DIAGNOSIS OFSTROKE AND STROKEMIMICS IN THEEMERGENCY SETTINGKevin M. Barrett, Joshua M. Levine, Karen C. Johnston

ABSTRACT

Patients with suspected stroke require urgent evaluation in order to identify thosewho may be eligible for time-sensitive therapies. A focused and systematicapproach to diagnosis improves the likelihood of identifying patients withprobable ischemic stroke and minimizes the chances of exposing patients withalternate diagnoses to potentially harmful treatment. This chapter emphasizes thehistorical, examination, and neuroimaging findings useful in the rapid evaluationand diagnosis of patients with suspected ischemic stroke. Other entities that maypresent with strokelike symptoms will also be discussed.

Continuum Lifelong Learning Neurol 2008;14(6):13–27.

INTRODUCTION

Stroke typically presents with thesudden onset of focal neurologic def-icits. Appropriate delivery of acutestroke therapies depends on accu-rately establishing the time of symp-tom onset, performing a focused bed-side assessment, and rapidly interpretingancillary tests. This chapter emphasizesa systematic diagnostic approach thatwill facilitate expeditious identificationof patients eligible for acute therapies.By necessity, the discussion that followsis presented sequentially. However, it isimportant to recognize that many ele-

ments of the stroke evaluation occur inparallel, depending on resource avail-ability or clinical circumstances. Specificevidence-based stroke therapies arecovered in a subsequent chapter.

BEDSIDE ASSESSMENT

History

The ability to treat acute ischemicstroke patients with IV thrombolysisor endovascular therapies depends onaccurately establishing the time ofsymptom onset. Under ideal circum-stances, the patient is able to provide a

13

Relationship Disclosure: Drs Barrett and Levine have nothing to disclose. Dr Johnston has received personalcompensation for activities with AstraZeneca, Boehringer Ingelheim Pharmaceuticals, Inc., DiffusionPharmaceuticals LLC, NIH–National Institute of Neurological Disorders and Stroke, Ono PharmaceuticalCo., Ltd, Remedy Pharmaceuticals, and sanofi-aventis. Dr Johnston has received personal compensation in aneditorial capacity from Up-to-Date and Neurology.Unlabeled Use of Products/Investigational Use Disclosure: Drs Barrett and Levine have nothing to disclose.Dr Johnston discusses the unlabeled use of investigational therapies.

Note: Text referenced in the Quintessentials Preferred Responses, which appearlater in this issue, is indicated in yellow shading throughout this chapter.

Copyright # 2008, American Academy of Neurology. All rights reserved.

KEY POINT

A The ability to

treat acute

ischemic stroke

patients with

IV thrombolysis

or endovascular

therapies

depends on

accurately

establishing

the time of

symptom onset.

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

detailed account of symptom onset.The history should focus on elicitinginformation that will establish eligibil-ity for thrombolytic therapy or identifypotential exclusionary conditions. Sit-uations may arise in which the precisetime of symptom onset may be diffi-cult to establish. The use of cues (eg,‘‘before or after lunch?’’ or ‘‘before orafter the evening news?’’) may behelpful in generating an estimatedtime of onset. If family or friends arepresent, it is wise to corroborate thepatient’s report of symptom onsetwith eyewitnesses. In some circum-stances, a precise time of symptomonset will prove impossible to deter-mine, and the line of questioning shouldthen shift to identifyingwhen the patientwas last neurologically normal. For thosepatients who awaken with symptoms,the time of onset becomes the time atwhich they went to bed (assuming theywere normal at that time). For patientswith heralding symptoms or TIA, it isnecessary to ensure complete resolutionbefore the clock can be ‘‘reset.’’

The nature of symptom onset shouldbe obtained when possible. Symptomsthat begin abruptly suggest a vascularetiology, whereas symptoms that beginin one region and gradually spread toinvolve other areas may support analternate etiology (ie, migraine). Inquiryshould be made regarding risk factorsfor vascular disease, as well as anyhistory of seizures, migraine, insulinuse, or drug abuse that may supportanother cause for the patient’s symp-toms. Information necessary for deci-sions regarding thrombolysis must beobtained in a structured fashion tominimize the possibility of overlook-ing critical information (Table 1-1).Accompanying symptoms, particularlyheadache, warrant further exploration.Ictal, or so-called thunderclap head-ache, should alert the clinician to thepossibility of subarachnoid hemorrhage.Small bleeds or sentinel leaks from un-

ruptured intracranial aneurysms may notbe evident on CT scan. Further evalua-tion with lumbar puncture to exclude thepresence of blood in the subarachnoidspace may be warranted (Edlow andCaplan, 2000).

The initial evaluation of the poten-tial stroke patient often occurs in a high-acuity area. Medical personnel respon-sible for establishing IV access, initiatingcardiorespiratorymonitoring, perform-ing blood draws, and performing elec-trocardiography compete for the patient’sattention. Additionally, the presenceof aphasia or neglect may limit thepatient’s ability to provide accurate

14

TABLE 1-1 ImportantHistoricalInformation inthe SuspectedStroke Patient

" History of the Present Illness

Time of symptom onset

Evolution of symptoms

Convulsion or loss ofconsciousness at onset

Headache

Chest pain at onset

" Medical History

Prior intracerebral hemorrhage

Recent stroke

Recent head trauma or lossof consciousness

Recent myocardial infarction

" Surgical History

Recent surgical procedures

Arterial puncture

" Review of Systems

Gastrointestinal orgenitourinary bleeding

" Medications

Anticoagulant therapy

Continuum: Lifelong Learning Neurol 2008;14(6)

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

information. To the physician perform-ing the initial assessment, these activi-ties pose significant challenges.

