Neue Seite 1

XVII. European Stroke Conference
Nice, France

Oral Session:

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 16:30 - 16:40 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

01
Donepezil improves executive functions in non-demented CADASIL patients
H. Chabriat H.S.Markus S. Salloway A . Verkkoniemi M. Moline Q. Wang H. Posner M. Dichgans

APHP, University Paris 7

FRANCE

Background CADASIL is a genetic form of subcortical ischemic vascular dementia. It represents a homogenous disease process. We performed a randomised double-blind placebo controlled trial to determine whether the ChEI donepezil improves cognition in cognitively impaired CADASIL patients in the absence or presence of dementia. Methods 168 patients with CADASIL (mean age 54.8 years) were randomly assigned to 10 mg of donepezil per day (n=86) or placebo (n=82) for 18 weeks. Inclusion criteria included a Mini-Mental State Examination (MMSE) score of 10–27 or a trail making test (TMT) B Time score at least 1.5 SD below the mean after adjustment for age and education. The primary endpoint was change from baseline in the score on the V-ADAS-cog (Vascular - AD Assessment Scale, cognitive subscale) at 18 weeks. Secondary endpoints included scores on the ADAS-cog, MMSE, TMT A Time & B Time, Stroop, EXIT25, CLOX, DAD, and CDR-SB. Findings There was no statistical difference between donepezil and placebo in the primary endpoint. In contrast, a significant treatment effect favouring donepezil was detected on the following secondary outcomes: TMT B Time (p=0.005); TMT A Time (p=0.01); and EXIT25 (p=0.02). In the trial, thirty-three (20%) patients were demented as defined by a global clinical dementia rating (CDR) score >1 plus an MMSE<26. Thirty-one percent of subjects had an MMSE score <26, and 81% had an abnormal Trails B score. No significant effect of donepezil was observed on the V-ADAS-cog in patients with MMSE > 26, but a significant treatment effect was detected on the TMT B (p=0.004). Conclusion. Donepezil had no effect on global cognitive scores in CADASIL patients with cognitive impairment. However, significant improvements were noted on several measures of executive function in the whole group of patients as well as in non-demented patients. The present results suggest that donepezil may improve executive functions before the occurrence of dementia in subcortical ischemic vascular diseases (ClinicalTrials.gov number, NCT00103948).

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 16:40 - 16:50 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

02
In vivo detection of cerebral amyloid deposition using 11C PiB PET in patients with Cerebral Amyloid Angiopathy Haemorrhage.
J.V.Ly G.A.Donnan V.L.Villemagne J.A.Zavala H. Ma G. O'Keefe U. Ackerman H. Tochon-Danguy C.C.Rowe

National Stroke Research Institute

AUSTRALIA

Background: Cerebral amyloid angiopathy (CAA) is an important cause of intracerebral haemorrhage (ICH). Although in vivo-diagnosis is usually inferred from clinical and CT or MRI features (such as number of microbleeds), post-mortem remains the only way to make a definite diagnosis. N-methyl-[11C]2-(4’-methylaminophenyl)-6-hydroxybenzothiazole (11C PiB) is a ligand which binds to beta-amyloid both in plaques and vessel walls and may be imaged with PET. We tested the hypothesis that patients with a clinical diagnosis of CAA related ICH (CAAH) will have increased 11C PiB PET uptake. Methodology: Patients with CAAH based on the Boston criteria were studied using 11C PiB PET and compared to age matched controls. Distribution Volume Ratio (DVR) maps were created using Logan graphical analysis and the cerebellar cortex as a reference. Differences between means were assessed by Kruskal_Wallis test. PiB binding in patients with CAAH was correlated with the number of microbleeds by Spearman correlation test. Results: Eleven patients with CAAH of mean age 73.5 yrs (58-93) were studied at a mean of 71 days (6-270) post-ICH and compared to 22 normal controls of mean age 71.8 yrs (59-83). PiB binding was higher in the Neocortical regions with a mean DVR of 1.67+/-0.28 SD in CAAH compared to 1.34+/-0.15 SD in controls (p=0.001). Ten patients had neocortical DVR higher than the 75% percentile of controls. There was good linear correlation between PiB uptake and number of microbleeds (R2=0.9, p=0.01). Conclusion: 11C PiB uptake is higher in patients with CAAH compared to normal aged matched controls. 11C PiB PET may assist the in-vivo diagnosis of CAAH. Equally important is the potential for PIB PET to serve as a surrogate marker for future therapeutic studies.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 16:50 - 17:00 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

03
Association between genetic variants in the lipoprotein receptor-related protein (LRP) and angiotensin-converting enzyme (ACE) genes and cerebral amyloid angiopathy
M. Mendioroz S. Domingues P. Delgado A. Del Río R. Arias M. Hernández E. Santamarina J. Alvarez-Sabín I. Fernández-Cadenas J. Montaner

Vall d'Hebron Hospital.

