Jonathon M. Sullivan

Cerebrovascular Disorders (11.1)

Cerebral Aneurysm (11.1.1.1)

Most intracerebral aneurysms (ICAs) are clinically silent, and have a benign clinical course. The cited prevalence of ICA ranges from 3.6 – 6%, and ICA is an incidental finding in up to 8.9% of autopsy cases. The most important clinical manifestation is rupture of the aneurysm, which results in intracranial hemorrhage (ICH), with potentially devastating results. The etiology of saccular ICA, the most common cause of aneurysm-related ICH, remains a matter of some controversy. Although the vast majority of aneurysms appear to be congenital, other risk factors such as uncontrolled hypertension, tobacco and alcohol use, and atherosclerosis appear to play a part in the natural history and risk of rupture of these lesions. Because most unruptured ICAs are asymptomatic, diagnosis and management are clinically challenging. Treatment options for ICA depend on location, size, and the condition and age of the patient, and include endovascular coiling or aneurysmal clipping to ablate the aneurysm.

Etiology/Anatomy/Pathophysiology

Saccular or “berry” aneurysms are the most common form of ICA, and account for over three-fourths of all cases of subarachnoid hemorrhage (SAH). The pathogenesis of these aneurysms remains unclear. A potentially confounding issue in understanding this debate is discriminating between factors that lead to the initial formation of the aneurysm and those that promote the development and eventual rupture of the aneurysm. Based on the available evidence, it seems clear that genetic factors predominate in the formation of saccular ICAs, while environmental and behavioral factors have a greater impact on development and risk of rupture. Risk factors invoked for the formation of saccular ICA include female sex and alterations in the genes for polycystin, fibrillin, collagen III, elastin, collagen IV, protease inhibitor or a1-antitrypsin and various cellular proteases. Recently, Farnham et al reported a linkage of ICA with chromosomal region 7q11 in a genomic search of 85 Japanese nuclear families with at least two effected siblings; this chromosomal region includes the elastin gene. Risk factors that may contribute to the development and eventual rupture of ICAs include smoking, arterial hypertension, alcohol use, and increasing age.

The most common sites for saccular intracranial aneurysms are at bifurcations of the anastomotic network known as the Circle of Willis. Basilar artery ICAs, at the bifurcations to the vertebral or posterior communicating arteries, account for 5 – 10%. Anterior circulation aneurysms are more common, at bifurcations of the internal carotid (40%), middle cerebral (20%), and anterior communicating arteries (30%). This observation has led several authors to hypothesize that hemodynamic stress at the intimal cusp of the arterial bifurcation, combined with predisposing genetic factors, leads weakening of the media, with consequent aneurysm formation. Futami et al found no evidence for such an etiology. However, their experimental design incorporated an animal model and artificially induced medial defects, and should be interpreted with caution.

The International Study of Unruptured Intracranial Aneurysms was a multicenter investigation of natural history and clinical outcome in 4060 patients with unruptured ICA. Their findings for the 5-year cumulative rupture rates for patients who did not have a history of SAH are summarized in Table 1. These results must be interpreted with some caution, as the population under study included both treated and untreated cohorts, limited followup for half the patients studied, and probably selection bias. The findings nevertheless emphasize that aneurysm size and location have a significant role in determining the risk of future rupture. The authors noted that in patients with unruptured intracranial aneurysms of less than 7mm in diameter without a history of previous SAH from a different aneurysm, the annual rupture rate is approximately 0.1%, which makes it difficult to improve on the natural history of these lesions, given the morbidity and mortality associated with currently available techniques of invasive therapy.

	< 7mm		7-12mm	13-24 mm	> 25 mm
	Gr 1	Gr 2
Cavernous Carotid artery	0	0	0	3.0%	6.5%
AC/MC/IC	0	1.5%	2.6%	14.5%	40%
Post –P comm	2.5%	3.4%	14.5%	18.4%	50%

Table 11-1. 5-year cumulative rupture rates according to size and location of unruptured ICA. Findings of the International Study of Unruptured Intracranial Aneurysms Investigators. Group 1 = patients without previous history of SAH. Group 2 = patients with history of previous SAH from a different aneurysm. AC=anterior communicating or anterior cerebral artery; IC=internal carotid artery (not cavernous carotid artery); MC=middle cerebral artery; Post—P comm=vertebrobasilar, posterior cerebral arterial system, or posterior communicating artery. Used by permission.

Some unruptured ICAs will produce symptoms by compressing neural or vascular structures, leading to headache, visual field deficits, or cranial nerve palsies. However, the vast majority of these lesions are clinically silent unless and until rupture occurs .

Emergency Department Evaluation

Although definitive data is lacking, it is probably safe to assume that a large percentage of ICAs diagnosed in the emergency department are incidental findings. The small number of patients who present with complaints related to unruptured ICA will usually have mild or vague symptoms, and clinical evaluation can be challenging. Although most aneurysms do not go on to rupture, those that do cause symptoms tend to be larger, with a higher risk of progressing to SAH. The emergency physician should maintain a high index of suspicion for intracranial pathology, including ICA, in patients presenting with headache and other neurological complaints.

Symptoms associated with ICA are wide-ranging and include headache, eye pain, diplopia or visual field deficits, blurry vision, facial pain, and cranial nerve palsies. Differential diagnosis can obviously be quite extensive for such presentations, including acute ischemic stroke, SAH or other intracranial hemorrhage, encephalitis, meningitis, Bell’s palsy, carotid artery dissection, temporal arteritis, myasthenia gravis, multiple sclerosis, botulism, retinal artery occlusion, and intracranial neoplasm. Patient presentation, combined with a careful history and thorough physical examination, will help to clarify the differential. However, evaluation for suspected intracranial lesions, including ICA, will ultimately rest on diagnostic imaging.

History should include questions regarding the onset and duration symptoms; what the patient was doing when symptoms were first noticed; associated symptoms such as fever, nausea and vomiting, and visual changes; history of head trauma; history of previous neurological or cardiovascular disorders; a family history of cerebrovascular disease including ICA, and whether the patient uses tobacco, alcohol or illicit drugs, especially cocaine and other sympathomimetics.

Physical examination should obviously emphasize a search for neurological deficits, and in particular the physician suspecting unruptured ICA should be on the lookout for cranial nerve palsies. A careful evaluation of extraocular movements is mandatory, as this examination may indicate possible aneurysmal compression of CNs III, IV and VI. The physician should also perform visual field testing by confrontation to determine whether quadrantanopsia or hemianopsia are present; these conditions indicate possible compression of the optic chiasm or optic tract, respectively. All neurologic examinations should incorporate testing of muscle strength, deep tendon reflexes, cerebellar function (Romberg, rapid alternating movements), and gait.

All patients with suspected intracranial lesions, including ICA, require brain imaging. CT angiography can be performed in most emergency departments, with sensitivity ranging from 94 – 98 % for ICAs larger than 5mm in diameter. However, for lesions smaller than 5mm, sensitivity drops to around 60%. Similar values hold for MR angiography. White et al performed a systematic review to determine the accuracy of CT angiography, MR angiography and transcranial Doppler ultrasonography in depicting intracranial aneurysms. They found that CT angiography and MR angiography had overall accuracies per aneurysm of 89% and 90% respectively, with similar accuracies for anterior and posterior circulation aneurysms. The authors went on to perform a clinical trial of CT angiography and MR angiography, using intra-arterial digital subtraction angiography as a gold standard, and found similar results.

