- Email: [email protected]
Leucoaraiosis and silent subcortical infarcts
revue neurologique 164 (2008) 801–804
Disponible en ligne sur www.sciencedirect.com
Re´union commune de la SFN et de la SFNV, 8–10 octobre 2008
Leucoaraiosis and silent subcortical infarcts Leucoaraiose et infarctus silencieux subcorticaux B. Norrving Department of Neurology, University Hospital, 22185 Lund, Sweden
article info
abstract
Available online 28 August 2008
Silent brain infarcts and leucoaraiosis are frequently observed in patients with transient ischemic attack (TIA) and ischemic stroke. Patients with silent brain infarcts and leucoa-
Keywords:
raiosis at baseline are a high-risk group with an increased long-term risk for recurrent
Silent brain infarcts
stroke, cognitive decline, and dementia. Effects on short-term outcomes are less clear, but
Leucoaraiosis
leucoaraiosis appear to increase the risk of early infarct growth as determined by MRI in
Stroke
mismatch areas of the brain. After acute thrombolytic therapy, presence of silent cerebro-
Thrombolytic therapy
vascular disease increases the risk of intracerebral hemorrhage, but the increased risk does
Anticoagulants
not negate the overall benefit of this therapy. Similarly, presence of leucoaraiosis is
Blood pressure
associated with an increased risk of intracerebral bleeding during long-term anticoagulant therapy, but because the risk–benefit ratio has not been well delineated, degree of leucoa-
Mots cle´s :
raiosis should not influence clinical decision-making at present. Carotid endarterectomy for
Infarctus ce´re´braux silencieux
symptomatic carotid stenosis appears equally effective across different degrees of baseline
Leucoaraiose
leucoaraiosis, despite an increased perioperative risk in those with moderate to severe
Infarctus ce´re´bral
leucoaraiosis. Long-term blood pressure lowering appears equally effective in patients with
He´morragie ce´re´brale
silent cerebrovascular disease at baseline, and there is some support that blood pressure
Thrombolyse
lowering may decrease the risk of progression of leucoaraiosis.
Anticoagulants
# 2008 Elsevier Masson SAS. All rights reserved.
Pression arte´rielle
r e´ s u m e´ Les infarctus silencieux et la leucoaraiose sont fre´quemment observe´s chez des patients ayant des accidents ische´miques transitoires ou des infarctus ce´re´braux. Les patients qui pre´sentent des infarctus silencieux ou une leucoaraiose sont des patients a` haut risque de re´cidive, de de´clin cognitif et de de´mence. Les effets sur le pronostic a` court terme ne sont pas clairement identifie´s, mais il semblerait que la leucoaraiose augmente le risque de croissance pre´coce de la zone infarcie en imagerie par re´sonance magne´tique, au sein de la zone de mismatch. En cas de thrombolyse, la pre´sence de ces le´sions silencieuses augmente le risque d’he´morragie ce´re´brale, mais n’alte`re pas le be´ne´fice du traitement. La pre´sence d’une leucoaraiose est associe´e a` un risque plus e´leve´ d’he´morragie ce´re´brale au cours d’un traitement anticoagulant. Toutefois, le rapport be´ne´fice–risque n’ayant pas e´te´ e´value´, il n’est donc pas justifie´ de tenir compte de la leucoaraiose pour la de´cision d’anticoaguler un patient. La chirurgie carotidienne pour ste´nose symptomatique est
E-mail address: [email protected]. 0035-3787/$ – see front matter # 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.neurol.2008.07.009
802
revue neurologique 164 (2008) 801–804
associe´e a` un risque pe´riope´ratoire plus e´leve´ en pre´sence d’une leucoaraiose se´ve`re. La re´duction de la pression arte´rielle pourrait dans certains cas ralentir la progression de la leucoaraiose. # 2008 Elsevier Masson SAS. All rights reserved.
1.
Introduction
The introduction of modern neuroimaging techniques allowed the identification of silent cerebrovascular disease during life. Silent brain infarcts (SBIs) were incidentally recognized during the 1980s in patients who underwent computerized tomography (CT) in the investigation of acute stroke. During the same decade, the frequency of abnormalities in the white matter were increasingly recognized on brain CT scans, and in 1986, Vladimir Hachinski and coworkers introduced the term leucoaraiosis (LA) (from the Greek leuko, white, and araiosis, rarefaction) to designate areas in the periventricular and centrum semiovale white matter that appeared hypodense on CT (ischemic white matter lesions). Pathological studies have linked LA with demyelination, gliosis, necrosis, and cavitation which are associated with atherosclerosis of small and large vessels. Similar findings are seen in arteriosclerotic vascular encephalopathy, or Binswanger disease. Silent cerebrovascular disease has subsequently been a topic of intense research efforts, and CT has largely been replaced by MRI, which is much more sensitive for this purpose. During the last one or two decades, SBIs and LA have evolved from innocent bystanders to small lesions of major importance for vascular and mental health in the population. LA and SBIs frequently coexist, and the two conditions share most risk factors and show a high degree of covariance. Studies of persons in the elderly general population have shown that silent cerebrovascular disease is strongly linked to cognitive impairment, dementia, and increased risk for stroke. Recently, genetical determinants of LA in the population have also been demonstrated. However, the prognostic and therapeutic implications of silent cerebrovascular disease in patients with transient ischemic attack (TIA) and ischemic stroke have been less well explored, and will be reviewed in this report.