Despite such barriers, critical ele-ments of the history may be obtainedindirectly. Emergency medical per-sonnel provide important informationregarding vital signs and blood glucoselevels obtained in the field. Observa-tions regarding level of consciousness,initial severity of deficits, and the pres-ence of bowel or bladder incontinenceat the scene provide useful clues tothe etiology of the presenting symp-toms. Family members provide impor-tant observations and additional his-tory. Reaching a family member bytelephone may be necessary if no oneis immediately available. In certain cir-cumstances, attempting to reach thepatient’s primary care provider mayprove useful. Documentation of prior

stroke, TIA, or other neurologic mor-bidity in medical records is valuable.Chronic or previously resolved deficitsmay potentially confound interpretationof neurologic examination findings inthe acute setting. The presence of ad-vanced dementia or other neurodegen-erative process may influence thedecision to pursue further aggressiveinterventions. Examination of medica-tion bottles, if they accompany thepatient, may provide clues to coexistingmedical conditions or anticoagulant use.

Neurologic Examination

The examination should focus onidentifying signs of lateralized hemi-spheric or brainstem dysfunction con-sistent with focal cerebral ischemia.Commonly encountered ischemic strokesyndromes are outlined in Table 1-2.The location of vascular occlusion and

15

TABLE 1-2 Common Ischemic Stroke Syndromes

Vascular Territory Signs and Symptoms

Left middle cerebral artery distribution Aphasia, right hemiparesis/hemisensory disturbance, righthomonymous hemianopia, left head and gaze preference

Right middle cerebral artery distribution Left hemispatial neglect, left hemiparesis/hemisensorydisturbance, left homonymous hemianopia, right headand gaze preference, anosognosia

Left posterior cerebral artery distribution Right visual field defect, impaired reading with intactwriting (alexia without agraphia), poor color naming, righthemisensory disturbance

Right posterior cerebral artery distribution Left visual field defect, visual neglect, left hemisensorydisturbance

Vertebrobasilar distribution Dizziness, vertigo,nausea,diplopia,quadriparesis, crossedmotoror sensory findings (ipsilateral face, contralateral body), truncalor limb ataxia, visual loss/dimming, impaired consciousness

Penetrating artery distribution(ie, lacunar syndromes)

(A) Internal capsule/corona radiata (A, B) Contralateral hemiparesis alone (pure motor stroke)OR contralateral hemiparesis + ataxia out of proportion toweakness (ataxic-hemiparesis); no cortical signs

(B) Ventral pons

(C) Thalamus (C) Contralateral sensory loss alone (pure sensory stroke);no cortical signs

Continuum: Lifelong Learning Neurol 2008;14(6)

KEY POINT

A The examination

should focus on

identifying

signs of

lateralized

hemispheric or

brainstem

dysfunction

consistent with

focal cerebral

ischemia.

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

the extent of collateral flow dictatewhether the complete or partial syn-drome is present. Frequently, impor-tant examination findings are observedwhile obtaining the history. Thus, levelof consciousness and the presenceof a gaze deviation, aphasia, neglect,or hemiparesis, may be establishedwithin minutes of the initial encounter.The NIH Stroke Scale (NIHSS) is avalidated 15-item scale that is used toassess key components of the stan-dard neurologic examination and mea-sure stroke severity (Lyden et al,1999). Although initially designed tomeasure clinical differences in experi-mental stroke therapy trials (Brott et al,1989), the NIHSS has gained wide-spread acceptance as a standard clinicalassessment tool. The scale assesseslevel of consciousness, ocular motility,facial and limb strength, sensory func-tion, coordination, language, speech, andattention. Scores range from 0 (normal)to 42 (maximal score) (Table 1-3). TheNIHSS may be performed rapidly andpredicts short-term and long-term neu-rologic outcomes (Adams et al, 1999).Even health care providers withoutexpertise in the neurologic examinationmay be trained to perform the assess-ment reliably with only a few hours ofinstruction (Goldstein and Samsa, 1997).Additionally, NIHSS severitymay provideinformation regarding the likelihoodof identifying a large vessel occlusionwith vascular imaging (Fischer et al,2005).

The NIHSS has important limita-tions. It does not include a detailedassessment of the cranial nerves, andrelatively low scores may occur inpatients with disabling brainstem orcerebellar infarction (Kasner, 2006).Likewise, milder deficits caused by fo-cal cerebral ischemia, such as impairedhand dexterity or fine finger move-ments, may escape detection if notspecifically tested. Stroke severity maynot be accurately reflected in nondom-

inant hemisphere syndromes as com-paredwith dominant hemisphere strokes(Woo et al, 1999), and a reliable scoreis often difficult to obtain in patientswith encephalopathy or cognitive dys-function. Clinically important changeson serial examination may not be re-flected as a measurable change on theNIHSS. Finally, one should recognizethat the presence of an abnormality onthe NIHSS does not support or re-fute a diagnosis of stroke. As discussedbelow, many other conditions maycause strokelike symptoms and NIHSSabnormalities.

Findings on general physical exam-ination may facilitate stroke diagnosisand influence treatment decisions. Asin any critically ill patient, the firstpriority is assessment and stabilizationof the patient’s airway, breathing, andcirculation. When performing the gen-eral examination, attention should befocused on the cardiovascular system.The presence of a cervical bruit, car-diac murmur, or irregularly irregularheart rhythm may provide an impor-tant clue to the underlying strokemechanism. Unequal extremity pulsesmay suggest aortic dissection or thepresence of concomitant peripheralarterial disease. Funduscopic examina-tion may reveal signs of chronic hy-pertensive arterial disease or endo-carditis. Signs of head or neck trauma(eg, contusions, lacerations) promptconsideration of occult cervical spineinjury. Other findings, such as rales onchest examination or bilateral asterixis,may suggest an alternate explanation(ie, pneumonia or metabolic enceph-alopathy) for the patient’s symptoms.

