Prevention and Management Issues in Stroke

December 31, 2006

Q:Based on the latest evidence, what constitutesoptimal management and follow-up for patientswith asymptomatic carotid stenosis?

Q:Based on the latest evidence, what constitutesoptimal management and follow-up for patientswith asymptomatic carotid stenosis?A:The long-term risk of stroke from asymptomaticstenosis is about 1% per year for stenoses of 60%or more, and less than 1% per year for stenoses of lessthan 60%. These risks remain stable during long-term follow-up.1 An asymptomatic bruit may indicate stenosis; however,many persons with carotid bruits do not have stenosis.Normally, the first step in follow-up would be carotidultrasonography, magnetic resonance angiography, orcomputed tomoangiography. If those procedures do notindicate a significant stenosis, no further evaluation is required.However, if 2 noninvasive procedures confirm ahigh-grade stenosis and the patient is a good surgical candidate,carotid endarterectomy (CEA) is now frequentlyperformed without previous angiography, simply becausethe latter procedure is associated with about a 1% risk ofstroke.If one of the noninvasive studies suggests the presenceof a high-grade stenosis that is not corroborated by anothernoninvasive study, conventional angiography may be indicatedfor clarification. However, it is generally preferableto avoid angiography, given the modest benefit of CEA inasymptomatic patients.Q:Would you be more specific about the benefitsand risks of CEA in asymptomatic patients?A:The Asymptomatic Carotid Atherosclerosis Study(ACAS)-the most definitive trial to date on this topic-showed that among asymptomatic patients, the optimalcandidate for CEA had carotid stenosis of 60% or more, wasrelatively young and healthy, was at low risk for complicationsfrom surgery, and had a surgeon with an excellenttrack record.2 Such candidates can derive significant benefitfrom surgery. The study didn't show any clear associationbetween the degree of stenosis and the degree of benefit.However, the authors found that even in patients whowere good surgical candidates, CEA was only slightly superiorto medical management. The gain was especiallymodest in elderly patients. (The absolute benefit is about 1%per year for 5 years compared with medical management.)In Europe, surgeons rarely perform CEA on persons withasymptomatic stenoses-even high-grade ones-becauseof the relatively modest benefit. Therefore, many cliniciansdo not aggressively recommend CEA for asymptomatic patients,particularly in view of its potential complications,which include stroke.Subgroup analysis of data from the ACAS studyshowed that men benefited from CEA more than women.In women, this surgery was not clearly more effective thanmedical therapy. However, the study was not powered to detecta benefit in women. Some clinicians have inferred thatthey should be a bit less aggressive with women. However,a final decision should be made based on the patient'srisk factor status.Q:Is CEA best reserved, then, for symptomaticpatients with a high-grade stenosis and atleast a 5-year life expectancy?A:Symptomatic patients with high-grade stenosis havea much greater benefit from surgery than those witha lower-grade stenosis. The degree of benefit from surgerygoes up substantially with increasing degrees of stenosis because the stroke risk with medical therapy is much higherwith higher-grade stenosis.3Q: Most guidelines state that thrombolytictherapy can be used up to 3 hours afteracute ischemic stroke. Do you agree with thisrecommendation?A:At present, the only FDA-approved treatment ofacute ischemic stroke is based on the National Instituteof Neurological Disorders and Stroke (NINDS) protocol,which used intravenous alteplase, a tissue plasminogenactivator (tPA).4 Most guidelines recommend criteriavery similar, if not identical, to the protocol used in thatstudy. However, this study enrolled patients up to 3 hoursafter symptom onset. Subsequent analysis has shown thatthe greatest benefit accrues to patients who are treatedwithin 90 minutes. The benefit at 3hours appears to be relatively smallfor most patients.One of the biggest concerns isthat patients are being treated towardthe end of the 3-hour window. In a recenttrial (the Standard Treatmentwith Alteplase to Reverse Stroke[STARS] study), my colleagues and Ievaluated the use of alteplase in communityand academic centers that hadstroke units.5 We found that most patientswere treated in the last 15 minutesof the 3-hour window and thatmany were being treated slightly beyondthat point. This is not becausethe patients all presented at the endof the window. Even for patients whopresented within 30 minutes of symptom onset, physiciansfrequently took up to 2/? hours to start treatment. If a patientpresented 2/? hours into the stroke, the clinicianssometimes initiated therapy within 30 to 40 minutes. Theprevailing notion seemed to be that it was acceptable touse the available time up to 3 hours-because alteplaseis approved for use up to 3 hours. But this is not a goodstrategy.Studies such as the one by Marler and colleagues6clearly show that the significant benefit seen early on tapersoff rapidly during the third hour after stroke onset. My colleaguesand I are advocating that physicians do whateverthey can to shorten the door-to-needle time.Of course, if a patient comes in at 2/? hours and can betreated within 30 minutes, he or she should still be treated.Future guidelines will emphasize the greater benefit withearlier treatment.Q:How significant is the risk of intracerebralhemorrhage (ICH) with tPA?A:The Cleveland