weak recommendation).
⚬ See Carotid Artery Stenosis Repair for a complete discussion of strategies.
● Additional risk factor reduction through statin therapy, blood pressure reduction, smoking cessation,
and avoidance of heavy alcohol use should be implemented to prevent stroke in patients who have
had a TIA (Strong recommendation). Consider advising moderate physical activity (Weak
recommendation). See Secondary Prevention of Stroke or Transient Ischemic Attack for a complete
discussion of strategies.

Related Summaries

● Stroke (Acute Management)

● Secondary Prevention of Stroke or Transient Ischemic Attack

● Antiplatelet Therapy for Secondary Prevention of Stroke or Transient Ischemic Attack

● Thromboembolic Prophylaxis in Atrial Fibrillation

● Carotid Artery Stenosis Repair

● Risk Factors for Stroke or Transient Ischemic Attack

General Information

Description

● TIA is a transient episode of neurologic dysfunction caused by focal ischemia of the brain, spinal cord,

or retina, and without detection of acute infarction on neuroimaging. 1

Also called

● acute neurovascular syndrome

Definitions

● TIA

⚬ American Heart Association/American Stroke Association 2009 tissue-based de nition of TIA -

transient episode of neurological dysfunction caused by focal brain, spinal cord, or retinal
ischemia, without acute infarction (Stroke 2009 Jun;40(6):2276 )
⚬ previous time-based de nition was reversible focal ischemic neurologic dysfunction lasting < 24

hours 1
– generally accepted for use in absence of computed tomography or magnetic resonance imaging
availability
– 30% of patients diagnosed with TIA using this de nition may in fact have infarcted brain

● ischemic stroke - episode of neurological dysfunction caused by focal cerebral, spinal, or retinal
infarction (Stroke 2013 Jul;44(7):2064 )

● nondisabling stroke - transient neurologic symptoms with presence of brain infarction but mild

(National Institutes of Health Score ≤ 3 points) or no persistent clinical de cits 1

Types
Amaurosis fugax

● amaurosis fugax

⚬ transient monocular blindness lasting seconds to minutes

⚬ caused by ischemia to retina, choroid, or optic nerve
⚬ considered a subform of TIA
⚬ common mechanism is embolism from the ipsilateral carotid nerve; another possible cause is giant
cell arteritis
⚬ less commonly reported causes include

– left atrial sarcoma (Am J Ophthalmol Case Rep 2016 Dec;4:24 full-text )
– antiphospholipid syndrome (Am J Ophthalmol Case Rep 2017 Sep;7:143 full-text )
– neurosarcoidosis (Clin Neurol Neurosurg 2018 Jun;169:103 )
– Trousseau syndrome (J Stroke Cerebrovasc Dis 2019 Jul;28(7):e92 )
– metastatic lung cancer (Can J Ophthalmol 2019 Jun;54(3):e131 )
⚬ history and physical for amaurosis fugax

– ask about history of transient monocular blindness

– perform retinal exam for retinal emboli
– perform neurological exam to rule out focal neurological features

⚬ diagnostic testing for amaurosis fugax

– perform lab testing for erythrocyte sedimentation rate and c-reactive protein levels for giant cell
arteritis
– perform ultrasound of carotid arteries to detect possible carotid stenosis
– consider neuroimaging, as patients may have asymptomatic infarctions

⚬ management of amaurosis fugax dependent on underlying cause

– consider carotid endarterectomy or carotid artery stenting for carotid artery stenosis (see
Carotid Artery Stenosis Repair )
– for treatment of giant cell arteritis, see Giant Cell Arteritis (Including Temporal Arteritis)

⚬ Reference - Clin Ophthalmol 2016;10:2165 full-text , Cerebrovasc Dis 2015;40(3-4):151

STUDY
⚬ SUMMARY
18.9% reported prevalence of ≥ 70% carotid stenosis in adults with amaurosis fugax

COHORT STUDY: Clin Ophthalmol 2016;10:2165 | Full Text

Details
– based on retrospective cohort study
– 302 adults (mean age 66 years, 54% women) with amaurosis fugax were assessed by ultrasound
for carotid artery stenosis at Sahlgrenska University Hospital in Gothenburg from 2004 to 2010
– 18.9% of adults had signi cant carotid stenosis (≥ 70% stenosis); 14.2% of adults underwent
carotid endarterectomy
– 1.7% of adults displayed retinal artery emboli on examination
– factors associated with increased risk of ≥ 70% carotid stenosis

● male sex (adjusted odds ratio [OR] 2.62, 95% CI 1.26-5.46)

● current smoking (adjusted OR 6.26, 95% CI 2.62-14.93)
● diabetes (adjusted OR 3.68, 95% CI 1.37-9.9)
● prior vasculitis (adjusted OR 10.78, 95% CI 1.36-85.5)

– Reference - Clin Ophthalmol 2016;10:2165 full-text

Epidemiology

Incidence/Prevalence

● reported prevalence of TIA in United States estimated at 2.3%, with reported incidence of 0.7-0.8 per

1,000 people 1

● incidence of TIA increases with age 1

STUDY
● SUMMARY
incidence of TIA 110 per 100,000 persons in southern Ohio and northern Kentucky region of
United States in 2010

CROSS-SECTIONAL STUDY: J Stroke Cerebrovasc Dis 2019 Sep;28(9):2468

Details
⚬ based on population-based cross-sectional study
⚬ 1.3 million people in southern Ohio and northern Kentucky were assessed for TIA via Greater
Cincinnati Northern Kentucky Stroke Study in 2010
⚬ incidence of TIA in 2010

– overall, 110 per 100,000 persons

– men, 117 per 100,000 persons
– women, 102 per 100,000 persons

⚬ Reference - J Stroke Cerebrovasc Dis 2019 Sep;28(9):2468

STUDY
● SUMMARY
0.3% event rate of TIA in Oxfordshire, United Kingdom from 2002 to 2005

COHORT STUDY: Lancet 2005 Nov 19;366(9499):1773

Details
⚬ based on population-based prospective cohort study
⚬ 91,106 persons of all ages registered with 63 family physicians in 9 general medical practices in
Oxfordshire, United Kingdom were evaluated
⚬ 1,657 persons (1.8%) developed 2,024 acute vascular events including

– 918 cerebrovascular events including 618 strokes and 300 transient ischemic attacks
– 856 coronary vascular events including 159 ST-elevation myocardial infarctions, 316 non-ST-
elevation myocardial infarctions, 218 unstable angina episodes, and 163 sudden cardiac deaths
– 188 peripheral vascular events including 43 aortic events, 53 embolic visceral or limb ischemic
events, and 92 critical limb ischemia events
– 62 unclassi able deaths

⚬ 12,886 persons > 65 years old had 735 cerebrovascular events, 623 coronary events, and 147
peripheral vascular events
⚬ 5,919 persons > 75 years old had 503 cerebrovascular events, 402 coronary events, and 105
peripheral vascular events
⚬ women had higher rate of strokes than men; men had higher rates of coronary and peripheral
vascular events than women
 ⚬ Reference - OXVASC study (Lancet 2005 Nov 19;366(9499):1773 ), editorial can be found in Lancet
2005 Nov 19;366(9499):1753

Risk factors

● common risk factors

⚬ history of transient ischemic attack (TIA)

⚬ cardiovascular disease such as

– hypertension
– myocardial infarction
– atrial brillation and atrial tachyarrhythmias
– left atrial enlargement

⚬ smoking
⚬ metabolic syndrome
⚬ heavy alcohol use
⚬ diabetes mellitus type 2
⚬ high cholesterol level
⚬ carotid artery stenosis

● other metabolic conditions

⚬ obesity
⚬ prediabetes

● other modi able lifestyle factors

⚬ dietary factors

– increased sodium intake

– increased intake of processed foods and lower intake of fruits, vegetables, sh, and whole
grains
⚬ poor physical function
⚬ substance abuse

● demographic factors such as lower income

● genetic factors

● biomarkers

⚬ high C-reactive protein (CRP) level

⚬ low estimated glomerular ltration rate (GFR)
⚬ lower 25-hydroxyvitamin D concentrations
⚬ proteinuria and microalbuminuria
⚬ hypokalemia
⚬ hyperuricemia

● hormone replacement therapy

● other possible risk factors

⚬ cerebral white matter lesions on magnetic resonance imaging (MRI)

⚬ embolic signals detected by transcranial Doppler
⚬ oral contraceptives
 ⚬ environmental tobacco smoke
⚬ headache
⚬ psychiatric conditions including depression and panic disorder
⚬ stress
⚬ periodontal disease
⚬ some infections
⚬ sleep disorders
⚬ ophthalmologic disorders

● factors not associated with increased risk

⚬ dietary factors

– fat intake
– egg consumption up to 1 egg/day
– high consumption of co ee and tea

⚬ medications - progestin-only contraceptives

● see Risk Factors for Stroke or Transient Ischemic Attack for details

Etiology and Pathogenesis

Causes

● 25%-50% of TIAs are cryptogenic 1

● possible causes of TIA

⚬ large artery atherosclerosis

– carotid artery stenosis or vertebral artery stenosis

– aortic stenosis

⚬ artery-to-artery embolism
⚬ bromuscular dysplasia
⚬ arteritis - Takayasu arteritis, temporal arteritis
⚬ cardioembolic

– atrial brillation
– cardiac thrombosis
– valve disease
– impaired left ventricular function (recent myocardial infarction - see also ST-elevation
Myocardial Infarction (STEMI))
– cardiomyopathies - dilated cardiomyopathy, hypertrophic cardiomyopathy, restrictive
cardiomyopathy, peripartum cardiomyopathy, Takotsubo syndrome
– infective endocarditis
– mechanical valvular prostheses
– septal aneurysm
– myxomas
– mitral stenosis
– mitral regurgitation
– nonbacterial thrombotic endocarditis
– calci cation of mitral annulus
– intracardiac defects with paradoxical embolism (patent foramen ovale, atrial septal defect
(ASD), Eisenmenger syndrome)
 ⚬ small artery occlusion

– intracranial atherosclerosis from hypertension or diabetes

– vasospasm
– moyamoya
– reversible cerebral vasoconstriction syndrome (RCVS)
– vasculitis

⚬ arterial dissection (common in young patients)

⚬ thoracic aortic aneurysm and dissection
⚬ hypercoagulable states

– inherited thrombophilias - see also Thrombophilia in Adults

– protein C de ciency
– protein S de ciency
– antithrombin de ciency
– plasminogen de ciency
– factor V Leiden mutation
– antiphospholipid antibody syndrome (APS)
– prothrombin mutation
– thrombocytosis
– hyperhomocystinemia
– polycythemia vera
– sickle cell disease - see also Sickle Cell Disease in Adults and Adolescents
– essential thrombocythemia
– heparin-induced thrombocytopenia
– thrombotic thrombocytopenic purpura (TTP)
– disseminated intravascular coagulation (DIC)
– myeloproliferative diseases

⚬ Reference - Emerg Med Clin North Am 2016 Nov;34(4):811 , Mayo Clin Proc 2004 Aug;79(8):1071

Pathogenesis

● primary mechanism of injury is permanent (stroke) vs. transient (TIA) focal deprivation of blood ow
to cerebral parenchyma
⚬ causes

– large-artery atherosclerosis (most common cause)

● accumulations of fatty material in arterial subintima form platelet clumps which attract
brin, thrombin, and erythrocyte debris
● amassed debris coagulates to a size large enough to pose stenotic risk to cerebral
vasculature
● certain areas of vasculature may be predisposed to atherosclerotic plaque development due
to local blood ow stagnation caused by low wall shear stress
● may be related to plaque that embolizes small fragments to cerebral circulation

– cardioembolism

● most commonly, emboli form in the heart and travel to cerebral circulation by way of atrial
brillation
– systemic embolization

● fatty material from the arterial system (emboli) travel to cerebral circulation
paradoxical embolization through patent foramen ovale; see Atrial Septal Defects for details
 – small-vessel
● occlusion (lacunae)
● small, penetrating arteries which run at right angles to major arterial branches become
blocked, commonly due to microatheroma or lipohyalinosis
● see Lacunar Stroke for details

– other causes of blockages include

● acute arterial dissection

● hematologic disorders, such as sickle cell anemia
● illicit drug use, such as cocaine or amphetamines
● Moyamoya disease

⚬ resulting thrombus deprives cerebral parenchyma cells of oxygen

– insu cient cerebral tissue perfusion (< 10 mL/100 g/minute) results in ischemia, preventing
cells from functioning properly and eventually resulting in cell death
– degree of severity and irreversibility of brain damage is proportional to duration of ischemia

● clinical manifestation of ischemia depends on location of infarction within vasculature

⚬ anterior circulation (including internal carotid artery)

– anterior circulation accounts for 80% of perfusion to the brain

– infarct within anterior circulation produces symptoms consistent with distribution, for example

● unilateral ischemia of ophthalmic artery results in monocular blindness (amaurosis fugax)

● ischemia of middle cerebral artery (in dominant hemisphere) results in marked contralateral
motor and sensory disturbances, agnosia, and aphasia
⚬ posterior/vertebrobasilar circulation

– posterior/vertebrobasilar circulation supplies blood to the brainstem

– infarcts within posterior/vertebrobasilar circulation are more frequently associated with loss of
consciousness
– resulting de cits may include impairment of cerebellar functions, dysphagia, dysarthria, vertigo,
vomiting, and cranial nerve palsies
⚬ degree and duration of ischemia is inversely associated with number of perfusion pathways in a
given area; the greater the collateral circulation in the area, the less severe the ischemia as tissue is
perfused through alternative paths

● Reference - Brain Circ 2017 Apr;3(2):66 full-text , Mayo Clin Proc 2004 Aug;79(8):1071

History and Physical

History

Chief concern (CC)

● sudden onset of focal neurologic de cit 2

● most TIAs last < 1 hour, symptoms typically resolve before patient evaluated 1

● most symptoms of TIA associated with loss of function ("negative symptoms"), can include 1

⚬ motor weakness
⚬ decreased or altered speech
⚬ diminished vision
⚬ decreased sensation (anesthesia)
● dizziness (may be indicative of TIA due to posterior circulation ischemia) 1

● pattern of event may help to determine site of vascular involvement

⚬ if anterior circulation event (carotid, middle cerebral artery, anterior cerebral artery), ndings may
include
– motor dysfunction of contralateral extremities, face, or both including

● weakness
● clumsiness
● paralysis

– loss of vision in ipsilateral eye (transient monocular blindness, amaurosis fugax)

– homonymous hemianopia
– speech disturbances

● aphasia (if involvement of dominant hemisphere)

● dysarthria (can be from motor involvement of either dominant or nondominant hemisphere)

