Clinical manifestations

Presentation and course

The onset of a stroke is typically sudden, although the precise onset may not be recognized in some individuals. Twenty-five percent of patients wake up with stroke symptoms (60). Ischemic stroke is frequently dichotomized into cortical and subcortical presentations and anterior and posterior circulation distributions.

Both cortical and subcortical strokes may manifest with focal weakness, sensory loss, and dysarthria. Therefore, these findings do not help distinguish location. Aphasia (most often a left hemisphere event) and neglect (most often a right hemisphere event) occur most frequently with cortical strokes. Gaze deviation away from the side of weakness suggests a cortical stroke ipsilateral to the side of gaze deviation. Gaze deviation toward the side of weakness suggests a subcortical event, typically in the pons contralateral to the side of gaze deviation. Occasionally, medial thalamic and supratentorial lesions produce ipsilateral version of the eyes toward the hemiparesis, known as “wrong-way deviation.” Epilepsy may also produce gaze deviation towards hemiparetic limbs but subtle nystagmoid jerks of the eyes may help distinguish this condition from stroke.

Distinguishing anterior (or carotid territory) from posterior (or vertebrobasilar territory) circulation strokes is important when deciding on the necessity and relevance of diagnostic testing (eg, carotid imaging). Approximately 80% of ischemic strokes involve the anterior circulation, and 20% involve the posterior circulation. Clinical features suggestive of ischemia in the anterior circulation are monocular visual loss, aphasia, and neglect. Posterior circulation infarction is suggested by diplopia, binocular visual loss, vertigo, hearing loss, lower motor neuron facial weakness, bilateral or alternating weakness, delirium or severely altered level of consciousness, and visual hallucinations. Focal weakness, focal sensory disturbance, dysarthria, and dysphagia are not helpful per se in distinguishing between anterior and posterior circulation ischemia. Headache occurs in approximately one third of patients and is less frequent in those with pure motor or sensory disturbance. The combination of headache and vomiting is more likely to be associated with intracerebral hemorrhage, especially when accompanied by rapidly progressing neurologic deficits.

Prognosis and complications

Mortality from stroke has substantially reduced over the last two decades, owing to advances in prevention and acute treatment (181). Major complications from stroke include aspiration pneumonia, venous thromboembolism, and cerebral edema (51). Because of advances in screening and treatment related to these conditions, mortality related to these events has significantly reduced. Risk of mortality increases with advancing age, and outcomes following interventions such as thrombolysis, mechanical thrombectomy, and decompressive hemicraniectomy are not as good in older patients. Older patients are also more likely to have care withdrawn than younger patients, which may also contribute to worse outcomes.

Clinical vignette

A 61-year-old right-handed man with hypertension, diabetes mellitus, and hypercholesterolemia presented with sudden-onset dysarthria and left face and arm weakness and tingling. Symptoms began one day prior to presentation. His baseline examination showed minor flattening of the left nasolabial fold, mild dysarthria, a left pronator drift, a mild left hemiparesis (grade 4/5), and decreased pinprick sensation in the left arm. National Institutes of Health Stroke Scale score was 4. The clinical diagnosis was sensorimotor lacunar syndrome due to a right subcortical infarction. Diffusion-weighted MRI showed a small (less than 15 mm) right corona radiata infarction.

Magnetic resonance angiography of the head and neck were negative.

Computed tomographic angiography of the head and neck were negative.

Transesophageal echocardiogram showed normal cardiac function with an ejection fraction of 55% to 65% and no evidence of mural thrombus or intracardiac shunt with agitated saline challenge. EKG showed normal sinus rhythm. Homocysteine level was 9 mg/d. Lipid profile was as follows: cholesterol 141 mg/dL, HDL 32 mg/dL, LDL 68 mg/dL. Liver function studies showed an AST of 20 IU/L and ALT of 21 IU/L. He was discharged home on aspirin 81 mg daily, simvastatin 10 mg daily, pioglitazone 5 mg daily, and lisinopril 5 mg daily.

At 1-month follow-up, the patient denied any symptoms, and his neurologic examination was normal. The average of three blood pressure measurements was 168/98 mmHg. Lisinopril was increased to 10 mg daily. A written prescription for exercise was given to him, and he was also given a copy of the DASH diet (32). A follow-up appointment was made for one month to recheck blood pressure. Repeat lipid profile and liver function testing were planned for 6 months after stroke onset.

Biological basis

Etiology and pathogenesis

Ischemic strokes can be due to local vascular occlusion (thrombus), occlusion from intravascular material that originates elsewhere (embolism), or poor perfusion through a narrowed vessel (hypoperfusion or watershed infarction). With large-artery atherosclerosis, there may be artery-to-artery embolization or formation of a thrombus at a site of preexisting stenosis. The heart, aorta, or deep venous system with an intracardiac shunt can be potential sources of embolism. A preexisting stenosis in the internal carotid artery can lead to a watershed infarction in the zones between the anterior, middle, and posterior cerebral arteries if blood pressure is suddenly and profoundly reduced (eg, cardiac arrest, surgery with hypotension).

The etiology of stroke depends on the age of the patient with atherosclerosis playing a more prominent role in those older than 55 years of age. Younger patients are more likely to have arterial dissections, right-to-left shunting, or coagulopathies associated with their stroke.

In a population-based series, the etiology of stroke was cardioembolic in about 30%, small vessel disease (also known as lacunar) in 20%, large vessel atherosclerosis in 15%, and undetermined in 35% (167; 128). The proportion of strokes attributable to small vessel disease is higher among African Americans (199) and Chinese populations (236), though the incidence rate of each subgroup, including large vessel atherosclerosis, is higher among African Americans as compared with whites (199). The INTERSTROKE study found that 10 risk factors were associated with 90.3% of all strokes (162). These risk factors included current hypertension (higher than 160/90 mmHg) or history of hypertension (percent attributable risk or PAR 36.8%), current smoking (PAR 18.9%), waist-to-hip ratio (PAR 26.5%), diet (PAR 18.8%), physical activity (PAR 28.5%), diabetes mellitus (PAR 5.0%), alcohol intake (PAR 3.8%), psychosocial stress and depression (PAR 5.0%), cardiac causes (PAR 6.7%), and ratio of apolipoproteins B to A1 (PAR 24.9%). By stroke subtype, all of these factors were significant for ischemic stroke, whereas hypertension, smoking, waist-to-hip ratio, diet, and alcohol intake were significant risk factors for intracerebral hemorrhage.

In a large series of stroke in young adults (ages 15 to 45), the etiologies of stroke were: large-artery atherosclerosis 9.7%, small artery occlusion 7.9%, cardioembolism 17.6%, other 30.4%, and undetermined 34.3% (06). The category of “other” includes prothrombotic states, arterial dissection, connective tissue diseases, and moyamoya disease. Children (younger than 15 years of age) are more likely to have prothrombotic states and less likely to have arterial dissection as a cause of their strokes (227).

A large proportion of strokes are without defined cause. There is likely a heritable explanation in many of these cases. In the Framingham Heart study, a family history of parental stroke before 65 years of age increased the risk of stroke in offspring by 3-fold even after adjustment for conventional risk factors (201).

Atrial fibrillation plays an increasingly important role among older patients. Up to 20% of events thought to be cryptogenic are, in fact, associated with atrial fibrillation with prolonged posthospitalization cardiac monitoring (194).

Pathogenesis. An ischemic stroke results when cerebral blood flow to an area of the brain is interrupted. Ischemia produces impaired energy metabolism and depolarization of cells that leads to an accumulation of calcium ions in the intracellular space, elevated lactate levels, acidosis, and production of free radicals. If the disruption is severe enough, cell death occurs. Activation of the N-methyl-D-aspartate receptor by an increase in glutamate leads to a cascade of chemical reactions that ultimately leads to cell death (“theory of excitotoxicity”). Modulators of this receptor include polyamines, glycine, magnesium, zinc, and phencyclidine. However, magnesium delivered in the first hour after stroke fails to improve outcome (198). Normal adult brain cerebral blood flow is 50 to 60 mL/100g/minute. When cerebral blood flow falls below 18 mL/100g/minute in baboons, sensory evoked potentials disappear (214). In the same experiment, when cerebral blood flow fell below 12 mL/100g/minute, infarction occurred. Therefore, cerebral blood flow between 10 and 20 mL/100g/minute is considered consistent with ischemic penumbra. Cerebral blood flow below 10 mL/100g/minute is considered compatible with infarction. These delineations are not absolute because time is also a factor in the fate of tissue. Cerebral blood flows of 5 mL/100g/minute result in infarction within 30 minutes, whereas those between 5 and 15 mL/100g/minute result in infarction after 1 to 3 hours (142).