Given this data, and with what is known about the natural history of ICA, it is reasonable to suggest that the clinical approach to patients with suspected ICA in the emergency department should incorporate CT angiography. CT is less expensive than MR and has comparable accuracy. Both studies will miss small aneurysms, but these lesions have much lower rates of rupture and available invasive interventional strategies are unlikely to positively impact upon their natural history. Stable patients with suspected unruptured ICA and negative CT angiography should be referred for further outpatient evaluation, which may include additional invasive or noninvasive imaging at the discretion of the neurologist or neurosurgeon.

Of those patients found to have ICA in the emergency department, many will not require hospitalization or urgent angiography. However, these decisions should be made in consultation with a neurosurgeon, and immediate referral is highly recommended.

Conventional therapy for unruptured ICA is occlusion of the aneurysm by means of vascular clipping at surgery. In the last decade, however, occlusion of ICA by percutaneous intravascular coiling has come into increasing use. The relative risks and benefits of these two procedures, and whether any invasive intervention should be used for aneurysms below a certain size, are subjects of ongoing debate and investigation.

Arteriovenous Malformation (11.1.1.2)

Arteriovenous malformations (AVMs) are congenital vascular anomalies in which the arterial and venous systems are connected directly, without an intervening capillary bed. These abnormal vascular networks are characterized by dilated torturous vessels and associated abnormal brain parenchyma. Most AVMs will hemorrhage, and most SAH arising from AVM occurs before the age of 40. As with unruptured ICA, the presentation of unruptured AVM can be subtle, and diagnosis of this lesion is clinically challenging.

Etiology/Anatomy/Pathophysiology

An AVM is an abnormal arteriovenous shunt through a plexus of dilated, tortuous vessels. Hypertrophic arterial feeding vessels connect directly to thickened, dilated veins without an intervening microvascular bed. Due in part to the inadequate perfusion in the territory of an AVM, the surrounding or underlying parenchyma is usually abnormal, with nonfunctional neurons and gliosis being prominent histological features. Both the abnormal parenchyma and the abnormal vascular morphology contribute to the spectrum of clinical effects produced by AVM, which include headache, neurological deficit, seizure and hemorrhage.

The natural history of AVM differs considerably from that of ICA. Although most ICAs never rupture, the majority of AVMs will bleed if untreated. Brown showed that the risk of hemorrhage from a previously unruptured AVM is approximately equal to 105 minus the patient’s age. For example, the lifetime risk of rupture in a 20 year-old with a newly diagnosed AVM is 85%. Furthermore, although rupture of ICA is more common with advancing age, most patients experience the first hemorrhage from AVM between the ages of 20 and 40. Accordingly, older patients with SAH are more like to have an aneurysmal etiology for the event.

The etiology of AVM, like that of ICA, is a matter of some controversy, although it is generally agreed that these lesions are congenital. The exact embryological and genetic origins have not been identified, and AVM is rarely familial. It should be noted that Ehlers-Danlos IV syndrome is associated with direct AV fistula, and Rendu-Osler Webers disease is associated with cavernoma, but these AV shunts are more uncommon than “true” AVM.

Emergency Department Evaluation

Unfortunately, primary presentation of unruptured AVM is uncommon. The most common revealing event for these lesions is SAH, which is discussed in the following section. AVM should be part of the differential diagnosis for any patient presenting with headache, unexplained seizure, or neurologic deficit. Many if not most unruptured AVMs diagnosed in the emergency department will be incidental findings on MRI or CT brain scanning performed for unrelated indications.

As always, the emergency physician will first direct attention to stabilization of the ABCs and the administration of supportive care. History and physical examination priorities are much as for ICA, although AVM is less likely than ICA to present with bulbar palsy. A thorough neurological examination is of course essential.

The emergency department evaluation of a suspected intracranial lesion hinges on diagnostic imaging. CT angiography and MRI angiography are both capable of identifying intracranial AVM, although MRI gives better vascular and parenchymal definition and may be more sensitive.

Prompt neurosurgical consultation should be obtained for all patients diagnosed with unruptured intracranial AVM. Most patients will require additional imaging with MRI and/or digital subtraction angiography. Treatment options include embolisation, surgery, and radiosurgical ablation.

Stroke Syndromes

Hemorrhagic Stroke (11.1.2)

Hemorrhagic stroke accounts for about 20% of all acute cerebrovascular accidents, with SAH accounting for 3-5% and ICH for 15-18%. However, these catastrophes have a disproportionate impact on mortality, disability and potential life-years lost. The vast majority of hemorrhagic strokes can be classified as intraparenchymal hemorrhage (IPH) or SAH. While both IPH and SAH reflect bleeding from intracranial vessels, the two entities are otherwise quite distinct in their etiology and pathophysiology. Patients with hemorrhagic stroke require rapid stabilization, diagnosis and referral for definitive neurosurgical management.

SPONTANEOUS Subarachnoid Hemorrhage

Etiology/Anatomy/Pathophysiology. Spontaneous (as opposed to traumatic) intracranial SAH is a potentially devastating vascular catastrophe that almost always results from rupture of an ICA or AVM. The incidence rate of SAH has remained stable for decades, at about 5-10 cases per 100,000 patient years. Although SAH accounts for only 3% of all strokes, it accounts for 25% of potential life-years lost to stroke, and 5% of all stroke deaths. Most patients are under 60 years of age. Women are 1.6 times more likely to have an SAH than their male counterparts, and African-Americans are more than twice as likely to suffer an SAH as whites.

A family history of SAH places one at a greater risk of cerebral hemorrhage. First-degree relatives of SAH victims have up to a 7-fold increased risk of SAH. Inherited connective tissue diseases such as autosomal dominant polycystic kidney disease, neurofibromatosis, and Ehlers-Danlos IV are also associated, to varying degrees, with increased risk of SAH. Modifiable risk factors include alcohol abuse, smoking, and hypertension. Evidence for increased risk of SAH from heavy coffee intake and oral contraceptives is mixed at best.

The vast majority (85%) of primary spontaneous SAH results from the rupture of a saccular ICA. AVM accounts for another 5-10% of cases, and the remaining 5-10%, including those with nonaneurysmal permisencephalic hemorrhage, have no demonstrable vascular abnormality at arteriography. These values do not incorporate data for secondary SAH, which results from the extension of a parenchymal hemorrhage of neoplastic, traumatic, hypertensive or coagulopathic etiology into the subarachnoid space. Furthermore, the values for AVM are potentially misleading, as many if not most cases of SAH associated with an AVM result from rupture of a saccular aneurysm on a feeding artery, rather than the AVM itself.

Emergency Department Evaluation. The diagnosis of SAH in the emergency department is fraught with peril. Misdiagnoses—and missed opportunities for patient salvage—are common. Up to one-third of all patients presenting with SAH are incorrectly diagnosed upon their first visit to a physician. Much of the traditional medical “wisdom” surrounding the evaluation of this condition is clearly wrong, and the emergency physician must maintain a high index of suspicion and low threshold for the performance of diagnostic testing to avoid misdiagnosis and inappropriate disposition.