Ancillary Testing

Laboratory and cardiac evaluation sup-plement the clinical impression de-rived from the bedside assessment.Some conditions that may presentwith strokelike symptoms may be iden-tified based on laboratory results (eg,

16

Continuum: Lifelong Learning Neurol 2008;14(6)

KEY POINTS

A The NIH

Stroke Scale is

a validated,

15-item scale,

which is utilized

to assess key

components of

the standard

neurologic

examination

in patients

with stroke.

A Cardiac enzymes

and a 12-lead

EKG are

recommended

for all patients

with stroke.

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

hypoglycemia). In addition, abnormallaboratory values may exclude patientsfrom receiving thrombolytic therapy.Recently published guidelines recom-mend routine laboratory testing ofblood glucose, electrolytes, completeblood count, prothrombin time, activat-ed partial thromboplastin time, interna-tional normalized ratio, and renalfunction (Adams et al, 2007). Testingfor stool guaiac is not routinely recom-mended unless an indication exists (eg,melena or hematochezia). Initiatingtreatment with IV recombinant tissue-type plasminogen activator (rt-PA) pri-or to obtaining results of coagulationstudies may be safe and feasible (Sattinet al, 2006). In patients otherwiseeligible for thrombolytic therapy, theAmerican Heart Association/AmericanStroke Association (AHA/ASA) guide-lines support the decision to initiatetreatment prior to results of plateletor coagulation studies, unless a bleed-ing disorder or thrombocytopenia issuspected.

Cardiac abnormalities are commonin patients with stroke. Cardiac en-zymes and a 12-lead EKG are recom-mended for all stroke patients.Myocardial infarction and atrial fibrilla-tion are common causes of cardioemb-olism and are readily identified in theacute setting. The utility of routine chestradiography as part of the acute strokeevaluation is limited (Sagar et al, 1996)and currently not routinely recom-mended. As discussed previously, unlesswarranted by the presence of sudden,severe headache, there is no role forroutine CSF examination. Urine toxicol-ogy screen, blood alcohol level, arterialblood gas, or pregnancy tests may beindicated when the clinical history islimited or uncertain.

NEUROIMAGING

Brain imaging is the only reliablemeansto differentiate between ischemic and

17

Continuum: Lifelong Learning Neurol 2008;14(6)

TABLE 1-3 NIH Stroke Scalea

Category Scale Definition

1a. Level of consciousness 0 = Alert1 = Not alert, arousable2 = Not alert, obtunded3 = Unresponsive

1b. Questions 0 = Answers both correctly1 = Answers one correctly2 = Answers neither correctly

1c. Commands 0 = Performs both tasks correctly1 = Performs one task correctly2 = Performs neither task

2. Gaze 0 = Normal1 = Partial gaze palsy2 = Total gaze palsy

3. Visual fields 0 = No visual loss1 = Partial hemianopsia2 = Complete hemianopsia3 = Bilateral hemianopsia

4. Facial palsy 0 = Normal1 = Minor paralysis2 = Partial paralysis3 = Complete paralysis

5a. Left motor arm 0 = No drift1 = Drift before 10 seconds2 = Falls before 10 seconds3 = No effort against gravity4 = No movement

5b. Right motor arm 0 = No drift1 = Drift before 10 seconds2 = Falls before 10 seconds3 = No effort against gravity4 = No movement

6a. Left motor leg 0 = No drift1 = Drift before 5 seconds2 = Falls before 5 seconds3 = No effort against gravity4 = No movement

6b. Right motor leg 0 = No drift1 = Drift before 5 seconds2 = Falls before 5 seconds3 = No effort against gravity4 = No movement

7. Ataxia 0 = Absent1 = One limb2 = Two limbs

continued on next page

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

hemorrhagic stroke and is thereforemandatory prior to thrombolytic ther-apy (Besson et al, 1995; Mader andMandel, 1998). This chapter will coverthe basics of standard CT and MRIused in the emergency setting. Moreadvanced vascular and physiologicimaging will be discussed in anotherchapter.

CT

Noncontrast head CT is the studymost readily available in most strokecenters. CT is sensitive to intracranialblood and may be rapidly performedas part of the acute stroke evaluation.CT is also inexpensive and less sus-ceptible than MRI to artifact intro-duced by patient movement. In the

acute setting, early ischemic change(EIC) may be apparent on CT. Loss ofdifferentiation of the gray-white mat-ter interface, particularly in the regionof the insular cortex or the lentiformnucleus, may be indicative of earlycerebral ischemia. Sulcal effacement,representing focal tissue edema, maybe appreciated in areas of relativehypoperfusion and may be anotherearly indicator of ischemia. Whetherthese changes are present in theminutes to hours after symptom onsetis probably related to the severity andextent of ischemia, collateral circula-tion, and presence of large vesselocclusion. Detection of EIC is variable(Grotta et al, 1999) and is likely relatedto reader experience. One study iden-tified EICs in 75% of patients present-ing within 3 hours of symptom onset(Barber et al, 2000), and an evenhigher prevalence was observedwithin 6 hours in patients with hemi-spheric strokes (von Kummer et al,1996). The presence of EICs involvinggreater than one-third of the middlecerebral artery (MCA) territory wasused as an exclusion criterion in earlyclinical trials of thrombolytics in aneffort to minimize the risk of hemor-rhagic complications (Hacke et al,1995). The presence of EICs, however,was not independently associated withadverse outcome after rt-PA treatmentin the National Institute of NeurologicalDisorders and Stroke (NINDS) trial andtherefore should not preclude throm-bolytic therapy in otherwise eligiblepatients (Patel et al, 2001).

