Search

Introduction: Sphingolipids are fundamental components of brain structure and function. They are located in the myelin sheath insulating neuronal axons and in the cell membranes of neurons, astrocytes, and microglia involved in signal transduction. Although circulating sphingolipid profiles reflect clinical neurodegeneration, few prospective studies have investigated their utility as pre-clinical risk markers in community-based cohorts. For heart disease and diabetes, themselves risk factors for dementia, associations with sphingolipids differed across species and the length of the saturated fatty acid carried. Hypothesis: We hypothesized that plasma levels of sphingolipid species containing palmitic acid (16:0) would be positively associated with incident dementia risk and cognitive decline, while those with arachidic acid (20:0), behenic acid (22:0), and lignoceric acid (24:0) would be associated with lower risk. Methods: We measured 18 sphingolipid species in plasma samples from 4612 participants of the Cardiovascular Health Study (CHS) (mean age 76.2 years; 40% male; and 25% APOE ε4 allele carriers). Cognitive decline was assessed across 6 years using annual 100-point modified Mini-Mental State Examination (3MSE) and Digital Symbol Substitution Test (DSST) scores. A total of 3088 non-demented participants were enrolled in the CHS Cognition Study and followed with adjudication for dementia. To correct the analyses for multiple comparisons, we applied a false discovery rate (FDR) correction and used a significance threshold of 0.05. Results: Higher baseline levels of lactosylceramide-24:0 (LacCer-24) were associated with a faster 6-year cognitive decline (time x sphingolipid interaction p=0.04 for 3MSE and 0.03 for DSST). Conversely, higher baseline plasma levels of hexosylceramide -16:0 (HexCer-16) and sphingomyelin-22:0 (SM-22) were associated with slower cognitive decline measured with DSST (time x sphingolipid interaction p=0.02 for HexCer and p=0.05 for SM-22). During a median follow-up of 4.5 years, we identified 366 incident dementia cases. In proportional Cox regression analyses, higher plasma levels of ceramide-16:0 (Cer-16) were positively associated with incident dementia upon adjustment for demographics, lifestyle factors, medical history, C-reactive protein, and lipids [hazard ratio per SD: 1.18 (95% CI, 1.06; 1.31)]. Conclusions: In older adults without dementia, higher plasma levels of Cer-16, a species previously associated with cardiovascular disease and mortality, were associated with higher risk of clinical dementia as we hypothesized. LacCer-24 was associated with faster, and HexCer-16 and SM-22 with slower, cognitive decline, but these species were not statistically associated with dementia risk. These novel and inconsistent findings will need to be replicated in other longitudinal studies.

Background: Higher visit-to-visit variability in fasting glucose (GV) and HbA1c (A1cV), indicating worse glycemic control, is associated with incident dementia in adults with type 2 diabetes (T2D). Few studies have investigated the relationship of GV and A1cV with cognitive decline in the general population. Hypothesis: Higher visit-to-visit GV and A1cV are associated with decline in global cognitive performance. Methods: Fasting glucose (FG) was measured in MESA participants at Exam 1 (2000-02) and each of 5 follow-up exams, and HbA1c was measured at Exam 2 (2002-04), Exam 5 (2010-12), and Exam 6 (2016-18). Global cognitive performance was assessed by the Cognitive Abilities Screening Instrument (CASI; scored 0-100) at Exam 5 and re-administered to a subset at Exam 6. We calculated GV and A1cV as the SD (SD GV ; SD A1c ), coefficient of variation (CV GV ; CV A1c ), and average real variability (ARV GV ; ARV A1c ), and defined global cognitive decline as a ≥1 race/ethnicity-specific SD decrease in CASI score from Exam 5 to Exam 6 (-5.9 to -8.2 points across 4 races/ethnicities). We report the CASI point-change and relative odds of decline (OR) from Exam 5 to Exam 6 (95% CI) associated with continuous log 2 -normalized and highest quartile (Q4 vs. quartiles 1-3 for potential floor effects) of GV and A1cV from inverse probability of censor-weighted linear and logistic regression adjusted for Exam 5 CASI score, age, sex, race/ethnicity, education, vascular risk factors, and APOE ε4 status. Results: A total of 1834 Exam 6 participants (aged 73±8 years; 47% men; 43% White; 7% with T2D at Exam 1; 13% with incident T2D over follow-up) had ≥3 FG measurements and repeat CASI data. The overall 6-year mean (SD) change in CASI was +0.1 (6.6) points. Q4 of SD GV , CV GV , and ARV GV was associated with 0.8 (0.1, 1.5), 0.9 (0.2, 1.6), and 0.7 (0.03, 1.4) point decreases in CASI, respectively, and Q4 of CV GV was associated with higher odds of global cognitive decline (OR = 1.5 [1.1, 2.1]); continuous GV indices were not associated with CASI change. Associations with SD GV and CV GV were driven by participants with incident T2D (interaction P A1c was associated with a 0.5 (0.1, 0.9) point decrease in CASI. However, in participants with incident T2D, 2-fold increments in SD A1c , CV A1c , and ARV A1c were associated with 1.3 (0.2, 2.4), 1.4 (0.2, 2.6), and 1.8 (0.6, 3.0) point decreases in CASI, respectively, and a 2-fold increment in ARV A1c was associated with greater odds of decline (OR = 2.1 [1.1, 3.8]) (interaction P all Conclusions: Poor glycemic control over 14-16 years is associated with modest cognitive decline over the subsequent 6-8 years, particularly among adults at high risk for T2D during mid- to late-life.

