P2-261 Plasma levels of amyloid beta-protein 42 are increased in women with mild cognitive impairment

$306

Poster Session P2: Epidemiology and Risk Factors of Alzheimer's Disease

complaint have been shown to increase the likelihood of conversion to DAT in non-demented, but cognitively impaired individuals, Objective: In a 3year longitudinal study, we aimed to investigate the predictive capacity of e4 and subjective memory complaint in asymptomatic individuals, Methods: Seventy individuals with no baseline memory impairment were examined comprising 30 participants with at least one e4 allele, and 40 with no e4 allele present. Participants were administered a memory complaint questionnaire, standard neuropsychological measures, and the Paired Associate Learning (PAL) subtest of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The latter is known to be sensitive to the memory changes observed in the preclinical stages of Dementia of the Alzheimer's Type, Results: The findings suggest that Apo E status is not associated with subjective memory complaint, and neither factor predicts memory decline over three years in healthy, non-depressed elderly individuals, Conclusions: Neither Apo E status nor memory complaint increase the risk of future memory decline in asymptomatic individuals. The results are presented in the context of normal ageing and Mild Cognitive Impairment.

•

VIENNA TRANSDANUBE AGING (VITA): PRECLINICAL ALZHEIMER DEMENTIA

Peter Fischer* 1,2, Susanne Jungwirth 2, Silvia Weissgram 2, Karl Heinz Tragl 2. 1Department of General Psychiatry, University Hospital for

Psychiatry, Vienna, Austria; 2Ludwig Boltzmann Institute forAging Research, V~enna,Austria. Contact e-mail: [email protected] Background: The definition of mild cognitive impairment MCI has 4 requirements: a) impaired memory performance, b) normal general cognitive functions, c) subjective memory complaints, and d) a) no dementia, i.e. no impairment of activities of daily living. We investigated these requirements in the prediction of cognitive decline and Alzheimer dementia (AD), particularly, in the setting of a community-based age-cohort study in Vienna, called VITA. Since age is by far the most certain predictor of cognitive decline in the elderly, VITA minimizes the variance of this already proven risk factor to 0.2 years. VITA investigates the whole cohort of elderly born between May, 1925 and June, 1926, living in the geographical districts on the left shore of the Danube (Vienna Transdanube). Objective(s): The main aim of the VITA is the prediction of incident cases of AD in a community-based age-cohort. Whether a patient suffers from AD is diagnosed in a longitudinal setting every 30 months. Methods: Prediction is tried multidisciplinary: short screening tests, extensive neuropsychologicai measures, quantitative data from MRT, blood biochemistry, selected genetic data, psychiatric symptoms, and medical data. Various variables can be compared with each other concerning sensitivity and specificity of early prediction with high statistical power. The presentation is based on results of 606 individuals who took part in the baseline investigation of the VITA study, 300 of whom underwent follow-up investigation up to now. Results: At baseline 141 subjects (23.3%) had cognitive impairment of more than 1,5 standard deviations below the mean of the age-cohort. About 15% of the cohort showed cognitive impairment without memory impairment, 8.3% had selective or non-selective memory impairment. Only 3 subjects (0.5%) fulfilled Petersen criteria "MCI-amnestic". Conclusions: AD was poorly predicted by the MCI-amnestic criteria of Petersen et al (2001). Small modifications of these criteria showed significantly higher predictive values.

~ - ]

DEPRESSIVE SYMPTOMS AND MILD COGNITIVE IMPAIRMENT IN THE CARDIOVASCULAR HEALTH STUDY

Deborah E. Barnes* 1, Lewis H. Kniler 2, Jeff D. Williamson 3, Kristine Yaffe 1.1 University of California, San Francisco, San Francisco,

