BACKGROUND: Mild cognitive impairment (MCI) is a clinical condition that is characterized by mild changes in cognition. The amnestic form of MCI (aMCI) primarily affects memory and is thought to represent a stage between healthy aging and Alzheimer’s disease (AD). The medial temporal lobe (MTL) and the limbic system are two areas of the brain that have been implicated in the amnestic form of MCI. While MCI represents a risk factor for AD, it does not always lead to dementias. Being a carrier of the APOE Ɛ4 allele has also shown to increase chances of progression from MCI to AD. OBJECTIVE: To determine whether the subnuclei of the amygdala, along with other specific regions within the MTL, can differentiate between cognitively normal individuals and age-matched subjects with aMCI. METHODS: T1-weighted magnetic resonance imaging (MRI) data from two sources, the Boston University Alzheimer’s Disease Center (BU-ADC) and the Alzheimer’s Disease Neuroimaging Initiative (ADNI), was compiled for cross-sectional analysis. 95 scans in total from 45 cognitively normal participants and 50 diagnosed with aMCI were analyzed and the volumes of interest were automatically generated by the developmental version of FreeSurfer v6.0. To evaluate how well the volumes could predict either group membership (i.e. control group or MCI group) or APOE Ɛ4 status (i.e. carrier or noncarrier), the variables were assessed by nominal logistic regression models. RESULTS: Six of the nine nuclei of the amygdala had significantly reduced volumes in the aMCI group compared to controls. The whole amygdala and the perirhinal cortex also demonstrated reduced volumes in the aMCI group compared to the control group. The whole amygdala was a good predictor of group membership (R2 = 0.1386, whole model test chi square = 18.21558, p = 0.0004), but none of the subnuclei were good predictors individually. A model containing the 9 nuclei, the entorhinal cortex, and the perirhinal cortex provided a good fit for predicting APOE Ɛ4 status fit (R2 = 0.3000, whole model test chi square = 36.29563, p = 0.0002) and the best predictor was the corticoamygdaloid transition area of the amygdala. CONCLUSIONS: The results of our study confirm previous findings of reduced whole amygdala volume and add to the limited literature of reduced perirhinal cortex and amygdaloid nuclei volumes in MCI compared to healthy controls. To the best of our knowledge, this was the first time the automatic segmentation atlas was used to analyze the volumes of nine subnuclei of the amygdala in a population of aMCI. Our model testing the volume of the whole amygdala accurately predicted aMCI subjects with 58% accuracy and controls with 70% accuracy; the accuracy rose to 69% when the entorhinal cortex and the perirhinal cortex were added to the model to predict aMCI subjects from controls. Additionally, the model for predicting APOE Ɛ4 status identified noncarriers of the allele at 85% accuracy. Future studies should consider increasing the sample size to better assess small ROIs and assess for differences in the separate hemispheres.

The search for reliable early indicators of age-related cognitive decline represents an important avenue in aging research. Most research on late-life development charts cognitive change as a function of chronological age (CA), however, although CA is a commonly used developmental index, it offers little insight into the mechanisms underlying cognitive decline. In contrast, biological age (BioAge), reflecting the vitality of essential biological processes, represents a promising operationalization of developmental time. My overall programmatic doctoral research interests involve the identification of biological risk factors that predict age-related cognitive decline, impairment and dementia. In this dissertation document, I present: an overview of my empirical contributions to the BioAge and cognitive aging literature throughout my doctoral training; the dissertation project which uses preliminary data from the PREVENT study and provides evidence that elevated plasma pro-inflammatory proteins are associated with cognitive status (healthy controls (HC) vs Alzheimer’s disease dementia (AD)), cognitive performance and are related to poorer cognitive performance in amnestic mild cognitive impairment (a-MCI); and a discussion on the broad implications of the project results and future directions in BioAge research. / Graduate