Classically, patients with SAH present with headache of explosive, catastrophic onset during strenuous activity, associated with syncope, nausea, vomiting, stiff neck, and obtundation. Patients with such a “textbook” presentation are easily recognized; however, not all patients with SAH will present with such clear-cut findings.

The emergency physician must recognize that SAH encompasses a broad spectrum of clinical manifestations. Half of all patients with SAH present with atypical features. Although headache from SAH tends to be sudden in onset, the headache may not be severe, and may not be associated with strenuous activity. Headache from SAH may be relieved by non-narcotic analgesics, or may resolve spontaneously. Nuchal rigidity, focal neurologic signs, and ophthalmologic manifestations may be absent, or so subtle as to escape cursory examination.

Furthermore, patients may present with the so-called “warning leak,” rather than a catastrophic hemorrhage. Such “leaks” are in fact relatively mild episodes of SAH, and are harbingers of more serious events to follow. These patients typically have abrupt onset of headache, which may be self-limited, in the absence of acute neurologic findings, meningismus, or ophthalmologic abnormalities. Failure to properly evaluate these patients is particularly tragic, as early diagnosis and management would result in complete salvage in the vast majority of cases.

SAH must therefore be part of the differential diagnosis for any patient presenting to the emergency department with a history of abrupt onset of headache, regardless of the severity, location, duration, response to non-narcotic analgesia, or presence or absence of associated findings such as retinal hemorrhage or meningismus. Patients with chronic headaches who present with “worst” headaches, or those presenting with atypical headaches, must also raise the suspicion of SAH. Finally, it must be emphasized that the emergency physician should never make a primary diagnosis of vascular or migraine headache. Not only does a first episode of severe headache fail to meet the criterion of the International Headache Society for diagnosis of vascular headache, but its consideration in the differential practically invites misdiagnosis of SAH.

The emergency physician should first address the ABCs, patient stabilization and supportive care. Obtunded or combative patients may require rapid sequence induction of anesthesia and tracheal intubation. IV access and cardiac monitoring are essential for all patients with a suspected stroke syndrome, including SAH. Seizure precautions and working suction at the bedside are highly recommended. The physician should obtain a careful medical history, including time of onset, duration, and location of the headache; associated activity; what analgesics may have been taken; family history of SAH or stroke; and patient use of tobacco, coffee, alcohol and illicit drugs. Physical examination should include a search for subhyaloid retinal hemorrhage, papilledema, cranial nerve palsies, meningismus, temporal artery tenderness, carotid bruits, and focal neurological deficits. An electrocardiogram may reveal an associated tachyarrhythmia or myocardial infarction.

Diagnosis of SAH is made at non-contrast CT or with lumbar puncture. CT scanning is up to 98% sensitive for SAH, although small amounts of subarachnoid blood can be subtle and difficult to detect. It is important to recognize that CT is to some degree an operator-dependent test, and subtle hemorrhage may escape detection by all but the most capable radiologists.

CT fails to identify 2-3% of patients with SAH, and therefore the traditional diagnostic algorithm for patients with suspected SAH has been to follow a negative CT study with lumbar puncture. Misinformation and controversy abound, however, and misinterpretation of lumbar puncture results presents another potential pitfall in the diagnosis of SAH. Visual inspection of spinal fluid for either blood or xanthochromia is completely unreliable. Operators have long ruled out “traumatic taps” by comparing the erythrocyte count in the first tube to that in the third or fourth tube; if the first tube shows a high erythrocyte count but the later tube has a lower count, the tap is said to be “traumatic” and SAH is excluded. This ostensibly reassuring and persistent article of medical folklore has no solid evidentiary basis. Diagnosis of SAH by lumbar puncture is made by the finding of erythrocytes or of xanthochromia, which is the staining of CSF by bilirubin and oxyhemoglobin due to erythrocytolysis. The relative accuracy of these two findings remains an issue of contention.

Some authors have suggested that, contrary to traditional practice, lumbar puncture should be performed prior to CT imaging in the patient with lone acute sudden headache (LASH). Patients meeting LASH criteria have normal neurological examinations, normal vital signs including temperature, and no meningismus. Several advantages are cited for the LP-first strategy. Proponents note that, although a negative CT should be followed by lumbar puncture, physicians (and patients) will frequently take inappropriate reassurance from a negative study and defer the LP. Performing the LP first would bypass this potentially catastrophic lapse in judgment. A negative LP is said to obviate the need for further evaluation for SAH in patients without focal neurological findings or altered level of consciousness, with consequent improvements in resource utilization. Finally, LP-first proponents correctly point out that lumbar puncture is safe in LASH patients.

Critics of this approach argue that the vast majority of CT scans will be positive in patients with SAH, obviating the need for a painful, invasive procedure that could theoretically precipitate rebleeding from a ruptured aneurysm. Critics also point out that CT is not merely a screen to determine whether the patient can safely undergo lumbar puncture. CT will delineate the location of bleeding and the presence or absence of intraventricular hemorrhage, and may also identify alternative sources for the headache, such as intracerebral bleeding and mass lesions.

At the time of this writing, emergency medicine practice appears to be shifting increasingly to the LP-first strategy, although strong evidentiary support is limited. Several caveats should be borne in mind. The LP-first strategy may not be appropriate for patients presenting less than twelve hours after the onset of headache. Patients must meet LASH criteria—patients with altered consciousness, fever and focal neurologic signs should undergo CT scanning as soon as possible. Finally, the physician should be aware that in a small number of LASH patients, a potentially serious intracranial lesion other than SAH may be present, and will be missed if CT scan is deferred.

Patients diagnosed with SAH should receive immediate neurosurgical consultation for additional evaluation, imaging and management. The patient’s vital signs and neurological status must be closely monitored, as deterioration can occur suddenly due to rebleeding, acute hydrocephalus, expanding subdural hematoma, or intraparenchymal hemorrhage. Hypoxia must be avoided at all costs. Management of hypertensive patients is difficult, and should be conducted in close consultation with a neurosurgeon. Changes in cerebral autoregulation after intracranial hemorrhage render the brain more dependent on arterial pressure to maintain perfusion. Lowering blood pressure in the hypertensive patient with SAH therefore entails a very real risk of inducing or exacerbating cerebral ischemia, and should generally be avoided in all but the most extreme scenarios.

Neuroprotective strategies in SAH include treatment with calcium channel blockers and antioxidant compounds. Calcium channel blockers were first employed on the rationale that they would prevent the cerebral vasospasm and resultant ischemia complicating SAH. However, nimodipine, the calcium channel blocker that seems to have the greatest salutary impact on outcome, also appears to be the one with the least effect on vasospasm. The mechanism for the limited neuroprotective effect (if any) of nimodipine therefore remains in doubt.

Tirilazad, which prevents iron-mediated lipid peroxidation, performed poorly in clinical trials. Similarly, N-propylenedinicotinamide, a free radical scavenger, failed to significantly improve outcome at 3 months after SAH.

