The trusted source for
healthcare information and
Part I: Risk Factors, Differential Diagnosis, and Prevention
Authors: Marcia A. Cort, MD, Assistant Professor, Division of Emergency Medicine, University of Maryland Medical Systems, Baltimore; Dick Kuo, MD, Assistant Professor, Division of Emergency Medicine, University of Maryland School of Medicine, Baltimore.
Peer Reviewers: Laurence Gavin, MD, Clinical Associate Professor of Emergency Medicine, University of Pennsylvania Health System-Presbyterian, Philadelphia; David Wright, MD, Assistant Director, Emergency Medicine Research Center, Emory University, Atlanta, GA.
Stroke is the third leading cause of death in the United States, surpassed only by heart diseases and malignant neoplasms.1 In 2000, stroke accounted for 167,661 deaths, yielding a population death rate of 60.9 per 100,000. Approximately 750,000 new strokes and more than 1 million hospitalizations for stroke occur annually in the United States. Eight percent of stroke victims die within 30 days of the event, and about 30% are dead within a year. Thirty-one percent of stroke survivors require assistance with activities of daily living and 16% require assisted-living institutions. The annual estimated economic impact of stroke on our society, i.e., the costs associated with health care and lost income, is $40.9 billion.2
Overall, the incidence of stroke is higher among males than among females; however, in younger age groups, the gender incidence is about the same.3,4 For people older than age 55, the incidence of stroke more than doubles in each successive decade. Nearly 30% of those who suffer a stroke are younger than age 65.5 Among the elderly, the number of female stroke victims is higher than the number of male stroke victims, in part because of the greater longevity of females. African-Americans and Hispanics have an increased annual age-adjusted relative risk of ischemic stroke compared with whites.1,6 Geographically, the southeastern United States has the highest incidence of stroke compared with other regions of the country.7
Part 1 of this series will cover the differential diagnosis of stroke, stroke mimics, and risk factors and prevention. Part II will cover the physical examination, laboratory investigations, imaging, and treatment of stroke.—The Editor
In 1980, the World Health Organization published what has become the generally accepted definition of stroke: "rapidly developing clinical signs of focal (at times global) disturbance of cerebral function, lasting more than 24 hours or leading to death with no apparent cause other than that of vascular origin."8 Transient ischemic attacks (TIAs) are neurologic deficits that occur as a result of loss of or decrease in perfusion of the brain, retina, or cochlea. These deficits resolve within 24 hours, most of them within one hour.9 Focal neurologic deficit lasting more than 24 hours but fewer than three weeks is classified as reversible ischemic neurologic deficit.10
Ischemic stroke accounts for 80-85% of all cerebrovascular accidents. It can result from thrombosis, embolism, or hypoperfusion (thrombosis is the most common etiology). It occurs as a result of neurons being deprived of oxygen and glucose. The brain does not store these substrates and, thus, requires a constant supply. When the blood supply to the cerebral microvasculature is disrupted, brain energy metabolism is disrupted, leading to loss of aerobic glycolysis, intracellular accumulation of sodium and calcium ions, release of excitotoxic neurotransmitters, elevation of lactate levels with local acidosis, free-radical production, cell swelling, overactivation of lipases and proteases, and cell death. Within 12-15 seconds of complete interruption of cerebral blood flow, electrical activity is suppressed, and within 2-4 minutes, synaptic excitability of cortical neurons is inhibited. After 4-6 minutes, electrical excitability is inhibited. When the cerebral blood flow decreases to about one-third of normal (18 mL/100 gm/min), the brain reaches a threshold for electrical failure but there still is a chance for recovery. This may occur in the area surrounding the core of the infarction, where tissue may be receiving flow from nearby collaterals. This area is known as the ischemic penumbra and is the target for stroke treatments. It is estimated that approximately six hours represents the window of opportunity available before the penumbra is lost and irreversible neurologic damage occurs. When cerebral perfusion further decreases to 8 mL/100 gm/min, as occurs in the core of the infarcted area, the result can be cell death. Between these two thresholds, there may be neurons that are functionally silent but structurally intact and potentially salvageable.2,11
Multiple clinical entities of various origins can resemble or be indistinguishable from an ischemic stroke or TIA. In the acute state, considerations of differential diagnosis apply equally to TIA and stroke. It is extremely important to make an accurate diagnosis rapidly, since the treatment of an ischemic stroke is most optimal when initiated within three hours of onset of symptoms. This treatment may be deleterious under other conditions. Several mimics of ischemic stroke and TIA are discussed below. (See Table 1.)
Hemorrhage. Intracranial hemorrhage, a mimic of cerebral ischemia, can occur by hypertensive and nonhypertensive mechanisms. Hypertension-related intracranial hemorrhage is caused by rupture of 50- to 300-mm diameter parenchymal perforating arteries damaged by local degenerative effects of hypertension. Focal deficits develop steadily over seconds to minutes. The possibility of intracranial hemorrhage is increased with the presence of coma on arrival, vomiting, severe headache, current warfarin therapy, systolic blood pressure greater than 200 mmHg, or glucose level greater than 170 mg/dL in a nondiabetic patient. The absence of these symptoms decreases the odds of hemorrhage by about one-third.12
The putamen is the area of the brain most commonly affected, followed by the cerebral lobes, thalamus, cerebellum, pons, and other parts of the brain. Putaminal hemorrhage is characterized by hemiparesis or hemiplegia (paralysis of half the body) involving the arm, face, or leg, accompanied by a hemisensory syndrome, hemianopsia (loss of vision on the same half of each visual field), and aphasia (loss or impairment of receptive or expressive language processing) if the dominant hemisphere is affected. Cerebral lobar hemorrhage occurs in the subcortical white matter of the cerebral hemispheres and frontal hemorrhage produces hemiparesis, particularly of the arm, along with behavioral changes and headache. Patients with thalamic hemorrhage present with complete contralateral hemisensory syndrome and capsular hemiparesis or hemiplegia. Cerebellar hemorrhages are sudden in onset, with nausea, vomiting, dizziness, ataxia, and facial and gaze palsy. Rapid worsening of the patient’s condition, progressing to coma, is common. Pontine hemorrhages are characterized by quadriplegia, decerebrate rigidity, ophthalmoplegia, and pinpoint pupils. Caudate hemorrhage is characterized by abrupt onset of nausea, vomiting, decreased level of consciousness, and transient hemiparesis or gaze palsy.13
Subarachnoid hemorrhage is caused by bleeding from arteries and veins close to the brain surface.13 Blood accumulates in the basal cisterns and subarachnoid space. The most common cause of subarachnoid hemorrhage is trauma; nontraumatic subarachnoid hemorrhages are caused primarily by rupture of congenital or berry lesions. Thirty percent of patients with subarachnoid hemorrhage present with loss of consciousness, which is associated with a higher mortality. Complaints also may include acute onset of severe ("thunderclap") headache with vision loss, diplopia, or facial pain as a result of cranial nerve involvement. Emotional upset and vigorous exercise have been noted to precipitate some cases of subarachnoid hemorrhage.
