Impact of Dietary Factors and Inflammation on Cognition among Older Adults

The Journal of Prevention of Alzheimer s Disease - JPAD Volume 2, Number 4, 2015 Impact of Dietary Factors and Inflammation on Cognition among Older Adults E.P. Handing 1, B.J. Small 1, S.L. Reynolds 1, N.B. Kumar 2 1. School of Aging Studies, University of South Florida, USA; 2. Moffitt Cancer Center, University of South Florida, USA Corresponding Author: Elizabeth Handing, School of Aging Studies, University of South Florida, Tampa, FL 33612, USA. Email: handing@mail.usf.edu J Prev Alz Dis 2015;2(4):220-226 Published online January 14, 2015, http://dx.doi.org/10.14283/jpad.2015.50 Abstract OBJECTIVE: This study examined the influence of age, nutrition (as measured through food diaries and serum/ plasma biomarkers) and inflammatory markers on cognitive performance in adults 60 years of age and older. DESIGN: A cross-sectional population based study, data from the National Health and Nutrition Examination Survey (NHANES; 2001-2002 wave). PARTICIPANTS: This study included 1,048 adults who had valid dietary data, blood biomarkers, were 60 years or older, completed the cognitive test, and had complete demographic information. METHOD: A series of regression models were used to examine the relationship between cognitive function as measured by the Digit Symbol Substitution Task (DSST), dietary factors/ biomarkers and inflammation. Mediation analyses were then utilized to examine whether individual nutrients accounted for the relationships between age and DSST performance. RESULTS: Dietary fat intake, serum vitamin E, serum folate, serum iron, plasma homocysteine, and serum vitamin D were significantly associated with better DSST performance. Elevated fibrinogen and C-reactive protein, were significantly associated with poorer cognitive function, but did not remain statistically significant after controlling for age, gender, education, ethnicity, income, and total calorie intake. Serum vitamin D and plasma homocysteine accounted for a portion of age-related variance in DSST. Specifically, higher levels of vitamin D were related to better DSST performance, while higher homocysteine resulted in poorer cognitive performance. CONCLUSION: Diet and nutrition are important modifiable factors that can influence health outcomes and may be beneficial to remediate age-related declines in cognition. Adequate nutrition may provide a primary preventive approach to healthy aging and maintenance of cognitive functioning in older adults. Key words: Nutrition, cognition, older adults. Introduction A s the aging population is steadily on the rise, it is estimated that the number of adults in the United States 65 and older will reach 88.5 million by 2050 (1). The majority of older adults experience some form of cognitive decline (2, 3) and this Received October 30, 2014 Accepted for publication December 18, 2014 220 has fueled research toward interventions aimed at ways for older adults to maintain and preserve their cognitive abilities. One potential intervention is through nutrition and dietary modification (2). A growing area of research in the aging field suggests that dietary components (antioxidant nutrients, fish, dietary fats, and B-vitamins) may play a role in the risk of age-related cognitive decline (4-6). This study will investigate the contribution of nutritional factors as well as inflammatory markers to cognitive function in older adults. The relationship between diet and cognitive decline has mainly been investigated using a single nutrient approach (7, 8). For example, Morris et al (9) found that higher Vitamin E intake (from food alone and supplementation) was related to less cognitive decline in older adults across four cognitive tasks. Llewellyn et al. (8) examined Vitamin D (as measured by levels of serum 25-hydroxyvitamin D) and risk of cognitive decline finding that Vitamin D deficiency was related to an increase in cognitive decline over 6 years. Recently, vitamin D deficiency has also been linked to increased risk of dementia and Alzheimer s disease (10). Although these studies provide insight into the link between diet and cognition, they are limited by viewing nutrients in isolation rather than considering nutrients from diet and have yet to include inflammation