2.
Methodological issues
There are several clinical caveats in designating SBIs as truly silent, i.e. not associated with previous acute focal neurological symptoms. Previous clinical symptoms may be forgotten by the patient, the clinical history may be incomplete or symptoms may have occurred, but were not recognized as possible stroke by the patient or physician. Furthermore, about half of all TIAs are associated with an infarct visualized by neuroimaging. Conceivably, a proportion of all TIAs may occur during sleep and pass unrecognized. However, in total, these caveats are likely to account only for a small proportion of all SBIs. Apparently, the majority of infarcts judged as silent were probably not associated with acute focal neurological symptoms, which were forgotten or misdiagnosed.
In patients with stroke and the finding of multiple ischemic areas on conventional imaging, it may sometimes be difficult or even impossible to determine which of the infarcts was symptomatic. Diffusion-weighted MRI (dw-MRI) is particularly helpful in separating acute lesions versus silent or remote. Dw-MRI has also demonstrated that ischemic abnormalities in patients with acute ischemic stroke are not always single: in unselected patient materials, multiple ischemic abnormalities within the territory of one major cerebral artery or in multiple arterial territories are seen in up to 40% of all patients. Large acute infarcts in one hemisphere may hamper evaluation of LA in the infarcted area: in such cases, assessment of LA needs to be limited to the contralateral hemisphere.
3. How frequent are LA and SBIs in patients with TIA and stroke? Several CT-based studies have documented that SBIs and LA are frequently seen in patients admitted with a first TIA or ischemic stroke. In the earlier studies, relatively low prevalences, between 11 and 13%, of SBI were reported, presumably related to the use of early generation CT equipments. In later studies (Jo¨rgensen et al., 1994; Corea et al., 2001), higher prevalences ranging from 23 to 38% were found. Large MRIbased studies of SBI in stroke patients are lacking. In patients with ischemic stroke, CT studies have reported a prevalence of LA ranging from 3 to 44%, whereas MRI-based studies have shown some degree of LA in up to 82% (Fu et al., 2005). Severe LA (grade 3/3) was present in about 30% of cases on MRI. As in healthy persons, SBIs in patients with TIA or stroke are most often small and deep, whereas cortical infarcts account for 10 to 24%, may more often involve the right hemisphere or the posterior cerebral artery territory, and may more often be linked to a cardioembolic cause. With respect to subtype of presenting stroke, SBIs appear to be more prevalent in patients with lacunar infarction than in those with atherothrombotic or cardioemblic infarcts. The presence of SBI and LA in stroke patients have been associated with a wide range of accepted cardiovascular risk factors, largely similar to those found in patients with symptomatic small vessel disease (lacunar infarcts).
4. Significance of SBIs and LA in the acute phase of stroke Several studies have reported that early morbidity and outcome were not influenced by the presence of SCIs or LA in cohorts of patients with stroke or TIA (Jo¨rgensen et al., 1994;
revue neurologique 164 (2008) 801–804
Jo¨rgensen et al., 1995; Wiszniewska et al., 2000; Corea et al., 2001). However, a recent study has provided interesting new insights on the effects of silent cerebrovascular disease in the very early phase of ischemic stroke. In a study of 61 patients with acute ischemic stroke, Ay et al. (2008) explored the relationship between the volume of white matter regions with LA on acute images and the proportion of dw-MRI imaging – mean transit time mismatch tissue that progressed to infarction (percentage mismatch lost). They found a significant correlation between percentage of mismatch lost and LA volume, suggesting that LA volume appears to be a composite predictive marker for the fate of acutely ischemic tissue. LA is associated with structural and functional vascular changes which may be more susceptible to tissue injury when infarct develops in adjacent areas. A similar mechanism may potentially explain the frequent occurrence of initially progressive deficits in patients with lacunar infarcts. These hypotheses warrant further study.
6.1.
