Identifying factors that moderate the onset and rate of prodromal cognitive impairment in dementia

Identifying factors that moderate the onset and rate of prodromal cognitive impairment in dementia

Oral O3-02: Neuropsychology healthy controls with respect to confrontation naming and semantic fluency (controlling for anxiety and depression scores ...

44KB Sizes 0 Downloads 2 Views

Recommend Documents

No documents
Oral O3-02: Neuropsychology healthy controls with respect to confrontation naming and semantic fluency (controlling for anxiety and depression scores and estimated premorbid intellectual capacity). When compared with NMCs at baseline, there was a trend for a significantly higher proportion of healthy controls defined as MCs at baseline to convert to the clinical diagnostic category of Mild Cognitive Impairment (MCI; objective functional impairment) 18 months later at followup (1.3% NMCs vs. 3.8% MCs). PIB PET analysis performed on n ¼ 81 NMCs and n ¼ 91 MCs at baseline showed that there was no significant difference in the proportion of NMCs and MCs above a designated threshold of brain amyloid deposition (33% NMCs vs. 30% MCs). Conclusions: Taken together, these findings have important implications for how we should conceptualize memory complaints vs. objective neuropsychological and neuroimaging findings in those over the age of 60. More specifically, subjective perception of memory status may a useful adjunct to other findings when seeking to identify those most at risk of age-related cognitive change. O3-02-02


Stuart W. S. MacDonald1, Valgeir Thorvaldsson1, Boo Johansson2, Lars Ba¨ckman3, 1University of Victoria, Victoria, BC, Canada; 2University of Gothenburg, Gothenburg, Sweden; 3Karolinska Institutet, Aging Research Center, Stockholm, Sweden. Contact e-mail: [email protected] Background: In the present study, we used data from two Swedish population-based longitudinal studies to (a) estimate time of onset and rate of accelerated cognitive decline across various domains during the prodromal phase of dementia, as well as (b) identify whether key factors linked to AD risk (years of education) further moderated both the onset and rate of accelerated decline. Methods: The H70 study has an age-homogeneous sample (113 cases and 272 controls), with cognitive performance initially assessed at 70 years of age, with up to 12 retest measurements spanning a 30-year period. Complementary data from the Kungsholmen Project (KP), including an ageheterogeneous sample (417 cases and 783 controls) with an average age of 82.14 years (SD ¼ 4.98) at initial assessment, were used to characterize cognitive change for up to 5 follow-up assessments across 13 years. We fit a series of linear mixed change-point models in 1-month increments to identify the average transition point that best characterized acceleration of decline prior to dementia onset (compared to normative age changes), and then evaluated how select moderators influenced the onset and rate of decline. Results: Despite differences in study samples and designs, our results were remarkably similar in demonstrating onset of accelerated decline around 10 years prior to diagnosis for fluid abilities (e.g., episodic memory, perceptual speed) and approximately 5 years pre-diagnosis for crystallized abilities (e.g., verbal ability, clock reading). Years of formal education moderated both the onset and rate of decline for select cognitive outcomes, with more education linked to later onset but accelerated decline. Conclusions: Protracted change points (>7 years) were observed for cognitive correlates of dementia pathology, but faster rates of decline were observed for tasks that incorporate both fluid and crystallized abilities. This distinction supports claims that well-preserved abilities may be most discriminative, but only in close proximity to incident diagnosis. Consistent with the cognitive reserve hypothesis, individuals who were more educated showed later disease onset but faster cognitive decline. This pattern may help explain the modest association between AD neuropathology on the one hand, and cognitive impairment on the other. O3-02-03


Jennifer H. Barnett1, Andrew Blackwell1, Philip Scheltens2, Gunhild Waldemar3, Peter Johannsen4, Magda Tsolaki5, Rik Vandenberghe6, Lars-Olof Wahlund7, Frans Verhey8, Pieter Jelle Visser8, 1University of Cambridge & Cambridge Cognition Ltd, Cambridge, United Kingdom; 2VU Medical Centre, Amsterdam, Netherlands; 3Rigshospitalet, Copenhagen University Hospital, Copenhagen,


