Primary And SecondaryPrevention Of StrokeRTF Cheung, PhD(West Ont). MRCP(UK), FHKCP, FHKAM(Med)

Y L Yu,* MD(HK), FRCP, FRCPE, FRACP

Division of NeurologyDepartment of Medicine

The University of Hong Kong

Summary

Results from recent clinical trials have shown that stroke is preventable both before and aftersymptomatic cerebrovascular disease through a change in lifestyle, medical therapy and surgical procedures.Risk factor control, antiplatelet agents, anticoagulation, and carotid endarterectomy have been proven to beeffective under appropriate clinical circumstances. In this update, we shall highlight the areas in whichscientific evidence is available. (HK Pract 1998;20:67-77)

Introduction

Stroke is a major neurologicaldisease and a leading cause of deathand disability in Hong Kong. Somebreakthrough has been achievedrecently in acute stroke treatment,1'2

but prevention remains the optimalstrategy for reducing the burden ofstroke in our community. Stroke is ahe te rogeneous syndrome wi thdifferent causes, types, and severity.Ischaemic stroke is the most common,intracerebral haemorrhage the second,

and subarachnoid haemorrhage theleast common among the three majorstroke types.3 Therapy should beformulated according to the clinicalcharacteristics and risk factor profileof individual patients. Primary strokeprevention refers to prevention ofstroke in people with no history ofcerebrovascular symptoms, andsecondary prevention applies topatients who have had a stroke ortransient ischaemic attack.

Epidemiological and cohorts t u d i e s have i d e n t i f i e d many

modifiable risk factors for stroke.Risk factors identification allows usnot only to recognize people at risk ofstroke but also to promote a healthylifestyle in the general population.Antithrombotic therapy should beconsidered in people at risk ofcerebral ischaemia, as thromboembo-lism is a major pathogenic mechanismfor ischaemic stroke. Finally,corrective procedures are availablefor carotid artery stenosis. Thesemeasures in general apply to bothp r i m a r y and secondary strokeprevention (Table 1).

* Address for correspondence: Dr YL Yu, Suite 1201, Lane Crawford House, 70 Queen's Road Central, Hong Kong.

(Continued on page 69)

67

Hong Kong Practitioner 20 (2) February 1998

UPDATE ARTICLE

Category

Modifiable risk factors

1. Hypertension

2. Cardiac diseases

3. Atrial fibrillation

4. Diabetes mellitus

5. Dyslipoproteinaemia

6. Cigarette smoking

7. Alcohol abuse

8. Obesity

9. Lack of exercise

Antithrombotic therapy

1. Antiplatelet agents

2. Anticoagulation

Carotid artery stenosis

1. Endarterectomy

2. Angioplasty and stenting

Modifiable Risk Factors

Hypertension

Both systolic and diastolic blood

pressures are independently associ-

ated with a 4 to 5 times increased risk

of stroke of all types.4 Hypertension

promotes atherosclerosis in large and

medium-sized arteries, arterioscle-

rosis in small penetrating arteries, and

Action

Screen for hypertension; keep systolic and

diastolic blood pressure around 140 and

85 mmHg, respectively

Check potential cardioembolic source;

screen for other causes of stroke; see

Tables 2 and 3Check ECG and echocardiography; see

Tables 3 and 4

Screen for undiagnosed diabetes mellitus;

aim at strict glycaemic control

Screen and control dyslipoproteinaemia

Advise to quit smoking; consider nicotine

gum or pad

Advise to reduce drinking

Reduce weight

Practise regular exercise

Start oral aspirin for secondary prevention

of stroke and other vascular events;

