Current Concepts in Cerebrovascular Prevention

Summary

Prevention remains a major therapeutic approach of stroke. In low-risk patients, primary prevention with medium-dose aspirin has provided no risk reduction of stroke and cannot be recommended, except on an individual basis when patients have many vascular risk factors. Antiplatelet drugs are the treatment of choice in the secondary prevention of stroke of arterioembolic origin. Aspirin remains the most commonly used antiplatelet at medium doses of 200–300 mg daily but with a modest stroke risk reduction of 20 to 25%. Unfortunately, till now, no other antiplatelet has an impressively greater effect than aspirin. Ticlopidine (500 mg daily) is slightly more effective than aspirin but its use is limited by rare (0.9%) but serious haematological adverse effects. Clopidogrel, a thienopyridine derivative like ticlopidine but without haematology toxicity, has provided a stroke risk reduction of about 7% against aspirin and will be an alternative to aspirin. The superiority of the combined therapy of dipyridamole and aspirin over aspirin alone must be confirmed by other trials because the dose of aspirin used was much lower than medium doses currently used. ESPRIT, being carried out now, will give an answer about efficacy and safety of oral anticoagulant (INR 2.0–3.0) versus aspirin and of the benefit of the combination of dipyridamole to aspirin versus aspirin alone. In NRAF, oral anticoagulants are clearly indicated in primary and secondary prevention of stroke, providing a 70% risk reduction of stroke. Independent risk factors for stroke in AF are important to recognize. In patients >75 years, probably at higher risk of intracranial bleeding, the intensity of anticoagulation must be carefully controlled and maintained to an INR of 3. Aspirin, though less effective, may be an alternative to oral anticoagulants when they are contraindicated.

Zusammenfassung

Prävention bleibt ein wichtiger therapeutischer Zugang zur Verhinderung des Hirnschlags. Im Gegensatz zu Hochrisikopatienten führte die primäre Prävention mit Aspirin in mittleren Dosen bei Patienten mit geringer Neigung zum Hirnschlag zu keiner Verminderung der aufgetretenen Hirnschläge und ist somit nicht zu empfehlen. Die Behandlung mit Aggregationshemmern ist Methode der Wahl in der sekundären Prävention bei Hirnschlägen arterioembolischen Ursprungs. Aspirin bleibt der meist verwendete Aggregationshemmer, mit einer durchschnittlichen Tagesdosierung von 200–300 mg, obwohl es nur zu einer mässigen Reduktion des Hirnschlagsrisikos um 20–25% führt. Leider hat bis zum jetzigen Zeitpunkt kein anderer Aggregationshemmer zu einer deutlich stärkeren Risikoverminderung geführt. Ticlopidin (500 mg täglich) ist etwas effektiver als Aspirin, aber sein Einsatz ist durch seltene (0,9%), aber erhebliche hämatologische Nebenwirkungen beschränkt. Clopidogrel, wie Ticlopidin ein Thienopyridin-Derivat, aber Ohne hämatologische Toxizität, führte zu einer weiteren Risikoverminderung um 7% verglichen mit Aspirin und stellt eine Alternative zu dieser Substanz dar. Der Vorteil einer Kombinationstherapie mit Clopidogrel und Aspirin gegenüber einer reinen Aspirin-Therapie muss erst durch andere Studien erwiesen werden, da die verabreichte Aspirin-Menge deutlich kleiner als die üblicherweise verwendete Dosis war. Die nun durchgeführte ESPRIT-Studie wird sowohl über die Effektivität und Sicherheit von oralen Antikoagulantien (INR 2,0–3,0) im Vergleich zu Aspirin als auch über den Vorteil einer Kombinationstherapie von Clopidogrel und Aspirin gegenüber einer reinen Aspirin-Therapie Auskunft geben. Die NRAF-Studie weist mit einer Risikoreduktion von 70% auf die deutlichen Vorteile von oralen Antikoagulantien in der primären und sekundären Prävention hin. Unabhängige Risikofaktoren für den Hirnschlag bei Vorhofflimmern müssen unbedingt berücksichtigt werden. Bei Patienten über 75 Jahre mit einem möglicherweise höheren RiSiko für Gehirnblutungen muss die Intensität der Behandlung mit oralen Koagulantien sorgfältig kontrolliert und auf eine INR von 3 eingestellt werden. Aspirin, obwohl weniger wirksam, ist eine Alternative zu oralen Antikoagulantien falls diese kontraindiziert sind.

