Long-term Prognosis of First-Ever Lacunar Strokes
A Hospital-Based Study
Background and Purpose Information concerning the long-term prognosis of lacunar strokes is still limited and has shown different results. The aim of this study was to investigate the long-term prognosis of first-ever lacunar strokes and the possible role of clinical prognostic factors and different pathogenic mechanisms.
Methods Between March 1990 and November 1993, a cohort of consecutive patients presenting with first-ever lacunar infarcts was prospectively evaluated after stroke onset at day 0 to 3 and/or day 7, every 3 months up to 1 year, and every 6 months thereafter. All patients were studied according to a protocol that included demographic and clinical data, neurological examination, Toronto Stroke Scale, Barthel Index, Rankin Scale, CT scan, routine laboratory workup, electrocardiogram, carotid duplex scanning, and echocardiogram. More recently, patients have also been evaluated with transcranial Doppler ultrasonography. Recurrent strokes, myocardial infarction, and death were registered through direct observation, chart review, or interviews with the attending physician or family members.
Results One hundred forty-five patients—94 (65%) with pure hemiparesis, 33 (23%) with sensorimotor stroke, 11 (8%) with ataxic hemiparesis, 5 (3%) with pure sensory stroke, and 2 (1%) with dysarthria-clumsy hand syndrome—were followed for a median period of 39 months. During follow-up there were 17 deaths (3 vascular), 30 first recurrent strokes (1 fatal), and 4 myocardial infarctions. Five-year survival rate free of recurrent stroke was 63% (95% confidence interval [CI], 52% to 73%), while 5-year survival rate was 86% (95% CI, 78% to 91%). Cox proportional hazards analysis showed that age (P=.02) was the only significant predictor of survival free of recurrent stroke. Age (P<.001) and the degree of neurological dysfunction and functional disability at 7 days after the index stroke measured by the Toronto Stroke Scale (P=.05) and a Barthel Index score <40 (P=.04) were the only significant predictors of death. The 5-year probability rate of stroke-free recurrence was 72% (95% CI, 60% to 81%). Sixty-three percent of the first recurrent strokes were lacunar infarcts. When clinical, laboratory, and CT data as well as possible etiopathogenic mechanisms of lacunar strokes were considered, Cox proportional hazards analysis could not identify any predictor of stroke recurrence.
Conclusions Our study confirms that lacunar infarcts are associated with low stroke recurrence and mortality rates. In our series, the majority of first recurrent strokes were also lacunar infarcts. Age, degree of neurological dysfunction, and functional disability at day 7 after the index stroke were significant predictors of death.
Although lacunar infarcts, ie, small subcortical infarcts due to single perforating artery occlusion, have been associated with low mortality and low stroke recurrence rates in both hospital- and community-based studies,1 2 3 4 5 6 7 8 9 10 11 data concerning long-term follow-up and type of recurrent stroke are still limited and have shown different results (Table 1⇓).
Furthermore, the role of extracranial12 13 14 15 16 17 18 19 and intracranial19 20 atherosclerotic carotid disease as well as cardioembolic disease12 15 18 19 21 in the pathogenesis of lacunar infarcts is still a matter of controversy, and the impact of these different pathogenic mechanisms on the risk of recurrent stroke is unknown. Recently, two clinical series22 23 have suggested that patients with single lacunes may represent an entity different from those with multiple lacunes with respect to the frequency of arterial hypertension and diabetes mellitus and the presence of leukoaraiosis. However, the possible impact of these two entities on stroke recurrence was not assessed by any of these studies.
In the present study we evaluated the long-term prognosis of first-ever lacunar infarcts with respect to stroke recurrence rate, type of recurrent stroke, and mortality rate and its relation to possible clinical prognostic factors and different pathogenic mechanisms.