consider ticlopidine as an alternative

See Table 3

See Table 5Await further studies

cardiac diseases of ischaemic and

hypertensive types. These patho-

logical processes may, in turn, cause

stroke through thrombosis, embolism,

occlusion of small vessels, or arterial

rupture. One meta-analysis revealed

a 42% reduction in the incidence of

stroke and a 45% reduction in the

incidence of fatal stroke with only 5

to 6mm Hg decrease in the mean

diastolic pressure following treat-

ment.5 The Systolic Hypertension in

the Elderly Program (SHEP) reported

a 36% decline in stroke risk with

11 mm Hg reduct ion in the mean

systolic pressure.6 In a local study,

74% of pa t i en t s presented wi th

cerebral haemorrhage were hyperten-

sive.7 Thus, early detection and good

control of hypertension is one of the

most important factors in the primary

prevention of stroke. The same

applies to secondary prevention. The

systolic and diastolic blood pressures

should be kept around 140 mmHg and

85 mmHg, respectively, to minimise

the risk of stroke and other complica-

tions.8, 9

Cardiac diseases

About one in f ive i schaemic

strokes are caused by cardioembo-

l i sm. 1 0 Table 2 summarises the

clinical, topographical, and neuro-

radiological features of cardioembolic

stroke. Nevertheless, the presence of

a potential cardioembolic source in a

patient with cerebral ischaemia may

be co-incidental . Other possible

causes of stroke, such as carotid

artery stenosis and prothrombotic

states, should also be considered.

Different cardiac conditions have

varying risks of causing strokes

(Table 3), and the chance of detecting

s u c h c o n d i t i o n s d e p e n d s o n

thoroughness of the evaluat ion.

Long-term antithrombotic therapy is

available to reduce the risk of embolic

stroke at the expense of an increased

r i sk of sys temic and cerebral

haemorrhages.

69

Primary and Secondary Prevention of Stroke

Table 2: Features of cardioembolic stroke

or ideomotor apraxia

Table 3: Causes and management of cardioembolic stroke*

70

Hong Kong Practitioner 20 (2) February 1998

UPDATE ARTICLE

Atrial fibrillation (AF) affects1% of the general population, 6% ofpeople aged above 65 years, and 10%of those over 75 years of age.11 AFincreases the risk of ischaemic strokethrough systemic embolization ofstasis-induced left atrial thrombi. Thecontinued decline in the frequency ofrheumatic heart disease will makenon-va lvu l a r a t r ia l f i b r i l l a t i o n(NVAF) the commonest cause of thisarrhythmia. Silent cerebral infarctson computed tomography of the brainare present in about 25% of patientswith NVAF, and are often multiple orb i l a t e r a l . 1 2 N e v e r t h e l e s s , t h eassociated cardiac abnormalities,coexistent cardiovascular diseasesand intrinsic cerebrovascular diseasesmay account for 25 to 33% of allstrokes associated with AF.13 Recentrandomised trials have revealed anincreased risk of stroke (about 5% peryear or 5 to 7 times as controls) inpatients with NVAF and have provedthe safety and benefit of long-termant i coagu la t ion with warfar in . 4

Nevertheless, the actual risk of strokeis affected by a number of clinical andechocardiographic factors (Table 4).Using meta-analysis , the overall

relative r isk reduction in strokeamong patients treated with warfarinas compared with placebo is 64%, andthe corresponding benefit of aspirin is22%.14 For patients with AF due torheumatic heart disease, the benefit oflong-term anticoagulation is proven.14

Long-term anticoagulation (withwarfarin) is therefore an establishedprac t i ce for secondary s t rokeprevention in patients with AF due tovalvular heart disease or non-valvularcauses (Table 3), and is increasinglyemployed in primary prevention.However, it must be emphasised thatcareful consideration of complianceand other factors must be givenbefore starting treatment and thatadequate instructions be communi-cated to patients and their relativesin order to reduce the r i s k ofhaemorrhagic complications. Low-intens i ty an t icoagula t ion (within terna t ional normal ized rat iosbetween 2 to 3) carries an extra 1 to2% risk of severe bleeding, includingan 0.3% a n n u a l i n c i d e n c e ofintracerebral haemorrhage.15 Olderpatients (more than 75 years) have ahigher risk of stroke but manifest

Table 4: Risk factors for stroke in non-valvular atrial fibrillation

1, Older than 652. Hypertension

5.

