Leukoaraiosis More Than Dementia Is a Predictor of Stroke Recurrence
Background and Purpose— It has been suggested that poststroke dementia is associated with a higher risk of stroke recurrence. Leukoaraiosis, however, might be a confounding factor because it is a risk factor for stroke recurrence and cognitive decline. Our aim was to determine the influence of prestroke and poststroke dementia on the 3-year risk of stroke recurrence.
Methods— We evaluated prestroke cognitive functions in 202 stroke patients ≥40 years of age using the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE), with a cutoff of 104 for the diagnosis of dementia. Patients were followed up for 3 years. Dementia was diagnosed on the basis of International Classification of Diseases, 10th revision, criteria in survivors who underwent a neurologist visit or from the IQCODE score in survivors who did not. The severity of leukoaraiosis was assessed with a visual rating scale on CT scans without contrast performed at the acute stage of stroke. At each follow-up contact, stroke recurrences were recorded.
Results— During 385 person-years of follow-up, a total of 29 patients developed 33 stroke recurrences, resulting in an incidence rate of 8.6 per 100 person-years. We did not find any influence of dementia on the risk of stroke recurrence. Leukoaraiosis, however, was a strong predictor of stroke recurrence within 3 years after stroke.
Conclusions— Dementia was not a predictor of stroke recurrence, but leukoaraiosis was strongly associated with stroke recurrence. Special attention in the secondary prevention of stroke must be given to patients with leukoaraiosis with or without dementia.
About 16% of patients admitted for stroke have preexisting dementia,1,2 and the incidence of poststroke dementia reaches 33% within 5 years.3,4 Poststroke dementia has been found to be associated with a poorer vital outcome3,5–7 and a higher risk of long-term stroke recurrence.3 The reason that demented patients have a higher risk of stroke recurrence remains unclear. Poststroke dementia and stroke recurrence may share common risk factors,3 and silent infarcts may play a role. A relationship has been suggested between silent infarcts and poststroke dementia4,8 and between silent infarcts and stroke recurrence9. Leukoaraiosis may be a confounding factor; it is a risk factor for stroke recurrence10 and cognitive decline.4,11 Finally, poststroke dementia may increase the risk of stroke recurrence because of an influence on prescription of antithrombotic drugs and on therapeutic observance.12 However, the only study3 to evaluate the relationship between dementia and stroke recurrence was conducted in a selected population with patients with hemorrhagic stroke, aphasic patients, and patients who died within 3 months excluded. This selection may have led to biases in the appreciation of the real influence of poststroke dementia on stroke recurrence. Moreover, prestroke dementia is frequent,1,2 and although the mortality rate is increased in patients with prestroke dementia,6,7 its influence on stroke recurrence has never been evaluated.
The aim of our study was to determine the influence of prestroke and poststroke dementia on the risk of stroke recurrence in 202 consecutive stroke patients.
