Neurocognition and Micronutrients in the Elderly

The concept of healthy cognitive aging is gaining importance as rapid population aging is taking place in the Western world. As a greater percentage of the population moves toward old age, their medical and primary care needs will be exacerbated, resulting in increased financial pressure on the health-care system. Furthermore, an aging population will result in an increased prevalence of age-related neurodegen-erative disorders such as dementia. In 2005, it was estimated that 24 million people worldwide were living with dementia, and it is predicted that this quantity will double every 20 years to reach 81 million by the year 2040 (Ferri et al., 2005).

Age constitutes the major risk factor for Alzheimer’s disease (AD), and it is estimated that in developed countries such as the United States, approximately 13% of individuals over the age of 65 and 48% of persons over the age of 85 are affected by this disease (Thies and Bleiler, 2011). The cost of caring for those with dementia constitutes a significant financial burden on tax payers and emotional burden on family members. With longer life expectancy anticipated, there are many benefits to maintaining both physical and cognitive well-being, including enhanced quality of life and a lower risk of dementia onset.

Cognitive deterioration occurs across the life span and is a feature not only of AD, but also the normal aging process. In many cases the experience of cognitive decline may be a precursor to the development of AD (Goedert and Spillantini, 2006). Consequently, to delay the onset or rate of dementia, it may be necessary to target interventions to individuals prior to the appearance of cognitive decline. Currently there is a scientific interest in the potential of health and lifestyle interventions to improve cognitive function or slow the rate of decline in the elderly. There is evi-dence from randomized controlled trials to indicate that aerobic exercise programs (Baker et al., 2010), mental training (Valenzuela and Sachdev, 2009), and dietary supplements consisting of omega-3 fatty acids (Yurko-Mauro et al., 2010), herbal preparations (Mix and Crews, 2002; Pipingas et al., 2008), or vitamins (Durga et al., 2007) may be capable of improving cognitive function and potentially serve as inter-ventions against cognitive decline.

Observations from prospective and epidemiological studies have demonstrated that maintaining adequate vitamin and nutritional status may be particularly impor-tant for cognitive function in the elderly. For instance, vitamin depletion has been shown to precede cognitive decline (Kado et al., 2005), and both intake of specific nutrients and circulating levels of vitamins in the blood have been correlated with cognitive function in healthy elderly (Perrig et al., 1997; Maxwell et al., 2005).

Vitamins and micronutrients are chemicals in the diet that do not belong to the major categories of fats, proteins, or carbohydrates and cannot be synthesized by the body in large enough quantities for normal requirements (Huskisson et al., 2007). Vitamins and mineral micronutrients must be introduced through food intake and this can become a problem for the elderly who are vulnerable to vitamin and mineral deficiencies due to decreased appetite caused by lower energy needs (Nieuwenhuizen et al., 2010) and reduced absorption in the gut (Baik and Russell, 1999). To date research in this area has focused on several important questions. First, can dietary interventions using micronutrients slow the rate of cognitive decline in those with dementia, and do they possess the potential to prevent the onset of dementia? Second, and of greater relevance to this post, can nutritional interventions with selected micronutrients improve cognitive performance or slow the rate of cognitive decline in healthy elderly? This post will address the latter question, with a focus on relationships between micronutrients and neurocognition in elderly humans. Predominantly, our discussion will center on recent studies that have investigated the association between cognition in healthy elderly and specific B group vitamins, antioxidant vitamins, and multivitamins.

Even in healthy elderly, changes in cognition occur across the life span, with memory decline widely held to be one of the hallmarks of advanced age. Decline occurs to processes that rely on complex, controlled, goal-oriented behavior, including per-formance monitoring, the generation of future goals, and the ability to adjust behav-ior in response to feedback (Budson and Price, 2005). These cognitive operations are referred to as executive function and are particularly important for effective working memory and episodic memory performance. In comparison, other more declarative forms of memory including vocabulary and verbal IQ remain relatively intact in older adults (Christensen, 2001). Crystallized intelligence, described as knowledge gained from cultural influences and experience tends to increase until age 60, rather than decrease with age (Jones and Conrad, 1933). Beyond the age of 70, crystallized abilities decline, albeit to a lesser degree than fluid intelligence (Christensen et al., 1994).

Disruptions in the frontal–striatal system and medial temporal cortical regions are thought to contribute to decline in executive function over the life span (Buckner, 2004). Evidence from the field of neuroimaging has corroborated these findings, demonstrating age-related changes in frontal neural activity during epi-sodic memory, semantic memory, and working memory activation tasks (Cabeza, 2002; Phillips and Andrés, 2010). Loss of brain volume can occur as neurons become more susceptible to the effects of mitochondrial dysfunction, changes in metabolic rate, oxidative stress, and neuroinflammation (Floyd and Hensley, 2002). This decrease in brain volume exerts effects upon cognitive function, with age-related atrophy shown to be a predictor of cognitive decline on IQ measures (Rabbit et al., 2008). Volumetric imaging studies indicate that the frontal white matter is preferentially vulnerable to aging (Raz et al., 2005; Firbank et al., 2007). Small infarcts induce anterior white-matter deterioration (Pugh and Lipsitz, 2002), which in turn may decrease the efficiency of frontally mediated executive pro-cesses (Buckner, 2005).

