Abstract
STROKE IS A MAJOR CAUSE OF MORBIDITY and mortality in an aging population. The current understanding of the pathophysiology of atherosclerotic diseases, the most common cause of stroke, and the evidence for existing therapeutic interventions for the prevention of stroke are presented. Specifically, we review the evidence for antiplatelet agents, anticoagulants, antihypertensive medications, lipid-lowering agents and carotid endarterectomy for stroke prevention.
Each year in Canada stroke occurs in 50 000 people and accounts for 7% of all deaths. Canada's population of stroke survivors numbers almost 300 000, of whom 30% remain permanently disabled.1,2,3 Care for stroke patients accounts for 2.1% of Canadian health care expenditures.4 Primary prevention of a first stroke and secondary prevention of recurrent events require rapid identification of risk factors and implementation of appropriate preventive measures.
The risk of stroke following an initial cerebrovascular event is high. Of patients presenting to an emergency department with a transient ischemic attack (TIA), 10.5% will have a stroke (half of these occurring in the first 2 days) and 2.6% will die within 90 days.5 Overall, 8.8% of stroke survivors will have a recurrent stroke within the first 6 months, and 15% within 5 years.6 In most cases (about 50%) the stroke is atherothrombotic in origin, with a further 25% attributable to small-vessel lacunar disease and 20% to cardioembolism. Data from a stroke registry reveal that patients who have an atherothrombotic stroke have the highest rates of recurrence within 30 days (18.5%), as compared with those who have a lacunar (1.4%) or cardioembolic (5.3%) stroke.7
In this review we present the current understanding of the pathophysiology of atherosclerotic disease, the most common cause of stroke. We also provide an overview of the available evidence for common therapeutic interventions used for stroke prevention: antiplatelet agents, anticoagulants, antihypertensive medications, lipid-lowering agents and carotid endarterectomy.
Atherogenesis
Atherosclerotic plaques begin to appear in the second and third decade of life. It is now widely agreed that inflammation at an endothelial level is the trigger for atherosclerosis.8 Leukocytes (predominantly monocytes and T-lymphocytes) localize in the earliest atherosclerotic lesions. They bind to vascular cell adhesion molecules (VCAM-1) on the vascular endothelium and migrate into the intima.9,10 This initiates and perpetuates a local inflammatory response. The monocytes mature into lipid-scavenging macrophages and subsequently foam cells. T-lymphocytes express inflammatory cytokines, which continue to stimulate macrophages and endothelial cells and further proliferation of smooth muscle cells11 (Fig. 1).
Sites that experience disturbed blood flow and increased wall stresses, such as arterial branch points, seem to be at highest risk of atherosclerosis and demonstrate impaired atheroprotective mechanisms.12 Nitric oxide production, which inhibits platelet aggregation, proliferation of arterial smooth muscle cells and expression of VCAM-1, may be reduced in these sites.13 Furthermore, smooth muscle cells at these sites produce increased proteoglycans that promote inflammation by facilitating the oxidation of lipoproteins.14
Acute rupture of atherosclerotic plaques, rather than gradual encroachment of the plaque causing luminal narrowing, is the main cause of vascular morbidity and mortality. Both immunologic and mechanical forces are thought to be responsible for plaque rupture, which then precipitates thrombosis. Macrophages produce proteolytic enzymes that degrade the collagen of the plaque's fibrous cap, which makes it susceptible to rupture,15 and produce procoagulant tissue factor, which triggers plaque thrombosis.16
An elevated C-reactive protein level, a marker of inflammation, inhibits nitric oxide release and thus promotes monocyte activation and adhesion to endothelium and increases low-density lipoprotein (LDL) uptake into existing plaques.17,18,19 It is associated with an increased risk of coronary and cerebrovascular events, independent of other known risk factors for vascular disease.20 Stroke risk in the highest quartile of C-reactive protein levels is almost double among men and threefold among women as compared with the lowest quartile.21 Therefore, it is hoped that interventions that reduce C-reactive protein levels may also reduce endothelial inflammation and vascular risk. ASA, for example, appears to offer its greatest protective effect in patients in the highest quartile of C-reactive protein levels, reducing the risk of myocardial infarction in this group of men by 55.7% compared with a nonsignificant 13.9% in the lowest quartile of C-reactive protein levels.22
Antiplatelet agents
ASA irreversibly blocks platelet cyclooxygenase and thus prevents the formation of thromboxane A2 (Fig. 2). As a primary preventive measure, antiplatelet agents do not reduce the risk of ischemic stroke among patients without vascular disease. They are associated with an increased risk of intracranial hemorrhage (odds ratio [OR] 1.35, 95% confidence interval [CI] 0.88–2.10)23 and major noncerebral hemorrhage (OR 1.73, 95% CI 1.14–2.63 ).24