as a potential mediator in this pathway. In these analyses, data from the National Health and Nutrition Examination Survey (NHANES) 2001-2002 wave were used to examine dietary macronutrients (fat, protein, and carbohydrates), and select blood serum/ plasma biomarkers (vitamin B12, vitamin D, vitamin E, folate, iron, and homocysteine) (6). The purpose of this study is to examine the association between age, nutrition, cognitive performance (as measured by the Digit Symbol Substitution Task), and inflammatory markers in adults 60 and older. Additional studies suggest an association between the pathogenesis of cognitive decline and inflammatory markers including C-reactive protein (CRP), ferritin, and fibrinogen (11, 12). The release of C- reactive protein (CRP) and other inflammatory markers may contribute to increased cognitive decline via the inflammatory pathway (13). Previous literature suggests that CRP

JPAD - Volume 2, Number 4, 2015 Original Article levels are associated with mild cognitive impairment, a prodromal stage of Alzheimer s Disease (14). Other biomarkers such as fibrinogen and ferritin have been related to cognitive decline (15), however few studies have analyzed the association between these inflammatory markers, cognitive function, and dietary factors. This study examined how nutritional factors measured through diet and serum/plasma along with inflammatory markers may be associated with cognitive decline and investigates potential mediators to cognitive performance. Method Participants This project used data from the National Health and Nutrition Examination Survey (NHANES) waves 2001-2002. NHANES conducts an extensive nutritional survey combining in-person interviews, questionnaires, and physical examinations on individuals varying from children to older adults. NHANES over-samples persons 60 and older, African Americans, and Hispanics to include a diverse sample population. Recruitment was performed using a stratified, multistage probability sample of non-institutionalized individuals living in the United States (16). Participation in the study required an in-home visit to administer questionnaires and cognitive testing, as well as a visit to a mobile examination center (MEC) for a comprehensive health examination. Our target population was adults 60 and older with valid dietary information and health measures. Measures Demographic variables Information on age, gender, ethnicity, education, and income was collected from a self-reported questionnaire. Age was coded as a continuous variable, gender was categorical, ethnicity was categorized as Non-Hispanic White, Non-Hispanic Black, Mexican American, other Hispanic, and other race, education was categorized as less than high school, high school, some college, or college graduate or above, and income was categorized as earning less than $14,999, earning $15,000-$49,000 or earning more than $50,000. Cognitive assessment Cognitive functioning was measured by the Digit Symbol Substitution Task (DSST) from a version of WAIS-IV. Participants were instructed to draw symbols corresponding to a number key, and the score is the number of correct symbols drawn within a period of 221 120 seconds. One point is given for each correctly drawn symbol completed within the time limit. The range in our sample was from 6-100 points, the average was 44.8 points. Dietary analysis Each participant completed a 24-hour dietary recall which was completed at the in-person interview. Participants were asked to describe their previous day s consumption of foods and beverages using detailed diagrams and pictures for accurate portion size and ingredients. The second dietary recall was collected by telephone and was scheduled 3 to 10 days later. Dietary information from both days was averaged for total nutrient values. NHANES 2001-2002 nutrient intakes were calculated using USDA s Food and Nutrient Database for Dietary Studies (FNDDS). The following dietary nutrients were included as predictors in the analyses: total fat (gm), protein (gm), and carbohydrates (gm). Serum samples Approximately 6 tablespoons of blood was drawn via venipuncture by a certified medical professional during the MEC visit. In the current study, the following serum/ plasma samples were and used for analyses: serum vitamin B12 (pg/ml), serum folate (ng/ml), plasma homocysteine (μmol/l), serum iron (μg /dl), vitamin E serum (μg/ml), and vitamin D serum (ng/ ml). Inflammatory markers CRP (mg/dl), fibrinogen (mg/dl), and ferritin (ng/ ml) were collected through the blood draw as part of the NHANES medical examination. Collection techniques and details can be found in the NHANES Laboratory/ Medical Technologists Procedures Manual (17). Statistical Analysis The analytic strategy consisted of regression models to examine the predictive value of age, nutrients, and inflammation to cognitive performance among older adults. First, a series of univariate linear regression models were used to examine individual dietary and serum biomarkers and inflammatory markers on DSST, independent of demographic characteristics. Second, a multiple regression model was conducted based upon significant predictors from the previous models. Finally, mediation analyses were used to examine the effects of dietary/ serum biomarkers and age on DSST performance. It is known that areas of cognition decline with age, however it is not known how diet may influence cognition. By examining age, nutrient status,

Original Article IMPACT OF DIETARY FACTORS AND INFLAMMATION ON COGNITION AMONG OLDER ADULTS and cognitive function in a mediation model, we are better able to examine these continuous variables in relation to each other. Mediation provides calculations of direct effects for the model X ->Y (where X represents age and Y represents DSST performance) as well as indirect effects X-> Z -> Y where Z represents nutrients as a mediator between age and DSST performance. The bootstrapping method (with bias correction and 5,000 iterations) was used to evaluate the direct and indirect mediating paths (18). All analyses were analyzed using SAS software (SAS Institute, Cary, NC) Version 9.3. Results A flowchart with the number of participants included in analyses is indicated in Figure 1. Participants were not included in analyses if they had missing dietary information (n=1,408), missing cognitive testing (n=8,361), were outliers by 3 standard deviations in cognitive testing (ie., score of <5; n=15), were outliers by 3 standard deviations in calorie intake (ie., <500 kcal/day or >5,000 kcal/day; n= 21), or missing demographic information (n=188). Our final analytic sample was 1,048 older adults. Table 1 displays the basic demographic characteristics of the sample. The mean age was approximately 71 years of age and mean calorie consumption was almost 1,800 calories per day. Figure 1. Flowchart of participants from the National Health And Nutrition Examination Survey (NHANES) 2001-2002 performance. Table 3 depicts the multivariate analyses of significant nutrient markers to DSST performance controlling for age, gender, education, ethnicity, income, and total calorie intake. Positive associations were found with higher intake of dietary fat, serum vitamin E, serum folate, serum iron, and serum vitamin D being related to better DSST performance, while a negative association was found for plasma homocysteine meaning higher values were related to lower DSST scores. There were no significant results for inflammatory factors. Table 1. Sample demographic characteristics NHANES Sample n= 1,048 Age, mean (SD) 70.85 (7.65) Gender, % Women 50.52% Ethnicity, % Non-Hispanic White 65.59% Non-Hispanic Black 13.93% Mexican American 15.83% Other Hispanic 2.56% Other Race 2.09% Education, % Less than high school 33.02% High school 24.76% Some college 21.92% College graduate or above 20.30% Income, % Less than $14,999 18.72% $15,000-$49,000 46.87% More than $50,000 34.41% Body Mass Index (BMI), % Underweight (16-18.5) 1.71% Normal (18.5-25) 26.82% Overweight (25-30) 41.42% Table 2 presents the results of the univariate regression models examining demographic variables, nutrients, and inflammatory markers as predictors for cognitive functioning. Statistically significant demographic factors included, age, gender, ethnicity, education, income, and total calorie intake. Being older, male, non-hispanic black, having a high school education or less, having less than $15,000 income, and a low calorie intake was related to worse DSST performance. All dietary macronutrients and serum biomarkers were significant predictors of DSST