The relationship between cerebrovascular disease (CVD) risk factors and cognitive impairment or dementia has been widely studied with significant variability in findings between groups. We hypothesized that chronic small vessel injury in the form of arteriolar sclerosis, measured quantitatively using MRI to measure total white matter hyperintensity (WMH) volumes, would identify specific association of CVD risk factors and patterns of cognitive decline, associated with mild cognitive impairment of the cerebrovascular type, that represent the core features of vascular cognitive impairment in our cohort. A Cross-sectional analysis of clinical and quantitative MRI data on 114 subjects with normal cognitive function (n=52) and mild cognitive impairment (MCI; n=62) was performed. Quantitative total WMH volumes were examined in relation to potentially causative CVD risk factors and resultant test scores across cognitive domains using linear regression models adjusted for age, gender, and education. Among CVD risk factors analyzed, age (p< 0.001), education (p= 0.003), hypertension (p= 0.012), and hyperlipidemia (p= 0.008) demonstrated the strongest associations with WMH volumes. Conversely, diabetes, smoking, history of heart attacks, atrial fibrillation, and history of stroke that have shown associations with CVD pathology on imaging in other studies were not statistically associated with increased WMH in this cohort. WMH volumes were associated with decrease performance on the Trial Making Test type A & B and long delayed free recall on the California Verbal Learning Test. Our findings suggest similarities and yet differences in comparison to other studies. Hypertension and hyperlipidemia appear to represent common shared risks across geographically disparate groups. Our findings, like others, suggest CVD pathology impact processing speed and executive function and provide further evidence for CVD effects on short-term memory in those at risk for cognitive decline and the future development of dementia in our cohort.

Dementia is a common cause of disability in the elderly and, in the absence of a successful long-term treatment, it is important to investigate possible lifestyle interventions to help reduce an individual s risk of developing the condition. This thesis investigated the relationship between physical activity and dementia risk, finding that not all research supports the link. The literature review presented in this thesis (Chapter 2) highlighted several possible mediating factors, specifically the type of physical activity performed, the cognitive domains being studied and participant characteristics. Women seemed most susceptible to the effect of physical activity and some other forms of midlife interventions, possible mechanisms for which were discussed in another review (Appendix A). The cognitive test battery to be used in later studies was evaluated for its relevance to dementia and treatment during a 6-month study of Alzheimer's disease patients and their carers (Chapter 3). Memory tasks were found to be especially sensitive to clinical outcomes of dementia treatment (Chapter 4). An observational study of Indonesian elderly found a positive relationship between physical activity and memory performance on the same tests. This effect was strongest in women and in those with no pre-existing cognitive impairment (Chapter 5). However, the relationship could be further modified by other demographic factors, such as education. Health was independently affected in this model by exercise and its association with engaging in physical activity in this cohort was further investigated in Chapter 6. A randomised controlled trial (Chapter 7) was conducted to assess the effect of a 12-week programme of non-aerobic physical activity in sedentary middle-aged adults. Results indicated that resistance training, but not flexibility exercises, influenced memory but not executive function. Overall, this thesis suggests that several types of physical activity may be effective at slowing cognitive decline in elderly groups who are at increased risk of dementia, such as those in middle age and elderly women (Chapter 8). These findings should be expanded with the aim to improve healthcare advice and influence policy-making.

Magister Curationis - MCur / Worldwide concerns have been raised about the presence and association of depressive symptoms, cognitive impairment, and dementia in older adults (60 years and older), which are often unrecognised and untreated in long-term care facilities (LTCF’s). The progression of cognitive impairment to dementia reduces quality of life with negative consequences of physical, mental, and psychosocial health. In many LTCF’s internationally, the standardised interRAI system is used to capture depressive symptoms and cognitive impairment. However, there is a fragmentation of systems for making evidence-based decisions to plan and manage care for residents with depressive symptoms, cognitive impairment, and dementia. This study, being the first of its kind in South Africa, addressed this gap, by describing a profile of depressive symptoms and cognitive impairment in residents, and analysing their coexistence, using the interRAI-LTCF in a LTCF in the Cape Metropole in South Africa. A quantitative, descriptive, and analytical cross-sectional secondary data analysis was conducted using the records of all 173 resident’s medical records of residents with a last interRAI-LTCF assessment from 2014 and 2016. The objectives were to determine the levels of depressive symptoms and cognitive impairment, and to assess variously associated demographics and clinical variables between depressive symptoms and cognitive impairment of the interRAI-LTCF in residents in a LTCF. Secondary data were analysed, using the IBM Statistical Package for Social Sciences (SPSS) software, version 25, to test any statistically significant relationship between the extracted variables (Significance was set as p˂0.05).