Gross pathology. The pathological characteristics of ischemic stroke are dependent on the mechanism of the stroke, the size of the obstructed artery, and the availability of collateral blood flow. There may be advanced changes of atherosclerosis visible within arteries.

The surface of the brain in the area of infarction appears pale. With ischemia due to hypotension or hemodynamic changes, the arterial border zones may be involved. A wedge-shaped area of infarction in the center of an arterial territory may result if there is occlusion of a main artery in the presence of collateral blood flow. In the absence of collateral blood flow, the entire territory supplied by an artery may be infarcted. With occlusion of a major artery, such as the internal carotid artery, there may be a multilobar infarction with surrounding edema. There may be flattening of the gyri and obliteration of the sulci with cerebral edema. A lacunar infarction in subcortical regions or the brainstem may be barely visible, with a size of 1.5 cm or less. Emboli to the brain tend to lodge at the junction between the cerebral cortex and the white matter. There may be early reperfusion of the infarct when the clot lyses, leading to hemorrhagic transformation. Over time the necrotic tissue is absorbed (leaving a cystic cavity) and is surrounded by a glial scar.

The initiation, progression, and activation of atherosclerosis are predominantly inflammatory conditions produced by a “response to injury” mechanism after exposure to certain injurious vascular risk factors. Individual genetic profiles can affect pathophysiological mediators of plaque development, symptomatic manifestations, and recovery from strokes associated with cerebrovascular atherosclerosis. A number of stimuli can initiate endothelial injury, which, in turn, leads to a cascade of events that result in lipid deposition and inflammatory cell migration (74; 233). This inflammatory process includes the increased expression of adhesion molecules, cytokines, chemokines, metalloproteinases, and antigen-mediated activation of macrophages and T-lymphocytes. As the plaque matures, platelet aggregation and clot formation with or without plaque rupture may ensue. The clinical result is an atherothrombotic ischemic stroke.

Microscopic pathology. Microscopic changes after infarction depend on the age of the infarction and may be delayed up to 6 hours after infarction. Initially there is neuronal swelling, followed by shrinkage, hyperchromasia, and pyknosis. Chromatolysis appears and the nuclei become eccentric. There is swelling and fragmentation of the astrocytes and endothelial swelling. Neutrophils infiltrate as early as four hours after the ischemia and become abundant by 36 hours. Within 48 hours, the microglia proliferate and ingest the products of myelin breakdown and form macrophages. There is neovascularity with proliferation of capillaries and increased prominence of the existing capillaries. The elements in the area of necrosis are gradually reabsorbed and a cavity, consisting of glial and fibrovascular elements, forms. In a large infarction, there are three distinct zones: an inner area of coagulative necrosis; a central zone of vacuolated neuropil, leukocytic infiltrates, swollen axons, and thickened capillaries; and an outer marginal zone of hyperplastic astrocytes and variable changes in nuclear staining.

Genetics. As with most diseases, stroke is a result of the interaction between genetics and environmental exposure (88). There are a number of genetic causes of stroke (154). These disorders often result in an early age of stroke onset (ie, younger than 40 years of age). Some inherited diseases predispose to accelerated atherosclerosis, such as the hereditary dyslipoproteinemias. A number of inherited diseases are associated with nonatherosclerotic vasculopathies, including cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) (117), Ehlers-Danlos (type 4) syndrome, Marfan's syndrome, Rendu-Osler-Weber disease, and Sturge-Weber syndrome. Inherited cardiac disorders that predispose to stroke include familial atrial myxomas, hereditary cardiomyopathies, and hereditary cardiac conduction disorders. Inherited hematologic abnormalities that are associated with venous stroke include deficiencies of protein C, S, and antithrombin III. Other hematologic abnormalities including mutation of factor V Leiden, polymorphism of thermolabile methylenetetrahydrofolate reductase, and G20210A mutation of the prothrombin gene have been associated with venous and arterial stroke. Sickle cell disease is a well-known cause of stroke and frequently leads to strokes during childhood (56). Finally, rare inherited metabolic disorders that can cause stroke include mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (165), Fabry disease (155), and homocystinuria. Other potential genetic factors, particularly for carotid atherosclerosis, include PDE4D (100), interleukin-1 receptor antagonist (IL-1ra) (232), toll-like receptor-4 (TL4) (176), 5-lipoxygenase (5-LO) (80), interleukin-6 (IL-6) (59), hepatic lipases (184), cyclooxygenase 2 (COX-2) (66), and matrix metalloproteinase polymorphisms (MMP) (01). More recently, the potential importance of certain microRNAs (miRNAs) has been demonstrated in the walls of unstable plaques (65).

Epidemiology

Approximately 6.6 million Americans 20 years and older have had a stroke (158). Stroke is the fifth leading cause of death in the United States and a leading cause of serious long-term disability. Approximately 795,000 people in the United States experience a stroke annually. About 610,000 of these strokes are first-time events, and 185,000 of the events represent recurrent events. Blacks have twice the risk of first-ever stroke compared with whites. Mexican Americans are twice as likely to suffer ischemic strokes as non-Hispanic Caucasians between the ages of 45 and 59 years (157). The direct and indirect cost of stroke for 2008 was .3 billion.

The prevalence of silent cerebral infarction in adults ranges between 6% and 28% depending on definition, populations studied, and imaging technique. The prevalence increases with age. An estimated 13 million people in the United States had a silent stroke in 1998 (158).

The average annual incidence rate of stroke based on data from a study of Rochester, Minnesota residents showed a decline from 183 to 102 per 100,000 between 1945 to 1954 and 1975 to 1984 (48). Postulated reasons for the decline in incidence include effective treatment of hypertension. Introduction of CT scanning increased the incidence of strokes from 1980 to 1984 compared with 1975 to 1979 by detecting smaller events. A study from the Greater Cincinnati/Northern Kentucky region found that the average age of stroke had declined from 71.3 years of age in 1993 to 1994 to 70.9 years of age in 1999 and 68.4 years of age in 2005 (125). The proportion of stroke patients who were younger than 45 years of age rose from 4.5% in 1993 to 1994 to 5.5% in 1999 and 7.3% in 2005. The incidence rate for stroke more than doubled in white patients aged 20 to 45 years old (from 12 per 100,000 to 25 per 100,000). The cause of this change is uncertain but may reflect changes in diet and physical activity. Stroke rates in those over 65 years of age declined, possibly reflecting improved treatment of hypertension in this group.

Prevention

The cornerstone of recurrent stroke risk reduction is modification of risk factors. The risk factors for stroke include the following:

Unmodifiable risk factors

Age. Age is the strongest determinant of stroke. The prevalence of silent infarction is 11%, 22%, 28%, 32%, and 40% between the ages of 55 and 64, 65 and 69, 70 and 74, 75 and 79, and 80 and 84 respectively (106). Most strokes occur in individuals older than 65, but approximately 25% occur below that age making it an important cause of lost income among those in the workforce.

Male gender. The incidence of stroke is greater in men when considering individuals younger than 85 years of age. After 85 years of age, strokes occur more often in women. Overall, incident rates are 1.25 times greater in men than women (158). However, 60,000 more women than men suffer from stroke each year, which may be attributable to the more common occurrence of stroke at older ages and greater longevity of life in women.