Increased CT attenuation withinan arterial segment, the ‘‘hyperdense’’artery sign, is an occasional finding. Itis observed most commonly in theMCA and is associated with occlusivethrombus within the vessel lumen(Tomsick et al, 1989) (Figure 1-1).Although the hyperdense MCA sign isfairly specific for vascular occlusion(Bastianello et al, 1991), the sensitivity

18

TABLE 1-3 Continued

Category Scale Definition

8. Sensory 0 = Normal1 = Mild loss2 = Severe loss

9. Language 0 = Normal1 = Mild aphasia2 = Severe aphasia3 = Mute or global aphasia

10. Dysarthria 0 = Normal1 = Mild2 = Severe

11. Extinction/inattention 0 = Normal1 = Mild2 = Severe

aThe full NIH Stroke Scale with instructions and scoring sheet isavailable online at www.ninds.nih.gov/doctors/NIH_Stroke_Scale.pdf.Accessed August 20, 2008.

Reprinted with permission from Adams HP Jr, del Zoppo G, Alberts MJ, et al. Guidelinesfor the early management of adults with ischemic stroke: a guideline from the AmericanHeart Association/American Stroke Council, Clinical Cardiology Council, CardiovascularRadiology and Intervention Council, and the Atherosclerotic Peripheral Vascular Diseaseand Quality of Care Outcomes in Research Interdisciplinary Working Groups: theAmerican Academy of Neurology affirms the value of this guideline as an educationaltool for neurologists [published errata appears in Stroke 2007;38(6):e38 and Stroke2007;38(9):e96]. Stroke 2007;38(6):1655–1711.Data fromNational Institute ofNeurological Disorders and Stroke. [accessed July 28, 2008]Available at www.ninds.nih.gov/doctors/NIH_Stroke_Scale.pdf. Accessed August 20.

Continuum: Lifelong Learning Neurol 2008;14(6)

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

of this finding is low. The MCA ‘‘dotsign’’ seen in the sylvian fissure withocclusion of distal MCA branches(Barber et al, 2001) and the hyper-dense basilar artery sign in patientswith basilar artery thrombosis (Ehsanet al, 1994) have also been described.A recent systematic review of studiesreporting on early CT signs in acuteischemic stroke found interobserveragreement to be moderate to poorand increased likelihood of poor func-tional outcome when early infarctionsigns were present (Wardlaw andMeilke, 2005). An example of earlyischemic changes is demonstrated inCase 1-1.

The appearance of ischemic changeson CT evolves over time. Within 12 to24 hours, an indistinct area of lowdensity becomes apparent in theaffected vascular distribution. After 24hours, the ischemic region becomesincreasingly hypodense and bettercircumscribed. Mass effect develops

and results in sulcal asymmetry orventricular distortion. The presence ofa clearly delineated area of hypoden-sity with associated mass effect should,therefore, prompt reassessment of thetime of symptom onset in patientsthought to be eligible for thrombolytictherapy, as distinct hypodensity is in-consistent with focal cerebral ischemiaof less than 3 hours’ duration.

Neurologists involved in the eval-uation of patients with acute strokemust be sufficiently skilled to identifyradiographic contraindications to throm-bolytic therapy. Generally speaking, a‘‘central to peripheral’’ approach tovisual inspection of CT images may helpto rapidly identify nonischemic causesof strokelike symptoms and to identifysubtle EIC. First, inspection of themidline structures (ie, ventricles andbasal cisterns) for shift may help rap-idly identify space-occupying mass le-sions, which may require urgent sur-gical intervention. The basal cisterns,

19

Continuum: Lifelong Learning Neurol 2008;14(6)

KEY POINTS

A Brain imaging is

the only reliable

means to

differentiate

between

ischemic and

hemorrhagic

stroke and is

therefore

mandatory prior

to instituting

thrombolytic

therapy.

A CT is sensitive

to intracranial

blood, is widely

available, and

may be rapidly

performed as

part of the acute

stroke evaluation.

A Loss of

differentiation

of the gray-white

matter interface,

particularly in the

region of the

insular cortex or

the lentiform

nucleus, may be

indicative of

early cerebral

ischemia.

A Sulcal

effacement,

representing

focal tissue

edema, may be

appreciated in

areas of relative

hypoperfusion

and may be

another early

indicator of

ischemia.

FIGURE 1-1 Noncontrast head CT from a 74-year-old woman with the abrupt onsetof dysarthria and left-sided weakness. The hyperdense right middle cerebralartery (arrows) is visible on these axial images at two sequential levels (A, B).Follow-up cerebral angiography confirmed occlusion of the proximal rightmiddle cerebral artery.

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

interhemispheric fissure, and sylvianfissures should be scrutinized forsubarachnoid blood. The fourth ven-tricle should be identified, particularlyin patients with posterior circulationsymptoms. Although CT evaluation ofthe posterior fossa may be challengingbecause of technical limitations, dis-tortion of the fourth ventricle mayresult from mass effect related tocerebellar infarction and often heraldsobstructive hydrocephalus due toevolving edema. Examination of theperiphery of the brain parenchyma forextraaxial collections should follow.