Surgical ablation of the ruptured aneurysm has been the mainstay of treatment for decades, but in the last 10 years endovascular coiling has come into increasing use. The International Subarachnoid Aneurysm Trial (ISAT) enrolled 2143 patients with ruptured aneurysms and randomized them to coiling or clipping. The investigators found significantly better 1-year outcomes with endovascular coiling, but the long-term advantages of coiling, especially with respect to aneurysmal recanalization and bleeding, are still unclear.

Intracerebral Hemorrhage

IPH usually results from long-standing uncontrolled hypertension, and occurs most frequently in the basal ganglia. Presentation is highly variable, ranging from simple headache with or without neurological deficit to full-blown coma, quadriparesis and cardiorespiratory depression. Emergency care emphasizes rapid stabilization and diagnosis, prompt neurosurgical consultation, and optimization of intracranial pressure, hemodynamics and metabolic parameters.

Etiology/Anatomy/Pathophysiology.  The most common cause of nontraumatic intracerebral hemorrhage is long-standing uncontrolled hypertension. In this setting, small cerebral arteries and arterioles undergo degenerative changes that lead to the development of microscopic defects, called Charcot-Broussard aneurysms. These micro-aneurysms are most commonly found in the basal ganglia, thalamus, internal capsule, cerebral white matter, cerebellum and brainstem, and it is these locations that IPH occurs with the greatest frequency, with basal ganglia hemorrhage predominating. In elderly patients, IPH may be due to cerebral amyloid angiopathy, in which deposition of proteinaceous material in the walls of leptomeningeal blood vessels leads to decreased vascular integrity. In younger patients and in those without chronic hypertension, arteriovenous malformation or abuse of sympathomimetic drugs (cocaine, methamphetamine) are more likely culprits. Other causes include iatrogenic coagulopathy, neoplasm, iatrogenic hemorrhage from stroke thrombolysis, and vasculitis.

Patients with intracerebral hemorrhage classically present with sudden headache, neurological deficit, syncope, vomiting, alteration of mental status, seizures or coma. As with SAH, however, the physician must beware of relying on such a “textbook” presentation, have a low index of suspicion for the diagnosis, and rely heavily on diagnostic imaging.

Neurological presentation of intracerebral hemorrhage is highly variable, depending on the location of hemorrhage, the structures affected, and the presence of edema or mass effect. Putamenal hemorrhage, which accounts for roughly 50% of IPH, presents with contralateral motor and sensory deficits, homonymous hemianopsia, aphasia, and ipsilateral gaze deviation (toward the lesion). This pattern can be difficult to distinguish from middle cerebral artery ischemic stroke. Pontine hemorrhage is a devastating injury that rapidly progresses to coma, pinpoint pupils, decerebrate posturing, and loss of oculovestibular reflexes. Cerebellar hemorrhage manifests as vomiting, vertigo, dysequilibrium and ataxia, and ataxia. Signs of pontine compression may be present, such as facial weakness and CN VI palsy. The presentation of cerebellar hemorrhage may be subtle and difficult to differentiate from more benign causes of vertigo and ataxia. This is an important point, because patients with posterior fossa hemorrhage respond well to surgical evaluation, but will deteriorate rapidly due to brainstem compression and herniation if the condition is not recognized and treated early.

Emergency Department Evaluation. The emergency physician begins with a rapid primary survey of the ABCs in conjunction with patient stabilization. Many patients with IPH will be comatose and require immediate endotracheal intubation. IV access, supplemental oxygen, cardiac monitoring, seizure precautions, and working suction at the bedside are essential. Early evaluation should also include measurement of capillary blood glucose, pulse oximetry, assessment for trauma, evidence of aspiration, and core body temperature. Rapid neurologic screening will include level of consciousness, pupillary examination, speech, motor function, and, if possible, gait.

Patients who require endotracheal intubation should be managed with rapid sequence induction of anesthesia, utilizing a protocol that incorporates intravenous lidocaine (1 mg/kg over 1 minute) to blunt the increase in intracranial pressure that accompanies manipulation of the airway. The use of lidocaine in this setting has been challenged recently, but this evidence is still preliminary and lidocaine remains appropriate for airway management in these patients.

Many patients with ICH will have delayed presentations, and may have been comatose or immobile for hours or even days. Dehydration should be treated with volume expansion, and the physician should search for evidence of trauma, limb ischemia due to compression, sepsis and rhabdomyolysis. Derangements in serum glucose should be corrected with dextrose or insulin, and every effort should be made to keep serum glucose levels below 200 mg/dl.

When the patient has been stabilized, a complete history and physical examination should be performed. History should focus on determining when the event occurred and under what circumstances, whether associated trauma occurred, pre-existing medical problems, medications taken, and use of illicit drugs such as cocaine or methamphetamine. Neurologic examination should include complete assessment of cranial nerves, pupillary reactivity, visual fields by confrontation, deep tendon reflexes, muscle strength, and sensation, including proprioception, fine touch, steriognosis and pinprick. The physician should look for and document degenerate reflexes or posturing. A fundascopic examination is mandatory. Special attention should be paid to assessing the patient’s gait if possible. A complete assessment of cerebellar function, including heel-shin maneuvers, rapid alternating movements, Romberg, and gait testing are extremely important and often overlooked.

Laboratory evaluation includes electrolytes, renal function, complete blood count, PT/PTT/INR, and serum glucose.

Diagnosis of acute ICH is made by nonenhanced CT, which will identify a well-circumscribed radiodense lesion that may be associated with edema, mass effect, hydrocephalus, or extension of the hemorrhage to the cerebral ventricles or subarachnoid space. Patients with suspected intracranial hemorrhage are at risk for sudden deterioration and must be accompanied to the CT suite by appropriate personnel and equipment for monitoring and resuscitation. Identification of ICH at CT mandates immediate neurosurgical referral.

Moderate hypertension is well-tolerated in patients with acute ICH, and the emergency physician must be judicious in his use of antihypertensive agents. Lowering blood pressure in the setting of ICH may exacerbate abnormal autoregulatory parameters and result in brain ischemia. In patients with severe hypertension (MAP > 130 mmHg or SBP >220 mmHg), in whom control of hypertension is deemed essential to prevent additional injury, cautious use of antihypertensive agents such as labetalol or esmolol may be warranted. Use of long-acting or oral agents is contraindicated.

Intracranial hypertension from mass effect and brain edema is the predominant threat to the patient with IPH. Cerebral perfusion pressure (CPP) is defined as the difference between MAP (normally about 90 mmHg) and ICP (normally about 15 mmHg). If CPP drops below 50 mmHg, brain ischemia ensues, resulting in additional injury. Rising intracranial pressure may also result in transtentorial herniation. This catastrophe manifests as decreased level of consciousness, a dilated and unreactive pupil which is usually (but not always) ipsilateral to the lesion, and altered respiratory pattern or apnea. If not corrected immediately, transtentorial herniation will lead to irreversible brain damage and death. Patients with late or unrecognized cerebellar hemorrhage will also progress to brainstem compression, apnea and death.