performance. All of the nutrients and biomarkers were positively related to DSST performance, except for homocysteine, which was related to worse cognitive performance. For the inflammatory markers, higher values of fibrinogen and CRP were related to worse DSST Obese (>30) 30.05% Energy (kcal), mean (SD) 1,778 (687) Note: NHANES= National Health And Nutrition Examination Survey The final set of analyses examined the potential for the statistically significant nutrient markers to mediate age-related differences in DSST performance (Figure 2). The results indicated that serum vitamin D and plasma homocysteine acted as full mediators of agerelated differences in performance (indirect effect and 95% CI in brackets): plasma homocysteine (-.036, CI [-.071 to -.01]), serum vitamin D (.017, CI [.004 to.038]). Homocysteine levels increased with age, but higher levels were associated with poorer performance. Vitamin D level increased with age and resulted in better DSST 222

JPAD - Volume 2, Number 4, 2015 Original Article Table 2. Univariate results for predictor variables with DSST performance as the outcome, n=1048 Covariates a Std β SE p-value R2 Age -0.288 0.068 <.001* 0.083 Female 0.108 1.075 <.001* 0.012 Non-Hispanic White 0.108 0.582 <.001* 0.012 > High school education 0.516 0.336 <.001* 0.265 > $15,000 income 0.306 0.143 <.001* 0.094 Calorie Intake 0.188 0.001 <.001* 0.035 BMI 0.022 0.100 0.483 <0.001 Nutrient b Dietary Fat 0.183 0.982 <.001* 0.034 Dietary Protein 0.146 1.180 <.001* 0.021 Dietary Carbohydrate 0.152 1.250 <.001* 0.023 Serum Vit. B 12 0.099 0.617 0.001* 0.009 Serum Vit. D 0.110 1.283 <.001* 0.012 Serum Vit. E 0.172 1.293 <.001* 0.030 Serum Folate 0.128 1.009 <.001* 0.016 Serum Iron 0.145 1.377 <.001* 0.021 Plasma Homocysteine -0.206 1.596 <.001* 0.042 Inflammatory Marker b Fibrinogen -0.117 2.667 <.001* 0.014 CRP -0.063 0.682 0.040* 0.004 Ferritin -0.059 0.563 0.058 0.003 Note: Vit= Vitamin, CRP= C-Reactive Protein, BMI= Body Mass Index; *=significance at the.05 alpha level; a. Estimates are based upon the covariates entered together on the first step; b. Estimates are based upon variables entered singly after controlling for covariates. Table 3. Regression model of nutrients and inflammation with DSST as outcome controlling for age, sex, education, ethnicity, income, and total calorie intake. An asterisk indicates significance at the.05 alpha level Nutrient Std β SE p-value Dietary Fat 0.090 1.305 0.025* Dietary Protein 0.025 1.418 0.488 Dietary Carbohydrate -0.030 1.691 0.475 Serum Vit. B 12 0.048 0.535 0.070 Serum Vit. D 0.098 1.012 <.001* Serum Vit. E 0.057 1.075 0.024* Serum Folate 0.088 0.826 <.001* Serum Iron 0.106 1.091 <.001* Plasma Homocysteine -0.086 1.361 <.001* Inflammatory Marker Fibrinogen <.0001 2.160 0.693 CRP -0.037 0.542 0.124 Ferritin -0.037 0.454 0.136 223

Original Article IMPACT OF DIETARY FACTORS AND INFLAMMATION ON COGNITION AMONG OLDER ADULTS performance. Several nutrients acted as partial mediators (indirect effect and 95% CI in brackets): dietary fat (.001, CI [-.006 to.012]), serum vitamin E (.002, CI [-.008 to.017]), serum folate (.017, CI [-.008 to.047]) and serum iron (-.001 CI [-.015 to.013]. In the case of iron, older age was associated with lower values but not statistically significant, however higher iron values resulted in better cognitive performance. Vitamin E, and folate increased with age, but higher levels were not significantly related to DSST performance. Figure 2. Multiple mediation with nutrient intake as mediators controlling for gender, education, ethnicity, income, and total calorie intake. The multiple mediation model of X -> Z -> Y (where X represents age, Z represents the nutrient, Y represents the Digit Symbol Substitution Task (DSST) performance. The bootstrapping method with bias corrected confidence estimates (based upon 5,000 iterations) was used to test the mediation hypothesis (18). Note: Vit= vitamin, hmcy= homocysteine, an asterisk indicates significance at the.05 alpha level. Discussion Findings from the univariate and multiple regression models suggest that several nutrients were associated with performance on the DSST including dietary fat, serum values for vitamin E, folate, iron, plasma homocysteine, and vitamin D. Mediation analyses further examined these