A subdural hematoma may form slowly and, thus, its symptoms may emerge gradually.13 Indeed, the precipitating trauma may have been forgotten since it may have occurred many days before. Patients may demonstrate fluctuating and false localizing signs. Xanthochromia may be noted in the cerebrospinal fluid, and computed tomography (CT) scan may show the hematoma. By the time of presentation, the hematoma may have become isodense with surrounding cerebral tissue. Chronic subdural hemorrhage has been recognized as a cause of stroke and TIA-like symptoms.14
Todd’s Paralysis. Partial motor or generalized seizures may precipitate postictal weakness, or Todd’s paralysis. Patients may present with a focal motor deficit, which may be weakness of an extremity or a complete hemiparesis. The preceding seizure activity may not have been witnessed or apparent, making this diagnosis difficult. The deficits may persist as long as 24 hours.
Hypoglycemia. The brain is particularly vulnerable to hypoglycemia since glucose is the primary energy source for its metabolism. Patients taking oral hypoglycemics or insulin with a resultant hypoglycemia (defined as a blood glucose concentration of < 45 mg/dL) may present with hemiplegia, hemiparesis, or aphasia with or without alteration of mental status. This syndrome also has been well described in alcoholics with hypoglycemia. Patients with a history of stroke may present with exacerbations of previous stroke symptoms upon becoming hypoglycemic or having other electrolyte abnormalities. The diagnosis can be made rapidly by bedside testing, and intravenous glucose can be administered to correct this disorder. The neurologic deficits may resolve immediately or over a number of hours.15-19
Migraine Headache. Migraine headaches are a common occurrence, with a higher incidence in females than in males. Acute hemiparesis may follow development of the headache ("complex migraine"). When this neurologic sign occurs in association with migraines, it usually does so in the initial stages of the headache. The diagnosis of migraine probably should not be considered seriously as an explanation for transient hemisphere attacks unless the patient is young, has repeat migraine headaches, experiences classic visual migraine auras at other times, and has a pounding headache contralateral to the sensory or motor symptoms in the hours after the attack.20 Familial syndromes of hemiplegic migraine are well recognized and related to a mutation of chromosome.19
Bell’s Palsy. This entity may present with an acute onset of isolated unilateral facial paralysis in the distribution of the seventh cranial nerve.21 Deficits usually are maximal within five days of onset. Symptoms include inability to fully close the affected eye, sagging of the lower eyelid, diminished lacrimation, or hypersensitivity to sound, gustatory dysfunction, drooling from the affected side of the mouth, and asymmetry of smile. This condition can be distinguished from cortical causes of facial weakness, as it causes ipsilateral paralysis of forehead muscles (cortical causes usually spare the forehead). CT scan of the head usually is normal. Testing for Lyme disease in endemic areas may be warranted. Treatment includes oral corticosteroids and protection of the affected eye.
Hyponatremia. A serum sodium concentration less than 120 mEq/L often is associated with central nervous system symptoms if it occurs acutely. Clinical findings may include mental status changes, focal neurologic deficits, ataxia, and seizures. In patients with hypovolemic hyponatremia, severe neurologic symptoms should be treated immediately with normal or hypertonic saline, depending on the degree of hyponatremia. The goal of treatment is to correct the sodium deficit by 0.5-1.0 mEq/L/hr. More rapid correction is associated with brain edema. Patients with euvolemic or hypervolemic hyponatremia should be treated with water restriction, diuretics, or demeclocycline.22
Hyperglycemia. Hyperglycemia with hyperosmolar state may be associated with focal neurologic deficits that mimic a stroke. Neurologic deficits include aphasia, homonymous hemianopsia, hemisensory deficits, hemiparesis, unilateral hyperreflexia, and the presence of the Babinski sign.19 The detrimental effect of hyperglycemia may be related to anaerobic glycolysis, leading to tissue acidosis and increased blood-brain barrier permeability.23
Meningitis/Encephalitis/Abscess. Patients with these conditions may present with change in mental status, headache, meningismus, and fever. However, encephalitis is more likely to present with focal neurologic deficits and seizures. Both may present with acute transient symptoms, which usually evolve over days or weeks. Seizures often occur before focal signs are evident, which distinguishes meningitis and encephalitis from stroke. CT scan in a patient with ischemic stroke usually is negative initially, but an enhancing mass may be seen on the CT scan of a patient with a cerebral abscess.20 Spinal fluid obtained after CT scan may be turbid and show a leukocytosis; organisms may be seen on Gram stain.
Hypertensive Encephalopathy. Hypertensive encephalopathy may occur as a result of an abrupt sustained increase in blood pressure that exceeds the limits of cerebral autoregulation. This occurs at a mean arterial pressure of 150-200 mmHg. At these pressures, cerebral autoregulation no longer can regulate blood flow, and marked vasospasm leads to ischemia, increased vascular permeability, punctate hemorrhages, and brain edema. The onset is acute and patients present with headache, vomiting, drowsiness, confusion, seizures, blindness, focal neurologic deficits, or coma. A careful neurologic examination will help distinguish hypertensive encephalopathy from acute ischemic stroke. Focal deficits in this disorder usually do not follow anatomic patterns. They usually are patchy in distribution or occur on the opposite side of the body. CT scan usually is normal. Hypertensive encephalopathy is a true emergency and is treated with rapid reduction of blood pressure (an action that may be deleterious in acute stroke). Intravenous nitroprusside or labetalol may be used. Symptoms usually resolve rapidly with treatment of blood pressure.24
Carotid Dissection. This entity may occur after forced hyperextension or neck trauma. Affected patients may present with focal neurologic deficits. The condition can be diagnosed by magnetic resonance angiography.