– sensory de cit of contralateral extremities, face, or both

● numbness or sensory loss

● cortical sensory loss (extinction of sensation, agnosias)

⚬ if posterior circulation event (vertebrobasilar circulation), ndings may include

– motor dysfunction of ipsilateral face and/or contralateral extremities (crossed pattern) including

● weakness
● clumsiness
● paralysis

– loss of vision of 1 or both homonymous visual elds (including cortical blindness and Anton
syndrome [lack of awareness of blindness])
– sensory de cit of ipsilateral face and/or contralateral extremities (crossed pattern)

● paresthesias
● numbness or sensory loss

– other focal neurologic ndings with posterior circulation includes associated signs and
symptoms often occurring in patterns together
● ataxia of gait, trunk, extremities
● vertigo (with other associated neurologic ndings)
● diplopia
● dysarthria
● dysphagia

⚬ Reference - Mayo Clin Proc 2004 Aug;79(8):1071

● American Stroke Association FAST signs for spotting stroke or TIA

⚬ Face dropping - Does one side of the face droop or is it numb? Ask the person to smile. Is the
person's smile uneven or lopsided?
⚬ Arm weakness - Is one arm weak or numb? Ask the person to raise both arms. Does one arm drift
downward?
⚬ Speech di culty - Is speech slurred? Is the person unable to speak or hard to understand? Ask the
person to repeat a simple sentence, like "The sky is blue." Is the person able to correctly repeat the
words?
⚬ Time to call emergency - If someone shows any of these symptoms, even if the symptoms go away,
call emergency and say, "I think this is a stroke" to help get the person to the hospital immediately.
 Time is important! Don't delay, and also note the time when the rst symptoms appeared.
Emergency responders will want to know.
⚬ other signs may include

– sudden confusion
– trouble speaking or understanding speech
– sudden numbness or weakness of face, arm or leg, especially on one side of the body
– sudden trouble seeing in one or both eyes
– sudden trouble walking
– sudden bout of dizziness or loss of balance or coordination
– sudden severe headache with no known cause

⚬ Reference - American Stroke Association

History of present illness (HPI)

● ask about nature and distribution of symptoms 1 , 2

⚬ symptoms of TIA typically focal and unilateral

⚬ nonfocal symptoms such as loss of consciousness, confusion, lightheadedness, generalized
weakness, or incontinence less often associated with TIA

● ask about onset of symptoms 1 , 2

⚬ vascular event with TIA usually sudden onset without prodrome, with maximal de cit at time of
onset
⚬ slow gradual migration of symptoms from one body part to another more frequently symptom of
migraine

● ask about duration of symptoms 1

⚬ most TIAs last < 1 hour

⚬ 50% of patients with symptoms lasting > 1 hour show radiologic evidence of infarct
⚬ consider that duration of symptoms lasting ≤ 24 hours does not accurately identify which patients

will have infarction and not TIA (AHA/ASA Class III, Level A) 3

● ask about headaches which may occur during TIA 1

● ask about recurrence of symptoms 1 , 2

⚬ recurrent stereotyped events may indicate alternative diagnosis such as seizure

⚬ be aware that "stuttering" or "crescendo" symptoms recurring over hours to days may warn of
highly unstable plaque

Medication history

● ask about history of oral contraceptive use or hormone replacement therapy, see Risk Factors for
Stroke or Transient Ischemic Attack

STUDY
● SUMMARY
interruption of aspirin therapy may be associated with increased risk of ischemic stroke or
transient ischemic attack in patients with cardiovascular risk factors for stroke

CASE-CONTROL STUDY: Arch Neurol 2005 Aug;62(8):1217

 Details
⚬ based on case-control study
⚬ 309 patients with ischemic stroke or TIA while on long-term aspirin therapy compared with 309
matched controls with similar antiplatelet therapy but no ischemic stroke in previous 6 months
⚬ aspirin therapy discontinuation in 4 weeks preceding ischemic cerebral event (cases) or interview
(controls) reported in 13 cases (4.2%) vs. 4 controls (1.3%)
⚬ Reference - Arch Neurol 2005 Aug;62(8):1217 , commentary can be found in Arch Neurol 2006
Feb;63(2):300

Past medical history (PMH)

● ask about history of

⚬ cardiovascular disease such as

– hypertension
– myocardial infarction
– atrial brillation and atrial tachyarrhythmias
– left atrial enlargement

⚬ metabolic syndrome
⚬ diabetes mellitus type 2
⚬ high cholesterol level
⚬ carotid artery stenosis
⚬ prediabetes
⚬ migraine - see also Migraine in Adults
⚬ recent head or neck trauma possibly causing cervical artery dissection
⚬ see Risk Factors for Stroke or Transient Ischemic Attack for other associated conditions

Family history (FH)

● ask about family history of stroke

● ask about family history of inherited thrombophilias 1

Social history (SH)

● ask about lifestyle factors, including

⚬ dietary factors such as high sodium intake, high consumption of processed foods or red meat
⚬ poor physical tness
⚬ smoking
⚬ heavy alcohol use
⚬ amphetamine and cocaine abuse

Physical

General physical

● physical examination likely to be normal in patient with TIA 1

● check for elevated blood pressure (Am Fam Physician 2012 Sep 15;86(6):521 , full-text )

Neck

● check for carotid bruits 1

Cardiac

● evaluate for cardiac arrhythmias 1

Neuro

● treat presence of neurologic de cits as evidence of stroke until proven otherwise 1

● perform neurologic examination, checking for focal neurologic de cits including

⚬ cranial nerve de cits

– diplopia
– hemianopia
– monocular blindness
– disconjugate gaze
– facial drooping
– lateral tongue movement
– dysphagia
– vestibular dysfunction

⚬ cerebellar system de cits

– nystagmus
– de cits in nger-to-nose or heel-to-shin movements
– past-pointing
– dystaxia
– ataxia

⚬ motor de cits in upper or lower extremities, face, or tongue, including

– spasticity
– clonus
– rigidity
– unilateral weakness

⚬ Reference - Am Fam Physician 2012 Sep 15;86(6):521 full-text

Diagnosis

Making the diagnosis

● diagnosis based on 1

⚬ transient episode of neurologic dysfunction

⚬ absence of acute infarction found on neuroimaging, preferably magnetic resonance imaging
including di usion-weighted sequences

● previous time-based diagnosis of TIA 1

⚬ reversible focal ischemic neurologic dysfunction lasting < 24 hours

⚬ generally accepted for use in absence of computed tomography or magnetic resonance imaging
availability
⚬ 30% of patients diagnosed with TIA using this de nition may in fact have infarcted brain

Differential diagnosis

● ischemic or hemorrhagic stroke 1

● other mimics of TIA include

⚬ emergent or urgent causes, including

– hypoglycemia
– hypokalemia
– hypertensive encephalopathy
– central nervous system (CNS) infection (meningitis, encephalitis)
– cerebral venous thrombosis
– chronic or subacute subdural hematoma
– CNS mass lesions
– CNS vasculitis
– CNS in ammatory disease (multiple sclerosis)
– falls or trauma
– systemic infections
– seizure
– subarachnoid hemorrhage

⚬ less urgent causes, including

– migraine
– transient global amnesia
– psychogenic causes (conversion disorder, somatization, hyperventilation)
– syncope
– acute vestibular syndrome due to peripheral cause
– peripheral nervous system lesion or compression
– drug toxicity (such as lithium, phenytoin, carbamazepine)
– vertigo (central or peripheral)

⚬ Reference - Emerg Med Clin North Am 2016 Nov;34(4):811 , Am Fam Physician 2012 Sep
15;86(6):521 full-text

STUDY
● SUMMARY
35% of patients referred for TIA have noncerebrovascular diagnosis

SYSTEMATIC REVIEW: Health Technol Assess 2014 Apr;18(27):1 | PDF

Details
⚬ based on systematic review limited by heterogeneity
⚬ systematic review of 16 studies evaluating frequency of stroke mimics in 14,542 patients referred
for suspected TIA or minor stroke
⚬ among patients referred for suspected TIA or minor stroke, proportion with noncerebrovascular
diagnosis was 34.9% (95% CI 28.4%-42%) in summary estimate of 16 studies (17 cohorts) with
14,542 patients, results limited by heterogeneity
⚬ mean proportions of most frequent noncerebrovascular diagnoses in patients referred for
suspected TIA or minor stroke
– migraine in 14.4% (10 studies)
– cardiac disorders in 8.8% (1 study)
– syncope in 6.6% (7 studies)
– seizure/epilepsy in 5% (11 studies)
– psychiatric in 4.4% (7 studies)
– vertigo in 4% (9 studies)
– peripheral nerve condition in 3.1% (8 studies)
 ⚬ other less frequent noncerebrovascular diagnoses included postural hypotension, transient global
amnesia, toxic or metabolic causes (such as drugs or hypoglycemia), brain tumor, subdural
hemorrhage, and subarachnoid hemorrhage
⚬ Reference - Health Technol Assess 2014 Apr;18(27):1 PDF

Testing overview

● perform blood tests for evaluation of patients with suspected TIA, including (AHA/ASA Class IIa, Level
B)
⚬ complete blood count with platelet count evaluating for in ammation, thrombocytosis, and
myelodysplastic disease
⚬ chemistry panel
⚬ prothrombin time (INR) and activated partial thromboplastin time (aPTT) to rule out coagulopathy
⚬ fasting lipid panel to assess for risk of atherosclerotic vascular disease
⚬ fasting glucose
⚬ consider HbA1c if diabetes suspected
⚬ cardiac biomarkers (troponin), serial measurements if acute coronary syndrome suspected

● consider specialized coagulation studies to assess for hypercoagulability in younger patients with TIA

● perform neuroimaging in patients with suspected TIA within 24 hours of symptom onset (AHA/ASA
Class I, Level B)
⚬ magnetic resonance imaging (MRI), including di usion-weighted imaging (DWI), is preferred
⚬ if MRI not available, perform computed tomography (CT)

● perform electrocardiogram as soon as possible after TIA (AHA/ASA Class I, Level B)

● consider prolonged cardiac monitoring (Holter monitor or continuous telemetry) in patients with TIA

of unclear origin after initial brain imaging and electrocardiogram AHA/ASA Class IIa, Level B 3

● perform noninvasive vascular imaging of cervical vessels (AHA/ASA Class I, Level A) and intracranial
vessels in patients with suspected TIA (AHA/ASA Class I, Level A); options include
⚬ ultrasound
⚬ computed tomographic angiography
⚬ magnetic resonance angiography

● consider transthoracic echocardiogram if TIA cause has not been identi ed (AHA/ASA Class IIa, Level

B; CSBPR Evidence Level C) 3 , 4

Clinical prediction rules

● National Institute for Heath and Care Excellence (NICE) guidelines for prompt recognition of
symptoms of stroke and transient ischemic attack in patients with sudden onset neurological
symptoms
⚬ – exclude hypoglycemia as cause of symptoms
– use validated tools to screen patients for stroke or TIA, such as

● Face Arm Speech Test (FAST) outside of hospital setting

● Recognition of Stroke in the Emergency Room (ROSIER) test for patients admitted to
emergency department
 ⚬ Reference - NICE 2019 May:NG128 PDF

● see Diagnosis in Stroke (Acute Management) for description of ROSIER and other stroke clinical
prediction tools

STUDY
● SUMMARY
ROSIER has moderate sensitivity to rule out stroke and transient ischemic attack in patients
presenting to emergency room (ER) DynaMed Level 1

COCHRANE REVIEW: Cochrane Database Syst Rev 2019 Apr 9;(4):CD011427

Details
⚬ based on Cochrane review of diagnostic cohort studies with methodological limitations
⚬ systematic review of 23 diagnostic cohort studies evaluating stroke recognition scales in
prehospital setting or ER to screen for stroke and transient ischemic attack in 9,230 adults
suspected of having stroke
– 6 studies evaluated scales in ER and 17 studies in prehospital setting (16 studies in eld and 1
study in primary care)
– reference standard was nal diagnosis of stroke or transient ischemic attack by neurologist or
stroke physician
⚬ prevalence of stroke and transient ischemic attack ranged from 16% to 92%
⚬ ranges for sensitivity and speci city reported if number of studies per test per setting was small,
risk of bias was high or uncertain, or results were highly heterogenous
⚬ performance of stroke scales to identify stroke or transient ischemic attack

– ROSIER in analysis of 5 studies in ER had

● sensitivity 88% (95% CI 84%-91%) (83% sensitivity reported in 1 high-quality trial)

● speci city range 25%-93%

– Los Angeles Prehospital Stroke Screen (LAPSS) in analysis of 5 studies in prehospital setting had

● sensitivity 83% (95% CI 75%-89%)

● speci city 93% (95% CI 88%-96%)

– Cincinnati Prehospital Stroke Scale (CPSS) in 11 studies (10 studies in prehospital setting, 1 study
in ER) had
● sensitivity range 44%-95%
● speci city range 21%-79%

– Face Arm Speech Test (FAST) in 5 studies (3 in prehospital setting and 2 in ER) had

● sensitivity range 64%-97%

● speci city range 13%-92%

– Melbourne Ambulance Stroke Screen (MASS) in 3 studies in prehospital setting had

● sensitivity range 74%-90%

● speci city range 67%-86%

– Ontario Prehospital Stroke Screening Tool (OPSS) in 1 study had

● sensitivity 92% (95% CI 88%-94%)

● speci city 86% (95% CI 80%-90%)

– Medical Prehospital Assessment for Code Stroke (MedPACS) in 1 study had

● sensitivity 74% (95% CI 67%-80%)

● speci city 33% (95% CI 27%-39%)
 – PreHospital Ambulance Stroke Test (PreHAST) in 1 study had

● sensitivity 100% (95% CI 87%-100%)

● speci city 40% (95% CI 25%-56%)

⚬ Reference - Cochrane Database Syst Rev 2019 Apr 9;(4):CD011427

⚬ Recognition of Stroke in the Emergency Room scale may help identify acute stroke in
patients presenting to emergency department
DynaMed Level 2

– based on diagnostic study with independent derivation and validation cohorts and unclear
interval between index and reference tests
– derivation cohort included 343 adults (mean age 71 years) presenting to emergency department
with suspected stroke and admitted to hospital
– 51.3% had stroke or transient ischemic attack
– 7-item ROSIER scale (range -2 to 5) developed using common clinical features of stroke in
derivation cohort
● -1 point for

⚬ loss of consciousness or syncope

⚬ seizure activity

● 1 point for new acute onset (or on awakening from sleep) of

⚬ asymmetrical facial weakness

⚬ asymmetrical arm weakness
⚬ asymmetrical leg weakness
⚬ speech disturbance
⚬ visual eld defect