Background: Diabetes is a well-known risk factor for dementia. Despite emerging evidence that higher glucose variability is associated with cognitive decline and hypoglycemia is associated with increased risk for dementia, few studies have investigated the relationship of lower fasting glucose and glucose variability with cognitive performance in the general population. Hypothesis: Diabetic and lower levels of fasting glucose and higher visit-to-visit glucose variability over 10 years are associated with worse cognitive performance. Methods: Participants in the Multi-Ethnic Study of Atherosclerosis (N = 4,591; mean age 69.7 ± 9.4 years) completed a cognitive assessment at Exam 5 including the Cognitive Abilities Screening Instrument (CASI), Digit Symbol Coding (DSC), and Digit Span (DS, Forward and Backward combined) tests measuring global cognitive performance, executive function/processing speed, and working memory, respectively. We defined fasting glucose categories as low (β [95% confidence limits]) from regression models adjusted for age, sex, race/ethnicity, education, income, cardiovascular risk factors, and APOE genotype. Results: Relative to normal glucose, participants with diabetic fasting glucose performed significantly worse on the CASI (Exam 1 β = -0.75 [-1.36, -0.15]; Exam 5 β = -0.30 [-0.83, 0.23]), DSC (Exam 1 β = -2.97 [-4.20, -1.74]; Exam 5 β = -1.26 [-2.34, -0.18]), and DS (Exam 1 β = -0.39 [-0.72, -0.07]; Exam 5 β = -0.29 [-0.58, -0.003]). In contrast, participants with low Exam 1 fasting glucose performed significantly better on the DSC (Exam 1 β = 1.55 [0.45, 2.65]) and similarly on the CASI and DS compared to those with normal fasting glucose. Adjusting for glucose lowering medication attenuated associations with the DSC and DS but not the CASI. Accounting for heterogeneity in the low glucose category by hypoglycemia (β = -0.36 [-0.60, -0.12]) and DSC ( β = -1.14 [-1.63, -0.65]). Associations were consistent across race and sex. Conclusions: Results support the hypothesis that hyperglycemia and higher glucose variability are associated with worse cognitive performance. Despite previous suggestion of a link between hypoglycemia and incident dementia, lower antecedent fasting glucose may be associated with better executive function/processing speed in this population-based sample.