The study population was recruited between November 1995 and May 1996. All consecutive patients ≥40 years of age admitted to the Stroke Unit of the Lille University Hospital were eligible. The protocol and first results of the Lille Stroke/Dementia Study have been published elsewhere.1,4,7
Patients were examined according to a standardized procedure.1,4,7 We prospectively collected the following data: age; sex; education level; presence of arterial hypertension (defined as systolic blood pressure >160 mm Hg, diastolic blood pressure >90 mm Hg, or current treatment with antihypertensive drugs either before stroke onset or lasting >1 month after stroke onset); diabetes mellitus (defined as serum glucose level >1.20 g/L or current use of antidiabetic drugs); dyslipidemia (defined as fasting serum level of triglycerides >1.5 g/L or fasting cholesterol serum level >2.3 g/L); cigarette smoking (>10 cigarettes a day or cessation <5 years earlier); alcohol consumption >300 g/wk; and history of stroke, coronary heart disease, and peripheral arterial disease. Stroke subtypes were defined according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria.13
On CT scans without contrast performed at admission, we determined the presence of silent infarcts, leukoaraiosis, and cerebral atrophy. Leukoaraiosis was defined according to the criteria of Inzitari et al14 and scored by means of the 0-to-3-point rating scale of Blennow et al.15 The extension of leukoaraiosis was graded as 0 (no decrease in the attenuation of white matter), 1 (decreased attenuation of white matter at the margins of the frontal and occipital horns of the lateral ventricles), 2 (decreased attenuation of white matter around the frontal and occipital horns of the lateral ventricles with some extension toward the semiovale center), and 3 (decreased attenuation of white matter extending around the whole lateral ventricles and coalescing in the semiovale center). The severity of leukoaraiosis was graded as 0 (none), 1 (mild), 2 (moderate), or 3 (marked decrease in the attenuation of white matter). The leukoaraiosis score was the mean value between the extension and severity scores. Leukoaraiosis was assessed on the hemisphere opposite the unilateral focal vascular lesion, if any, and on the right hemisphere in the remainder. Leukoaraiosis score was determined on a CT scan performed at the acute stage of stroke in all patients. Cerebral atrophy was scored according to the method of Leys et al.16 This assessment provided cerebral atrophy scores of 0 (none), 1 (mild), 2 (moderate), or 3 (severe). The location of the index stroke was determined as a function of the territory involved (hemispheric territory or posterior fossa)8 on a delayed CT (87 patients) or MRI (82 patients) scan. The main reason that patients had CT and not MRI was availability of the machine. In 33 patients, delayed CT or MRI scan was not performed because of early death (n=27) or refusal (n=6); the location of the index stroke was determined from clinical data.
Pre-existing dementia was assessed by means of a French translation of the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE),17 with a cutoff of 104 for the diagnosis of dementia.
Patients were followed up at 6 months and then annually for 3 years by a visit with a neurologist or by telephone contact with the patient’s family or general practitioner (GP). At each visit, the occurrence of a recurrent stroke was recorded. The diagnosis of recurrent stroke was considered when the neurological deficit was clearly different from that of the index stroke, when a different anatomic or vascular territory was involved, or when the recurrent stroke was of a subtype different from that of the index stroke according to TOAST criteria.13 When possible, we obtained hospital charts to document the diagnosis of stroke recurrence. When a patient was aphasic or demented, information concerning stroke recurrence was obtained from the family and the GP. When a patient died before the planned follow-up visit, we called the family or the GP to determine whether death was secondary to a stroke recurrence and whether the patient had experienced stroke between the last contact and death. True dates for stroke recurrence were recorded; for the patients who experienced >1 recurrent stroke, we used the date of the first recurrence for statistical analysis.
Patients who went to the neurologist underwent a battery of neuropsychological tests.4 Dementia was diagnosed according to International Classification of Diseases, 10th revision, criteria18; Alzheimer’s disease (AD), according to the NINCDS-ADRDA criteria19; and vascular dementia (VaD), according to the NINDS-AIREN criteria.20 For patients who could not or refused to attend the visit with the neurologist or the neuropsychological testing (because of severe physical deficit, aphasia, or severe cognitive deficit) or who died >6 months after the last contact, cognitive evaluation was performed with the IQCODE.17
We recorded the treatments prescribed at discharge, the treatments received by the patients before each follow-up visit, and those prescribed by the neurologist at the end of the visit. We took into account antithrombotic drug and antihypertensive, hypoglycemic, and hypolipidemic agents. We considered antithrombotic treatments appropriate (quoted 1) when the prescription was as follows: (1) antiplatelet drugs for atheromatous and lacunar stroke, cardioembolic stroke in patients with contraindication to anticoagulation, and arterial dissection; (2) anticoagulant for cardioembolic stroke and arterial dissection; and (3) either antiplatelet drug or anticoagulant for stroke of undetermined origin. The absence of antithrombotic treatment in patients with contraindication to antithrombotic drugs and in patients with cerebral hemorrhage was also considered appropriate. We considered antithrombotic treatments inappropriate (quoted 0) when the prescription was as follows: (1) anticoagulant for atheromatous and lacunar stroke, (2) antiplatelet drug for cardioembolic stroke in patients without contraindication to anticoagulation, and (3) anticoagulant or antiplatelet drug in patients with cerebral hemorrhage. The absence antithrombotic treatment in patients with ischemic stroke but without contraindication to antithrombotic treatment was also considered inappropriate.