Functional Hemiplegia. Functional hemiplegia may be the result of either a conversion disorder or malingering.19 In a conversion disorder, the disability may be a "cry for help" or the patient may be trying to avoid a certain situation. These patients often appear strangely calm and unconcerned despite their disability. Malingerers usually have secondary gain to their disability, e.g., litigation or avoiding work. Functional hemiplegia can be diagnosed by careful physical examination. The functional deficits rarely occur in an anatomic distribution. The patient’s weakness may diminish with coaching, and paralysis will be present in absence of tone or reflex changes. The patient also may demonstrate weakness on foot extension but be able to walk on his/her toes. Functional hemiplegia is a diagnosis of exclusion and must be made only after detailed investigation.
Temporal Arteritis. Temporal arteritis, an uncommon disease entity with an elderly predominance, may be mistaken for a stroke. Patients may present with headache, blurred or loss of vision, and, occasionally, aphasia and hemiparesis. The acute vision loss can be differentiated from visual loss from a posterior cerebral artery occlusion, since the latter gives a homonymous visual field defect that does not violate the vertical midline. Temporal arteritis also presents with scalp tenderness and other systemic symptoms, including malaise, fatigue, arthralgias, and weight loss. The most important sequela of temporal arteritis is permanent blindness; thus, early diagnosis and treatment are critical. A head CT scan will be normal. Definitive diagnosis is by temporal artery biopsy. Laboratory studies may show a sedimentation rate above 100 mm/hr. Treatment is with oral or intravenous steroids, depending on the severity of the symptoms.
Air Embolism. Air embolism should be suspected in patients who have been exposed to acute changes in barometric pressure. The condition also can be caused iatrogenically by inappropriate technique for obtaining venous or arterial access. Apart from the abrupt temporally related onset of focal neurologic symptoms, cardiac auscultation would reveal a harsh new murmur throughout the precordium. These patients should be placed in the left lateral decubitus position to trap the air within the right atrium, where it can be aspirated by Swan-Ganz catheterization. Alternatively, hyperbaric treatment can be used to decrease the size of the bubbles in the cerebral circulation, thus decreasing the area of affected cerebral circulation.25
Multiple Sclerosis. The most common presenting symptom of patients subsequently diagnosed with multiple sclerosis is visual or oculomotor disturbances, followed by weakness or a sensory disturbance in one or more limbs.26 Symptoms generally develop over hours to days, plateau, then decline, but occasionally they may be maximal within seconds to minutes. In this case, vascular causes must be ruled out. Overall, magnetic resonance imaging (MRI) is the most sensitive paraclinical study in the diagnosis of multiple sclerosis. Lesions most frequently are detected with proton density, weighted images, and the fluid attenuated inversion recovery sequence.
Dementia. Posterior cerebral strokes initially present with visual field cuts, an inability to read, and poor ability to manipulate objects and easily could be mistaken for senility in an elderly person. Patients with non-dominant hemispheric lesions initially may experience right-left confusion, leading to disorientation in the home environment. Careful history taking will reveal that these symptoms were of a progressive or insidious onset rather than an abrupt onset.27
Diurnal and Seasonal Variations
Cerebral infarction has been positively associated with physical activity, catecholamine levels, blood viscosity, platelet function, coagulability, and fibrinolytic activity. These factors are present with circadian variation. Chronologic patterns of ischemic stroke in the early morning appear to be independent of risk factors and clinical subtypes of stroke. Studies have noted the morning predominance of ischemic stroke between 6 and 10 a.m. and speculate it is correlated with rising changes in heart rate and blood pressure, decreased vagal tone, increase in catecholamines, and activation of the renin-angiotensin system, which may make atheroscleotic plaques more likely to rupture and embolize. Another peak in stroke occurrence has been noted between noon and 4 p.m.—the etiology for this is unknown. The fewest stroke events occur between midnight and 4 a.m.28-31 More strokes are noted during the colder months.
Risk Factors and Prevention
A number of risk factors reliably have been identified as contributing to the development of stroke.32 Fortunately, some of the factors can be modified by pharmacologic therapy or surgical intervention (see Table 2); others cannot be modified (see Table 3). Secondary prevention of stroke is aimed at decreasing the proliferation of atherosclerotic disease since ischemic stroke is a cerebrovascular disease.
Modifiable Risk Factors. Hypertension. At least 25% of the adult population in the United States has hypertension, making it the single most modifiable risk factor for stroke. Hypertension accelerates atherosclerosis and, thus, increases the risk for stroke four-fold. Blood pressure treatment resulting in a reduction in systolic blood pressure of 10-12 mmHg and diastolic blood pressure of 5-6 mmHg is associated with a 38% reduction of stroke incidence.33 The Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure (JNC-7) recommends lower blood pressure limits than previously recommended. Goals for hypertension treatment now are a blood pressure of less than 140/90 mmHg, or less than 130/80 mmHg in patients with diabetes or chronic kidney disease.34 The benefit of lower blood pressure is helpful even in hypertensive patients with previous stroke or TIA: A reduction of blood pressure by 9/4 mmHg reduces the risk by one-quarter.2,35
Atrial Fibrillation. This dysrhythmia is the most common rhythm disturbance in adults and is associated with more than 120,000 strokes per year in the United States. Patients with atrial fibrillation have a five-fold increase in the risk of stroke and increased mortality compared with patients without atrial fibrillation. The risk of stroke further is increased in patients with recent congestive heart failure, hypertension, high atrial rate, and prior thromboembolism. Atrial high-rate events, defined as an atrial rate more than 220 beats per minute for more than 10 consecutive beats in patients with sinus node dysfunction, is associated with doubling the risk of dying or having a stroke. Anticoagulation in patients with atrial fibrillation decreases the relative risk of suffering a stroke by two-thirds.2,36