Objective: We investigated two hypotheses: First, incident CHD, defined as myocardial infarction or definite angina, leads to faster long-term cognitive decline. Second, among those with CHD, treatment with CABG surgery or PCI leads to slower long-term cognitive decline. Methods: The Cardiovascular Health Study is a cohort of US adults aged 65+. Global cognitive ability was assessed annually up to 9 times from 1990 to 1998 with the 100-point Modified Mini-Mental State Examination (3MS). We estimated trajectories of 3MS scores in the absence of stroke, adjusting for demographics, health behaviors, and comorbidities. For hypothesis 1, we compared 3MS trajectory after incident CHD with 3MS trajectory in the absence of CHD, censoring at first receipt of CABG/PCI. For hypothesis 2, among participants with CHD, we compared 3MS trajectory after first receipt of CABG/PCI with 3MS trajectory without CABG/PCI. Results: For hypothesis 1, of 4,122 participants, 398 had incident CHD during a mean of 5.9 years of follow-up. Figure Panel A shows model-predicted mean 3MS trajectories without CHD (blue) and after incident CHD (red) diagnosed at ages 70, 75, 80, or 85. Model-predicted 3MS score declined faster after incident CHD, especially for CHD diagnosed at age 80 or later. For example, after incident CHD at age 85, predicted 5-year decline in mean 3MS score through age 90 was 13.9 points (95% CI: 11.0, 16.7) versus 8.9 points (95% CI: 8.1, 9.7) among those without CHD. For hypothesis 2, of 1,183 participants who had prevalent or incident CHD, 118 had their first CABG/PCI during a mean of 4.1 years of follow-up. Model-predicted 3MS score declined faster after first receipt of CABG/PCI ( Figure Panel B ). Conclusions: Older adults diagnosed with incident CHD had faster average cognitive decline than those without CHD. However, treatment with CABG/PCI did not slow cognitive decline among those with CHD. This finding may be due to adverse effects of CABG/PCI on brain health or CABG/PCI recipients having more severe CHD or more cerebral atherosclerosis.

Background: Plasma free fatty acids (FFAs) are a byproduct of lipolysis largely derived from adipose tissue. High plasma FFA levels have toxic effects on a variety of organs central to cardiometabolic disease. Whether FFAs associates with cognitive decline or dementia remains unknown. Objective: To assess the association of plasma FFAs with risk of cognitive decline and dementia. Methods: Plasma FFAs were measured in participant samples from the 1992-1993 study visit of the Cardiovascular Health Study (CHS) (mean age 74 yrs; 59% female; 14% African American [AA]; 24% APOE4 carriers). A total of 3,242 non-demented participants were followed with adjudication for dementia through 1998-1999 (n=456 cases). In addition, cognitive decline was assessed annually among all CHS participants using the 100-point modified Mini-Mental State examination (3MSE; n=4,417) and the Digital Symbol Substitution Test (DSST; n=4,254). Results: Higher plasma FFAs levels were positively associated with risk of dementia (Figure: Log-rank p-value=0.04). In Cox regression analysis adjusted for demographics, FFAs remained associated with risk of dementia (HR per SD [0.20 mEq/L] =1.10; 95% CI, 1.00-1.22). In fully adjusted models that included lifestyle factors, medical history, C-reactive protein, LDL-C, HDL-C, and 3MSE at baseline, the hazard ratio per SD was 1.13 (1.01-1.26), with a particularly strong association observed among AA participants (HR=1.49 [1.10-2.01], p interaction=0.08). FFA levels did not differ by APOE genotype, and adjustment for genotype did not influence results. FFA levels from the baseline visit were also associated with a decline in both cognitive assessments over 6 years of follow-up, with significant time x FFA interactions (p=0.04 for 3MSE and p=0.002 for DSST). We observed no interactions with APOE genotype or race for cognitive outcomes. Conclusions: In non-demented older men and women, higher plasma FFA levels are associated with faster cognitive decline and higher risk of dementia over the subsequent years.

Introduction: Vascular risk factors (VRFs) predict incident dementia. Coronary artery calcium (CAC), capturing cumulative VRFs exposure, may predict dementia beyond VRFs. Hypothesis: In a community-based sample, baseline CAC is associated with incident dementia independent of VRFs assessed at baseline and APOE -ε 4 genotype. Methods: Among 6,814 participants of the Multi-Ethnic Study of Atherosclerosis (MESA), aged 45-84 years at baseline (2000-2002), we first identified candidate cases for incident all-cause dementia using hospital and death certificate ICD codes. For the analyses, we limited to cases whose diagnosis was validated using following criteria: 1) verifiable medical records available, 2) unlikely alternative diagnosis (e.g. delirium), and 3) written dementia diagnosis. Baseline CAC was categorized according to Agatston score (0, > 0 to -ε 4, and incident stroke prior to dementia diagnosis. Results: We had complete data on 6,295 individuals for analysis with a mean follow-up of 11.1 years (208 validated cases). Crude rates of dementia were greater with higher baseline CAC categories. Adjusted HR of dementia per one higher CAC category was 1.14 (95%CI 1.03, 1.26) (see Table). The associations were similar when excluding interim incident coronary heart disease, or cardiovascular diseases, or using a model that accounted for competing risk of death with no dementia diagnosis. Conclusion: Higher CAC score measured at baseline was independently and significantly associated with increased risk of dementia. This finding is consistent with an important role for vascular injury in development of dementia.