Kaplan-Meier survival analysis with the log-rank test was used to determine the proportion of patients surviving free of stroke recurrence at the end of the 3-year follow-up in patients with and without prestroke dementia. The influence of potential predictors of recurrence was assessed with the log-rank test for qualitative variables. Relative risk (RR) was computed through Cox proportional-hazards analysis, which was also used for quantitative variables and multivariate analysis. Only variables with a value of P<0.20 in the previous analysis were considered for multivariate analysis.21 The analyses were computed with SPSS-9.0 for Windows.
We used the same statistical methodology to evaluate the influence of prestroke dementia and of new-onset poststroke dementia on the risk of delayed stroke recurrence among the 6-month survivors.
Using the χ2 test, we evaluated the relationship between prestroke dementia and treatments prescribed at discharge and between dementia at 6 months and 1, 2, and 3 years and treatments received by the patient before the visit and prescribed by the neurologist after the visit. The following variables were evaluated: appropriate antithrombotic treatment, presence of nontreated arterial hypertension, presence of nontreated diabetes mellitus, and presence of nontreated dyslipidemia.
During the 28-week recruitment period, 258 stroke patients ≥40 years of age were admitted to the Stroke Unit. Fifty-six (27.7%) were excluded because of lack of informant. Thus, the study population consisted of 202 patients. Baseline characteristics of the patients are detailed in Table 1.
Death occurred in 89 patients (44%). After exclusion of patients dead before each follow-up visit, the proportion of survivors who were examined was 110 of 142 patients (77%) at 6 months, 96 of 132 patients (73%) at 1 year, 71 of 122 patients (58%) at 2 years, and 74 of 113 patients (65%) at 3 years. Among survivors who did not undergo the visit, the IQCODE score was obtained in all patients but 1 at 6 months and in all but 3 at subsequent contacts. Information on cognitive status before death was obtained for all patients who died within the follow-up period. The presence of dementia could be assessed in all patients but 1 at 6 months and in all but 3 at subsequent contacts. The presumed cause of dementia could be determined only in demented patients who completed at least 1 visit with the neurologist. Data concerning stroke recurrence were obtained for all patients at 6 months and in all but 2 at subsequent contacts. Information concerning the occurrence of stroke recurrence between the last contact and death was obtained for all patients who died between 2 visits.
Of the 202 patients, 33 had preexisting dementia. At 6 months, 44 of the 142 survivors were demented; of those, 15 had prestroke dementia and 29 had new-onset poststroke dementia. The presumed cause of dementia was determined in 39 patients: 19 had AD and 20 had VaD.
Incidence of Stroke Recurrence
During 385 person-years of follow-up, 29 patients developed 33 stroke recurrences (incidence rate, 8.6 per 100 person-years); 2 patients experienced 2 new strokes within the end of the follow-up period, and 1 experienced 3 new strokes. Hospital charts were obtained for all patients but 4 with stroke recurrence. For these 4 patients, the diagnosis of stroke recurrence was based on the neurologist’s opinion. For 2 patients who died before the 6-month visit, information were obtained from the GP; in 2 patients who were followed up until year 3, the sequelae of the stroke recurrence could be assessed during the visit with the neurologist (hemiplegia in 1 patient, cerebellar ataxia in 1 patient).
Mechanisms of Stroke Recurrence
The 33 stroke recurrences consisted of 24 ischemic strokes and 5 cerebral hemorrhages. In 4 patients, we did not obtain enough information to precisely identify the mechanism of stroke recurrence. The 3 patients with an initial cerebral hemorrhage had a recurrent cerebral hemorrhage.
