C ognitive impairment in the context of stroke or transient ischemic attack (TIA) is a prototype of vascular cognitive impairment (VCI). Our recent in vivo study using carbon-11-labeled Pittsburgh compound B ( 11 C-PiB) positron emission tomography (PET) found that ≈30% of subjects with poststroke/TIA cognitive impairment harbored Alzheimer's pathology.1 Although some studies suggested that in patients with clinical Alzheimer's disease (AD), comorbid cerebrovascular disease was associated with a more rapid cognitive decline, 2,3 it is still unknown whether concurrent presence of Alzheimer's pathology is associated with a faster cognitive deterioration in patients with poststroke/TIA cognitive impairment.In this study, we compared the longitudinal cognitive changes between VCI patients with and without AD-like amyloid-beta (Aβ) deposition measured using 11 C-PiB PET. We hypothesized that over 3 years, PiB-positive VCI (mixed VCI [mVCI]) subjects would experience a more rapid and continuous course of cognitive deterioration, resulting in more severe cognitive impairment than those who were PiB-negative (pure VCI [pVCI]).
Methods
SubjectsSubjects of this study were participants of the ongoing Stroke Registry Investigating cognitive Decline (STRIDE) study. 1 In the Background and Purpose-We hypothesized that comorbid amyloid-beta (Aβ) deposition played a key role in long-term cognitive decline in subjects with stroke/transient ischemic attack. Methods-We recruited 72 subjects with cognitive impairment after stroke/transient ischemic attack to receive Carbon-11-labeled Pittsburgh compound B positron emission tomography. We excluded subjects with known clinical Alzheimer's disease. Those with and without Alzheimer's disease-like Aβ deposition were classified as mixed vascular cognitive impairment (mVCI, n=14) and pure VCI (pVCI, n=58), respectively. We performed Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment to evaluate global cognition and cognitive domains (memory, visuospatial function, language, attention, and executive function) at 3 to 6 months (baseline) and annually for 3 years after the index event. We compared cognitive changes between mVCI and pVCI using linear mixed models and analysis of covariance adjusted for age and education. Results-Over 3 years, there were significant differences between mVCI and pVCI on change of MMSE score over time (group×time interaction, P=0.007). We observed a significant decline on MMSE score (P=0.020) in the mVCI group but not in the pVCI group (P=0.208 6 were performed on subjects at 3 to 6 months (baseline) after the index event to index general cognition. Then subjects were further invited to return for annual follow-up. This study analyzed the data collected ≤3 years.This substudy included a sample of 72 subjects with cognitive impairment (CDR grade ≥0.5) from the STRIDE study. Clinical diagnoses were made by 2 neurologists specialized in dementia (V.C.T.M. and L.A.). Thirty-six subjects had CDR grade of ≥1 and met the criteria for dementia acco...