Monitoring and treatment of elevated ICP should be aggressive and conducted in consultation with a neurosurgical specialist. Direct measurement via ICP monitor, which is usually placed ipsilateral to the lesion, is an indispensable guide to management. The goal of therapy is to maintain the patient’s ICP in the range of 15-20 mmHg. Options for the management of ICP include diuretic and osmotic therapy, fluid management, patient positioning, CSF drainage, and proper patient positioning. Of all these, elevating the patient’s head 30-40 degrees is the most benign, although the degree of benefit is questionable. The effectiveness and safety of osmotic therapy and hyperventilation have come under question in recent years, and should probably be used aggressively only in the most severe cases or where brain herniation is imminent, and in close consultation with a neurosurgeon. Hyperventilation to a PaCO₂ of 25-30 mmHg has a direct vasoconstrictor effect on cerebral arterioles and lowers ICP, but may also exacerbate concomitant cerebral ischemia. Mannitol is administered intravenously in a dose 1g/kg, and repeat doses of 0.25 mg/kg may be given every three to four hours for persistent elevations of ICP.

All patients with acute ICH require immediate neurosurgical consultation and intensive hemodynamic, neurological and ICP monitoring. Many patients, especially those with posterior fossa hemorrhages, will go to immediate craniotomy for evacuation. Indications for evacuation of lobar, putamenal, thalamic and brainstem hematomas are variable and controversial, and depend on the location, accessibility, and size of the lesion, as well as the presence of comorbidities, response of intracranial hypertension to medical management, and the threat of impending herniation. All patients with ICH require admission to an intensive care setting.

Ischemic Stroke (11.1.3)

The brain has no long-term stores of oxidizable substrates, and undergoes rapid loss of high-energy phosphate in the setting of ischemia. Furthermore, neurons are terminally differentiated, nonmitotic cells, and cannot be replaced once lost. Brain tissue and brain function are therefore exquisitely sensitive to circulatory impairment. Cerebral ischemia is a complication of subarachnoid hemorrhage (secondary to vasospasm), traumatic brain injury, or any condition that diminishes CPP, such as cerebral edema or rapidly expanding cerebral hematoma. Transient global brain ischemia is a devastating consequence of prolonged cardiac arrest. However, the following discussion focuses on focal ischemic stroke arising from either embolic or thrombotic occlusion of part of the brain’s blood supply. Focal ischemic stroke effects at least 600,000 patients per year in the US alone. The monetary cost of stroke exceeds $20 billion per year, and the human cost in disability and suffering is incalculable. Recent developments in the pharmacologic treatment of ischemic stroke have given the evaluation and management of this condition a new urgency, and emergency physicians have an increasingly central role in the acute care of patients with focal cerebral ischemia. For most patients, however, supportive care continues to be the cornerstone of management, with special attention directed to the maintenance of cerebral perfusion and oxygenation, control of serum glucose, and the prevention of complications and recurrent injury. The distinction between thrombotic and embolic stroke is somewhat hazy, in terms of both pathophysiology and clinical evaluation and management. Much of the material covered in the following discussion of embolic stroke (11.1.3.1) is therefore equally applicable to the setting of thrombotic stroke (11.1.3.2).

Embolic Stroke (11.1.3.1)

Etiology/Anatomy/Pathophysiology The brain’s blood supply enters the skull as the anterior and posterior circulations. The anterior circulation is supplied by the internal carotid arteries, which enter the skull through the petrous portion of the temporal bone. The internal carotids give rise to the ophthalmic arteries and then branch into the anterior cerebral, middle cerebral and posterior communicating arteries. The posterior circulation arises from the vertebral arteries, which enter the cranial vault through the foramen magnum. The two vertebral arteries unite to form the basilar artery, which lies on the ventral surface of the pons. At the level of the midbrain, the basilar artery bifurcates to form the two posterior cerebral arteries.

The anterior and posterior circulations are integrated into an anastomotic structure, the Circle of Willis (Figure), which lies on the ventral surface of the cerebrum, surrounding the pituitary gland and optic chiasm. The two circulations are joined by the posterior communicating arteries, which link the posterior cerebral arteries to the middle cerebral arteries. Moreover, the right and left anterior circulations are linked by the anterior communicating artery.

This anastomotic structure confers upon the brain a robust protection against injury arising from occlusion of the proximal arteries. In the presence of adequate collateral circulation, such an occlusion may result in minimal damage or even be clinically silent. On the other hand, anatomic variations of the Circle of Willis may leave the brain catastrophically vulnerable to a lesion that would be of little consequence in an individual with normal anatomy.

Embolization of thrombi to the cerebrovascular system accounts for roughly 15-25% of ischemic strokes. Cardiogenic emboli are the most frequent offenders, and may arise from diseased cardiac valves, prosthetic valves, atrial thrombi in the setting of atrial fibrillation, or ventricular thrombi in the setting of myocardial infarction. It is not unusual for acute ischemic stroke to be the presenting symptom of new-onset atrial fibrillation or “silent” myocardial infarction. Bacterial endocarditis, congenital heart disease, and cerebrovascular disease also place the patient at increased risk of embolic stroke.

Most focal ischemic strokes (as distinguished from the global cerebral ischemia that occurs during cardiac arrest) are characterized by a central “core” territory, in which ischemia is complete or near-complete, surrounded by a “penumbra” of tissue in which some collateral perfusion is preserved. Neurons in the ischemic core tend to die early and demonstrate frankly necrotic death phenotypes. In the penumbra, delayed neuronal death through apoptosis (programmed cell death or “cell suicide”) is more predominant. This is important because, unlike necrosis, apoptosis is a highly regulated phenomenon, subject to modulation by growth factors, inhibition of proteases, and changes in genetic expression. Accordingly, therapeutic interventions of value in stroke must focus on limiting the size of the ischemic core, optimizing perfusion of the penumbra, and promoting a biochemical milieu that favors survival signaling over programmed cell death processes.

The presentation of acute ischemic stroke depends on the vascular territory effected, and encompasses a broad spectrum of clinical syndromes. The reader is referred to Table 11-2 for a summary of the most commonly encountered stroke scenarios and their corresponding vascular lesion. In patients who have “showered” emboli to multiple cerebrovascular territories, the clinical picture will naturally be more complex.

ANTERIOR CIRCULATION
Vascular Lesion	Clinical Features
Complete MCA occlusion	Global aphasia (dominant hemisphere), neglect (nondominant hemisphere), homonymous hemianopia
Occlusion of Lentriculostriate branches of MCA	Lacunar Infarction. Internal Capsule: Hemiparesis or hemiplegia with sensory deficit, affecting contralateral face, arm and leg, without cortical deficit or visual loss.
Anterior MCA occlusion	Broca’s aphasia (dominant hemisphere), contralateral weakness of the face and arm.
Posterior MCA occlusion	Wernicke’s (fluent) aphasia (dominant hemisphere) and hemianopia without motor deficit.
ACA occlusion	Contralateral lower extremity weakness and sensory deficit; upper extremity “drift” on sustained extension, impairment of judgment and insight, confusion, gait apraxia, bowel and bladder incontinence.
POSTERIOR CIRCULATION
Vascular Lesion	Clinical Features
PICA or PICA branch occlusion	Cerebellar infarction: Ipsilateral ataxia, dysequilibrium, vertigo. Cerebellar edema at 2-4 days may be life-threatening. Lateral Medullary infarction: ipsilateral ataxia, decreased pain/temperature sensation on contralateral arm, trunk or leg; vertigo, nystagmus, Horner’s syndrome, hoarseness, hiccups, possible airway compromise.
Basilar artery occlusion	Massive brainstem infarction: Pontine and midbrain infarction results in coma, quadriparesis, meiosis, respiratory ataxia, apnea, death. Locked-in syndrome: Pontine infarction with midbrain sparing results in quadriparesis/quadriplegia, loss of horizontal gaze, facial paralysis and anarthria, with preservation of consciousness. Anton’s syndrome: Embolus to distal BA occludes both PCAs, resulting in cortical blindness.
Penetrating branches of basilar artery	Pontine lacunar infarction syndromes: pure motor hemiparesis, clumsy hand-dysarthria syndrome, ataxic hemiparesis.
Circumferential branches of basilar artery	Pontine-midbrain infarction. Crossed findings--contralateral extremity weakness, with ipsilateral facial weakness, bulbar palsy, ataxia, nystagmus, vertigo.
Posterior cerebral artery occlusion	Occipital lobe infarction: contralateral homonymous hemianopia. Involvement of thalamogeniculate arteries produces an accompanying contralateral hemisensory deficit. Alexia without agraphia: Left PCA occlusion produces inability to read with preservation of writing ability.

