brand logo

Am Fam Physician. 1999;60(8):2329-2336

A more recent article on recurrent ischemic stroke is available.

See related patient information handout on preventing ischemic strokes and treating transient ischemic attacks, written by the authors of this article.

Stroke is the third most common overall cause of death and the leading cause of adult disability in the United States. New therapeutic interventions instituted in the period immediately after a stroke have revolutionized the approach to ischemic cerebrovascular disease. Recognition of a transient ischemic attack provides an opportunity to prevent a subsequent stroke. Specific stroke prevention treatment depends on the cause of the transient ischemic attack, its cerebrovascular localization and the presence of associated coexisting medical problems. Modification of stroke risk factors is the principal therapeutic approach. Antiplatelet agents and anticoagulants have been shown to be effective in reducing the occurrence of stroke in certain populations. Several well-designed studies have recently demonstrated the effectiveness of carotid endarterectomy in preventing strokes related to extracranial carotid artery disease.

Approximately 550,000 new strokes and 150,000 stroke-related deaths occur each year in the United States.1 About 80 percent of strokes are due to ischemic cerebrovascular disease, and the rest are attributable to hemorrhagic causes such as subarachnoid or intracerebral hemorrhage.1 Even though great strides have been made in the identification and treatment of risk factors for stroke and the development of new therapeutic interventions, ischemic stroke continues to be a significant public health problem.

A completed stroke is caused by irreversible brain injury secondary to the interruption of blood flow. In contrast, a transient ischemic attack (TIA) is a temporary focal neurologic deficit caused by the brief interruption of local cerebral blood flow. The prevalence of TIAs ranges from 1.6 to 4.1 percent, depending on gender and age. Stroke occurs in one third of patients who have a TIA.2

The duration of a focal neurologic deficit that leads to cerebral infarction has arbitrarily been determined to be 24 hours or greater. Any focal neurologic deficit that resolves completely within 24 hours is considered a TIA.3 However, a TIA in the carotid territory typically lasts only seven to 10 minutes.3 The diagnosis of a TIA indicates that no irreversible neurologic injury has occurred and provides an excellent opportunity to prevent permanent damage.

Differential Diagnosis and Symptoms

The first step in the evaluation of a patient with possible TIA is to determine if the event in question actually represents a TIA. Table 1 lists conditions that can present with focal neurologic deficit and are sometimes confused with TIAs. These conditions should be ruled out before the diagnosis of TIA is made. Excluding other diagnoses reduces the possibility of inappropriately labeling a patient with the diagnosis of cerebrovascular disease and launching into a course of costly and potentially dangerous diagnostic testing.

Todd's paralysis (postepileptic paralysis)
Complicated migraine
Conversion disorder or malingering
Brain tumor
Drug overdose
Bell's palsy

The symptoms of a TIA depend on the region of the brain that is supplied by the transiently occluded cerebral artery. Precise localization may be difficult because anatomic variations, especially in the arteries that form the circle of Willis, are the rule rather than the exception. The clinical findings most frequently associated with ischemia in various arterial distributions are listed in Table 2. However, other symptoms may occur, and many other clinical stroke syndromes have been described.

Anterior cerebral artery
Weakness in contralateral leg
Sensory loss in contralateral leg, with or without weakness or numbness in proximal contralateral arm
Middle cerebral artery
Contralateral hemiparesis
Deviation of head and eyes toward side of lesion
Contralateral hemianesthesia
Contralateral hemianopia
Aphasia (if dominant hemisphere is affected)
Unawareness of stroke (if nondominant hemisphere is affected)
Lenticulostriate arteries
Pure motor hemiparesis (lacunar syndrome)
Posterior cerebral artery
Visual field disturbance
Contralateral sensory loss
Vertebrobasilar arteries
Nausea and vomiting

If a TIA is recognized, steps can be taken to prevent future ischemic stroke. All TIAs should be promptly investigated because the risk of ischemic stroke is highest soon after a TIA (i.e., 5 percent in the first month).4

Diagnostic Evaluation


The goals of diagnostic testing are to identify or exclude causes of TIAs that require specific therapy, to assess modifiable risk factors and to determine the prognosis.4 The initial tests recommended for all patients with TIA are listed in Table 3.4