Subjects and Methods
Between March 1990 and November 1993, we included in a prospective registry a cohort of consecutive patients, regardless of age, with first-ever strokes seen at either the emergency department or outpatient clinic or admitted to the neurology or medical wards with the following characteristics: (1) examined within 7 days after stroke onset; (2) presenting with one of the five classic lacunar syndromes (pure motor hemiparesis, sensorimotor stroke, pure sensory stroke, ataxic hemiparesis, and dysarthria-clumsy hand syndrome)2 ; and (3) presenting with a normal CT scan or a subcortical hypodense circular or oval lesion with a diameter ≤15 mm,24 hereafter designated as a lacune.
Patients were evaluated according to a protocol that included demographic data, medical history, vascular risk factors (definitions in the “Appendix”), and the Toronto Stroke Scale25 at days 0 to 3. At day 7 after stroke onset, all patients were evaluated according to the Toronto Stroke Scale,25 modified Rankin Scale,26 and Barthel Index.27 Additional studies included brain CT scan, carotid duplex, two-dimensional and M-mode echocardiograms, electrocardiogram, and routine blood work. More recently, patients have also been evaluated with transcranial Doppler ultrasonography.
All CT scans performed within 72 hours after stroke onset were repeated. All scans were performed with the use of a Phillips Tomoscan LX (n=88; 61%) or a Siemens Somaton HQ (n=42; 29%) system, with consecutive slices of 10 mm thickness parallel to the orbitomeatal line. In 15 patients (10%), CT was performed with other third generation scans. CT scan reading with respect to the number of lacunes and the presence of leukoaraiosis28 was done by two of the authors blinded to the patient’s clinical presentation. Interobserver agreement was 69% for the presence of single or multiple lacunes and 81% for the presence of leukoaraiosis. Differences in CT interpretation were solved by consensus. Cases in which CT showed single lacunes not appropriate to the clinical syndrome were classified as multiple lacunes. In cases with single lacunes, lacunar size was measured according to the method described by Nelson et al.13
To determine the possible role of the different pathogenic mechanisms on the development of end points, echocardiographic and electrocardiographic results were used to distinguish patients with high versus medium or low risk of cardioembolism according to the criteria of Kittner et al.29 Carotid duplex scanning was performed with a 7.5-MHz Toshiba SS 140 A or 7.5-MHz ATL Ultramark 9 system. Carotid duplex results were used to distinguish patients with and without evidence of atherosclerotic carotid disease, ipsilateral or contralateral to the symptomatic hemisphere, determined by the presence of stenosis irrespective of its degree. A carotid stenosis >50% was diagnosed when (1) peak systolic velocity was >1.25 m/s and (2) diastolic flow was increased to at least double that of the primitive carotid artery and showed marked spectral broadening. More recently, transcranial Doppler results have been used to distinguish patients with or without middle cerebral artery disease ipsilateral or contralateral to the symptomatic hemisphere and vertebrobasilar artery disease documented by the presence of increased systolic velocity in the relevant vessels.30
Patients were evaluated by one of the authors (A.V.S.) at the outpatient clinic at 3-month intervals up to 12 months after stroke onset and every 6 months thereafter. To determine the occurrence of the primary end points, ie, death and recurrent stroke (definitions in the “Appendix”), patients or family members were asked to report immediately to the Department of Neurology. The type of recurrent stroke was based on both clinical and CT findings, whenever possible. In patients unable to come to the outpatient clinic, information concerning the occurrence of end points during follow-up was obtained through review of medical charts or interview with the attending physician or family members. Recurrent strokes were considered disabling if there was a sustained 2-point increment in the Rankin Scale from the previous follow-up score. Myocardial infarctions were also recorded.
Scores at 7 days after the index stroke on the Toronto Stroke Scale were used as a continuous variable, whereas scores on the Barthel Index were divided into two classes with a cutoff of 4031 and scores on the Rankin Scale were divided with a cutoff of 3.
Kaplan-Meier curves were used to estimate survival free of recurrent stroke, survival, and probability of stroke recurrence. Potentially significant predictors of death and stroke-free recurrence were evaluated with Cox proportional hazards regression analysis with a forward stepwise procedure. Statistical significance was established at the P<.05 level.