Longitudinal studies have shown that cognitive change in healthy elderly is not a unitary process and that advancing age is accompanied by greater interindi-vidual variance, particularly in the domains of memory (Rabbitt et al., 2004) and cognitive speed (Christensen, 2001; Wilson et al., 2004). The diversity of these trajectories indicates that cognitive aging does not occur at the same rate for all individuals and some will experience greater decline than others. Individual dif-ferences stemming from genetic risk factors (Alexander et al., 2007), education (Ardila et al., 2000), cardiovascular function (Beeri et al., 2009), physical activity (Larson et al., 2006), depression (Depp and Jeste, 2006), and nutritional status (Moreiras et al., 2007) represent a few of the more widely researched predictors of cognitive decline in the elderly. In some cases, poorer memory performers may be in the prodromal stage of an age-related neurodegenerative disease such as AD as many of the same cognitive domains to be affected by the normal aging pro-cess are exacerbated in AD (Collie and Maruff, 2000). AD is the most common form of dementia in the elderly and is characterized by the DSM-IV as a marked decline from previous functioning in short-term memory and a severe disruption to language, planning, or visual processing (American Psychiatric Association, 2000). Risk factors for AD include advanced age, genetic factors, vascular dis-ease, hypercholesterolemia, hypertension, atherosclerosis, coronary heart disease, smoking, diabetes, and obesity (Carr et al., 1997; Duron and Hanon, 2008; Lange-Asschenfeldt and Kojda, 2008).

Computed tomography (CT) and magnetic resonance imaging (MRI) have shown that the neuropathological process of AD includes enlargement of ventricles due to substantial loss of brain tissue, nerve cells, synapse, and dendrites caused by the presence of neurofibrillary tangles and beta amyloid plaques (Rusinek et al., 1991; Petrella et al., 2003; Kidd, 2008). Pathologically, AD is characterized by beta amyloid plaques (AP) and neurofibrillary tangles caused by the deposition of abnor-mal proteins. Neutritic or senile plaques are extracellular deposits of Ap, whereas tangles are intracellular aggregates formed by a hyperphosphorylated form of the microtubule-associated protein tau (Blennow et al., 2006). The pathogenic mecha-nism of AD has been suggested to be an imbalance between the production and clearance of Ap in the brain, leading to neuronal degeneration and dementia (Hardy and Higgins, 1992). Damage initially occurs to the large cortical neurons in the temporal lobe and later in the association areas (Braak and Braak, 1991; Norfray and Provenzale, 2004).

Cholinesterase inhibitors are the most commonly used treatment for AD and are used to prevent the breakdown of acetylcholine (ACh), a neurotransmitter involved in learning, memory, and attention at the synaptic junctions (Lleo, 2007). By increasing the brain synaptic availability of acetylcholine, remaining neurons are able to function more effectively, but cognitive benefits fade as deterioration worsens and cholinesterase inhibitors are unable to reverse the process of this dis-ease (Kidd, 2008).

Mild cognitive impairment (MCI) is a term used to define individuals who have experienced a decline in cognition that is greater than expected for their age and edu-cation level but is not severe enough to meet the criteria for dementia (De Mendonca et al., 2004). In many cases, MCI represents a preclinical stage of AD, particu-larly in those with the amnestic form of MCI who have been observed to develop dementia at a rate of approximately 10%–15% per year as compared with healthy controls who convert to dementia at a rate of 1%–2% per annum (Petersen, 2007). Trials of cholinesterase inhibitors in patients with MCI have shown only limited cognitive improvements and a high rate of adverse events suggesting that they are not suitable to serve as interventions in the dementia process (Birks and Flicker, 2006; Allain et al., 2007; Sobów and Kloszewska, 2007). Research has shown that low levels of B vitamins (Clarke et al., 1998) and antioxidants (Rinaldi et al., 2003; Mecocci, 2004) exist in cases of MCI and AD indicating that low vitamin status may be associated with the neuropathological process of AD and possibly precede this age-related condition. Consequently, it has been posited that individuals at high risk of developing cognitive decline or dementia may experience benefits from dietary interventions consisting of vitamin supplements (Jelic and Winblad, 2003). Further evidence reviewed in the subsequent sections of this post may indicate that vita-mins including B vitamins and antioxidants are also important for the maintenance of cognition in healthy elderly.