Among the 26 patients with an initial ischemic stroke, 1 patient had 3 new ischemic strokes, 1 had 2 strokes of unknown origin, 1 had 1 stroke of unknown origin and 1 ischemic stroke, and 23 patients had 1 recurrent stroke (20 ischemic strokes, 2 cerebral hemorrhages, and 1 stroke of unknown origin). The presumed cause of the index stroke in the 2 patients who had cerebral hemorrhage as stroke recurrence was lacunar in 1 and cardioembolic in 1.
Delay of Stroke Recurrence
The recurrence occurred within the first 6 months in 12 of 202 patients (6%). Among the 142 survivors at 6 months, 10 (7%) had a stroke between 6 months and 1 year. Among the 132 survivors at 1 year, 8 (6%) had stroke between years 1 and 2. Among the 122 survivors at 2 years, 3 (2.5%) had a stroke between years 2 and 3.
Influence of Prestroke Dementia on Risk of Stroke Recurrence
Within the follow-up period, 3 of 33 patients (9%) with prestroke dementia and 26 of 169 (15%) patients without prestroke dementia had at least 1 stroke recurrence. The cumulative proportion of survivors without stroke recurrence at the end of the follow-up period was 81.5% in patients with prestroke dementia and 80.9% in patients without prestroke dementia (RR, 0.86; 95% confidence interval [CI], 0.26 to 2.85).
Results of the univariate analysis are given in Tables 2 and 3⇓. Multivariate analysis found leukoaraiosis score to be the only independent predictor of stroke recurrence (RR, 1.70; 95% CI, 1.23 to 2.36).
Influence of Poststroke Dementia on Risk of Stroke Recurrence
Among the survivors at 6 months, 6 of 44 demented patients (14%) and 12 of 98 nondemented patients (12%) had at least 1 stroke recurrence. The cumulative proportion of survivors without stroke recurrence at the end of follow-up was 84.8% in demented patients and 87.4% in nondemented patients (RR, 1.25; 95% CI, 0.47 to 3.33).
The proportion of patients who underwent at least 1 stroke recurrence was 2 of 15 patients (13%) with prestroke dementia, 4 of 29 patients (13%) with new-onset poststroke dementia, 2 of 19 patients (11%) with AD, and 3 of 20 patients (15%) with VaD.
The results of the univariate analysis are given in Tables 4 and 5⇓. Multivariate analysis found leukoaraiosis score to be the only independent predictor of stroke recurrence (RR, 1.82; 95% CI, 1.20 to 2.75).
Relation Between Dementia and Treatment
The relation between dementia and treatment at discharge was evaluated in the 175 survivors at discharge; of these 175, 23 had prestroke dementia. There was no relation between prestroke dementia and treatment prescribed at hospital discharge.
The relationships between dementia at 6 months and 1, 2, and 3 years and treatments received before the visit and prescribed during the visit with the neurologist were evaluated in patients who visited the neurologist. We did not find any relationship between treatments and dementia.
Our study has shown that ≈15% of patients had at least 1 stroke recurrence within 3 years after the index stroke and that the only independent predictor of recurrence was leukoaraiosis score, not prestroke or poststroke dementia.
To the best of our knowledge, this is the first study to examine the influence of prestroke dementia on stroke recurrence. Only 1 study3 has evaluated the influence of poststroke cognitive decline on the long-term risk of stroke recurrence, revealing a 2.7 RR in patients demented 3 months after stroke compared with nondemented patients, with an influence of dementia on stroke prevention treatment. Our results did not confirm this finding. This discrepancy may be partly explained by the fact that in demented patients the true incidence of stroke recurrence has been underestimated because of a less valid and reliable diagnosis of stroke based on self-report in these patients. However, in all patients considered unable to answer (because of severe cognitive decline or aphasia), we obtained complementary information from either from the family or the GP. Moreover, to decrease the risk of bias, we did not take into account transient ischemic attack. The discrepancy might also be partly explained by differences in the way treatments were prescribed. Our study did not reveal any difference concerning stroke recurrence prevention between demented and nondemented patients: however, our data are partial because they are available only for patients who underwent the visit with the neurologist and we have no information on the observance, which could be influenced by the cognitive status. Moreover, the sample size is small, especially for some of the subgroups. This could have led to a lack of statistical power for detecting differences between groups and may be a reason for the lack of association described.