Between March 1990 and November 1993, 145 patients were included in the study: 93 men (64%) and 52 women (36%), with a mean±SD age of 64.8±10.8 years (range, 34 to 93 years). One hundred thirty-seven patients (94%) were first examined up to day 3, and 8 (6%) were first examined between days 4 and 7 after stroke onset. The median time between stroke onset and CT scan was 4 days (range, 3 to 29 days). Among the 81 patients with single or multiple lacunes on CT scan, 4 (5%) had infarcts involving the brain stem. Patients’ clinical characteristics and CT results are shown in Tables 2⇓ and 3⇓, respectively.
Fourteen patients (10%) had a potential cardiac source of embolism: atrial fibrillation in 7, a left ventricular hypokinesia in 6, and both in 1. Carotid duplex was normal in 81 patients (56%) and showed bilateral stenosis in 37 (26%), ipsilateral stenosis to the symptomatic hemisphere in 16 (11%), and contralateral stenosis in 11 (8%). Carotid stenosis >50% was present ipsilateral to the symptomatic hemisphere in 2 patients (1%) and contralateral in 3 (2%). Nine of the 14 patients (64%) with a cardiac source of embolism had concomitant carotid disease documented by duplex (ipsilateral to the symptomatic hemisphere in 3, bilateral in 4, and contralateral in 2).
Transcranial Doppler ultrasonography was performed in 65 patients. Five patients (3%) presented an increased velocity in the middle cerebral artery ipsilateral to the symptomatic hemisphere, 6 (4%) in both middle cerebral arteries, 3 (2%) in the contralateral middle cerebral artery only, and 3 (2%) in the vertebrobasilar artery system.
All patients were discharged from the hospital on antiaggregant agents except for 3 who were discharged on warfarin. Two of these patients had a history of atrial fibrillation before the index stroke and continued on warfarin throughout the study period. The other patient was given warfarin during a 6-month period because of the stepwise progression of the neurological deficit at the onset of the index stroke.
End Points During Follow-up
During a median follow-up period of 39 months (range, 1 to 60 months), information concerning development of end points was obtained through direct clinical examination (n=134) and chart review and through information given by the attending physician (n=5) or family members (n=6). Only one patient was lost to follow-up, 1 month after the index stroke.
End points that occurred during follow-up are listed in Table 4⇓.
Survival Free of Recurrent Stroke
When we combined first recurrent stroke or death, whichever happened first, in a single end point (n=40; 27.6%), the survival rate free of recurrent stroke was 90% (95% confidence interval [CI], 84 to 94) at 12 months, 81% (95% CI, 73 to 86) at 24 months, 71% (95% CI, 62 to 78) at 36 months, 66% (95% CI, 56 to 74) at 48 months, and 63% (95% CI, 52 to 73) at 60 months (Fig 1⇓). Univariate Cox proportional hazards analysis showed that age (hazard ratio [HR], 1.03; 95% CI, 1.0 to 1.07; P=.02) was the only variable that predicted the development of either of these two end points.
During the same period there were 17 deaths (12%): 7 due to different medical complications, ie, diabetes or infection, 5 due to cancer, 2 after surgery, 1 occurring 24 hours after coronary artery bypass surgery, 1 due to a fatal subcortical hematoma, 1 sudden death, and 1 due to a motor vehicle accident. The survival rate was 95% (95% CI, 90 to 98) at 12 months, 92% (95% CI, 86 to 96) at 24 months, 88% (95% CI, 80 to 92) at 36 months, 86% (95% CI, 78 to 91) at 48 months, and 86% (95% CI, 78 to 91) at 60 months (Fig 2⇓). Univariate Cox proportional hazards analysis identified the variables that significantly predicted death as age (HR, 1. 09; 95% CI, 1.03 to 1.1; P<.001), the Toronto Stroke Scale score (HR, 1.1; 95% CI, 1.0 to 1.2; P=.03), and a Barthel Index score <40 (HR, 0.2; 95% CI, 0.08 to 0.6; P=.003) 7 days after the index stroke. In the multivariable stepwise model, age, the Toronto Stroke Scale score, and a Barthel Index score <40 at 7 days after the index stroke maintained a significant statistical relationship with the risk of death (Table 5⇓).