Asymptomatic Carotid Artery Disease. This term encompasses non-stenosing atherosclerotic plaque and carotid stenosis. About 15% of ischemic strokes are caused by internal carotid artery stenosis. Patients with an asymptomatic carotid bruit have an estimated risk of stroke of 1.5% at the first year and 7.5% at five years. Asymptomatic carotid artery stenosis less than 75% carries a stroke risk of 1.3% annually; with stenosis greater than 75%, the combined TIA and stroke risk is 10% in one year. Carotid endarterectomy of high-grade lesions (70-99%) has been proven to prevent stroke. Absolute risk reduction from carotid endarterectomy is 15% for all strokes.37,38 The composition of the plaque also is a factor in determining stroke risk. Ulcerated, echolucent, and heterogenous plaques with a soft core, as seen on ultrasound, are at increased risk of arterioarterial embolism.39
Previous TIA or Cerebrovascular Accident. Ten to fifteen percent of patients suffer TIAs prior to their cerebrovascular accident. Patients with one or more TIAs have a 10-fold increase in the risk of subsequent CVA. Location of TIAs also is predictive of stroke risk: Hemispheric TIAs carry a greater risk of stroke than retinal TIAs. Previous stroke also is a significant risk for recurrence and for increased morbidity and mortality. This is especially true in the period immediately following the stroke. Recurrent risk varies by stroke subtype, with the greatest risk occurring in patients with atherosclerotic infarction.32
Diabetes Mellitus. Diabetic patients have a two- to four-fold increased risk of developing ischemic stroke compared with non-diabetics. Diabetes is associated with development of atherosclerotic disease. In combination with hypertension or hyperlipidemia, the risk of stroke markedly is increased. High insulin levels associated with Type II diabetes increase the risk of atherosclerosis and thus stroke. Diabetics with retinopathy and autonomic neuropathy also appear to be at further increased risk of developing stroke. At this time evidence does not exist that tighter diabetic control decreases the risk of stroke or its recurrence.2,40
Hypercholesterolemia. High total cholesterol and low-density lipoprotein levels are associated with atherosclerosis. There is, however, an inconsistent relationship between serum cholesterol levels and death from ischemic stroke; thus, different subtypes of cholesterol confer different levels of risk. Lipid-lowering agents may decrease the development of atherosclerotic plaques and, therefore, possibly cause a regression of plaque formation, leading to a decrease in risk for stroke.2,41
Cigarette Smoking. This is associated with a relative risk of brain infarction of 1.7 and also is related to development of subarachnoid hemorrhage in a dose-response fashion. It is a direct risk factor for development of carotid plaque. Smoking cessation always should be encouraged, although more than five years of cessation may be required before risk is decreased appreciably.2
Alcohol Consumption. Heavy alcohol consumption, defined as five or more drinks per day, is associated with increased risk for ischemic stroke. Consumption of up to two drinks per day actually was found to decrease the risk of ischemic stroke (comparing drinkers with non-drinkers). Moderate alcohol intake is associated with an increase in serum high-density lipoproteins. There is, therefore, a J-shaped relationship between alcohol consumption and ischemic stroke.2,42
Cocaine and Amphetamine Use. Illicit drug use must be investigated, especially in a young individual presenting with symptoms of stroke. Cocaine is the leading cause of drug-related stroke in the United States. Ischemic stroke can be seen in crack cocaine and amphetamine users, secondary to the drugs’ powerful vasoconstrictive effects. In intravenous cocaine users, embolization from infected cardiac valves or injected foreign material can cause cerebral ischemia.
Physical Inactivity. There is a direct relationship between leisure time physical activity and decreased risk of stroke in both sexes and in the Hispanic, Caucasian, and African-American populations. Benefit can be derived from even light physical activity such as walking.43
Females and Estrogen. Women who smoke and take high-dose oral contraceptive pills are at increased risk for stroke. The risk of thrombosis increases in the postpartum female, with the highest risk within the first six weeks after delivery.
The Women’s Health Initiative44 was designed to examine a number of factors affecting the health of postmenopausal women. One arm of this study was a randomized trial that assessed the effect of estrogen plus progestin on ischemic and hemorrhagic stroke. In all examined age groups and categories of baseline stroke risk, the risk of ischemic stroke was increased among women taking estrogen and progestin supplementation.
Hemostatic Factors. Risk for development of ischemic stroke is increased in patients with elevated hemoglobin, hematocrit, and blood viscosity. Patients with increased levels of fibrinogen, factor VIII, Von Willebrand’s factor, antithrombin III, and lower mean levels of protein C have an increased risk of stroke. Patients younger than age 50 with increased antiphospholipid antibodies are at elevated risk for stroke, as are patients with anticardiolipin antibodies. The reason is unknown.11
Pharmacologic Agents for Prevention of Stroke. Antihypertensives. Blood pressure reduction is effective for primary and secondary prevention of stroke. Initially, angiotensin-converting enzyme (ACE) inhibitors and ACE receptor blockers (ARB) were thought to have beneficial effects independent of their direct effects on blood pressure. This effect (possibly related to neuroprotective mechanisms) has not yet been proven definitively. Multiple trials have compared the relative efficacies of different classes and combinations of antihypertensives.45 Thiazide diuretics, beta-adrenergic antagonists, ACE inhibitors, ARBs, and long-acting calcium channel blockers have been found to reduce the incidence of stroke. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack trial compared high-risk hypertensive patients randomized to receive ACE inhibitors, calcium channel blockers, or thiazide diuretics. In that trial, thiazide diuretics were found to be the most effective antihypertensive in stroke prevention. The relative inexpensive cost of thiazide diuretics also makes them an attractive option for stroke prevention.46,47 In the Losartan Intervention for Endpoint Reduction in Hypertension trial, losartan was found to be superior to the beta-adrenergic antagonist atenolol for stroke prevention.48
Antilipidemic Agents. Treatment with 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (HMG-CoA reductase inhibitors ["statins"]) has been associated with a 25% reduction in the risk of fatal and nonfatal stroke.
