Clinical manifestations

Presentation and course

The diagnosis of transient ischemic attacks is subjective, as only 10% of spells are actually witnessed by the diagnosing physician (20; 28). The kappa value for interobserver agreement on the diagnosis of transient ischemic attack is only 0.65 (28).

A transient ischemic attack is defined as an acute focal neurologic deficit that is localized to one vascular territory, lasting less than 24 hours, with abrupt onset and abrupt or gradual resolution. Because the symptoms may be frightening, the history may be unreliable. Fully 40% of transient ischemic attacks last more than one hour, and 90% last less than 6 hours, despite the common description that they usually last only minutes in duration. The mean duration is 3 to 6 hours if patients are examined during a spell. The severity of symptoms varies from very mild to extreme, and the frequency may vary from a solitary attack to many per day over a period of years (20).

Symptoms of left carotid transient ischemic attack include transient dysarthria, right limb weakness, right face weakness, left eye or right field visual impairment, dysphasia, and right body or face numbness or paresthesia. Right carotid transient ischemic attacks produce similar symptoms on the left side of the body, but dysphasia is rare (20).

Transient ischemic attacks of the eye, also known as amaurosis fugax or transient monocular blindness, are described by patients as a unilateral fog, blur, cloud, mist, shade, or curtain (52). A shade or curtain is no more predictive of carotid artery disease than are the other symptoms, any of which usually lasts less than 10 to 15 minutes. About one fourth of all patients with transient ischemic attacks complain of headache during the spell.

Atypical carotid transient ischemic attacks include transient monocular blindness occurring only after exposure to bright light, limb shaking, orthostatic transient ischemic attacks, transient anosognosia, and transient loss of pitch perception (20).

One third of patients who report "transient ischemic attacks" and are then examined during an attack are believed by their physicians to have not had a transient ischemic attack. Patients who complain of a single symptom often exhibit other signs on examination. Language and motor deficits are common, but sensory deficits during a spell are rare. Visual field deficits, when detected, are typically unnoticed by the patient (20). During a spell of transient monocular blindness, retinal vessels typically appear normal, but this does not mean that ischemia is not present. The careful examiner may see white or grayish material, presumably platelet complexes, in retinal vessels, but it is rare to see a patient during an attack. A careful ophthalmologic examination can be crucial to the diagnosis of transient monocular blindness: narrow anterior chamber angle suggests recurrent angle-closure as the cause of symptoms, rather than ischemia; conjunctivitis, uveitis, and scleritis suggest vasculitis; papilledema suggests intracranial hypertension; and one swollen disc suggests ischemia, inflammation, or infiltration. Cotton wool and Roth spots suggest arteritis or blood dyscrasia, whereas a pale swollen disc suggests inflammation or lipohyalinotic ischemia of the optic disc.

A carotid bifurcation bruit is heard in about two thirds of transient ischemic attack patients with severe ipsilateral carotid stenosis. These patients may have enhanced facial artery pulses due to collateral flow downstream from severe internal carotid artery bifurcation disease (10).

Prognosis and complications

The outcome of carotid transient ischemic attack is believed to be more likely benign in young patients with frequent episodes, a prolonged course, and no severe cardioembolic or carotid disease (54). Yet, many patients with transient ischemic attack have an impaired quality of life and occupational status, with fatigue or memory problems even if a stroke did not occur (20). Observation suggests that the risk of stroke is highest soon after the first event or with multiple transient ischemic attacks.

The stroke risk varies according to the etiology of the initial event. In a 5-year follow-up of 3847 patients from 21 countries after an initial transient ischemic attack diagnosis, the rate of stroke was 9.5%, with about half of the events occurring in the first year and half during years 2 to 5 (03). During the 5-year period, rates of death from any cause, death from vascular causes, and major bleeding were 10.6%, 2.7%, and 1.5%, respectively. Individuals with index transient ischemic attack due to large artery atherosclerosis or cerebral embolism were at higher risk. Patients with an elevated ABCD2 score of 4 or higher were also at higher stroke risk. Retinal infarction after transient monocular blindness, presumably due to retinal branch or central retinal artery occlusion, is uncommon (48).