Personal vascular history. Approximately 10% of patients with transient ischemic attack will suffer from stroke within 90 days of the attack; half of which occur within the first days of the event (115). Transient ischemic attack precedes stroke in approximately 25% of cases (183). The risk of stroke is greater with hemispheric than with retinal transient ischemic attacks (36).

History. A prior history of stroke increases the risk of recurrent stroke. In Rochester, Minnesota, the 5-year probability of recurrent stroke is about 29%, with 12% of events occurring in the first year (166).

Race. African Americans have a greater incidence of stroke than whites, even after controlling for hypertension, diabetes mellitus, and age (127). Mexican Americans also have a higher risk of stroke relative to white patients and are twice as likely to suffer from stroke between the ages of 45 and 59 (157). A study of African Americans, Hispanics, and whites in Northern Manhattan found that African Americans had a substantially higher incidence of stroke whereas Hispanics and whites had more similar incidences (187). Individuals who are American Indian or Alaska native may have the highest prevalence of stroke overall, but data are considered unreliable (221).

Left atrial size. In one study, for every 10 mm increase in left atrial size on echocardiography, the relative risk of stroke is 2.4 in men and 1.4 in women (38).

Late-life epilepsy. One study found that a new onset of seizures after the age of 60 increases the risk of subsequent stroke by 2.89 times (67).

Heredity. An increased risk of stroke is seen in persons with a family history of stroke among first-degree relatives (122; 226). A number of genetic disorders increase the risk of stroke, such as hereditary dyslipoproteinemias and hereditary cardiac disorders. Some connective tissue disorders such as Ehlers-Danlos syndrome are associated with nonatherosclerotic vasculopathies. Rare inherited hematologic disorders such as deficiencies of protein C, protein S, and antithrombin III are associated with an increased risk of stroke.

Modifiable risk factors

Hypertension. Hypertension increases the risk of stroke approximately 2- to 4-fold (73; 229).

Physical inactivity. Physical fitness is associated with a reduced risk of ischemic and hemorrhagic stroke (131; 134). Low VO(2) max, a marker of physical fitness may be as important a risk factor as hypertension (131). Men who have a VO(2) max less than 25.2 mL/kg/minute have a relative risk of ischemic stroke of 3.5 compared to men whose VO(2) max is greater than 35.3 mL/kg/minute.

Atrial fibrillation. Atrial fibrillation increases the risk of stroke approximately 5-fold. The attributable risk of stroke rises from 1.5% for those between the ages of 50 and 59 to 23.5% for those between the ages of 80 and 89 (228).

Diabetes mellitus. Diabetes increases the risk of stroke approximately 2-fold (73; 229).

Cigarette smoking. The risk of stroke in heavy smokers (more than 40 cigarettes per day) is twice that of light smokers (less than 10 cigarettes per day). The risk of stroke for smokers is the same as that of nonsmokers five years after quitting (230).

Abnormalities of serum lipids. Abnormalities of serum lipids are a stronger risk factor for coronary artery disease than for stroke. The degree and progression of carotid arthrosclerosis are directly related to cholesterol and low-density lipoprotein and inversely related to high-density lipoprotein (174; 190). Data regarding the effect of cholesterol on the risk of stroke are conflicting. In the Framingham study, LDL cholesterol had no relationship with stroke in men but was associated with an increased risk in women (96). In contrast, the Multiple Risk Factor Intervention Trial and the Honolulu Heart Study showed a relationship between cholesterol levels and stroke (109; 37).

Alcohol. Though there are conflicting reports regarding the association between alcohol consumption and stroke, most studies support such an association when consumption is in excess of 40 grams per day (35). Though some studies have suggested a possible protective effect of moderate alcohol consumption, problems with study methodology have limited any firm conclusions (224).

Obesity. Several studies demonstrate that abdominal obesity is a risk factor for stroke (107; 226; 45; 89; 212).

Obstructive sleep apnea. Obstructive sleep apnea is an independent risk factor for stroke and death and increases the risk nearly 2-fold. Increasing severity of sleep apnea is associated with an increasing risk of stroke and death (235).

Other cardiac diseases. Cardiac disorders such as valvular heart disease, coronary artery disease, acute myocardial infarction, congestive heart failure, infective endocarditis, and left ventricular hypertrophy increase the risk of stroke.

Aortic arch atheromas. Protruding aortic arch atheroma is associated with strokes that are otherwise unexplained. The association of aortic arch atheromatosis is particularly strong with mobile and thick atherosclerotic plaques measuring greater than 4 mm in thickness (13).

Homocysteine. Elevated homocysteine levels are a risk factor for stroke (44). Lowering homocysteine levels with vitamin B6, vitamin B12, and folate supplementation does not appear to reduce stroke risk (217; 138).

C-reactive protein. Elevated levels of C-reactive protein increase the risk of stroke (130). Statin drugs, particularly at doses above 40 mg daily, substantially reduce C-reactive protein levels (177). However, no trial of statins specifically targeting lower C-reactive levels in stroke patients has been done yet.

Oral contraceptives. Oral contraceptives increase the risk of stroke, but because the incidence of stroke is so low, the absolute event rate is one per 24,000 women (93).

Hematologic disorders. Alterations in hemostasis are associated with an increased risk of vascular events, particularly those of an ischemic nature. An elevated serum fibrinogen level is an independent risk factor for ischemic stroke. It is thought that an elevated fibrinogen level may reflect the progression of atherogenesis. In addition, fibrinogen levels are closely associated with other stroke risk factors such as cigarette smoking, arterial hypertension, diabetes mellitus, obesity, hematocrit levels, and spontaneous echocardiographic contrast. Antiphospholipid antibodies are antibodies to protein that bind to phospholipids. In small case series, antiphospholipid antibodies have been associated with arterial and venous thrombosis, recurrent pregnancy loss, and cardiac and neurologic involvement. Based on randomized stroke trials, isolated antiphospholipid antibodies may not be a risk factor for recurrent arterial stroke, particularly in older patients (136). It is unknown whether elevated cardiolipin, phosphatidylserine, or phosphatidylethanolamine antibodies are a cause of stroke or are only a marker of the presence of other precursors of stroke.

Differential diagnosis

Intracerebral hemorrhage may mimic ischemic stroke because the symptoms and signs are often indistinguishable. A severe headache, neck pain, nuchal rigidity, coma, and vomiting are more likely with a hemorrhage. The differential diagnosis of ischemic stroke also includes a brain tumor, which can usually be differentiated by a more gradual onset of symptoms with an accumulation of deficits. Seizures can usually be differentiated by the presence of associated involuntary motor movements with a march of sensory or motor symptoms. Migraines can be differentiated by the presence of fleeting positive visual phenomena, such as scintillating scotomata. However, the association of focal neurologic symptoms in association with a headache, as seen in complicated migraines, may warrant neuroimaging. Occasionally, some individuals with migraine headaches and focal neurologic symptoms are found to have new ischemic infarction. Other conditions in the differential diagnosis include cerebral venous occlusive disease, cerebral abscess, syncope, and conversion disorder.

Diagnostic workup

The goals for the diagnostic evaluation are to establish the diagnosis of ischemic stroke as a cause of the patient's symptoms and to determine the underlying cause of the event.

A basic evaluation that should be considered in all patients with ischemic stroke includes a complete blood count with differential and platelet count, prothrombin time with international normalized ratio, partial thromboplastin time, serum chemistries including plasma glucose level, blood urea nitrogen, and serum creatinine; lipid analysis; liver function tests; electrocardiography; brain imaging with CT or MRI; and vascular imaging with computed tomographic angiography, Doppler, or magnetic resonance angiography. Critical elements in the initial evaluation include identification of cerebral hemorrhage, which would preclude treatment with antiplatelet agents and anticoagulants; identification of atrial fibrillation, which would be a strong indication for anticoagulation; and identification of severe stenosis in the extracranial internal carotid artery ipsilateral to hemispheric symptoms, which would be a strong indication for invasive repair. Additional testing for cardiac sources of embolism includes transthoracic echocardiography and transesophageal echocardiography; the latter is considered to be more sensitive, particularly because of a superior ability to evaluate the left atrial appendage. However, both modalities are considered to be low yield (identification of potential management altering findings in less than 10% of patients studied). Routine transesophageal echocardiography may be cost effective, but routine transthoracic echocardiography is not (150). The utility of transesophageal echocardiography may be increased in patients younger than 60 years of age without other obvious cause of stroke, such as carotid stenosis. Holter monitoring and 7-day ambulatory monitoring, when used in the general stroke population, are of low yield when trying to identify atrial fibrillation beyond that uncovered by electrocardiography (110). In patients with cryptogenic stroke, prolonged ambulatory monitoring for 30 days or greater identifies up to 20% more people with atrial fibrillation (194).