Introduction

During the last decades, substantial reduction in cerebrovascular disease mortality was observed in most Western European countries, due to better control of vascular risk factors, mainly hypertension and probably other more specific treatments (antithrombotics, carotid surgery). Nevertheless, stroke remains the third leading cause of death and an important cause of long-term disability in industrialized countries. Incidence rate of stroke is age-related and increases from about 20 per 10 000 for people aged 55–64 years up to more than 200 per 10000 for people aged 85 and over [1,2]. Although new promising therapies of acute stroke are emerging, prevention remains a major therapeutic approach of stroke. The actual concepts of cerebrovascular prevention are discussed in this review.

Primary prevention

Only two studies [3,4] examined the benefit of antiplatelet therapy on the incidence of vascular events in low risk patients. In the meta-analysis of both studies by the Antiplatelet Trialists’ [5], the absolute benefit on all vascular events seems small and statistically uncertain with only about four vascular events averted for every 1000 patients treated for five years. However, considering separately myocardial infarction and stroke prevention, a still unexplained great difference was observed. In the American Physician’s Health Study [3], a 44% risk reduction of myocardial infarction was demonstrated in the group assigned to aspirin and the meta-analysis of the British and American studies confirmed an absolute reduction of 5 myocardial infarctions for 1000 patients treated. Yet, none of the studies proved the benefit of aspirin in the primary prevention of stroke and this was not because of the rarity of events. Moreover, a slight but non significant increase of cerebral haemorrhages was observed in the aspirin groups. From these data, antiplatelets cannot be recommended systematically in primary prevention except in individual patients with many vascular risk factors.

Non rheumatic atrial fibrillation (NRAF)

Epidemiological studies ([6,7,8,9] and reviewed in [10]) have reported age-specific prevalence of NRAF. Overall prevalence is estimated at about 0.9%. The median age of individuals with NRAF is approximately 75 years. Atrial fibrillation is present in 2.3% of people older than 40 years, increases to 5.9% of those older than 65 years and could be as high as 10% in those older than 80 years. 70% of people with NRAF are between age 65 and 85 years. In the Framingham Study [6], there was a near fivefold excess of stroke when NRAF was present. Yet, attributable risk of stroke from NRAF increased significantly with age, rising from 1.5% for those aged 50—59 years to 23.5% for those aged 80—89 years. For people aged 80-89 years, NRAF was the sole cardiovascular condition to exert an independent effect on stroke incidence. From 1989 to 1992, five randomised clinical trials ([11,12,13,14,15] and analysis of pooled data in [16]) of anticoagulation versus placebo clearly demonstrated the effectiveness of oral anticoagulant in the prevention of stroke in NRAF. Annual incidence of thromboembolism, mainly stroke, decreased from 5.5% to 2% in AFASAK, from 6.3% to 2.3% in SPAF I, from 2.98% to 0.41% in BAATAF, from 5.2% to 3.5% in CAFA and from 3.6% to 2% in SPINAF. Target range for intensity of anticoagulation varied from 1.4 to 2.8 in SPINAF to 2.0 to 4.5 in SPAF. Yet, the effectiveness of oral anticoagulant was consistent among trials reducing the frequency of all strokes by 68%. Analysis of pooled data from the five randomised controlled trials [16] revealed that the INR at the time of stroke was most commonly below 1.7 or the patients were not taking warfarin at that time.

In the pooled data of the five randomised controlled trials [16], four independent risk factors for stroke were identified by multivariate analysis. These were a history of previous TIA or stroke (RR: 2.5), diabetes (RR: 1.7), history of hypertension (RR: 1.6) and increasing age (RR: 1.4 for each decade). Cardiac diseases are not independent risk factors but congestive heart failure or coronary artery disease associated with AF increased the risk of stroke nearly threefold. Echocardiographic findings predictive of thromboembolism (left auricular size, left ventricular dysfunction and spontaneous intraauricular contrast) were examined only in SPAF study [17]. The type of AF, constant or paroxysmal, and the length of time the patient was in AF had no effect on the stroke rate. Thus, in the pooled data, the annual risk of stroke varied from 1% in patients younger than 65 years and with no other risk factors (“lone AF”), to 8.1% in patients older than 75 years who had one or more of the other risk factors. Warfarin appears to decrease the risk of stroke in all subgroups of patients except those younger than 65 years with no other risk factors, in whom the risk of stroke was low.