In the North American Symptomatic Carotid Endarterectomy Trial, in which patients with carotid stenosis and transient ischemic attack or minor stroke were studied, the stroke risk in patients with transient visual loss was approximately half that of patients with hemispheric transient ischemic attack (hazard ratio 0.53) (06). There was a 14.3% risk reduction with endarterectomy in patients with at least three of the following risk factors: (1) age of 75 years or more, (2) male sex, (3) history of hemispheric transient ischemic stroke or stroke, (4) history of intermittent claudication, (5) 80% to 94% angiographic stenosis, and (6) absence of collateral circulation.

In a study that did not preselect patients with carotid stenosis, it was found that the presence of age greater than 65 years, diabetes, hypertension, coronary disease, and stroke rather than transient ischemic attack as a first symptom dramatically increased the stroke and death rate at two years from 10% to 59% (40).

A study of patients seen in the emergency room who were given a diagnosis of transient ischemic attack found that the short-term outlook was slightly worse than earlier reports. In this analysis of 1707 patients, there was a 10.5% incidence of stroke at 90 days, with half of the strokes occurring in the first two days (35). Age greater than 60 years, diabetes, symptom duration of greater than 10 minutes, weakness, and speech impairment were associated with a higher risk. A Canadian study that followed 2285 patients diagnosed in the emergency room with a transient ischemic attack in the province of Alberta identified a similar 90-day rate of stroke of 9.5% (30). In this study, the 1-year stroke risk was 14.5%, and hypertension, diabetes, and older age were predictors of increased stroke risk at one year. In more modern studies, such as the TIA.org Registry, the 1-year stroke risk is now lower due to widespread use of more intensive medical therapy.

The short-term risk of stroke can be reduced with rapid evaluation and institution of treatment with antiplatelet agents, statins, and antihypertensives (61). In the Oxford Vascular Study, patients referred to a rapid evaluation stroke and TIA clinic had a 2.1% risk of stroke at 90 days, compared to a 10.3% risk with the “standard” evaluation system (p=0.0001). For patients with symptomatic carotid stenosis, modern medical therapy appears to be associated with a reduction in the stroke rate. In a study of 397 patients with symptomatic carotid stenosis, the 30-day risk of stroke was 7.5% (66).

Even in patients with a non-recent transient ischemic attack, there is an increased risk of stroke and cardiac events. In an analysis of patients from the Oxford community stroke project who had a transient ischemic attack at a median of 3.8 years previously, there was a 18.8% risk of stroke over 10 years and a 27.8% risk of myocardial infarction or cardiac death over 10 years (15).

A simple scoring system can help clinicians identify patients at increased risk for stroke following transient ischemic attack. The ABCD acronym signifies A (age of greater than 60 years), B (blood pressure greater than 140/90), C (clinical features: unilateral weakness, speech disturbance), and D (duration of symptoms in minutes). In the Oxford Vascular Study, the seven-day risk of stroke was 0.4% in patients with a score of less than 5, 12.1% in patients with a score of 5, and 31.4% in patients with a score of 6 (62). The ABCD2 has been suggested as a refinement of the original score, with diabetes as a factor (36). In a statement from the American Stroke Association, it is recommended that patients be admitted to the hospital if they experience a transient ischemic attack in the preceding 72 hours and if the ABCD2 score is 3 or higher or if the score is 0 to 2 and there is uncertainty about whether diagnostic testing can be completed within two days as an outpatient (19).

One third to one half of patients with transient ischemic attacks will have a small ischemic lesion on MRI using DWI (19). A multi-center study of 4574 patients with early imaging after TIA found that DWI lesions can help identify a high-risk group (27). For patients with a DWI lesion, the 7-day rate of stroke was 7.1% versus 0.4% in patients without a DWI lesion (p< 0.0001). Some have advocated that the ABCD2 score plus early MRI can help with identifying low-risk patients who can be discharged home with outpatient investigations (27).