The brain parenchyma should approx-imate the inner table of the skull in afairly symmetric fashion, acknowledg-ing that atrophy may occur in a regionaldistribution. As with ischemic changes,the appearance of subdural collectionsevolves with time and these may appearisodense to brain parenchyma. Finally,careful inspection for early ischemicchanges (see above) may be guidedby localizing information from theclinical examination. Effective andtimely administration of acute stroketherapy relies on accurate bedside as-sessment and exclusion of intracranial

20

Continuum: Lifelong Learning Neurol 2008;14(6)

KEY POINTS

A Increased CT

attenuation

within an

arterial

segment, the

‘‘hyperdense’’

artery sign, is

an occasional

finding in acute

ischemic stroke.

A The appearance

of ischemic

changes on

CT evolves

over time.

Case 1-1A 48-year-old man was seen by the acute stroke intervention team after hehad developed left-sided weakness. He was last known to be normal2 hours earlier. The patient reported mild headache but otherwise deniedany problems.

Neurologic examination demonstrated a right head and gazepreference, left homonymous hemianopsia, left lower facial weakness,dysarthria, and dense left hemiparesis with absent sensation in the left armand leg. The patient did not acknowledge the presence of his left-sidedweakness. His NIHSS score was 14. Noncontrast head CT demonstrated alarge area of subtle hypodensity with loss of the gray-white interface in theinsular ribbon (white arrow) consistent with right MCA distributionischemia. Sulcal effacement and blurring of the deep nuclei (arrowhead)are also seen (Figure 1-2).

Comment. Thiscase exemplifiesearly ischemicchanges that maybe seen with CTin the early hoursafter stroke onset.The clinicalsyndrome fitwell with anondominanthemisphereischemic event,and the earlyischemic changesappreciable onnoncontrast CThelped confirmthe clinical assessment. The patient underwent emergent vascular imagingand was found to have a right internal carotid artery occlusion.

FIGURE 1-2

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

hemorrhage. Subtle changes of earlyischemia, while helpful in confirmingclinical suspicion, are often identifiedwith the assistance of a radiologist andshould not directly influence therapeu-tic decisions.

MRI

Evaluation of patients with acutestroke with MRI has clear advantages.Compared with CT, MRI with diffu-sion-weighted imaging (DWI) se-quences is more sensitive for acutecerebral ischemia and improves diag-nostic accuracy (Fiebach et al, 2002).DWI may detect abnormalities withinminutes after onset of cerebral ische-mia (Hjort et al, 2005) and delineatesthe location, size, and extent ofhyperacute ischemia. MRI better eval-uates the posterior fossa and improvesvisualization of small cortical infarc-tions (Figure 1-3).

Historical concerns about the ability ofMRI to identify acute intracerebral hem-orrhage have been addressed by sev-eral studies. Conventional T1-weightedand T2-weighted MRI pulse sequencesare able to identify subacute andchronic blood, but they are less sensi-tive for parenchymal hemorrhage dur-ing the first 6 hours after symptomonset. Susceptibility-weighted MRI, orgradient-recalled echo (GRE), se-quences have improved sensitivity forrecently extravasated blood products(Patel et al, 1996). A prospective studyof MRI and CT performedwithin 6 hoursof stroke symptom onset demonstratedthat the accuracy of GRE sequencesfor acute hemorrhage is equal to thatof CT (Kidwell et al, 2004). A smallermulticenter study found a similarlyhigh accuracy of GRE for identifica-tion of acute intracerebral hemorrhage(Fiebach et al, 2004).

Some centers have developed ex-tensive experience with MRI in acutestroke and have adopted the use ofMRI protocols for routine evaluation

of patients with stroke. A recent pro-spective study performed at such acenter demonstrated the superiorityof MRI for detection of acute stroke inthe full spectrum of patients whopresented for emergency assessmentof strokelike symptoms (Chalela et al,2007). Based on these results, theauthors have advocated the use ofMRI as the sole modality in the eval-uation of patients with suspectedstroke. The utility of MRI in detectingsubarachnoid hemorrhage has notbeen rigorously evaluated, and onlylimited data are available (Wiesmannet al, 2002). The previous studies ofMRI for the acute evaluation of strokewere performed using 1.5 tesla (T)magnets. Further studies will be need-ed to clarify the role of newer gener-ation 3T MRI scanners for hyperacutestroke diagnosis.

21

Continuum: Lifelong Learning Neurol 2008;14(6)

FIGURE 1-3 CT and MRI images from a 39-year-old manwho developed sudden speech difficulty. TheCT images (top row) demonstrate hypodensity

in the posteroinferior left frontal (top left, white arrow) lobe andadjacent superior left temporal lobe as well as loss of gray-whitejunction and sulcal effacement in the posterior parietal region(top right, white arrow). MR diffusion-weighted images (bottomrow) demonstrate the abnormalities clearly (bright signal).

KEY POINTS

A Diffusion-

weighted imaging

is more sensitive

for detection of

acute cerebral

ischemia and

improves

diagnostic

accuracy.

A A prospective

study ofMRI and

CT performed

within 6 hours

of stroke

symptom onset

demonstrated

that the

accuracy of

gradient-recalled

echo sequences

for acute

hemorrhage is

equal to that

of CT.

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

The vast majority of emergencydepartments lack the resources nec-essary to perform emergent MRI.The costs of the technology, includingaround-the-clock technician support,are prohibitive for many centers andhave limited widespread implemen-tation. Abbreviated stroke MRI proto-cols have been developed to ad-dress concerns about additional timeneeded to acquire MRI images com-pared with CT. This issue is importantgiven the association between time toinitiation of thrombolytic therapy andlikelihood of an excellent neurologicoutcome (Marler et al, 2000). MRI iscontraindicated in patients with pace-makers or other metallic hardwareand is further limited by its suscepti-bility to motion artifact in agitatedpatients.