The choice of screening studies for carotid stenosis depends, in part, on the quality of imaging equipment and technologist acquiring data, particularly in the case of duplex ultrasonography. In one study, compared with catheter angiography, misclassification rates of stenosis, severity was 28% for carotid duplex alone, 18% for magnetic resonance angiography alone, and 8% when both studies were concordant (114). In another study by the same institution, contrast-enhanced magnetic resonance angiography, thought to give better resolution than unenhanced magnetic resonance angiography had a misclassification rate of 24% alone and 17% when concordant with carotid duplex (113). At a separate institution, the misclassification rate of contrast-enhanced magnetic resonance angiography was 15%, but the authors found that clinical decision making would have been altered in 6% (219). In general, magnetic resonance angiography tends to overestimate stenosis because of signal dropout with increasing stenosis. Carotid duplex tends to underestimate stenosis because of occasional difficulty sampling the area of greatest stenosis. Similar to contrast-enhanced magnetic resonance angiography, computed tomographic angiography requires the injection of a contrast agent for visualization of vessels. In one study, agreement between computed tomographic angiography and digital subtraction angiography was 96% using a 70% stenosis cutoff value (116). The study was small, with only 5 of 81 vessels having greater than 70% stenosis. Potential pitfalls with computed tomographic angiography include failure to provide an adequate view of the carotid bulb when there is heavy calcification. Digital subtraction angiography remains the gold standard for cerebrovascular imaging. Complication rates related to the procedure range between 0.7% and 1%. Digital subtraction angiography does not provide information regarding vascular plaque characteristics that may be potentially important. In clinical practice, some clinicians advocate for digital subtraction angiography for all patients considered for carotid endarterectomy based on the misclassification rates described above. Other clinicians are satisfied with two concordant studies because of the 0.7% to 1% complication rate associated with angiography.

Noninvasive intracranial vascular imaging choices include computed tomographic angiography, magnetic resonance angiography, and transcranial Doppler ultrasonography. The SONIA trial found that both transcranial Doppler ultrasonography and magnetic resonance angiography both have substantial negative predictive value compared with conventional angiography for 50% to 99% stenosis (86% and 91%, respectively) but have substantially poorer positive predictive values (36% and 59%, respectively) (84). Computed tomographic angiography has the greatest accuracy with respect to noninvasive diagnosis of intracranial stenosis. The gold standard for intracranial arterial stenosis is digital subtraction angiography. Most recently, arterial wall imaging with MRA has been introduced to identify vulnerable plaque within the intracranial circulation.

Leptomeningeal biopsy is reserved for extraordinary circumstances when the diagnosis of isolated central nervous system angiitis or vasculitis is considered. Sedimentation rate, lumbar puncture, and cerebral angiography are insensitive of nonspecific for the diagnosis. Even leptomeningeal biopsy may be falsely negative if the area sampled is not affected.

Management

Because of differences in approach, management of ischemic stroke will be divided into acute treatment and secondary prevention. The definition of what constitutes the acute management period and the period of secondary prevention is arbitrary. In actuality, these periods overlap because certain strategies for secondary prevention may be effectively introduced when the patient is acutely hospitalized.

The focus of acute management includes recanalization of occluded vessels, preventing expansion of the ischemic tissue volume, preventing secondary complications such as pneumonia and deep venous thrombosis, and enhancing recovery.

Proven acute medical treatment includes intravenous recombinant tissue plasminogen activator (rt-PA). Alteplase can be administered on the basis of plain head CT alone within 4.5 hours of stroke onset (22; 102). After 4.5 hours, advanced imaging with MRI showing a diffusion-FLAIR mismatch (216) or CT perfusion imaging showing a threshold ischemic core-penumbra difference up to nine hours (140) can be used to support treatment with alteplase. Recent analyses and guidelines also support the use of tenecteplase as an alternative to alteplase (172). Studies including ACT (153) and TRACE-2 (222) have demonstrated noninferiority of tenecteplase to alteplase, and many centers have made the transition to tenecteplase because of the ease of use (ie, single bolus delivered over 5 seconds). Patients treated within 90 minutes of symptom onset are more likely to recover than those treated afterwards (144). Posttreatment monitoring is essential to treat blood pressure elevations that may increase the risk of intracranial hemorrhage. A blood pressure of less than 180 mmHg is targeted. Recommended blood pressure-lowering drugs include labetalol, intravenous push, and nicardipine drip. Ideally, patients should be monitored in a dedicated stroke or medical intensive care unit. Surgery has been attempted for patients with intracranial hemorrhage following rt-PA, but results have not been favorable.

Endovascular treatment of stroke has been a popular form of treatment since the early 2000s. Initial randomized trials failed to show a benefit over standard treatment alone (SYNTHESIS, MR RESCUE, IMS III) (47; 64; 124). However, modifications to trial design resulted in statistically and clinically significant results favoring the addition of endovascular treatment in select patients (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, REVASCAT) (40; 52; 98; 197; 118). Changes in approach to treatment included selection of patients with CTA angiography who had a proximal clot (mainly in the M1 segment of the middle cerebral artery), shorter onset to treatment times, and use of stent retriever devices. Some of these studies also excluded patients who had evidence of early major tissue damage (ESCAPE, EXTEND-IA, SWIFT PRIME, REVASCAT). The odds of functional independence were significantly increased with intraarterial therapy. Of approximately four patients who received intraarterial therapy in ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT, one more had a functionally independent outcome. In addition, mortality was reduced from approximately 20% to 10% at 90 days in these trials. Symptomatic intracranial hemorrhage was no more frequent in patients receiving intraarterial therapy than standard treatment and ranged from approximately 0% to 3%. Based on the results of these trials, endovascular treatment up to 6 hours was endorsed by guidelines. In 2018, two trials were published showing the advantage of extending treatment beyond 6 hours. DAWN showed benefit of endovascular treatment up to 24 hours in carefully selected patients who had large mismatches between core infarcted tissue (average less than 10 cc) and penumbra (160). DEFUSE 3, using different criteria for mismatch, established benefit up to 16 hours in patients with small infarct cores (average less than 10 cc) (10). Based on the results of these studies, guidelines endorsed mechanical thrombectomy for ischemic stroke up to 24 hours using specific selection criteria according to the DAWN and DEFUSE 3 studies (172). Planned trials will evaluate whether patients with larger infarct cores might benefit from endovascular treatment beyond 6 hours. Among patients with large core ischemic infarctions (> 50 mL), thrombectomy results in better outcomes than medical management alone, as demonstrated in the SELECT-2 trial (195).

For those unable to receive rt-PA or endovascular treatment, aspirin is a recommended treatment. Approximately one patient out of 100 will be dead or severely disabled at 6 months after receiving this medication. Aspirin should not be given within 24 hours of tPA based on the results of a randomized trial that showed no statistically significant improvement in outcomes at 3 months but a statistically significant increase in symptomatic intracerebral hemorrhage (239).

Despite widespread use in the 1980s and 1990s, fractionated or unfractionated heparin has not been proven to improve outcomes in patients with arterial ischemic stroke. Two meta-analyses of large numbers of patients showed no advantage of immediate anticoagulation after presumed noncardioembolic stroke (n = 2487 patients) and strokes of all types (n = 23,748 patients) (191; 192). Anticoagulation in cerebral venous thrombosis may be of benefit based on a meta-analysis of two trials (n = 79 patients) (210).