SPAIN

Background: sporadic cerebral amyloid angiopathy (CAA) is characterized by the presence of Abeta peptide in the wall of cerebral cortical and leptomeningeal vessels. Patients develop recurrent lobar hemorrhages and dementia at advanced age. Currently, there is no prognostic test available to predict CAA outcome. Therefore, our aim was to study the relationship between genetic markers in haemostatic and blood-brain barrier-related pathways with clinical and radiological progression variables in CAA. Methods: in 68 patients with intracerebral hemorrhages who met Boston criteria for CAA (possible 23, probable 45) we genotyped four SNPs (Thr325Ile in TAFI gene, C766T in LRP gene, I/D in ACE gene, C-1562T in MMP-9 gene) by direct sequencing and agarose gel electrophoresis. Leukoaraiosis was assessed by Fazekas scale and the presence of siderosis/number of microhemorrhages was determined on T2-weighted gradient-echo MRI images. Hemorrhage recurrence follow-up was performed during a year after patient recruitment. Results: CT genotype of the C766T LRP gene polymorphism was significantly associated with leukoaraiosis degree (p=0.013) and associated to siderosis (p=0.079). In the ACE gene, the DD genotype showed an association with leukoaraiosis (p=0.004) and the number of hemorrhages (p=0.007). Time to a new recurrent hemorrhagic event was significantly shorter for DD genotype as shown by Kaplan Meier survival analysis (log rank= 0.002). Conclusions: Genetic variation in LRP (C766T) and ACE (I/D) might play a role in CAA progression and therefore become useful prognostic tools. Moreover, it would be interesting to perform functional studies to determine the involvement of these polymorphisms in CAA pathophysiology.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 17:00 - 17:10 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

04
Understanding the heterogeneity in reported rates of pre and post-stroke dementia: systematic review and meta-analysis
S.T.Pendlebury P.M.Rothwell

Stroke Prevention Research Unit, Department of Clinical Neurology, Oxford

UNITED KINGDOM

INTRODUCTION: Up to 30% of dementia cases show vascular pathology at post mortem and stroke increases the risk of dementia. However, estimates of the prevalence of post-stroke dementia (PSD) vary considerably and the reasons for this heterogeneity are uncertain. We therefore did the first quantitative review of studies of PSD. METHODS: Studies of PSD were identified from MEDLINE, EMBASE (to November 2007), relevant reference lists and reviews. Only studies of consecutive cohorts were considered. Pooled estimates of prevalence and heterogeneity between studies were calculated, and analyses were stratified by study setting and other methodological factors. RESULTS: 26 studies were included (22 hospital and 4 population-based) with a total of 4404 stroke patients (median study size = 251) of which 869 had PSD. In hospital-based studies, prevalence of PSD 3-6 months after stroke ranged from 20 to 31.8% (pooled rate = 27.8%, 95% CI 24.8-30., heterogeneity – p=0.01), and from 9.2 to 23.1% (17.9%; 14.6-21.1, p=0.00003) excluding patients with pre-stroke dementia. In population-based studies, prevalence tended to be lower: 17.2% (9.2-25.3, p=0.11) and 9.8% (7.3-12.3, p=0.00001) respectively. Heterogeneity between studies was largely accounted for by variations in patient characteristics, inclusion/exclusion criteria and definitions of dementia. CONCLUSION: Reported rates of pre and post-stroke dementia vary considerably, but much of the heterogeneity can be accounted for by methodological factors. Most studies were too small to reliably quantify risk factors for PSD or differences between stroke subtypes. Larger studies with consistent methodology would facilitate understanding of the epidemiology and determinants of PSD.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 17:10 - 17:20 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

05
Examination of depression in those with mild stroke: Results from the Secondary Prevention of Small Subcortical Strokes (SPS3) Trial
C.L.White L.A.McClure P. Wallace J. Braimah A. Roldan O.R.Benavente
for the SPS3 Investigators