Safety of oral anticoagulation is a crucial point. Reluctance to anticoagulation remains among physicians, particularly for old patients, because intracranial haemorrhage is widely believed to increase with age. In the analysis of the pooled data [16], the annual frequency of major bleeding events was low and quite similar in controls (1%), in warfarin-treated patients (1.3%) and in aspirin-treated patients (1%). Age was not a statistically significant risk factor for intracranial bleeding. This last point was debated [18] for two reasons. First, the mean age of the population studied in the five atrial fibrillation trials was 69 years, and only 25% were older than 75 years, whereas it is estimated that approximately half of the patients with AF in the United States are older than 75 years [8]. Second, the follow-up of the patients in the trials lasted 2—3 years. Therefore the long-term safety of anticoagulation in very elderly patients remains uncertain. In SPAF II trial [19,20] comparing warfarin (INR 2.0–4.5) with aspirin 325 mg daily for prevention of ischaemic stroke and systemic embolism in two parallel randomised trials involving 715 patients aged 75 years or less and 385 patients older than 75 years followed-up for a mean of 3.1 years, age and intensity of anticoagulation were independent risks for bleeding at any site during anticoagulation. Rate of major haemorrhage were 1.7% per year in patients ≤75 years who received warfarin versus 4.2% per year in those >75 years (RR 2.6). Rates for intracranial bleeding were 0.6% per year ≤75 years and 1.8% per year >75 years (RR 3.2). Although warfarin remained better than aspirin, most of the benefit of warfarin over aspirin for the prevention of ischaemic stroke in these elderly patients was cancelled by bleeding complications. Yet, most patients had relatively high intensity of anticoagulation at the time of intracranial haemorrhage (INR >3). Another recent study confirmed that intensity of anticoagulation therapy and deviation from target INR may be stronger risk factors for bleeding than age [21]. Optimal level of anticoagulation for primary [22] and secondary prevention [23] of stroke in AF has been determined at an INR around 3. INR <2 were of no efficacy and INR >4 to 5 were dangerous. The benefit of aspirin in the prevention of stroke in AF was examined in a few trials [11,12,19,24]. In AFASAK study [11], the risk reduction of ischaemic stroke was 18%, statistically not significant, with a dose of aspirin of 75 mg daily. In SPAF study [12], the risk reduction was 44% with a dose of 325 mg daily. A subgroup analysis suggested that aspirin was not effective for patients older than 75 years of age. The EAFT [24] documented a mild but statistically not significant benefit of aspirin over placebo for the prevention of stroke (risk reduction 16%). Taken together these results suggest that aspirin provides a stroke risk reduction of 20 to 25% in AF. Direct comparison of warfarin to aspirin [11,19,24] demonstrated that anticoagulation therapy is approximately 50% more effective than aspirin therapy for prevention of ischaemic stroke in AF. SPAF III [25] compared a combination of low-intensity, fixed dose warfarin (INR 1.2–1.5) and aspirin (325 mg daily) to adjusteddose warfarin (INR 2.0–3.0) in the prevention of stroke in AF. The combination therapy was clearly insufficient compared to adjusted-dose warfarin (INR 2.0–3.0).

In summary, oral anticoagulation, with an INR at 3.0, is clearly indicated in primary and secondary prevention of stroke in patients with AF, especially when specific risk factors (previous TIA or stroke, diabetes, hypertension and older age) are associated. Aspirin is less effective than anticoagulants but is an alternative when anticoagulants are contraindicated. The patients younger than 60 years with no other risk factors (“lone AF”) have a low risk of stroke (<0.5%/year) and can be given aspirin therapy. Age and intensity of anticoagulation are risk factors for intracranial bleeding. Thus, for the patients >75 years who certainly benefit of anticoagulation because they are frequently at high risk of stroke, the treatment must be carefully monitored with a target INR at 3.0.