Supplemental oxygen is of no benefit in nonhypoxemic stroke patients. The SO2S Study, which included 8000 patients, randomly allocated patients to no supplemental oxygen, nocturnal supplemental oxygen, and continuous supplemental oxygen for 72 hours (178). There was no difference in clinical outcomes at 90 days. Use of 100% oxygen with or without hyperbaric treatment should be avoided at this time given the negative effects reported in two trials (180; 185). The use intravenous fluids and bed positioning maneuvers is optional. However, none of these interventions has been clearly shown to benefit patients with acute ischemic stroke. At this time, there are no approved neuroprotective therapies.

Additional management of acute ischemic stroke includes hospitalization in a dedicated stroke unit. Of patients hospitalized in an acute stroke unit, approximately 14% fewer will be dead and 18% fewer will be dead or severely disabled at one year compared with patients hospitalized on a general medical unit with or without a subspecialty stroke consultation, based on a meta-analysis of 5592 patients (211).

A variety of neurologic and medical complications can occur after ischemic stroke, including cerebral edema, pneumonia, deep venous thrombosis, seizures, cardiac arrhythmias, myocardial infarction, electrolyte disturbances, decubitus ulcers, falls, and urinary tract infections (132). In one study, pneumonia occurred in about one of every 14 patients admitted with ischemic stroke (121). Patients with more severe stroke and features suggesting general frailty were at greatest risk. Pneumonia increased the risk of death at 30 days 3-fold. A swallowing assessment is recommended for most stroke patients. The gag reflex is an unreliable indicator of aspiration risk (42; 133). Certain patients require modified barium swallow assessments depending on bedside findings of a trained assessor. A temporary enteral feeding tube is placed if there is evidence of oropharyngeal dysfunction to minimize the risk of aspiration. Good pulmonary hygiene, including chest percussion, suctioning, drainage, and frequent turning is recommended for bedridden patients. Clinically detected deep venous thrombosis occurs in approximately every 200 patients with ischemic stroke (99). Using magnetic resonance venography, a pelvic deep venous thrombosis may be detected in as many as 12% of patients and in 20% of patients with cryptogenic stroke (69). Prophylaxis with subcutaneous heparin 5000 units subcutaneously twice daily reduces the rate of deep venous thrombosis and should be considered for all nonambulatory patients (92). In a randomized trial of different regimens of venous thromboembolism among stroke patients unable to walk without assistance, the prevalence of venous thromboembolism over a 10-day period was 10% among patients given enoxaparin 40 mg subcutaneously daily and 18% among patients given unfractionated heparin 5000 U subcutaneously every 12 hours (203). Among patients with an NIH stroke scale score of 14 or more, the prevalence of venous thromboembolism was 16% and 30% respectively among enoxaparin-treated and heparin-treated patients, respectively; for those with an NIH stroke scale score less than 14, the prevalence was 8% and 14% respectively. The rates of symptomatic intracranial hemorrhage and major extracranial hemorrhage were 1% for both groups. Compression boots or compression stockings may also be considered, though their value in patients with ischemic stroke is unknown (05). A variety of cardiac arrhythmias can complicate ischemic stroke, especially if there is involvement of the right insula (63). Continuous cardiac monitoring during the first 24 hours is suggested, though no definitive studies show a benefit in terms of outcomes. Because of the risk of urosepsis, indwelling urinary catheters should be placed only if absolutely necessary and should be removed at the earliest possible time to avoid urinary infection. Approximately 20% of patients develop pressure sores after a stroke (132). Steps that may reduce this complication include frequent inspection of the skin, skin cleansing, frequent turning, use of special mattresses and protective dressings, maintaining adequate nutritional status, and early mobilization. Up to 25% of patients with stroke have falls after their stroke with 5% resulting in serious injury (132). Assessments of fall risk should be made at regular intervals during acute hospitalization. Measures should be instituted to reduce the occurrence of falls.

It is not unusual for the blood pressure to be transiently elevated after a stroke with up to 75% of patients exhibiting blood pressures above 140 mmHg systolic. Within a few days, the blood pressure may return to pre-stroke levels. The appropriate management of elevated blood pressure in the first 24 hours is controversial. The most current guidelines by the American Heart Association recommend avoiding treatment unless the systolic blood pressure is greater than 220 mmHg (05) or greater than 180 mmHg in the case of patients treated with thrombolytic therapy. The most important objective is to maintain adequate cerebral blood flow in the presence of impaired autoregulation. The SCAST study found no advantage of candesartan use within the first week of stroke with respect to vascular death, myocardial infarction, and stroke in the first 6 months (approximately 11.5% in both groups) (193). In addition, there was also a suggestion of worse clinical outcomes as measured by the modified Rankin scale at 6 months.

Attention to nutritional status is important. A large proportion of patients with ischemic stroke may be underfed, especially those with severe strokes (57). The FOOD trial found a nonsignificant reduction in risk of death of 5.8% (p=0.09) and a nonsignificant risk of poor outcome of 1.2% (p=0.7), whereas percutaneous endoscopic gastrostomy was associated with a nonsignificant increase in risk of death of 1.0% (p=0.9) and a borderline significant increase of poor outcome of 7.8% (p=0.05) (76).

Rehabilitation after stroke begins as soon as the diagnosis of stroke is established and as soon as any life-threatening neurologic or medical complications have been stabilized. Meta-analyses show that task-oriented exercise training, especially when applied intensively and early after stroke onset, improves outcomes (220). No specific type of physiotherapy treatment has been shown to be superior to another (170). However, physiotherapy, using a mix of components from different approaches, is significantly more effective than no treatment or placebo in the recovery of functional independence following stroke (170). The EXCITE trial showed that compared with usual care constraint-induced movement therapy (wearing a restraining mitt on the less-affected hand while engaging in repetitive task practice and behavioral shaping with the hemiplegic hand) among patients who had suffered from a stroke within the previous 3 to 9 months produced significant improvements in arm motor function that persisted for at least one year (231). Current areas of research interest include robotic-assisted movement therapy, and neurorestorative treatments. When patients are screened for rehabilitation, criteria used to make the decision regarding eligibility include the medical status, neurologic impairment, social and environmental factors (21). Prosthetic devices, canes, or walkers should be used when appropriate.

Secondary stroke prevention consists of lifestyle modification, pharmacological management, and surgical and revascularization procedures.

Lifestyle modification includes improving cardiovascular fitness, altering diet, and smoking cessation. Patients with a low index of physical fitness, as measured by maximum oxygen uptake (or VO2 max), are at higher risk for all stroke subtypes including ischemic and hemorrhagic stroke (131). Those with a VO2 max less than 25 are three times as likely to suffer from stroke compared with those whose VO2 max is greater than 35. Given the prevalence of physical inactivity in society, the attributable risk associated with this condition makes it as or almost as important as hypertension, which is considered the single most important modifiable risk factor for stroke. Although no randomized trials have assessed encouraging stroke patients to pursue regular physical activity following stroke, it seems reasonable to recommend this for all patients who are able to participate in a program. A written prescription for exercise yields a better result than verbal counseling alone; one more patients out of 11 will perform regular physical activity when employing the former strategy (213). A target of 30 minutes daily is recommended (27). For those who do not have time to do 30 minutes of exercise at one time, exercise may be done throughout the day with the goal of achieving a total of 30 minutes. Patients should be encouraged to pursue routines that they find enjoyable such as walking, running, bicycling, swimming, and dancing. For those who have not exercised for many years, starting with a short duration and intensity of exercise with gradual increase in both is recommended. Patients should be reminded that it may take weeks to months to achieve significant changes to avoid early discouragement.

Patients should be encouraged to follow balanced diets rich in fruits and vegetables and avoid excessive salt and fat. Though no diets have specifically been tested in stroke patients, randomized trials of diets rich in alpha-linoleic acid (also known as the Mediterranean diet) in patients following acute myocardial infarction show a statistically and clinically significant reduction in recurrent myocardial infarction and death (75; 205). Physicians may consider prescribing this diet. The DASH diet, similar to the Mediterranean diet, combined with aggressive sodium restriction, is associated with an 11 point reduction in systolic blood pressure in hypertensive patients and a 7-point reduction in nonhypertensives (189). Because hypertension is a significant risk factor for stroke recurrence, patients should be encouraged to consider this diet.