relationships and found vitamin D and homocysteine acted as significant mediators between age and cognitive performance. Mediation for biomarkers and cognitive performance Recently Littlejohns and colleagues (10) reported that low serum vitamin D levels (< 25nmol/L) resulted in a two-fold increase in risk for dementia and Alzheimer s Disease. Additional studies have found evidence suggesting a relationship between insufficient serum vitamin D and cognitive decline (19, 20). Our results indicate that higher serum vitamin D was related to better cognitive performance on the DSST. A study by Bowman and colleagues (21) examining multiple biomarkers and cognitive function found that a dietary nutrient biomarker pattern high in antioxidant vitamins B, C, D, and E, was related to better executive function, attention, visuospatial function, and global cognition. Our findings of vitamin D and E as important nutrients for brain health and function are supported. Additionally, plasma homocysteine was found to negatively mediate the relationship between age and cognitive performance. With age, homocysteine increased while higher values negatively affected DSST scores. Elevated plasma homocysteine concentrations have been consistently associated with both cognitive impairment and dementia and found to negatively mediate the relationship between age and cognitive performance (22). Research from the Framingham study found an association between higher homocysteine levels (>14 µmol per liter) and nearly a two-fold increase in risk of Alzheimer s disease (23). Homocysteine is a sulfur-containing amino acid generated through the demethylation of the essential amino acid methionine. It is largely catabolized by trans-sulfuration to cysteine, but it may also be remethylated to methionine. Deficiencies in the homocysteine re-methylation cofactors cobalamin (B12) and folate, as well as the trans-sulfuration cofactor vitamin B6, are commonly seen in the elderly population, with a resultant increase in homocysteine with advancing age (24). Increasing number of studies have demonstrated that high red cell folate levels were associated with worse long-term episodic memory, total episodic memory, and global cognition (25-27). In a large population-based sample of elderly people, the association between high homocysteinemia and decreased cognition was only seen in participants with low folate levels (26). Similarly, Blasko et al (27) reported higher levels of homocysteine predictive of moderate/severe global brain atrophy at five years while folate demonstrated a protective ability to reduce conversion to dementia in moderately cognitively impaired patients. Based on these studies, hyperhomocysteinemia continues to be consistently associated with an increased risk of cognitive impairment in the elderly with more recent studies suggesting that folate levels may also influence the course of cognitive decline. These observations may have significant implications for future interventions with nutritional cofactors for proper functioning of the methionine cycle which may ultimately improve methylation and protect the brain from damage. 224

JPAD - Volume 2, Number 4, 2015 Original Article Interpretation of partial mediators Partial mediation was found for serum vitamin E and folate, which positively increased with age. Dietary fat intake and serum iron were not related to age, but were significantly related to better cognitive performance. Fat intake can be beneficial for cognition, but this interpretation should be evaluated with care. Presently, omega 3 fatty acids via dietary intake (i.e. higher amounts of fish) have been associated with a reduced risk for dementia (28). In a study by Morris et al. (5) older adults who consumed 1 or more fish meals per week compared with those with less than weekly consumption showed a reduced rate of decline by 10% to 13% as indicated by the Mini Mental Status Exam score. Conversely, high intake of saturated fat (found in cheese, red meat, and whole milk) has been related to an increased risk of dementia with the greatest effect for vascular type dementia (Relative Risk Ratio =2.9, 95% CI (0.6-13.8) p =0.01) (29). In our study, we further investigated the type of fat and found no significant effect for saturated fat, monounsaturated