| Dementia is a main cause of dependency in stroke survivors. Poststroke dementia (PSD) includes any dementia after a stroke, irrespective of its causes, e.g. vascular dementia (VD), Alzheimer's disease (AD) or mixed dementia. A huge increase in prevalence and burden of PSD is likely to happen because of the decline in mortality after stroke and ageing of populations in the coming decades worldwide. In this thesis, we reported the risk factors for early PSD and delayed poststroke cognitive decline, and the prevalence of concurrent amyloid pathology as identified by Pittsburgh compound B (PIB) positron emission tomography (PET) in PSD. Understanding the risk factors of PSD will help to devise preventive and treatment strategies that may reduce the burden of PSD.;Objectives.;Study 1: Risk factors of early PSD. Mechanisms explaining poststroke early and delayed cognitive decline are complex. We set up a STroke Registry IVEstigating COGnitive decline (STRIVE-COG) among Chinese stroke survivors. Study 1 reported the findings on risk factors for early PSD.;Study 2: Prevalence and predictors for delayed (15-18 months) cognitive decline after stroke. Having a stroke increases the risk of delayed dementia. However, mechanisms accounting for the cognitive decline are uncertain. We investigated the predictors for delayed poststroke cognitive decline.;Subjects and Methods.;Study 1: We recruited consecutive stroke or transient ischemic attack (TIA) patients admitted to our acute stroke unit over 1 year. We performed neuropsychological assessment 3-6 months poststroke. We investigated the association between clinical and structural neuroimaging features with early PSD. We performed PIB positron PET among a subset of subjects with early PSD.;Study 2: We performed neuropsychological assessment at baseline (i.e. 3-6 months poststroke) and at 15-18 months poststroke. We defined cognitive decline as a drop of ≥ 3 points in the mini-mental state examination and/or increment in ≥ 1 grading of the clinical dementia rating scale. We investigated the association between cognitive decline with baseline clinical, neuropsychological, and neuroimaging features (white matter changes [WMC] severity, old lacunar infarct, global atrophy, frontal lobe atrophy [FLA], parietal lobe atrophy, medial temporal lobe atrophy). Among a subset of subjects (n=18) with PSD at baseline, we investigated the influence of AD-like PIB retention upon the rate of cognitive decline.;Results.;Study 1: Prevalence of early PSD among all recruited subjects (n=549) was 15.3% (n=84). Apart from age and female gender, multivariate regression analyses showed that risk factors for early PSD were presence of acute lacunar (odds ratio [OR] 2.725, 95% confidence interval [CI] 1.364-5.434, P=0.004) or non-lacunar infarct (OR 2.809, 95%CI 1.124-6.410,P=0.014) over no acute infarct apparent on neuroimaging, WMC severity (OR 1.120, 95% CI 1.037-1.210, p=0.004), FLA (OR 2.596, 95% CI 1.080-6.241, p=0.033), and global brain atrophy (4th quartile vs 1st quartile, OR 3.096, 95% CI 1.374-6.993, p=0.006). Among 19 subjects with early PSD who had PIB PET, 6 (31.6%) had AD-like PIB retention.;Study 2: Among 452(82.3%) subjects who had completed the study, cognitive decline occurred in 73 (16.2%) subjects. Age, education, and multiple old lacunar infarcts independently predicted cognitive decline. Recurrent stroke occurred only in 5.1% and was not associated with cognitive decline. Progressive cognitive decline occurred in 41.7% and 33.3% of PIB negative (n=12) and PIB positive (n=6) PSD patients, respectively (p=0.731).;Conclusion.;Study 1: Apart from age, female gender, and presence of acute infarct evident in neuroimaging, chronic brain changes (WMC, global brain atrophy, FLA, and concurrent AD pathology) are associated with early PSD.;Study 2: Age, education, and multiple old lacunar infarcts predicted cognitive decline at 15-18 months poststroke. Concurrent AD-like lesion is not necessary associated with a rapid cognitive decline. |