University of Texas Health Sciences Center at San Antonio

USA

Background: Small subcortical strokes represent more than 25% of all ischemic strokes. Most cause mild disability and until recently have been understudied. Consequently poststroke depression (PSD) has seldom been examined in this specific subgroup. The multicenter international SPS3 trial is testing antiplatelet therapy and targets of blood pressure control in a well-defined cohort of patients with subcortical strokes to determine the best strategy to prevent stroke recurrence and cognitive decline. Using data from this cohort, our main objective was to build a predictive model of PSD. Methods: Depression is evaluated using the Patient Health Questionnaire. A multivariable logistic regression model was fitted, to examine the relationship between the covariates of interest and depression, adjusted for time between the index stroke and the 3 month follow-up (approx. 6 months post-stroke). Results: There are 1421 subjects with a 3-month follow-up, mean age of 63 years, 63% male, 33% diabetic and 66% show minimal functional disability (Rankin 0-1). The prevalence of depression at 6 months was 18%. At 15 and 27 months post-stroke, the prevalence was 16% and 13% respectively. In the fully adjusted model, age, gender, diabetes, and disability were all independent risk factors for depression. For each year increase in age at time of stroke, the odds of depression decreased by 0.96 times (95% CI: 0.95, 0.97). Males were less likely than females to be depressed (OR=0.77; 95% CI: 0.58, 1.02). Those with a Rankin of 2-3 were 1.53 times (95% CI: 1.11, 2.11) more likely to be depressed compared to those with a Rankin of 0-1. Finally, diabetics were 1.55 times (95% CI: 1.16, 2.07) more likely to be depressed than non-diabetics. Conclusions: Despite minimal functional disability, community dwelling subcortical stroke survivors report depression that is sustained over time and is at a higher proportion than those in the community without stroke. Recognition of risk factors for depression over time in this sub-group is critical for optimizing post-stroke care.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 17:20 - 17:30 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

06
Management strategies for depression after stroke: is there new evidence since 2004?
M.L.Hackett A.O.House C.S.Anderson

The George Institute for International Health

AUSTRALIA

Background: It is not surprising that depression after stroke is often undiagnosed or inadequately treated given the lack of evidence to guide clinical practice shown in our Cochrane reviews from 2004. Methods: We updated 2 reviews of therapies to manage depression in patients with stroke. We searched 9 electronic databases and 10 clinical trail registers for randomised controlled trials comparing pharmaceutical agents (with placebo), or psychotherapy (with standard care or attention control) to treat or prevent depression in patients with stroke. The primary endpoint was the proportion of patients who met the diagnostic criteria for depression applied in the trials at the end of treatment. Secondary outcomes included depression or mood scores on standard scales, disability or physical function, death, recurrent stroke and adverse effects. Results: 28 trials involving 3010 participants were included. Data were available for 21 pharmaceutical (23 comparisons) and 7 psychotherapy trials. Analyses were complicated by a lack of standardised diagnostic and outcome criteria, and differing analytic methods. There was no clear effect of pharmacological therapy on the prevention of depression or other endpoints. A significant improvement in mood and the prevention of depression was evident for psychotherapy (OR 0.64; 95% CI 0.42-0.98). In the treatment trials there was evidence of benefit of pharmacotherapy in terms of complete remission of depression (OR 0.46, 95% CI 0.33-0.63) and a reduction (improvement) in scores on depression rating scales, but there was also evidence of an associated increase in adverse events (OR 1.67, 95% CI 1.30-1.26). There was no evidence of benefit of psychotherapy in treating depression after stroke. Conclusions: A small but significant effect of psychotherapy (but not pharmacotherapy) was identified for preventing depression after stroke. While there is now evidence to support the use of pharmacotherapy (but not psychotherapy) for treating depression after stroke it should be used with caution in light of the increase in adverse events.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 17:30 - 17:40 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

07
Personal factors as predictors of health-related quality of life (HRQOL) and depression after stroke
C. Donnellan A. Hickey D. Hevey D. O'Neill
Healthy Ageing Research Programme (HARP)

Trinity College Dublin

IRELAND

Background: HRQOL and patterns of depression after stroke have mainly been explained by clinical factors. The evidence regarding personal factors as determinants of stroke outcome remains limited. This longitudinal study examined the influence personal factors have in conjunction with clinical factors on depression and HRQOL after stroke. Methods: Patients (n=153, 49% male, mean age 71 years +/-13.4) were interviewed within 4 weeks of admission (T1) and followed up at 12 months (T2). Personal factors assessed were adaptive strategies (Selection, Optimisation and Compensation 15-item questionnaire (SOC-15)), perceived control (Recovery Locus of Control questionnaire (RLOC)) and socio-demographics. Clinical factors included stroke severity (Orpington Prognostic Score (OPS)) and functional ability (The Nottingham Extended Activities of Daily Living (NEADL)). Outcome measures were the Stroke Specific Quality of Life Questionnaire (SS-QoL) and the Depression Subscale of the Hospital Anxiety and Depression Scale (HADS-D). Results: Univariate analyses showed significant relationships between HRQOL at T2 and HRQOL at T1 (r=.62, p<.001), depression at T1 (r=.41, p<.001) and stroke severity (r=-.30, p<.01). Significant relationships were also found between depression at T2 and depression at T1 (r=.05, p<.001), stroke severity (r=.30, >/=001), perceived control at T1 (r=-.22, p<.05) and functional ability at T1 (r=-.29, p<.01). Multivariate analyses showed that socio-economic status (β21, p<.05) and HRQOL at T1 β62, p<.001) were significant predictors of HRQOL at T2. Depression at T1 β49, p<.001) was a significant predictor for depression at T2. Discussion: This current study indicates that an individual’s initial HRQOL and socio-economic status are important factors in determining HRQOL one year after stroke. Findings also indicate that there is a high probability of depression at one year as a result of depression within the acute phase after stroke.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 17:40 - 17:50 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