A recent trial evaluated the additive benefits of pravastatin and aspirin to decrease risks of cardiovascular disease. In randomized trials of secondary prevention, pravastatin sodium and aspirin reduced risks of cardiovascular disease. Pravastatin has a predominantly delayed antiatherogenic effect, and aspirin has an immediate antiplatelet effect, raising the possibility of additive clinical benefits. In five randomized trials of secondary prevention with pravastatin (40 mg/d), comprising 73,900 patient-years of observation, aspirin use also was prescribed in varying frequencies, and data were available on a large number of confounding variables. Two large clinical trials (Long-term Intervention With Pravastatin in Ischaemic Disease trial and the Cholesterol and Recurrent Events trial) compared the clinical benefits of combined pravastatin plus aspirin therapy with pravastatin alone. A meta-analysis of these two trials and three smaller angiographic trials was performed. In all analyses, multivariate models were used to adjust for a large number of cardiovascular disease risk factors. Both the individual trials and the meta-analyses demonstrated similar additive benefits of pravastatin and aspirin on cardiovascular disease. In meta-analysis, the relative risk reductions for fatal or nonfatal myocardial infarction were 31% for pravastatin plus aspirin vs. aspirin alone and 26% for pravastatin plus aspirin vs pravastatin alone. For ischemic stroke, the corresponding relative risk reductions were 29% and 31%, respectively. For the composite end point of coronary heart disease death, nonfatal myocardial infarction, coronary artery bypass graft, percutaneous transluminal coronary angioplasty, or ischemic stroke, the relative risk reductions were 24% and 13%. All relative risk reductions were statistically significant. The investigators concluded that more widespread and appropriate combined use of statins and aspirin in secondary prevention of cardiovascular disease will avoid large numbers of premature deaths.49
Other lipid-lowering therapies (e.g., dietary measures, resins, fibrates) have not been shown to decrease the risk of stroke.47 The National Cholesterol Education Program has developed guidelines for the treatment of hyperlipidemia. Patients with diabetes mellitus or atherosclerotic disease require more aggressive lowering of lower-density-lipoprotein (LDL) cholesterol. The goal for therapy in these patients is an LDL cholesterol level of less than 100 mg/dL. In patients with hyperlipidemia, one or no cardiovascular risk factors, and no evidence for atherosclerotic disease, the goal of treatment is 160 mg/dL.47
Antiplatelet Agents. Aspirin (acetylsalicylic acid) is the most widely studied and prescribed antiplatelet drug for patients at high risk of vascular disease. It affects a single pathway in the platelet activation process and provides incomplete protection against cardiovascular events. Adenosine diphosphate receptor antagonists, by blocking an alternate pathway of platelet activation, are slightly more effective than aspirin in reducing serious vascular events in patients at high risk, with similar results for the subset of TIA/ischemic stroke patients. Clopidogrel is an effective and safe alternative in patients who do not tolerate aspirin, in diabetics, in hypercholesterolaemic patients, or in those with a previous history of cardiac surgery. Moreover, antiplatelet combination therapy using agents with different mechanisms of action is an attractive preventive approach.
Aspirin, though beneficial for the primary prevention of myocardial infarction (MI), has not been found to be effective for primary prevention of stroke. In fact, the risk of stroke marginally is increased in patients taking aspirin to prevent first events of stroke. Many studies have demonstrated, however, that aspirin is effective in secondary prevention of stroke.50 Various doses have been studied. Lower dosages have been found to be better tolerated by the gastrointestinal tract and have fewer bleeding complications while providing a 20-25% relative risk reduction for recurrent stroke. Current Food and Drug Administration guidelines recommend 50-325 mg per day of aspirin for secondary stroke prevention.48 In the Ticlopidine Aspirin Stroke Study, ticlopidine reduced the rate of occurrence of stroke by 21% at three years.51 Ticlopidine has the significant side effects of diarrhea, rash, and, rarely, neutropenia, which has limited its use in stroke prevention.
Clopidogrel, which is better tolerated by patients and has a more favorable side effect profile, has emerged as the antiplatelet agent of choice for stroke prevention. When clopidogrel was compared with aspirin in patients with stroke, MI, or peripheral vascular disease (the Clopidogrel vs. Aspirin in Patients at Risk of Ischemic Events [CAPRIE] trial),52 an 8.7% relative risk reduction was demonstrated. At two years, the absolute risk reduction favoring clopidogrel was only 0.5%.53
However, more recent studies have investigated the amplified benefit of clopidogrel over aspirin in patients with a history of ischemic events.54 The goal of this study was to examine the influence of preexisting symptomatic atherosclerotic disease on subsequent ischemic event rates and compare the efficacy of clopidogrel vs aspirin (acetylsalicylic acid) in patients with such disease. Using the CAPRIE database, a group from the Cleveland Clinic performed multivariate analyses for patients who had symptomatic atherosclerotic disease (ischemic stroke or MI) in their medical history before enrollment in the CAPRIE trial. Two composite end points were used: 1) Ischemic stroke, MI, or vascular death; and 2) Ischemic stroke, MI, or rehospitalization for ischemia. In the CAPRIE population, as would be expected, a previous history of ischemic stroke and MI were historical risk factors that predicted subsequent ischemic events. Compared with the overall population, patients with documented symptomatic atherosclerotic disease had elevated event rates for the end point of ischemic stroke, MI, or vascular death; three-year rates were 20.4% with clopidogrel and 23.8% with aspirin (absolute risk reduction, 3.4%; 95% CI, -0.2 to 7.0; number needed to treat, 29; relative risk reduction, 14.9%; P=0.045). Similar results were obtained for the end point of ischemic stroke, MI, or rehospitalization for ischemia; three-year event rates were 32.7% with clopidogrel and 36.6% with aspirin (absolute risk reduction, 3.9%; 95% CI, -0.4 to 8.1; number needed to treat, 26; relative risk reduction, 12.0%; P = 0.039).
