Poster Session PI : Diagnosis and Disease Progression - Clinical dementia (p < 0.01) and a lower cognitive impairment (p = 0.01), were more indipendent in ADLs and IADLs (p < 0.01) and were less likely to have a high comorbidity (p < 0.01). Gender distribution, level of education, type and duration of disease did not differ significantly between survivors and deceased. The multiple logistic regression showed that only ADL results were independently associated with mortality. During the observation period seven patients were admitted to nursing home (institutionalization rate 8.5%). Gender, level of education, duration and severity of disease, cognitive impairment, comorbidity, behavioural disturbances, age of caregiver did not significantly differ between institutionalized and living-at-home patients. Demented institutionalized patients were significantly older (10 < 0.01), more dependent in ADLs and IADLs (p < 0.05) and assisted by caregivers with a higher level of stress (p < 0.01). The multiple logistic regression showed that only age of patients and caregiver stress were independently associated with institutionalization. Conclusions: Increasing functional disability decreases survival in old demented people, Caregiver distress is one of the most important factors that threaten continuing care at home.
SURVIVALFROM M E M O R Y S Y M P T O M ONSET: A C O M P A R I S O N OF INDIVIDUALS W I T H D E M E N T I A AND C O G N I T I V E IMPAIRMENT. T H E CACHE COUNTY STUDY
Elizabeth Klein t , Chris Corcoran* 1, JoAnn Tschanz I , Maria Norton ~, Kathleen Welsh-Bohmer2, John Breitner 3, Peter Zandi 4, Constantine Lyketsos 4. 1Utah State University, Logan, UT, USA; 2Duke
University, Durham, NC, USA; SUniversity of Washington, Seattle, WA, USA; 4The Johns Hopkins University, Baltimore, MD, USA. Contact e-mail: [email protected]
Background: Dementia is an important risk factor for mortality, with approximately 8-year survival post dementia onset in prevalence samples. Data from prevalence samples, however, are difficult to interpret due to left censoring. Mortality estimates from incidence samples may better explain survival after disease onset. Objective(s): In a population panel of incident dementia, we examined survival duration from time of memory onset to death or censoring, as compared with survival in those with Cognitive Impairment No Dementia (CIND). To reduce the impact of left censorship, we supplemented our panel of dementia and CIND subjects with those ascertained via an informant clinical interview completed after death. Methods: The Cache County Study is a longitudinal population-based study of Alzheimer's and other dementias. Cases of incident dementia (n = 287) and were identified from clinical examination at each study wave or if they did not survive, via an informant based clinical interview querying about cognitive symptoms and limitations in dally living activities. The CIND group (n = 270) was comprised of geropsychiatrist/neuropsychologist ratings of living and post-mortem dementia interviews. We examined Kaplan-Meier (K-M) plots for the participant groups stratified by 10-year age groups. To examine the effect of participant group (dementia/CI), age, education, presence of E4 at Apolipoprotein E (APOE), gender, and selected medical conditions on survival, we used Cox proportional hazards models. Results: Individuals with dementia had significantly shortened survival times compared with those with CIND [logrank p < 0.00001;median survival dementia = 6.41 years (95% CI = 5.89-6.94); CIND = 14.07 years (95%CI = 7.77-20.37)]. However, in those aged 85 and above, CIND had shorter survival than dementia; but the pattern reversed at six years. Cox models revealed shortened survival from memory onset for those with dementia (Risk Ratios (RR) = 1.99; 95%C1 = 1.48-2.68). Older age group (RR = 2.4;95 %CI = 1.56-3.72), coronary heart disease (RR = 1.68; 95%CI = 1.22-2.32) and Parkinson's Disease (RR = 2.37; 95%CI = 1.37-4.11) also shortened survival. Conclusions:These results are consistent with shortened survival for those with dementia, but suggest that in the oldest age group, any cognitive impairment may also reduce survival. These results warrant further study of factors that may influence mortality risk in the oldest-old.