The benefits of ASA in the secondary prevention of stroke have been well documented in the period immediately after a stroke25 (Table 1). Long-term antiplatelet treatment after stroke shows an even more impressive reduction of 25 nonfatal strokes and 36 serious vascular events per 1000 treated over a 29-month follow-up period.25 There appears to be no difference in efficacy between low (50 mg) and higher doses (up to 1500 mg) of ASA.29,30
The combination of dipyridamole and ASA has been shown to reduce the relative risk of recurrent stroke compared with ASA alone26 (Table 1). Dipyridamole inhibits platelet aggregation by increasing levels of cyclic adenosine monophosphate and cyclic guanosine monophosphate. The limiting factor in the use of the ASA–dipyridamole combination is the latter's potential effects on coronary perfusion. Dipyridamole can cause coronary vasodilation, which results in increased blood flow to nonstenosed coronary arteries. The result is that myocardial ischemia may be provoked during exercise. As such, the current American College of Cardiology/American Heart Association guidelines recommend that dipyridamole not be used in patients with chronic stable angina.31 However, this recommendation was based on short-acting dipyridamole, which also reduced myocardial ischemia in a similar number of patients in the study quoted by these guidelines.32 With the use of sustained-release dipyridamole, no increase in cardiac events was observed in subjects with prior coronary artery disease.33
Thienopyridines block adenosine - diphosphate– mediated platelet aggregation. Ticlopidine was the first of this class of drug to be studied for stroke prevention. One clinical trial demonstrated that patients treated with ticlopidine had similar rates of stroke as those treated with ASA28 (Table 1). However, serious granulocytopenia associated with ticlopidine has limited its use. Clopidogrel is a second-generation thienopyridine that has also shown similar efficacy to ASA in stroke prevention27 (Table 1), without any increased risk of major hemorrhage or granulocytopenia. Among patients with acute coronary syndromes, vascular events occurred in 9.3% of those given clopidogrel and ASA, as compared with 11.4% of those given placebo and ASA.34 Although there were significantly more patients with major hemorrhage in the clopidogrel group than in the placebo group (3.7% v. 2.7%), there were not significantly more patients with life-threatening or intracranial hemorrhage. To date, there are no published studies specifically examining the ASA–clopidogrel combination in stroke prevention, although several are ongoing.
Summary
ASA remains a first-line therapy in the secondary prevention of atherothrombotic stroke after a first episode, although combination therapy with ASA and dipyridamole is also justified in the absence of coronary artery disease. Clopidogrel is recommended if the patient is ASA-intolerant. Patients with recurrent TIA or stroke while taking ASA should be re-evaluated to exclude other pathogeneses. If none is found, ASA–dipyridamole combination therapy should be given if not already prescribed. Further events, intolerance to dipyridamole or the presence of coexisting coronary or peripheral vascular disease should merit consideration of an ASA–clopidogrel combination therapy.
Anticoagulants
Anticoagulation is achieved primarily with heparin and warfarin. Unfractionated heparin binds with antithrombin III and thus accelerates its neutralizing effects on thrombin and inactivates clotting factors. Warfarin, a vitamin K antagonist, prevents the vitamin K-dependent activation of clotting factors II, VII, IX and X.
There is no evidence supporting anticoagulation in the treatment of acute stroke. Although heparin decreases the incidence of recurrent ischemic stroke in the short term, the benefits are offset by increases in the rates of hemorrhagic stroke35,36,37 (Table 2). The remainder of this section examines the role of warfarin in modifying various clinical risk factors for the long-term prevention of stroke.
Atrial fibrillation
As previously mentioned, ischemic stroke is due to cardioembolism in about 20% of cases. Of these, the majority occur because of a mural thrombus in patients with atrial fibrillation. The annual rate of stroke with atrial fibrillation is about 4.5%. The risk of stroke is increased among patients who have hypertension, diabetes mellitus, left ventricular dysfunction, prior TIA or stroke and are older than 65. Anticoagulation reduces the annual risk to 1.4%. Patients younger than 65 with atrial fibrillation and no other risk factors had only a 1% rate of stroke, even without treatment.42 Long-term treatment with warfarin to achieve an international normalized ratio (INR) above 2.0 decreases not only the risk of stroke, but also its clinical severity and the risk of death.43 ASA has also been shown to be more effective than placebo in this group of patients but less effective than warfarin38 (Table 2).