– validation cohort included 160 patients (mean age 71 years) presenting to emergency
department with suspected stroke
– 63.1% had stroke or transient ischemic attack
– performance of ROSIER scale at cuto score > 0 for diagnosis of acute stroke in validation cohort
● sensitivity 93%
● speci city 83%
● positive predictive value 90%
● negative predictive value 88%

– Reference - Lancet Neurol 2005 Nov;4(11):727 , editorial can be found in Lancet Neurol 2005
Nov;4(11):691

Blood tests

● blood tests used to assess risk for TIA and to eliminate possible mimics of TIA 1

● perform blood tests for evaluation of patients with suspected TIA, including(AHA/ASA Class IIa, Level

B; CSBPR Evidence Level C) 1 , 3 , 4

⚬ complete blood count with platelet count evaluating for in ammation, thrombocytosis, and
myelodysplastic disease
⚬ chemistry panel including electrolytes, creatinine, and estimated-glomerular ltration rate to
assess for metabolic disorder or renal dysfunction
⚬ prothrombin time (INR) and activated partial thromboplastin time (aPTT) to rule out coagulopathy
⚬ fasting lipid panel to assess for risk of atherosclerotic vascular disease
 ⚬ capillary glucose level
⚬ consider HbA1c measurement if diabetes suspected

● consider specialized coagulation studies to assess for hypercoagulability in younger patients with TIA,

especially in patients without vascular risk factors or underlying identi able cause 3
⚬ activated protein C resistance/factor V Leiden
⚬ anticardiolipin antibody
⚬ D-dimer
⚬ factor VIII
⚬ brinogen
⚬ endogenous tissue plasminogen activator activity
⚬ homocysteine
⚬ lupus anticoagulant
⚬ protein C, protein S, antithrombin III activity levels
⚬ prothrombin gene G20210A mutation
⚬ plasminogen activator inhibitor-1
⚬ Von Willebrand factor

Imaging studies

● perform initial assessment with brain imaging and noninvasive vascular imaging of carotid arteries in
all patients with suspected TIA or nondisabling ischemic stroke who are not being considered for
thrombolytic or endovascular therapy (CSBPR Evidence Level B) 4

Neuroimaging

● neuroimaging used to evaluate for cerebral infarction and to rule out structural causes of symptoms

(such as subdural hematoma, intracranial hemorrhage, or brain tumor) 1 , 2

● perform neuroimaging within 24 hours of symptom onset in patients with TIA (AHA/ASA Class I, Level

B) 3
⚬ magnetic resonance imaging (MRI), including di usion-weighted imaging (DWI), is preferred brain
imaging modality
⚬ head computed tomography (CT) should be done if MRI not available

● di usion-weighted imaging MRI more sensitive than CT for di erentiating stroke from TIA 1

● National Institute for Heath and Care Excellence (NICE) guidelines

⚬ do not o er CT brain scanning to patients with suspected TIA unless clinical suspicion of alternative
diagnosis that could be detected by CT
⚬ consider MRI (including di usion-weighted and blood sensitive sequences) in patients with
suspected TIA after specialist assessment in TIA clinic to
– determine territory of ischemia
– detect possible hemorrhage
– detect alternative pathologies

⚬ Reference - NICE 2019 May:NG128 PDF

● if infarction on neuroimaging in patient with clinical presentation of TIA, classi cation changes to

nondisabling stroke 1
● perfusion CT, perfusion MRI, and uid-attenuated inversion recovery MRI techniques 1

⚬ can increase speci city and sensitivity of diagnosis

⚬ in case of stroke, can identify area of brain salvageable by reperfusion therapies

STUDY
⚬ SUMMARY
perfusion-weighted MRI abnormalities at presentation associated with eventual presence of
lesions in patients initially diagnosed TIA who had no lesions on initial diffusion-weighted
MRI

COHORT STUDY: J Clin Neurol 2017 Apr;13(2):129 | Full Text

Details
– based on cohort study
– 45 patients in prospective cohort and 42 patients in retrospective cohort with TIA and no acute
ischemic lesions on initial di usion-weighted MRI were assessed by perfusion-weighted MRI
– patients underwent di usion-weighted and perfusion-weighted multimodal MRI within 72 hours
of TIA, and follow-up di usion-weighted MRI 3 days later
– perfusion-weighted MRI abnormalities at initial MRI

● 18% of patients had focal perfusion abnormalities

● 32% of patients had territorial (large area) perfusion abnormalities

– perfusion-weighted MRI abnormalities associated with appearance of di usion-weighted MRI

lesions at 3 days follow-up with
● focal perfusion abnormalities (odds ratio [OR] 15.1, 95% CI 3.6-62.9)
● territorial perfusion abnormalities (OR 3.7, 95% CI 1.2-11.5)

– territorial perfusion abnormalities also associated with presence of relevant vessel stenosis (OR
31, 95% CI 7-137.9)
– Reference - J Clin Neurol 2017 Apr;13(2):129 full-text

● review of brain imaging in TIA can be found in J Neuroradiol 2015 Feb;42(1):3

● see Neuroimaging for Acute Stroke for further descriptions

Noninvasive vascular imaging

● carotid ultrasonography or transcranial Doppler ultrasonography, magnetic resonance angiography,

and computerized tomography angiography used to assess for treatable atherosclerotic lesions (Am
Fam Physician 2012 Sep 15;86(6):521 ) full-text

● American Heart Association/American Stroke Association (AHA/ASA) recommendations 3

⚬ perform noninvasive imaging of cervical vessels as part of routine evaluation of patients with
suspected TIA (AHA/ASA Class I, Level A)
– initial evaluation of extracranial vasculature may include (AHA/ASA Class IIa, Level B)

● carotid ultrasound or transcranial Doppler

● magnetic resonance angiography
● computerized tomography angiography
● speci c test may depend on availability of studies and expertise, and patient variables

– con rm consistent ndings in 2 di erent kinds of noninvasive studies if only noninvasive testing
is performed prior to carotid endarterectomy, otherwise consider catheter angiography
 (AHA/ASA Class IIa, Level B)
⚬ perform noninvasive imaging of intracranial vessels if knowledge of intracranial stenosis or
occlusion will change management (AHA/ASA Class I, Level A)
– computed tomography angiography (CTA) or magnetic resonance angiography (MRA) can
exclude proximal intracranial stenosis and/or occlusion
– reliable determination or presence and degree of intracranial stenosis requires invasive
catheter angiography for con rmation
– phrasing for this recommendation revised in 2013 (Stroke 2013 Mar;44(3):870 )
⚬ signi cance of plaque characteristics and detection of microembolic signals not yet clear (AHA/ASA
Class IIb, Level B)

● Canadian Stroke Best Practice Recommendations (CSBPR) 4

⚬ conduct extracranial vascular imaging to identify extracranial carotid stenosis; if positive, refer
patient for possible carotid revascularization (CSBPR Evidence Level A)
⚬ conduct CTA at time of brain CT to assess extracranial and intracranial circulation (CSBPR Evidence
Level B)
⚬ CTA recommended for visualizing intracranial circulation, posterior circulation, and aortic arch to
identify stroke etiology and guide management (CSBPR Evidence Level C)
⚬ consider alternatives to CTA including MRA or carotid ultrasound (for extravascular imaging)
depending on clinical situation and immediate availability (CSBPR Evidence Level C)
⚬ for CTA or MRA, consider including extracranial and intracranial vasculature ("aortic arch-to-
vertex") (CSBPR Evidence Level C)

● NICE guidelines recommend urgent carotid imaging in patients with TIA considered candidates for
carotid endarterectomy (NICE 2019 May:NG128 PDF )

Cardiac imaging

● transthoracic echocardiography can be used to detect possible 1

⚬ cardiac hypertrophy
⚬ ventricular hypokinesis or thrombosis
⚬ mitral stenosis
⚬ endocarditis

● consider transthoracic echocardiogram if TIA cause has not been identi ed (AHA/ASA Class IIa, Level

B; CSBPR Evidence Level C) 3 , 4

● consider transesophageal echocardiography if identi cation of following conditions will alter

management (AHA/ASA Class IIa, Level B) 3

⚬ patent foramen ovale
⚬ aortic arch atherosclerosis
⚬ valvular disease

● American Society of Neuroradiology/American College of Radiology/Society of NeuroInterventional

Surgery (ASNR/ACR/SNIS) imaging recommendations for TIA are consistent with above
recommendations (AJNR Am J Neuroradiol 2013 Nov-Dec;34(11):E117 full-text ), commentary can
be found in AJNR Am J Neuroradiol 2013 Nov-Dec;34(11):2053
● ACR Appropriateness Criteria for cerebrovascular disease can be found at ACR 2016 PDF

Electrocardiography (ECG)

● ECG used to assess possible cardioembolic mechanism of TIA, such as atrial brillation, ventricular

hypertrophy, or cardiac ischemia 1

● conduct ECG in all patients with suspected TIA or nondisabling ischemic stroke to (CSBPR Evidence

Level C) 4
⚬ assess baseline cardiac rhythm
⚬ look for evidence of structural heart disease such as previous myocardial infarction or left
ventricular hypertrophy

● conduct ECG as soon as possible after suspected TIA AHA/ASA Class I, Level B 3

● consider prolonged cardiac monitoring in select patients

⚬ prolonged cardiac monitoring (inpatient telemetry or Holter monitoring) is reasonable in patients

with etiology not identi ed after initial neuroimaging and ECG (AHA/ASA Class IIa, Level B) 3
⚬ prolonged cardiac monitoring (up to 30 days) is recommended to assess for paroxysmal atrial
brillation if no signs of atrial brillation on 24- to 48-hour ECG monitoring and cardioembolic
mechanism suspected (CSBPR Evidence Level B) 4

Management

Management overview

● evaluate patients with suspected TIA as soon as possible after event

● consider diagnosis of stroke rather than TIA if patient does not return to baseline, see Stroke (Acute
Management)

● appropriate care setting depends on timing of symptoms and patient's risk for stroke

⚬ triage similar to patient with stroke if presenting within 3 hours of symptom onset
⚬ exclude hypoglycemia as cause of symptoms
⚬ use validated tools such as Face Arm Speech Test (FAST) or Recognition of Stroke in the Emergency
Room (ROSIER) test to screen and diagnose patients for stroke or TIA; exclude hypoglycemia as
cause of symptoms
⚬ transport to stroke-ready emergency department and strongly consider admission to hospital or
observation unit to expedite workup if symptoms occurred > 3 hours but < 24 hours ago
⚬ reasonable to hospitalize patients with TIA if presentation within 72 hours of event and

– unclear if outpatient workup can be completed within 2 days

– suspected focal ischemia

⚬ perform same-day evaluation for TIA or minor stroke associated with reduced risk of recurrent
stroke at 90 days if not admitting patient to hospital

● acute management should include

⚬ aspirin 81-325 orally mg once daily (unless contraindicated) to be started immediately

 ⚬ immediate referral for patients with suspected TIA for specialist assessment and investigation
within 24 hours of symptom onset
⚬ additional diagnostic evaluation, including brain imaging (magnetic resonance imaging preferred),
vascular imaging (if carotid involvement suspected), blood tests, electrocardiography, and risk
factor assessment

● implement measures to prevent stroke following diagnosis of TIA, including

⚬ risk factor reduction through statin therapy, blood pressure reduction, smoking cessation, and
avoidance of heavy alcohol use; consider advising moderate physical activity
⚬ antiplatelet agents recommended over anticoagulation for patients with nonembolic TIA
⚬ in patients with atrial brillation, oral anticoagulation recommended
⚬ in patients with carotid artery stenosis

– perform carotid endarterectomy (CEA) for patients with a perioperative morbidity and mortality
risk estimated to be < 6% who have had a TIA or nondisabling ischemic stroke within the last 6
months and have either
● ipsilateral severe (70%-99%) stenosis documented by noninvasive imaging
● ipsilateral moderate (50%-69%) stenosis documented by catheter-based imaging

– consider carotid artery stenting (CAS) as an alternative to CEA for symptomatic patients with an
average or low risk of complications associated with endovascular intervention plus stenosis >
70% by noninvasive imaging or > 50% by catheter angiography

Acute management

● evaluate patients with suspected TIA

⚬ as soon as possible after event (AHA/ASA Class I, Level B) 3

⚬ using healthcare professional with expertise in stroke care to determine risk for recurrent stroke

and begin appropriate testing and management 4

● consider diagnosis of stroke rather than TIA if patient does not return to baseline, see Stroke (Acute
Management)

● use validated tools to screen patients for stroke or TIA, such as Face Arm Speech Test (FAST) or
Recognition of Stroke in the Emergency Room (ROSIER) test; exclude hypoglycemia as cause of
symptoms (NICE 2019 May:NG128 PDF )

● o er antiplatelet agent (usually aspirin 81-325 mg orally once daily) to all patients with TIA except

those being anticoagulated for atrial brillation 2

● National Institute for Heath and Care Excellence (NICE) guidelines for initial management of suspected
and con rmed TIA
⚬ immediately o er aspirin 300 mg orally once daily (unless contraindicated) in patients with
suspected TIA
⚬ immediately refer patients with suspected TIA for specialist assessment and investigation within 24
hours of symptom onset
⚬ do not use scoring systems such as ABCD2 score to assess risk of subsequent stroke or inform
urgency for referral in patients with suspected or con rmed TIA
– risk prediction scores without imaging parameters used in isolation poor at discriminating risk
of stroke
 – appropriate imaging not available in general practice or for paramedics

⚬ o er secondary prevention (in addition to aspirin) as soon as possible after diagnosis of TIA
con rmed
⚬ Reference - NICE 2019 May:NG128 PDF

● appropriate care setting depends on timing of symptoms and patient's risk for stroke

⚬ Canadian Stroke Best Practice Recommendations (CSBPR) 4

– for presentation ≤ 48 hours after suspected TIA or nondisabling ischemic stroke

● if transient, uctuating, and/or persistent unilateral weakness (face, arm, and/or leg), or
speech disturbance
⚬ patient is considered HIGHEST RISK for recurrent stroke (CSBPR Evidence Level B)
⚬ immediately send patient to emergency department with capacity for advanced stroke
care including on-site brain/vascular imaging and, if possible, access to tissue
plasminogen activator (tPA) (CSBPR Evidence Level C)
⚬ conduct urgent brain and noninvasive vascular imaging (CSBPR Evidence Level B)

– for brain imaging, use computed tomography (CT) or magnetic resonance imaging
(MRI)
– for vascular imaging, use CT or MR angiography from arch to vertex

⚬ conduct urgent electrocardiogram (CSBPR Evidence Level B)