Vitamin B deficiency may exert more direct effects on cognitive function via the role of these vitamins in the production of neurotransmitters in the brain. In order to maintain normal levels of homocysteine, vitamin B6, B12, and folic acid are required in the methylation of homocysteine to methionine. The amino acid methionine is essential to one-carbon metabolism, a series of biological processes crucial for DNA synthesis, DNA repair, and various methylation reactions (Mattson and Shea, 2003). In the central nervous system, methionine is required in the syn-thesis of S-adenosylmethionine (SAM), the sole donor for methylation reactions in the brain. The neurotransmitters dopamine, norepinephrine, and serotonin, as well as proteins, phospholipids, DNA, and myelin, are the products of these reactions (Selhub et al., 2000). Additionally, SAM is essential for maintenance of choline in the CNS as well as the production of acetylcholine and the antioxidant glutathione (Tchantchou et al., 2008). In accordance with the hypomethylation hypothesis, some loss of cognitive function in the elderly may be the end result of a lower produc-tion of SAM and consequently neurotransmitters due to B12 and folate deficiency (Calvaresi and Bryan, 2001).

A second hypothesis has proposed that impaired neurocognitive function in the elderly may stem from the injurious effects of elevated levels of homocysteine (Calvaresi and Bryan, 2001). In the brain, elevated homocysteine increases oxidative stress and DNA damage, triggers apoptosis, and imparts excitotoxic effects (Sachdev, 2005). In vitro, homocysteine disrupts neuronal homeostasis by multiple routes including N-methly-D-aspartate (NMDA) channel activation leading to excessive calcium influx and glutamate excitotoxicity (Ho et al., 2002). Hyperhomocysteinemia may also induce gene expression and interact with specific targets including cellular receptors, intracellular proteins, and molecules of nitric oxide, leading to neuropa-thology (McCaddon, 2006).

Elevated homocysteine also exerts a detrimental effect on the cardiovascu-lar system via pro-coagulant actions toward platelets and vascular endothelium (De Koning et al., 2003). Vascular damage may be related to reduced bioavailability of endothelial nitric oxide, which is a powerful vasodilator (Obeid and Herrmann, 2006). Furthermore, homocysteine-induced damage to endothelial cells promotes arteriosclerosis, which is in turn associated with poorer cognitive function (Breteler et al., 1994; Knopman et al., 2001; Seshadri, 2006). Collectively these findings sug-gest homocysteine may affect cognitive function in the elderly, via direct effects on the brain or through indirect mechanisms operating on the cardiovascular system.

It is thought that free radicals and oxidative stress contribute to the neuronal changes responsible for cognitive decline in normal and pathological aging. Oxidative stress represents a disturbance in the equilibrium status of prooxidant reactions involving oxygen free radicals and those involving antioxidants (Valko et al., 2007). It is the maintenance of this prooxidant/antioxidant balance via redox homeostasis that is vital for healthy cellular function. Vitamin A, C, E, selenium, and coenzyme Q10 possess potent antioxidant actions and protect neural tissue from aggression by free radicals (Bourre, 2006). Beta-carotene is a precursor to vitamin A and is a powerful antioxidant. Vitamin C is known to interact synergistically with B complex vitamins and is essential for the metabolism and utilization of folic acid (Huskisson et al., 2007). Vitamin C levels are particularly high in the brain and this antioxidant is also necessary for the production of some neurotransmitters and the transforma-tion of dopamine into noradrenalin (Bourre, 2006). Vitamin E is a lipid soluble chain-breaking antioxidant, which exercises neuroprotective effects against reactive oxygen species injuries (Cantuti-Castelvetri et al., 2000). Vitamin E interacts syner-gistically with selenium possibly increasing the antioxidant capacity of this vitamin (Bourre, 2006). Coenzyme Q10 is also a lipid soluble molecule that acts as an anti-oxidant and coenzyme for mitochondrial enzymes (Boreková et al., 2008).

According to the free radical hypothesis of aging, detrimental age-related changes take place in the brain as the result of an inability to cope with oxidative stress that occurs throughout the life span (Beckman and Ames, 1998). Oxidative stress can be defined as an excessive bioavailability of reactive oxygen species (ROS) caused by an imbalance between production of ROS and destruction of ROS by antioxi-dants (Kregel and Zhang, 2007). ROS are produced in the mitochondria in aero-bic cells and cause damage to mitochondrial components and initiate degradative processes including damage to lipids, proteins, and DNA (Floyd and Carney, 1992; Cadenas and Davies, 2000). In particular, the brain is vulnerable to the effects of oxidative stress as it possesses reduced free radical scavenging ability and requires large quantities of oxygen (Floyd and Carney, 1992; Cantuti-Castelvetri et al., 2000). In the body, oxidative stress can lead to lipid peroxidation, a deleterious process that modifies the fluidity and permeability of neuronal membranes leading to an altera-tion of cellular functioning and damaged membrane bound receptors and enzymes (Mariani et al., 2005). Vitamin E exerts antioxidant activity in cell membranes and can inhibit lipid peroxidation (Isaac et al., 2008). Vitamins C and E have been dem-onstrated to decrease oxidative DNA damage markers (Boothby and Doering, 2005). Markers of oxidative stress were reduced, including high sensitivity C reactive pro-tein, LDL oxidation F2-isoprostanes, and monocyte superoxide anion concentrations following supplementation with vitamin E for a period of 2 years (Devaraj et al., 2007). Subsequently, neuroprotective effects of antioxidants against oxidative dam-age in the brain may represent a mechanism that can enhance cognition or delay the rate of cognitive decline in the elderly.