Smoking cessation should be strongly encouraged. The average smoker requires seven attempts at stopping before achieving success. Varenicline has been linked to an increased risk of serious adverse cardiovascular events (206).

Pharmacological options for secondary stroke prevention include antiplatelet therapy, anticoagulant therapy, antihypertensives, and medications to treat lipid abnormalities.

Antiplatelet choices in North America include aspirin, clopidogrel, extended release dipyridamole combined with aspirin, and ticlopidine. In CSPS-1, cilostazol was shown efficacious for secondary stroke prevention in comparison to placebo (97). The CSPS-2 study found an annual recurrent stroke rate (cerebral infarction, cerebral hemorrhage, or subarachnoid hemorrhage) of 2.76% with cilostazol versus 3.71% with aspirin (p=0.0357) (204). Systemic or intracerebral hemorrhage was less common with cilostazol (0.77% versus 1.78%, p < 0.01), but headache, diarrhea, palpitations, dizziness, and tachycardia were more common. The medication is approved for secondary stroke prevention in some countries for this purpose but has not been specifically approved for this indication by the Food and Drug Administration. Aspirin results in approximately an 18% reduction in risk of ischemic stroke when used for secondary prevention (19). For approximately every 29 ischemic stroke patients treated with aspirin, one less will have a stroke compared with no antiplatelet therapy over a 2-year period (79). Higher doses (eg, 1300 mg per day) do not provide more protection than lower doses (eg, 50 mg per day), but complications such as bleeding ulcers do increase (11). Clopidogrel 75 mg daily, in comparison with aspirin, prevents one more stroke, death, or myocardial infarction for every 200 patients treated over approximately two years (54). The practice of adding aspirin to clopidogrel for long-term stroke prevention does not reduce the risk of ischemic stroke compared with clopidogrel alone, results in a greater risk of life threatening bleeding, and is, therefore, not recommended (78). Likewise, adding clopidogrel to aspirin long-term does not provide additional protection against cardiovascular disease and is harmful for certain subgroups of patients (41). Short-term combination of aspirin with clopidogrel after transient ischemic attack or minor stroke may be of benefit.

In the CHANCE trial, patients were randomly assigned to aspirin plus placebo or aspirin plus clopidogrel within 24 hours of a transient ischemic attack or minor stroke (223). After 30 days, patients continued antiplatelet monotherapy until 90 days. At 90 days, the rate of stroke was 11.7% in the aspirin and clopidogrel group versus 8.2% in the aspirin and placebo group, a result that was statistically significant. Important caveats are: (1) that the rate of stroke at 90 days was much higher in either group as compared with recent reports using rapid evaluation and treatment (less than 3.2% at 90 days), and (2) the study was conducted exclusively in China, where genetic polymorphisms of increased responsiveness to clopidogrel are more prevalent.

The POINT trial, which randomized a similar population in chance to aspirin monotherapy versus dual antiplatelet therapy within 12 hours of transient ischemic attack or minor stroke and similarly found a reduction in recurrent stroke at 90 days (5.0% versus 6.5%, respectively) (112). An analysis of both trials combined found that the risk was primarily reduced within the first 21 days and the risk of major hemorrhage was increased with dual antiplatelet therapy after that. Therefore, a duration of no more than 21 days with dual antiplatelet therapy is recommended after minor stroke or transient ischemic attack. For those with bleeding ulcers, clopidogrel is not a safer option than aspirin (58). Adding omeprazole to aspirin results in significantly fewer bleeding complications but reduces the efficacy of clopidogrel and is not recommended if clopidogrel is used.

THALES found a similar reduction in ischemic stroke at 90 days with dual antiplatelet therapy that included aspirin and ticagrelor versus aspirin monotherapy (111). The reduction was similar to that seen in the POINT trial (5.5% versus 6.6%). A rare side effect associated with clopidogrel is thrombotic thrombocytopenic purpura, which occurs in approximately 11 per million treated compared with a baseline risk of four per million (39). Such patients respond to plasma exchange. Extended-release dipyridamole 200 mg twice daily combined with aspirin, in comparison with aspirin alone, prevents about one more stroke for every 100 patients treated per year (79; 83).

In randomized studies, the dose of aspirin used in combination with extended-release dipyridamole was typically less than 100 mg. Important side effects of combination therapy include gastrointestinal upset and headache. Both side effects are significantly increased by the dipyridamole portion of the combination. Tolerance to the headache usually occurs by the fourth day of therapy but not in all patients requiring discontinuation of therapy. Addition of acetaminophen to the combination does not prevent the headache (137). Aspirin at 50 mg daily has not been proven to be cardioprotective. Despite the superiority of aspirin combined with extended-release dipyridamole over aspirin alone and the apparent equivalence of clopidogrel to aspirin for secondary stroke prevention, a large randomized trial failed to show a difference between extended-release dipyridamole combined with aspirin compared to clopidogrel (186). Ticlopidine 250 mg twice daily, in comparison with aspirin, prevents one more stroke for every 33 patients treated over 3 years (104). Important side effects include neutropenia and thrombotic thrombocytopenic purpura. A complete blood count must be drawn every two weeks for the first 3 months of therapy to monitor for neutropenia. Monthly costs of therapy in 2010 for aspirin, cilostazol, clopidogrel, extended release dipyridamole, and ticlopidine are .33, 9, 0, and 6 respectively.

Until 2010, the only FDA-approved anticoagulant for atrial fibrillation was warfarin, a vitamin K antagonist. Ximelagatran, a direct thrombin inhibitor, was shown to be equivalent to warfarin with fewer bleeding complications in two randomized trials (163; 09). However, FDA approval was denied in September 2004 primarily because of three cases of fatal hepatotoxicity and an increase in cardiac events (28). Dabigatran, another direct thrombin inhibitor, has shown similar efficacy with less bleeding risk as warfarin at 110 mg twice daily and greater efficacy with similar bleeding risk at 150 mg twice daily (169). The medication was approved for use by the Food and Drug Administration at a dose of 150 mg twice daily. There is an increased risk of gastrointestinal hemorrhage, and caution is advised in patients over 80 years of age and in those with renal impairment. Rivaroxaban, an Xa inhibitor, has also been approved for the treatment of atrial fibrillation (179). Apixiban, another Xa inhibitor, has also been approved (68). The combination of aspirin and clopidogrel is inferior to warfarin for stroke prevention in atrial fibrillation with one or more risk factors (03). For patients with atrial fibrillation who are unable to receive warfarin, aspirin and clopidogrel provides a modest benefit in stroke reduction (1% per year) over aspirin alone (04).

Warfarin inhibits the vitamin K-dependent clotting factors II, VII, IX, and X. The main proven indication of warfarin for stroke prevention is atrial fibrillation based on numerous randomized studies (200). A target INR of 2.0-3.0 is recommended. Target INRs above 3 are associated with a significantly increased risk of intracranial hemorrhage and are generally not recommended (23). Patients with mechanical mitral valves require anticoagulation with a higher target INR of 2.5 to 3.5.

Uncertain indications for warfarin include arterial dissection, cardiac mural thrombus, antiphospholipid antibodies, and venous sinus thrombosis. The CADISS trial, which enrolled 250 patients with cervical arterial dissection, found no difference in the risk of stroke at 3 months in patients assigned antiplatelet versus anticoagulants (143). The rate of stroke or death was 2% versus 1% respectively, which was not statistically significant. In the TREAT-CAD study, which enrolled 194 patients with cervical arterial dissection, aspirin was not shown to be noninferior to anticoagulation in patients with cervical arterial dissection (82). The primary endpoint, which included a combination of clinical and MRI events at 7 and 90 days, occurred in 23% versus 15%, respectively. Eight percent of aspirin and 0% of warfarin-treated patients had ischemic strokes. At this time, treatment with antiplatelet therapy or anticoagulation for cervical arterial dissection is uncertain. Treatment with dual antiplatelet therapy was not tested in these trials. Though it is common practice to anticoagulate patients with mural thrombi, no randomized trials have been conducted to evaluate the superiority of warfarin over antiplatelet therapy.