Clinical vignette

A 66-year-old man presented to the emergency room with the acute onset of expressive speech difficulty, which lasted for 20 minutes and then resolved. The patient and his family did not notice weakness, loss of consciousness, or seizure activity. The past medical history was notable for diabetes mellitus and hypertension for 10 years each. Medical history was negative for coronary artery disease or peripheral vascular disease.

On examination, the patient was in sinus rhythm with a blood pressure of 150/80 mm Hg. His exam revealed intact speech, no facial weakness, no limb weakness, and flexor plantar responses. There was a left carotid bruit. ECG was normal.

Subsequent testing revealed 80% to 99% stenosis of the left internal carotid artery on duplex. This was confirmed as severe stenosis on MRA. MRI of the brain showed two small, deep corona radiata infarcts on the left side. Successful left carotid endarterectomy was performed two days after presentation.

Biological basis

Etiology and pathogenesis

Ischemia of the internal carotid artery territory and its branches via embolism, thrombosis, or hemodynamic insufficiency is the underlying basis for carotid transient ischemic attacks. Rarely, vasospasm may be responsible for the symptoms of transient monocular blindness (72).

The underlying pathology for the ischemia of transient ischemic attacks may be atherothrombotic, embolic, inflammatory, or hematologic. The primary process could reside in the small or large arteries of the brain (within the carotid territory), the heart, or a coagulation disorder in the blood (50; 52). The specific transient ischemic attack symptoms rarely distinguish among these possibilities, which must be identified or excluded through laboratory testing.

Atherosclerosis of the internal carotid artery can appear as early as 25 to 40 years of age. It is usually patchy or focal rather than diffuse, and most commonly occurs at the carotid bifurcation or 2 cm from the internal carotid artery origin. Intracranial lesions are more frequent in African Americans and Asians; the cavernous segment of the internal carotid artery is the most often affected downstream from the bifurcation.

Atherosclerosis may progress over months or remain static for years, even if severe. About one third of atherosclerotic lesions progress and one third of these become symptomatic (52). After endarterectomy, redevelopment of greater than 50% stenosis appears in 14 months in 19% of patients, whereas 9% develop greater than 80% stenoses. About one half of these patients become symptomatic a second time. In the first two years after surgery, the mechanism of restenosis is believed to be myointimal hyperplasia, whereas atherosclerosis is believed to be more prevalent in restenoses that develop after this period (18).

Atherosclerotic lesions may produce symptoms by a hemodynamic mechanism, when the residual lumen is restricted to less than 2 mm (08; 46). Occlusion usually results from thrombosis superimposed on atherosclerosis. A hemodynamic mechanism may logically explain brief, stereotyped, especially orthostatic transient ischemic attacks. The deficit vanishes as blood flow is restored. Carotid plaques from patients with symptomatic cervical lesions have been reported to have less collagen staining and greater matrix metalloproteinase activity compared to asymptomatic plaques (70).

Embolism can also occur as platelet-fibrin thrombi commonly superimpose themselves on the atheroma and may easily be washed downstream to occlude smaller arteries and produce symptoms. The presence of fresh thrombus in an endarterectomy specimen distinguished recently symptomatic patients, 80% of whom had thrombus, from patients with remote symptoms or asymptomatic patients in whom thrombus was uncommon (30% to 40%) (68). Ulceration with no thrombus and plaque hemorrhage did not distinguish these groups of patients.

Cardiac embolism leading to carotid transient ischemic attack most often arises from nonvalvular atrial fibrillation. Other causes include postmyocardial infarction wall motion disorder, aortic atherosclerosis, cardiomyopathy, mitral valve prolapse and other valve diseases, patent foramen ovale, calcified aortic disease especially during catheterization or valve procedures, endocarditis, valve prostheses, and myxoma (69).

Less common causes of transient ischemic attack include dissection, fibromuscular dysplasia, kinking, extracranial aneurysm, local tumor invasion of the carotid, radiation therapy accelerated atherosclerosis, and postendarterectomy restenosis. Also, primary thrombi may develop in hypercoagulable blood (29).

Finally, vasospasm has been reported to occur in the eye and cause transient monocular blindness. The vasospasm may be primary or secondary to another disease, such as temporal arteritis or cocaine abuse (45).