In older adults, low intake and peripheral levels of antioxidants appear to be asso-ciated with greater risk of vascular disease (Jialal and Devaraj, 2003). Oxidative stress has been further implicated in the pathophysiological process of cardiovascu-lar disease and stroke (Mariani et al., 2005). Throughout the progression of cardio-vascular disease, low density lipoprotein (LDL) accumulates in the sub-endothelial space in arteries where it becomes oxidized leading to foam cell formation, endo-thelial dysfunction and injury, and generation of atherosclerotic lesions (Diaz et al., 1997). Cellular antioxidants protect against the cytotoxic effects of oxidized LDL, with vitamin C suggested to affect atherogenesis, with a high intake of this antioxidant protective against cerebrovascular disease (Gale et al., 1996).

Calcium, magnesium, and zinc are essential for optimum neural function-ing, although their relationship to cognitive processes in the elderly has not been researched as extensively as the B vitamins or antioxidants. To summarize the roles of these minerals in the central nervous system, magnesium is essential for enzymes requiring vitamin B1 as a cofactor, and is needed for the synthesis and action of ATP (Bourre, 2006). Neurotransmission is regulated by calcium, and zinc is required as a structural component of many proteins, hormones, hormone receptors, and neuro-peptides (Huskisson et al., 2007). Zinc deficiencies are relatively common and can impair neuropsychological function even at a mild to moderate level of deficiency (Sandstead, 2000). Zinc increases antioxidant activity, and supplementation has been demonstrated to increase the activity of zinc-dependent antioxidant enzymes in healthy elderly subjects (Mariani et al., 2008).

Vitamin D is a steroid hormone that maintains levels of phosphorus, calcium, and bone mineralization. More pertinent to cognition, vitamin D may also play a biologi-cal role in neurocognitive function as receptors are located in regions of the brain important for planning, processing, and forming new memories (Buell and Dawson-Hughes, 2008). Although research into the importance of vitamin D for cognition in normal and pathological aging is only beginning to gain momentum, vitamin D has been suggested to exert protective effects against cardiovascular and cerebrovascu-lar disease, peripheral artery disease inflammation, and to promote neuronal health (Buell and Dawson-Hughes, 2008; Buell et al., 2010). Low sunlight exposure, age-related decreases in cutaneous synthesis, and diets low in vitamin D contribute to the high prevalence of vitamin D inadequacy in the elderly (Holick, 2006).

Before turning to the literature examining the nature of the relationship between vitamin status and cognitive function, this post will cover a brief description of the measures used to assess cognition in the elderly. Most commonly, within large-scale epidemiological studies, cognitive performance has been measured in one of two ways. The first way of assessing cognitive performance has involved investigat-ing individual cognitive domains using standardized cognitive or neuropsychologi-cal measures. The second way has relied on global cognitive measures such as the mini mental state examination (MMSE; Folstein et al., 1975) or the 3MS modi-fied version of the MMSE. These assessments have been used to provide an overall estimate of an individual’s level of functioning, without reference to any specific cognitive domain. Scores derived from these measures can then be correlated with reported dietary intake of vitamins or levels of vitamins taken from the serum or plasma. Other global measures of cognition have comprised a composite perfor-mance score consisting of an averaged or summed score across a range of indi-vidual tests from different cognitive domains. Most commonly global measures have been administered together with, at minimum, a verbal memory and verbal fluency assessment of executive function. As a separate entity from cognitive performance, cognitive decline has also been investigated using these instruments, not withstand-ing that decline can only be measured longitudinally.