● if transient, uctuating, or persistent symptoms (such as hemibody sensory loss, acute

monocular visual loss, or binocular diplopia) without motor weakness or speech disturbance
⚬ patient is considered HIGH RISK for recurrent stroke (CSBPR Evidence Level C)
⚬ send patient for same-day assessment at closest stroke prevention clinic or emergency
department with capacity for advanced stroke care (CSBPR Evidence Level B)
– for presentation 2-14 days after suspected TIA or nondisabling ischemic stroke

● if transient, uctuating, and/or persistent unilateral weakness (face, arm, and/or leg), or
speech disturbance
⚬ patient is considered at INCREASED RISK for recurrent stroke (CSBPR Evidence Level B)
⚬ patient should have comprehensive assessment by healthcare professional with stroke
expertise as soon as possible and ≤ 24 hours after rst contact with healthcare system
(CSBPR Evidence Level B)
● if transient, uctuating, or persistent symptoms (such as hemibody sensory loss, acute
monocular visual loss, or binocular diplopia) without motor weakness or speech disturbance
⚬ patient is considered at INCREASED RISK for recurrent stroke (CSBPR Evidence Level C)
⚬ patient should have comprehensive assessment by healthcare professional with stroke
expertise as soon as possible and ≤ 14 days after rst contact with healthcare system
(CSBPR Evidence Level B)
– for presentation > 14 days after suspected TIA or nondisabling ischemic stroke (CSBPR Evidence
Level C)
● patient is considered at LOWER risk for recurrent stroke
● patient should be assessed by neurologist or stroke specialist as soon as possible and ≤ 30
days after symptoms onset
– for atypical sensory symptoms (such as patchy numbness and/or tingling) (CSBPR Evidence
Level C)
● patient is considered at LOWER risk for recurrent stroke
● patient should be assessed by stroke specialist as required
⚬ consider hospitalizing patients with TIA if presenting within 72 hours of event with any of 3

– unclear if outpatient workup can be completed within 2 days

– other evidence indicates event caused by focal ischemia

⚬ e cacy for inpatient vs. outpatient management

STUDY
– SUMMARY
if not admitting patient to hospital, same-day evaluation for transient ischemic attack or
minor stroke associated with reduced risk of recurrent stroke at 90 days DynaMed Level 2

BEFORE AND AFTER STUDY: Lancet 2007 Oct 20;370(9596):1432

Details
● based on before-and-after study
● 591 patients not admitted to hospital for TIA or minor stroke were evaluated before and after
government policy change
⚬ before April 2002 patients seen in outpatient clinic (including scheduling delays in
obtaining appointments and referrals)
⚬ between 2002 and 2004 patients were seen mostly on same day and had neuroimaging
same day or soon afterward
● median delay to assessment in study clinic decreased from 3 days to < 1 day (p < 0.0001)
● median delay to rst prescription of treatment decreased from 20 days to 1 day (p < 0.0001)
● 90-day risk of recurrent stroke was 10.3% before vs. 2.1% after policy change (p = 0.0001,
NNT 13)
● Reference - EXPRESS study (Lancet 2007 Oct 20;370(9596):1432 ), editorial can be found in
Lancet 2007 Oct 20;370(9596):1398
● policy change associated with reduced fatal or disabling stroke (p = 0.0005), hospital bed-
days (p = 0.017), and acute costs (p = 0.028) in subsequent multivariate analysis (Lancet
Neurol 2009 Mar;8(3):235 ), editorial can be found in Lancet Neurol 2009 Mar;8(3):218

STUDY
– SUMMARY
outpatient management of low-risk TIA (ABCD2 score < 4) associated with 1.5% stroke
rate within 90 days

COHORT STUDY: J Clin Neurosci 2014 Jan;21(1):47

Details
● based on prospective cohort study
● 200 patients (mean age 68 years) with initial diagnosis of low-risk TIA were treated with
antiplatelet therapy and discharged with management instructions and referral for medical
evaluation within 7 days, and were followed for 90 days
● low-risk TIA de ned by

⚬ ABCD2 score < 4 (111 patients)

⚬ absence of high-risk features such as known carotid artery disease, atrial brillation, or
crescendo TIA
⚬ absence of hemorrhage on computed tomography
⚬ neurologist judgment (65 patients with ABCD2 score ≥ 4)

● 11 patients lost to follow-up

 ● nal diagnosis in 150 patients with follow-up by neurologist
⚬ stroke in 18 patients
⚬ probable or possible TIA in 70 patients
⚬ other diagnosis in 40 patients, most frequently, anxiety, migraine, transient global
amnesia, or presyncope
⚬ clinical impression not recorded in 72 patients

● 3 patients (1.5%) had stroke within 90 days

● Reference - J Clin Neurosci 2014 Jan;21(1):47

– arguments for and against hospital admission for TIA or minor stroke can be found in Stroke
2006 Apr;37(4):1137 , Stroke 2006 Apr;37(4):1139

● if not already obtained, diagnostic testing should include 3

⚬ brain imaging with MRI (preferred) or CT within 24 hours of symptom onset

⚬ noninvasive vascular imaging
⚬ blood tests
⚬ electrocardiography with telemetry
⚬ echocardiography (if cardioembolic stroke suspected)

Prevention of stroke after TIA

● implement secondary prevention measures to prevent stroke as soon as diagnosis of TIA con rmed
(Royal College of Physicians 2016 PDF )

● most evidence and recommendations for reducing the risk of stroke in patients with recent TIA are
based on studies including patients with minor stroke or TIA, or with underlying atrial brillation or
carotid artery stenosis
⚬ for a primary evidence and a complete discussion of strategies, see

– Secondary Prevention of Stroke or Transient Ischemic Attack

– Antiplatelet Therapy for Secondary Prevention of Stroke or Transient Ischemic Attack
– Thromboembolic Prophylaxis in Atrial Fibrillation
– Carotid Artery Stenosis Repair

⚬ a short overview of strategies is presented below

● antiplatelet agents recommended over anticoagulation for noncardioembolic TIA; options for most
patients include:
⚬ combination aspirin plus clopidogrel for up to 21 days, with single antiplatelet thereafter
⚬ aspirin 81-325 mg orally once daily
⚬ combination aspirin 25 mg plus extended-release dipyridamole 200 mg orally once daily
(Aggrenox)
⚬ clopidogrel 75 mg orally once daily
⚬ see Antiplatelet Therapy for Secondary Prevention of Stroke or Transient Ischemic Attack for a
complete discussion of strategies and graded recommendations from professional organizations

● thromboembolic prophylaxis in patients with atrial brillation and TIA

⚬ use oral anticoagulation over antiplatelet therapy or no therapy for patients with high risk for
stroke based on
– history of cardioembolic stroke or TIA
 – CHA2DS2-VASc score ≥ 2 in men and score ≥ 3 in women
– valvular atrial brillation (AF)

⚬ see Thromboembolic Prophylaxis in Atrial Fibrillation for a complete discussion of strategies and
graded recommendations from professional organizations

● management for patients with symptomatic carotid artery stenosis and TIA

⚬ consider revascularization within 2 weeks of symptomatic TIA for patients without

contraindications to early intervention
⚬ perform carotid endarterectomy (CEA) for patients with a perioperative morbidity and mortality
risk estimated to be < 6% who have had a TIA or nondisabling ischemic stroke within the last 6
months and have either
– ipsilateral severe (70%-99%) stenosis documented by noninvasive imaging
– ipsilateral moderate (50%-69%) stenosis documented by catheter-based imaging

⚬ consider carotid artery stenting (CAS) as an alternative to CEA for symptomatic patients with an
average or low risk of complications associated with endovascular intervention plus stenosis > 70%
by noninvasive imaging or > 50% by catheter angiography
⚬ see Carotid Artery Stenosis Repair for a complete discussion of strategies and graded
recommendations from professional organizations

● other risk factor reduction strategies

⚬ consider advising moderate-to-vigorous aerobic physical activity for an average of 40 minutes 3-4
times/week
⚬ use statin therapy for TIA of presumed atherosclerotic origin or evidence of aortic arch atheroma
⚬ initiate blood pressure therapy several days after TIA in previously untreated patients with blood
pressure ≥ 140/90 mm Hg or previously treated patients with known hypertension
⚬ other lifestyle changes

– smoking cessation
– reduction or discontinuation of alcohol consumption of alcohol if history of heavy drinking
– reduction in sodium intake to < 2.4 g/day, or < 1.5 g/day for greater blood pressure reduction
– Mediterranean-type diet (instead of a low-fat diet) that emphasizes vegetables, fruits, and whole
grains and includes low-fat dairy products, poultry, sh, legumes, olive oil, and nuts, and limits
sweets and red meats
⚬ see Secondary Prevention of Stroke or Transient Ischemic Attack for a complete discussion of
strategies and graded recommendations from professional organizations

Follow-up

● if clinical evidence of ischemic stroke and patient not admitted to hospital, assess for functional

impairment when appropriate, including (CSBPR Evidence Level B) 4

⚬ cognitive evaluation
⚬ depression
⚬ tness to drive
⚬ assessments for potential rehabilitation treatment

Complications

● stroke

⚬ 3%-10% risk of stroke in rst 48 hours after TIA

 ⚬ 5%-17% risk of stroke in rst 90 days after TIA

● see Prognosis for details

EVIDENCE SYNOPSIS

TIA and ischemic stroke are associated with an increased long-term risk for myocardial infarction
and vascular death

STUDY
⚬ SUMMARY
0.9% annual risk of myocardial infarction in patients with TIA or ischemic stroke after 2005

SYSTEMATIC REVIEW: J Am Heart Assoc 2018 Jan 18;7(2):

Details
– based on systematic review of observational studies
– systematic review of 25 observational studies and observational data from 26 randomized
trials evaluating association between TIA or ischemic stroke and myocardial infarction in
131,299 patients followed for 1-10 years
– annual risk of myocardial infarction after TIA or ischemic stroke 1.67% (95% CI 1.36%-1.98%) in
analysis of 46 studies with 100,786 patients
● risk of myocardial infarction after TIA or ischemic stroke signi cantly decreased comparing
newer studies to older studies (p < 0.015)
⚬ 0.9% (95% CI 0.49%-1.32%) in analysis of 6 studies with 5,914 patients enrolled after
2005
⚬ 1.53% (95% CI 1.14%-1.91%) in analysis of 26 studies with 83,884 patients enrolled 1990-
2005
⚬ 2.32% (95% CI 1.63%-3%) in analysis of 14 studies with 10,988 patients enrolled prior to
1990
● factors associated with increased risk of myocardial infarction after TIA or ischemic stroke

⚬ male sex (rate ratio [RR] 1.89, 95% CI 1.32-2.74) in analysis of 3 studies with 1,393
patients
⚬ hypertension (RR 1.53, 95% CI 1.07-2.23) in analysis of 3 studies with 1,393 patients
⚬ coronary artery disease (RR 2.32, 95% CI 2.01-2.68) in analysis of 5 studies with 26,493
patients
⚬ peripheral artery disease (RR 2.91, 95% CI 1.29-5.68) in analysis of 2 studies with 1,005
patients
– annual risk of other events after TIA or ischemic stroke

● recurrent stroke 4.26% (95% CI 3.43%-5.09%) in analysis of 34 studies with 73,184 patients
● cardiac death 1.38% (95% CI 0.96%-1.81%) in analysis of 11 studies with 15,050 patients
● vascular death 2.17% (95% CI 1.75%-2.59%) in analysis of 38 studies with 76,501 patients

– Reference - J Am Heart Assoc 2018 Jan 18;7(2):

STUDY
⚬ SUMMARY
average annual incidence of myocardial infarction after TIA reported to be 0.95%
 COHORT STUDY: Stroke 2011 Apr;42(4):935 | Full Text

Details
– based on cohort study
– 456 patients in United States with TIA between 1985 and 1994 in Olmsted County, Rochester,
Minnesota were followed for incidence of myocardial infarction with median follow-up of 10.2
years
– increased risk of myocardial infarction after TIA compared to general population

● in all patients (relative risk 2.09, 95% CI 1.52-2.81)

● in patients < 60 years old (relative risk 15.1, 95% CI 4.11-38.9)

– factors associated with increased risk of myocardial infarction after TIA

● use of lipid-lowering therapy (hazard ratio [HR] 3.1 95% CI 1.2-8)

● male sex (HR 2.19, 95% CI 1.14-2.01)
● increased age (per 10 years) (HR 1.51, 95% CI 1.14-2.01)

– Reference - Stroke 2011 Apr;42(4):935 full-text

STUDY
● SUMMARY
posttraumatic stress disorder (PTSD) reported in about 10%-25% of patients following stroke or
transient ischemic attack

SYSTEMATIC REVIEW: PLoS One 2013;8(6):e66435 | Full Text

Details
⚬ based on systematic review of 9 studies reporting prevalence of PTSD after stroke or transient
ischemic attack in 1,138 patients
⚬ prevalence of PTSD after stroke or transient ischemic attack (all results limited by signi cant
heterogeneity)
– 13% (95% CI 11%-16%) overall
– 23% (95% CI 16%-33%) within 1 year of event in analysis of 8 studies
– 11% (95% CI 8%-14%) after 1 year of event in analysis of 2 studies

⚬ Reference - PLoS One 2013;8(6):e66435 full-text

STUDY
● SUMMARY
TIA associated with ongoing fatigue, psychological impairment, and cognitive impairment

COHORT STUDY: Eur J Neurol 2016 Nov;23(11):1642

Details
⚬ based on retrospective cohort study
⚬ 9,419 patients with TIA (median age 74 years) and 46,511 age-matched controls from The Health
Improvement Network (THIN) database in United Kingdom were followed for up to 60 months
⚬ patients with fatigue, psychological impairment (depression, anxiety, or PTSD), or cognitive
impairment at index date were excluded from analysis
⚬ comparing patients with TIA vs. controls, TIA associated with increased risk of

– fatigue in 9.8% vs. 5.8% (adjusted hazard ratio [HR] 1.43, 95% CI 1.33-1.54)
– psychological impairment in 34% vs. 24% (adjusted HR 1.26, 95% CI 1.2-1.31)
 – cognitive impairment in 3.8% vs. 2.2% (adjusted HR 1.45, 95% CI 1.28-1.65)

⚬ Reference - Eur J Neurol 2016 Nov;23(11):1642

STUDY
● SUMMARY
mild cognitive impairment reported in 29%-68% and severe cognitive impairment reported in
8%-13% of patients 5 years after TIA

SYSTEMATIC REVIEW: Cerebrovasc Dis 2016;42(1-2):1 | Full Text

Details
⚬ based on systematic review of observational studies
⚬ systematic review of 2 cross-sectional studies, 1 case control study, 1 case series, and 9 cohort
studies reporting prevalence of cognitive impairment in patients with TIA
⚬ cognitive function assessed using a neuropsychological test battery (NPS), Mini-Mental State
Examination (MMSE), Montreal Cognitive Assessment (MoCA), or modi ed Telephone Interview for
Cognitive Status (TICSm)
⚬ prevalence of mild cognitive impairment