The investigators concluded that patients in the CAPRIE study who had known symptomatic atherosclerotic disease were at higher risk for developing neurovascular ischemic events, and that the absolute benefit of clopidogrel over aspirin seemed to be amplified this high risk subgroup. Other recent clinical trial results with antiplatelet therapy also have yielded important implications in stroke prevention, especially as this relates to combination antiplatelet therapy with aspirin and clopidogrel.55 This approach is supported scientifically by the observation that ASA and clopidogrel inhibit different pathways, and this dual inhibition may confer additional prophylactic benefits in stroke prevention. The CURE (Clopidogrel in Unstable Angina to Prevent Recurrent Events) study evaluated the efficacy and safety of clopidogrel in addition to acetylsalicylic acid vs standard therapy (including aspirin) in more than 12,000 patients with unstable angina or non-ST-segment elevation MI. In combination with aspirin, clopidogrel reduced the relative risk of the combined atherothrombotic endpoint of cardiovascular death, MI, or stroke by 20% (95% CI 0.72-0.90; p < 0.001) and the absolute risk of this composite endpoint by 2.1%. While the study was not powered or designed to demonstrate a reduction in stroke, there was a 14% reduction in stroke risk (p > 0.05). This approach was associated with an acceptable 1% increase in the incidence of major bleeding events (p = 0.001). In PCI-CURE, a prespecified substudy of patients who underwent percutaneous coronary intervention (PCI) during CURE, the benefits of clopidogrel therapy also was demonstrated.55
CREDO (Clopidogrel for Reduction of Events During Observation) was a randomized, double-blind, placebo-controlled trial. In this study of about 2100 patients, continuation of clopidogrel in addition to standard therapy including aspirin for 12 months after PCI was assessed, as was the benefit of a preprocedural clopidogrel loading dose. At one year, there was a 27% reduction in the risk of stroke, MI, or death with long-term clopidogrel therapy (p = 0.02). There was a consistent benefit of long duration clopidogrel therapy for each component of the composite endpoint, with a 25.1% relative risk reduction for all-cause stroke. In patients who received clopidogrel six or more hours prior to PCI, there was a 39% reduction in the risk of death, MI, or urgent target-vessel revascularization at 28 days (p = 0.051). CREDO data suggests an early loading dose of clopidogrel in patients undergoing stenting may be beneficial, and use of a loading dose followed by long-term continuation of clopidogrel in other high-risk atherothrombotic patients such as those with TIA or ischemic stroke may also be effective in prevention of thrombotic events.55
Antithrombotic Agents. In patients with atrial fibrillation, warfarin therapy has been found to be efficacious in preventing strokes. In patients with nonvalvular atrial fibrillation, warfarin therapy is recommended for patients older than 65 years with or without major risk factors (defined as previous stroke, systemic embolism, TIA, hypertension, poor left ventricular function).3 Close monitoring is required to keep the patient’s international normalized ratio (INR) between 2 and 3. The benefit of treating all patients with valvular atrial fibrillation has been well established. Patients should be screened for risk of bleeding before warfarin therapy is initiated.50
Nonmodifiable Risk Factors. Age. Age is the most important determinant of the likelihood of stroke. After the age of 55, increments of 10 years are correlated with a doubling of stroke rate in both sexes.4 As the population ages, the prevalence and economic impact of stroke will increase.
Race. African-Americans and Hispanics have a greater stroke incidence than whites across all age groups. There also is an increased mortality from stroke in these minority groups. The increased mortality can be explained only partially by socioeconomic and environmental factors.4,56,57
Gender. Overall, the incidence of stroke among males is 30% higher than among females. However, in younger individuals, males and females have an equal incidence of stroke. The increased incidence in young females probably is due to pregnancy and hormonal factors. Males develop ischemic strokes at higher rates than women up to the age of 75 years.
Heredity. A family history of stroke among first-degree relatives is associated with increased risk of cerebral infarction. Parental history is an independent risk factor for stroke. Some inherited diseases can predispose individuals to atherosclerotic and non-atherosclerotic vasculopathies, coagulopathies, and embolisms; examples include Marfan’s syndrome, Osler-Weber-Rendu disease, Sturge-Weber syndrome, and Ehlers-Danlos syndrome. Stroke has been reported in patients of all ages with sickle cell disease. Ischemic stroke is more likely in children, and hemorrhagic stroke is more likely in adults, although either type can occur in any age group. Stroke in this population is caused by small or large vessel occlusion, not atherosclerosis. Risk factors for stroke in sickle cell disease are as follows: 1) previous TIA; 2) low steady-state hemoglobin; 3) more frequent or recent acute chest syndrome; and 4) elevated systolic blood pressure. Sixty-seven percent of sickle cell patients with stroke had a recurrent stroke within 12-24 months.58,59
TIAs are temporary, focal neurologic deficits of acute onset lasting fewer than 24 hours. Ninety percent of TIAs actually last fewer than 10 minutes. They are a significant warning of ischemic stroke. They can occur from the carotid distribution (90%), the vertebrobasilar distributions (7%), or both (3%).60
Carotid TIAs are associated with ischemia to the ipsilateral eye or brain. Symptoms may include any or any combination of the following symptoms:
Vertebrobasilar TIAs occur as a result of disrupted blood supply to the brainstem, cerebellum, and visual cortex. Patients will present with the following symptoms:
Anterior Cerebral Artery Syndrome. (See Table 4.)2,61,62 Clinically recognizable symptoms from anterior cerebral artery occlusion depend on the location of the occlusion and the patency of anastomotic channels. Occlusion proximal to the medial striate artery may produce no discernable symptoms if there is good collateral supply from the opposite anterior cerebral artery. Occlusion proximal to the callosomarginal branch can lead to infarction of a large segment of the frontal lobe, leading to altered mentation, confusion, impaired insight, sucking reflex, and apraxia (inability to perform simple tasks despite full strength in the requisite muscle groups). Paralysis mainly of the opposite leg and mild arm involvement may occur with paralleling sensory deficits. Bowel and bladder incontinence and gait apraxia (wide base of support, short stride, and shuffling gait) also are a part of this syndrome.
Middle Cerebral Artery Syndrome. Middle cerebral artery infarction is one of the most common manifestations of cerebrovascular disease. Symptoms vary depending on the location of the occlusion and the presence of collateral flow. When there is occlusion prior to the lenticulostriate branches, extensive infarctions will occur and involve the internal capsule and the opercular region. Patients experience a dense hemiplegia with paralleling sensory disturbance. Arm and face deficits are worse than those of the leg. Blindness may occur in half the visual field. Right-left confusion, agraphia (loss or impairment of the ability to produce written language), acalculia (impairment of previous ability to perform simple mathematic calculations), aphasia (if the dominant hemisphere is affected), and agnosia (inability to recognize known objects) also can occur.
Posterior Cerebral Artery Syndrome. The posterior cerebral artery supplies portions of the parietal and occipital lobes; therefore, occlusion leads to vision and thought-processing impairment. Patients may experience blindness in half the visual field, third-nerve paralysis, visual agnosia, altered mental status with impaired memory, and cortical blindness.