The probability rate of stroke-free recurrence was 93% (95% CI, 87 to 96) at 12 months, 86% (95% CI, 79 to 91) at 24 months, 79% (95% CI, 70 to 85) at 36 months, 75% (95% CI, 65 to 82) at 48 months, and 72% (95% CI, 60 to 81) at 60 months (Fig 3⇓).
During follow-up, 30 patients (21%) had at least 1 recurrent stroke, 5 (3%) had 2, and 2 (1%) had 3. First recurrent strokes were ischemic in 26 (lacunar in 19, cortical in 3, large subcortical in 2, brain stem in 2) and hemorrhagic in 2 (subcortical hematoma and lobar hematoma in 1 each). In 2 patients CT scan was not performed. Lacunar infarcts (63%) were more common than all other nonlacunar strokes (37%). Second recurrent strokes were ischemic in 3 (cortical in 2 and lacunar in 1) and hemorrhagic in 2 (subcortical hematoma and cerebellar hematoma in 1 each). Two patients had a third recurrent stroke, both ischemic (lacunar and brain stem in 1 each). Only 7 (23%) of the 30 first recurrent strokes were in the same vascular distribution of the index stroke. Only 1 of the 5 patients with a carotid stenosis >50% had a recurrent stroke during follow-up, the event occurring in the carotid artery distribution contralateral to the stenosis. Thirteen first recurrent strokes (17%) occurred in 77 patients with no cardiac or carotid source of embolism, 7 (54%) being lacunar, 4 (31%) cortical or subcortical, and 2 (15%) hematomas, while 17 (25%) first recurrent strokes occurred in 68 patients with a cardiac and/or carotid source of embolism, 12 (71%) being lacunar, 3 (18%) cortical or large subcortical, and 2 (12%) of unknown type. Fifteen (50%) of the 30 first recurrent strokes were disabling.
Cox proportional hazards regression analysis failed to show any clinical or CT predictor of recurrent stroke. The presence of either ipsilateral (HR, 1.0; 95% CI, 0.4 to 2.6; P=.9) or contralateral carotid stenosis (HR, 0.8; 95% CI, 0.3 to 2.2; P=.7) or cardioembolic source (HR, 2.1; 95% CI, 0.8 to 5.9; P=.1) was not associated with an increased risk of stroke recurrence. The same was true for the presence of increased intracranial arterial velocity on transcranial Doppler (HR, 0.7; 95% CI, 0.08 to 5.8; P=.7).
To our knowledge, this is the first prospective series of patients with first-ever strokes presenting with classic lacunar infarcts evaluated according to a protocol that included CT scan, echocardiogram, and carotid duplex, in which information concerning stroke recurrence, myocardial infarction, and death during a median follow-up of 39 months was available.
In our study the risk of stroke recurrence after first-ever lacunar stroke was 7%/y until the third year and stabilized at 3%/y until the fifth year of follow-up, which results in a mean annual stroke rate of 7% and a cumulative risk of stroke recurrence of 14% at 24 months and 28% at 60 months. Although series evaluating the long-term follow-up of lacunar strokes have not used the same inclusion criteria, all studies, including ours, have described a mean annual stroke rate between 4% and 7%,1 5 11 an early (24 months) cumulative recurrence rate between 7% and 14%, and a late (48 to 72 months) cumulative recurrence rate between 26% and 34%.1 4 6 11
When compared with other hospital-based series of patients with lacunar strokes, the mean age of our patient population (65 years) was comparable to that of the series of Gandolfo et al1 and that of Boiten and Lodder9 but lower than that of Miyao et al (71 years),8 which was similar to the one in a community-based study (70 years).6 In Portugal there is some epidemiological evidence that the percentage of acute strokes admitted to the hospital decreases with age,32 and therefore our series is only representative of patients with lacunar strokes referred to a large community hospital.