CA, USA; 2 University of Pittsburgh, Pittsburgh, PA, USA; 3Wake Forest University School of Medicine, Winston-Salem, NC, USA. Contact e-mail: barnes @medicine,ucsf edu Background: Several prior studies have found that depressed eiders have an elevated risk of developing dementia, but it is not clear whether this association extends to mild cognitive impairment (MCI), which is often a

precursor to dementia. Some have hypothesized that vascular disease may underlie both mood and cognitive disorders in the elderly. Objective: To determine whether depressive symptoms are associated with an increased risk of MCI among Cardiovascular Health Study (CHS) participants and, if so, whether the association is attributable to vascular disease. Methods: CHS is a prospective, muhiceuter study of risk factors for coronary heart disease and stroke. Our analyses included the 2,813 white participants aged >65 years without dementia or cognitive impairment (defined as Modified Mini-Mental score <80) at baseline. High number of depressive symptoms was defined as shortened Center for Epidemiologic Studies-Depression (CES-D) scale score >8. MCI was diagnosed after approximately 10 years of follow-up based on standard clinical criteria (e.g., neurological exam, neuropsychologicai testing and brain MRIs) that were adjudicated by a committee of experts. In our analyses, vascular disease was defined as clinical cerebrovascular disease (history of stroke or transient ischemic event), subclinical disease (carotid artery stenosis >50%) or cortical infarcts or white matter grade >3 on brain MRIs. Results: Of the 2,813 study participants, 457 (16.3 %) had high depressive symptoms at baseline and 350 (12.4%) developed MCI during the 10-year follow-up period. Depressive symptoms were more common at baseline among subjects who developed MCI compared to those who remained coghitively normal (22.9% vs. 14.9%, p < 0.001). In logistic regression analyses, depressed subjects had significantly higher odds of developing MCI (OR = 1.57; 95% CI: 1.162.13) after adjustment for age, sex, education, baseline cognitive function and APO-E status. In addition, this association was stronger in women (OR = 1.83; 95% CI: 1.28-2.62) than men (OR = 1.10; 95% CI: 0.60-2.00). The association between depressive symptoms and MCI was not attributable to vascular disease measures. Conclusions: In a large, prospective study of elders, depressive symptoms at baseline were associated with an elevated risk of developing MCI. The association was not explained by clinical or subclinical vascular disease.

•

PLASMA LEVELS OF AMYLOID BETA-PROTEIN 42 ARE INCREASED IN WOMEN WITH MILD COGNITIVE IMPAIRMENT

Luca Giliberto* 1, Andrea Assini 1, Sergio Cammarata 2, Monica Colucci 1, Roberta Borghi 1, Alessandra Piccini 1 Patrizio Odetti 3, Franca Dagna-Briccarelli 4 , Alessandra Argusti 4, Massimo Tabaton 1.

1University of Genoa, Department ofNeurosciences, Ophtalmology and Genetics, Genoa, Italy; 2Department of Neurology, Galliera Hospital, Genoa, Italy; 3University of Genoa, Department of Internal Mdicine, Genoa, Italy; 4Department of Human Genetics, Galliera Hospital, Genoa, Italy. Contact e-mail: [email protected] Background: Accumulation in the brain of soluble aggregates of betaamyloid 42 (A~42) is the major pathogenic event of Alzheimer's disease (AD). In familial early-onset AD this event is likely the result of A~42 overproduction, but in the most common sporadic late-onset form of the disease the mechanisms of AIM2 accumulation are unknown and various factors are suspected to produce its accumulation, through increased aggregation and defective clearance. Elderly subjects that developed a sporadic AD-type dementia within four years had higher amounts of plasma AI342 than controls, suggesting that elevated plasma levels of A~42 is an early sign of the pathological process, while those levels normalize in already demented patients. Some evidence indicates that Mild Cognitive Impairment characterized by isolated memory loss (anmestic MCI) is incipient AD: MCI patients develop AD-type dementia at a higher rate than normal controls and all autopsied amnestic MCI cases show cerebral lesions consistent with the pathologic diagnosis of AD. Thus, amnestic MCI is the ideal condition to study the relationship between plasma levels of Af3 species and late-onset sporadic AD. Objectives: We studied plasma levels of AI~ species in amnestic MCI patients, compared to aged-matched controls, to verify their relevance in pathogenesis of AD. Methods: We measured the plasma concentrations of A1342 and A1340 in 88 MCI patients and in 72 age-matched normal controls, paying attention to variables such as plasma cholesterol, creatinine and apolipoprotein E genntype, known to influence AI~ production, aggregation and excretion. Conclusions: We found a significant