..6..

Previous history of stroke.- .Transient ischaemic attack

- dysfunction .

greater toxicity to warfarin. Patientsat lower risks for cardioembolism andthose with contraindications forwarfarin should be given low-doseaspirin at 325 mg per day to achievestroke prevention.4,14'15 Other medicaland surgical therapies are also usefulin individual cardiac diseases.

Glucose tolerance

Diabetes mellitus is associatedwi th c e r e b r o v a s c u l a r d i s ea se(including atherosclerosis and/orsmall vessel disease) directly andindirectly via coronary artery disease,hypertension, abnormal lipids, andobesity.16 The re la t ive r i sks ofischaemic stroke and stroke of alltypes are elevated by two-folds inboth diabetic men and women.4 In alocal study of 176 Chinese patientswi th acute s troke, the overa l lprevalence of diabetes mellitus andimpaired glucose tolerance was33.5% and 21.0% respectively, with ahigher prevalence being found inpatients with cerebral infarction.1 7

Forty percent of those with diabetesm e l l i t u s w e r e p r e v i o u s l yundiagnosed.

Since undiagnosed diabetesmellitus is a distinct health problemin our elderly population, it may beimportant to screen for diabetesmellitus as part of primary health careof the elderly particularly those whoare at a higher risk. Treatment shouldaim at both strict glycaemic control aswell as control of all risk factors forcardio- and cerebrovascular diseases,although the influence of the formeron the risk of stroke remainsunknown.

71

Primary and Secondary Prevention of Stroke

Cigarette smoking

Cigarette smoking increases therisk of ischaemic stroke by 2 to 3times t h rough the induc t ion ofatherosclerosis and ischaemic heartdisease.4 In addition, cigarettesmoking increases blood levels offibrinogen and other clotting factors,promotes platelet aggrega t ion ,elevates the haematocrit, reduces thelevel of high-density l ipoproteincholesterol, produces acute rise inarterial blood pressure, and enhancesthe breakdown of elastic tissue withinarteries.16 The latter two effects maylead to arterial rupture and intracere-bral or subarachnoid haemorrhage.18

The risk of stroke for smokers ishigher in women than men, becomesless with increasing age, and isproport ional to the number ofcigarettes smoked per day.19 The riskof stroke of all types decreasespromptly towards the level of non-smokers within five years of quittingin both men and women."

Alcohol abuse

Recent studies have clarified therole of alcohol consumption in stroke.Low levels of alcohol intake (one totwo drinks per day) appears to reducethe risk for ischaemic stroke viareducing the risk of ischaemic heartdisease, improving the lipid pattern,increasing the level of prostacyclin,and a c t i v a t i n g the f i b r i n o l y t i csystem.16 Higher levels of drinking,especially following acute intoxica-tion, increases the risk for ischaemicand haemorrhagic strokes by 3 to 4times via haemoconcentration (pro-ducing hyperviscosity), hypertension

(increasing the risk for intracerebralbleeding), rebound thrombocytosis(causing a prothrombotic state),cardiac arrhythmias (generat ingthromboemboli or hypoperfusion),and interactions with smoking (assmoking is more frequent amongheavy drinkers).16

Blood lipids

High level of total and low-densi ty l ipoprotein cholesterols,reduced l eve l o f h i g h - d e n s i t ylipoprotein cholesterol, and increasedrat io of low- to h i g h - d e n s i t ylipoprotein cholesterols constituted y s l i p o p r o t e i n a e m i a , w h i c h isassociated with premature atheroscle-rosis and ischaemic heart disease.20

The association between dyslipopro-te inaemia and stroke has beenestablished recently. Lipid loweringagents have been found to reduce therisk of fatal and non-fatal strokes aswell as retard the progression ofcarotid atherosclerosis.4

Obesity

Obesity is p robab ly not anindependent factor but related todyslipoproteinemia, hypertension,hyperinsulinaemia, glucose intoler-ance, and lack of exercise.16 Inexperimental animal studies, caloricrestriction prolongs survival.21

Lack of exercise

Exercise reduces blood pressure,body weight, and platelet aggregationbut improves lipid pattern and

sensitivity to insulin.16 These effectsmay explain the association betweensedentary lifestyle and strokes of alltypes in both men and women.