Few studies have addressed the type of recurrent stroke after lacunar infarcts. While in the series of Miyao et al,8 Boiten and Lodder,9 and ours most of the recurrent strokes were lacunar infarcts, in others1 6 10 11 the type of recurrent stroke was of different subtypes, ie, lacunar or cortical. The reason for these different results may be explained by the small number of recurrent strokes as well as different methodologies in study design, namely, inclusion of patients with history of prior stroke10 11 and unspecific lacunar syndromes.1 11 Also of note and as previously described by others,9 the majority of the first recurrent strokes occurred in a different vascular distribution compared with the index stroke, and only half were associated with significant increasing disability.
As in the series of Gandolfo et al1 and Clavier et al,11 we were not able to find any clinical or CT predictor of the risk of recurrent stroke. However, these results should be interpreted with caution, since in our series 23% of the scans were normal even 72 hours after stroke onset, which limits the use of lacunar size and location as possible predictors of the risk of recurrent stroke. Likewise, and as previously described,11 we were unable to demonstrate any increased risk of stroke recurrence when we considered different subtypes of lacunar stroke determined by the presence of hypertension and/or diabetes and multiple lacunes or leukoaraiosis, as suggested by some authors.8 22 23 Although Fisher33 suggested that duration of hypertension played a role in determining the type of small-vessel vasculopathy, we were unable to find any relationship between the duration of hypertension or diabetes and the risk of recurrent stroke.
The role of potential cardioembolic sources, carotid artery disease, and intracranial atherosclerotic disease in the pathogenesis of lacunar strokes is still a matter of controversy.12 13 14 15 16 17 18 19 20 21 In our series there was no relation between the presence of any of these different pathogenic mechanisms and the risk of recurrent stroke. Furthermore, in our study the frequency of the different types of first recurrent stroke, although small in number, was similar when patients with and without a cardiac or carotid source of embolism were compared, the majority being lacunar in both groups. Both of these findings suggest that the presence of a cardiac or carotid embolic source in patients with lacunar stroke may only represent a concomitant disease or a marker of diffuse atherosclerosis, as previously postulated by others.14 16 17 18 21 23 Furthermore, in patients with lacunar stroke the most important vascular pathogenic mechanism appears to be widespread intracranial small-vessel disease.
In our series, the risk of myocardial infarction was 3%, a figure significantly lower than that described in patients with other subtypes of stroke.34 Although this finding may be due to the relatively young age of our sample and a lower prevalence of myocardial ischemic disease in Portugal, where the mortality rate due to myocardial infarction is half that of stroke,35 it may also suggest that the underlying vascular pathology in lacunar stroke is different from other stroke subtypes, particularly the association with coronary artery disease, such as in the case of large-vessel disease.
In our study survival rate decreased steadily until the third year of follow-up, when it reached 88%, and was stable at 86% until the fifth year. These results are similar to those previously reported by others1 2 4 6 11 Most of the deaths were caused by several medical complications, namely, diabetes, infection, or cancer. As in previous studies,1 2 4 6 7 8 9 in our series only one patient died as a direct consequence of a recurrent stroke. In patients with lacunar stroke, death due to myocardial infarction or heart failure has been described in none to 20% of the cases,1 2 4 6 7 8 9 while in our study only one patient died within 24 hours of a coronary bypass graft surgery and another from sudden death.
In the series of Gandolfo et al,1 age and degree of functional disability were predictors of death, while in our study age, the Toronto Stroke Scale score, and a Barthel Index score <40 at 7 days after the index stroke were predictors of death. Both age and the degree of disability or neurological dysfunction predispose subjects to a greater risk of medical complications, which accounted for a significant number of deaths (35%). However, unlike the results of Gandolfo et al, we were unable to demonstrate any statistically significant relationship between hypertension and the risk of death or the risk of combined stroke and death, although the definition of hypertension was not the same in both series. Despite the same frequency of patients with diabetes, unlike the series of Clavier et al11 and that of Brainin et al,36 in our study diabetes was not a predictor of death or combined stroke and death. Likewise and contrary to the series of Clavier et al, in our study cigarette smoking was not a predictor of death. This difference may be due in part to the fact that we had half the number of patients who were current smokers.