fat, or polyunsaturated fat on cognitive performance (data not shown). Therefore, our finding that dietary fat intake was associated with cognitive performance should be interpreted within a holistic dietary pattern, rather than individual fat values. In our study, serum iron was predictive of better cognitive performance. The study population had an average of 86.36 µg/dl (normal range 60-170 µg/dl) meaning most older adults in our sample were receiving adequate iron from their diet and/or supplements. This finding is provocative in light of the association of the role of iron and other metals like copper in inducing oxidative stress and should be more thoroughly examined with additional studies, in particular its association with homocysteine. The involvement of homocysteine involving iron dysregulation and oxidative stress designated as the ferric cycle has also been implicated in AD (30). Strengths & Limitations of interest given our aim was to examine cognitive performance in older adults. Our study also used selfreported dietary records, which are susceptible to over/ under estimation. Conclusion In summary, age accounts for some of the decline in DSST performance, but a poorer performance may be exacerbated by an unhealthy diet. Initially, inflammatory markers showed an association with poorer cognitive performance, however after controlling for age, sex, education, ethnicity, income, and total calorie intake, results were no longer significant which limited the potential for examining them as mediators. From our mediation results from nutrients, we found that serum vitamin D served as a full mediator and was related to better cognitive performance. A significant negative meditation of plasma homocysteine was related to worse cognitive performance. Future studies are warranted to provide longitudinal analysis of the effects of age, dietary, and inflammatory factors on cognitive functioning. Multiple cognitive tests should be included to examine whether nutrients affect certain parts of the brain more than others, which could have significant clinical public health implications such as revising diet recommendations for older adults. Dietary factors and nutrient values may play a pivotal role in brain health among older adults and can be translated into lifestyle modifications to promote healthy aging and remediate cognitive decline. Conflict of interest: Authors have no conflicts of interest. Ethical standards: Institutional Review Board (IRB) approval and documented consent was obtained from all participants. Funding: NHANES is funded by the U.S. government, National Center for Health Statistics, Centers for Disease Control and Prevention (CDC). The sponsors had no role in the design and conduct of study; in the collection, analysis, and interpretation of the data; in the preparation of the manuscript; or in the review and approval of the manuscript. References Strengths of this study include the detailed dietary record information on over 1,000 older adults. NHANES over-recruits older adults with a focus on dietary assessment and health status of Americans. It is important to note that our results are from a cross sectional sample and longitudinal data would strengthen the results. Our study is limited by containing a single assessment of cognition which may not capture the specific changes that occur with age. We were unable to examine the potential for reverse causation (i.e. lower cognition impacting dietary choices) which could be addressed in a longitudinal model. Previous studies show that processing speed changes with age and this may be 225 1. Vincent GK, Velkoff VA. The next four decades: The older population in the United States: 2010 to 2050.2010. Population estimates and projections. Washington, D.C., U.S. Dept. of Commerce, Economics and Statistics Administration, U.S. Census Bureau: p 14 2. Deary IJ, Corley J, Gow AJ et al. Age-associated cognitive decline. Brit Med Bull. 2009; 92:1 135-152 3. Salthouse T. What and when of cognitive aging. Curr Dir Psychol Sci. 2004; 13:4 140-144 4. Spencer J. Flavonoids and brain health: Multiple effects underpinned by common mechanisms. Genes Nutr. 2009; 4: 243-250 5. Morris MC, Evans DA, Tangney CC, Bienias JL, Wilson RS. Fish consumption and cognitive decline with age in a large community study. Arch Neurol. 2005; 62:12 1849-1853 6. Gu Y, Scarmeas N. Dietary patterns in Alzheimer s disease and cognitive aging. Curr Alzheimer Res. 2011; 8:5 510-519 7. Kang JH, Cook N, Manson J, Buring JE, Grodstein F. A randomized trial of vitamin E supplementation and cognitive function in women. Arch Intern Med. 2006; 166:22 2462-2468 8. Llewellyn DJ, Lang IA, Langa KM et al. Vitamin D and risk of cognitive decline in elderly persons. Arch Intern Med. 2010; 170:13 1135-1141 9. Morris MC, Evans DA, Bienias JL, Tangney CC, Wilson RS. Vitamin E and cognitive decline in older persons. Arch Neurol. 2002; 59:7 1125-1132