Lacunar Syndromes. These occur when small penetrating arteries become occluded. They may be single, multiple, symptomatic, or asymptomatic. Multiple lacunar infarctions are associated with progressive cognitive decline. Five syndromes of lacunar infarcts have been recognized:
Vertebrobasilar Artery Syndrome. The posterior inferior cerebellar artery supplies various areas; thus, there are different patterns of infarction. Medial branch territory infarction affecting the vermis and vestibulocerebellum causes vertigo, ataxia, and nystagmus. With lateral cerebellar hemispheric involvement, patients experience vertigo, gait ataxia, nausea, vomiting, gaze palsies, miosis, and dysarthria. With large infarctions, the patient may have altered consciousness or be confused. Patients with anterior inferior cerebellar artery syndrome may have ipsilateral deafness, vertigo, nausea, vomiting, nystagmus, ipsilateral facial hypalgesia (decreased sensitivity to pain), thermoanesthesia, and corneal hypesthesia (increased sensitivity to touch).
Anterior Choroidal Artery Syndrome. This syndrome is rare and can be asymptomatic. However, occlusion of the anterior choroidal artery may produce contralateral hemianopia, hemiplegia, and hemihypalgesia. Because the distribution of this artery is quite variable, symptoms of its disruption show considerable variation. Visual disturbances include a homonymous hemianopsia of variable density with sloping margins. If the posterior limb of the internal capsule is infarcted, contralateral hemimotor or sensory loss may occur. If the retrocapsular sensory and visual radiations are infarcted, a superior homonymous quadrantopia may occur. Middle cerebral artery syndrome may be mimicked by the dysphasia, apraxia, and hemineglect that can be associated with the anterior choroidal artery syndrome.
1. Anderson RN. Deaths: Leading causes from 1999. National Vital Statistics Reports, Vol. 49, No. 11, Oct. 12, 2001.
2. Biller J, Love BB. Vascular diseases of the nervous system: Ischemic cerebrovascular disease. In: Bradley WG et al, eds. Neurology in Clinical Practice: The Neurological Disorders, 3rd edition. Boston: Butterworth Heinemann; 2000:1125-1231.
3. Elkind MS. Stroke in the elderly. Mt. Sinai Journal of Medicine 2003;70:27-37.
4. Jacobs BS, Boden-Abala B, Lin IF, et al. Stroke in the young in the northern Manhattan Stroke Study. Stroke 2002;33:2789-2793.
5. Stroke: Hope Through Research. National Institute of Neurologic Disorders and Stroke, National Institutes of Health, www.ninds.nih.gov/health_and_medical/pubs/stroke_hope_through_research.htm, May 1999 (accessed Nov. 7, 2003).
6. Miniño AM, Arias E, Kochanek KD, et al. Deaths: Final data for 2000. National Vital Statistics Reports, Vol. 50, No. 15, Sept. 16, 2002.
7. Kurtze JF, Wallin MT. Neuroepidemiology. In: Bradley WG, et al, eds. Neurology in Clinical Practice: The Neurological Disorders, 3rd edition. Boston: Butterworth-Heinemann; 2000:759-776.
8. Aho K, Harmsen P, Hatano S, et al. Cerebrovascular disease in the community: Results of a WHO collaborative study. Bull World Health Organ 1980;58:113-130.
9. Borg KT, Pancioli AM. Transient ischemic attacks: An emergency medicine approach. Emerg Med Clin North Am 2002;20:597-608.
10. Kelly RE. Cerebral vascular disorders. In: Weiner WJ, Shulman LM, eds. Emergency and Urgent Neurology, 2nd edition. Philadelphia: Lippincott Williams & Wilkins; 1999:23-52.
11. Thurman RJ, Jauch EC. Acute ischemic stroke: Emergent evaluation and management. Emerg Med Clin North Am 2002;20:609-630.
12. Panzer RJ, Feibel JH, Barker WH, et al. Predicting the likelihood of hemorrhage in patients with stroke. Arch Intern Med 1985;145: 1800-1803.
13. Samuels MA, Feske SK, eds. Office Practice of Neurology, 2nd edition. Philadelphia: Churchill Livingston; 2003.
14. Moster ML, Johnston DE, Reinmuth OM. Chronic subdural hematoma with transient neurologic deficits: A review of 15 cases. Ann Neurol 1983;14:539-542.
15. Pazevic JP, Hambrick RL, Roskam SA. Hypoglycemic hemiplegia. J Am Osteopath Assoc 1990;90:539-541.
16. Carter F, Taylor C. Transient hypoglycemic hemiparesis. J Natl Med Assoc 2002;94:999-1001.
17. Spiller HA, Schroeder SL, Ching DS. Hemiparesis and altered mental status in a child after glyburide ingestion. J Emerg Med 1998;16: 433-435.
18. Wallis WE, Donaldson I, Scott RS. Hypoglycemia masquerading as cerebrovascular disease (hypoglycemic hemiplegia). Ann Neurol 1985;18:510-512.
19. Huff JS. Stroke mimics and chameleons. Emerg Med Clin North Am 2002;20:583-595.
20. Elkind MS, Mohr JP. Differential diagnosis of stroke. In: Rowland LP, ed. Merritt’s Neurology, 10th edition. Philadelphia: Lippincott Williams & Wilkins; 2000:247-249.
21. Little N. Special neurologic problems. In: Rosen P, Barkin R, Danzl D, et al, eds. Emergency Medicine: Concepts and Clinical Practice, 4th edition. St. Louis: Mosby; 1998:2212-2226.
22. Gibbs MA, Wolfson AB, Tayal VS. Electrolyte disturbances. In: Rosen P, Barkin R, Danzl D, et al, eds. Emergency Medicine: Concepts and Clinical Practice, 4th edition. St Louis: Mosby; 1998: 2432-2456.
23. Adams HP, Adams RJ, Brott T, et al. Guidelines for the early management of patients with ischemic stroke: A scientific statement from the stroke council of the American Stroke Association. Stroke 2003;34:1056-1083.
24. Matthews J. Hypertension. In: Rosen P, Barkin R, Danzl D, et al, eds. Emergency Medicine: Concepts and Clinical Practice, 4th edition. St. Louis: Mosby; 1998:1755-1769.
25. Feied C. Pulmonary embolism. In: Rosen P, Barkin R, Danzl D, et al, eds. Emergency Medicine: Concepts and Clinical Practice, 4th edition. St. Louis: Mosby; 1998:1770-1805.