Vitamin B12 depletion is common in old age (Wolters et al., 2003). The effects of atrophic gastritis are thought to contribute to insufficient B12 absorption in the elderly, rather than a lack of the micronutrient in the diet per se (Selhub et al., 2000). Observations of psychiatric symptoms associated with B12 deficiency have been documented since 1849 and over the past 50 years low B12 status has been related to memory impairment, personality change, and psychosis (McCaddon, 2006), many of which overlap with dementia (American Psychiatric Association, 2000). More recent evidence has shown that folate and homocysteine are equally impor-tant for mental function. Interestingly relationships between MCI and low serum folate levels (Quadri et al., 2004) and elevated plasma total homocysteine have been reported (Quadri et al., 2005). Based on these findings, the authors of these studies proposed that low folate and elevated homocysteine may predate dementia onset and that maintenance of micronutrient levels through dietary supplementation may aid in dementia prevention. While this is a promising assertion, it is still under debate as to whether deficient B12 and folate may contribute to the neurodegenerative process associated with AD, or whether deficiencies may reflect a consequence of the disease process (Seshadri, 2006).

Numerous studies have also investigated the association between B vitamins, homocysteine and cognitive function in healthy elderly. In the cross-sectional, Maine-Syracuse study of 812 young through to elderly subjects, a relationship was identified between vitamin B6 and multiple cognitive domains encompassing visual–spatial organization, working memory, scanning-tracking, and abstract reasoning (Elias et al., 2006). Outcomes from the Third National Health and Nutrition Examination Survey revealed that individuals aged over 60 years, with elevated homocysteine accompanied by low folate levels, demonstrated poorer story recall than those with normal levels of homocysteine (Morris et al., 2001). Findings from the Singapore Longitudinal Ageing study, which investigated 451 high functioning Chinese elders, revealed that higher levels of folate were associated with better scores on a verbal learning instrument (Feng et al., 2006). A study by Riggs et al. (1996) revealed that middle aged to elderly men from the Boston Veterans Affairs Normative Aging study who had higher concentrations of plasma homocysteine and low concentrations of B12 and folate also displayed poorer spatial copying skills. Follow-up analysis from this study revealed that plasma folate became the strongest predictor of spatial copy-ing ability, independent of homocysteine, 3 years later (Tucker et al., 2005).

Other trials have examined the relationship between B vitamins or homocysteine and cognition using a longitudinal design. Findings from the MacArthur Studies of Successful Aging showed that low folate levels in individuals aged in their 70s were predictive of cognitive decline 7 years later (Kado et al. 2005). It was found by Nurk et al. (2005) that elevated homocysteine at baseline predicted memory deficit after

6 years in those aged 65–67 who took part in the Hordaland Homocysteine study. Although a historical connection between B12 and cognition has been established, several epidemiological studies have failed to find an association between B12 con-centration and cognitive status (Kado et al., 2005; Mooijaart et al., 2005; Nurk et al., 2005; Feng et al., 2006). Alternatively, measurement of methylmalonic acid (MMA), a product of amino acid metabolism, may provide a more useful diagnostic tool for B12 deficiency (Moretti et al., 2004). Results from the Oxford Healthy Aging Project showed that when holotranscobalamin (holoTC), the biologically active fraction of vitamin B12, and MMA were used as measures of vitamin B12 status, each were asso-ciated with a more rapid cognitive decline on the MMSE over 10 years (Clarke et al., 2007). When considered together, outcomes from these longitudinal studies point to a more causal role of B12 and folate deficiency in cognitive decline.

Antioxidant vitamin status in the blood has been associated with cognitive func-tion in the elderly. The Third National Health and Nutrition Examination Survey investigated blood vitamin levels taken from a multiethnic sample of 4809 elderly residents of the United States (Perkins et al., 1999). Findings from this trial revealed that higher serum levels of vitamin E were associated with better memory recall of a three sentence story. Interestingly, no correlations between vitamin A, beta-carotene, selenium, or vitamin C with memory performance were identified. By contrast, in a sample of Swiss elderly aged 64–95 years, both past and current levels of ascorbic acid and beta-carotene were associated with free recall, recognition, and vocabu-lary performance (Perrig et al., 1997). In the Cognitive Change in Women study, a comprehensive neuropsychological test battery was utilized to assess the domains of memory, executive function, language, attention, and visual function in 526 women aged 60 or above with cognitive impairment (Dunn et al., 2007). Baseline results from this study revealed that low serum alpha-tocopherol status was cross section-ally associated with increased odds ratio of memory and mixed cognitive impair-ments. In contrast, previous vitamin E supplement intake was not associated with any type of cognitive impairment.

Intake of antioxidant vitamins from the diet has also been linked to cognitive function in older adults. The Chicago Health and Aging Project investigated the dietary habits of 2889 community residents, aged 65–102 using a food frequency questionnaire (Morris et al., 2002). The results of this study revealed that higher intake of vitamin E from the diet or from supplements was associated with a slower rate of cognitive decline over 3 years as measured by a combined cognitive score derived from tests of immediate and delayed story recall, a measure of perceptual speed and the MMSE. In contrast, findings from the Rotterdam study revealed that intake of beta-carotene, and not vitamin C or E, was associated with performance on the MMSE in elderly aged 55–95 years (Warsama Jama et al., 1996).