08
Telmisartan attenuated brain damage after focal ischemia with an additional effect through PPAR gamma activation in type-2 diabetes model, KKAy mice
M. Masaki J.-M.Li K. Tsukuda J. Iwanami L.-J.Min A. Sakata T. Fujita M. Iwai M. Horiuchi

Ehime University Graduate School of Medicine

JAPAN

Objectives: Angiotensin II type 1 (AT1) receptor blockers are known to prevent onset of stroke by clinical trials and neural damage after stroke in animal model. Telmisartan is unique ARB with agonistic action of peroxisome proliferator-activated receptor gamma (PPARgamma). PPAR gamma is expressed in neurons and microglia, and its activation prevents inflammation. Therefore, we investigated possibility that telmisartan could attenuate ischemic brain damage via both AT1 receptor blockade and PPARgamma activation, using type 2 diabetes mouse model, KKAy. Methods: Eight week-old male KKAy mice were subjected to middle cerebral artery occlusion (MCAO) by microfilament technique. Telmisartan was administrated in water at a low dose (0.375 mg/kg/day) with or without PPARgamma antagonist, GW9662, for 2 weeks before MCAO. Twenty-four hours after MCAO, neurological deficit was measured using neurological score, and brain samples were obtained and stained with 2,3,5-triphenyltetrasodium chloride for evaluating ischemic size. Results: Significant blood pressure changes were not observed in no-treated and telmisartan-, GW9662-, and telmisartan with GW9662-treated groups. KKAy mice exhibited a larger ischemic size compared with wild-type, C57BL6 mice after MCAO. Treatment with telmisartan significantly decreased ischemic size in KKAy mice. This effect was attenuated with co-administration of GW9662. Moreover, KKAy mice exhibited a higher neurological deficit compared with wild-type mice. Administration of telmisartan significantly improved neurological deficit and this effect was decreased by co-administration of GW9662. Conclusions: These results suggested that even a low dose of telmisartan has a protective effect on brain ischemic damage through AT1 receptor blockade and activation of PPARgamma.

Dementia and cognition /Behavior and mood
Date: Wednesday 14 May 2008
Time: 17:50 - 18:00 - Room: Rhodes
Chair: L. Pantoni, Italy and O. Benavente, USA

09
Severity of Cognitive Impairment without dementia (CIND) predicts for incidence of dementia after stroke
K. Narasimhalu S. Ang M.C.Wong H.M.Chang C. Chen

National Neuroscience Institute, Singapore General Hospital Campus

SINGAPORE

Background: Studies have examined the prognostic utility of post-stroke cognitive performance, namely, dementia, cognitive impairment no dementia (CIND), and no cognitive impairment (NCI). However, few studies have examined CIND severity in predicting dementia after stroke. Methods: Consenting patients underwent a baseline neuropsychological assessment between 3-4 months after stroke and were followed up annually for up to 5 years. A diagnosis of dementia was reached by using DSM-IV criteria. Cox proportional regression was performed to determine hazards ratios (HR). Results: 419 non-demented patients were followed for a mean of 3.2 years. 25 patients developed dementia. In agreement with previous studies, CIND patients had a significantly higher risk of dementia. A median split of CIND patients resulted in two subsets: CIND with 1 or 2 domains impaired (CIND-mild), or CIND with 3-6 domains impaired (CIND-severe). Nineteen CIND-severe (N=106) patients developed dementia compared to four CIND-mild (N=101) and two NCI patients (N=212). In multiple regression analysis including NCI patients, age alone (HR 1.14, p=0.03) was an independent predictor of dementia in CIND-mild patients, while age (HR 1.10, p=0.001), previous stroke (HR 3.03, p=0.02), and cognitive impairment (HR 8.91, p=0.001) were independent predictors of dementia in CIND-severe patients. In analysis restricted to CIND patients, older age (HR 1.07, p=0.008) and CIND-severe (HR: 4.17, p=0.005) were independent predictors of dementia. Conclusions: Patients with CIND-severe are at higher risk of developing post-stroke dementia whilst NCI and CIND-mild patients share a minimal risk.