Others

Higher-dose oral contraceptiveshave been found to increase the riskof stroke in women over 35 years ofage, smokers, hypertensives, andmigraineurs.1 6 Although low-doseoral contraceptives do not increasethe risk of stroke,22 oral contracep-tives should be avoided in women athigh risk for stroke. Post-menopausaloestrogen replacement may reducecardio- and cerebrovascular eventsand prevent osteoporosis, but there isan increased risk of endometrial andb r e a s t cance r . 2 2 P r o s p e c t i v erandomised tr ials are awaited toclar i fy the benefits of oestrogenreplacement, antioxidant vitamins,dietary potassium, or polyunsaturatedfatty acids in stroke prevention.16

Antithrombotic therapy

Anticoagulation

Prophylactic anticoagulation forpa t i en t s wi th a high risk ofcardioembolism has been discussed(see Table 3).

Antiplatelet agents

Use of aspirin in the primaryp reven t ion of stroke r e m a i n scontroversial among healthy subjects.Studies of healthy subjects takingaspirin of various doses have reported

(Continued on page 74)

72

Primary and Secondary Prevention of Stroke

UPDATE ARTICLE

no benefit in the risk of ischaemicstroke, strokes of all types, fatalstroke, or fatal vascular events.4 '14

While low-dose aspirin protectsagainst first myocardial infarctions,there may be a small increase in therisk of intracerebral haemorrhage.

Meta-analysis of data from largeclinical trials on antiplatelet agentsshows that aspirin reduces vascularmortali ty by 15% and non-fatalv a s c u l a r e v e n t s ( s t r o k e a n dmyocardial infarction) by 30%.23'24

The dosage is less important; 300 mgdaily is as effective as 1.3 gm daily,but produces less gastrointestinalupset.25

Two major multicentre trials26'27

have shown that ticlopidine 250 mgtwice daily is as effective as aspirinin secondary stroke prophylaxis. It isnot ulcerogenic but may cause skinrash, leucopenia, aplastic anaemia andb l e e d i n g t endency , and rarelycho les ta t i c j a u n d i c e or raisedtransaminases. In general, it is givento pa t ien ts who cannot tolerateaspirin.

Carotid artery stenosis

Ext rac ran ia l ca ro t id ar terystenosis is an established risk factorfor symptomatic and silent stroke.T h e i m m e d i a t e p a t h o g e n i cmechanisms for cerebrovascularischaemic symptoms and/or signs incarotid artery stenosis are two:thromboembolic (especially with anulcerated plaque) and haemodynamic(when there is near or total occlusion,especially with poor collaterals and/or systemic hypotension). The risk

for stroke is non-uniform but affectedby factors such as occurrence ofsymptoms , degree of s t enos i s ,presence of ulceration, contralateralocclusion, presence of collaterals,number of other risk factors andcerebral infarction on computedtomography of the brain.28 Standardcontrast angiography remains the goldstandard for p rov id ing accurateimages of the carotid arteries, theproximal vessels and the intracranialvascula ture . Diameter stenosis(comparing the diameter of theinternal carotid artery above thestenosis as the denominator with thediameter of the narrowest part as thenumerator) should be used.29 This isthe method used in the Nor thAmerican Symptomatic CarotidEndarterectomy Trial. Contemporarydata on the prevalence of carotidartery stenosis is not available inHong Kong Chinese. In addition, therisk of first and recurrent strokes isu n k n o w n in local pa t ien ts wi thmoderate or severe carotid arterystenosis.

Carotid artery stenosis can becorrected by carotid endarterectomyor p e r c u t a n e o u s t r a n s l u m i n a langioplasty with or without stenting.Carotid endarterectomy carries as igni f icant perioperative risk ofmorbidity and mortality of 5 to 7% ormore.29 The preoperative carotidarteriography also contributes anadditional risk to the procedure.