Poster Session P2: Epidemiology and Risk Factors of Alzheimer's Disease increase of AIM2 plasma levels in women with MCI, in comparison to the affected males and to cognitively normal age-matched subjects. The elevation of A1342 plasma levels in women with MCI was independent of variables as education, apolipoprotein E genotype, cholesterol and creatinine plasma concentrations, as well as of hemoglobin content. Increase of AI342 plasma concentrations in women with MCI links the overproduction of this peptide with sporadic late-onset AD. Furthermore, these findings represent the biological explanation for the gender-dependent increased incidence of late-onset AD in women reported by epidemiological studies, and suggest a role of estrogens in AIM2 accumulation.

•

P R E V A L E N C E O F C O G N I T I V E I M P A I R M E N T IN VISUALLY IMPAIRED ELDER TEXANS

Patricia Heyn* 1, Rosa A. Tang 1, Steven S. Owen 2, Syed Azhar 1, Mukalla Raji I . 1University of Texas Medical Branch, Galveston, IX, USA;

2University of Texas Health Science Center, San Antonio, TX, USA. Contact e-mail: paheyn @utmb.edu Currently Editing: Abstract, Control # 04-A-1774-ALZ Background: Mild Cognitive Impairment (MCI) is associated with poor health outcomes including Alzheimer's disease (AD); recent studies have also demonstrated associations between visual impairments and increased risk of MCI. Objecfives: The purpose of the present study is to examine visual and cognitive impairment in a community sample of older adults from Texas who attend the University of Texas Medical Branch for their health care. A primary goal of our study is to explore and detect the prevalence of MCI as well as the potentially modifiable non-genetic factors, such as heart disease, diabetes, high cholesterol, diet, levels of physical activity/function, body mass index, hypertension, depression, and other related factors to AD. Methods: Study outcomes are based on 1) biographic information, 2) physical and physiological values, 3) Cognitive Status and Function, 4) Mood, 5) Quality of Life, 6) Lifestyle Factors, and 7) Visual Data Battery. Results: Descriptive statistics will summarize demographics, frequencies, percentages, and means. The central group of analyses will use structural equation modeling (Figure) to evaluate proposed causal pathways to quality of life. In this model, Visual Function is seen to have not only a direct influence on QOL, but also indirect pathways mediated through other variables. AMOS 5.0 will be used for confirmatory factor analysis and SEM. Using MacCallum, Browne, and Sugawara's SAS program, a power analysis was performed for the SEM. Assuming alpha = 0.05, power = 0.80, root mean square approximation of error (RMSEA) for H0 = 0.00, RMSEA for HA = 0.08, then minimum n = 70. To account for attrition, we will oversample by about 12%, so that N = 80. Conclusion: Identifying a subset of visual symptoms and signs that are highly predictive of MCI could potentially yield useful information, upon which intervention programs can be developed at early stage for those adults at highest risk of dementia.

~

HIGH SERUM TOTAL CHOLESTEROL IN MID- AND LATE-LIFE ASSOCIATED W I T H INCIDENT D E M E N T I A IN YOUNGER, BUT NOT OLDER COHORTS OF W O M E N

Michelle M. Mielke .1 , Peter P. Zandi 1, Deborah Gustafson 2, Karin Sjogren 2, Xinxin Guo 2, Ingmar Skoog ~. 1Johns Hopkins University,

Baltimore, MD, USA; 21nstitute of Clinical Neuroscience, Sahlgrenska University Hospital, Goteborg, Sweden. Contact e-mail: mmielke @jhsph.edu Background: Several studies have examined the association between high total serum cholesterol and dementia, with conflicting results. These observed differences may be due to the timing of the cholesterol measurements, i.e. mid-fife versus late-life. Objective: To examine the importance in the timing of high cholesterol exposure and incident dementia over a 32-year follow-up. Methods: A population-based sample of 1467 Swedish women aged 38, 46, 50, 54, and 60 years in 1968 were followed for 32 years. Anthropometric, nettropsychiatric, laboratory, and other assessments were conducted in 1968, 1974, 1980, 1992, and 2000. Dementia from 1992-2000