– 36%-57% from 1 week to 1 month after TIA in 5 studies

– 30%-57% from 4 months to 1 year after TIA in 5 studies
– 29%-68% at 5 years after TIA in 2 studies

⚬ prevalence of severe cognitive impairment after TIA was 22% at 6 months in 1 study, 8%-13% at 5
years in 2 studies
⚬ Reference - Cerebrovasc Dis 2016;42(1-2):1 full-text

Prognosis

Risk of stroke after TIA

● 20% of patients with ischemic stroke present with TIA in hours or days preceding stroke 2

STUDY
● SUMMARY
stroke risk 5.2% at 7 days, 6.7% at 90 days, and 11.3% at > 90 days after TIA

SYSTEMATIC REVIEW: Health Technol Assess 2014 Apr;18(27):1 | Full Text

Details
⚬ based on systematic review
⚬ systematic review of 53 studies evaluating risk of stroke after TIA in 30,558 patients
⚬ pooled cumulative stroke rate

– 5.2% at 7 days (95% CI 3.9%-5.9%) in analysis of 28 studies with 12,332 patients who had 974
strokes
– 6.7% at 90 days (95% CI 5.2%-8.7%) in analysis of 34 studies with 19,803 patients who had 1,567
strokes
– 11.3% after 90 days (95% CI 7.5%-16.6%) in analysis of 8 studies with 8,699 patients who had 927
strokes, follow-up ranged from 6 months to 14 years
⚬ Reference - Health Technol Assess 2014 Apr;18(27):1 full-text
⚬ similar 7-day results in systematic review of 18 cohort studies with 10,126 patients (Lancet Neurol
2007 Dec;6(12):1063 )
STUDY
● SUMMARY
stroke incidence 1.5%-5.1% within 1 year after TIA and 9.5% at 5 years

COHORT STUDY: N Engl J Med 2016 Apr 21;374(16):1533 | Full Text

Details
⚬ based on prospective cohort study
⚬ 4,583 patients with TIA or minor stroke in previous 7 days were followed for up to 5 years
⚬ 91% completed ≥ 1 year of follow-up and were included in analysis
⚬ stroke incidence

– 1.5% at 2 days
– 2.1% at 7 days
– 2.8% at 30 days
– 3.7% at 90 days
– 5.1% at 1 year

⚬ factors associated with increased risk of stroke at 1 year included multiple infarctions on brain
imaging, large-artery atherosclerosis, and ABCD2 score 6-7 (p < 0.005 for each)
⚬ Reference - N Engl J Med 2016 Apr 21;374(16):1533 full-text , editorial can be found in N Engl J
Med 2016 Apr 21;374(16):1577
⚬ stroke incidence 9.5% at 5-year follow-up of trial above (3,847 patients) with increased risk
associated with ipsilateral large-artery atherosclerosis, cardioembolism, and baseline ABCD2 score
≥ 4 (N Engl J Med 2018 Jun 7;378(23):2182 )

STUDY
● SUMMARY
8%-12% risk of stroke over 1 year in patients with intracranial atherosclerotic disease

COHORT STUDY: Arch Neurol 2008 Jun;65(6):733

Details
⚬ based on cohort study of patients from WASID study
⚬ 569 patients with TIA or nondisabling stroke within past 3 months who had corresponding
50%-99% stenosis of major intracranial artery on angiography were followed for 2 years
⚬ risk of ischemic stroke in arterial territory in patients with 50%-99% stenosis

– 90-day risk

● 6.9% after TIA

● 4.7% after stroke

– 1-year risk

● 7.9% after TIA

● 11.7% after stroke

– 2-year risk

● 10.7% after TIA

● 15.5% after stroke

⚬ Reference - Arch Neurol 2008 Jun;65(6):733

Risk scores for predicting stroke after TIA

● ABCD2 score is most widely used tool for risk strati cation of patients for stroke after TIA 1
● National Institute for Heath and Care Excellence (NICE) guidelines for initial management of suspected
and con rmed TIA do not recommend use of scoring systems such as ABCD2 score to assess risk of
subsequent stroke or inform urgency for referral in patients with suspected or con rmed TIA (NICE
2019 May:NG128 PDF )

● ABCD2 score calculated by

⚬ addition of 5 di erent factors

– A for Age 60 years or older (1 point)

– B for Blood pressure with systolic blood pressure > 140 mm Hg and/or diastolic blood pressure
> 90 mm Hg (1 point)
– C for Clinical features (2 points for unilateral weakness, 1 point for speech disturbance without
weakness)
– D for Duration of symptoms (2 points for 60 minutes or more, 1 point for 10-59 minutes)
– D for Diabetes (1 point)

⚬ score of 7-6 considered high risk of stroke

⚬ score of 4-5 considered moderate risk
⚬ score 0-3 considered low risk
⚬ Reference - Lancet 2007 Jan 27;369(9558):283 , editorial can be found in Lancet 2007 Jan
27;369(9558):251 , commentary can be found in Lancet 2007 Mar 31;369(9567):1082 , ACP J
Club 2007 May-Jun;146(3):79 , Am Fam Physician 2007 Sep 15;76(6):865

● limitations of ABCD2 score include 1

⚬ approximately 20% of patients with low-risk score have high-risk etiology such as atrial brillation
or carotid stenosis
⚬ up to 41% of patients with high score (> 4) may have TIA mimic instead of TIA
⚬ clinical measurements of score favor anterior circulation strokes, may miss posterior circulation
ndings such as dizziness or sensory loss

● newer iterations of scoring incorporate 1

⚬ brain imaging ndings (ABCD2-I)

⚬ previous TIAs, brain imaging ndings, and presence of carotid artery stenosis (ABCD3-I

● prognostic performance of risk scores

EVIDENCE SYNOPSIS

The ABCD2 score may help to identify the risk of stroke following a TIA, but it cannot be relied
upon in isolation to make clinical decisions.

STUDY
– SUMMARY
ABCD2 score cutoff ≤ 3 (low risk)/4-7 (high risk) has moderate-to-high sensitivity but
low specificity to identify patients at low risk of stroke 2, 7, and 90 days after TIA
DynaMed Level 1
SYSTEMATIC REVIEW: Neurology 2015 Jul 28;85(4):373
SYSTEMATIC REVIEW: Health Technol Assess 2014 Apr;18(27):1 | Full Text
SYSTEMATIC REVIEW: Neurology 2012 Sep 4;79(10):971

Details
● based on 3 systematic reviews
● systematic review of 29 studies evaluating use of ABCD2 score with cuto ≥ 4 for predicting
risk of stroke recurrence in 13,766 patients with TIA or mild stroke
⚬ stroke recurrence was assessed at 7 days (17 studies), at 90 days (22 studies), or at > 90
days after TIA or mild stroke (3 studies)
⚬ 26 studies included only patients with de nite or con rmed TIA by neurologist or stroke
physician and excluded TIA mimics
⚬ 10 studies with 4,443 patients evaluated stroke at both 7 days and 90 days after TIA or
minor stroke, pooled analyses limited by signi cant heterogeneity

Table 1. Stroke Rates at Days 7 and 90 After TIA in Patients Classified

Using ABCD2 Score

Time Score ≥ 4 Score < 4

7 days 5.2% (95% CI 1.4% (95% CI

2.8%-9.4%) 0.7%-3.1%)

90 days 8.9% (95% CI 2.4% (95% CI

5.3%-14.5%) 1.3%-4.4%)

⚬ diagnostic performance of ABCD2 score with cuto ≥ 4 for predicting high risk of stroke
recurrence at 7 days
– sensitivity 86.7% (95% CI 81.4%-90.7%)
– speci city 35.4% (95% CI 33.3%-38.3%)

⚬ similar prognostic performance at 90 days after TIA

⚬ Reference - Neurology 2015 Jul 28;85(4):373 , editorial can be found in Neurology 2015
Jul 28;85(4):304
● systematic review of 26 studies evaluating use of ABCD2 score to predict risk of stroke after
TIA or minor stroke in 12,586 patients
⚬ 9 studies with 4,291 patients evaluated stroke at both 7 days and 90 days after TIA or
minor stroke, pooled analyses limited by signi cant heterogeneity

Table 2. Pooled Risk of Stroke Based on ABCD2 Score

Time Score ≥ 4 Score < 4

 Time Score ≥ 4 Score < 4

7 days 4.7% (95% CI 1.6% (95% CI

2.4%-8.7%) 1%-3.4%)

90 days 8.2% (95% CI 2.7% (95% CI

4.7%-14%) 1.5%-4.7%)

⚬ Reference - Health Technol Assess 2014 Apr;18(27):1 full-text

● systematic review of 33 studies evaluating ABCD2 score for predicting risk of stroke at 2, 7,
or 90 days after TIA in 16,070 patients
⚬ pooled stroke prevalence was 4.3% at 7 days
⚬ pooled performance of ABCD2 score for prediction of stroke at 7 days after TIA

– with cuto ≤ 3 (low risk)/4-7 (high risk)

– sensitivity 89% (95% CI 87%-91%)
– speci city 34% (95% CI 33%-35%)
– positive predictive value 8% (95% CI 7%-9%)
– negative predictive value 98% (95% CI 98%-98%)

– with cuto ≤ 4 (low risk)/5-7 (high risk)

– sensitivity 66% (95% CI 62%-70%)
– speci city 59% (95% CI 58%-60%)
– positive predictive value 9% (95% CI 8%-10%)
– negative predictive value 96% (95% CI 96%-97%)

⚬ very similar performance at 2 days and 90 days after TIA

⚬ assuming 5% baseline stroke risk

– ABCD2 > 3 associated with 2% increase in 7-day stroke risk

– ABCD2 ≤ 3 associated with 2.9% decrease in 7-day stroke risk

⚬ Reference - Neurology 2012 Sep 4;79(10):971 , editorial can be found at Neurology

2012 Sep 4;79(10):958 , commentary can be found in Ann Intern Med 2013 Apr
16;158(8):JC12

STUDY
– SUMMARY
as ABCD2 score cutoff increases, the sensitivity of the score decreases and specificity
increases with no cutoff having both high sensitivity and specificity DynaMed Level 2

COHORT STUDY: CMAJ 2011 Jul 12;183(10):1137 | Full Text

Details
● based on prospective cohort study without blinding of carers to ABCD2 score

● 2,056 patients (mean age 68 years) presenting to emergency department with TIA were
assessed for ABCD2 score and followed for 90 days
● of 2,032 patients included in analysis, 38 patients (1.9%) had stroke at 7 days and 65
patients (3.2%) had stroke at 90 days
● ABCD2 scores > 2 had high sensitivity but low speci city
 ⚬ 94.7% sensitivity and 12.5% speci city for stroke at 7 days
⚬ 96.9% sensitivity and 12.7% speci city for stroke at 90 days

● ABCD2 score > 3 had moderate speci city

⚬ 92.1% sensitivity and 32.7% speci city for stroke at 7 days

⚬ 92.3% sensitivity and 33% speci city for stroke at 90 days

● ABCD2 score > 4 had moderate sensitivity and speci city

⚬ 65.8% sensitivity and 57.2% speci city for stroke at 7 days

⚬ 63.1% sensitivity and 57.4% speci city for stroke at 90 days

● ABCD2 score > 5 had low sensitivity (31.6% at 7 days and 29.2% at 90 days)

● Reference - CMAJ 2011 Jul 12;183(10):1137 full-text

STUDY
– SUMMARY
20% patients with ABCD2 score ≤ 3 may still have high-risk disease requiring immediate
treatment

COHORT STUDY: Stroke 2009 Sep;40(9):3091 | Full Text

Details
● based on retrospective cohort study
● 1,176 patients admitted to hospital clinic with de nite or possible TIA or minor stroke were
evaluated
● 697 patients had ABCD2 score ≤ 3

● immediate consideration for emergency treatment required in 20% of patients with ABCD2
< 4 and 31.6% of patients with score ≥ 4
● reasons for immediate treatment in patients with ABCD2 score ≤ 3

⚬ symptomatic internal carotid stenosis ≥ 50% in 9.1%

⚬ symptomatic intracranial stenosis in 5%
⚬ atrial brillation in 5.9%
⚬ other major cardiac source of embolism in 2.1%

● Reference - Stroke 2009 Sep;40(9):3091 full-text

STUDY
– SUMMARY
in patients with transient symptoms with infarction (TSI), ABCD2 score does not appear
useful in predicting 7-day risk of stroke DynaMed Level 2

COHORT STUDY: Stroke 2011 Aug;42(8):2186 | Full Text

Details
● based on retrospective cohort study
● 257 patients with TSI diagnosed by di usion-weighted imaging (DWI) within 24 hours of
symptom onset were analyzed
● 16 patients (6.2%) had stroke within 7 days
● area under receiver-operating curve 0.57 (95% CI 0.45-0.69, p = 0.331) showing ABCD2 score
was not predictive of 7-day risk of stroke
● Reference - Stroke 2011 Aug;42(8):2186 full-text
⚬ ABCD2 score with addition of imaging (ABCD2-I)

STUDY
– SUMMARY
addition of infarction on imaging (diffusion-weighted magnetic resonance imaging
[DWMRI] or computed tomography [CT]) to ABCD2 score may improve prediction of stroke
within 7 and 90 days after TIA DynaMed Level 2

COHORT STUDY: Neurology 2011 Sep 27;77(13):1222

COHORT STUDY: Stroke 2009 Oct;40(10):3252

Details
● based on 1 retrospective data review and 1 prospective cohort study
● retrospective data review of unpublished data from ABCD2 publication authors regarding
patients who also had DWMRI or CT imaging at time of TIA (using time-based de nition)
⚬ data on 4,574 patients with TIA who had DWMRI or CT scan and 7 days of follow-up were
analyzed
– of 3,206 patients with DWMRI, 884 patients (27.6%) had acute infarction
– of 1,368 patients with CT, 327 patients (23.9%) had new or old infarction
– 145 patients (3.2%) had stroke within 7 days of TIA
– 7-day risk of stroke comparing patients with vs. without infarction on imaging

– 8.7% with vs. 1.2% without infarction on DWMRI or CT

– 7.1% with vs. 0.4% without infarction on DWMRI (adjusted odds ratio 14.9, 95% CI
7.4-30.2)
– 12.8% with vs. 3% without infarction on CT (adjusted odds ratio 4.2, 95% CI 2.6-6.9)