Epidemiology

Epidemiological data related to carotid transient ischemic attacks are not always separated in the literature from carotid stroke, or from strokes or transient ischemic attacks in the vertebrobasilar territory. Patients with carotid transient ischemic attack can roughly be divided into groups over and under 40 years of age. The older group usually has underlying atherothrombosis, whereas younger patients often have cardioembolic, migrainous, or idiopathic spells. The epidemiology of transient ischemic attacks has been nicely summarized (23).

Prevention

In a cohort of patients with transient ischemic attacks (including vertebrobasilar spells), 68% were male, 47% smoked, 42% had hypertension, 21% had coronary disease, 15% had valvular disease, 4% had atrial fibrillation, 5% had congestive heart failure, 16% had cardiomegaly, 17% had peripheral vascular disease, 5% had diabetes, 42% had elevated cholesterol, and 18% had a history of migraine (28). A similar distribution and prevalence of risk factors have been reported in a cohort of patients with carotid hemisphere transient ischemic attacks (excluding patients with transient monocular blindness) (07).

Differential diagnosis

Confusing conditions

In one study in which thousands of transient ischemic attacks were reported by patients through questionnaire, subsequent evaluation of the patients by a physician rejected most of them as not being transient ischemic attacks, but as peripheral neuropathy, psychoneuroses, Bell palsy, migraine, cerebral infarction with an examination that failed to detect residual signs, seizures, conversion disorders, and structural lesions of the orbit. Migraine and seizures are important alternatives to the diagnosis of transient ischemic attack.

In migraine, one half of the patients with neurologic accompaniments do not experience headache. A buildup in severity of symptoms and a migration, especially of scintillations, flashing lights, colors, and fortification spectra across the visual field strongly suggest the diagnosis of migraine. An anatomic march of symptom type, from visual symptoms to aphasia to hemiparesis, for example, and the benign recurrence of symptoms over years, with or without episodic headache also suggest migraine. Some physicians require exclusion of appropriately located atherosclerosis before offering the diagnosis of migraine.

In seizures, convulsive activity precedes or follows the paralysis. These patients usually have structural lesions causing seizures, no vascular risk factors, and a therapeutic response to anticonvulsants. As with migraine, patients with seizures have positive phenomena that may migrate and leave negative phenomena in their wake.

Structural lesions may produce transient ischemic attack-like symptoms via an unclear mechanism. Such lesions include tumors, subdural hematomas, aneurysms, amyloid angiopathy, multiple sclerosis, hypertensive encephalopathy, hypoglycemia, and hyperventilation (20).

Associated or underlying disorders

The most common associated conditions are underlying vascular risk factors such as hypertension, diabetes, hypercholesterolemia, and smoking. Common cardiac causes for transient ischemic attacks include atrial fibrillation, valvular heart disease, and decreased cardiac ejection fraction.

Diagnostic workup

Numerous noninvasive technologies capable of identifying carotid bifurcation atherosclerosis reveal a disease prevalence similar to that found with angiography. MR angiography and color duplex ultrasound frequently compete as the screening study of choice for carotid disease. These tests seek to identify severe stenosis, both to establish a mechanism for transient ischemic attack and to identify surgical candidates. The positive and negative predictive values for surgical disease (70% to 99% angiographic stenosis) for color duplex are 0.84 and 0.98, respectively, and for MR angiography 0.79 and 0.81, respectively. MR angiography often displayed a flow gap (zone of signal intensity loss due to slow flow) in patients with severe disease, giving a false impression of complete occlusion. Color duplex performed better at distinguishing severe stenosis from complete occlusion in this cohort (65). Other physicians support the use of MR angiography over duplex, citing its higher interobserver reliability, the fact that it is less operator-dependent, and the fact that the entire vessel is seen. Ultrasound is probably cheaper and more readily available than MR angiography (51). In a study of 167 patients, contrast-enhanced MRA was associated with a stenosis misclassification rate of 15% (67). There is increasing interest in CT angiography as well, and a study of 81 vessels found that CT angiography had high negative predictive value (37). In eight vessels where CT angiography indicated greater than 70% stenosis, however, only five out of eight (63%) were greater than 70% on angiography, suggesting a tendency for overestimation of stenosis. Use of PET using the FDG ligand can also identify “hot” carotids. With use of PET imaging, patients with carotid stenosis at elevated risk of stroke recurrence can be identified (39). Such information can potentially be used to guide timing of carotid revascularization.