In our study, follow-up was performed by either a visit with the neurologist or telephone contact with the patient, the patient’s family, or the GP when the patient could not or refused to undergo the visit, leading to a very small number of patients lost to follow-up. Moreover, the use of survival curve analysis decreases the risk of bias related to the high mortality rate observed in this population.
The diagnosis of dementia may be difficult in stroke patients. Previous studies on poststroke dementia have thus excluded from 6% to 69% of patients,3,8,22–24 leading to a risk of incorrect determination of the real prevalence and incidence of poststroke dementia.25 Use of an informant’s questionnaire gives high specificity and good sensibility of the diagnosis of poststroke dementia.26 In our study, we used an informant’s questionnaire, the IQCODE, for patients who did not undergo the neuropsychological testing because of refusal, severe aphasia, or severe physical or cognitive impairment. However, to improve sensibility, we used results of the neuropsychological evaluation when possible.
Leukoaraiosis appears to be an important prognostic factor in stroke patients. Although the risk of early death after stroke was not increased in patients with leukoaraiosis,27 it was found to increase the risk of death or dependency 3 months after an ischemic stroke,28 to increase the risk of death in patients with lacunar stroke,29 and to adversely influence the cardiovascular prognosis in patients with history of stroke with a higher incidence of cardiac failure, repolarization disturbances, and atrial fibrillation.29 In patients with transient ischemic attack or minor stroke, the risk of stroke recurrence was increased by 60% in patients with leukoaraiosis, with a similar proportion territorial and lacunar stroke.10 Other studies conducted in patients with lacunar stroke also revealed a higher risk of stroke recurrence in patients with leukoaraiosis.29 Our results confirm that white matter changes are associated with a higher risk of stroke recurrence after stroke. Leukoaraiosis was assessed on CT scan performed at the acute stage of stroke because CT scan was available for all patients. Even though MRI is more sensitive for the detection of leukoaraiosis, MRI was not always available in emergency situations, was contraindicated (pacemaker), or could not be performed because of the patient’s agitation, with a risk of bias related to the exclusion of some patients.
Contrary to previously reported results,3 we did not find any influence of dementia on the risk of stroke recurrence or any clear relationship between dementia and treatment of vascular risk factors or use of stroke prevention drugs. The adverse influence of dementia on the risk of stroke recurrence observed in a previously reported study4 might be related to a different therapeutic approach related to patient’s cognitive status. Leukoaraiosis is an independent predictor of stroke recurrence within 3 years after stroke. This finding emphasizes that special attention in the secondary prevention of stroke should focus on patients with leukoaraiosis with or without dementia.
This study was supported by the CH&U de Lille (grant 9306), Association Pour la Recherche et l’Enseignement en Pathologie Neurovasculaire (APREPAN), Association d’Étude et de Recherche sur la Maladie d’Alzheimer (ADERMA), ADRINORD, and an MENRT grant from the French Ministry of Education, Research and Technology for the Research Group on Cognition in Degenerative and Vascular Disorders (grant EA 2691). We thank J. Campos for technical assistance and S. Pollet and C. Mouly for neuropsychological assistance.
- Received March 24, 2003.
- Revision received August 13, 2003.
- Accepted August 28, 2003.
Hénon H, Pasquier F, Durieu I, Godefroy O, Lucas C, Lebert F, Leys D. Preexisting dementia in stroke patients: baseline frequency, associated factors and outcome. Stroke. 1997; 28: 2429–2436.
Moroney JT, Bagiella E, Tatemichi TK, Paik MC, Stern Y, Desmond DW. Dementia after stroke increases the risk of long-term stroke recurrence. Neurology. 1997; 48: 1317–1325.