Introdução: O Questionário do Informante sobre o Declínio Cognitivo do Idoso (IQCODE) é um instrumento de rastreio que se baseia nas informações fornecidas por familiares ou cuidadores acerca de um possível declínio cognitivo do paciente. Embora tenha boa sensibilidade para a identificação de casos suspeitos de demência, poucos estudos avaliaram as propriedades diagnósticas do IQCODE no rastreio do comprometimento cognitivo leve (CCL). O CCL corresponde a uma condição de risco para o desenvolvimento de demência, sendo caracterizado pela presença de alterações cognitivas que podem ser mensuradas objetivamente, indicando um declínio em relação ao desempenho esperado para indivíduos da mesma faixa etária e nível de instrução. Tais alterações cognitivas (ou déficits) são insuficientes para o diagnóstico de demência, no caso de um funcionamento cognitivo global preservado e da capacidade de desempenhar as atividades da vida diária (Winblad, 2004). Objetivos: Examinar as propriedades diagnósticas do IQCODE no rastreio do CCL, identificando os pontos de corte do teste IQCODE que melhor separam indivíduos idosos cognitivamente normais dos indivíduos com CCL; correlacionar os resultados obtidos com outros testes de rastreio cognitivo amplamente utilizados em nosso meio, como o Mini-Exame do Estado Mental (MEEM), o Teste do Desenho do Relógio (TDR) e o Teste Cognitivo de Cambridge (CAMCOG); identificar entre os 26 itens do IQCODE os agrupamentos (clusters) que contribuem para a identificação dos casos de CCL. Métodos: Estudo de corte transversal em amostra de 167 indivíduos idosos (Controles n=51, CCL n=58 e Demência de Alzheimer (DA) n=58) acompanhados no Ambulatório de Psicogeriatria do LIM-27, Instituto de Psiquiatria do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. O diagnóstico do estado cognitivo (padrão-ouro) estabelecido por meio de consenso multidisciplinar, levando-se em consideração as informações clínicas e o desempenho em testes neuropsicológicos: A idade média dos indivíduos de cada grupo foi, respectivamente, de 67,5 (±5,6), 70,2 (±6,1) e 75,5 (±8,4) anos, e a escolaridade média foi de 12,6 (±5,4), 9,6 (±5,7) e 8,5 (±5,5) anos. Análises de curvas ROC (Receiver Operating Characteristics) foram realizadas para avaliar a acurácia diagnóstica do IQCODE e demais testes comparativos na separação dos pacientes de cada grupo diagnóstico, comparados dois a dois (CCL versus Controles, CCL versus DA, DA versus Controles); por meio de regressão logística, avaliou-se o potencial do uso combinado do IQCODE em conjunto com os demais instrumentos de rastreio para predizer a ocorrência de CCL e DA; finalmente, por meio de análise de clusters, avaliou-se a distribuição dos diferentes itens do IQCODE nos pacientes com CCL e seus subtipos. Resultados: Os pontos de corte do IQCODE para a separação dos grupos diagnósticos foram: (a) DA versus Controles: 3,3 (AUC=0,90; sensibilidade: 84,5%; especificidade: 82,4%); (b) CCL versus Controles: 3,1 (AUC=0,73; sensibilidade: 77,6%; especificidade: 60,8%); (c) CCL versus DA: 3,4 (AUC=0,81; sensibilidade: 79,3%; especificidade: 70,7%). O IQCODE apresentou melhor correlação com o CAMCOG (=0,542; p<0,001). Com base na análise de cluster, estimou-se que o agrupamento que contém itens relacionados à memória episódica foi o mais relevante para identificar os pacientes portadores de CCL amnéstico. Conclusões: O uso do IQCODE obteve melhores resultados para diferenciar idosos cognitivamente normais de CCL quando utilizado em conjunto com o CAMCOG. A análise de cluster do IQCODE melhor prediz CCL e seus subtipos / Introduction: The Informant Questionnaire of Cognitive Decline in the Elderly is a screening diagnostic instrument which is based on given information from family members and caregivers regarding a possible patients cognitive impairment. Despite its good sensitivity for suspected dementia caseness, few studies have been carried out using the diagnosis properties of the IQCODE to screen for Mild Cognitive Impairment (MCI). MCI corresponds to a condition of a risk factor for dementia outcome and is characterized by the presence of cognitive changes measured objectively, indicating an impairment in comparison with the expected performance for individuals at the same age and years of schooling. These deficits are insufficient for dementia diagnosis in case of preserved global cognitive functioning as well as in the capacity to perform daily activities (Winblad, 2004). Objectives: Examine diagnostic properties of the IQCODE in identifying cut-off scores which best distinguish the cognitively normal elderly from those with MCI; to correlate these results with other widely used cognitive tests, such as the Mini Mental State Examination (MMSE), the Clock Drawing Test (CDT) and the Cambridge Cognitive Test (CAMCOG); to identify among the 26 items in the IQCODE those clusters which best contribute to the identification of the cases. Methods: Cross-sectional study in a sample of 167 elderly subjects (Controls: n=51, MCI: n=58 and Alzheimer Disease (AD): n=58) followed at the Psychogeriatric Clinic of the Laboratory of Neuroscience (LIM-27), Department and Institute of Psychiatry, Faculty of Medicine, University of São Paulo. The cognitive diagnosis was reached by consensus at expert multi-disciplinary meetings (gold standard), taking into account clinical and neuropsychological evaluation. The mean age in each group was respectively: 67.5(±5.6), 70.2 (±6.1) and 75.5 (±8.4) years, and mean of years of schooling were 12.6(±5.4), 9.6(±5.7) and 8.5(±5.5) years. ROC (Receiver Operating Characteristics) Curve analysis were carried out to determine diagnostic accuracy of the IQCODE and the comparative tests in paired sets (MCI versus Controls, MCI versus AD, AD versus Controls); by logistic regression analysis it was evaluated the prediction of MCI and AD with the IQCODE and its combined usage with the comparative tests; finally by cluster analysis it was evaluated the different distribution of the IQCODE items in MCI patients and its subtypes. Results: The IQCODE cut-off scores for diagnostic groups separation were: (a) AD versus Controls: 3.3 (AUC=0.90; sensitivity: 84.5%; specificity: 82.4 %( b) MCI versus Controls: 3.1 (AUC=0.73; sensitivity: 77.6%; specificity%: 60.8); (c) MCI versus AD: 3.4 (AUC=0.81; sensitivity: 79.3%; specificity: 70.7%). The IQCODE had the best correlation with the CAMCOG (=0.542; p<0.001). According to cluster analysis, the episodic memory grouping was the most relevant in identifying amnestic MCI. Conclusions: The IQCODE achieved best results to discriminate cognitively unimpaired elderly from MCI when combined with the CAMCOG. Cluster analysis of the IQCODE better predicts MCI and its subtypes.