The role of leukoaraiosis as a predictor of death is uncertain. In the series of Miyao et al,8 death was significantly associated with the presence of leukoaraiosis in patients with lacunar stroke, while in the series of Clavier et al11 and a recent stroke community study37 as well as in our study, the same association was not found.
In conclusion, in our long-term follow-up study lacunar strokes were associated with low stroke recurrence and mortality rates as well as a low risk of myocardial infarction. We were unable to find any relationship between clinical and CT variables or between the coexistence of cardioembolic or artery-to-artery intracranial or extracranial embolic source and an increased risk of recurrent stroke. Age and degree of neurological dysfunction and functional disability 7 days after the index stroke were reliable predictors of death, which was rarely due to recurrent stroke.
Hypertension was defined as systolic blood pressure values >160 mm Hg and/or diastolic blood pressure values >95 mm Hg on two different occasions before stroke onset or after the first week after the stroke or use of antihypertensive medication; duration of hypertension was defined as the interval in years from the time the patient was first informed that he suffered from hypertension to the date of the index stroke; diabetes mellitus was defined as fasting glucose >140 mg/dL or use of antidiabetic medication; duration of diabetes was defined as the interval in years from the time the patient was first informed that he suffered from diabetes to the date of the index stroke; hyperlipidemia was defined as fasting serum cholesterol level >240 mg/dL and/or fasting serum triglyceride level >200 mg/dL; elevated hematocrit was defined as >45%; cigarette smoking was defined as smoking at the time of the stroke; alcohol abuse was defined as daily intake >120 g; and peripheral vascular disease was defined as intermittent claudication due to atheromatous arterial disease.
Recurrent stroke was defined as a focal neurological deficit occurring suddenly in a vascular territory, lasting >24 hours and occurring at any time after the acute phase of the index stroke.38 Myocardial infarction was defined as at least two of the following: typical pain, new electrocardiographic changes, and enzyme elevation. Fatal recurrent stroke was considered when this condition was the immediate cause of death and clinical and CT scan features confirmed that death was the result of increased intracranial pressure or brain stem compression.39
This study was supported by Junta Nacional de Investigacão Científica e Tecnológica grant STRADA/C/SAU353/92.
- Received June 29, 1995.
- Revision received December 14, 1995.
- Accepted January 5, 1996.
- Copyright © 1996 by American Heart Association
Bamford J, Sandercock P, Jones L, Warlow C. The natural history of lacunar infarction: the Oxfordshire Community Stroke Project. Stroke. 1987;18:545-551.
Arboix A, Marti-Vilalta JL, Garcia JH. Clinical study of 227 patients with lacunar infarcts. Stroke. 1990;21:842-847.
Hier DB; Foulkes MA, Swiontoniowski M, Sacco RL, Gorelick PB, Mohr JP, Price TR, Wolf PA. Stroke recurrence within 2 years after ischemic infarction. Stroke. 1991;22:155-161.
Sacco SE, Whisnant JP, Broderick JP, Phillips SJ, O’Fallon WM. Epidemiological characteristics of lacunar infarcts in a population. Stroke. 1991;22:1236-1241.
Landi G, Cella E, Boccardi E, Musicco M. Lacunar versus non-lacunar infarcts: pathogenetic and prognostic features. J Neurol Neurosurg Psychiatry. 1992;55:441-445.
Miyao A, Teramoto J, Takahashi A. Leukoaraiosis in relation to prognosis for patients with lacunar infarction. Stroke. 1992;23:1434-1438.
Boiten J, Lodder J. Prognostic for survival, handicap and recurrence of stroke in lacunar and superficial infarction. Cerebrovasc Dis. 1993;3:221-226.
Clavier I, Hommel M, Besson G, Noèlle B, Perret F. Long-term prognosis of symptomatic lacunar infarcts: a hospital-based study. Stroke. 1994;25:2005-2009.
Pullicino P, Nelson RF, Kendall BE, Marshall J. Small deep infarcts diagnosed on computed tomography. Neurology. 1980;30:1090-1096.
Nelson RF, Pullicino P, Kendall BE, Marshall J. Computed tomography in patients presenting with lacunar syndromes. Stroke. 1980;11:256-261.