A clear algorithm for the optimal noninvasive evaluation of patients with carotid stenosis is not currently available. Physicians tailor their approach depending on the strengths and weaknesses of individual laboratories at their particular institution.

Transcranial Doppler ultrasound can detect hemodynamic changes distal to a carotid stenosis, which help assess the severity of bifurcation stenosis, improving the selection of patients for angiography. Transcranial Doppler can also assess the severity of intracranial disease and can detect emboli distal to internal carotid artery or cardiac disease in patients with transient ischemic attack. Emboli are abnormally high-pitched unidirectional signals, less than 0.1 second in duration that appear randomly in the cardiac cycle and make an easily heard chirping sound (26).

CT or MRI of the brain is performed in patients with transient ischemic attack to rule out mass lesions. A focal hypodensity appropriately located for the symptoms of the transient ischemic attack is found in approximately 34% of patients, who are then considered to have had cerebral infarction with transient signs. Their transient ischemic attacks usually last longer than average and are more often embolic in origin (52). MRI with diffusion-weighted imaging may be especially useful in patients with transient ischemic attacks. A study of 42 patients who were imaged with diffusion-weighted imaging within three days of symptom onset found that 48% of patients had a conspicuous lesion (41). Of these, approximately one half had corresponding infarcts on follow-up MR imaging. Diffusion-weighted imaging in patients with transient ischemic attacks may also help clarify the ischemia localization and the symptom mechanism.

The fact that many transient ischemic attack patients will have an associated small infarct on imaging studies has led to a new definition of transient ischemic attack from the American Stroke Association (19). In an analysis of pooled data from 10 MRI studies that enrolled 818 patients, it was observed that an MRI lesion is present in about one third of patients with symptoms lasting less than six hours and one half of patients with symptom duration of 6 to 24 hours (19). Based on this observation, the suggested definition of transient ischemic attack is “a transient episode of neurologic dysfunction caused by focal brain, spinal cord, or retinal ischemia, without acute infarction.”

High-resolution plaque imaging with MRI is being used occasionally to identify high risk carotid stenosis. A study of patients with 50% to 99% symptomatic stenosis found that patients with intraplaque hemorrhage on MRI had an increased risk for stroke (23% annual risk), whereas the stroke risk was very low (0.6%) for patients without intraplaque hemorrhage (24). It has also been observed that men have more frequent intraplaque hemorrhage compared to women.

In patients with normal or mildly diseased carotid arteries or those at high risk of suspected embolism, cardiac testing is initiated. Cardiac monitoring is performed when an arrhythmia is suspected, such as intermittent atrial fibrillation. Studies with CTA have found a higher frequency of thick plaques (> 5 mm) ipsilateral to cryptogenic stroke, raising the possibility of a link between the nonstenotic plaque and stroke (16).

In patients where carotid and cardiac tests are unrevealing, systemic and coagulation abnormalities that can cause transient ischemic attack are sought. Atherosclerotic disease of the intracranial vessels should also be considered. The physician may perform tests for antiphospholipid antibodies and other hypercoagulable states, perform toxic drug screening, and perform serological tests for vasculitis.