Complete blood cell count with platelet count
Chemistry profile (including cholesterol and glucose levels)
Prothrombin time and activated partial thromboplastin time
Erythrocyte sedimentation rate
Syphilis serology
Cranial computed tomography (particularly with hemispheric transient ischemic attack)
Noninvasive arterial imaging (ultrasonography, magnetic resonance angiography)

The complete blood count with differential rules out profound anemia, polycythemia, leukocytosis, thrombocytopenia and thrombocytosis as hematologic causes of stroke or as factors that may influence therapy. The chemistry profile demonstrates hypoglycemia that can present with focal neurologic deficits or hyperglycemia that can worsen the outcome after stroke.5

A prothrombin time and an activated partial thromboplastin time are needed to rule out coagulopathies. The erythrocyte sedimentation rate serves as a screening test for autoimmune disorders, and syphilis serology screens for neurosyphilis.

An electrocardiogram (ECG) should be obtained in all patients with TIA or stroke. The ECG is used to detect arrhythmias (e.g., atrial fibrillation) as the cause of ischemia. Computed tomographic (CT) scanning of the head is necessary to rule out intracranial bleeding or tumors. CT evidence of old infarcts, if present, may reveal the vascular distribution of previous ischemic events.

Cerebrovascular ultrasound studies are recommended in all patients with TIA symptoms. These tests are noninvasive but have some limitations. Carotid duplex studies (e.g., Doppler plus B-mode imaging) detect extracranial carotid disease well but may miss intracranial carotid artery disease, vertebral artery disease and complete occlusion.


One or more of the tests listed in Table 44 may be performed in patients with TIA. Transthoracic and transesophageal echocardiographic examinations do not usually reveal a cause for TIA unless the patient has clinical heart disease. Nonetheless, transthoracic echocardiography is almost always employed in younger patients and in patients for whom no other cause of TIA can be found.4 This examination may also be helpful in identifying atrial thrombus in patients with atrial fibrillation. Transesophageal echocardiography may be useful in confirming suspected atrial appendage thrombus, valvular defects, atrial septal defects, mitral valve vegetation and atrial septal aneurysms.

Transthoracic echocardiography
Transesophageal echocardiography
Transcranial Doppler ultrasonography
Magnetic resonance angiography
Cerebral arteriography
Measurements of antiphospholipid antibodies and other markers for prothrombotic states
Ambulatory electrocardiographic monitoring
Cerebrospinal fluid examination

Transcranial Doppler ultrasonography can reveal intracranial stenosis of the trunks of the middle cerebral or posterior cerebral arteries. However, stenotic lesions in smaller branches may be missed.

More recently, magnetic resonance angiography has been used to detect stenosis in extracranial or intracranial cerebral arteries. Arteriography is usually reserved for special situations such as intracranial vasculitis or arterial dissection.4 This study is also performed when cerebrovascular surgery is being considered.4 Although arteriography is expensive and invasive, it is the gold standard for defining occlusive cerebrovascular disease.

Special testing for hypercoagulable states should be reserved for use in patients less than 50 years of age, patients with a history of thrombotic disease and patients in whom no other cause of TIA is found.4 Holter monitoring is recommended for use in patients who had palpitations close in time to the TIA and patients who have an enlarged left atrium.

Lumbar puncture is not routinely recommended as part of the evaluation of patients with TIA. However, this study may be warranted if central nervous system infection is suspected or the presenting symptoms suggest subarachnoid hemorrhage but the CT scan is negative.



Current theories on the pathogenesis of TIA suggest that effective measures to prevent stroke should also prevent the recurrence of TIA. The initial approach is to modify risk factors that are amenable to treatment (Table 5).631 In this section, emphasis is placed on risk factors that have a definite correlation with the incidence of stroke.

The rightsholder did not grant rights to reproduce this item in electronic media. For the missing item, see the original print version of this publication.

The Stroke Council of the American Heart Association has recommended aggressive treatment of chronic hypertension to maintain the systolic blood pressure below 140 mm Hg and the diastolic blood pressure below 90 mm Hg.4 However, even modest reductions in blood pressure (i.e., 9 mm Hg systolic and 5 mm Hg diastolic) can reduce the relative risk of stroke by about one third.32 Comprehensive recommendations for the treatment of hypertension have been made by the sixth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.7

Cigarette smoking is associated with an increased risk of stroke. Hence, patients should be strongly encouraged to stop smoking.4

The association of alcohol intake with stroke risk is dose-dependent. Some investigators have described a J-shaped curve for stroke risk, with a slightly increased risk of stroke in patients who are abstinent compared with those who consume moderate amounts of alcohol. Conversely, the risk of stroke is doubled in patients who consume three or more alcohol drinks per day.33 Although absolute statements on alcohol consumption cannot be made based on the available data, heavy drinking should be discouraged.