Table 2 . Stroke syndromes and corresponding vascular lesions.

Emergency Department Evaluation. As always, the emergency physician’s first priority is to perform an expeditious primary survey and stabilize the patient. Critically ill patients with airway compromise will require prompt intubation, but a rapid neurological examination should be performed prior to sedation and paralysis whenever possible. Hypoglycemia, narcotic overdose, Todd’s post-ictal paresis and other readily reversible causes of altered mental status and neurological deficit should also be considered prior to rapid sequence induction and intubation. All patients require cardiac monitoring, evaluation of pulse oximetry, vascular access, seizure precautions, and working suction at the bedside.

Administration of supplemental oxygen in patients without respiratory compromise or desaturation is an issue of some controversy. Some authorities worry that maintaining high cerebral oxygen concentrations will promote the elaboration of reactive oxygen species, which damage neuronal membranes, induce organelle stress responses, and promote the activation or maintenance of cell death processes. Ronning and Guldvog found that supplemental oxygen was of no benefit and might be slightly detrimental in acute ischemic stroke. However, patients in this study were given 100% oxygen. While cerebral hyperoxia presents certain theoretical concerns, hypoxia is indisputably an aggravating factor in the setting of stroke. At present, it seems reasonable to treat non-intubated stroke patients with low-flow supplemental oxygen, and to avoid excessively high oxygen fractions in intubated patients.

A thorough history and physical examination, with emphasis on cardiovascular and neurological findings, should be performed as soon as possible. It is important to determine, if possible, when symptom onset occurred, or when the patient was last observed to be functioning normally. The physician should search for meningismus, temporal artery tenderness, irregular pulses, carotid bruits, and cardiac murmurs, gallops or clicks that might suggest valvular disease or myocardial ischemia. Fundoscopic examination and a careful assessment of the cranial nerves is crucial. The neurological examination should also include an assessment of mental status, judgment, insight and language, visual fields by confrontation, cerebellar function, motor strength of all four extremities, deep tendon reflexes, sensation, and the presence of degenerate reflexes. Gait testing, if possible, can be extremely useful in localizing the putative ischemic lesion. A frequently neglected aspect of neurological assessment is repeated examination, which will often identify progression or resolution of deficits, clarify the presentation, and guide management and disposition.

Rapid imaging at CT is of paramount importance, especially in patients who present early, as the results of this study are crucial in identifying those patients who are suitable candidates for thrombolysis. CT will delineate subacute strokes and identify focal edema in many cases. CT may offer alternative diagnoses, such as intracranial mass or hemorrhage.

All patients should have blood drawn for basic laboratory studies, including complete blood count, electrolytes and creatinine, and coagulation profile. A 12-lead EKG is mandatory, to evaluate for arrhythmia or occult myocardial injury. Prompt neurological consultation is essential for all patients with suspected cerebral ischemia.

Unless a specific contraindication exists, aspirin should be administered early to all patients with suspected acute ischemic stroke, as this is the single most effective pharmacological intervention available for these patients. Administration of heparin and heparinoids to patients with acute ischemic stroke not known to be cardioembolic in origin does not improve outcome, presents considerable risk to the patient, and cannot be recommended. Patients with documented atrial fibrillation or other embolic source certainly may require anticoagulation to prevent additional injury; this intervention is ideally undertaken after echocardiography, and in consultation with a neurologist and cardiologist.

A growing body of data underscores the importance of controlling serum glucose levels in the setting of acute ischemic stroke and other forms of brain injury. Elevated glucose levels are strongly associated with increased mortality and neurological deficit. Severe hyperglycemia in the stroke patient should be treated aggressively, and every effort should be made to maintain serum glucose levels below 200 mg/dl. The neuroprotective effect of serum glucose control in acute ischemic stroke may have less to do with brain glucose levels than with the intrinsic growth factor properties of insulin, which is a potent inhibitor of programmed cell death. Post-stroke hyperglycemia may therefore be a biomarker for decreased “insulin tone” that places vulnerable neurons in the penumbra at greater risk.

The physician must avoid the temptation to aggressively treat hypertension in the patient with acute ischemic stroke. Cerebral autoregulation may be impaired in the stroke patient, and lowering the blood pressure may disproportionately compromise collateral flow to salvageable neurons in the penumbra, needlessly expanding stroke volume and worsening outcome. Control of profound hypertension (MAP > 130 mmHg or SBP >220 mmHg) in the patient with acute ischemic stroke should entail the use of agents that preserve cerebral autoregulation, such as labetalol, or agents that are subject to rapid titration, such as esmolol or nitroprusside. Oral antihypertensive agents are contraindicated.

Hypotension poses a dire threat to salvageable neurons in the penumbra, and should be corrected aggressively with volume expansion. Pressors should be used with great caution, and only when euvolemia has been established, because these agents can actually impair cerebral blood flow. Furthermore, vasopressors such as dopamine and norepenephrine probably “counter-signal” intrinsic growth-factor mediated cell survival processes, and may promote apoptosis in vulnerable neurons.

Until recently, emergency department treatment of ischemic stroke was nonspecific and largely limited to supportive care. The advent of thrombolytic therapy for acute ischemic stroke has had a tremendous impact on the emergency physician’s approach to such patients, not least by imparting a new sense of urgency to their evaluation and management. Most protocols exclude patients from thrombolytic intervention if stroke symptoms have been present for more than three hours. It is therefore essential that patient evaluation, neurological consultation and CT imaging be performed as expeditiously as possible in eligible patients if thrombolysis is contemplated.