study showed a higher rate of ICH(about 15%) with tPA when all hospitals in the areawere considered than in clinical trials (about 6%).7 However,one of the authors of that study indicated recently thatafter additional training of community hospital personnel,the study was repeated and the rate of ICH dropped byabout half.It makes sense that with increased training, hospitalsand stroke centers can achieve acceptable ICH rates of 6%or less. In the STARS study, the ICH rate in stroke centerswas just 3.3%. In the Canadian Activase for Stroke EffectivenessStudy (CASES), which was much larger thanSTARS, the ICH rate was 4%.8Most hospitals do not have personnel trained to administertPA; consequently, hospitalsare not treating nearly the number ofpatients they could. (Fewer than 2% ofstroke patients are treated with tPA ingeneral hospitals. The number is higherin most stroke centers.) When patientsdo undergo thrombolysis, mistakesare often made; for example,the medication may be administeredmuch later than would be optimal orthe wrong dose may be administered.It is not realistic to expect thatevery hospital will be able to administertPA. The ideal would be a systemthat certifies stroke centers at certainhospitals within a city or region. Paramedicscould then take patients withpresumed stroke to those centers.Currently, in most areas, paramedics take stroke patients tothe closest hospital-even if that hospital has seldom ornever used tPA. These patients may be more likely to beharmed than helped in this setting if tPA is used.It's appalling, actually, that we've had approved, effectivestroke therapy for 6 years now and it is still being usedvery inefficiently. A new study has confirmed higher ratesof in-hospital mortality and serious extracranial hemorrhageassociated with major protocol deviations.9Q: For which patients is thrombolytic therapyinappropriate?A:Most physicians choose not to treat nursing homeresidents who are extremely elderly, debilitated, ordisabled. Contraindications to tPA use include blood pressurehigher than 185/110 mm Hg, recent stroke or surgery, and bleeding disorder. A subgroup analysis of the NINDSdata on patients older than 75 years who had had majorstrokes-the group with the worst prognosis-revealedthat these patients appeared to derive modest benefit. Manywere left with mild to moderate disability, rather than a severedisability, and were able to return home rather than beadmitted to a long-term-care facility. The benefit in older patientswith severe stroke is much less dramatic than inyounger patients who experience a moderate stroke, manyof whom recover fully.There is no absolute upper age limit for thrombolytictherapy. We've all seen patients who are active and in goodshape at age 90. These patients may benefit substantiallyfrom treatment if they have a stroke.Q:What constitutes a cost-effective workup forpersons with suspected TIA?A:A:In a very good recent study, Johnston and associates10found that half of the strokes that occurred inthe first few months after a TIA actuallyoccurred in the first 2 days. Consequently,an expedient workup is vital.In some health care systems, patientswith a TIA can have ultrasonographydone the same day they areseen in the ED. But in other systems,an outpatient workup can take up to aweek or more; in this case, I believe itis best to consider admitting the patientfor a 24-hour short stay for evaluationof a TIA.You want to ascertain if the patient has a high-gradecarotid stenosis, because a TIA with a stenosis greater than70% is a clear indication for CEA. (With stenosis of 50% to69%, there's only a small benefit from surgery, so it's not ascritical to act urgently with these patients.) I also recommendthat you determine whether the patient has a highriskcardiac source of embolism and whether he will requireechocardiography. You also want to order routine laboratorystudies, identify risk factors that need treatment,and look for unusual medical conditions that may havecaused the TIA.The requirement for brain imaging is controversial.Most clinicians choose this option, however. Although thechance of finding an abnormality on a CT scan in a patientwith a TIA is low, sometimes other conditions that canmimic a TIA-such as a subdural hematoma or a tumor-are identified. CT may also show evidence of previousstrokes or TIAs.CT is the least expensive modality. MRI offers a higherdegree of resolution, which can help identify small ischemic lesions or detect TIA mimics. MRI also allows examinationof the intracranial circulation and carotid arteries. CT dye,which is required for imaging of blood vessels, is slightlymore toxic than the dye used to image vessels with MRI.Q:How do you explain the failure of manypromising therapies for acute stroke? Andwould you comment on the newer imaging techniques?A:One of the main reasons for the failure of neuroprotectiveagents is that brain tissue dies within thefirst few hours after stroke onset. The only class of drugthat's ever been tested in an extremely early time window-ie, within 3 hours of stroke onset-is tPA. An NIHstudy has been proposed by the UCLA Stroke Center toevaluate the use of a neuroprotective agent by paramedicsin the ambulance.A number of drugs have been effective in animal modelsof stroke, but generally only within the first 2 hours ofstroke onset. My colleagues and I are doing a study involvingtPA that uses new MRI techniquesto try to identify patients whose braintissue can be salvaged at 3 to 6 hours.The rate at which brain tissue diesvaries from patient to patient, dependingon which blood vessel is occludedand how efficient the collateral