Background The influence of physical activity (PA) and fitness (PF) at older ages on changes in telomere length (TL), repetitive DNA sequences that may mark biologic aging, are not well established. Prior studies in older adults were cross-sectional, and few evaluated PF. Aim To investigate both cross-sectional and prospective associations of PA and PF with Leukocyte TL in older adults. Methods Among 582 adults age 73±5 y at baseline in CHS having serial LTL measures (leucocyte mean terminal restriction length) in 1992 and 1997, PA (walking pace, blocks, combined walking score; leisure time activity) and PF (15-ft walk, chair stands, grip strength) were assessed multiple times from 1989 to 1997. Cross-sectional associations were assessed using multivariable repeated-measures regression of cumulative average serial PA and PF measures from 1989 to 1992 against LTL in 1992; and cumulative averaged PA and PF from 1993 to 1997 against LTL in 1997. Longitudinal analyses assessed cumulative averaged PA and PF from 1989-92 against later changes in LTL from 1992-97; and changes in cumulative averaged PA and PF from 1989-92 to 1993-97 against changes in LTL from 1992-97. We evaluated all subjects, those without poor/fair self-reported health, and those without limited ADLs. Results Cross-sectionally, greater walking distance and chair test performance, but not other PA and PF measures, each associated with longer LTL ( P trend 0.007, 0.04 respectively). Lab measures of DNA integrity suggested measurable degradation that markedly attenuates ability to detect differences in LTL changes over time, especially over only 5 y. For example, baseline PA and PF from 1989-92 were unassociated with later changes in LTL. In contrast, changes in leisure time activity and chair test performance from 1992-97 were each inversely associated with LTL shortening. Conclusions Even late in life, certain PA and PF metrics are associated with longer LTL cross-sectionally and, even with likely substantial measurement error, with less LTL shortening over time.

Objective. We sought to determine whether atrial fibrillation (AF) was associated with cognitive decline in the absence of stroke. Methods. We included Cardiovascular Health Study participants aged 65+ who did not have AF or prior stroke at baseline and had at least one Modified Mini-Mental State Examination (3MSE) during the study. AF was identified by hospital discharge diagnoses and annual study ECGs. The 3MSE (score range 0–100) was administered annually up to nine times from 1990/91 to 1998/99. Mixed effects linear regression was used to characterize the relationship of cognitive decline with AF and age, adjusted for birth year, sex, race, education, smoking, alcohol use, diabetes, hypertension, systolic blood pressure, coronary heart disease, and heart failure. Interaction terms for AF and age variables with incident clinical stroke were used to allow estimation of cognitive decline in the absence of stroke. Results. Analyses included 5,199 participants with a mean of 6.7 3MSE scores. During follow-up 11% of participants (554 of 5,199) had AF identified. Panel A of the Figure shows mean 3MSE score trajectories for participants without AF (solid line) and after AF was identified at ages 70, 75, 80, and 85 (dashed lines). Panel B shows the trajectory for participants without AF set to zero (solid horizontal line) and differences in mean trajectories after AF was identified relative to participants without AF (dashed lines). Gray shading shows 95% CIs. Having newly-identified AF was not associated with a difference in 3MSE score relative to participants of the same age without AF. However, after AF was identified, 3MSE scores declined 2.3 points faster per five years, relative to participants without AF. For example, after AF was identified at age 80, mean 3MSE score declined 9.1 points by age 85, versus 6.8 points for participants without AF. Conclusions. In older adults without clinical stroke, the average rate of cognitive decline was greater after AF was identified.