Atrial fibrillation is one of the strongest independent risk factors for stroke.33 The literature contains a great deal of information on the use of warfarin (Coumadin) or acetylsalicylic acid (aspirin) for stroke prevention in patients with atrial fibrillation.3436 The choice of agent is determined by additional risk factors for stroke and the patient's ability to comply with therapy.3436 Unfortunately, evidence also exists that these treatment options continue to be under-used in up to 75 percent of patients.37

Mildly elevated serum homocysteine levels have been shown to hasten the development of atherosclerosis and thereby increase the risk of ischemic stroke.38 Elevated homocysteine levels can be normalized through the administration of vitamin B6, vitamin B12 or folic acid.39 However, the link between stroke reduction and the administration of these vitamins remains to be proved.

Other potential risk factors for ischemic stroke include hyperlipidemia and diabetes mellitus.13,4046 However, controversy exists regarding the degree of stroke risk associated with these conditions and the action or treatment, if any, that can reduce the risk.

Oral contraceptives may also be associated with an increased risk of stroke. However, several studies have shown that the incidence of stroke is not increased in women who have taken either estrogen or estrogen-progesterone combinations as hormone replacement therapy.44,47


Once risk factors have been modified, the physician, in collaboration with the patient and family, must decide whether medical or surgical treatment is to be used. The North American Symptomatic Carotid Endarterectomy Trial (NASCET)4,48,49 demonstrated the advantage of endarterectomy over medical management in patients with greater than 70 percent stenosis and neurologic symptoms referrable to the stenosed artery.

Recently published data from NASCET II50 have helped to clarify decision-making in the area of carotid endarterectomies. This trial randomized symptomatic patients (TIA or nondisabling stroke) with carotid stenosis of less than 70 percent to either medical management or carotid endarterectomy. Failure rates for the two treatment groups did not differ significantly in patients with carotid stenosis of less than 50 percent (14.9 percent in the endarterectomy group versus 18.7 percent in the medical management group, P = 0.16). Patients with carotid stenosis of 50 to 69 percent benefited from surgery (failure rate of 15.7 percent in the endarterectomy group versus 22.2 percent in the medical management group, P = 0.045).

The NASCET II investigators have urged physicians to consider several points, such as the surgical risk to the individual patient, the technical skill of the surgeon and the overall complication rate of the institution, when they are contemplating endarterectomy in patients with 50 to 69 percent carotid stenosis.50

Carotid endarterectomy guidelines can be summarized as follows:

  1. Surgery should be considered in symptomatic patients with 70 percent carotid stenosis.48,49

  2. Surgery may be considered in symptomatic patients with carotid stenosis of 50 to 69 percent. The risks and benefits of surgery should be carefully considered in these patients.50

  3. Surgery should not be considered in patients with carotid stenosis of less than 50 percent.50

Some investigators have advocated vertebral artery surgery for patients with vertebrobasilar disease who have continued symptoms despite maximal medical therapy and are good surgical candidates.4 Information on favorable outcomes after vertebral artery surgery for stroke prevention is derived from anecdotal case reports and nonrandomized case series. Hence, this surgical treatment can only be recommended cautiously and in selected patients.4


Some form of stroke prevention therapy must be provided for all patients with TIA. Agents appropriate for this use include aspirin, ticlopidine (Ticlid), clopidogrel (Plavix) and warfarin. When choosing a treatment, the physician should give particular attention to the patient's previous use of prophylactic agents for TIA, if any, and the patient's compliance with these treatments.

Aspirin. Thromboxane A2 is a potent inducer of platelet aggregation. Because aspirin inactivates cyclooxygenase activity for the life of platelets, thromboxane A2 cannot be produced.47 Low and high doses of aspirin appear to reduce thomboxane A2 production. However, prostacyclin, which is thought to have some intrinsic antiplatelet effects, is inhibited in a greater proportion at higher dosages of aspirin.51 The clinical application of this biochemical situation involves finding the appropriate dosage of aspirin for stroke prevention.52,53

Older studies demonstrated benefit from high-dose aspirin therapy (975 to 1,300 mg per day).54,55 Studies such as the Dutch TIA Trial53 and the Swedish Aspirin Low-Dose Trial (SALT)56 have examined the effects of low-dose aspirin therapy.