Long-term prognostic significance
Whereas silent cerebrovascular disease has no or limited effects on early clinical outcomes, both SBIs and LA at baseline appear to have important effects on long-term prognosis. A recent study (Fu et al., 2005) found that patients with severe LA had increased risk of deep subcortical stroke and a higher risk of recurrent stroke. The prognostic associations with SBIs and LA have in particular been demonstrated in patients with lacunar infarcts, in whom silent cerebrovascular disease at baseline along with risk factors have been shown to be independent predictors of future stroke risk. Presence of SBIs and LA have also been found to be independent predictors of worse functional outcome, suggesting that more advanced small artery disease may limit the possibility of functional recovery of the brain after a small focal lesion. There is also mounting evidence that SCI and LA are an important risk factor for mild cognitive impairment and dementia after stroke, in particular in the longer perspective. Silent cerebrovascular disease is not static. As in the general population, SBIs and LA frequently progress with time also in patients with TIA and stroke. In patients with launar infacts, the subgroup in which this issue has been mainly studied, new silent infarcts develop at a rate of at least twice that of new symptomatic lacunae, and after about three years, almost half of all patients had developed SBIs as well as progression of LA. However, there are as yet no data on the clinical effects of progressing silent cerebrovascular disease with time in these patient groups.
6. Interactions with acute and long-term secondary preventive therapies There is a paucity of prospective randomized controlled clinical trial data on therapeutic issues with respect to silent cerebrovascular disease in patients with TIA and stroke. Existing data are mainly based on posthoc analyses that were not prespecified.
Thrombolytic therapy
Two studies (Neumann-Haefelin et al., 2006; Palumbo et al., 2007) have identified silent cerebrovascular disease as an independent risk factor for intracerebral hemorrhage after thrombolysis with rtPA in acute stroke. In the German multicenter study (Neumann-Haefelin et al., 2006), the rate of hemorrhage was 10.5% among the quarter of the total cohort who had moderate to severe LA versus 3.8% in those without, corresponding to an odds-ratio of 2.9. In the Canadian study (Palumbo et al., 2007), LA and multiple lacunes were both independent predictors for intracerebral hemorrhage (oddsratios 2.7 and 3.4, respectively), but no difference was noted for good outcome rate at 90 days. These observations should not influence the use of rtPA; several other factors have also been associated with increased bleeding rates after rtPA, but none of these have been clearly shown to reverse the overall benefit of rtPA.
6.2.
5.
803
Antiplatelet therapy
Although there is evidence that silent cerebrovascular disease at baseline affects prognosis, it is unknown if more aggressive secondary preventive approaches than those generally applied are warranted. Benefits of combining antithrombotic regimens may well be offset by increased adverse effects, as illustrated by the recent MATCH-trial of clopidogrel plus aspirin versus clopidogrel only. The ongoing Secondary Prevention of Small Subcortical Strokes (SPS3) study will provide more data on this issue; in a factorial design that also involves different blood pressure targets (see below), the trial is testing clopidogrel and aspirin as compared with aspirin alone in the prevention of stroke recurrence and cognitive decline in patients with recent symptomatic small subcortical stroke.
6.3.
Anticoagulants
In the SPIRIT trial, which compared warfarin with target INR values of 3.0 to 4.5 to aspirin in patients with TIA and ischemic stroke of presumed arterial origin, the presence of LA was associated with a markedly elevated hazard (9.2) for intracerebral hemorrhage. This effect remained significant when controlled for age, blood pressure, and INR. A small casecontrolled study has also suggested that LA may be an independent risk factor for intracerebral hemorrhage also in patients with commonly employed range of anticoagulation (INR 2.0 to 3.0). However, until further data become available on both the risk and benefits of anticoagulation in the presence of LA, these data are insufficient to incorporate silent cerebrovascular disease in the decision-making of individual patients. The exact mechanism by which anticoagulants and rtPA increase the risk of intracerebral hemorrhage in patients with silent cerebrovascular disease has not been determined.
6.4.
Carotid endarterectomy
A posthoc analysis of the NASCET study showed that in patients with symptomatic carotid artery disease the presence
804
revue neurologique 164 (2008) 801–804
of LA was associated with an increased risk of any stroke and of disabling or fatal stroke (Streifler et al., 2002). The 30-day perioperative risk was also increased, from 5.3% (no LA), to 10.6% (restricted LA) and 13.9% (widespread LA). However, despite higher perioperative risk, endarterectomy reduced the absolute three-year risk of ipsilateral stroke irrespective of degree of LA.
6.5.
Blood pressure lowering
Blood pressure control plays a major role both in primary and in secondary stroke prevention. However, theoretical concerns have been raised that too aggressive blood pressure lowering may have adverse effects in patients with advances small vessel disease and impaired autoregulation. A CT substudy of the PROGRESS trial on blood pressure lowering therapy in patients with stroke found no significant effect on therapy on the rate on SBIs over 3.9 years (Hasegawa et al., 2004). However, a MRI substudy showed that the blood pressure-lowering regimen stopped or delayed the progression of white matter lesions (Dufouil et al., 2005). The greatest beneficial effect of blood pressure lowering on LA was observed in the group of patients who had severe leucoaraiosis at study entry. In the ongoing SPS3 study, the role of lowering systolic blood pressure two different levels (< 130 mmHg versus < 140 mmHg) is tested in a factorial design.