Patent foramen ovale (PFO) is common in the general population. In one study, prevalence of PFO in the general population may be as high as 25% (152). An early study of patent foramen ovale and stroke found that patients with an age range of 15 to 55 who had a PFO and stroke were no more likely to have a recurrent stroke than those without a PFO unless there was an associated atrial septal aneurysm (145). The 4-year recurrence rates were less than 5% for those with or without an isolated PFO and 15% for those with a PFO and associated atrial septal aneurysm. Subsequently, three trials in 2012 to 2013 did not show a statistically significant benefit of PFO closure in intention-to-treat populations with stroke. The CLOSURE-I, RESPECT, and PC trials showed no benefit of PFO closure in addition to the best medical therapy over a period of follow-up that ranged from 2 to 4 years (90; 55; 151). In 2017, three trials were published simultaneously showing benefit of PFO closure in specific stroke patients: CLOSE (147), REDUCE (207), and RESPECT (196). One of the trials represented a longer-term follow-up of patients (RESPECT) (196). The other two trials restricted entry of patients to those who had cryptogenic cortical strokes (CLOSE, REDUCE). The rate of benefit varied between trials depending on how specific inclusion criteria were, and ranged from an absolute benefit of approximately 0.5% per year over 6 years to 1% per year over 5 years). Larger shunts or those associated with atrial septal aneurysm appeared to benefit most consistently from closure. All patients were aged less than 60 years. In general, PFO closure was associated with a higher risk of atrial fibrillation, though many of these events were transient. The DEFENSE-PFO trial further confirmed benefit of PFO closure and included only patients with atrial septal aneurysm, hypermobility of the septum (> 10 mm), and PFO size greater than 2 mm (135). Absolute risk reduction was 5% per year over two years in that trial.

Patients with stroke and serum positivity for antiphospholipid antibodies are no more likely to suffer from recurrent stroke than those without such antibodies based on a substudy of a randomized controlled trial (136). Further, warfarin confers no advantage over aspirin for prevention of stroke recurrence. Patients with stroke and evidence of a true antiphospholipid antibody syndrome (ie, evidence of premature miscarriages, prior venous thromboses, livedo reticularis, etc.) do benefit from anticoagulation and be treated at an INR of 2-3 (71). Small randomized trials have shown benefit of early anticoagulation with unfractionated heparin in patients with venous sinus thrombosis, regardless of the presence of hemorrhagic transformation (210). Frequently, such patients are converted to warfarin for 3 to 6 months of therapy, but no randomized trials have been conducted to evaluate the benefit of such a treatment scheme.

Conditions in which warfarin provides no additional benefit and increases bleeding risk include nonatrial fibrillation stroke (eg, lacunar stroke), patients with stroke while on antiplatelet therapy, and those with intracranial atherosclerosis (156; 62). For the latter condition, warfarin therapy is also associated with a higher rate of myocardial infarction (62).

Randomized trials with subgroups of stroke patients or those with stroke patients exclusively have shown that reduction in blood pressure is associated with a reduction in the risk of ischemic and hemorrhagic stroke (237; 26; 72). One study suggests that angiotensin-converting enzyme inhibitors confer special cardioprotective and cerebroprotective properties (237), whereas another failed to show such a benefit (173). In head to head trials of antihypertensive agents in patients at risk for stroke, angiotensin receptor antagonists prevent more events than beta blockers (72), but angiotensin-converting enzyme inhibitors do not confer an advantage over thiazide diuretics or calcium channel blockers (26). From a practice standpoint, these data may be interpreted to suggest that angiotensin converting enzyme inhibitors, angiotensin receptor blockers, calcium channel blockers, and thiazide diuretics are all reasonable first-line agents for the management of blood pressure in patients with stroke. Data to this point suggest that beta blockers may not confer the same cardiovascular and cerebrovascular benefit in spite of similar reductions in blood pressure. Aside from benefit, other factors to consider include side effects and cost. Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers tend to be the best tolerated agents but are generally the most expensive. For many patients, the point may be moot because the average patient with hypertension requires four agents for control (91). Most patients deemed “resistant” actually have suboptimal medication regimens. The recommended target blood pressure after ischemic stroke is less than 130 mmHg systolic. In the SPS3 trial, patients with lacunar stroke were randomized to a blood pressure target of less than 130 mmHg systolic versus 130-149 mmHg systolic (209). After a mean follow-up of 3.7 years, the rate of stroke was 2.25% per year and 2.77% per year, respectively, which was not statistically significant. The mean number of blood pressure medications required was 1.8 in the former group and 2.4 in the latter, a statistically significant difference. The annual rate of stroke in both groups was less than half of what was projected by the investigators.

Although elevations of total and LDL cholesterol are strong risk factors for cardiovascular disease, their impact on stroke recurrence is considerably less. Nevertheless, treatment of lipid abnormalities is part of a comprehensive stroke treatment strategy. By far, the most commonly used agents are the HMG-CoA reductase inhibitors, more commonly referred to as statins. These agents have demonstrated significant reductions in stroke recurrence in patients with cardiovascular and cerebrovascular disease. In the British Heart Protection study, the reduction in stroke recurrence was 25% over 5 years with benefit apparent after the first year (105). Patients (including those with diabetes mellitus, coronary disease, occlusive disease of noncoronary arteries, or treated hypertension) were treated with total serum cholesterols as low as 140 mg/dL. All patients showed benefit regardless of baseline serum cholesterol. Side effects of common concern, including liver and muscle enzyme disturbance, were trivially elevated in those receiving statins compared with those treated with placebo. Overall, these events occurred in less than 2% of patients. No increase in the risk of cancer or hemorrhagic stroke was seen; the latter observation was in contrast to a prior observational study that suggested an increase in the risk of hemorrhagic stroke in patients with low total serum cholesterol (119). In the SPARCL study, patients with either stroke or TIA within the preceding 1 to 6 months with an LDL cholesterol of 100 to 190 mg/dL and no known coronary heart disease were randomly assigned to atorvastatin 80 mg daily or placebo (12). The 5-year absolute risk reduction in stroke was 2.2% in favor of atorvastatin. Hemorrhagic stroke was more frequent in the atorvastatin group. The 5-year absolute risk reduction in major cardiovascular events was 3.5% in favor of atorvastatin. Serious adverse events, including rhabdomyolysis were similar between groups. In spite of their designated mechanism of action, namely reduction of total and LDL cholesterol, statins are also potent reducers of C-reactive protein. In cardiovascular disease, reductions of C-reactive protein correlate with reductions in recurrent cardiovascular events independent of effects on LDL cholesterol. The greatest reductions in recurrent events are seen when LDL cholesterol is reduced to less than 70 mg/dL and C-reactive protein is reduced to less than 2.0 mg/dL (177). The highest risk is associated with both levels being elevated above the aforementioned thresholds and intermediate risk is observed when one of the markers is above the threshold. The effects of reducing C-reactive protein in patients with stroke are unknown, and routine testing for C-reactive protein in stroke patients is currently not a standard.

Targeting an LDL less than 70 mg/dL after acute ischemic stroke results in fewer cerebrovascular and cardiovascular events than a target of 90 to 110 mg/dL (14). Trials using niacin (AIM-HIGH) (159), fenofibrate (ACCORD lipid study) (02), and torcetrapib (ILLUMINATE) (34) failed to show clinical benefit despite raising HDL and/or lowering triglycerides.

Revascularization procedures include arterial endarterectomy, arterial angioplasty, arterial angioplasty and stenting, and cerebrovascular bypass.