Hénon H, Durieu I, Guerouaou D, Daems C, Lebert F, Pasquier F, Leys D. Post-stroke dementia: incidence and relationship with pre-stroke cognitive decline. Neurology. 2001; 57: 1216–1222.
Tatemichi TK, Paik M, Bagiella E, Desmond DW, Pirro M, Hanzawa LK. Dementia after stroke is a predictor of long-term survival. Stroke. 1994; 25: 1915–1919.
Barba R, Morin MD, Cemillan C, Delgado C, Domingo J, Del Ser T. Previous and incident dementia as risk factors for mortality in stroke patients. Stroke. 2002; 33: 1993–1998.
Tatemichi TK, Foulkes MA, Mohr JP, Hewitt JR, Hier DB, Price TR, Wolf PA. Dementia in stroke survivors in the Stroke Data Bank cohort: prevalence, incidence, risk factors, and computed tomographic findings. Stroke. 1990; 21: 858–866.
EAFT Study Group. European Atrial Fibrillation Trial: silent brain infarction in nonrheumatic atrial fibrillation. Neurology. 1996; 46: 159–165.
Van Swieten JC, Kappelle LJ, Algra A, van Latum JC, Koudstaal PJ, van Gijn J, for the Dutch TIA Trial Study Group. Hypodensity of the cerebral white matter in patients with transient ischemic attack or minor stroke: influence on the rate of subsequent stroke. Ann Neurol. 1992; 32: 177–183.
Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE III, for the TOAST Investigators. Classification of subtypes of acute ischemic stroke: definitions for use in a multicenter clinical trial. Stroke. 1993; 24: 35–41.
Inzitari D, Giordano GP, Ancona AL, Pracucci G, Mascalchi M, Amaducci L. Leukoaraiosis, intracerebral hemorrhage, and arterial hypertension. Stroke. 1990; 21: 1419–1423.
World Health Association. International Classification of Diseases. 10th revision. Geneva, Switzerland: World Health Organization; 1989; 25–31.WHO typescript document MNH/MEP/87.1.
McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984; 34: 939–944.
Roman GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC, Garcia JH, Amaducci L, Orgogozo JM, Brun A, Hofman A. Vascular dementia: diagnostic criteria for research studies: report of the NINDS-AIREN International Workshop. Neurology. 1993; 43: 250–260.
Huguier M, Flahaut A. Biostatistiques au Quotidien. Paris, France: Elsevier; 2000.
Pohjasvaara T, Erkinjuntti T, Vataja R, Kaste M. Dementia three months after stroke: baseline frequency of different definitions of dementia in the Helsinki Stroke Aging Memory Study (SAM) cohort. Stroke. 1997; 28: 785–792.
Moroney JT, Bagiella E, Desmond DW, Paik MC, Stern Y, Tatemichi TK. Risk factors for incident dementia after stroke: role of hypoxic and ischemic disorders. Stroke. 1996; 27: 1283–1289.
Censori B, Manara O, Agostinis C, Camerlingo M, Casto L, Galavotti B, Partziguian T, Servalli MC, Cesana B, Belloni G, Mamoli A. Dementia after first stroke. Stroke. 1996; 27: 1205–1210.
Inzitari D, Di Carlo A, Pracucci G, Lamassa M, Vanni P, Romanelli M, Spolveri S, Adriani P, Meucci I, Landini G, Ghetti A. Incidence and determinants of poststroke dementia as defined by an informant interview method in a hospital-based stroke registry. Stroke. 1998; 29: 2087–2093.
Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS. Leukoaraiosis in stroke patients: the Copenhagen Stroke Study. Stroke. 1995; 26: 588–592.
Hénon H, Godefroy O, Leys D, Mounier-Vehier F, Lucas C, Rondepierre P, Duhamel A, Pruvo JP. Early predictors of death and disability after acute cerebral ischemic event. Stroke. 1995; 26: 392–398.
Miyao S, Takano A, Teramoto J, Takahashi A. Leukoaraiosis in relation to prognosis for patients with lacunar infarction. Stroke. 1992; 23: 1434–1438.