Patients with severe asymptom-atic carotid artery stenosis have a 1.5to 2% annual stroke rate.4 Results offour randomised clinical trials ofcarotid endarterectomy in patientswith severe asymptomatic carotid

disease were published.14'30 Only theAsymptomatic Carotid Atherosclero-sis Study found a marginal benefitf rom surgery in a symptomat icstenosis of 60% or more by diameter:1% per year absolute reduction instroke risk over a five-year projectedperiod despite a remarkably lowperioperative risk of 2.3%.30 Never-theless, there was no reduction in themajor strokes or deaths. There wasno benefit in females, and there wasno relation between the benefit andthe degree of stenosis. At present,scientific data do not unequivocallysupport carotid endarterectomy as aneffective way of pr imary strokeprevention in patients with asymp-tomatic carotid artery stenosis of anyseverity so the prophylactic measuresare antithrombotic therapy with low-dose aspirin and modification of riskfactors.29

In gene ra l , p a t i e n t s w i thtransient ischaemic attacks related tosevere (70 to 99% by diameter)carotid artery stenosis have a 12 to13% rate of stroke in the first yearand a cumulative rate of 30 to 35% infive years, whereas those presentingwith strokes have a 5 to 9% annualrate of recurrence and a five-year rateof 25 to 45%.29 After successfulcarotid endarterectomy, patients whopresented with transient ischaemicattacks continue to have a 1 to 2%annual risk of recurrent stroke, whilea presenting history of stroke carriesa 2 to 3% per year rate of subsequentstroke. Surgery has been proven to beinferior to the best medical manage-ment if the symptomatic stenosis isless than 30%.29 Contemporaryguidelines of carotid endarterectomyare summarised in Table 5.

74

Hong Kong Practitioner 20 (2) February 1998

UPDATE ARTICLE

Table 5: Contemporary guidelines of carotid endarterectomy*

Category Carotid endarterectomy

Symptomatic 1.

2.

3.

4,

Indicated in patients with 70 to 99% stenosis if

peri-operative morbidity and mortality rate <6%

Not indicated in patients with < 70% stenosis

even if surgical risk <6%

Not indicated when surgical risk approaches

10% irrespective of degree of stenosis

Considered acceptable in patients with < 70%

stenosis in an ongoing prospective contemporary

randomized trial plus <6% surgical risk

Asymptomatic 1. Not definitely indicated in any patients

2. Considered acceptable in patients with > 75%

stenosis plus < 3% surgical risk

3. Not indicated when surgical risk is >3%

: : :; irrespective of degree of stenosis

* degree of stenosis as defined by the North American Symptomatic Carotid

Endarterectomy Trial

The o p t i m a l s t r a t e g y fo rmanaging patients with combinedcoronary artery and carotid diseaseremains unresolved. Meta-analysis ofthe available reports indicates thatcombined carot id and coronarysurgery carr ies the same peri-operative stroke rate (6.2%) as whencarotid endarterectomy precedescoronary artery bypass g ra f t ing(stroke rate = 5.3%) and that thestroke rate is h ighes t (10.0%) ifcarotid surgery follows coronarys u r g e r y . 2 9 N e v e r t h e l e s s , t h eprobability of myocardial infarction(11.5%) and death (9.4%) is greaterwhen carotid surgery precedescoronary artery bypass grafting thanwhen the order is reversed (myocar-

dial infarction rate = 2.7% and deathrate - 3.6%). Simultaneous surgeryhas a 4.7% probability of myocardialinfarction and a 5.6% probability ofdeath. Thus, a prospective ran-domised trial is needed to resolve thecontroversy. Meanwhile, combinedcarotid and coronary surgery appearsto be acceptable for patients in whomboth procedures are of proven benefit.