Secondary prevention

Atheroembolic stroke

General considerations

In this situation, the efficacy of antiplatelet therapy and particularly aspirin, the most commonly studied substance, has been demonstrated by the recent meta-analysis of the Antiplatelet Trialists Collaboration [5]. 145 randomised trials of prolonged antiplatelet therapy versus control, including 70 000 high risk patients divided into four main categories (with acute myocardial infarction MI, past history of MI, past history of stroke or TIA and others with angina, peripheral vascular disease, arterial bypass surgery or angioplasty) have been included. The odds reduction in vascular events were about one quarter in each of the four above mentioned main categories and statistically significant independently of factors like age, sex, hypertension or diabetes. In particular, taking together all high risk patients, reductions reached about one third in non-fatal MI, about one third in non-fatal stroke and about one sixth in vascular death. Among more than 10 000 patients with a prior history of stroke or TIA, antiplatelet therapy for a mean duration of 33 months produced an absolute reduction of 20 non-fatal stroke for 1000 patients treated.

Aspirin and about dose

Aspirin blocks the production of thromboxane A2 via permanent inhibition of platelet oxygenase. Experimentally, a dose of aspirin as low as 30 mg daily completely inhibits the platelet oxygenase while preserving the antithrombotic effect of prostacyclin produced by endothelial cells. Nevertheless, in practice, controversies remain about the optimal dose of aspirin. Two trials only studied low dose aspirin, less than 300 mg daily, in the secondary prevention of stroke. The Swedish Aspirin Low-Dose Trial (SALT) was the first placebo-controlled study to show a beneficial effect of daily aspirin dose of less than 300 mg. The risk of stroke or death was reduced by 18% with a daily dose of 75 mg compared to placebo [26]. The Dutch TIA Trial showed that a daily dose of 30 mg or 283 mg of aspirin has a similar efficacy in the prevention of vascular events (death from vascular causes, non-fatal stroke or non-fatal MI) in patients with a TIA or minor stroke [27]. The UK TIA Aspirin Trial is the only trial providing a direct randomised comparison of 300 mg versus 1200 mg aspirin versus placebo after TIA or minor stroke [28]. Results were disappointing because, owing to the small number of patients included, neither the low dose nor the high dose of aspirin taken separately were better than placebo. Only the combination of both aspirin groups showed a reduction of 15% of the odds of major stroke, MI and vascular death. The conclusion was drawn that there was no significant difference in efficacy between the 300 mg and 1200 mg daily doses of aspirin.

Antiplatelet Trialists Collaboration meta-analysis [5] tried to answer this question. No significant difference was demonstrated between high doses of aspirin (500—1500 mg daily) and medium doses (75–325 mg daily) in the secondary prevention of vascular events. Noteworthy, direct randomised comparisons were based on a limited number of vascular events in a few trials, mostly heart studies, and indirect comparisons between different doses of aspirin used in different trials should be interpreted with caution because the groups of patients, the treatment modalities, the followup and the end points of the studies may be different. Thus, although medium doses of 300 mg daily are effective and recommended by most neurologists in European countries, the limited data cannot rule out a benefit of higher doses. Indeed no prospective, randomised, blinded trial has ever compared the efficacy of lowmedium doses to high doses (≥1 g) of aspirin in secondary prevention of stroke [29,30]. Yet, the studies agree about the dose dependent side-effects of aspirin [26,27,28]. In the UK TIA Aspirin Trial [28] there was a statistically significant dose response effect for upper gastrointestinal symptoms with an odds ratio for the high dose group compared with the low dose group of 1.5. Gastrointestinal haemorrhages were significantly more frequent within the group taking aspirin and dose-related (3/1000 patients/year in the placebo group; 7/1000 patients/year in the low dose group and 11/1000 patients/year in the high dose group). The odds ratio for GI haemorrhages comparing the high dose with the low dose aspirin group was 1.62 and comparing the low dose with placebo group was 2.57. There was also an excess but not statistically significant risk of intracranial haemorrhage in those allocated to aspirin.

Other antiplatelet drugs

Aspirin, independently of the dose used, has only a modest efficacy and the need for more potent antiplatelet drugs is obvious. Yet, differences between one antiplatelet regimen and another in their effects on vascular events are unlikely to be large, therefore, direct randomised comparisons may be needed that are of sufficient size with a total of a few thousand vascular events to detect the superiority of other antiplatelet treatment against aspirin.