– 7-day risk of stroke also increased (in patients with and without infarction) with
increasing ABCD2 scores (adjusted odds ratio 6.2, 95% CI 4.2-9)
– ABCD2 score and presence of infarction on DWMRI or CT scan were independently
predictive of stroke at 7 days
– optimal prediction score was ABCD2-I score which added 3 points for infarction on
neuroimaging
⚬ data analyzed for 3,700 patients with 90-day follow-up

– 176 patients (4.8%) had stroke within 90 days of TIA

– 90-day risk of stroke comparing patients with vs. without infarction on imaging

– 12% with vs. 2.2% without infarction on DWMRI or CT

– 7.7% with vs. 1% without infarction on DWMRI
– 22.6% with vs. 5.1% without infarction on CT

– 90-day risk of stroke also increased (in patients with and without infarction) with
increasing ABCD2 scores
⚬ Reference - Neurology 2011 Sep 27;77(13):1222 , Stroke 2010 Sep;41(9):1907
● prospective cohort study of 944 patients hospitalized for TIA had DWMRI within 24 hours of
admission
⚬ patients excluded if DWMRI not within 24 hours or ABCD2 could not be calculated
⚬ 4% patients had disabling ischemic stroke within 90 days
⚬ sensitivity for prediction of disabling ischemic stroke within 90 days

– 100% for combination of low-risk ABCD2 score of ≤ 3 and negative early DWMRI
 – 92% for combination of moderate- to high-risk ABCD2 score of 4-7 and negative early
DWMRI
⚬ Reference - Stroke 2009 Oct;40(10):3252

STUDY
⚬ SUMMARY
ABCD3-I score predicts risk of stroke with better discrimination of risk than ABCD2 or
ABCD2-I score DynaMed Level 1

PREDICTION RULE: Lancet Neurol 2010 Nov;9(11):1060

PREDICTION RULE: Lancet Neurol 2016 Nov;15(12):1238

Details
– based on 4 prognostic cohort studies with independent derivation cohort and 4 validation
cohorts
– derivation cohort included 2,654 patients with TIA, validation cohort included 1,232 similar
patients
● 2 risk scores derived

⚬ ABCD3 score (0-9 points) de ned as ABCD2 score plus 2 points for earlier TIA within 7 days
of index TIA (dual TIA)
⚬ ABCD3-I score (0-13 points) de ned as ABCD3 score plus

– 2 points for stenosis of ≥ 50% on carotid imaging

– 2 points for abnormal DWI

● scores based on ndings were validated in 1,232 patients with TIA from 2 population-based
cohorts
● in validation cohort

⚬ 7.5% had stroke by 7 days and 11.6% had stroke by 90 days

⚬ using classi cation by categories (high, intermediate, or low risk), no signi cant
di erences (measured as net reclassi cation improvement) at 7 or 90 days comparing
ABCD3 score or ABCD3-I vs. ABCD2 score
⚬ using risk as a continuous variable, ABCD3-I score associated with improved

discrimination of stroke risk compared to ABCD2 score at 7 days (p = 0.045), 28 days (p =

0.033), and 90 days (p = 0.007)
● Reference - Lancet Neurol 2010 Nov;9(11):1060

– validation cohort included 1,899 patients (median age 68 years) with TIA from 16 centers in
Europe, United States, and Asia followed for 90 days
● rate of strokes in patients after TIA (rates lower than previous studies likely due to patients
receiving early treatment)
⚬ 1% day 2
⚬ 2% day 7
⚬ 3% day 28
⚬ 4% day 90

● patients were strati ed into low-, medium-, and high-risk stroke categories using

⚬ ABCD2 score (low 0-3 points, medium 4-5 points, high 6-7 points)

⚬ ABCD2-I score (ABCD2 categories plus 3 points for infarction, low 0-3 points, medium 4-7
points, high 8-10 points)
 ⚬ ABCD3-I score (low 0-3 points, medium 4-7 points, high 8-13 points)

● stroke rates at days 2 and 7 after TIA in patients classi ed using ABCD2, ABCD2-I, and ABCD3-I
scores

Table 3. Stroke Rates at Days 2 and 7 for Patients With TIA Classified by
ABCD2, ABCD2-I, and ABCD3-I Scores

Risk ABCD2 ABCD2-I ABCD3-I

Classi
catio Numb Stroke Numb Stroke Numb Stroke
n er of er of er of
Patien Patien Patien
ts ts ts

Stroke at Day 2

Low 671 0.5% 516 0.2% 407 0.3%

Mediu 892 2.5% 1,039 1.9% 1,108 0.7%

m

High 253 2% 261 3.5% 301 7%

Stroke at Day 7

Low 680 1% 516 0.6% 408 0.5%

Mediu 902 3.5% 1,054 2.9% 1,126 1.6%

m

High 253 3.5% 265 5.7% 301 9.3%

● discrimination for predicting stroke after TIA using ABCD2, ABCD2-I, and ABCD3-I scores

Table 4. Discrimination for Predicting Stroke After TIA Using ABCD2,

ABCD2-I, and ABCD3-I Scores

C-statistics

ABCD2 ABCD2-I ABCD3-I

 C-statistics

ABCD2 ABCD2-I ABCD3-I

Day 2 stroke 0.64 (95% CI 0.74 (95% CI 0.84 (95% CI

0.56-0.71) 0.67-0.8) 0.76-0.9)

Day 7 stroke 0.61 (95% CI 0.71 (95% CI 0.76 (95% CI

0.53-0.67) 0.64-0.77) 0.69-0.83)

Day 28 stroke 0.59 (95% CI 0.7 (95% CI 0.76 (95% CI

0.53-0.65) 0.64-0.77) 0.69-0.82)

Day 90 stroke 0.61 (95% CI 0.71 (95% CI 0.76 (95% CI

0.55-0.67) 0.65-0.76) 0.71-0.82)

⚬ p < 0.01 comparing ABCD2 vs. ABCD2-I, p < 0.001 comparing ABCD3-I vs. ABCD2, and p <

0.05 comparing ABCD3-I to ABCD2-I for all time points

● Reference - Lancet Neurol 2016 Nov;15(12):1238

● similar results in validation cohort of 239 patients with TIA (Stroke 2013 May;44(5):1244 )
and in prospective validation cohort of 693 patients with TIA (Stroke 2014 Feb;45(2):418 )

Other prognostic factors for stroke

Nondisabling stroke with resolving symptoms

● diagnosis of TIA using classical time-based de nition with presence of acute infarction now de ned as

nondisabling stroke 1

● assessing presence vs. absence of infarction in patients diagnosed with TIA using time-based
de nition

STUDY
⚬ SUMMARY
lesions detected by DWI associated with increased long-term risk of stroke in patients with
TIA diagnosed using time-based definition

COHORT STUDY: Neurology 2019 May 21;92(21):e2455 | Full Text

Details
– based on population-based cohort study
– 1,033 patients with TIA (61.2%) or minor stroke (38.8%) were assessed by DWMRI and followed
for 10 years
● TIA de ned as acute onset of neurologic de cit persisting < 24 hours
● minor stroke de ned as acute onset of neurologic de cit persisting >24 hours, with National
Institutes of Health stroke scale (NIHSS) score ≤ 3
– DWI-detected lesions in 13.9% of patients with TIA, 40% of patients with minor stroke
 – 10-year stroke risk comparing DWI-lesion-positive vs. DWI-lesion-negative patients with

● TIA in 11.5% vs. 4.6% (adjusted hazard ratio [HR] 2.54, 95% CI 1.21-5.34)
● minor stroke and NIHSS score 0-1 points in 14.4% vs. 4.8% (adjusted HR 2.99, 95% CI 1.27-
7.03)
● cryptogenic TIA or minor stroke in 11.2% vs. 2.5% (adjusted HR 4.59, 95% CI 1.65-12.81)

– no signi cant association between DWI lesions and risk of recurrent stroke in patients with
minor stroke and NIHSS score 2-3
– Reference - Neurology 2019 May 21;92(21):e2455 full-text
– see summary for 7-day and 90-day risk of stroke with lesions detected by DWI

STUDY
● SUMMARY
abnormal DWI results may predict short-term risk of recurrent TIA or stroke in patients with
resolution of neurologic symptoms within 24 hours DynaMed Level 2

COHORT STUDY: Arch Neurol 2007 Aug;64(8):1105

Details
⚬ based on retrospective study
⚬ 196 patients with discharge diagnosis of TIA or ischemic stroke, resolution of neurologic symptoms
within 24 hours, and MRI within 48 hours were evaluated
⚬ risk of recurrent TIA or stroke

– 27% among 37 patients with TIA and abnormal DWI (transient symptoms associated with
infarction), based on 4 recurrent TIAs and 6 strokes
– 2.8% among 109 patients with TIA and normal DWI, based on 3 recurrent TIAs and no strokes
– 2% among 50 patients with ischemic stroke, based on no TIAs and 1 recurrent stroke

⚬ Reference - Arch Neurol 2007 Aug;64(8):1105 , editorial can be found in Arch Neurol 2007
Aug;64(8):1080

● Recurrence Risk Estimator (RRE) for predicting stroke risk in patients with TSI

⚬ online calculator can be found at Recurrence Risk Estimator (RRE)

STUDY
⚬ SUMMARY
in patients with TSI, RRE may predict 7-day risk of stroke DynaMed Level 2

COHORT STUDY: Stroke 2011 Aug;42(8):2186 | Full Text

Details
– based on retrospective cohort study
– 257 patients with TSI diagnosed by DWI within 24 hours of symptom onset were analyzed
– 16 patients (6.2%) had stroke within 7 days
– 7-day risk of stroke by RRE score

● 0% with RRE score 0 or 1

● 3% with RRE score 2
● 15% with RRE score 3
● 22% with RRE score ≥ 4

– Reference - Stroke 2011 Aug;42(8):2186 full-text

Carotid artery stenosis

STUDY
● SUMMARY
carotid artery stenosis associated with increased risk of stroke after TIA or minor stroke

SYSTEMATIC REVIEW: Health Technol Assess 2014 Apr;18(27):1 | Full Text

Details
⚬ based on systematic review
⚬ systematic review of 21 studies evaluating clinical prediction models after TIA and e ect of adding
imaging (brain CT or carotid imaging)
⚬ carotid artery stenosis associated with increased risk of stroke after TIA (hazard ratio 3.02, 95% CI
1.18-7.75) in analysis of 4 studies with 1,944 patients, results limited by signi cant heterogeneity,
including di ering de nitions of stenosis
⚬ Reference - Health Technol Assess 2014 Apr;18(27):1 full-text

● see Carotid Artery Stenosis for details

Transient monocular blindness

STUDY
● SUMMARY
transient monocular blindness associated with lower stroke risk than hemispheric TIA

COHORT STUDY: N Engl J Med 2001 Oct 11;345(15):1084

Details
⚬ based on cohort study of medically treated patients enrolled in carotid endarterectomy trial
⚬ 198 patients with transient monocular blindness had half the 3-year ipsilateral stroke risk
compared to 417 patients with hemispheric transient ischemic stroke
⚬ among patients with transient monocular blindness, risk factors for stroke were

– age > 75 years

– male sex
– history of hemispheric TIA or stroke
– history of intermittent claudication
– 80%-94% stenosis of internal carotid artery
– absence of collateral circulation

⚬ Reference - N Engl J Med 2001 Oct 11;345(15):1084

Survival

STUDY
● SUMMARY
TIA in patients aged 18-50 years associated with increased risk of 20-year all-cause mortality

COHORT STUDY: JAMA 2013 Mar 20;309(11):1136

Details
⚬ based on prospective cohort study
⚬ 959 patients aged 18-50 years with rst-time acute TIA, ischemic stroke, or intracerebral
hemorrhage in Netherlands were followed for mean 11.1 years
 ⚬ 262 patients had TIA and 99.6% survived to 30 days
⚬ cumulative risk of all-cause mortality in 30-day survivors of TIA

– 1.2% at 1 year
– 2.5% at 5 years
– 9.2% at 10 years
– 24.9% at 20 years

⚬ TIA associated with increased risk of 20-year all-cause mortality compared to general population
(standardized mortality ratio 2.6, 95% CI 1.8-3.7)
⚬ Reference - FUTURE study (JAMA 2013 Mar 20;309(11):1136 ), editorial can be found in JAMA 2013
Mar 20;309(11):1171

STUDY
● SUMMARY
myocardial infarction after TIA and comorbid diabetes associated with increased risk of
mortality in patients with TIA

COHORT STUDY: Stroke 2011 Apr;42(4):935 | Full Text

Details
⚬ based on cohort study
⚬ 456 patients in United States with TIA between 1985 and 1994 in Olmsted County, Rochester,
Minnesota were tracked for incidence of myocardial infarction with median follow-up of 10.2 years
⚬ factors associated with increased risk of death during follow-up after TIA

– myocardial infarction after TIA (hazard ratio [HR] 3.11, 95% CI 2.11-4.57)
– increased age (per 10 years) (HR 2.3, 95% CI 2-2.63)
– diabetes (HR 1.54, 95% CI 1.05-2.28)
– smoking (HR 1.3, 95% CI 0.98-1.73), results limited by borderline signi cance
– hypertension (HR 1.28, 95% CI 0.97-1.68), results limited by borderline signi cance

⚬ Reference - Stroke 2011 Apr;42(4):935 full-text

Prevention and Screening

Prevention

● treat patients with atrial brillation with anticoagulation (Stroke 2014 Jul;45(7):2160 full-text )

● control risk factors including

⚬ hypertension
⚬ elevated lipids
⚬ tobacco use
⚬ alcohol use
⚬ overweight
⚬ inactivity
⚬ Reference - Stroke 2014 Jul;45(7):2160 full-text

● see also

⚬ Thromboembolic Prophylaxis in Atrial Fibrillation

⚬ Cardiovascular Disease Prevention Overview
⚬ Secondary Prevention of Stroke or Transient Ischemic Attack
Screening

● not applicable

Quality Improvement

Quality and Outcomes Framework Indicators

● STIA1. Contractor establishes and maintains a register of patients with stroke or transient ischemic
attack (TIA)

● STIA3. Percentage of patients with history of stroke or transient ischemic attack (TIA) in whom last
blood pressure reading (measured in preceding 12 months) is ≤ 150/90 mm Hg

● STIA9. Percentage of patients with stroke or transient ischemic attack (TIA) who have had in uenza
immunization in the preceding 1 August to 31 March

● STIA7. Percentage of patients with stroke shown to be nonhemorrhagic, or history of transient

ischemic attack (TIA), who have a record in preceding 12 months that antiplatelet agent or
anticoagulant is being taken

● STIA8. Percentage of patients with stroke or transient ischemic attack (TIA) diagnosed on or after 1
April 2014 who have a record of referral for further investigation between 3 months before or 1
month after date of latest recorded stroke or rst TIA