Carotid endarterectomy is the oldest and most tested revascularization procedure. Randomized trials versus placebo in symptomatic and asymptomatic patients have been conducted with reproducible results. The North American Symptomatic Carotid Endarterectomy Trial and the European Carotid Surgery Trial evaluated patients with symptomatic stenosis and found that those with greater than 70% stenosis had the clearest benefit from surgery (161; 18; 20; 24b). In the North American Symptomatic Carotid Endarterectomy Trial, roughly one out of every six patients with greater than 70% stenosis who underwent endarterectomy was spared a stroke at two years compared to best medical therapy (defined as aspirin 1300 mg daily at the time the study was performed). Each decile of increasing stenosis confers increasing benefit such that those with 90% to 99% stenosis have the greatest benefit. The greatest yield in terms of prevention occurs when endarterectomy is performed with two weeks of the neurologic event (182). Octogenarians benefited even more than those in younger age groups (07). For those with less with 50% to 69% stenosis, one out of 15 patients had to undergo endarterectomy to prevent one stroke at five years (33). Subgroup analysis found that males with a hemispheric transient ischemic attack without diabetes have the greatest benefit. Those with less than 50% stenosis have no benefit from surgery (33; 24). Surgery for asymptomatic stenosis greater than 60% confers considerably less benefit with approximately 20 surgeries required to prevent one stroke over five years in two separate randomized trials (20; 103). In contrast to surgery for symptomatic stenosis, women do not benefit from asymptomatic endarterectomy, and increasing stenosis does not confer increasing benefit. Surgical and angiographic risk must be less than 2% to avoid negating the small benefit of surgery; however, most institutions do not know their own complication rates (94). An important consideration when evaluating management of the patient with carotid stenosis is that best medical therapy in these trials consisted primarily of antiplatelet treatment and that most were conducted in an era before statin agents and newer antihypertensives were widely used.

Endarterectomy of vertebral and basilar arteries has been performed but is technically more challenging because of the arteries’ location and critical branches that emanate from them. No randomized trials have evaluated endarterectomy’s utility relative to medical therapy. In the rare patient with subclavian steal syndrome who presents with brainstem symptoms with ipsilateral arm activity, surgical revascularization often, but not always, leads to rapid improvement in symptoms (43).

Angioplasty and stenting of vessels has undergone less testing and long-term evaluation compared with surgery. CAVATAS tested carotid angioplasty alone versus carotid endarterectomy and found that outcomes at three years were similar but that periprocedural complications were lower and restenosis rates at one year were higher (25). SAPPHIRE combined angioplasty with stenting and distal emboli protection device; it found that in so-called high-risk patients, stenting is associated with less periprocedural complication rates and was noninferior at one year with respect to the occurrence of stroke, death, or myocardial infarction (234). The majority of patients in this study had asymptomatic stenosis. This study did not include a best medical therapy treatment arm; therefore, it is difficult to know what, if any, advantage there is over a noninterventional approach. The EVA-3S (146) and SPACE (208) trials both found no advantage of stenting over endarterectomy with respect to periprocedural complication rates. In the EVA-3S study, the rates of death and stroke were lower with endarterectomy at 1 and 6 months. Supporters of stenting criticized the study for having inexperienced operators performing stenting procedures. The CREST trial found equivalence of stroke, death, and myocardial infarction over a 4-year period when comparing carotid endarterectomy and stenting in patients with symptomatic and asymptomatic disease (49). More strokes occurred in patients undergoing stenting and more myocardial infarctions occurred in patients undergoing endarterectomy. Younger patients (less than 70 years of age) benefited more from stenting, whereas older patients benefited more from endarterectomy. The International Carotid Stenting Study trial found that the rate of stroke, death, and myocardial infarction at 120 days was 5.2% in endarterectomy patients and 8.5% in stenting patients (108). Long-term follow-up up to three years will continue in this study. Because of the observed reduction in stroke rates in the last 10 years in patients on updated best medical therapy, the ongoing CREST 2 study is comparing best medical therapy alone with carotid stenting or carotid endartectomy and best medical therapy.

Initial reports suggested that the periprocedural risks for intracranial stenting may be higher than for carotid revascularization (101) and lower with extracranial posterior circulation stenting (141; 95). At least one report has suggested that stent without angioplasty can be performed with gradual dilation of the vessel because of radial forces of the device (139). However, the SAMPRISS study was stopped in 2011 because of an increased risk of stroke and death within 30 days of stenting (61). In the stenting group, 14.7% of patients experienced a stroke or died within 30 days compared to 5.8% treated with medical therapy alone. The percentages for stenting were higher and percentages for medical therapy were lower than projected, suggesting publication bias in previous reports. Patients treated with medical therapy alone had better outcomes out to almost three years (77). Similar results were observed in the VISSIT trial (238).

Cerebrovascular bypass was tested in the 1980s as a means of providing collateral flow around an occluded internal carotid artery. The EC-IC bypass study found no difference in recurrence rates between medically and surgically managed patients (17). The Carotid Occlusion Surgery Study randomized patients with impaired oxygen extraction fraction (OEF) on PET to best medical therapy versus bypass treatment and was halted due to futility (171). Two-year ipsilateral stroke rates were 21% in the surgical group and 23% in the nonsurgical group (no statistical difference). Surgical graft patency was 96% at two years. Ipsilateral/contralateral OEF ratios improved from 1.258 to 1.109 in surgery patients. The stroke rate in the nonsurgical group was approximately half of what was projected. Other bypass procedures include encephaloarteriosynangiosis and encephalomyosynangiosis for patients with Moyamoya syndrome who have an abnormal circle of Willis with poor collateral flow (218). A graft is placed on the surface of the brain in the hopes of augmenting blood supply. There are no randomized trials evaluating its efficacy.

Special considerations

Pregnancy

The Baltimore-Washington Cooperative Young Stroke Study found that the risk of ischemic stroke or intracerebral hemorrhage during pregnancy and the first six weeks postpartum was 2.4 times greater than for nonpregnant women of similar age and race (126). The risk of ischemic stroke during pregnancy was not increased during pregnancy but was 8.7 times greater during the six weeks postpartum. Intracerebral hemorrhage was 2.5 times greater during pregnancy and was 28.3 times greater in the six weeks postpartum. The excess risk of stroke (all types except subarachnoid hemorrhage) attributable to the combined pregnant and postpregnant period was 8.1 per 100,000 pregnancies. The authors concluded that the extremely high relative risk of stroke during the postpartum period suggests a causal role for the large decrease in blood volume or the rapid changes in hormonal status that follow a live birth or stillbirth. Preeclampsia and eclampsia do not fully explain the much stronger associations with stroke found for the postpartum state than for pregnancy itself. Another study of acute care hospitals in California found that the risk of stroke was increased up to 12 weeks postpartum (120).

Anesthesia

Because cerebral autoregulation or cerebrovascular reactivity may be lost in ischemic or oligemic brain tissue, it may be sensitive to sudden changes in systemic arterial pressure. Therefore, in patients with compromised cerebral blood flow who undergo a necessary procedure, the anesthesiologist and operator should be advised to avoid excessive reductions in blood pressure. Unfortunately, precise parameters for the regulation of blood pressure based on scientific evidence are not available.

Guidelines regarding the use of warfarin prior to a planned surgery as can be found in the Seventh American College of Chest Physicians Consensus Conference on Antithrombotic Therapy (30). The recommendation is to stop warfarin therapy four days prior to surgery whether the patient is at low, medium, or high risk for thromboembolism. Coverage with unfractionated heparin or low molecular weight heparin is then individualized depending on risk stratification. Warfarin is started postoperatively in all patients. In patients undergoing dental procedures with a need to control local bleeding, tranexamic acid mouthwash or epsilon amino caproic acid mouthwash without interrupting, anticoagulant therapy is suggested by the consensus authors.

Antiplatelet agents are frequently used periprocedurally, and discontinuation of these medications may not be necessary. In fact, rethrombosis may be reduced when combinations of antiplatelets are used during stenting procedures. One case-control study suggested an increased risk of stroke over a 4-week period when aspirin was discontinued (148). The Acetylsalicylic Acid and Carotid Endarterectomy trial found that stroke, death, and myocardial infarction at three months occurred in 2% fewer patients undergoing endarterectomy when given 325 mg of aspirin or less versus 650 mg or more (215).