Ticlopidin

Ticlopidin is a thienopyridine that inhibits platelet function by interacting with platelet glycoprotein IIb/IIIa in a way to inhibit the binding of fibrinogen to activated platelets. Two randomised, double-blind trials have proved the efficacy of ticlopidin at a dose of 500 mg daily in the secondary prevention after stroke [31,32]. The Canadian American Ticlopidine Study [31] in thromboembolic stroke demonstrated that ticlopidin was better than placebo with a relative risk reduction of stroke, MI and vascular death of 30.2% in the ticlopidin group compared to the placebo group. Analysis by intention to treat gave a smaller but still substantial risk reduction of 23.3% for stroke, MI and vascular death. The Ticlopidine Aspirin Stroke Study [32] compared the effects of ticlopidine hydrochloride (500 mg daily) with those of aspirin (1300 mg daily) on the risk of stroke or death after recent transient or mild persistent focal cerebral or retinal ischaemia. Ticlopidine was found to be slightly more effective than aspirin with a 10% risk reduction for stroke and vascular death rates at three years. If myocardial infarction is included in the analysis together with stroke and vascular death, the risk reduction is only a 6% marginal advantage in favour of ticlopidine [33]. Yet, risk of side effects were greater in the ticlopidine group: diarrhea were reported in 20%, skin rash in 14% and severe but reversible neutropenia in 0.9%. Subgroups analysis of TASS [34], which should be interpreted with caution, showed a greater benefit of ticlopidine for women, those failing aspirin therapy, those with vertebro-basilar symptoms, hypertension, diabetes and no severe carotid stenosis. The Antiplatelet Trialists Collaboration meta-analysis, including three trials comparing the effect of ticlopidine to aspirin, confirmed a risk reduction of stroke, MI and vascular death of 8% in favour of ticlopidine. Ticlopidine use, limited by potentially severe side-effects, remains an alternative treatment to aspirin in patients allergic or intolerant to aspirin or when aspirin fails.

Clopidogrel

Clopidogrel, a new thienopyridine derivative, chemically related to ticlopidine blocks activation of platelets by adenosine diphosphate by selectively and irreversibly inhibiting the binding of this agonist to its receptor on platelets, thereby affecting ADP-dependent activation of the GpIIB-IIIa complex, the major receptor for fibrinogen present on the platelet surface. Clopidogrel (75 mg daily) was compared to aspirin (325 mg daily) in CAPRIE [35], a large randomised, double-blind international trial based on 19185 patients with atherosclerotic disease like recent ischaemic stroke or myocardial infarction or symptomatic peripheral arterial disease and followed up for 1—3 years. The trial demonstrated a slight superiority of clopidogrel over aspirin with a statistically significant risk reduction of 8.7% on a composite outcome cluster of ischaemic stroke, MI and vascular death. Yet, a difference in the true benefit of clopidogrel among the three clinical subgroups remained unexplained. Indeed, the relative risk reduction was highest (23.8%) for patients with peripheral arterial disease, lower (7.3%) for patients with stroke and there was a relative risk increase (3.7%) for patients with MI. The lack of observed benefit of clopidogrel over aspirin in the MI subgroup prompted a single additional analysis. There were 2144 patients in the stroke and peripheral arterial disease groups who had past history of MI. When this cohort was combined with the 6302 patients who presented with MI as the qualifying event, the overall relative-risk reduction was 7.4% in favour of clopidogrel, consistent with the observed benefit in the rest of the CAPRIE cohort. Concerning adverse effects, there was no major difference between clopidogrel and aspirin in term of safety. Clopidogrel is safer than ticlopidine since there was no excess neutropenia in the clopidogrel group. Rash and diarrhea occurred respectively in 0.23% and 0.26% in the clopidogrel group, twice as frequently as in the aspirin group but balanced by the extent of gastrointestinal discomfort and haemorrhage with aspirin. Clopidogrel is thus moderately more effective and as safe as medium dose aspirin in the secondary prevention of atherothrombotic disease.