STROKE MIMICS

When acute ischemic stroke is sus-pected, it is crucial to consider and toexclude alternative diagnoses, espe-cially intracranial hemorrhage. Manyconditions, including systemic abnor-malities and other nervous systemdiseases, present with focal neurologicdeficits that ‘‘mimic’’ acute ischemicstroke. Table 1-4 lists commonlyencountered stroke mimics. Somestroke mimics may be discovered earlyduring the course of evaluation (eg,hypoglycemia), but others may re-quire more extensive investigationand/or neuroimaging (Case 1-2). Thehistory and examination help deter-mine the probability of a stroke mimicas the cause of neurologic dysfunc-tion. Distinguishing features of somestroke mimics are highlighted inTable 1-5.

Early studies found the frequency ofstroke mimics range from 1% (O’Brienet al, 1987) to 19% (Libman et al, 1995)in patients with suspected or initiallydiagnosed stroke. The time of patientassessment relative to symptom onset,

examiner experience, and availabilityof imaging results at the time of initialdiagnosis were variable in these stud-ies. A recent prospective study of morethan 300 patients who presented toan urban teaching hospital with sus-pected stroke found mimics in 31%at the time of final diagnosis (Handet al, 2006). The most frequent mimicswere postictal deficits (21%), sepsis(13%), and toxic-metabolic disturban-ces (11%). Seventy-five percent ofmimics in the study were neurologicdisorders, and 42% of patients with amimic had experienced a previousstroke. Eight variables independentlyassociated with a correct diagnosiswere identified. The most powerfulpredictors of an accurate stroke diag-nosis were ‘‘definite history of focalneurologic symptoms’’ (odds ratio[OR] 7.21; 95% confidence interval[CI], 2.48–20.93) and an NIHSS scoregreater than 10 (OR 7.23; 95% CI,2.18–24.05). In patients with knowncognitive impairment, the likelihoodof having a stroke was markedlyreduced (OR 0.33; 95% CI, 0.14–0.76).

While studies such as these improveour general understanding of the

22

Continuum: Lifelong Learning Neurol 2008;14(6)

TABLE 1-4 Common AcuteStroke Mimics

" Postictal deficits(Todd paralysis)

" Hypoglycemia

" Migraine(hemiplegic, with aura)

" Hypertensive encephalopathy

" Reactivation of prior deficits

" Mass lesions

" Subarachnoid hemorrhage

" Peripheral vestibulopathy

" Conversion reaction

KEY POINTS

A When acute

ischemic stroke

is suspected,

it is crucial to

consider and

to exclude

alternative

diagnoses,

especially

intracranial

hemorrhage.

A Many conditions,

including

systemic

abnormalities

and other

nervous system

diseases, present

with focal

neurologic

deficits that

‘‘mimic’’ acute

ischemic stroke.

A A recent

prospective

study of more

than 300

patients who

presented to an

urban teaching

hospital with

suspected

stroke found

mimics in 31%

at the time of

final diagnosis.

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Case 1-2An 82-year-old woman was seen by the acute stroke intervention team forthe sudden onset of speech difficulty 90 minutes earlier. She had beenworking with a physical therapist at home when she became unable tospeak. There was no associated weakness, alteration of consciousness, orheadache. Per report, she had experienced a minor ‘‘stroke’’ approximately2 weeks earlier but had made some improvement.

The woman was afebrile, her initial blood pressure was 142/72 mm Hg,and the finger-stick glucose level was 188 mg/dL. She was awake, alert,and appropriate. Language examination was remarkable for impairedfluency with the ability to say only fragments of words. She was able tofollow simple midline commands but was unable to follow complexcommands. She was unable to repeat, read, or name objects. There wasno limb weakness or sensory disturbance. Her NIHSS score was 6.

Noncontrast head CT demonstrated a subtle hyperdense lesion with masseffect involving the left temporoparietal region (Figure 1-4, top, arrows). Giventhe radiographic findings suggestive of an underlying structural lesion, thepatient did not receive thrombolytic therapy. Follow-up MRI demonstrated anill-defined enhancing lesion involving the white matter and cortex of the leftparietal lobe suggestive of a low-grade neoplasm (Figure 1-4, bottom, arrows).

Comment. This case is an example of a stroke mimic. The abrupt onsetof symptoms might not prompt initial consideration of an underlyingstructural lesion as a potential etiology. However, one study found that 6%of patients with brain tumors presenting to an emergency departmenthad symptoms of less than 1 day’s duration (Snyder et al, 1993). Suddenonset of focal symptoms in patients with either diagnosed or undiagnosedtumors may result from seizures, hemorrhage into the tumor, or obstructivehydrocephalus caused by increasing mass effect.

FIGURE 1-4

23

Continuum: Lifelong Learning Neurol 2008;14(6)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

frequency and nature of strokemimics, the results are less applicableto individual patients. As emphasizedpreviously, the diagnosis of stroke isbased on a composite of informationobtained from the history and thepattern of findings on physical exam-ination. A single symptom or signcannot be used to rule in or rule outthe diagnosis (Goldstein, 2006).

STROKE CHAMELEONS

The clinician should also be aware ofcommon atypical stroke presenta-tions. Recognition of these entities isimportant so as not to miss an op-portunity to offer treatment to anotherwise eligible patient with stroke.These patients may not be triagedinto acute stroke pathways and, there-fore, may be at higher risk of mis-diagnosis. The term stroke chameleonhas been aptly used to describe an

atypical stroke presentation that ap-pears to mimic another disease pro-cess (Huff, 2002). The clinician shouldsuspect such problems when symptomonset is abrupt or occurs in patientswith risk factors for cerebrovasculardisease.