Denmark; 4Rigshospitalet, Copenhagen University Hospital, Copenhagen, United Kingdom; 5Aristotle University of Thessaloniki, Thessaloniki, Greece; 6Universitair Ziekenhuis Gasthuisberg, Leuven, Belgium; 7Karolinksa Institute, Stockholm, Sweden; 8University of Maastricht, Maastricht, Netherlands. Contact e-mail: [email protected] Background: Episodic memory impairments are cardinal features of Alzheimer’s disease (AD) and other dementias. Sensitive assessment of memory and other cognitive function may help predict deterioration into dementia in individuals with mild cognitive impairment (MCI). Large scale, prospective studies are needed to assess the utility of cognitive assessment alongside other biomarkers in predicting and characterising cognitive decline in elderly populations. Methods: Cognitive function was assessed in volunteers taking part in EDAR, a Europe-wide prospective longitudinal study assessing the potential of biomarkers, including cognitive assessment, in the early diagnosis of Alzheimer’s disease. Memory, attention and executive functions were measured in more than 200 participants at baseline and after 9 and 18 months using the computerised CANTAB battery. Results: Patients with AD, but not those with MCI, showed significant impairments in motor control, choice reaction time, and object naming at baseline. On a visuospatial associative learning test, the CANTAB Paired Associates Learning, the AD and ‘other dementia’ groups made more errors than the MCI group, who themselves made more errors than healthy controls. These differences remained after controlling for age, however the poorer performance of AD patients in planning and working memory were explained predominantly by age effects. No cognitive differences were found at baseline between AD and the ‘other dementia’ group. Longitudinal follow-up of cognitive function demonstrated that associative learning appears particularly sensitive to the pathological process of AD, with significant worsening over time in the AD group and in a subset of MCI patients. Conclusions: Measures of episodic memory and attention differentiate dementia patients from both healthy elderly volunteers and individuals with mild cognitive impairment. In contrast these data suggest a relative sparing of executive function in patients with AD and other dementias. Decline in associative learning reflects a rapid worsening of cognitive function in MCI patients that may represent the prodromal stages of AD. O3-02-04


Joshua R. Steinerman1, Charles B. Hall1, Martin J. Sliwinksi2, Wendy Ramratan1, Herman Buschke1, Richard B. Lipton1, 1Albert Einstein College of Medicine, Bronx, NY, USA; 2Pennsylvania State University, University Park, PA, USA. Contact e-mail: [email protected] edu Background: Development of therapeutic interventions to delay or prevent the onset of Alzheimer’s Disease (AD) is hampered by the modest reliability of traditional neuropsychological outcomes and imprecise estimation of cognitive decline. These limitations could be overcome by modeling changes in optimal performance, a latent construct which is approximated but never achieved. Methods: In a subset of 40 dementia-free Einstein Aging Study participants, we administered an experimental cognitive battery on 3 sessions within a one-week measurement burst. Herein, we focus on across-session performance on the N-back task, including the simple 1-back variant and complex 2-back variant gauging working memory and executive attention reaction times over 64 trials. We separately considered four combinations of task complexity and performance level: Simple-peak (90th percentile speed over all correct trials for each 1-back session); complex-peak (90th percentile, 2-back); simple-poor (10th percentile, 1-back); and complex-poor (10th percentile, 2-back). For each combination, we determined the coefficient of variation (CV) , intraclass correlation coefficient (ICC) and the pattern of session-to-session group-level change, hypothesized to reflect a balance of neuroplasticity and state fluctuation. Results: Mean age was 79.5 (SD ¼ 6.4) and the sample was 60% female. Measures of peak performance demonstrated low CV and high ICC, whereas complex-poor performance had the highest CV and lowest ICC (Table). Furthermore, we found that simple-optimal performance did not change across sessions. Complex-optimal performance improved steadily from session 1 to 3. Complex-poor performance