7.
Conclusions
In patients with ischemic stroke and TIA, there is a growing amount of evidence that presence and extent of silent cerebrovascular disease at the time of the first event have significant prognostic implications in the long-term for almost all outcomes. Short-term effects on clinical outcomes have not been clearly demonstrated, but on MRI presence of severe LA appears to increase the susceptibility to early infarct growth. Silent cerebrovascular disease increases the risk of intracerebral hemorrhage after acute thrombolytic therapy as well during long-term anticoagulant therapy, but has not been clearly shown to reverse the overall benefit of these therapies. At present, presence of SBI and LA should therefore not be used to influence clinical decisions. Carotid endarterectomy appears equally effective in patients with and without LA. Theoretical concerns of detrimental effects of blood pressure lowering in the presence of cerebral small vessel disease has not been substantiated by clinical trials. Blood pressure lowering after TIA and ischemic stroke has not been linked to increased risk of progression of silent cerebrovascular disease. Blood pressure lowering appears effective across different stroke subtypes, including patients with SBIs and LA at baseline. Should the management of stroke patients be modified in the presence of SBIs or LA at the time of the index event? At present, there is no data to support that any of the established
acute or long-term secondary preventive therapies should be withheld for this reason. Furthermore, there are no data to support the use of more intense therapeutic regimens like combined use of antiplatelets, or more intense blood pressure lowering than what is generally recommended in clinical guidelines. However, given that the presence of SBIs and LA at baseline implies an increased long-term risk for recurrent stroke, cognitive decline and dementia, careful attention to risk factor control, life style modifications, and long-term compliance with secondary preventive measures is essential in this high-risk group.
references
Ay H, Arsava M, Rosand J, Furie KL, Singhal AB, Schaefer PW, et al. Severity of leucoaraiosis and susceptibility to infarct growth in acute stroke. Stroke 2008;39:1409–13. Corea F, Henon H, Pasquier F, Leys D. Silent infarcts in stroke patients: patient characteristics and effects on 2-year outcome. J Neurol 2001;248:271–8. Dufouil C, Chalmers J, Coskun O, Besancon V, Bousser MG, et al. Effects of blood pressure lowering on cerebral white matter hyperintensities in patients with stroke. The PROGRESS (Perindopril protection against recurrent stroke study) Magnetic resonance imaging substudy. Circulation 2005;112:1644–50. Fu JH, Lu CZ, Hong Z, Dong Q, Luo Y, Wong KS. Extent of white matter lesions is related to acute subcortical infarcts and predicts further stroke risk in patients with first ever ischaemic stroke. J Neurol Neurosurg Psychiatry 2005;76:793–6. Hasegawa Y, Yamaguchi T, Omae T, Woodward M, Chalmers J. PROGRESS CT Substudy Investigators. Effects of perindoprilbased blood pressure lowering and of patient characteristics on the progression of silent brain infarcts: the Perindopril Protection against Recurrent Stroke Study (PROGRESS) CT Substudy in Japan. Hypertens Res 2004;27:147–56. Jo¨rgensen HS, Nakayama H, Raaschou HO, Olsen TS. Silent infarction in acute stroke patients. Prevalence, risk factors, and clinical significance: the Copenhagen Stroke Study. Stroke 1994;25:97–104. Jo¨rgensen HS, Nakayama H, Raaschou HO, Olsen TS. Leucoaraiosis in stroke patients. The Copenhagen Stroke Study. Stroke 1995;36:588–92. Neumann-Haefelin T, Hoelig S, Berkfeld J, Fiehler J, Gass A, Humpich M, et al. Leucoaraiosis is a risk factor for symptomatic intracerebral hemorrhage after thrombolysis for acute stroke. Stroke 2006;37:2463–6. Palumbo V, Boulanger JM, Inzitari D, Buchan AM, on behalf of the CASES Investigators. Leucoaraiosis and intracerebral hemorrhage after thrombolysis in acute stroke. Neurology 2007;68:1020–4. Streifler JY, Eliaziw M, Benavente OR, Alamowitch S, Fox AJ, et al. Prognostic importance of leucoaraiosis in patients with symptomatic internal carotid artery stenosis. Stroke 2002;33:1651–5. Wiszniewska M, Devuyst G, Bogousslavsky J, Ghika J, van Melle G. What is the significance of leucoaraiosis in patients with acute ischemic stroke? Arch Neurol 2000;57:967–73.