Dipyridamole

Dipyridamole is an antiplatelet and vasodilator agent which interferes with platelet function by increasing the cellular concentration of cyclic AMP. This effect is mediated by inhibition of cyclic nucleotide phosphodiesterase and/or by blockade of uptake of adenosine, which acts at A2 receptors for adenosine to stimulate platelet adenylyl cyclase. The question about the benefit of adding dipyridamole to aspirin in secondary prevention of stroke has been re-actualised by the results of the second European Stroke Prevention Study ESPS 2 [36]. Two previous trials, the French study “AICLA” [37] and the American-Canadian study “ACCSG” [38], did not prove any superiority of the combination of aspirin (1000— 1300 mg daily) to dipyridamole (225—300 mg daily) over aspirin alone. ESPS 1 [39] which compared only the combined treatment of aspirin (975 mg daily) and dipyridamole (225 mg daily) to placebo, showed a 33% benefit in favour of the treated group. ESPS 2, a randomised, placebo-controlled, double-blind trial, compared the efficacy of aspirin alone (50 mg daily), dipyridamole alone (400 mg daily), both treatments in combination and placebo in 6602 patients with prior stroke or TIA. Conclusions were that low dose aspirin and high dose dipyridamole were equally effective for the secondary prevention of ischaemic stroke and that the combination was significantly superior to each drug alone. Risk of stroke and death was reduced by 13% with aspirin alone, 15% with dipyridamole alone and 24% with the combination. Yet, the results of this trial are controversial because the dose of aspirin was much lower than the medium dose of 200—300 mg commonly used. So we cannot exclude that the combination of low dose aspirin and dipyridamole could be less effective than higher dose of aspirin alone. In fact, the stroke and death event rate reduction for the dipyridamole and aspirin group in ESPS 2 was not as great as that in the ESPS 1 when 975 mg of aspirin a day was given instead of 50 mg. So, in clinical practice the applicability of the results of ESPS 2 are debated. ESPRIT, a randomised trial after TIA or minor stroke, and including three arms: oral anticoagulants, aspirin alone and the combination of aspirin and dipyridamole, is now being carried out and should give more information about the benefit of combining dipyridamole to aspirin in the secondary prevention of stroke.

Oral anticoagulants

If the indication of anticoagulation in primary and secondary stroke prevention in atrial fibrillation is well documented by the results of many trials, no double-blind randomised trial studied the benefit of anticoagulants in stroke of atheroembolic origin. Data from secondary prevention trials after myocardial infarction indicate that the efficacy of anticoagulation may be higher than aspirin with about 40% relative risk reduction [40,41,42,43]. Thus, atherosclerosis being the common cause of MI and ischaemic stroke, the risk/benefit ratio of anticoagulation is now studied. SPIRIT compared anticoagulation (INR 3.0–4.5) with aspirin 30 mg daily in patients after cerebral ischaemia of presumed arterial origin. This trial was stopped early because of an excess of major bleeding complications in the anticoagulation group. Data from this trial indicate that the bleeding risk at an INR of 2.0–3.0 is lower and comparable to that for other indications. The aim of ESPRIT (Stroke Prevention in ReverSible Ischemia Trial), being carried out now, is to compare the efficacy and safety of (a) anticoagulation (INR 2.0–3.0) versus aspirin (in any dose between 30 mg and 325 mg daily) and (b) the combination of aspirin (in any dose between 30 mg and 325 mg daily) and dipyridamole (400 mg daily) versus aspirin alone in patients after TIA or minor ischaemic stroke. 4500 patients will be included with a followup of 3 years.

Non rheumatic atrial fibrillation (NRAF)

The European Atrial Fibrillation Study EAFT [24] demonstrated the efficacy of anticoagulation in decreasing the risk of recurrent vascular event by two-thirds after a recent TIA or minor ischaemic stroke. The results are similar to those of the five primary prevention studies of NRAF patients [16]. Yet an important difference between primary and secondary prevention is the much higher absolute risk of recurrent stroke with an annual incidence of 12% in the placebo-treated group which is almost three times as much as in the placebo-treated group of the primary prevention studies (4.5%). In absolute numbers, anticoagulation prevents 90 vascular events, mainly strokes, if 1000 patients are treated for one year. Aspirin is a safe, effective alternative, though less than warfarin, when anticoagulants are contraindicated. Aspirin prevents 40 vascular events of all types for 1000 patients treated.