circulationis. Diffusion- and perfusionweightedMRI provides importantpathophysiologic information in acutebrain ischemia and can help differentiatesalvageable ischemic brain tissuefrom irreversibly injured tissue. A recent Australian studyshowed that a group of patients whose treatment was guidedby this MRI strategy had very good outcomes with tPA,even though many of these patients were treated beyondthe 3-hour window.11Q:Would you comment on advances in secondaryprevention of stroke?A:We've had much more success in secondary preventionthan in primary prevention. We now have severalclasses of medications and usually several choices withineach class that are effective for patients who have had a TIAor a stroke and those who are at very high risk because ofother vascular risk factors.Although most primary prevention studies have focusedon MI, secondary outcomes of studies of therapiesused primarily for MI prevention have shown that these interventionsmay also help prevent stroke. A number ofstudies of statins have shown substantial reductions in the incidence of stroke. Important trials of angiotensin-convertingenzyme (ACE) inhibitors-including the Heart OutcomesPrevention Evaluation Study (HOPE)12 and thePerindopril Protection Against Recurrent Stroke Study(PROGRESS)13-have shown that these agents effectivelylower the risk of stroke or stroke recurrence. The recentLosartan Intervention For Endpoint reduction in hypertension(LIFE) study found substantial stroke reduction withangiotensin II receptor blockade.14 Extended-release dipyridamolein combination with aspirin is effective for secondarystroke prevention.15 Clopidogrel has shown somebenefit over aspirin. A major stroke study-the Managementof Atherothrombosis with Clopidogrel in High-RiskPatients with Recent Transient Ischaemic Attack or IschaemicStroke (MATCH) trial-is now under way to evaluatethe combination of clopidogrel plus aspirin in secondaryprevention.Q:What prophylactic measures do you recommendto reduce the likelihood of stroke in patientsat risk?A:All patients at risk for stroke-regardless of whetherthey undergo surgery or whether they have high- orlow-grade stenosis-are good candidates for risk factor control(Table). Recent data suggest that very aggressive controlof blood pressure (to below 120/80 mm Hg) reducesthe risk of stroke.16 Antiplatelet therapy-usually aspirin-is also recommended. Smoking cessation, control of diabetes,exercise, and weight control are beneficial. A numberof studies have shown that lipid-lowering therapy reducesstroke rates, although this hasn't been specifically studiedin patients with asymptomatic carotid stenosis.Of course, a primary stroke prevention study would requirethat the participants be elderly persons with vascular risk factors. To date, the focus in such trials has been theprevention of MI, not stroke. Nevertheless, one of the greatsuccess stories in primary prevention of stroke is anticoagulationtherapy for atrial fibrillation.Q:What are your thoughts about blood pressurecontrol in elderly persons who are at risk forstroke or who have already had a stroke?A:Many physicians are cautious about controlling bloodpressure aggressively in older persons because, forexample, they fear that if a patient becomes hypotensive, hemight fall and break a hip. But in fact, more strokes are attributableto hypertension than any other risk factor. Aggressivecontrol of blood pressure is therefore absolutelyessential.Clinical trials consistently show that the lower theblood pressure, the lower the risk of stroke. ThePROGRESS study compared combination ACE inhibitorplus diuretic therapy with placebo in patients with a historyof stroke or TIA. Even in patients with relatively normalblood pressure after the event, combination therapy reducedthe pressure substantially-12 mm systolic on average-and lowered stroke risk by 43%. Patients with thegreatest blood pressure reductions had the greatest strokerisk reductions. This was true even for the oldest patientsin the study.With newer antihypertensive medications-such asangiotensin receptor blockers-that have fewer side effects,it should be relatively simple to treat hypertensive elderlypatients with agents they can tolerate and get their bloodpressure down toward 120/80 mm Hg. The idea that ablood pressure measurement of 140/90 mm Hg can be considerednormal-particularly in elderly patients-has notbeen borne out in recent trials. Persons whose pressure isat this level have much higher event rates than those withlower pressure. A recent study showed that "high-normal"pressure (systolic pressure of 130 to 139 mm Hg, diastolicpressure of 85 to 89 mm Hg) was associated with a muchhigher risk of cardiovascular events than pressure lowerthan 120/80 mm Hg.16Q:Is a higher dose of a single drug preferable tolower doses of a combination?A:As the PROGRESS study showed, many patients getbetter results with combination therapy. In that study,there was no discernible reduction in the risk of stroke inpatients who took only a single drug.In clinical practice, however, the decision for single versuscombination therapy depends on the adverse effects. Ifa higher dose of a single agent controls blood pressure without sequelae, this would usually be the choice. However,most clinical trials are not designed for single-drug dosetitration; if a single agent does not reduce pressure, a seconddrug is usually added early on. In actual practice, mostclinicians usually raise the dose of a single agent until adverseeffects develop or maximum response is attained. Ifthe pressure is still not controlled, a second drug must beadded.




