- Email: [email protected]
REPRODUCTIVE HORMONES AND COGNITIVE IMPAIRMENT AMONG COMMUNITY-DWELLING OLDER MEN: THE CONCORD HEALTH AND AGEING IN MEN PROJECT
Poster Presentations: P2 Background: CAIDE Dementia Risk Score is a tool for estimating dementia risk in the general population. Its associations with Alzheimer or vascular neuropathology are not known.Our objective was to explore the associations between CAIDE Dementia Risk Score at baseline and neuritic plaques, neurofibrillary tangles, micro- and macroinfarcts, and cerebral amyloid angiopathy (CAA) after up to 10 years follow-up in the Vantaa 85+ population. Methods: Study population included 116 participants aged 85 years, without dementia at baseline, and with available clinical and autopsy data. Methenamine silver staining was used for b-amyloid and modified Bielschowsky method for neurofibrillary tangles and neuritic plaques. Macroscopic infarcts were identified from cerebral hemispheres, brainstem, and cerebellum slices. Standardized methods were used to determine microscopic infarcts and CAA. Two versions of the CAIDE risk score were calculated: one based on age, gender, BMI, total cholesterol, Systolic Blood Pressure and physical activity (range 0-15 points); and one additionally including APOE4 carrier status (range 0-18 points). Results: A CAIDE score above 12 points (version including APOE) was associated with b-amyloid accumulation and CERAD score: OR (95%CI) was 2.5 (1.1-5.3) and 2.6 (1.1-5.9), respectively; it also tended to relate to Braak stage: OR (95% CI) 2.0 (0.9-4.6). No associations were found with infarcts or CAA. Conclusions: In a population of elderly aged 85 years, CAIDE Dementia Risk Score was associated with increased Alzheimer-type pathology, particularly amyloid-b accumulation
P2-299
REPRODUCTIVE HORMONES AND COGNITIVE IMPAIRMENT AMONG COMMUNITY-DWELLING OLDER MEN: THE CONCORD HEALTH AND AGEING IN MEN PROJECT
Benjumin Hsu1, Robert Cumming1, Vasi Naganathan1, Fiona Blyth1, David Handelsman2, 1University of Sydney, Sydney, Australia; 2ANZAC Research Institute, Sydney, Australia. Contact e-mail: bhsu3843@uni. sydney.edu.au Background: The main objective of this study was to examine associations in older men between serum reproductive hormones and changes in cognitive function over 5-years. Methods: 955 men aged 70 years and older from the Concord Health and Ageing in Men Project (CHAMP) were assessed at baseline (2005-2007) and at 5-years follow-up (20112013). At baseline, total testosterone (TT), dihydrotestosterone (DHT), estradiol (E2), and estrone (E1) were measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS), and sex hormone-binding globulin (SHBG), luteinizing hormone (LH), and follicle-stimulating hormone (FSH) were measured by immunoassay. The calculated free testosterone (cFT) was computed using an empirical formula derived from the measured TT and SHBG. Dementia diagnoses were obtained at baseline by clinical assessment and review by a specialist panel. Cognitive function was measured by the Mini Mental State Examinations (MMSE) and Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE), which were conducted at both baseline and follow-up. Cognitive impairment was defined using a cutoff of MMSE<24 and/or IQCODE>3.31. Results: Low levels of serum TT were significantly associated with cognitive decline over time. Men in the lowest TT quartile were 2.52-fold (95%CI: 1.15-5.52) more likely to exhibit cognitive decline compared to the highest TT quartile. Similarly, men in the lowest cFT were 2.71-fold (95%CI: 1.26-5.84) more likely to exhibit cognitive decline when compared to the men in the highest cFT quartile. There were significant linear trend (p<0.05) across both TT and cFT quartiles in association with decline in cognitive function. Adjustment for age, ed-
P587
ucation, body mass index, smoking, and depression in the multivariable model did not affect the observed associations. There were no consistent associations observed for the other studied reproductive hormones and glycoproteins. Conclusions: Previous longitudinal studies have reported conflicting findings for the relationship between testosterone and cognitive function. Our study is the first to measure serum TT with the current gold standard measurement, the LC-MS/MS. Low levels of serum TT and cFT were strong predictors of cognitive decline over time in older men. Large randomized placebo-controlled trials would be needed to determine whether testosterone treatment protects against cognitive decline in older men or is a biomarker of cognitive decline.