The risk of hemorrhagic complications is often a concern when considering long-term anticoagulation for atrial fibrillation. However, one study39 demonstrated that the annual risk of ischemic stroke or systemic embolism was greater among patients assigned to low-intensity anticoagulation combined with ASA than among those given full anticoagulation (7.9% v. 1.9%), with no difference in major bleeding (Table 2).
Until recently, conversion of atrial fibrillation to sinus rhythm with the use of electrical or pharmacological methods was felt to be beneficial in reducing long-term embolic complications. However, 2 studies have challenged this concept, showing no significant difference in mortality or stroke reduction between a strategy of restoring and maintaining sinus rhythm and another aimed at rate control of chronic atrial fibrillation. Most strokes occurred only when anticoagulation was stopped or was subtherapeutic.44,45
Other causes of cardioembolic stroke
Patients with mechanical prosthetic heart valves are at high risk of stroke from thromboembolism and require anticoagulation at a level determined by the type of valve. If systemic embolism occurs despite adequate anticoagulation with a target INR of 3.0, 80–100 mg of ASA should be added. Patients with bioprosthetic valves require anticoagulation only for the first 3 months after valve insertion. Thereafter, ASA alone is sufficient prophylaxis.46
Despite treatment with thrombolytic agents, left ventricular thrombus may occur in 28% of patients who have an acute anterior myocardial infarction, with even higher rates in the presence of apical hypokinesis.47,48 The high risk of subsequent embolic events (OR 5.45, 95% CI 3.02–9.83) may be significantly reduced with anticoagulation (OR 0.14, 95% CI 0.04–0.52)49 and has been suggested for 1–3 months after anterior myocardial infarction50 especially when complicated by severe left ventricle dysfunction or previous emboli.
The presence of aortic atheroma of 4 mm or more in thickness causes a 3- to 9-fold increased risk of stroke.51,52,53 This risk is further increased if the atheroma is mobile53,54 or noncalcified.52 The limited evidence available53,54 suggests that anticoagulation is superior to antiplatelet treatment in preventing a combined end-point of embolic events and death for atheroma that are 4 mm or more in thickness or are mobile.
The presence of patent foramen ovale (PFO) may increase the risk of ischemic stroke via paradoxical embolism through right-to-left shunting. PFO is found at autopsy in 26% of the general population and 46% of patients with cryptogenic stroke under the age of 55.55 A meta-analysis of treatments showed that anticoagulation was superior to antiplatelet therapy for secondary stroke prevention in the presence of PFO (OR 0.37, 95% CI 0.23–0.60) and that surgical PFO closure was comparable to anticoagulation (OR 1.19, 95% CI 0.62–2.27).56 Definitive percutaneous transcatheter closure of PFO has been shown to be successful in preventing neurologic events in previously symptomatic patients, in addition to being safe and minimally invasive.57 However, a recent study found no difference in 2-year recurrent event rates (stroke or TIA, death) between those with or without PFO and atrial septal aneurysm,58 which raises the question of whether aggressive anticoagulation or surgical treatment is truly warranted.
Recurrent stroke of noncardioembolic origin
Anticoagulation may seem an attractive option for patients with recurrent cerebrovascular ischemia without an identifiable cardiac source in whom antiplatelet options have been exhausted. However, one study demonstrated no difference between ASA and warfarin (target INR 1.4–2.8) in reducing the risk of recurrent noncardioembolic stroke41 (Table 2). A similar study also found that anticoagulation (target INR 3.0–4.5) was associated with an increased risk of overall bleeding compared with ASA (OR 1.43, 95% CI 0.96–2.13 for each 0.5 unit increase in INR) largely attributable to an increase in intracranial hemorrhage40(Table 2).
The presence of antiphospholipid antibody is an independent risk factor for first ischemic stroke, occurring in 9.7% of stroke patients versus 4.3% of control subjects in one study (adjusted OR 2.31).59 Although ASA is sufficient prophylaxis as primary prevention,60 for patients who have had a previous thrombotic event, anticoagulation greatly reduces the risk of recurrent events, both venous and arterial (hazard ratio [HR] 0.32, 95% CI 0.15–0.7 for INR < 3) as compared with ASA (HR 0.93, 95% CI 0.51–1.68).61
Summary
Heparin should be avoided in the treatment of acute stroke. Treatment with warfarin for patients with atrial fibrillation without prior TIA or stroke is outlined in Table 3. Warfarin is also indicated as primary stroke prevention for patients with prosthetic heart valves and for recent acute myocardial infarction (< 3 months) with apical hypokinesis or thrombus and as secondary prevention in the presence of antiphospholipid antibody. It may also be indicated for recurrent cryptogenic stroke with PFO.