Nadareishvili ZG, Rothwell PM, Beletsky V, et al. Long-term risk of strokeand other vascular events in patients with asymptomatic carotid artery stenosis.

Arch Neurol

. 2002;59:1162-1166.


Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomyfor asymptomatic carotid artery stenosis.


. 1995;273:1421-1428.


Barnett HJ, Taylor DW, Eliasziw M, et al, for the North American SymptomaticCarotid Endarterectomy Trial Collaborators. Benefit of carotid endarterectomy inpatients with symptomatic moderate or severe stenosis.

N Engl J Med

. 1998;339:1415-1425.


The National Institute of Neurological Disorders and Stroke (NINDS) rt-PAStroke Study Group. Tissue plasminogen activator for acute ischemic stroke.

N Engl J Med

. 1995;333:1581-1587.


Albers GW, Bates VE, Clark WM, et al. Intravenous tissue-type plasminogenactivator for treatment of acute stroke: the Standard Treatment with Alteplase toReverse Stroke (STARS) study.


. 2000;283:1145-1150.


Marler JR, Tilley BC, Lu M, et al. Early stroke treatment associated with betteroutcome: the NINDS rt-PA stroke study.


. 2000;55:1649-1655.


Katzan IL, Furlan AJ, Lloyd LE, et al. Use of tissue-type plasminogen activator foracute ischemic stroke: the Cleveland area experience.


. 2000;283:1151-1158.


Hill MD, Buchan AM, for the CASES Investigators. Methodology for theCanadian Activase for Stroke Effectiveness Study (CASES).

Can J Neurol Sci



Bravata DM, Kim N, Concato J, et al. Thrombolysis for acute stroke in routineclinical practice.

Arch Neurol

. 2002;162:1994-2001.


Johnston SC, Gress DR, Browner WS, Sidney S. Short-term prognosis afteremergency department diagnosis of TIA.


. 2000;284:2901-2906.


Parsons MW, Barber PA, Chalk J, et al. Diffusion- and perfusion-weightedMRI response to thrombolysis in stroke.

Ann Neurol

. 2002;51:28-37.


Yusuf S, Sleight P, Pogue J, et al, for the Heart Outcomes Prevention EvaluationStudy Investigators. Effects of an angiotensin-converting enzyme inhibitor, ramipril,on cardiovascular events in high-risk patients.

N Engl J Med.



PROGRESS Collaborative Group. Randomised trial of a perindopril-basedblood-pressure-lowering regimen among 6,105 individuals with previous strokeor transient ischaemic attack.




Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity andmortality in the Losartan Intervention For Endpoint reduction in hypertensionstudy (LIFE): a randomised trial against atenolol.


. 2002;359:995-1003.


Sivenius J, Cunha L, Diener HC, et al, for the ESPS2 Working Group. SecondEuropean Stroke Prevention Study: antiplatelet therapy is effective regardless ofage.

Acta Neurol Scand.



Vasan RS, Larson MG, Leip EP, et al. Impact of high-normal blood pressureon the risk of cardiovascular disease.

N Engl J Med.



Pearson TA, Blair SN, Daniels SR, et al. AHA guidelines for primary preventionof cardiovascular disease and stroke: 2002 update.