Kappelle LJ, Koudstaal PJ, Van Gijn J, Ramos LMP, Keunen JEE. Carotid angiography in patients with lacunar infarction. Stroke. 1988;19:1093-1096.
Ghika J, Bogousslavsky J, Regli F. Infarcts in the territory of the deep perforators from the carotid system. Neurology. 1989;39:507-512.
Norrving B, Cronqvist S. Clinical and radiologic features of lacunar versus nonlacunar minor stroke. Stroke. 1989;20:59-64.
Tegeler CH, Shi F, Morgan T. Carotid stenosis in lacunar stroke. Stroke. 1991;22:1124-1128.
Boiten J, Lodder J. Lacunar infarcts: pathogenesis and validity of the clinical syndromes. Stroke. 1991;22:1374-1378.
Horowitz DR, Tuhrim S, Weiberger JM, Rudolph SH. Mechanisms in lacunar infarction. Stroke. 1992;23:325-327.
Lodder J, Bamford JM, Sandercock PAG, Jones LN, Warlow CP. Are hypertension or cardiac embolism likely causes of lacunar infarction? Stroke. 1990;21:375-381.
Boiten J, Lodder J, Kessels F. Two clinically distinct lacunar infarct entities? Stroke. 1993;24:652-656.
Mast H, Thompson JLP, Lee S-H, Mohr JP, Sacco RL. Hypertension and diabetes mellitus as determinants of multiple lacunar infarcts. Stroke. 1995;26:30-33.
Norris JW. Comments on ‘Study Design of Stroke Treatments.’ Stroke. 1982;13:527-528.
van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJA, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19:604-607.
van Swieten JC, Hijdra A, Koudstaal PJ, van Gijn J. Grading white matter lesions on CT and MRI: a simple scale. J Neurol Neurosurg Psychiatry. 1989;53:1080-1083.
Kittner SJ, Sharkness CM, Price TR, Plotnick GD, Dambrosia JM, Wolf PA, Mohr JP, Hier DB, Case CS, Thurim S. Infarcts with a cardiac source of embolism in the NINCDS Stroke Data Bank: historical features. Neurology. 1990;40:281-284.
De Witt LD, Rosengart A, Teal PA. Transcranial Doppler ultrasonography: normal values. In: Babikian VL, Wechsler LR, eds. Transcranial Doppler Ultrasonography. St Louis, Mo: Mosby; 1993.
Orgogozo JM. Evaluation of treatments in ischemic-stroke patients. In: Amery WK, Bousser M-G, Clifford Rose F, eds. Clinical Trial Methodology in Stroke. London, England: Baillière Tindall; 1989.
Furtado Silva DJM. Acidentes vasculares cerebrais notificados em Portugal em 1990 pelos Médicos Sentinela. Rev Port Clin Geral. 1992;9:81-87.
Fisher CM. The underlying lesions underlying lacunes. Acta Neuropathol (Berl). 1969;12:1-15.
Dennis MS, Burn JS, Sandercock PAG, Bamford JM, Wade DT, Warlow CP. Long-term survival after first-ever stroke: the Oxfordshire Community Stroke Project. Stroke. 1993;24:796-800.
Elementos Estatisticos: Saúde/94. Lisbon, Portugal: DEPS, Ministério da Saúde; 1994:77.
Brainin M, Foulkes MAA, Pauly E, Seiser A, Dastmaltschi J, Steinert M. The role of hypertension and diabetes for the survival of lacunar stroke patients. Cerebrovasc Dis. 1992;2:202. Abstract.
Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS. Leukoaraiosis in stroke patients: the Copenhagen Stroke Study. Stroke. 1995;26:588-592.
Meissner I, Whistnant JP, Garraway WM. Hypertension management and stroke recurrence in a community (Rochester, Minnesota, 1950-1979). Stroke. 1988;19:459-463.
Anderson CS, Jamrozik KD, Broadhurst RJ, Stewart-Wynne EG. Predicting survival for 1 year among different subtypes of stroke: results of the Perth Community Stroke Study. Stroke. 1994;25:1935-1944.