Disponible en ligne sur www.sciencedirect.com
Re´union commune de la SFN et de la SFNV, 8–10 octobre 2008
Leucoaraiosis and silent subcortical infarcts Leucoaraiose et infarctus silencieux subcorticaux B. Norrving Department of Neurology, University Hospital, 22185 Lund, Sweden
article info
abstract
Available online 28 August 2008
Silent brain infarcts and leucoaraiosis are frequently observed in patients with transient ischemic attack (TIA) and ischemic stroke. Patients with silent brain infarcts and leucoa-
Keywords:
raiosis at baseline are a high-risk group with an increased long-term risk for recurrent
Silent brain infarcts
stroke, cognitive decline, and dementia. Effects on short-term outcomes are less clear, but
Leucoaraiosis
leucoaraiosis appear to increase the risk of early infarct growth as determined by MRI in
Stroke
mismatch areas of the brain. After acute thrombolytic therapy, presence of silent cerebro-
Thrombolytic therapy
vascular disease increases the risk of intracerebral hemorrhage, but the increased risk does
Anticoagulants
not negate the overall benefit of this therapy. Similarly, presence of leucoaraiosis is
Blood pressure
associated with an increased risk of intracerebral bleeding during long-term anticoagulant therapy, but because the risk–benefit ratio has not been well delineated, degree of leucoa-
Mots cle´s :
raiosis should not influence clinical decision-making at present. Carotid endarterectomy for
Infarctus ce´re´braux silencieux
symptomatic carotid stenosis appears equally effective across different degrees of baseline
Leucoaraiose
leucoaraiosis, despite an increased perioperative risk in those with moderate to severe
Infarctus ce´re´bral
leucoaraiosis. Long-term blood pressure lowering appears equally effective in patients with
He´morragie ce´re´brale
silent cerebrovascular disease at baseline, and there is some support that blood pressure
Thrombolyse
lowering may decrease the risk of progression of leucoaraiosis.
Anticoagulants
# 2008 Elsevier Masson SAS. All rights reserved.
Pression arte´rielle
r e´ s u m e´ Les infarctus silencieux et la leucoaraiose sont fre´quemment observe´s chez des patients ayant des accidents ische´miques transitoires ou des infarctus ce´re´braux. Les patients qui pre´sentent des infarctus silencieux ou une leucoaraiose sont des patients a` haut risque de re´cidive, de de´clin cognitif et de de´mence. Les effets sur le pronostic a` court terme ne sont pas clairement identifie´s, mais il semblerait que la leucoaraiose augmente le risque de croissance pre´coce de la zone infarcie en imagerie par re´sonance magne´tique, au sein de la zone de mismatch. En cas de thrombolyse, la pre´sence de ces le´sions silencieuses augmente le risque d’he´morragie ce´re´brale, mais n’alte`re pas le be´ne´fice du traitement. La pre´sence d’une leucoaraiose est associe´e a` un risque plus e´leve´ d’he´morragie ce´re´brale au cours d’un traitement anticoagulant. Toutefois, le rapport be´ne´fice–risque n’ayant pas e´te´ e´value´, il n’est donc pas justifie´ de tenir compte de la leucoaraiose pour la de´cision d’anticoaguler un patient. La chirurgie carotidienne pour ste´nose symptomatique est
E-mail address: [email protected]. 0035-3787/$ – see front matter # 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.neurol.2008.07.009
802
revue neurologique 164 (2008) 801–804
associe´e a` un risque pe´riope´ratoire plus e´leve´ en pre´sence d’une leucoaraiose se´ve`re. La re´duction de la pression arte´rielle pourrait dans certains cas ralentir la progression de la leucoaraiose. # 2008 Elsevier Masson SAS. All rights reserved.
1.
Introduction
The introduction of modern neuroimaging techniques allowed the identification of silent cerebrovascular disease during life. Silent brain infarcts (SBIs) were incidentally recognized during the 1980s in patients who underwent computerized tomography (CT) in the investigation of acute stroke. During the same decade, the frequency of abnormalities in the white matter were increasingly recognized on brain CT scans, and in 1986, Vladimir Hachinski and coworkers introduced the term leucoaraiosis (LA) (from the Greek leuko, white, and araiosis, rarefaction) to designate areas in the periventricular and centrum semiovale white matter that appeared hypodense on CT (ischemic white matter lesions). Pathological studies have linked LA with demyelination, gliosis, necrosis, and cavitation which are associated with atherosclerosis of small and large vessels. Similar findings are seen in arteriosclerotic vascular encephalopathy, or Binswanger disease. Silent cerebrovascular disease has subsequently been a topic of intense research efforts, and CT has largely been replaced by MRI, which is much more sensitive for this purpose. During the last one or two decades, SBIs and LA have evolved from innocent bystanders to small lesions of major importance for vascular and mental health in the population. LA and SBIs frequently coexist, and the two conditions share most risk factors and show a high degree of covariance. Studies of persons in the elderly general population have shown that silent cerebrovascular disease is strongly linked to cognitive impairment, dementia, and increased risk for stroke. Recently, genetical determinants of LA in the population have also been demonstrated. However, the prognostic and therapeutic implications of silent cerebrovascular disease in patients with transient ischemic attack (TIA) and ischemic stroke have been less well explored, and will be reviewed in this report.