P e r c u t a n e o u s t r a n s l u m i n a langioplasty with or without stentingin cerebral artery stenosis is at theexperimental stage.14 This procedurecarries the risk of cerebral or retinalembolization. While prospective dataare being collected to build up oure x p e r i e n c e on the p rocedure ,

randomised control trials are awaitedto confirm its usefulness.

Conclusion

Recent clinical studies haveprovided important data to guideprimary and secondary prevention ofstroke. Family physic ians cantherefore play an important role instroke prevention. They are in abetter position than specialists inscreening and identifying risk factorsof stroke, promoting a healthy life-style and providing routine manage-ment of modifiable risk factors. Inmost patients with ischaemic stroke,the pathogenic mechanism is athero-thrombosis, and antiplatelet therapy isuseful in secondary prevent ion .Family physicians should be able tomonitor the long-term antiplatelettherapy, whereas complicated orunusual cases should be referred toneurologists. Anticoagulation isindicated for prevention of cardi-oembolic stroke provided contraindi-cations are absent. Since long-termanticoagulation is a complex issue, itw o u l d be b e t t e r h a n d l e d byspecialists. If the perioperative riskis acceptable, carotid endarterectomyfor severe stenosis is beneficial forsymptomatic patients but is contro-versial in asymptomatic patients.Patients with symptomatic carotidartery stenosis should be referred tospecialists for assessment as carotidendarterectomy or angioplasty mayhave to be considered. Finally,ongoing clinical trials will helpclarify the uncertain areas in strokeprevention. •

(Continued on page 77)

75

Hong Kong Practitioner 20 (2) February 1998

UPDATE ARTICLE

1. Stroke is a major neurological disease and a leading cause of death and disability. Family physicians play an important

role in stroke prevention.

2. Stroke is a heterogeneous syndrome, and specific therapy varies according to the clinical setting and risk factor profile

of individuals.

3. Epidemiological and cohort studies have identified hypertension, atrial fibrillation and other types of cardiac diseases,

diabetes mellitus, dyslipoproteinaemia, smoking and alcohol abuse, obesity and lack of exercise as the modifiable

risk factors for stroke.

4. Antithrombotic therapy should be considered in people at risk of cerebral ischaemia. Antiplatelet therapy is effective

in secondary prevention of atherothrombotic stroke. Anticoagulation prevents cardioembolic stroke, but

contraindications should be noted.

5. Corrective procedures are available for carotid artery stenosis. If the perioperative risk is acceptable, carotid

endarterectomy for severe stenosis is beneficial for symptomatic patients but is controversial in asymptomatic patients.

6. Complicated or unusual cases should be referred to neurologists.

References

1. The N a t i o n a l I n s t i t u t e of N e u r o l o g i c a lDisorders and Stroke rt-PA Stroke StudyGroup. Tissue plasminogen activator for acuteischemic stroke. N Engl J Med I995;333:1581-1587.

2. Kay R, Wong KS, Yu YL, et al. Low-molecular-weight heparin for the treatment ofacute ischemic stroke. N Engl J Med 1995;333:1588-1593.

3. H u a n g CY, Chan FL, Yu YL, el al.Cereb rovascu l a r d i sease in Hong K o n gChinese. Stroke 1990;21:230-235.

4. Feinberg WM. Primary and secondary strokeprevention. Curr Opin Neurol 1996;9:46-52.

5. Collins R, Peto R, MacMahon S, et al.Blood pressure, stroke, and coronary heartdisease. Part 2, short-term reductions in bloodpressure: overview of randomised drug trialsin their epidemiological context. Lancet1990;335:827-838.

6. SHEP C o o p e r a t i v e R e s e a r c h G r o u p .Prevention of stroke by antihypertensive drugtreatment in older persons with isolatedsystolic hypertension. JAMA 1991;265:3255-3264.