Use of vitamin E and C supplements by elderly men has been linked to better cognitive function at follow up 3–5 years later (Masaki et al., 2000). Results from the Cache County study revealed that participants aged 65 years or older who had a lower a intake of vitamin C, vitamin E, and carotene also had a greater acceleration of the rate of cognitive decline on the 3MS over 7 years compared to those with a higher use. The conclusions from this study were that higher antioxidant dietary vita-min C, vitamin E, and carotene may delay cognitive decline in the elderly (Wengreen et al., 2007). Separate analyses from the same trial demonstrated that those with the APOE e4 allele using vitamin C, E, or multivitamin supplements in combination with nonsteroidal anti-inflammatory drugs showed less cognitive decline over an 8 year period than those with the E4 allele who were taking antioxidant supplements (Fotuhi et al., 2008). It was suggested from these results that those at a higher genetic risk of developing AD may benefit from use of anti-inflammatories and antioxidant supplementation.

Thus, the use of combined vitamin C and E supplements appears to influence cognition in the elderly. Long-term consumption (>10 years) of vitamin C and E supplements was associated with better performance on cognitive tests includ-ing verbal fluency, digit span backward, and a telephone administered version of the MMSE, in a large sample of community-dwelling elderly women who participated in the Nurse’s Health Study (Grodstein, 2003). Women taking both supplements displayed equivalent cognitive function to individuals 2 years younger. Similarly, elderly subjects from the Canadian Study of Health and Ageing using combined vitamin C and E and supplements were less likely to experience significant cognitive decline on the 3MS after 5 years (Maxwell et al., 2005). Collectively, results from these studies suggest that chronic use of antioxidant supplements may help slow the rate of cognitive decline in the elderly.

Few studies have investigated the relationship between zinc and cognition in older adults. In older adults recruited from European countries including Italy, Greece, Germany, France, and Poland, plasma zinc status was correlated with global cogni-tive functioning as measured by the MMSE (Marecllini et al., 2006). The observation that inhabitants from countries with diets rich in zinc exhibited superior cognitive function suggests that zinc supplementation may have beneficial effects on cognition in elderly who are deficient.

In a large sample of middle aged to elderly men free from dementia, levels of vita-min D have been associated with digit symbol substitution performance (Lee et al., 2009). Findings from the Nutrition and Memory in Elders study revealed a posi-tive correlation between levels of vitamin D and measures of executive function and attention processing speed in over 1000 elderly subjects (Buell et al., 2009). As this relationship remained robust after adjustment for a number of factors including homocysteine, apoE4 allele, plasma B vitamins, and multivitamin use, it is possible that vitamin D may also represent an important predictor of cognitive function in the elderly. In a sample of elderly French community-dwelling women, dietary intake of vitamin D, as estimated from a food frequency questionnaire, has been associated with performance on a global cognitive measure (Annweiler et al., 2010), indicating intake of vitamin D may also be important for cognition.

In summary, the findings from population studies have demonstrated that in the elderly cognitive function is related to dietary intake, long-term supplement use, and circulating levels of individual vitamins. The results from these studies indicate that dietary intake of B vitamins, antioxidants, and selected minerals may comprise important predictors of cognitive function, particularly as individuals enter the later stages of the life span.

When considering the neurocognitive effects of dietary factors, it may be useful to examine the association between vitamin status and brain structural parameters. Most commonly structural MRI has been used for this purpose, with more com-prehensive examinations undertaken at post mortem. In a fascinating study, blood nutrient levels were correlated with brain pathology in elderly nuns who lived in the same convent, ate at the same kitchen, and consequently had comparable environ-mental factors and overall lifestyle (Snowdon et al., 2000). Blood was collected and analyzed for nutrients, lipids, and nutrient markers. Following the death of 30 elderly nuns, a neuropathologist examined the brains for signs of atrophy, AD lesions (neurofibrillary tangles, senile plaques, and neutritic plaques), and atherosclerosis in the major arteries at the base of the brain. The results demonstrated that serum folate levels were negatively related to atrophy of the neocortex, particularly in those with a significant number of AD lesions in the neocortex. It was proposed that the association between low folate levels and cortical atrophy may not be entirely due to the deleterious effects of vascular disease, as folate was negatively correlated with subgroups of participants with minimal signs of vascular neuropathology such as arteriosclerosis and brain infarcts. Similarly, De Lau et al. (2009) have posited that neuropathology related to low B12 levels may arise from other nonvascular factors. The larger-scale, population-based, Rotterdam Scan study provided evidence that poorer vitamin B12 status in the normal range was significantly associated with greater severity of white-matter lesions, in particular periventricular white-matter lesions in healthy elderly. However, as B12 levels were not related to cerebral infarcts, these researchers hypothesized that the association between B12 levels and white-matter lesions may due to effects on myelin integrity in the brain, rather than through vascular mechanisms alone.