A small proportion of patients withstroke may present with symptomssuggestive of an acute confusional state(eg, delirium). While encephalopathytypically reflects diffuse hemisphericdysfunction, a ‘‘pseudo-encephalopathy’’may occur with focal cerebral ischemiainvolving the limbic cortex or orbito-frontal regions. ‘‘Confusion’’ may alsobe reported in patients with fluentaphasia or neglect syndromes withoutaccompanying motor deficits. System-atic neurologic examination shouldidentify these focal features and in-crease the clinical suspicion of stroke.Likewise, examination of visual fieldswill avoid overlooking patients with

24

Continuum: Lifelong Learning Neurol 2008;14(6)

TABLE 1-5 Characteristics of Common Stroke Mimics

Diagnosis Comments

Seizure (postictal) Focal deficits likely are caused by seizure-inducedneuronal dysfunction (reversible). May occur with simplepartial or generalized seizures. Clinical seizure is oftenunwitnessed or unrecognized. Spontaneous resolutionoccurs over hours (may last up to 48 hours).

Hypoglycemia Aphasia or hemiplegia may be present. Variabledrowsiness or obtundation. Blood glucose usually<45 mg/dL. Resolution of symptoms (immediate!hours)with IV glucose.

Metabolicencephalopathy

Etiologies include hyperosmolar hyperglycemia,hyponatremia, and hepatic encephalopathy. May beassociated with altered level of consciousness, poorattention, or disorientation (eg, delirium) asterixis.

Conversion reaction Diagnosis of exclusion. Conversion disorder is the mostcommon psychiatric diagnosis. Comorbid psychiatricproblems are common. Paresis, paralysis, and movementdisorders are common.

Reactivation ofprior deficits

Imaging evidence or history of remote stroke is oftenapparent. Previous deficit may have resolved completely.

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

cortical blindness or visual neglectsyndromes.

Chest pain or discomfort mimickingmyocardial ischemia has been reportedin patients with infarction of the thal-amus, corona radiata, or lateral medulla(Gorson et al, 1996). In some of thesepatients, the sensory symptoms werepart of a more extensive stroke syn-drome, but the clinician should beaware of this possibility. Distal armparesis with patterns of weaknessconforming to peripheral nerve distri-butions may result from focal cerebralischemia. Radial or ulnar involvementhas been reported with small corticalinfarction of the motor cortex (Gasset al, 2001). Again, abrupt onset andthe presence of vascular risk factorsshould alert the astute clinician.

CONCLUSION

Timely and accurate diagnosis of acuteischemic stroke in the emergency set-ting relies on eliciting a focused his-tory, performing an efficient, thorough,neurologic examination, and inter-preting the results of laboratory andneuroimaging studies. The suddenonset of focal neurologic symptomsin a recognizable arterial distributionis the hallmark of stroke. Accuratelydetermining time of symptom onset,measuring the severity of neurologicdeficit, and excluding nonischemiccauses of strokelike symptoms maylead to identification of patients eligi-ble for acute stroke therapy. An effi-cient and systematic approach tostroke diagnosis facilitates evidence-based treatment.

REFERENCES

Adams HP Jr, Davis PH, Leira EC, et al. Baseline NIH Stroke Scale score strongly predictsoutcome after stroke: a report of the Trial of Org 10172 in Acute Stroke Treatment(TOAST). Neurology 1999;53(1):126–131.

Adams HP Jr, del Zoppo G, Alberts MJ, et al. Guidelines for the early management ofadults with ischemic stroke: a guideline from the American Heart Association/AmericanStroke Council, Clinical Cardiology Council, Cardiovascular Radiology and InterventionCouncil, and the Atherosclerotic Peripheral Vascular Disease and Quality of CareOutcomes in Research Interdisciplinary Working Groups: the American Academy ofNeurology affirms the value of this guideline as an educational tool for neurologists[published errata appear in Stroke 2007;38(6):e38 and Stroke 2007;38(9):e96]. Stroke2007;38(6):1655–1711.

Barber PA, Demchuk AM, Hudon ME, et al. Hyperdense sylvian fissure MCA ‘‘dot’’ sign:A CT marker of acute ischemia. Stroke 2001;32(1):84–88.

Barber PA, Demchuck AM, Zhang J, Buchan AM. Validity and reliability of a quantitativecomputed tomography score in predicting outcome of hyperacute stroke beforethrombolytic therapy. ASPECTS Study Group. Alberta Stroke Programme Early CT Score.Lancet 2000;355(9216):1670–1674.

Bastianello S, Pierallini A, Colonnese C, et al. Hyperdense middle cerebral artery CTsign. Comparison with angiography in the acute phase of ischemic supratentorialinfarction. Neuroradiology 1991;33(3):207–211.

25

Continuum: Lifelong Learning Neurol 2008;14(6)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Besson G, Robert C, Hommel M, Perret J. Is it clinically possible to distinguishnonhemorrhagic infarct from hemorrhagic stroke? Stroke 1995;26(7):1205–1209.

Brott T, Adams HP Jr, Olinger CP, et al. Measurements of acute cerebral infarction: aclinical examination scale. Stroke 1989;20(7):864–870.

Chalela JA, Kidwell CS, Nentwich LM, et al. Magnetic resonance imaging andcomputed tomography in emergency assessment of patients with suspected acutestroke: a prospective comparison. Lancet 2007;369(9558):293–298.

Edlow JA, Caplan LR. Avoiding pitfalls in the diagnosis of subarachnoid hemorrhage.N Engl J Med 2000;342(1):29–36.