The primary objective of this study was to assess cognitive functioning in participants with relapsing remitting multiple sclerosis (RRMS) using the MicroCog and to compare their performance to that of a demographically matched, healthy control group. It was hypothesized that as a group, participants with RRMS would have worse cognitive function than healthy controls on all Level 1, 2, and 3 Index scores of the MicroCog. Twenty-six participants with RRMS and twenty-nine sex and education matched healthy controls were administered the MicroCog (Standard Form) along with measures of depression and clinical status, and paper-pencil tests of processing speed (Symbol Digit Modalities Test; SDMT and Paced Auditory Serial Addition Test; PASAT). A series of ANCOVAs with depression as a covariate was performed to determine between group differences for each MicroCog Level 3 Index score (General Cognitive Proficiency (GCP) and General Cognitive Functioning (GCF)), Level 2 Index score (Information Processing Accuracy (IPA) and Information Processing Speed (IPS)), and Level 1 Index score (Attention/Mental Control, Memory, Reasoning/Calculation, Spatial Processing, and Reaction Time). Pearson's and point biserial r correlations were calculated in order to assess the degree to which Level 2 and 3 Index scores correlated with clinical and demographic factors (sex, disease duration, depression, and clinical status) and to correlate the MicroCog IPS index score with traditional measures of processing speed. Eight RRMS and two control participants met criteria for cognitive impairment on the MicroCog. ANCOVA results indicated there were significant differences between RRMS and control performance for two MicroCog scores (GCF and IPS). There were not significant differences for GCP, IPS, and all Level 1 scores. A post-hoc analysis performed for the same hypothesis with a group of age equivalent participants suggested a significant RRMS by depression interaction for Level 3 scores. RRMS was not predictive of Level 2 scores after controlling for depression in the age equivalent sample. Correlations for clinical and demographic factors with cognitive outcomes indicated significant relationships for clinical status and depression. There was not a significant relationship detected for disease duration or sex. MicroCog and processing speed measures were significantly related. Post-hoc analyses supported that the criterion validity of the MicroCog is comparable to other cognitive screening tools in RRMS. The results and limitations of our study are discussed, in addition to recommendations for future research.