Vitamin B12 levels in healthy community-dwelling elderly have also been associ-ated with brain volume loss over a 5 year period (Vogiatzoglou et al., 2008). The results of this study revealed that the decrease in brain volume was greater among those with lower vitamin B12 and holoTC levels. Specifically, those in the bottom ter-tile for B12 (<308 pmol/L) at baseline experienced the greatest rate of brain-volume loss. Based on these findings, the authors concluded that plasma vitamin B12 status may provide an early marker of brain atrophy and consequently may represent a potentially modifiable risk factor for cognitive decline in the elderly. Findings from a recent randomized trial indicated that 2 years B vitamin supplementation was capable of slowing the rate of brain atrophy (Smith et al., 2010). In this study elderly with MCI were assigned to a treatment of combined vitamins B6, B12, and folate or a placebo. Volumetric MRI scans revealed that the vitamin treatment reduced the rate of atrophy by approximately 30%, and biochemical measures indicated this was accompanied by a reduction in homocysteine. These findings indicate that reducing homocysteine via B vitamin supplementation may exert protective effects on brain structural parameters.

In summary, evidence from intervention studies suggests that supplementation with vitamin B12 or folate may provide only limited improvements to processing speed and memory in the elderly.

The effects of selected antioxidant supplementation on cognitive measures in elderly adults have been investigated in several large-scale placebo-controlled stud-ies. Within some of these studies the primary aim was not to measure cognitive performance changes with supplementation and in some studies no cognitive data was available prior to supplementation. Participants in the Age-Related Eye Disease study were randomly assigned to receive daily antioxidants (vitamin C, 500 mg; vitamin E, 400 IU; beta-carotene, 15 mg), zinc and copper (zinc, 80 mg; cupric oxide, 2 mg), antioxidants plus zinc and copper, or placebo. After approximately 7 years, there was no difference between the treatment groups on any of the cog-nitive tests (Yaffe et al., 2004). A cognitive testing component including assess-ment of general cognition, verbal memory, and category fluency was added to the Physician’s Health study, where elderly men supplemented their diet with 50 mg beta-carotene or a placebo on alternate days (Grodstein et al., 2007). There was no impact of treatment with beta-carotene for 3 years or less on cognitive performance, whereas treatment duration of at least 15 years provided significant benefits for ver-bal memory, cognitive status, and a composite of these measures. These findings indicate that long-term interventions, implemented at early stages of brain aging may provide cognitive benefits. Using the same outcome measures, subjects in the Women’s Health study received vitamin E supplementation (600 IU) on alternate days or placebo (Kang et al., 2006). The results of this study showed no effect of the treatment on a global composite score after 9 12 years and it was concluded that long-term use of vitamin E supplements did not provide cognitive benefits among generally healthy older women.

In brief, there appears to be some advantage to long-term beta-carotene supple-mentation in the elderly; however, the mainly telephone-administered cognitive tests used in these trials may not be sensitive to the neurocognitive effects of antioxidant supplementation.

Fewer studies have utilized RCT methodology to explicitly evaluate the cogni-tive effects of trace minerals such as zinc. In a trial of healthy adults aged 55–87 years, subjects received either 15 or 30 mg of zinc per day (Maylor et al., 2006). Treatment effects were assessed at 3 and 6 months using measures of visual mem-ory, working memory, attention, and reaction time from the Cambridge Automated Neuropsychological Test Battery. At the 3 month testing period, spatial working memory was improved at both dosages; however, a detrimental effect of the lower dose was observed on an attentional measure, indicating that the beneficial treatment effects of zinc may be limited to specific cognitive domains.

Multivitamins are a combination formula of the B vitamins and antioxidant vita-mins described previously in this post, in addition to minerals such as calcium, magnesium, zinc, and iron. Dietary supplementation with multivitamins is rela-tively common in the elderly, with data from the U.S. National Health and Nutrition Examination Survey showing that 63% of individuals aged 60 years or above had used a dietary supplement in the past month and 40% had specifically used a mul-tivitamin supplement (Radimer et al., 2004). In adults above the age of 75 years, this figure appears to be somewhat higher, with another study reporting that over 59% of this demographic supplemented their diet with multivitamins (Nahin et al., 2006). While these supplements are widely used in the older population, there are not a large number of studies that have used randomized controlled designs to investigate the effects of multivitamin supplementation on cognitive function in the elderly.

The findings from several RCTs investigating the effects of multivitamins in the elderly have not uncovered cognitive benefits. For instance, Cockle et al. (2000) did not identify improvements to reaction time, short-term memory, or recogni-tion memory, in seniors following up to 24 weeks of multivitamin supplementation. Similarly, there were no improvements on measures of pattern recognition, IQ or symbol search after 24 weeks of multivitamin supplementation in women aged 60 and above (Wolters et al., 2005). In a 12 month trial conducted by McNeill et al. (2007), there were no alterations in verbal fluency or digit span performance in indi-viduals aged 65 years or older. Once the age groups were separated into younger and older subjects, there was a small beneficial effect of the multivitamin treatment on verbal fluency for subjects over 75 years of age.