● SMOK2 (NM38). Percentage of patients with any or any combination of the following conditions:
coronary heart disease, peripheral artery disease, stroke or transient ischemic attack, hypertension,
diabetes, chronic obstructive pulmonary disease (COPD), chronic kidney disease, asthma,
schizophrenia, bipolar a ective disorder, or other psychoses who have a record of smoking status in
the preceding 12 months

● SMOK5 (NM39). Percentage of patients with any or any combination of the following conditions:
coronary heart disease, peripheral artery disease, stroke or transient ischemic attack, hypertension,
diabetes, chronic obstructive pulmonary disease (COPD), chronic kidney disease, asthma,
schizophrenia, bipolar a ective disorder, or other psychoses who smoke who have a record of an
o er of support and treatment within preceding 12 months

● see Quality and Outcomes Framework Indicators for additional information

Efficacy

STUDY
● SUMMARY
electronic decision support tool reported to provide guideline-compliant TIA management
DynaMed Level 3

MODELING STUDY: N Z Med J 2013 Apr 5;126(1372):25

Details
⚬ based on comparative study without patient outcomes
⚬ 7 hypothetical patient cases were evaluated by 12 physicians with expertise in stroke care, 12
internists without special training or experience in stroke care, 10 general practitioners, and
electronic TIA/stroke decision support tool
 ⚬ electronic TIA/stroke decision support tool included computer software with single page patient
data entry form and algorithm to provide guidance based on New Zealand TIA treatment guidelines
⚬ reference standard was New Zealand TIA treatment guidelines and expert consensus
⚬ software tool guidance was 100% correct for each of diagnosis, localization, triage advice, and
immediate treatment, and provided most complete lifestyle management advice
⚬ physician performance (proportion with correct guidance)

– stroke expert performance 92%-98% for diagnosis, localization, and treatment

– internist performance 79% for diagnosis, 58% for localization, 59% for triage, and 31% for
immediate treatment
– general practitioner performance 76% for diagnosis, 9% for localization, 84% for triage, and 27%
for immediate treatment general practitioners
⚬ Reference - N Z Med J 2013 Apr 5;126(1372):25

Guidelines and Resources

Guidelines

United States Guidelines

● American Academy of Neurology (AAN) practice advisory on recurrent stroke with patent foramen
ovale (update of practice parameter) can be found in Neurology 2016 Aug 23;87(8):815 full-text ,
previous version can be found in Neurology 2004 Apr 13;62(7):1042

● American Heart Association/American Stroke Association (AHA/ASA)

⚬ AHA/ASA scienti c statement on de nition and evaluation of transient ischemic attack (TIA) can be
found in Stroke 2009 Jun;40(6):2276 ; American Academy of Neurology (AAN) a rms the value of
this statement as an educational tool for neurologists
⚬ AHA/ASA recommendations on management of cerebral and cerebellar infarction with swelling can
be found in Stroke 2014 Apr;45(4):1222
⚬ American Heart Association/American Stroke Association (AHA/ASA) guideline on prevention of
stroke in patients with stroke or transient ischemic attack can be found in Stroke 2014
Jul;45(7):2160 , correction can be found in Stroke 2015 Feb;46(2):e54 , commentary can be
found in Stroke 2015 Apr;46(4):e85
⚬ American Heart Association/American Stroke Association (AHA/ASA) guideline on early
management of adults with ischemic stroke can be found in Stroke 2018 Mar;49(3):e46 ,
corrections can be found in Stroke 2018 Mar;49(3):e138 and Stroke 2018 Jun;49(6):e233
⚬ American Heart Association/American Stroke Association (AHA/ASA) statement on updated
de nition of stroke for 21st century can be found in Stroke 2013 Jul;44(7):2064 PDF

● American College of Emergency Physicians (ACEP) clinical policy on critical issues in the evaluation of
adult patients with suspected transient ischemic attack in the emergency department can be found in
Ann Emerg Med 2016 Sep;68(3):354

● Institute for Clinical Systems Improvement (ICSI) guideline on diagnosis and initial treatment of
ischemic stroke can be found at ICSI 2019 Dec PDF

● University of Michigan Health System (UMHS) guideline on secondary prevention of ischemic heart
disease and stroke in adults can be found at UMHS 2014 May PDF
● American College of Radiology (ACR) appropriateness criteria on cerebrovascular disease can be
found at ACR 2016 PDF

● American College of Cardiology Foundation/American Heart Association (ACCF/AHA) practice guideline

on management of patients with extracranial carotid and vertebral artery disease can be found in
Circulation 2011 Jul 26;124(4):e54 , corrections can be found in Circulation 2011 Jul 26;124(4):e146,
Circulation 2012 Jul 10;126(2):e26

● American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice Guidelines for
Antithrombotic Therapy and Prevention of Thrombosis (Ninth Edition) guideline and expert panel
report on antithrombotic therapy for atrial brillation can be found in Chest 2018 Nov;154(5):1121

● American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice guideline on

antithrombotic and thrombolytic therapy (9th ed.) recommendation on antithrombotic and
thrombolytic therapy for valvular disease can be found in Chest 2012 Feb;141(2 Suppl):e576S full-
text , commentary can be found in Chest 2013 May;143(5):1513

● American College of Chest Physicians (ACCP) evidence-based clinical practice guideline (ninth edition)
on antithrombotic and thrombolytic therapy for ischemic stroke can be found in Chest 2012 Feb;141(2
Suppl):e601S full-text

United Kingdom Guidelines

● Royal College of Physicians (RCP) national clinical guideline on stroke can be found at RCP 2016 Oct 3
PDF

● National Institute for Health and Care Excellence (NICE) guideline on diagnosis and initial
management of acute stroke and transient ischemic attack (TIA) can be found at NICE 2019
May:NG128

Canadian Guidelines

● Canadian Cardiovascular Society (CCS)

⚬ focused 2018 update of CCS atrial brillation guideline: management of atrial brillation can be
found in Can J Cardiol 2018 Nov;34(11):1371
⚬ focused 2016 update of CCS atrial brillation guideline: management of atrial brillation can be
found in Can J Cardiol 2016 Oct;32(10):1170
⚬ focused 2014 update of CCS atrial brillation guideline: management of atrial brillation can be
found in Can J Cardiol 2014 Oct;30(10):1114 , corrections can be found in Can J Cardiol 2014
Dec;30(12):1495 and Can J Cardiol 2015 Oct;31(10):1302
⚬ focused 2012 update of CCS atrial brillation guideline: recommendations on stroke prevention
and rate/rhythm control can be found in Can J Cardiol 2012 Mar-Apr;28(2):125 , correction can
be found in Can J Cardiol 2012 May;28(3):396
⚬ CCS 2010 atrial brillation guidelines on

– etiology and initial investigations can be found in Can J Cardiol 2011 Jan-Feb;27(1):31

– management of recent-onset atrial brillation and utter in emergency department can be

found in Can J Cardiol 2011 Jan-Feb;27(1):38 , correction can be found in Can J Cardiol 2012
Mar-Apr;28(2):244
– rate and rhythm management can be found in Can J Cardiol 2011 Jan-Feb;27(1):47 , correction
can be found in Can J Cardiol 2011 May-Jun;27(3):388
 – catheter ablation for atrial brillation/atrial utter can be found in Can J Cardiol 2011 Jan-
Feb;27(1):60
– surgical therapy can be found in Can J Cardiol 2011 Jan-Feb;27(1):67
– prevention of stroke and systemic thromboembolism in atrial brillation and utter can be
found in Can J Cardiol 2011 Jan-Feb;27(1):74
– prevention and treatment of atrial brillation following cardiac surgery can be found in Can J
Cardiol 2011 Jan-Feb;27(1):91

● Canadian Stroke Best Practice Recommendations (CSBPR) on

⚬ stroke recognition and response can be found at CSBPR 2018 Jun

⚬ secondary prevention of stroke can be found at CSBPR 2017 Oct

⚬ hyperacute stroke care can be found at CSBPR 2018 Jun
⚬ acute inpatient stroke care can be found at CSBPR 2015 Oct or in Int J Stroke 2016 Feb;11(2):239

⚬ stroke rehabilitation can be found at CSBPR 2015 or in Int J Stroke 2016 Jun;11(4):459
⚬ transitions of care can be found at CSBPR 2016 Jul

⚬ mood, cognition, and fatigue can be found at CSBPR 2019

⚬ Telestroke can be found at CSBPR 2017 Apr or in Int J Stroke 2017 Oct;12(8):886

● Registered Nurses Association of Ontario (RNAO) guideline on stroke assessment across continuum of
care can be found at RNAO 2005 Jun PDF , supplement can be found at RNAO 2011 Aug PDF

European Guidelines

● S3 Leitline Sekundärprophylaxe ischämischer Schlaganfall und transitorische ischämische Attacke

nden Sie unter AWMF 2015 PDF [Deutsch]

● European Society of Cardiology (ESC) guideline on management of atrial brillation can be found at
ESC 2020 Aug

Asian Guidelines

● Singapore Ministry of Health (SMOH) guideline on use of intravenous recombinant tissue plasminogen
activator (rtPA) in ischaemic stroke patients can be found at SMOH 2013 Dec PDF

● Japan Stroke Society (JSS) guideline on management of stroke can be found at JSS 2019 PDF
[Japanese 日本語]

Mexican Guidelines

● Mexican Association of Cerebral Vascular Disease (Asociación Mexicana de Enfermedad Vascular

Cerebral) guideline on
⚬ secondary prevention treatment in cerebrovascular diseases after acute phase can be found in Rev
Invest Clin 2010 Mar-Apr;62(2):130 [Spanish] PDF
⚬ antiplatelet treatment for secondary prevention of ischemic stroke or transient ischemic attack can
be found in Rev Invest Clin 2010 Mar-Apr;62(2):135 [Spanish] PDF
⚬ anticoagulation for the secondary prevention of ischemic stroke can be found in Rev Invest Clin
2010 Mar-Apr;62(2):141 [Spanish] PDF
⚬ hypertension treatment in patients with cerebrovascular diseases after acute phase can be found
in Rev Invest Clin 2010 Mar-Apr;62(2):152 [Spanish] PDF
 ⚬ statins for the secondary prevention of ischemic stroke can be found in Rev Invest Clin 2010 Mar-
Apr;62(2):162 [Spanish] PDF
⚬ surgical (endarterectomy) and endovascular treatment (angioplasty with distal protection and
stenting) for secondary prevention treatment of ischemic stroke secondary to carotid
atherosclerotic disease can be found in Rev Invest Clin 2010 Mar-Apr;62(2):170 [Spanish] PDF
⚬ lifestyle and primary and secondary prevention for cerebrovascular disease can be found in Rev
Invest Clin 2010 Mar-Apr;62(2):181 [Spanish] PDF

Central and South American Guidelines

● Brazilian Stroke Society/Brazilian Academy of Neurology (BSS/BAN) guidelines on acute ischemic

stroke treatment
⚬ part I can be found in Arq Neuropsiquiatr 2012 Aug;70(8):621 full-text [Portuguese, English]
⚬ part II can be found in Arq Neuropsiquiatr 2012 Nov;70(11):885 full-text [Portuguese, English]

Australian and New Zealand Guidelines

● New Zealand Ministry of Health (NZMOH) guideline on cardiovascular risk assessment and
management for primary care can be found at NZMOH 2018 PDF

● Stroke Foundation clinical guideline on stroke management can be found at Stroke Foundation

African Guidelines

● South African guideline on management of ischemic stroke and transient ischemic attack:
recommendations for resource-constrained healthcare setting can be found in Int J Stroke 2011
Aug;6(4):349

Review articles

● review can be found in Ann Intern Med 2011 Jan 4;154(1):ITC11

● review of epidemiology of TIA can be found in Front Neurol Neurosci 2014;33:69

● review of etiology and pathogenesis of TIA and stroke can be found in Brain Circ 2017 Apr;3(2):66

full-text

● review of posterior circulation ischemic stroke can be found in BMJ 2014 May 19;348:g3175

● review of symptoms of TIA can be found in Front Neurol Neurosci 2014;33:82

● review of diagnosis and evaluation of TIA can be found in Am Fam Physician 2012 Sep 15;86(6):521

full-text

● review of neuroimaging in evaluation of TIA can be found in J Neuroradiol 2015 Feb;42(1):3

● review of management of TIA in the emergency department can be found in Ann Emerg Med 2018
Mar;71(3):409

● review of prognostic risk scores for stroke after TIA can be found in Front Neurol Neurosci 2014;33:41
● review of risk factor modi cation and treatment of TIA can be found in Am Fam Physician 2012 Sep
15;86(6):527

● review of secondary prevention after transient ischemic attack or ischemic stroke can be found in Am
J Med 2014 Aug;127(8):728

● case presentation of secondary prevention after ischemic stroke or transient ischemic attack can be
found in N Engl J Med 2012 May 17;366(20):1914

MEDLINE search

● to search MEDLINE for (Transient ischemic attack) with targeted search (Clinical Queries), click therapy
, diagnosis , or prognosis

Patient Information

● handout from EBSCO Health Library or in Spanish

● handout from Patient UK PDF

● information from American Stroke Association

● information from National Institute of Neurological Disorders and Stroke

ICD Codes

ICD-10 codes

● G45 transient cerebral ischaemic attacks and related syndromes

⚬ G45.0 vertebro-basilar artery syndrome

⚬ G45.1 carotid artery syndrome (hemispheric)
⚬ G45.2 multiple and bilateral precerebral artery syndromes
⚬ G45.3 amaurosis fugax
⚬ G45.4 transient global amnesia
⚬ G45.8 other transient cerebral ischaemic attacks and related syndromes
⚬ G45.9 transient cerebral ischaemic attack, unspeci ed

References

General References Used

1. Duca A, Jagoda A. Transient Ischemic Attacks: Advances in Diagnosis and Management in the
Emergency Department. Emerg Med Clin North Am. 2016 Nov;34(4):811-835

2. Coutts SB. Diagnosis and Management of Transient Ischemic Attack. Continuum (Minneap Minn). 2017
Feb;23(1, Cerebrovascular Disease):82-92

3. Easton JD, Saver JL, Albers GW, et al. De nition and evaluation of transient ischemic attack: a scienti c
statement for healthcare professionals from the American Heart Association/American Stroke
Association Stroke Council; Council on Cardiovascular Surgery and Anesthesia; Council on
 Cardiovascular Radiology and Intervention; Council on Cardiovascular Nursing; and the
Interdisciplinary Council on Peripheral Vascular Disease. The American Academy of Neurology a rms
the value of this statement as an educational tool for neurologists. Stroke. 2009 Jun;40(6):2276-93