Management

Patients with greater than 70% carotid stenosis and a recent ipsilateral carotid transient ischemic attack should be treated surgically if the perioperative stroke and death rate is less than 6% (53). In the North American Symptomatic Carotid Endarterectomy Trial (NASCET), carotid endarterectomy reduced the ipsilateral stroke rate in patients with transient ischemic attack or stroke and 70% to 99% stenosis from 26% to 9% after two years follow-up (53). The beneficial effect was greater for more severe stenoses. Of note, patients in the North American Symptomatic Carotid Endarterectomy Trial had become symptomatic within the last four months, whereas those in the European trial became symptomatic in the last six months. If symptoms have been present for a longer period, optimal management has not been determined. For patients with 50% to 69% stenosis in the North American Symptomatic Carotid Endarterectomy Trial, there was a modest benefit from surgery (absolute risk reduction 1.3% per year) (04), but the benefit was heterogeneous. Women and patients with retinal ischemia did not derive benefit. Below 50%, there was no benefit seen in the North American Symptomatic Carotid Endarterectomy Trial. It should be noted that the degree of stenosis measured as percent stenosis is not strictly comparable across the two clinical trials mentioned above.

A combined analysis of the North American and European symptomatic endarterectomy studies also evaluated the outcome in patients with a collapsed distal internal carotid artery, an angiographic finding referred to as “near occlusion” or a carotid string sign. In these patients external carotid filling is more brisk, and the external carotid is larger than the collapsed internal carotid. The combined analysis found no benefit of surgery in patients with carotid near occlusion, with a 1.7% trend for recurrent stroke favoring medical treatment at five years (59).

Physicians must keep in mind that the perioperative complication rate at their institution may be higher than the less than 6% rate preselected by the North American Symptomatic Carotid Endarterectomy Trial (13). This information may dampen their enthusiasm for the procedure at their institution. Higher complication rates are seen in surgeons with low annual volumes, especially less than 10 endarterectomies per year (43).

The perioperative complication rate for carotid endarterectomy may be increased by certain clinical and angiographic variables. In one analysis, two or more of the following factors doubled the risk: age greater than 75 years, diastolic blood pressure greater than 110 mm/Hg, history of angina, performance of endarterectomy in preparation for coronary artery bypass grafting, thrombus in the internal carotid artery lumen, or a downstream siphon stenosis (49). In 16% of endarterectomies, angiography or ultrasound performed at the completion of the procedure detects a correctable problem such as a kink, flap, thrombus, or spasm. Immediate correction of these difficulties lowers the perioperative complication rate (18). An analysis from a Scandinavian registry found that endarterectomy can be safely performed within 2 weeks of the patient having received intravenous thrombolysis (58). This study found the stroke/death rate to be 3.5% for patients with surgery following thrombolysis.

The American Academy of Neurology has published a previous statement on carotid endarterectomy (14). The statement includes information from the pooled analysis mentioned above for symptomatic patients. The absolute benefit at 5 years is 16% for patients with 70% to 99% stenosis and 4.6% for patients with 50% to 69% stenosis (60). In the pooled analysis, men, patients enrolled in the trial within two weeks of their last symptomatic event, and patients above age 75 years showed greater benefit. There were relatively few patients in the multi-center trials above age 80 years; however, some caution is needed in this group.

For patients at increased risk for endarterectomy, the SAPPHIRE study compared carotid artery stenting with distal protection versus endarterectomy (73). About one-third of the 334 patients were symptomatic. In the total cohort, the primary endpoint (stroke, death, or myocardial infarction within 30 days plus ipsilateral stroke between days 31 through 365) occurred in 12.2% of the stent patients and 20.1% of the endarterectomy patients (p=0.05). For the symptomatic patients, there was no difference with the primary endpoint occurring in 16.8% of the stent patients and 16.5% of the endarterectomy patients (p=0.95). There was not a medically treated control group, so it is difficult to judge if either intervention was better than medical therapy alone. The Center for Medicare Services has approved the reimbursement of carotid stenting for symptomatic patients with 70% to 99% stenosis who are at increased risk for endarterectomy.

Three European studies have compared carotid endarterectomy to carotid artery stenting (11). In a pooled analysis of these trials, within 120 days of study entry there was an increased risk of stroke or death in those patients assigned to carotid artery stenting (8.9%) compared to those who received carotid endarterectomy (5.8%, hazard ratio 1.53, p=0.0006). However, a significant interaction was noted with age. In patients 70 years of age or older, the risk with carotid artery stenting was magnified (12.0% with carotid artery stenting, 5.9% with carotid endarterectomy, HR 2.04). However, no difference was noted in the 120-day risk of stroke or death in patients younger than 70 years of age (5.8% carotid artery stenting vs. 5.7% carotid endarterectomy).