7. Huang CY, Yu YL, Woo KW, et al.Cerebral haemorrhage in a urban Chinesepopulation. Functional Neurol 1986;1:213-221.

8. Marmot MG, Poul te rNR. Primary preventionof stroke. Lancet 1992:339:344-347.

9. Warlow C. Secondary prevention of stroke.Lancet 1992;339:724-726.

10. Davis PH, Hachinski VC. The cardiac factorin stroke. Curr Opin Neurol 1992;5:39-43.

11. Feinberg WM, Blackshear JL, Laupacis A, etal. Prevalence, age distribution, and genderof patients with atrial f ib r i l l a t ion . Analysisand implications. Arch Intern Med 1995;155:469-473.

12. Feinberg WM, Seeger JF, Carmody RF, etal. Epidemiologic features of asymptomaticcerebral infarction in patients with nonvalvularatrial fibrillation. Arch Intern Med 1990; 150:2340-2344.

13. Hart RG. Halperin JL. Atrial fibrillation andstroke - revisiting the di lemmas. Stroke1994:25:1337-1341.

14. Barnett HJM, Eliasziw M, Mcldrum HE.Drugs and surgery in the p reven t ion ofischemic stroke. N Engl J Med 1995;332:238-248.

15. Solomon DH. Hart RG. Ant i thrombot ictherapies for stroke prevention. Curr OpinNeurol 1994:7:48-53.

16. BronnerLL, KanterDS, Manson JE. Primaryprevention of stroke. N Engl J Med 1995;333:1392-1400.

17. Lam KSL, Ma JTC, Woo E, et al. Highprevalence of undiagnosed diabetes amongChinese pa t ien ts with ischaemic stroke.Diabetes Res Clin Pract 1991;14:133-138.

18. Longstreth WT, Nelson LM, Koepsell TD, etal. Cigarette smok ing , a l c o h o l use andsubarachnoid hemorrhage. Stroke 1993;23:1242-1249.

19. Shinton R, Beevers G. Meta-analysis ofrelation between cigarette smoking and stroke.BMJ 1989:298:789-794.

20. Havel RJ, Rapaport E. Management ofprimary hyperlipidemia. N Engl J Med 1995;332:1491-1498.

21. Olshansky SJ, Carncs BA, Cassel CK. Theaging of the human species. Sci Am 1993;268:46-52.

22. Petitti DB, Sidney S, Bernstein A, et al.Stroke in users of low-dose oral contracep-tives. N Engl J Med 1996;335: 8-15.

2 3 . A n t i p l a t e l e t T r i a l i s t s ' C o l l a b o r a t i o n .Secondary prevention of vascular disease byprolonged antiplatelet treatment. BMJ 1988;296:320-331.

24 . A n t i p l a t e l e t T r i a l i s t s ' C o l l a b o r a t i o n .Collaborative overview of randomised trials ofantiplatelet therapy - I: Prevention of death,myocardial infarction, and stroke by prolongedantiplatelet therapy in various categories ofpatients. BMJ 1994:308:81-106.

25. UK-TIA Study Group. The United Kingdomtransient ischaemic attack (UK-TIA) aspirintrial: f inal results . J Neurol NeurosurgPsychiatry 1991;54:1044-1054.

26. Gent M, Blakely JA, Easton JD, et al. TheCanadian American Ticlopidine Study (CATS)in thromboembolic stroke. Lancet 1989;1:1215-1220.

27. Hass WK, Easton JD, Adams HP, et al. Arandomised t r i a l compar ing t ic lop id inehydrochloride with aspirin for the preventionof stroke in high-risk patients. N Engl J Med1989:321:501-507.

28. Barnet t HJM. Status report on the NorthAmerican Symptomatic Carotid Surgery Trial.J Mal Vasc 1993:18:202-208.

29. Moore WS, Barnett HJM, Beebe HG, et al.Guidel ines for carotid endarterectomy - amultidisciplinary consensus statement from theAd Hoc C o m m i t t e e , A m e r i c a n H e a r tAssociation. Stroke 1995:26:188-201.

30. Executive Committee for the AsymptomaticCarotid Atherosclerosis Study. Endarterec-tomy for asymptomatic carotid artery stenosis.JAMA 1995:273:1421-1428.

77