Introduction: Leukocyte telomere length (LTL) is a complex genetic trait that is linked to higher risks of CVD, diabetes and other age-related disease. Little is known about the relationship between the intake of specific foods and LTL. Hypothesis: Foods that generally increase cardiometabolic risk, such as processed meat, will be associated with shorter LTL and higher rate of LTL attrition; foods that decrease risk, such as seafood, will have inverse associations with LTL and LTL attrition. Methods: We evaluated the relation between LTL and specific foods in 1,515 US adults age 65+ y participating in the Cardiovascular Health Study. Usual dietary habits were assessed by detailed food frequency questionnaires (FFQs) at baseline and year 6. Leukocyte DNA was extracted, and LTL measured via Southern blots of terminal restriction fragments, in blood samples collected at years 3 and 8. Clustered linear regression analysis was used to assess the multivariable-adjusted, cross-sectional relationship between specific foods and LTL (baseline FFQ vs. year 3 LTL and year 6 FFQ vs. year 8 LTL). Baseline diet was also evaluated in relation to the rate of LTL attrition over time. Results: In multivariate analyses, higher red and processed meat intakes were each associated with shorter LTL (p-trend Table ). Intakes of seafood and vegetables were associated with non-significant trends toward longer LTL In the subset with both baseline diet and change in LTL (n=617), a significant association between processed meats and greater 5-year LTL attrition was observed (p-trend=0.03). Higher nut intake was also associated with less LTL attrition across quintiles (p=0.05), although not monotonically (p-trend=0.35). Conclusions: Diet, in particular red meat and processed meat consumption, may play a role in telomere dynamics. Our findings support the need for further investigation of prospective relations in larger samples. Mean LTL LTL difference Q5 vs. Q1 (SE) Wald p-value * Q5 vs. Q1 p-trend * Q1 Q2 Q3 Q4 Q5 Fruits 6.33 6.35 6.30 6.35 6.28 −0.07 (0.04) 0.10 0.11 Vegetables 6.26 6.31 6.35 6.30 6.39 0.06 (0.04) 0.16 0.33 Nuts 6.34 6.31 6.33 6.32 6.27 −0.06 (0.06) 0.20 0.27 Whole grains 6.33 6.31 6.31 6.33 6.32 −0.04 (0.06) 0.39 0.63 Dairy 6.36 6.33 6.30 6.32 6.30 −0.07 (0.04) 0.11 0.09 Red meats 6.42 6.32 6.30 6.27 6.27 −0.14 (0.05) Processed meats 6.43 6.34 6.30 6.28 6.23 −0.15 (0.05) Seafood 6.30 6.28 6.32 6.34 6.37 0.05 (0.04) 0.29 0.16 * Adjusted for age, sex, race, education, smoking status, body mass index, and physical activity.

Introduction: Hypertension in Viet Nam is becoming an increasing source of mortality and morbidity. While data collection on hypertension in resource poor environments has been improving through the use of standardized surveys, little is known about how well treatments are being applied. Knowledge of gaps in diagnosis and treatment is necessary if appropriate public health programs are to developed. Using a community based observational sample in the Da Nang province of Viet Nam, we studied both the prevalence of undiagnosed hypertension, as well as the types of treatment those with hypertension received. Methods: Randomized cluster sampling methods were used to identify adults age 35 years and older in six communes of Da Nang province, Viet Nam. Using World Health Organization guidelines, data were collected on subject’s demographics, medical history, blood pressure and health behaviors. Subjects were also instructed to bring all medications they take regularly which were then recorded. Separate multivariate logistic regressions were used to identify both factors associated with awareness of hypertension as well as factors influencing whether one received medical therapy. Results: The initial survey examined 1,621 adults with 557 (34.5%) having hypertension. Of those, only 232 (41.7%) were aware of the diagnosis. Being male (OR 0.52, 95% CI: 0.36–0.75), older (OR 1.03, 95% CI: 1.02–1.05 per year) and having a low income (OR 1.11, 95% CI: 1.05–1.16 per million dong/month) were each independently associated with being unaware of one’s hypertension. Of those who were aware of their hypertension, 212 (91.4%) had been recommended some behavioral modification (exercise, decreased sodium intake, smoking cessation, or weight loss). Two-thirds (152/229) reported having taken antihypertensives in the past two weeks. Reported use of medication differed by region, 109/150 (72.7%) in urban regions versus 43/79 (54.4%) in rural/mixed-urban regions (p Conclusions: Undiagnosed hypertension remains a major health problem in the Da Nang province of Viet Nam. The vast majority of those aware of their hypertension were receiving at least some level of behavioral modification with a smaller, but substantial, majority receiving medications. Screening programs focused on young and poor men may be a rich target for improving hypertension control in Viet Nam.