P2-300
RISK PREDICTION MODELS FOR DEMENTIA: A SYSTEMATIC REVIEW
Blossom Christa Maree Stephan1, Eugene Tang1, Stephanie L. Harrison1, Mark Gordon2, Pieter Jelle Visser3, Gerald Novak4, Louise A. Robinson1, Carole Dufouil5, Fiona E. Matthews6, Carol Brayne7, 1Newcastle University, Newcastle upon Tyne, United Kingdom; 2Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut, United States; 3Maastricht University, Maastricht, Netherlands; 4Janssen R&D, Titusville, New Jersey, United States; 5INSERM U708 / CIC-EC7, Bordeaux, France; 6MRC Biostatistics Unit, Cambridge, United Kingdom; 7Cambridge University, Cambridge, United Kingdom. Contact e-mail: [email protected] Background: Accurate identification of individuals at high risk of dementia is important to guide clinical care (e.g., individually tailor risk factor management and intervention), and has research implications regarding clinical trial recruitment and development of preventative strategies. Numerous models for predicting dementia, and more specifically Alzheimer’s Disease (AD) in older adults, have been developed. To evaluate these models we undertook a systematic review in 2009 (and updated this in 2013), including a critique of the variables selected for inclusion and an assessment of model prognostic performance. Methods: A systematic literature search was conducted in MEDLINE, Embase, Scopus, and ISI Web of Science of all English-language articles published from database inception till 9 September 2013. Longitudinal, population-based studies examining dementia risk prediction models, with measures of sensitivity/specificity, area under the receiver operating characteristic curve (AUC), or the c-statistic were included. Results: In total, 35 articles were eligible. Dementia risk models can be broadly divided into three categories: (1) neuropsychological based models (incorporating cognitive test scores, with or without subjective memory/cognitive complaints and demographic information such as age and educational attainment); (2) health based models (incorporating self-reported or objectively measured health status such as cardio-metabolic and neurological features); and, (3) multifactorial models (typically combining demographic and cognitive measures with health status or genetic variables). The AUC estimates ranged from poor (AUC¼0.50) to high (AUC¼0.87). Model development has been primarily undertaken in Caucasians, over variable follow-up times (1 year to >20 years), and with various outcomes (e.g., all cause dementia, AD). Only one model has been externally validated, raising questions of model transportability outside the populations from which the particular model was developed. Conclusions: No dementia risk prediction model can be currently recommended for use within a screening program. Within the field of dementia, a key research priority is the development of an externally validated, cost-effective, consensus tool for early identification of individuals at high risk of dementia. Such a tool will improve dementia risk prediction and more targeted strategies for risk factor reduction in older populations.