$307

was diagnosed according to DSM-III-R. Information on those lost to followup was collected from case records, the hospital linkage system, and death certificates. Fasting serum total cholesterol was obtained at all assessments. Results: We found a cohort effect in the cholesterol-dementia relationship. Among women born in 1918, 1922, and 1930, increasing cholesterol in 1968 (OR: 1.21, 95%CI: 0.97-1.52), 1974 (OR: 1.05, 95%CI: 0.84--1.32), 1980 (OR: 1.30, 95%CI: 1.05-1.62), and 1992 (OR: 1.25, 95%CI: 0.941.65) was associated with incident dementia between 1992 and 2000 using logistic regression and controlfing for age. However, among older women born in 1908 and 1914, we did not find an association between cholesterol and dementia risk. This cohort effect was not explained by differences in education. Conclusion: We observed that increasing total cholesterol levels both in mid- and late-fife were associated with higher risk of dementia, but only among those born in the later cohorts. Thus, cohort effects, perhaps due to differences in socioeconomic or nutritional status, may therefore be related to the conflicting reports concerning the cholesterol-dementia association. These results suggest further research is warranted concerning how cohort effects may modify the influence of high cholesterol, and other cardiovascular risk factors, on dementia risk. This work was supported by the Alzheimer's Association, Stephanie B. Overstreets Scholars IIRG-00-2159.

•

S M O K I N G A F F E C T S THE P H E N O T Y P E OF ALZHE1MER'S DISEASE

Marwan N. Sahbagh *~ , Suzanne L. Tyas 2, Shawn C. Emery 3, Lawrence A. Hansen 3, Michael E Alford 3, Richard T. Reid 4, Pietro Tiraboschi 3, Leon J. Thai 3. 1Sun Health Research Institute, Sun City, AZ, USA;

:University of Kentucky, Lexington, KY USA; 3University of California-San Diego, La JoUa, CA, USA; 4Banek Clinical Research Center, San Diego, CA, USA. Contact e-mail: [email protected] Background: A possible link between smoking and Alzheimer's disease (AD) has generated much interest and research. Epidemiological evidence suggests smoking increases the risk of AD while nicotine appears to be neuroprotective in vitro. Objective(s): To determine how cigarette smoking impacts the clinical and neuropathological phenotype of AD. Methods: We performed a clinico-pathological analysis to compare the phenotype of AD in 46 never smokers, 47 former smokers, and 15 active smokers who were prospectively characterized and longitudinally followed to autopsy. Active smoking is defined as routinely smoking cigarettes at the time of onset of AD symptoms. Mean disease duration; scores on the Mini-Mental State Exam, Mattis Dementia Rating Scale, and Blessed Information, Memory, and Concentration Scale; Braak stage; and counts of neuritic plaques, total plaques, and neurofibrillary tangles were examined by smoking history. The effect of apolipoprotein E (APOE) genotype on these associations was also analyzed. Parameters were assessed in bivariate tests and multivariate models controlling for age, gender, education, and presence of an APOE-e4 allele. Results: Active smokers were significantly younger at onset of AD and at death. After adjusting for covariates in multivariate models, age at death remained significantly associated with smoking status, an effect specific to APOEe4 non-carriers. There was no association between smoking status and any cognitive or pathological variable in either bivariate or multivariate analyses and results were not influenced by APOE-e4 status. Conclusions: Tobacco use may alter the disease characteristics of AD in the absence of any differences in cognitive performance or neuropathological markers.

•

EVALUATION O F S E L E C T I O N BIAS IN A COMMUNITY-BASED D E M E N T I A AUTOPSY CASE SERIES

Kate L. Simpson*. University of Washington, Seattle, WA, USA. Contact

e-mail: [email protected] Background: Definite Alzheimer's disease (AD) can only be diagnosed neuropathologically. However, neuropathological frequency estimates of dementia may be biased if the autopsy subjects are not representative of all dementia subjects within a target population. Objective(s): To identify