2.
Methodological issues
There are several clinical caveats in designating SBIs as truly silent, i.e. not associated with previous acute focal neurological symptoms. Previous clinical symptoms may be forgotten by the patient, the clinical history may be incomplete or symptoms may have occurred, but were not recognized as possible stroke by the patient or physician. Furthermore, about half of all TIAs are associated with an infarct visualized by neuroimaging. Conceivably, a proportion of all TIAs may occur during sleep and pass unrecognized. However, in total, these caveats are likely to account only for a small proportion of all SBIs. Apparently, the majority of infarcts judged as silent were probably not associated with acute focal neurological symptoms, which were forgotten or misdiagnosed.
In patients with stroke and the finding of multiple ischemic areas on conventional imaging, it may sometimes be difficult or even impossible to determine which of the infarcts was symptomatic. Diffusion-weighted MRI (dw-MRI) is particularly helpful in separating acute lesions versus silent or remote. Dw-MRI has also demonstrated that ischemic abnormalities in patients with acute ischemic stroke are not always single: in unselected patient materials, multiple ischemic abnormalities within the territory of one major cerebral artery or in multiple arterial territories are seen in up to 40% of all patients. Large acute infarcts in one hemisphere may hamper evaluation of LA in the infarcted area: in such cases, assessment of LA needs to be limited to the contralateral hemisphere.
3. How frequent are LA and SBIs in patients with TIA and stroke? Several CT-based studies have documented that SBIs and LA are frequently seen in patients admitted with a first TIA or ischemic stroke. In the earlier studies, relatively low prevalences, between 11 and 13%, of SBI were reported, presumably related to the use of early generation CT equipments. In later studies (Jo¨rgensen et al., 1994; Corea et al., 2001), higher prevalences ranging from 23 to 38% were found. Large MRIbased studies of SBI in stroke patients are lacking. In patients with ischemic stroke, CT studies have reported a prevalence of LA ranging from 3 to 44%, whereas MRI-based studies have shown some degree of LA in up to 82% (Fu et al., 2005). Severe LA (grade 3/3) was present in about 30% of cases on MRI. As in healthy persons, SBIs in patients with TIA or stroke are most often small and deep, whereas cortical infarcts account for 10 to 24%, may more often involve the right hemisphere or the posterior cerebral artery territory, and may more often be linked to a cardioembolic cause. With respect to subtype of presenting stroke, SBIs appear to be more prevalent in patients with lacunar infarction than in those with atherothrombotic or cardioemblic infarcts. The presence of SBI and LA in stroke patients have been associated with a wide range of accepted cardiovascular risk factors, largely similar to those found in patients with symptomatic small vessel disease (lacunar infarcts).
4. Significance of SBIs and LA in the acute phase of stroke Several studies have reported that early morbidity and outcome were not influenced by the presence of SCIs or LA in cohorts of patients with stroke or TIA (Jo¨rgensen et al., 1994;
revue neurologique 164 (2008) 801–804
Jo¨rgensen et al., 1995; Wiszniewska et al., 2000; Corea et al., 2001). However, a recent study has provided interesting new insights on the effects of silent cerebrovascular disease in the very early phase of ischemic stroke. In a study of 61 patients with acute ischemic stroke, Ay et al. (2008) explored the relationship between the volume of white matter regions with LA on acute images and the proportion of dw-MRI imaging – mean transit time mismatch tissue that progressed to infarction (percentage mismatch lost). They found a significant correlation between percentage of mismatch lost and LA volume, suggesting that LA volume appears to be a composite predictive marker for the fate of acutely ischemic tissue. LA is associated with structural and functional vascular changes which may be more susceptible to tissue injury when infarct develops in adjacent areas. A similar mechanism may potentially explain the frequent occurrence of initially progressive deficits in patients with lacunar infarcts. These hypotheses warrant further study.
6.1.