Antihypertensive medications
Observational studies demonstrate the strong positive association of stroke risk with rising blood pressure in patients with or without previous cerebrovascular events.63 Each 7.5-mm Hg rise in diastolic blood pressure doubles the risk of stroke,64 while isolated systolic hypertension confers an even greater risk, especially among middle-aged men, who have an almost 5-fold risk of stroke.65 A meta-analysis of some of the largest primary prevention trials of thiazides and β-blockers, totalling nearly 50 000 patients, demonstrated a 38% relative risk reduction in the rates of primary stroke with a difference of 5–6 mm Hg in diastolic blood pressure over 5 years.66 Numerous guidelines have been developed for the treatment of hypertension67,68,69,70 that emphasize a target blood pressure of less than 140/90 mm Hg and less than 130/85 mm Hg for patients who have had a previous cerebrovascular event and have diabetes or other vascular disease.
The question of which antihypertensive medication should be the first choice remains controversial. Thiazide diuretics, angiotensin-converting-enzyme (ACE) inhibitors, angiotensin receptor blockers and β-blockers have all been studied (Table 4). Although thiazides and ACE inhibitors are both widely accepted as first-line therapy, choosing one over the other remains a contentious issue. While some studies have favoured thiazides71 and others ACE inhibitors,73,74 others have found no difference72 (Table 4). Practically, many patients require more than 1 antihypertensive to achieve normal blood pressure, which makes the need to choose one over the other irrelevant. Indeed, in one study the combination of ACE inhibitors and the diuretic indapamide was shown to be more effective than monotherapy for secondary stroke prevention.77 The LIFE study75 (Table 4) suggests a role for angiotensin-receptor blockers (ARBs) in primary stroke prevention, demonstrating a 25% reduction in stroke for losartan when compared with atenolol. A substudy76 of patients with left ventricular hypertrophy and isolated systolic hypertension demonstrated an even more impressive 40% stroke reduction.
Although restoration of normal blood pressure remains central to preventing strokes, studies of ACE inhibitors or angiotensin-receptor blockers have demonstrated reductions in rates of stroke with very modest reductions in blood pressure.74,78 Angiotensin-II is known to increase VCAM-1 expression and the production of inflammatory cytokines that promote atherogenesis79,80 (Fig. 1). Therefore, inhibition of angiotensin-II may have protective effects beyond blood pressure control.
Summary
Treatment of hypertension significantly reduces the risk of stroke. Both thiazides and ACE inhibitors have been shown to reduce this risk. Although there is controversy in the choice of the first-line medication, many patients require a combination of 2 or more antihypertensive drugs to normalize blood pressure.70 Therefore, thiazides, ACE inhibitors or a combination of both should all be considered as first-line therapy. Additional therapies should depend on the clinical circumstances. Patients intolerant of ACE inhibitors should be treated with an angiotensin-receptor blocker, particularly in the presence of diabetes or leftventricular hypertrophy for renal protective effects and cardiac afterload reduction respectively. β-Blockers may be preferred in the presence of diabetes, coronary artery disease or congestive heart failure.67
Lipid-lowering agents
LDL plays an important role in the formation of atherosclerotic plaques and the endothelial inflammatory pathway. Traditionally, the rationale of treating hyperlipidemia was to reduce the amount of substrate available for plaque formation. However, there is increasing evidence that agents used to treat hyperlipidemia, in particular the HMG–CoA reductase inhibitors (statins), also have anti-inflammatory and endothelial protective effects that are independent of lipid lowering and confer protection against vascular events. For example, statins have been shown to reduce levels of C-reactive protein, which results in a decreased risk of coronary events independent of lipid lowering.81,82
Unlike coronary artery disease, there is surprisingly little evidence to support a correlation between elevated serum cholesterol levels and overall cerebrovascular events. Prospective observational cohorts, with 450 000 patients, demonstrated no correlation except among patients under the age of 45.83 This may be in part because a positive association with nonhemorrhagic stroke is offset by an observed inverse relation with intracranial hemorrhage.84,85
Meta-analyses have shown that statins reduce the risk of stroke.86,87,88 For primary prevention, an earlier study of pravastatin did not demonstrate a reduced risk of stroke among hyperlipidemic subjects.89 However, atorvastatin was recently shown to decrease the risk of stroke among patients not traditionally deemed hyperlipidemic (total cholesterol < 6.5 mmol/L) but who had multiple vascular risk factors.90 Among patients who have had previous coronary artery disease, both simvastatin and pravastatin have been shown to decrease the risk of stroke without any increase in the risk of hemorrhagic stroke91,92 (Table 5). The Heart Protection Study (HPS)93 (Table 5) has now lowered the lipid treatment threshold even further in high-risk groups, demonstrating marked reductions in all vascular events with baseline LDL as low as 2.5 mmol/L, irrespective of age, sex and other treatments.