Original Article IMPACT OF DIETARY FACTORS AND INFLAMMATION ON COGNITION AMONG OLDER ADULTS 10. Littlejohns TJ, Henley WE, Lang IA et al. Vitamin D and the risk of dementia and Alzheimer disease. Neurology. 2014; 83 1-10 11. Luciano M, Marioni RE, Gow AJ, Starr JM, Deary IJ. Reverse Causation in the Association Between C-Reactive Protein and Fibrinogen Levels and Cognitive Abilities in an Aging Sample. Psychosom Med. 2009; 71:4 404-409 12. Schiepers OJG, van Boxtel MPJ, de Groot RHM et al. Serum Iron Parameters, HFE C282Y Genotype, and Cognitive Performance in Older Adults: Results From the FACIT Study. J Gerontol. A Biol Sci Med Mci. 2010; 65:12 1312-1321 13. Teunissen CE, van Boxtel MP, Bosma H et al. Inflammation markers in relation to cognition in a healthy aging population. J Neuroimmun. 2003; 134:1-2 142-150 14. Schmidt R, Schmidt H, Curb JD et al. Early inflammation and dementia: a 25-year follow-up of the Honolulu-Asia Aging Study. Ann Neurol. 2002; 52:2 168-174 15. Marioni RE, Stewart MC, Murray GD et al. Peripheral levels of fibrinogen, C-reactive protein, and plasma viscosity predict future cognitive decline in individuals without dementia. Psychosom Med. 2009; 71:8 901-906 16. National Health and Nutrition Examination Survey: Survey Operations. Available from: http://www.cdc.gov/nchs/nhanes.htm 17. National Health and Nutrition Examination Survey Laboratory/Medical Technologists Procedures Manual (2002); Available from: http://wwwn.cdc. gov/nchs/nhanes/search/nhanes01_02.aspx 18. Preacher KJ, Hayes AF. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behav Res Methods Instrum Compt. 2004; 36:4 717-731 19. Annweiler C, Montero-Odasso M, Llewellyn DJ et al. Meta-analysis of memory and executive dysfunctions in relation to vitamin D. J Alzheimers Dis. 2013; 37:1 147-171 20. Balion C, Griffith LE, Strifler L et al. Vitamin D, cognition, and dementia: a systematic review and meta-analysis. Neurology. 2012; 79:13 1397-1405 21. Bowman GL, Silbert LC, Howieson D et al. Nutrient biomarker patterns, cognitive function, and MRI measures of brain aging. Neurology. 2012; 78:4 241-249 22. Quadri P, Fragiacomo C, Pezzati R et al. Homocysteine, folate, and vitamin B-12 in mild cognitive impairment, Alzheimer disease, and vascular dementia. Am J Clin Nutr. 2004; 80:1 114-122 23. Seshadri S, Beiser A, Selhub J et al. Plasma homocysteine as a risk factor for dementia and Alzheimer s disease. New Eng J Med. 2002; 346:7 476-483 24. Miller AL. The methionine-homocysteine cycle and its effects on cognitive diseases. Altern Med Rev. 2003; 8:1 7-19 25. Faux NG, Ellis KA, Porter L et al. Homocysteine, vitamin B12, and folic acid levels in Alzheimer s disease, mild cognitive impairment, and healthy elderly: baseline characteristics in subjects of the Australian Imaging Biomarker Lifestyle study. J Alzheimers Dis. 2011; 27:4 909-922 26. Vidal JS, Dufouil C, Ducros V, Tzourio C. Homocysteine, folate and cognition in a large community-based sample of elderly people--the 3C Dijon Study. Neuroepidemiology. 2008; 30:4 207-214 27. Blasko I, Hinterberger M, Kemmler G et al. Conversion from mild cognitive impairment to dementia: influence of folic acid and vitamin B12 use in the VITA cohort. J Nutr Health Aging. 2012; 16:8 687-694 28. Barberger-Gateau P, Letenneur L, Deschamps V et al. Fish, meat, and risk of dementia: cohort study. BMJ. 2002; 325:7370 932-933 29. Kalmijn S, Launer LJ, Ott A et al. Dietary fat intake and the risk of incident dementia in the Rotterdam Study. Ann Neurol. 1997; 42:5 776-782 30. Dwyer BE, Takeda A, Zhu X, Perry G, Smith MA. Ferric cycle activity and Alzheimer disease. Curr Neurovasc Res. 2005; 2:3 261-267 226