More recently, positive results have been obtained from a trial that investigated a combined multivitamin and herbal supplement. Specifically, benefits to verbal flu-ency and recall were identified in elderly aged 50–75 years after 4 months treatment with a complex antioxidant blend consisting of 34 vitamins, minerals, amino acids, lipids, and herbal extracts, all with antioxidant properties (Summers et al., 2010). Intake of herbs contributes significantly to intake of plant antioxidants (Dragland et al., 2003) and may contribute to synergistic benefits when in the presence of other antioxidant vitamins (Cantuti-Castelvetri et al., 2000). Subsequently, it is conceiv-able that combined multivitamin and herbal supplements may exert greater effects on cognition than formulas consisting of vitamins alone.

Several other smaller trials have examined the cognitive effects of multivitamin supplementation. In a study of only 20 subjects, there were no treatment related improvements to the MMSE following 12 months supplementation with a multivita-min or placebo in elderly with cognitive impairments (Baker et al., 1999). A recently published 3 month study of 47 young adults compared the cognitive effects of supplementation with a multivitamin, a poly-herbal formula, and a placebo. Despite small, unequal treatment group sizes, there appeared to be some memory enhancements associated with the multivitamin and poly-herbal treatments (Shah and Goyal, 2010). Replication of this study with larger, equal group sizes may be required to confirm these results. In a trial of young to elderly subjects, improved performance on a digit memory task and the trails making test have been documented after time periods of 2 weeks and 3 months treatment with a dietary supplement consisting of folic acid, B12, vitamin E, S-adenosylmethionine, N-acetylcysteine, and acetyl-L-carnitine (Chan et al., 2009). Interestingly, in an open label extension, memory benefits were diminished when the treatment was withdrawn for 3 months and reemerged when treatment was reinstated for a further 3 months. As only a subset of elderly subjects were shown to respond to the treatment, and only a fraction of the initial sample remained at the 12 month conclusion of this study, replication of this study may be necessary.

Haskell et al. (2008) have argued that the outcome measures previously employed in randomized controlled trials to investigate the effects of multivitamins on cogni-tive function in children have relied on measures of IQ rather than true cognitive performance measures. These authors proposed that measures of attention and non-verbal IQ would be more sensitive to the effects of multivitamins, due to the potential modulating effect on neurotransmitters involved in these processes. However, in the elderly, the onset of other age-related neural and cognitive changes that are not appli-cable to children may complicate this interpretation.

In the elderly, several factors may influence the cognitive domains that respond to dietary supplementation with vitamins or minerals. First, it can be postulated that the domains most vulnerable to the deleterious effects of aging will respond best to intervention. For example, chronic folic acid supplementation has been shown to improve performance on tests that measure information processing speed and working memory (Durga et al., 2007), domains that are known to decline with age (Babcock and Salthouse, 1990; Salthouse, 1990, 1996). Recent studies have also shown that supplementation with powerful antioxidants known as flavonoids can enhance mental functions on working memory–related indices (Pipingas et al., 2008; Ryan et al., 2008).

Second, it could be hypothesized that the cognitive processes mediated by corti-cal structures most vulnerable to vitamin deficiency in the elderly may show the greatest improvement with supplementation. Gray-matter volume in the parietal regions has been correlated with vitamin B12 intake (Erickson et al., 2008), and ele-vated homocysteine has been shown to selectively compromise the hippocampus (Williams et al., 2002; Den Heijer et al., 2003). Subsequently the cognitive domains supported by these regions, such as memory, may prove to be more vulnerable to vitamin depletion than others and may even respond better to dietary supplementa-tion. Randomized controlled trials integrating both structural and functional brain imaging techniques may have the potential to address this research question.

Within the elderly population, there is a need to identify subgroups of people who would experience the most cognitive benefits from dietary supplementation with vitamins. For example, the results of a study conducted by Bunce et al. (2005) showed that possession of the APOE e4 allele, in combination with low B vitamin levels, impeded retrieval memory performance in demanding face recognition con-ditions. These findings were independent of dementia up to 6 years later, indicating that these individuals were not in a preclinical phase of AD. Based on the results of this study, the authors proposed that in healthy elderly, genetic risk factors for AD such as the APOE e4 allele may predispose some brain structures and processes to be more vulnerable to the deleterious effects of low B12 levels. In addi-tion to genetic factors, elderly experiencing ongoing illness, stress, poor appetite, and vision impairments are also at risk of vitamin and micronutrient deficiencies (Payette et al., 1995).