Long-term prognostic significance
Whereas silent cerebrovascular disease has no or limited effects on early clinical outcomes, both SBIs and LA at baseline appear to have important effects on long-term prognosis. A recent study (Fu et al., 2005) found that patients with severe LA had increased risk of deep subcortical stroke and a higher risk of recurrent stroke. The prognostic associations with SBIs and LA have in particular been demonstrated in patients with lacunar infarcts, in whom silent cerebrovascular disease at baseline along with risk factors have been shown to be independent predictors of future stroke risk. Presence of SBIs and LA have also been found to be independent predictors of worse functional outcome, suggesting that more advanced small artery disease may limit the possibility of functional recovery of the brain after a small focal lesion. There is also mounting evidence that SCI and LA are an important risk factor for mild cognitive impairment and dementia after stroke, in particular in the longer perspective. Silent cerebrovascular disease is not static. As in the general population, SBIs and LA frequently progress with time also in patients with TIA and stroke. In patients with launar infacts, the subgroup in which this issue has been mainly studied, new silent infarcts develop at a rate of at least twice that of new symptomatic lacunae, and after about three years, almost half of all patients had developed SBIs as well as progression of LA. However, there are as yet no data on the clinical effects of progressing silent cerebrovascular disease with time in these patient groups.
6. Interactions with acute and long-term secondary preventive therapies There is a paucity of prospective randomized controlled clinical trial data on therapeutic issues with respect to silent cerebrovascular disease in patients with TIA and stroke. Existing data are mainly based on posthoc analyses that were not prespecified.
Thrombolytic therapy
Two studies (Neumann-Haefelin et al., 2006; Palumbo et al., 2007) have identified silent cerebrovascular disease as an independent risk factor for intracerebral hemorrhage after thrombolysis with rtPA in acute stroke. In the German multicenter study (Neumann-Haefelin et al., 2006), the rate of hemorrhage was 10.5% among the quarter of the total cohort who had moderate to severe LA versus 3.8% in those without, corresponding to an odds-ratio of 2.9. In the Canadian study (Palumbo et al., 2007), LA and multiple lacunes were both independent predictors for intracerebral hemorrhage (oddsratios 2.7 and 3.4, respectively), but no difference was noted for good outcome rate at 90 days. These observations should not influence the use of rtPA; several other factors have also been associated with increased bleeding rates after rtPA, but none of these have been clearly shown to reverse the overall benefit of rtPA.
6.2.
5.
803
Antiplatelet therapy
Although there is evidence that silent cerebrovascular disease at baseline affects prognosis, it is unknown if more aggressive secondary preventive approaches than those generally applied are warranted. Benefits of combining antithrombotic regimens may well be offset by increased adverse effects, as illustrated by the recent MATCH-trial of clopidogrel plus aspirin versus clopidogrel only. The ongoing Secondary Prevention of Small Subcortical Strokes (SPS3) study will provide more data on this issue; in a factorial design that also involves different blood pressure targets (see below), the trial is testing clopidogrel and aspirin as compared with aspirin alone in the prevention of stroke recurrence and cognitive decline in patients with recent symptomatic small subcortical stroke.
6.3.
Anticoagulants
In the SPIRIT trial, which compared warfarin with target INR values of 3.0 to 4.5 to aspirin in patients with TIA and ischemic stroke of presumed arterial origin, the presence of LA was associated with a markedly elevated hazard (9.2) for intracerebral hemorrhage. This effect remained significant when controlled for age, blood pressure, and INR. A small casecontrolled study has also suggested that LA may be an independent risk factor for intracerebral hemorrhage also in patients with commonly employed range of anticoagulation (INR 2.0 to 3.0). However, until further data become available on both the risk and benefits of anticoagulation in the presence of LA, these data are insufficient to incorporate silent cerebrovascular disease in the decision-making of individual patients. The exact mechanism by which anticoagulants and rtPA increase the risk of intracerebral hemorrhage in patients with silent cerebrovascular disease has not been determined.
6.4.
Carotid endarterectomy
A posthoc analysis of the NASCET study showed that in patients with symptomatic carotid artery disease the presence
804
revue neurologique 164 (2008) 801–804
of LA was associated with an increased risk of any stroke and of disabling or fatal stroke (Streifler et al., 2002). The 30-day perioperative risk was also increased, from 5.3% (no LA), to 10.6% (restricted LA) and 13.9% (widespread LA). However, despite higher perioperative risk, endarterectomy reduced the absolute three-year risk of ipsilateral stroke irrespective of degree of LA.
6.5.
Blood pressure lowering
Blood pressure control plays a major role both in primary and in secondary stroke prevention. However, theoretical concerns have been raised that too aggressive blood pressure lowering may have adverse effects in patients with advances small vessel disease and impaired autoregulation. A CT substudy of the PROGRESS trial on blood pressure lowering therapy in patients with stroke found no significant effect on therapy on the rate on SBIs over 3.9 years (Hasegawa et al., 2004). However, a MRI substudy showed that the blood pressure-lowering regimen stopped or delayed the progression of white matter lesions (Dufouil et al., 2005). The greatest beneficial effect of blood pressure lowering on LA was observed in the group of patients who had severe leucoaraiosis at study entry. In the ongoing SPS3 study, the role of lowering systolic blood pressure two different levels (< 130 mmHg versus < 140 mmHg) is tested in a factorial design.