Patients with low or normal LDL (< 3.6 mmol/L), low high-density lipoprotein (HDL) levels (< 0.81 mmol/L) and coronary artery disease may also benefit from gemfibrozil, a fibrate drug, which has been shown to reduce the risk of stroke by 31%94 (Table 5). However, the safety and efficacy of gemfibrozil in combination with statins are unclear.
A meta-analysis of nonstatin interventions showed that a treatment strategy to reduce total cholesterol below 6 mmol/L with fibrate drugs, niacin and diet reduced the risk of stroke.88
Summary
All patients with prior TIA, ischemic stroke or with high risk factors for vascular disease as defined by HPS criteria should be treated with a statin irrespective of their serum cholesterol level. Fibrate drugs may also reduce the risk of stroke; however, the safety and efficacy of the combination of the fibrate and statin drugs are not clear.
Carotid endarterectomy
Carotid endarterectomy of a symptomatic severe stenosis of an internal carotid artery remains one of the most effective methods of preventing recurrent stroke, reducing the risk by up to two thirds95,96 (Table 6). The number-needed-to-treat (NNT) to prevent 1 stroke at 2 years is 8 for high grade stenosis (≥70%) and 20 for moderate stenosis (50%–69%).99
Endarterectomy for asymptomatic stenosis of the internal carotid artery remains controversial. Although one study demonstrated a 53% relative risk reduction in ipsilateral stroke and death over 5 years, the number of events was small, with a higher NNT and men appeared to benefit considerably more than women.97,99 Long-term benefits may also be outweighed by the early risks of excess perioperative stroke or death (relative risk [RR] 6.52, 95% CI 2.66–15.96) and are influenced by the complication rates of individual surgeons.100 Guidelines suggest that surgery should be considered only for asymptomatic carotid disease if the complication rate is less than 3% and the stenosis is greater than 60%.101 The age and health of the patient, plaque stability and presence of coexisting cerebral artery disease should also be considered.
Summary
Carotid endarterectomy remains the definitive treatment in patients with symptomatic stenosis of the internal carotid artery of 70% or higher and in selected patients with a stenosis of 50%–69%. We do not currently recommend surgery for asymptomatic disease, preferring to treat proven vascular risk factors aggressively with immediate follow-up in the event of any stroke symptoms.
Conclusion
We have presented the current understanding of the pathophysiology of atherosclerosis, the most important cause of stroke and the evidence available for the medical and surgical interventions for the prevention of stroke. Based on the evidence, it is clear that the use of an antiplatelet agent or warfarin and the correction of hypertension and hyperlipidemia all reduce the risk of stroke. In patients with symptomatic and severe internal carotid stenosis, it is also clear that carotid endarterectomy is the treatment of choice for stroke prevention.
The research in stroke prevention is ongoing with the evaluation of new drugs, refinement of knowledge of current medications and the tailoring of treatments for appropriate patient groups. Although beyond the scope of this review, the modification of lifestyle factors such as smoking and obesity remain crucial components to the prevention of stroke. Stroke risk may be reduced by almost 50% with regular exercise alone.102
The management of risk factors after stroke in general is not optimal103 and requires an increase in both the awareness of these risk factors and the appropriate, evidence-based use of existing therapeutic interventions. We have attempted to present the findings of important clinical trials to ascertain these risks and potential benefits, and we believe there is a need for more aggressive monitoring and treatment by both primary care physicians and specialists alike.
Footnotes
This article has been peer reviewed.
Contributors: Fintan O'Rourke, as the main author, was responsible for the writing, conception, intellectual content and design of the manuscript. Naeem Dean, Naveed Akhtar and Ashfaq Shuaib contributed to the intellectual content and design of the manuscript. All authors critically reviewed and approved the final version of the manuscript.
Competing interests: None declared for Fintan O'Rourke and Naveed Akhtar. Ashfaq Shuaib has received speaker fees and travel assistance from Sanofi-Synthelabo, Boehringer Ingelheim, Roche and Pfizer. Naeem Dean has received speaker fees from Sanofi-Synthelabo and Servier.
Correspondence to: Fintan O'Rourke, Dept. of Elderly Care Medicine, Barnet General Hospital, Wellhouse Lane, Barnet, UK EN5 3DJ
References
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