It is important, however, to cast a judicious eye on the risks, benefits and effectiveness of stroke thrombolysis. The landmark study conducted by the National Institutes of Neurological Disease and Stroke, published in 1995, demonstrated a 12% increase in the number of patients with minimal or no disability compared to placebo, but this effect was accompanied by a ten-fold increase in the rate of symptomatic intracranial hemorrhage, from 0.6% to 6.4%. All subsequent studies with t-PA in the setting of acute ischemic stroke have demonstrated a similarly narrow therapeutic margin for this intervention. It is both instructive and sobering to compare the number-needed-to-treat and the number-needed-to-harm obtained from various clinical trials of stroke thrombolysis. In a meta-analysis of 6 studies of rt-PA for stroke, Lindbloom derived aggregate values for NNT and NNH at 17 and 19, repsectively; the number-needed-to-kill (from iatrogenic intracranial hemorrhage) was 40. Furthermore, an objective evaluation of the clinical effectiveness of this therapy when administered in the community setting reveals that thrombolysis is unlikely to have a major impact on outcomes in stroke populations, primarily because most patients who present with acute ischemic stroke will fail to meet rigid inclusion criteria. An additional problem is the inappropriate administration of thrombolytics to patients who do not actually have a stroke. In one study, 19% of patients evaluated by a stroke team were misdiagnosed as having acute ischemic stroke. Finally, it is worth noting that a study by Shriger et al suggested that many emergency physicians, neurologists and radiologists will fail to detect subtle radiographic signs of pre-existing cerebral hemorrhage—an absolute contraindication to thrombolytic therapy—on a screening CT examination, which casts additional doubt on the safety of this intervention.

Both physician and patient must be aware of the profound risk associated with thrombolytic therapy. Indeed, the issue of informed consent is an important and often confounding factor to consider when contemplating thrombolysis for stroke. A patient suffering from acute ischemic stroke may not be competent to render informed consent, raising difficult ethical and medicolegal issues. Up to 20% of patients presenting to an emergency department with stroke may be unable to understand the relevant information sufficiently to render informed consent. Institutions using thrombolysis for acute ischemic stroke would be well-advised to have in place explicit protocols for dealing with these eventualities.

Multiple studies have confirmed that the risk of symptomatic intracranial hemorrhage from stroke thrombolysis is increased when deviations from established criteria for t-PA administration occur. Table 3 outlines the generally accepted criteria for stroke thrombolysis, based on those utilized in the NINDS trial. If stroke thrombolysis is contemplated, it is absolutely imperative that the emergency physician insist on strict observation of these criteria, as any deviation from them is likely to place the patient in even greater jeopardy from an already risky therapy.

Inclusion Criteria

1. Patient or legal surrogate capable of giving informed consent.

2. Age > 18 years.

3. Onset of symptoms of hemispheric stroke within the last 3 hours.

4. Clinically apparent hemispheric ischemic stroke with neurological deficit.

5. NIHSS stroke severity score > 4.

6. CT scan complete prior to expiration of 3-hour limit.

Exclusion Criteria

1. Total anterior circulation syndrome: coma or severe obtundation with fixed gaze deviation and/or complete hemiplegia, or NIHSS score >22.

2. Evidence of rapid improvement or resolution of symptoms.

3. NIHSS score < 4.

4. Previous stroke within past 6 weeks.

5. Seizure with onset of stroke symptoms.

6. Any clinical evidence or suspicion of ICH, even in the face of a normal CT examination.

7. History of prior intracranial hemorrhage, AVM, aneurysm, or intracranial neoplasm.

8. SBP >185 mmHg or diastolic BP >110 mmHg despite acute antihypertensive therapy.

9. Known or suspected septic embolus.

10. Myocardial infarction within last 30 days.

11. History of moderate-severe trauma within last 30 days.

12. History of surgery within last 30 days.

13. Recent arterial puncture at a noncompressible site.

14. History of hemorrhagic diathesis or anticoagulant therapy with INR > 1.5.

15. Pregnancy, lactation or parturition with the last 30 days.

16. Thrombocytopenia (platelet count < 100 x 10⁹/L)

17. Evidence of intracranial hemorrhage on CT.

18. CT evidence of infarction encompassing more than 1/3 of MCA territory.

Table 3. Suggested inclusion and exclusion criteria for administration of rt-PA to patients with acute ischemic stroke.

Consideration of potential medicolegal consequences has no part in the decision to use rt-PA for stroke. “Thrombolitigation” may ensue whether thrombolysis is employed or not.

At least one study has suggested that emergency physicians are qualified to administer thrombolytics for acute ischemic stroke. In practice, however, the decision to use thrombolytics should almost always be made in consultation with a neurologist who can assist in evaluation, patient selection and education, optimization of metabolic and hemodynamic parameters, and disposition.

Clinical trials with putative neuroprotective agents have been disappointing, and at present no such agent can be recommended for routine clinical practice. For the time being, the best “neuroprotective” strategy is meticulous supportive care, with special attention directed to maintaining cerebral perfusion and oxygenation, controlling serum glucose levels, rigorous adherence to criteria for thrombolysis, prevention of recurrent stroke, and protecting the patient from comorbidities such as aspiration, infection, and thromboembolic events.

A large body of data confirms that patients with acute ischemic stroke are much more likely to enjoy a better outcome and early discharge when admitted to multidisciplinary “stroke units.” These units, which ideally occupy a fixed location within the institution and a committed multidisciplinary staff, focus on early diagnostics, appropriate control of blood pressure and serum glucose, early mobilization and rehabilitation, and close follow-up and social support.

Thrombotic Stroke (11.1.3.2)

Approximately half of all acute ischemic stroke is due to atherothrombotic disease of the cerebrovascular system, primarily in the extracranial and large intracranial vessels. Atherosclerotic disease can result in brain ischemia either by complete or near-complete thrombotic occlusion of the vessel itself, or by embolization of thrombus fragments from the diseased vessel to the distal cerebral circulation. Patient presentation, management and disposition are much the same as for embolic stroke, discussed in the preceding section.

Etiology/Anatomy/Pathophysiology. Most thrombotic strokes arise from atherosclerotic disease, which can progress to the formation of luminal thrombus in a manner not unlike that known to occur in acute coronary thrombosis. The most common sites for atherosclerotic plaque formation are at the bifurcation of the common carotid artery, the bifurcation of the middle and anterior cerebral arteries, and in the vertebrobasilar system. Ischemic stroke may then result either from occlusion by the thrombus, or by embolization of thrombus fragments to the cerebral circulation. Less common causes of thrombotic stroke include vasculitis, syphilis, systemic lupus erythematosis, carotid or vertebral artery dissection, and hypercoagulable states such as those seen in polycythemia, sickle cell disease, or deficiency of antithrombin III, proteins C and protein S.

Presentation of thrombotic stroke will vary depending on the site of occlusion and resultant brain injury. The reader is referred to the foregoing discussion of embolic stroke and Table for more information.

Emergency Department Evaluation.  Evaluation and management of acute thrombotic stroke is essentially identical to that discussed in the preceding section on acute embolic stroke. Rapid assessment of ABCs, patient stabilization and supportive care, cardiac monitoring, careful history and physical examination, diagnostic imaging, and early neurological consultation are the cornerstones of care. All patients should receive aspirin unless a specific contraindication exists. Anticoagulation is not indicated. The use of thrombolytics may considered in eligible patients, but the issue of informed consent must be addressed, and the emergency physician must insist on strict adherence to selection criteria and treatment protocols.

Transient Ischemic Attack (11.1.4)

The emergency physician may usefully regard transient ischemic attack (TIA) as the cerebrovascular analogue of unstable angina: an indicator of advanced cerebrovascular disease and harbinger of acute ischemic stroke. TIA represents an opportunity to initiate preventive therapy before permanent damage occurs, and as such its identification, management and appropriate disposition are exceedingly important.