Ehsan T, Hayat G, Malkoff MD, et al. Hyperdense basilar artery. An early computedtomography sign of thrombosis. J Neuroimaging 1994;4(4):200–205.

Fiebach JB, Schellinger PD, Gass A, et al. Stroke magnetic resonance imaging is accurate inhyperacute intracerebral hemorrhage: a multicenter study on the validity of strokeimaging. Stroke 2004;35(2):502–506.

Fiebach JB, Schellinger PD, Jansen O, et al. CT and diffusion-weighted MR imagingin randomized order: diffusion-weighted imaging results in higher accuracy andlower interrater variability in the diagnosis of hyperacute ischemic stroke. Stroke2002;33(9):2206–2210.

Fischer U, Arnold M, Nedeltchev K, et al. NIHSS score and arteriographic findings inacute ischemic stroke. Stroke 2005;36(10):2121–2125.

Gass A, Szabo K, Behrens S, et al. A diffusion-weighted MRI study of acute ischemicdistal arm paresis. Neurology 2001;57(9):1589–1594.

Goldstein LB. Improving the clinical diagnosis of stroke. Stroke 2006;37(3):754–755.

Goldstein LB, Samsa GP. Reliability of the National Institutes of Health StrokeScale: extension to non-neurologists in the context of a clinical trial. Stroke1997;28(2):307–310.

Gorson KC, Pessin MS, DeWitt LD, et al. Stroke with sensory symptoms mimickingmyocardial ischemia. Neurology 1996;46(2):548–551.

Grotta JC, Chiu D, Lu M, et al. Agreement and variability in the interpretation ofearly CT changes in stroke patients qualifying for intravenous rtPA therapy. Stroke1999;30(8):1528–1533.

Hacke W, Kaste M, Fieschi C, et al. Intravenous thrombolysis with recombinant tissueplasminogen activator for acute hemispheric stroke: the European Cooperative AcuteStroke Study (ECASS). JAMA 1995;274(13):1017–1025.

Hand PJ, Kwan J, Lindley RI, et al. Distinguishing between stroke and mimic at thebedside: the brain attack study. Stroke 2006;37(3):769–775.

Hjort N, Christensen S, Solling C, et al. Ischemic injury detected by diffusion imaging11 minutes after stroke. Ann Neurol 2005;58(3):462–465.

Huff JS. Stroke mimics and chameleons. Emerg Med Clin N Am 2002;20(3):583–595.

26

Continuum: Lifelong Learning Neurol 2008;14(6)

"DIAGNOSIS AND MIMICS

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Kasner SE. Clinical interpretation and use of stroke scales. Lancet Neurol2006;5(7):603–612.

Kidwell CS, Chalela JA, Saver JL, et al. Comparison of MRI and CT for detection ofacute intracerebral hemorrhage. JAMA 2004;292(15):1823–1830.

Libman RB, Wirkowski E, Alvir J, Rao TH. Conditions that mimic stroke in theemergency department. Implications for acute stroke trials. Arch Neurol1995;52(11):1119–1122.

Lyden P, Lu M, Jackson C, et al. Underlying structure of the National Institutesof Health Stroke Scale: results of a factor analysis. NINDS tPA Stroke Trial Investigators.Stroke 1999;30(11):2347–2354.

Mader TJ, Mandel A. A new clinical scoring system fails to differentiate hemorrhagicfrom ischemic stroke when used in the acute care setting. J Emerg Med 1998;16(1):9–13.

Marler JR, Tilley BC, Lu M, et al. Early stroke treatment associated with better outcome:the NINDS rt-PA stroke study. Neurology 2000;55(11):1649–1655.

O’Brien PA, Ryder DQ, Twomey C. The role of computed tomography brain scan in thediagnosis of acute stroke in the elderly. Age Ageing 1987;16(5):319–322.

Patel MR, Edelman RR, Warach S. Detection of hyperacute primary intraparenchymalhemorrhage by magnetic resonance imaging. Stroke 1996;27(12):2321–2324.

Patel SC, Levine SR, Tilley BC, et al. Lack of clinical significance of early ischemicchanges on computed tomography in acute stroke. JAMA 2001;286(22):2830–2838.

Sagar G, Riley P, Vohrah A. Is admission chest radiography of any clinical value inacute stroke patients? Clin Radiol 1996;51(7):499–502.

Sattin JA, Olson SE, Liu L, et al. An expedited code stroke protocol is feasible and safe.Stroke 2006;37(12):2935–2939.

Snyder H, Robinson K, Shah D, et al. Signs and symptoms of patients with brain tumorspresenting to the emergency department. J Emerg Med 1993;11(3):253–258.

Tomsick TA, Brott TG, Olinger CP, et al. Hyperdense middle cerebral artery: incidence andquantitative significance. Neuroradiology 1989;31(4):312–315.

von Kummer R, Nolte PN, Schnittger H, et al. Detectability of cerebral hemisphereischaemic infarcts by CT within 6 h of stroke. Neuroradiology 1996;38(1):31–33.

Wardlaw JM, Mielke O. Early signs of brain infarction at CT: observer reliabilityand outcome after thrombolytic treatment—systematic review. Radiology2005;235(2):444–453.

Wiesmann M, Mayer TE, Yousry I, et al. Detection of hyperacute subarachnoidhemorrhage of the brain by using magnetic resonance imaging. J Neurosurg2002;96(4):684–689.

Woo D, Broderick JP, Kothari RU, et al. Does the National Institutes of Health StrokeScale favor left hemisphere strokes? NINDS t-PA Stroke Study Group. Stroke1999;30(11):2355–2359.

Continuum: Lifelong Learning Neurol 2008;14(6)

27

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.