For conventional risk subjects, the Carotid Revascularization Endarterectomy vs. Stent Trial (CREST) results were reported (31; 09). The trial included both symptomatic and asymptomatic subjects. A total of 1321 symptomatic patients were enrolled with either more than 50% stenosis on angiography or more than 70% stenosis on ultrasound. Patients were required to have had an ipsilateral transient ischemic attack or nondisabling stroke within 6 months of study entry.

Patients were followed for the primary endpoint of stroke, myocardial infarction, or death within 30 days and ipsilateral stroke after 30 days. Within the 30-day periprocedure period, the primary endpoint did not differ between the carotid artery stenting and carotid endarterectomy groups (09). The rate was 6.7% in the carotid artery stenting group and 5.4% in the carotid endarterectomy group (HR 1.26, p=0.30). However, some have criticized the inclusion of myocardial infarction in the primary endpoint (17). If myocardial infarction is removed and the traditional stroke or death endpoint is used, then symptomatic patients did better with carotid endarterectomy (6.0% stroke/death with carotid artery stenting, 3.2% with carotid endarterectomy, HR 1.89, p=0.02). The CREST investigators also reported a higher event rate with carotid artery stenting compared to carotid endarterectomy in patients more than 70 years old (71). For the primary endpoint, there was 1.77 times increase in the rate of the primary endpoint for each 10-year increment in age. At the current time in North America, carotid artery stenting has not been approved for conventional risk patients. Some of the differences between the European studies and CREST and other issues pertaining to carotid revascularization are highlighted in a review (56).

A newer revascularization option is transcarotid artery revascularization (TCAR), in which direct access is achieved in the common carotid artery and flow reversal is instituted from the internal carotid artery to the femoral vein. One advantage of TCAR compared to transfemoral stenting is that TCAR avoids the aorta, which can be a source of embolic debris and periprocedural stroke. Analysis of a large national registry found a lower complication rate with TCAR compared to transferal stenting (64). A multicenter register also evaluated TCAR. The single arm ROADSTER 2 registry found a 5.0% stroke/death rate in symptomatic patients and a 1.4% rate in asymptomatic patients (38). Registry data showed that 1-year outcomes were similar between TCAR and carotid endarterectomy, but the rate of complete data availability was limited (44). There is not randomized trial information available on TCAR, and as a result, recent American Heart Association/American Stroke Association guidelines felt that the utility of TCAR in symptomatic patients is not well established (42).

Guidelines recommend early (less than 7 days) carotid revascularization for patients with carotid transient ischemic attack or nondisabling stroke (55). However, in actual practice, only a minority of patients receive early revascularization. In a study of 4138 patients, only 11% to 14% received the procedure within seven days of symptoms (57). In this same analysis, endarterectomy appears to provide better results than stenting, with a 1.3% rate of stroke/death with endarterectomy and an 8.3% risk of stroke/death with stenting.

The mechanism of stroke following carotid revascularization procedures is diverse. Analysis of 85 patients with postoperative stroke in a clinical trial found the following mechanisms: 14 carotid-embolic cases, 20 hemodynamic cases, 15 patients with occlusion of the carotid, nine hyperperfusion cases, five cardioembolic patients, and three patients with multiple causes (32).

In patients with complete occlusion of the internal carotid artery, external carotid collaterals commonly reconstitute the ophthalmic artery and intracerebral vessels. Stenosis and ulcers in the external carotid or internal carotid cul-de-sac in such patients can cause eye and brain symptoms. Two thirds of these patients have transient monocular blindness as the indication for surgery. External carotid endarterectomy can be effective for these embolic sources at the external carotid origin (63).

Patients with noncardioembolic stroke or transient ischemic attack should be treated with antiplatelet therapy, such as aspirin, clopidogrel, or low-dose aspirin and dipyridamole (42). In addition, aggressive risk factor modification is important. This would include smoking cessation and treatment of hypertension and diabetes. Anticoagulation is employed in patients with sources of recurrent cardiac embolism. Dietary modification and regular physical activity are also recommended for patients with transient ischemic attack (42).