P2-299
REPRODUCTIVE HORMONES AND COGNITIVE IMPAIRMENT AMONG COMMUNITY-DWELLING OLDER MEN: THE CONCORD HEALTH AND AGEING IN MEN PROJECT
Benjumin Hsu1, Robert Cumming1, Vasi Naganathan1, Fiona Blyth1, David Handelsman2, 1University of Sydney, Sydney, Australia; 2ANZAC Research Institute, Sydney, Australia. Contact e-mail: bhsu3843@uni. sydney.edu.au Background: The main objective of this study was to examine associations in older men between serum reproductive hormones and changes in cognitive function over 5-years. Methods: 955 men aged 70 years and older from the Concord Health and Ageing in Men Project (CHAMP) were assessed at baseline (2005-2007) and at 5-years follow-up (20112013). At baseline, total testosterone (TT), dihydrotestosterone (DHT), estradiol (E2), and estrone (E1) were measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS), and sex hormone-binding globulin (SHBG), luteinizing hormone (LH), and follicle-stimulating hormone (FSH) were measured by immunoassay. The calculated free testosterone (cFT) was computed using an empirical formula derived from the measured TT and SHBG. Dementia diagnoses were obtained at baseline by clinical assessment and review by a specialist panel. Cognitive function was measured by the Mini Mental State Examinations (MMSE) and Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE), which were conducted at both baseline and follow-up. Cognitive impairment was defined using a cutoff of MMSE<24 and/or IQCODE>3.31. Results: Low levels of serum TT were significantly associated with cognitive decline over time. Men in the lowest TT quartile were 2.52-fold (95%CI: 1.15-5.52) more likely to exhibit cognitive decline compared to the highest TT quartile. Similarly, men in the lowest cFT were 2.71-fold (95%CI: 1.26-5.84) more likely to exhibit cognitive decline when compared to the men in the highest cFT quartile. There were significant linear trend (p<0.05) across both TT and cFT quartiles in association with decline in cognitive function. Adjustment for age, ed-
P587
ucation, body mass index, smoking, and depression in the multivariable model did not affect the observed associations. There were no consistent associations observed for the other studied reproductive hormones and glycoproteins. Conclusions: Previous longitudinal studies have reported conflicting findings for the relationship between testosterone and cognitive function. Our study is the first to measure serum TT with the current gold standard measurement, the LC-MS/MS. Low levels of serum TT and cFT were strong predictors of cognitive decline over time in older men. Large randomized placebo-controlled trials would be needed to determine whether testosterone treatment protects against cognitive decline in older men or is a biomarker of cognitive decline.
P2-300
RISK PREDICTION MODELS FOR DEMENTIA: A SYSTEMATIC REVIEW
Blossom Christa Maree Stephan1, Eugene Tang1, Stephanie L. Harrison1, Mark Gordon2, Pieter Jelle Visser3, Gerald Novak4, Louise A. Robinson1, Carole Dufouil5, Fiona E. Matthews6, Carol Brayne7, 1Newcastle University, Newcastle upon Tyne, United Kingdom; 2Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut, United States; 3Maastricht University, Maastricht, Netherlands; 4Janssen R&D, Titusville, New Jersey, United States; 5INSERM U708 / CIC-EC7, Bordeaux, France; 6MRC Biostatistics Unit, Cambridge, United Kingdom; 7Cambridge University, Cambridge, United Kingdom. Contact e-mail: [email protected] Background: Accurate identification of individuals at high risk of dementia is important to guide clinical care (e.g., individually tailor risk factor management and intervention), and has research implications regarding clinical trial recruitment and development of preventative strategies. Numerous models for predicting dementia, and more specifically Alzheimer’s Disease (AD) in older adults, have been developed. To evaluate these models we undertook a systematic review in 2009 (and updated this in 2013), including a critique of the variables selected for inclusion and an assessment of model prognostic performance. Methods: A systematic literature search was conducted in MEDLINE, Embase, Scopus, and ISI Web of Science of all English-language articles published from database inception till 9 September 2013. Longitudinal, population-based studies examining dementia risk prediction models, with measures of sensitivity/specificity, area under the receiver operating characteristic curve (AUC), or the c-statistic were included. Results: In total, 35 articles were eligible. Dementia risk models can be broadly divided into three categories: (1) neuropsychological based models (incorporating cognitive test scores, with or without subjective memory/cognitive complaints and demographic information such as age and educational attainment); (2) health based models (incorporating self-reported or objectively measured health status such as cardio-metabolic and neurological features); and, (3) multifactorial models (typically combining demographic and cognitive measures with health status or genetic variables). The AUC estimates ranged from poor (AUC¼0.50) to high (AUC¼0.87). Model development has been primarily undertaken in Caucasians, over variable follow-up times (1 year to >20 years), and with various outcomes (e.g., all cause dementia, AD). Only one model has been externally validated, raising questions of model transportability outside the populations from which the particular model was developed. Conclusions: No dementia risk prediction model can be currently recommended for use within a screening program. Within the field of dementia, a key research priority is the development of an externally validated, cost-effective, consensus tool for early identification of individuals at high risk of dementia. Such a tool will improve dementia risk prediction and more targeted strategies for risk factor reduction in older populations.