26. Mitchell G. Multiple sclerosis. In: Biller J, ed. Practical Neurology, 2nd edition. Philadelphia: Lippincott Williams & Wilkins; 2002: 518-531.
27. Short R, Graff-Radford N. Approach to the patient with dementia. In: Biller J, ed. Practical Neurology, 2nd edition. Philadelphia: Lippincott Williams & Wilkins; 2002:19-26.
28. Bhalla A, Singh R, Sachdev A, et al. Circadian pattern in cerebrovascular disorders. Neurol India 2002;50:526-527.
29. Hill M, Newcommon N. Diurnal variance in stroke onset. Circulation 2003;107:589-590.
30. Kario K, Pickering T, Umeda Y. Morning surge in blood pressure as a predictor of silent and clinical cerebrovascular disease in elderly hypertensives: A prospective study. Circulation 2003;107: 1401-1406.
31. Kaplan N. Morning surge in blood pressure. Circulation 2003;107:1347-1348.
32. Johnston SC, Sidney S, Bernstein AL, et al. A comparison of risk factors for recurrent TIA and stroke in patients diagnosed with TIA. Neurology 2003;60:280-285.
33. MacMahon S, Rodges A. Primary and secondary prevention of stroke. Clin Exp Hyperten 1996;18:537-546.
34. Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. U.S. Department of Health and Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute, www.nhlbi.nih.gov/guidelines/hypertension/express.pdf, May 2003 (accessed November 7, 2003).
35. Johnston SC. Transient ischemic attack. N Engl J Med 2002;347:1687-1692.
36. Glotzer TV, Hellkamp AS, Zimmerman J, et al. Atrial high rate episodes detected by pacemaker diagnostics predict death and stroke: Report of the Atrial Diagnostics Ancillary Study of the MOde Selection Trial (MOST). Circulation 2003;107:1614-1619.
37. Haynes RB, Taylor DW, Sackett DL, et al. Prevention of functional impairment by endarterectomy for symptomatic high-grade carotid stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. JAMA 1994;271:1256-1259.
38. Barnett HJ, Meldrum HE, Eliasziw M. North American Symptomatic Carotid Endarterectomy Trial (Collaborators). The appropriate use of carotid endarterectomy. CMAJ 2002;166:1169-1179.
39. Mohr IP. Medical and surgical management of carotid stenosis. In: Toole JF, ed. Cerebrovascular Disorders, 5th edition. Philadelphia: Lippincott Williams & Wilkins; 1999:40-59.
40. Sahay BK, Sahay RK. Neurological emergencies: Diabetes management. Neurol India 2001;49 (suppl I):S31-S36.
41. Challa V. Atherosclerosis of the cervicocranial arteries. In: Toole JF, ed. Cerebrovascular Disorders, 5th edition. Philadelphia: Lippincott Williams & Wilkins; 1999:18-39.
42. Reynolds K, Lewis LB, Nolen JD, et al. Alcohol consumption and risk of stroke: A meta-analysis. JAMA 2003;289:579-588.
43. Sacco RL, Boden-Abala B, et al. Stroke incidence among white, black and Hispanic residents of an urban community: The Northern Manhattan Stroke Study. Am J Epidemiol 1998;147:259-268.
44. Wassertheil-Smoller S, Hendrix SL, Limacher M, et al. WHI Investigators. Effect of estrogen plus progestin on stroke in postmenopausal women: The Women’s Health Initiative: A randomized trial. JAMA 2003;289:2673-2684.
45. Naidech AM, Weisberg LA. Treatment of chronic hypertension for the prevention of stroke. South Med J 2003;96:359-362.
46. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002;288:2981-2997.
47. Ezekowitz JA, Straus SE, Majumdar SR, et al. Stroke: Strategies for primary prevention. Am Fam Phys 2003;68:2379-2386.
48. Kirshner HS. Medical prevention of stroke, 2003. South Med J 2003;96:354-358.
49. Urbano LA, Bobousslavsky J. Antiplatelet drugs in ischemic stroke prevention: From monotherapy to combined treatment. Cerebrovasc Dis 2004;17 Suppl 1:74-80.
50. Straus SE, Majumdar SR, McAlister FA. New evidence for stroke prevention: Clinical applications. JAMA 2002;288:1396-1398.
51. Hass WK, Easton JD, Adams HP Jr, et al. A randomized trial comparing ticlopidine hydrochloride with aspirin for the prevention of stroke in high-risk patients: Ticlopidine Aspirin Stroke Study Group. N Engl J Med 1989;321:501-507.
52. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). CAPRIE Steering Committee. Lancet 1996;348:1329-1339.
53. Barnett HJ. Stroke prevention in the elderly. Clin Exp Hypertens 2002;24:563-571.
54. Ringleb PA, Bhatt DL, Hirsch AT, et al. Clopidogrel Versus Aspirin in Patients at Risk of Ischemic Events Investigators. Benefit of clopidogrel over aspirin is amplified in patients with a history of ischemic events. Stroke 2004;35:528-532.
55. Teal PA. Recent clinical trial results with antiplatelet therapy: Implications in stroke prevention. Cerebrovasc Dis 2004;17 Suppl 3:6-10.
56. Ruland S, Raman R, Chaturvedi S, et al. Awareness, treatment and control of vascular risk factors in African Americans with stroke. Neurology 2003;60:64-68.
57. Broderick J, Brott T, Kothari R, et al. The Greater Cincinnati/Northern Kentucky Stroke study: Preliminary first-ever and total incidence rates of stroke among blacks. Stroke 1998;29:415-421.
58. Duncan GE. Ischemic stroke in sickle cell disease: A review. Tenn Med 1997;90:498-499.
59. Klausner H, Lewandowski C. Infrequent causes of stroke. Emerg Med Clin North Am 2002;20:657-670.
60. Fisher CM. Transient ischemic attacks. N Engl J Med 2002;347: 1642-1643.
61. Carotid territory syndromes. In: Toole JF, ed. Cerebrovascular Disorders, 5th edition. Philadelphia: Lippincott Williams & Wilkins; 1999:129-159.
62. Barsan W, Kothari R. Stroke. In: Rosen P, Barkin R, Danzl D, et al, eds. Emergency Medicine: Concepts and Clinical Practice, 4th edition. St. Louis: Mosby; 1998:2184-2198.