Etiology/Anatomy/Pathophysiology A TIA is defined as a transient focal neurological deficit that resolves within 24 hours, although most resolve within 1-2 hours. Reversible deficits lasting longer than 24 hours are commonly referred to as RIND, or reversible ischemic neurologic deficit. These two conditions represent one end of a spectrum that progresses from reversible deficit to complete ischemic stroke, and are probably more reflective of clinical presentation than actual pathophysiology. Many if not most TIAs and RINDs will result in some small, albeit clinically silent, loss of brain tissue. Moreover, the definition of TIA seems to be undergoing a quiet evolution in the age of thrombolytic therapy--thrombolytics are not indicated for TIA, yet it is impossible to discriminate between TIA and stroke during the therapeutic window in which thrombolysis may be contemplated.

The overwhelming majority of TIAs arise from atherosclerotic lesions in the extracranial carotid vessels. TIA may occur either because of critical reversible thrombotic occlusion, or embolisation of thrombus fragment from the carotid lumen to a distal vessel, with subsequent resolution due to distal displacement of the fragment or disintegration of the fragment by hemodynamic forces or intrinsic fibrinolytic processes. Neurological deficits arising from TIA correspond to those arising from complete ischemic stroke in the same vascular distribution (Table 11-2 ).

Amaurosis fugax (AF) is a painless, monocular visual loss, which may be partial or complete, lasting seconds to hours, with complete resolution. AF is usually thromboembolic in etiology, and generally occurs without other neurological deficits. AF is clinically equivalent to other TIAs: a sign of atherosclerotic cerebrovascular disease and a harbinger of acute ischemic stroke.

Emergency Department Evaluation

There is no way to determine which patients with acute focal neurological deficits will resolve. Therefore, all such patients should be evaluated and managed as for acute ischemic stroke. Patients who have resolve their deficits prior to presentation, or who resolve during their clinical course, will also require similar evaluation. Patient stabilization is, as usual, the first priority. All patients without contraindications should receive aspirin early in their emergency department course. IV access, cardiac monitoring, measurement of capillary blood glucose and pulse oximetry and, if necessary, supplemental oxygen should be instituted, and an EKG should be obtained early. History should focus on time of onset and duration of ischemic symptoms, associated symptoms such as nausea or vertigo, previous episodes if any, presence of vascular disease or atherosclerotic risk factors, medications, and use of tobacco, alcohol or illicit drugs. A thorough physical examination should be performed, including a complete neurological assessment, documenting the status of cranial nerves, presence or absence of carotid bruits, motor and sensory function, deep tendon reflexes, gait, and cerebellar testing. Fundascopic exam should be performed in all patients, although visualization of retinal emboli in patients with AF is uncommon. Basic laboratory evaluation should be conducted as for acute ischemic stroke. All patients should undergo unenhanced CT imaging.

Patients with new TIA, those with crescendo TIA, and those with more than three episodes in 72 hours, require prompt neurological evaluation and hospitalization for observation and further evaluation. Although some authors have suggested that many patients with TIA can be managed on an outpatient basis, convincing data of the safety and cost-effectiveness of this approach is lacking. The risk of full-blown stroke after TIA is highest soon after the event, approaching 7-9% in the first week, and 10-12% in the first 90 days. Given these data, admitting virtually all patients with new or crescendo TIA seems not only justifiable, but medically prudent and ethically compulsory. Admission allows for rapid performance of ancillary studies such as carotid duplex and cardiac echocardiography, immediate institution of appropriate antiplatelet and cholesterol-lowering therapies, anticoagulant therapy when indicated, optimization of hemodynamics and metabolic parameters, careful monitoring during the early high-risk interval, and expeditious triage to definitive therapy such as carotid endarterectomy (CEA) for selected patients.

In patients with carotid stenosis of 70% or greater, CEA may be warranted, substantially lowering the risk of recurrent ipsilateral TIA or acute ischemic stroke. Patients with less severe stenosis are probably best managed medically, with blood pressure control, cholesterol-lowering drugs, and aspirin or other antiplatelet agents. Anticoagulant therapy may be of benefit in patients with TIA due to atrial fibrillation. The current evidence does not support anticoagulation of TIA patients without atrial fibrillation, however, and most patients should be treated with aspirin or another anti-platelet agent.

Selected Reading

International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003; 362:103-10.

Fleetwood IG, Steinberg GK. Arteriovenous malformations. Lancet 2002; 359:863-73.

Isaksen J, Egge A, Waterloo K, Romner B, Ingebrigsten T. Risk factors for aneurysmal subarachnoid hemorrhage: the Tromsǿ study. J Neruol Neurosurg Psychiatry 2002;73:185-7.

Edlow JA, Caplan LR. Avoiding pitfalls in the diagnosis of subarachnoid hemorrhage. NEJM 2000;342:29-36.

International Subarachnoid Aneurysm Trial (ISAT) Collaborative Group. International subarachnoid aneurysm trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002;360:1267-74.

Gopal AK, Whitehouse JD, Simel DL, Corey GR. Cranial computed tomography before lumbar puncture. Arch Intern Med 1999;159:2681-5.

White BC, Sullivan JM, DeGracia DJ, O'Neil BJ, Neumar RW, Grossman LI, Rafols JA, Krause GS. Brain ischemia and reperfusion: molecular mechanisms of neuronal injury. J Neurol Sci 2000 Oct 1;179(S 1-2):1-33.

Ronning OM, Guldvog B. Should stroke victims routinely receive supplemental oxygen? A quasi-randomized controlled trial. Stroke 1999;30:2033-7.

Hankey GJ. Heparin in acute ischaemic stroke: the t-wave is negative and it’s time to stop. Med J Australia 1998;169:534-7.

Adams JG, Chisholm CD; SAEM Board of Directors. The Society for Academic Emergency Medicine position on optimizing care of the stroke patient. Acad Emerg Med 2003;10:805.

National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. NEJM 1995;333:1581-7.

Ciccone A, Bonito V; Italian Neurological Society's Study Group for Bioethics and Palliative Care in Neurology. Thrombolysis for acute ischemic stroke: the problem of consent. Neurol Sci 2001;22:339-51.

Henneman PL, Lewis RJ. Is admission medically justified for all patients with acute stroke or transient ischemic attack? Ann Emerg Med 1995;25:458-63.

FIG. 11–1. A: Early subarachnoid hemorrhage resulting from leakage of a known basilar artery aneurysm. B: Later findings after delayed, more extensive rebleeding from the same aneurysm. (Courtesy of Mary Burry, MD, Portland VA Medical Center.)

FIG. 11–2. Computed tomography of the brain showing a large intracranial hemorrhage in the basal ganglia, with intraventricular extension, surrounding edema, mass effect, and effacement of the lateral ventricle. (Courtesy of Richard Harper, MD, Portland VA Medical Center.)

FIG. 11–3. A: Subtle findings of a left hemispheric stroke approximately 6 hours after onset of symptoms, showing effacement of the sulcal pattern and minimal edema. B: The same patient 3 days later, revealing massive loss of brain parenchyma. (Courtesy of Mary Burry, MD, Portland VA Medical Center.)