The randomized, double-blind Dutch study53 compared the use of low-dose aspirin (30 mg per day) and medium-dose aspirin (283 mg per day). No difference between the two treatments was found for the primary outcomes of death from all vascular causes, non-fatal stroke or nonfatal myocardial infarction.

SALT56 compared low-dose aspirin therapy (75 mg per day) with placebo. In this study, treatment with low-dose aspirin decreased the incidence of stroke and death by 18 percent compared with placebo.

No studies have directly compared low-dose, medium-dose and high-dose aspirin therapies for stroke prevention. Thus, the American College of Chest Physicians currently recommends aspirin in a dosage of 50 to 325 mg per day for stroke prevention.11

Ticlopidine. Adenosine diphosphate (ADP) is a nucleotide involved in platelet aggregation and fibrinogen binding to the glycoprotein IIb/IIIa receptor. Ticlopidine prevents these actions, thereby inhibiting platelets for their entire life span.57,58 Studies have shown that ticlopidine (250 mg administered orally twice daily) is more effective than placebo or aspirin in preventing a second thromboembolic stroke.59,60

The adverse events noted in ticlopidine clinical trials were neutropenia, thrombocytopenia, diarrhea, rash, abnormal liver function tests and elevated cholesterol levels.59,60 The manufacturer of ticlopidine recommends close monitoring of complete blood counts with differentials for the first three months of therapy to detect blood dyscrasias.61

Clopidogrel. In early 1998, the U.S. Food and Drug Administration (FDA) labeled clopidogrel for use in the reduction of atherosclerotic events in patients with atherosclerosis documented by recent stroke, recent myocardial infarction or established peripheral arterial disease. By irreversibly inhibiting ADP-induced platelet aggregation, clopidogrel prevents clot formation in a similar fashion as ticlopidine.62

In the major trial leading to the FDA labeling of clopidogrel,63 19,185 patients were given either clopidogrel (75 mg per day) or aspirin (325 mg per day). Patients selected for the trial had previous ischemic stroke, myocardial infarction or atherosclerotic peripheral arterial disease. Efficacy was determined by the subsequent occurrence of ischemic stroke, myocardial infarction or vascular death. The incidence of any one of these outcomes was 5.32 percent per year in the clopidogrel-treated group and 5.83 percent per year in the aspirin-treated group, a small but statistically significant difference (P = 0.043).

Clopidogrel appears to have a more promising side effect profile than ticlopidine. In its major clinical trial,63 clopidogrel was more likely to cause rash and diarrhea than aspirin; conversely, aspirin caused nausea, vomiting and gastrointestinal hemorrhage more frequently than clopidogrel. The recommended dosage of clopidogrel is 75 mg per day taken with or without food.64

Warfarin. Few studies have investigated the use of warfarin therapy for stroke prevention outside the setting of atrial fibrillation. The Stroke Prevention in Reversible Ischemia Trial (SPIRIT)65 was halted early because of excess major bleeding complications, particularly intracerebral hemorrhage, in the anticoagulation group. According to investigators in this trial, high-intensity anticoagulation (i.e., International Normalized Ratio [INR] of 3.0 to 4.5) is a customary practice in many European countries. In the United States, an INR range of 2.0 to 3.0 is usually employed for ischemic stroke prevention. SPIRIT demonstrated a clear incremental relationship between hemorrhagic complications and increasing INR, with each 0.5-unit increase in the INR causing the incidence of bleeding to increase by a factor of 1.43.

It is unclear what a comparison of aspirin and warfarin at the standard U.S. INR ranges would show. A single retrospective study66 reviewed outcomes in patients with angiographically identified intracranial artery stenosis who were treated with either warfarin (n = 88) or aspirin (n = 63). A statistically significant difference in the number of recurrent events favored the use of warfarin over aspirin (P <0.01).

A prospective, double-blind randomized study comparing warfarin and aspirin in the prevention of stroke secondary to intracranial stenosis is currently being conducted.

Continue Reading

More in AFP

More in PubMed

Copyright © 1999 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See permissions for copyright questions and/or permission requests.