7.
Conclusions
In patients with ischemic stroke and TIA, there is a growing amount of evidence that presence and extent of silent cerebrovascular disease at the time of the first event have significant prognostic implications in the long-term for almost all outcomes. Short-term effects on clinical outcomes have not been clearly demonstrated, but on MRI presence of severe LA appears to increase the susceptibility to early infarct growth. Silent cerebrovascular disease increases the risk of intracerebral hemorrhage after acute thrombolytic therapy as well during long-term anticoagulant therapy, but has not been clearly shown to reverse the overall benefit of these therapies. At present, presence of SBI and LA should therefore not be used to influence clinical decisions. Carotid endarterectomy appears equally effective in patients with and without LA. Theoretical concerns of detrimental effects of blood pressure lowering in the presence of cerebral small vessel disease has not been substantiated by clinical trials. Blood pressure lowering after TIA and ischemic stroke has not been linked to increased risk of progression of silent cerebrovascular disease. Blood pressure lowering appears effective across different stroke subtypes, including patients with SBIs and LA at baseline. Should the management of stroke patients be modified in the presence of SBIs or LA at the time of the index event? At present, there is no data to support that any of the established
acute or long-term secondary preventive therapies should be withheld for this reason. Furthermore, there are no data to support the use of more intense therapeutic regimens like combined use of antiplatelets, or more intense blood pressure lowering than what is generally recommended in clinical guidelines. However, given that the presence of SBIs and LA at baseline implies an increased long-term risk for recurrent stroke, cognitive decline and dementia, careful attention to risk factor control, life style modifications, and long-term compliance with secondary preventive measures is essential in this high-risk group.
references
Ay H, Arsava M, Rosand J, Furie KL, Singhal AB, Schaefer PW, et al. Severity of leucoaraiosis and susceptibility to infarct growth in acute stroke. Stroke 2008;39:1409–13. Corea F, Henon H, Pasquier F, Leys D. Silent infarcts in stroke patients: patient characteristics and effects on 2-year outcome. J Neurol 2001;248:271–8. Dufouil C, Chalmers J, Coskun O, Besancon V, Bousser MG, et al. Effects of blood pressure lowering on cerebral white matter hyperintensities in patients with stroke. The PROGRESS (Perindopril protection against recurrent stroke study) Magnetic resonance imaging substudy. Circulation 2005;112:1644–50. Fu JH, Lu CZ, Hong Z, Dong Q, Luo Y, Wong KS. Extent of white matter lesions is related to acute subcortical infarcts and predicts further stroke risk in patients with first ever ischaemic stroke. J Neurol Neurosurg Psychiatry 2005;76:793–6. Hasegawa Y, Yamaguchi T, Omae T, Woodward M, Chalmers J. PROGRESS CT Substudy Investigators. Effects of perindoprilbased blood pressure lowering and of patient characteristics on the progression of silent brain infarcts: the Perindopril Protection against Recurrent Stroke Study (PROGRESS) CT Substudy in Japan. Hypertens Res 2004;27:147–56. Jo¨rgensen HS, Nakayama H, Raaschou HO, Olsen TS. Silent infarction in acute stroke patients. Prevalence, risk factors, and clinical significance: the Copenhagen Stroke Study. Stroke 1994;25:97–104. Jo¨rgensen HS, Nakayama H, Raaschou HO, Olsen TS. Leucoaraiosis in stroke patients. The Copenhagen Stroke Study. Stroke 1995;36:588–92. Neumann-Haefelin T, Hoelig S, Berkfeld J, Fiehler J, Gass A, Humpich M, et al. Leucoaraiosis is a risk factor for symptomatic intracerebral hemorrhage after thrombolysis for acute stroke. Stroke 2006;37:2463–6. Palumbo V, Boulanger JM, Inzitari D, Buchan AM, on behalf of the CASES Investigators. Leucoaraiosis and intracerebral hemorrhage after thrombolysis in acute stroke. Neurology 2007;68:1020–4. Streifler JY, Eliaziw M, Benavente OR, Alamowitch S, Fox AJ, et al. Prognostic importance of leucoaraiosis in patients with symptomatic internal carotid artery stenosis. Stroke 2002;33:1651–5. Wiszniewska M, Devuyst G, Bogousslavsky J, Ghika J, van Melle G. What is the significance of leucoaraiosis in patients with acute ischemic stroke? Arch Neurol 2000;57:967–73.