PROSPECTIVE, LONGITUDINAL BELGIAN STUDY ON BEHAVIORAL AND P S Y C H O L O G I C A L SIGNS AND S Y M P T O M S O F D E M E N T I A
Sebastiaan Engelborghs* 1,2, Karen Maertens 1, Ellen Vloeberghs 1, Peter Mari~n 2, Anoek Symons 2, Veerle Ketels 2, Nore Somers 2, Guy Nagels 1, Peter Paul De Deyn 1,2. t University of Antwerp, Antwerp,
Belgium; 2Middelheim General Hospital, Antwerp, Belgium. Contact e-mail: [email protected]
Objective(s): In order to better characterize behavioral and psychological signs and symptoms of dementia (BPSD) in AD and several non-AD dementias, a prospective and longitudinal study was set up. Methods: Patients with probable Alzheimer's disease (AD) (N = 205), mixed dementia (MXD) (N = 39), frontotemporal dementia (FTD) (N = 29), dementia with Lewy bodies (DLB) (N = 14) and Parkinson's disease dementia (PDD) (N = 9) were included. Behavior was assessed at baseline using a battery of behavioral assessment scales: Middelheim Frontality Score (MFS), Behavioral Pathology in Alzbeimer's Disease Rating Scale (Behave-AD), Cohen-Mansfield Agitation Inventory and Cornell Scale for Depression in Dementia. In the follow-up study, all patients underwent semi-annual behavioral and neuropsychological assessments. Results: With disease-long frequencies of more than 80%, activity disturbances and aggression were the most frequently occurring BPSD in AD patients. Aggression was often associated with other forms of BPSD like delusions (R = 0.361; P < 0.0001), hallucinations (R = 0.215; P = 0.002), activity disturbances (R = 0.450; P < 0.0001) and diurnal rhythm disturbances (R = 0.275; P < 0.0001). Psychotic symptoms were present in more than 60% of AD patients. Aggressiveness and frontal lobe symptoms increased as dementia progressed. No major differences in behavioral profiles between AD and MXD patients were observed. Although the Behave-AD systematically underestimated BPSD in FTD patients, frontal lobe symptoms (e.g.: apathy, aspontaneity, loss of insight, disinhibition, emotional bluntness and impaired control of emotions) were very frequent as reflected by significantly higher MFS scores (6.6 41.3) compared to AD patients (3.8 4- 1.7; P < 0.0001). Visual hallucinations discriminated PDD/DLB patients from the other disease groups. Affective and emotional disturbances were found in 65 and 87% of PDD/DLB patients respectively. The high disease-long prevalence of disinhibition (74%) might reflect frontal lobe involvement in PDD/DLB that increased as dementia progressed given the negative correlation between MFS and MMSE scores (R= -0.606; P = 0.002). Several disease-long behavioral scores correlated in AD and FTD, reaching high levels of statistical significance. Conclusions: With exception of MXD subjects who were behaviorally very comparable to AD patients, all disease groups presented with specific behavioral profiles. As several forms of BPSD seemed to cluster in AID and FTD patients, behavioral subgroups with distinct underlying pathophysiological mechanisms are likely to exist.
• P GAIT - ANALYSIS J - IN•D E M E N T I A S John Merory* i, Kate Webster 2, Joanne Wittwer 2 . tAustin Health, Heidelberg Heights, Australia; 2Latrobe University, Bundoora, Australia. Contact e-mail: [email protected]
Background: There is a paucity of information of gait analysis in patients with dementia. We sought to analyze the gait of patients having dementia with Lewy Bodies (DLB) and compare them to patients with Alzheimer's Disease (AD) and age and sex-matched controls. We hypothesized firstly that patients having DLB would have significantly worse gait parameters than the other two groups and secondly that these differences may be a useful addition in the diagnosis of dementia type. Objectives: to quantify gait patterns in patients with DLB and compare these to patients with AD and to control subjects using spatial and temporal measures. Methods: A total of 33 subjects comprising 3 groups (AD, DLB and controls) each with 11 (8 males and 3 females) subjects were recruited. The AD and DLB patients had dementia of mild to moderate severity. DLB was diagnosed according to the McKeith criteria (1996) and AD using the NINCDS-ADRDA classification. Gait was analyzed using an electronic walkway (GAITRite). The following parameters were measured: velocity, cadence, stride length, step width