4. Casaubon LK, Boulanger JM, Blacquiere D, et al; Heart and Stroke Foundation of Canada Canadian
Stroke Best Practices Advisory Committee. Canadian Stroke Best Practice Recommendations (CSBPR):
Hyperacute stroke care guidelines, update 2015. Int J Stroke 2015 Aug;10(6):924-40

Recommendation Grading Systems Used

● American Heart Association/American Stroke Association (AHA/ASA) grading system for

recommendations
⚬ classi cations of recommendations

– Class I - procedure or treatment should be performed or administered

– Class IIa - reasonable to perform procedure or administer treatment, but additional studies with
focused objectives needed
– Class IIb - procedure or treatment may be considered; additional studies with broad objectives
needed, additional registry data would be useful
– Class III - procedure or treatment should not be performed or administered because it is not
helpful or may be harmful
● Class III ratings may be subclassi ed as Class III No Bene t or Class III Harm

⚬ levels of evidence

– Level A - data derived from multiple randomized clinical trials or meta-analyses

– Level B - data derived from single randomized trial or nonrandomized studies
– Level C - only expert opinion, case studies, or standard of care

⚬ Reference

– AHA/ASA scienti c statement on de nition and evaluation of TIA (Stroke 2009 Jun;40(6):2276 )
– AHA/ASA guidelines on prevention of stroke in patients with stroke or TIA (Stroke 2014
Jul;45(7):2160 PDF )
– ACC/AHA focused update on management of patients with atrial brillation (update on
dabigatran) (Circulation 2011 Mar 15;123(10):1144 PDF )

● American College of Cardiology/American Heart Association (ACC/AHA) grading system for

recommendations
⚬ classi cations of recommendations

– Class I - procedure or treatment should be performed or administered

– Class IIa - reasonable to perform procedure or administer treatment, but additional studies with
focused objectives needed
– Class IIb - procedure or treatment may be considered; additional studies with broad objectives
needed, additional registry data would be useful
– Class III - procedure or treatment should not be performed or administered because it is not
helpful or may be harmful
● Class III ratings may be subclassi ed as Class III No Bene t or Class III Harm

⚬ levels of evidence

– Level A - high-quality evidence from > 1 randomized controlled trial or meta-analysis of high-
quality randomized controlled trials
 – Level B-R - moderate-quality evidence from ≥ 1 randomized controlled trial or meta-analysis of
moderate-quality randomized controlled trials
– Level B-NR - moderate-quality evidence from ≥ 1 well-designed nonrandomized trial,
observational studies, or registry studies, or meta-analysis of such studies
– Level C-LD - randomized or nonrandomized studies with methodological limitations or meta-
analyses of such studies
– Level C-EO - consensus of expert opinion based on clinical experience

⚬ Reference - ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline on prevention,

detection, evaluation, and management of high blood pressure in adults (Hypertension 2018
Jun;71(6):e13 )

● American College of Chest Physicians (ACCP) uses Grading of Recommendations, Assessment,

Development, and Evaluation (GRADE) approach to recommendations
⚬ recommendations

– Strong - bene ts outweigh risk and burdens, or vice versa; recommend

– Weak - conditional, bene ts closely balanced with risks and burden; suggest
– Ungraded - consensus based, uncertainty due to lack of evidence, expert opinion that bene ts
outweigh risk and burdens, or vice versa; insu cient evidence for a graded recommendation
⚬ quality of evidence

– High - con dence that true e ect lies close to estimate of e ect from the estimate of e ect
– Moderate - moderate con dence in e ect estimate, true e ect likely to be close to the estimate
of the e ect, but possibility it is substantially di erent
– Low - con dence in the e ect estimate is limited, true e ect may be substantially di erent from
estimate of e ect
– Very low - little con dence in the e ect estimate. true e ect is likely to be substantially di erent

⚬ Reference - ACCP Evidence-Based Clinical Practice Guidelines for Antithrombotic Therapy and
Prevention of Thrombosis (Ninth Edition) guideline and expert panel report on antithrombotic
therapy for atrial brillation (Chest 2018 Nov;154(5):1121 )

● American College of Chest Physicians (ACCP) grades

⚬ Grade 1 - strong recommendation based on clear risk/bene t balance

⚬ Grade 2 - weak recommendation based on unclear or close risk/bene t balance
⚬ Grade A - high-quality evidence based on consistent evidence from randomized trials without
important limitations or exceptionally strong evidence from observational studies
⚬ Grade B - moderate-quality evidence based on randomized trials with important limitations
(inconsistent results, methodologic aws, indirect or imprecise results) or very strong evidence
from observational studies
⚬ Grade C - low- or very low-quality evidence based on evidence for ≥ 1 critical outcome from
observational studies, case series, or randomized trials with serious aws or indirect evidence
⚬ Reference - ACCP evidence-based clinical practice guideline on methodology for development of
antithrombotic therapy and prevention of thrombosis (Chest 2012 Feb;141(2 Suppl):53S full-text
), commentary can be found in Chest 2013 Apr;143(4):1190

● Canadian Cardiovascular Society (CCS) grading system for recommendations

⚬ strength of recommendation

– Strong
– Conditional (weak)

⚬ quality of evidence
 – High - future research unlikely to change con dence in estimate of e ect; multiple well-
designed, well-conducted clinical trials
– Moderate - further research likely to have important impact on con dence in estimate of e ect
and may change estimate; limited clinical trials, inconsistency of results, or study limitations
– Low - further research very likely to have signi cant impact on estimate of e ect and is likely to
change estimate; small number of clinical studies or cohort observations
– Very Low - estimate of e ect is very uncertain; case studies or consensus opinion

⚬ References

– CCS atrial brillation guidelines 2010: rate and rhythm management (Can J Cardiol 2011 Jan-
Feb;27(1):47 , Can J Cardiol 2011 Jan-Feb;27(1):27 )
– CCS focused 2012 update of atrial brillation guidelines: recommendations for stroke
prevention and rate/rhythm control (Can J Cardiol 2012 Mar-Apr;28(2):125 )
– CCS focused 2014 update of atrial brillation guidelines: management of atrial brillation (Can J
Cardiol 2014 Oct;30(10):1114 )
– CCS focused 2016 update on atrial brillation guidelines: management of atrial brillation (Can J
Cardiol 2016 Oct;32(10):1170 )

● Canadian Stroke Best Practice Recommendations (CSBPR) levels of evidence

⚬ Evidence Level A

– meta-analysis of randomized controlled trials or consistent ndings from ≥ 2 randomized trials

– desirable e ects clearly outweigh undesirable e ects or vice versa

⚬ Evidence Level B

– single randomized controlled trial or consistent ndings from ≥ 2 well-designed nonrandomized

and/or uncontrolled trials, and large observational studies
– desirable e ects outweigh or are closely balanced with undesirable e ects or vice versa

⚬ Evidence Level C

– writing group consensus and/or supported by limited research evidence

– desirable e ects outweigh or are closely balanced with undesirable e ects or vice versa

⚬ Reference - CSBPR Overview and Methodology (CSBPR 2014 PDF )

● European Society of Cardiology (ESC) grading system for recommendations

⚬ classes of recommendations

– Class I - evidence and/or general agreement that given treatment or procedure is bene cial,
useful, and e ective
– Class II - con icting evidence and/or divergence of opinion about usefulness/e cacy of given
treatment or procedure
● Class IIa - weight of evidence/opinion in favor of usefulness/e cacy
● Class IIb - usefulness/e cacy less well-established by evidence/opinion
– Class III - evidence or general agreement that given treatment or procedure is not
useful/e ective, and in some cases may be harmful
⚬ levels of evidence

– Level A - data derived from multiple randomized clinical trials or meta-analyses

– Level B - data derived from single randomized trial or large nonrandomized studies
– Level C - consensus of opinion of experts and/or small studies, retrospective studies, registries

⚬ References
 – ESC guideline on diagnosis and management of acute pulmonary embolism (Eur Heart J 2014
Nov 14;35(43):3033 full-text ), corrections can be found in Eur Heart J 2015 Oct
14;36(39):2642 and Eur Heart J 2015 Oct 14;36(39):2666, commentary can be found in Rev Esp
Cardiol (Engl Ed) 2015 Jan;68(1):10
– ESC guideline on management of atrial brillation (Eur J Cardiothorac Surg 2016 Nov;50(5):e1

full-text )

Synthesized Recommendation Grading System for DynaMed Content

● The DynaMed Team systematically monitors clinical evidence to continuously provide a synthesis of
the most valid relevant evidence to support clinical decision-making (see 7-Step Evidence-Based
Methodology ).

● Guideline recommendations summarized in the body of a DynaMed topic are provided with the
recommendation grading system used in the original guideline(s), and allow users to quickly see
where guidelines agree and where guidelines di er from each other and from the current evidence.

● In DynaMed content, we synthesize the current evidence, current guidelines from leading authorities,
and clinical expertise to provide recommendations to support clinical decision-making in the Overview
& Recommendations section.

● We use the Grading of Recommendations Assessment, Development and Evaluation (GRADE) to

classify synthesized recommendations as Strong or Weak.
⚬ Strong recommendations are used when, based on the available evidence, clinicians (without
con icts of interest) consistently have a high degree of con dence that the desirable consequences
(health bene ts, decreased costs and burdens) outweigh the undesirable consequences (harms,
costs, burdens).
⚬ Weak recommendations are used when, based on the available evidence, clinicians believe that
desirable and undesirable consequences are nely balanced, or appreciable uncertainty exists
about the magnitude of expected consequences (bene ts and harms). Weak recommendations are
used when clinicians disagree in judgments of relative bene t and harm, or have limited
con dence in their judgments. Weak recommendations are also used when the range of patient
values and preferences suggests that informed patients are likely to make di erent choices.

● DynaMed synthesized recommendations (in the Overview & Recommendations section) are
determined with a systematic methodology:
⚬ Recommendations are initially drafted by clinical editors (including ≥ 1 with methodological
expertise and ≥ 1 with content domain expertise) aware of the best current evidence for bene ts
and harms, and the recommendations from guidelines.
⚬ Recommendations are phrased to match the strength of recommendation. Strong
recommendations use "should do" phrasing, or phrasing implying an expectation to perform the
recommended action for most patients. Weak recommendations use "consider" or "suggested"
phrasing.
⚬ Recommendations are explicitly labeled as Strong recommendations or Weak
recommendations when a quali ed group has explicitly deliberated on making such a
recommendation. Group deliberation may occur during guideline development. When group
deliberation occurs through DynaMed Team-initiated groups:
– Clinical questions will be formulated using the PICO (Population, Intervention, Comparison,
Outcome) framework for all outcomes of interest speci c to the recommendation to be
 developed.
– Systematic searches will be conducted for any clinical questions where systematic searches
were not already completed through DynaMed content development.
– Evidence will be summarized for recommendation panel review including for each outcome, the
relative importance of the outcome, the estimated e ects comparing intervention and
comparison, the sample size, and the overall quality rating for the body of evidence.
– Recommendation panel members will be selected to include at least 3 members that together
have su cient clinical expertise for the subject(s) pertinent to the recommendation,
methodological expertise for the evidence being considered, and experience with guideline
development.
– All recommendation panel members must disclose any potential con icts of interest
(professional, intellectual, and nancial), and will not be included for the speci c panel if a
signi cant con ict exists for the recommendation in question.
– Panel members will make Strong recommendations if and only if there is consistent
agreement in a high con dence in the likelihood that desirable consequences outweigh
undesirable consequences across the majority of expected patient values and preferences.
Panel members will make Weak recommendations if there is limited con dence (or
inconsistent assessment or dissenting opinions) that desirable consequences outweigh
undesirable consequences across the majority of expected patient values and preferences. No
recommendation will be made if there is insu cient con dence to make a recommendation.
– All steps in this process (including evidence summaries which were shared with the panel, and
identi cation of panel members) will be transparent and accessible in support of the
recommendation.
⚬ Recommendations are veri ed by ≥ 1 editor with methodological expertise, not involved in
recommendation drafting or development, with explicit con rmation that Strong
recommendations are adequately supported.
⚬ Recommendations are published only after consensus is established with agreement in phrasing
and strength of recommendation by all editors.
⚬ If consensus cannot be reached then the recommendation can be published with a notation of
"dissenting commentary" and the dissenting commentary is included in the topic details.
⚬ If recommendations are questioned during peer review or post publication by a quali ed
individual, or reevaluation is warranted based on new information detected through systematic
literature surveillance, the recommendation is subject to additional internal review.

DynaMed Editorial Process

● DynaMed topics are created and maintained by the DynaMed Editorial Team and Process .

● All editorial team members and reviewers have declared that they have no nancial or other
competing interests related to this topic, unless otherwise indicated.

● DynaMed content includes Practice-Changing Updates, with support from our partners, McMaster
University and F1000.

Special acknowledgements

● DynaMed topics are written and edited through the collaborative e orts of the above individuals.
Deputy Editors, Section Editors, and Topic Editors are active in clinical or academic medical practice.
Recommendations Editors are actively involved in development and/or evaluation of guidelines.

● Editorial Team role de nitions

 Topic Editors de ne the scope and focus of each topic by formulating a set of clinical
questions and suggesting important guidelines, clinical trials, and other data to be
addressed within each topic. Topic Editors also serve as consultants for the internal
DynaMed Editorial Team during the writing and editing process, and review the nal
topic drafts prior to publication.

Section Editors have similar responsibilities to Topic Editors but have a broader role
that includes the review of multiple topics, oversight of Topic Editors, and systematic
surveillance of the medical literature.

Recommendations Editors provide explicit review of Overview and

Recommendations sections to ensure that all recommendations are sound,
supported, and evidence-based. This process is described in "Synthesized
Recommendation Grading."

Deputy Editors oversee DynaMed internal publishing groups. Each is responsible for
all content published within that group, including supervising topic development at
all stages of the writing and editing process, nal review of all topics prior to
publication, and direction of an internal team.

How to cite

National Library of Medicine, or "Vancouver style" (International Committee of Medical Journal Editors):

● DynaMed [Internet]. Ipswich (MA): EBSCO Information Services. 1995 - . Record No. T116640,
Transient Ischemic Attack (TIA); [updated 2018 Nov 30, cited place cited date here]. Available from
https://www.dynamed.com/topics/dmp~AN~T116640. Registration and login required.

Published by EBSCO Information Services. Copyright © 2020, EBSCO Information Services. All rights reserved. No part of
this may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying,
recording, or by any information storage and retrieval system, without permission.

EBSCO Information Services accepts no liability for advice or information given herein or errors/omissions in the text. It is
merely intended as a general informational overview of the subject for the healthcare professional.