For patients with severe carotid stenosis awaiting surgery, early use of both aspirin and clopidogrel has been associated with a reduced rate of microemboli and lower rate of recurrent stroke (05).

In the 2021 American Stroke Association (ASA) stroke prevention update, a few items are noteworthy. First, aspirin alone (50 to 325 mg), clopidogrel, or aspirin plus extended-release dipyridamole are all acceptable for initial therapy (42). Administration of a statin with intensive lipid-lowering effects is recommended in patients with an atherosclerotic stroke or transient ischemic attack, even in the absence of overt ischemic heart disease (47). The ASA document further comments that for patients with an atherosclerotic stroke or transient ischemic attack, and without known coronary heart disease, it is reasonable to target a reduction of LDL of at least 50% or an LDL of less than 70 mg/dl. For patients over the age of 70, carotid endarterectomy is preferred over carotid stenting. Finally, it is worth noting that the NASCET data are 30 years old, and new trials are needed to compare intensive medical therapy with carotid revascularization (12).

The POINT study (Platelet Oriented Inhibition in New TIA and minor ischemic stroke, NCT00991029) evaluated the use of dual antiplatelet therapy (DAPT) for 90 days in patients with high-risk transient ischemic attacks or minor strokes (34). It enrolled patients with an ABCD score of 4 or greater or NIH stroke scale score of 3 or less who could be treated within 12 hours. Subjects received either aspirin alone or aspirin plus clopidogrel (600 mg loading dose followed by 75 mg per day) for a 90-day treatment period, with the primary endpoint being stroke, myocardial infarction, or vascular death. The primary endpoint was reduced by 25% in patients who received DAPT with aspirin and clopidogrel. Further analysis stratified by duration of treatment suggests that three weeks of DAPT appears to be the optimal treatment duration.

There also appears to be benefit in using DAPT for patients presenting within 72 hours of a transient ischemic attack or minor stroke. A study from China (INSPIRES) compared single antiplatelet therapy with aspirin with DAPT combining aspirin and clopidogrel (25). Patients were enrolled with a high-risk transient ischemic attack or minor stroke with an NIH stroke scale score of 5 or less. Trial participants were also required to have more than 50% stenosis of an ipsilateral vessel. Over the study period, 6100 patients were enrolled. The stroke rate within 90 days was 7.3% with DAPT versus 9.2% with aspirin alone, a significant difference. Moderate to severe bleeding occurred in 0.9% of the DAPT group and 0.4% of the aspirin group.

Another antiplatelet option for patients with carotid transient ischemic attacks is ticagrelor. In the THALES trial, patients with minor stroke or high risk transient ischemic attack who were evaluated within 24 hours were assigned to either ticagrelor (180 mg loading dose, then 90 mg bid) plus aspirin or aspirin alone. Over 11,000 patients were enrolled and the ischemic stroke rate within 30 days was 5.0% with aspirin plus ticagrelor versus 6.3% with aspirin alone (33). Serious bleeding was higher with dual antiplatelet therapy (0.5% vs. 0.1%). In a subgroup analysis of patients with atherosclerotic stroke, the benefit of aspirin plus ticagrelor was greater (8.1% vs. 10.9%, 27% relative reduction) (01).

Outcomes

With modern medical therapy, the risk of stroke following a transient ischemic attack has decreased. A multicenter registry analyzed patients who were evaluated following a transient ischemic attack (02). The 1-year risk of stroke was 5.1%; this is lower than the 10% to 15% rate observed in earlier decades. Patients with large vessel atherosclerosis and those with multiple lesions on diffusion-weighted imaging were at increased risk for stroke.

Special considerations

Pregnancy

Pregnancy may produce a hypercoagulable state that can lead to stroke, but these strokes are not commonly preceded by transient ischemic attacks.

Anesthesia

Head and neck trauma with carotid dissection could occur during anesthesia secondary to neck manipulation, but this is rare.