Alzheimer's disease (AD) is the most common form of dementia and results in progressive cognitive decline, particularly in regards to memory (National Institute on Aging, 2012). Prior research has shown sex differences in brain-atrophy rates of AD patients, with women experiencing a higher rate of progression in volume reduction (Skup et al., 2011). This suggests that there may also be differences in cognitive functioning between sexes, particularly in the rate of cognitive decline with a more rapid disease progression for dementing females compared to dementing males. The current study monitored memory function longitudinally in approximately 200 total participants, 100 with Mild Cognitive Impairment (MCI) or probable AD and 100 healthy controls enrolled in an aging study through the Arizona Alzheimer's Disease Research Consortium. Memory performance was evaluated with two memory tests, the Rey Auditory Verbal Learning Test (RAVLT; Rey, 1941) and the Brief Visuospatial Memory Test-Revised (BVMT-R; Benedict, 1997). Memory function was evaluated in participants with at least three data points over a five-year span. A multivariate regression model was used that includes controls for disease severity, age, age at disease onset, education, ethnicity, and medical comorbidities. Results indicated that females in the MCI and AD groups initially performed better than the males, but that over time, female scores had dropped significantly lower than male scores, suggesting a more rapid decline in females. Significant sex differences in cognitive decline may yield a deeper understanding of the development and progression of AD and aid in more effective and sex-specific treatment.

Although there are ample studies confirming that memory self-efficacy (MSE) declines with age, less is known about what factors account for the variation in MSE among older adults. The purpose of this study was to examine the relationship between MSE, diagnostic and clinical characteristics, and subsequent episodic memory performance in older adults. A nonprobability sample of 200 cognitively normal and older adults with mild cognitive impairment (MCI) participating in a longitudinal population-based study investigating the incidence, prevalence and risk factors for MCI completed a questionnaire about self-referent beliefs of MSE. Bandura's (1989) selfefficacy theory and the Integration Model (Whittemore, 2005) informed the descriptive study. Pearson product-moment correlations, a general linear model and a multiple linear regression analysis were conducted. The difference in MSE ratings between the cognitively normal group and the MCI group tested as a whole was significant when adjusting for age, gender and educational attainment (p < .001; ES= 0.585). The overall regression model explained 17 % of the variance of MSE (p < .001) and included age, gender, educational attainment, APOE 4 genotype, family history of dementia, cognitive diagnosis and depressive symptoms. After controlling for age and the other variables of interest, cognitive classification and depression were significant predictors of MSE. Higher MSE ratings were correlated with better episodic memory performance for both groups (r = .273, p < .001). Memory training that capitalizes on the benefits accruing from higher MSE is needed for cognitively normal older adults and older adults with MCI.

This dissertation is a critical examination of the relationship between sleep problems and/or disorders, particularly Obstructive Sleep Apnea (OSA) and Alzheimer Disease (AD). First, I conducted an exhaustive systematic review of existing literature, and identified gaps in research that led to specific research aims. For the first aim, I conducted the first ever-published meta-analysis examining sleep, cognitive decline and AD, providing an aggregate effect of sleep on AD. Second, focusing on OSA, I conducted a study examining OSA’s effect on longitudinal changes on AD biomarkers in cognitive normal, MCI and AD subjects, using data from the Alzheimer Disease Neuroimaging Initiative (ADNI). Lastly, I conducted a review, integrating over 3 decades of research examining OSA and cognition; OSA and subsequent cognitive decline; and OSA and AD; with particular focus in appreciating the heterogeneity of OSA and its outcomes in distinct age groups. Results and implications from my research indicate that ample evidence exists linking sleep impairments and circadian regulating mechanisms directly to clinical symptoms in AD. Sleep problems and/or disorders increases your risk of cognitive decline and AD. OSA is associated with increased AD biomarker burden over time, and effects longitudinal changes in these biomarkers, such that OSA subjects progress faster than non-OSA subjects do. OSA may be age-dependent in older adults (60 – 70 years old) and the elderly (70 years and above) and is associated with neurodegenerative diseases particularly, cognitive decline and AD. Intermittent hypoxia and sleep fragmentation are two main processes by which OSA induces neurodegenerative changes. Therefore, clinical interventions aimed at OSA, such as treatment with CPAP or dental appliances, in cognitive normal and MCI patients, could possibly slow the progression of cognitive impairment to AD.