| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
a Commonwealth Scientific and Industrial Research Organisation, Division of Health Sciences and Nutrition, Adelaide, South Australia
Janet Bryan, Commonwealth Scientific and Industrial Research Organisation, Division of Health Sciences and Nutrition, P.O. Box 10041, Adelaide BC, South Australia, 5000 E-mail: janet.bryan{at}dhn.csiro.au.
Decision Editor: Margie E. Lachman, PhD
 |
Abstract |
|---|
 TOP Abstract B Vitamins and Brain...General SummaryReferences |
|---|
COGNITIVE performance in older age can be influenced by a number of factors, and the association between nutrition and cognition has become a topic of increasing scientific and public interest (Bellisle et al. 1998
). The impetus for much of this research is the increasingly longer life span and the need to maintain functional well-being, particularly cognitive function, a major determinant of the quality of life in older age (Rosenberg and Miller 1992). The link between nutrition and cognitive performance is based on the knowledge that the central nervous system (CNS) depends heavily on a constant supply of glucose and almost all of the essential nutrients for effective functioning (Selhub, Bagley, Miller, and Rosenberg 2000). In particular, recent research has focused on the effect of the B vitamins, especially folate, B12, and B6, on the brain and on cognition (Bottiglieri, Crellin, and Reynolds 1995), especially in older adults. This work stems from the assumption that nutrition may be an important modifiable lifestyle factor in age-related cognitive decline and the growing interest in the development of nutritional supplements as therapeutic agents aimed at enhancing or maintaining cognitive function or halting cognitive decline (Riedel and Jorissen 1998).
Although most of the evidence for the link between B vitamins and cognitive performance is based on studies involving participants with clinical vitamin deficiencies, the effect of these vitamins might also be important for a broader population because milder, subclinical vitamin deficiencies are not uncommon, especially among older adults (Rosenberg and Miller 1992; Selhub et al. 2000). In particular, a mild to moderate folate deficiency is thought to be relatively common in the general population (Mazza 1998), although the prevalence of clinical and subclinical levels of folate and B12 deficiency has been found to be higher in the elderly (Mazza 1998; Ortega et al. 1996; Parnetti, Bottiglieri, and Lowenthal 1997; Sauberlich 1991). For example, Joosten and colleagues 1993 measured serum concentrations of folate, B12, and B6 and four metabolites of these vitamins in younger, healthy older and hospitalized older adults. They found evidence of tissue deficiencies of all three B vitamins, demonstrated by increased metabolite concentrations, in the healthy older adults and particularly in the hospitalized older adults. Stabler, Lindenbaum, and Allen 1997 estimate that low serum B12 concentrations might be evident in 5% to 15% of the elderly population. They suggest there may be multiple causes of B12 deficiency in older adults: pernicious anemia, atrophic gastritis causing B12 malabsorption, or previous gastric or intestinal surgery. These findings support the indications suggesting that older adults in particular might be at risk for subclinical deficiencies in these vitamins, and evidence is continuing to mount indicating that higher intakes and serum concentrations of certain vitamins, particularly folate and B12, might influence cognitive performance (Lindeman et al. 2000).
|
B Vitamins and Brain Function |
|---|
|
TOPAbstractB Vitamins and Brain...General SummaryReferences |
|---|
; La Rue et al. 1997; Rosenberg and Miller 1992; Selhub et al. 2000 for reviews). Folate (folic acid or folacin), B12 (cobalamin) and B6 (pyridoxine) are water-soluble B vitamins that play an important role in a variety of critical brain-metabolic pathways (Alpert and Fava 1997; Bottiglieri 1996). Research to date indicates two interlinked neurochemical mechanisms by which folate, with Vitamins B12 or B6 as catalyzing cofactors, exerts an influence on cognitive performance and mood via its role in methylation in the CNS (Alpert and Fava 1997; Bottiglieri et al. 1995; Hankey and Eikelboom 1999). It could be hypothesized that the B vitamins effect CNS function in two interrelated ways: a direct, and possibly acute influence via hypomethylation, and a more indirect, longer term influence on homocysteine levels resulting in structural changes in the brain (Clark et al. 1998).
The hypomethylation hypothesis proposes that folate, B12, and B6 have direct effects on the functioning of the brain through their role in the one-carbon cycle, essential to many transmethylation reactions within the CNS (Bottiglieri et al. 1995; Rosenberg and Miller 1992). Folate is converted to 5-methyltetrahydrofolate (5-MTHF) and then combines with homocysteine, catalyzed by vitamin B12-dependent methionine synthetase, to produce L-methionine, a precursor to S-adenosylmethionine (SAMe), a methyl donor in a variety of reactions in the CNS (Alpert and Fava 1997; Bailey and Gregory 1999; see Fig. 1). Low levels of folate lead to hypomethylation, inhibiting the synthesis of methionine and SAMe, which in turn inhibits many methylation reactions throughout the CNS involving proteins, membrane phospholipids, DNA, the metabolism of neurotransmitters such as the monoamines (e.g., dopamine, norepinephrine, serotonin), and melatonin, all of which are crucial to neurological and psychological status (Alpert and Fava 1997; Bottiglieri 1996; Bottiglieri et al. 1995; Fenech, Aitken, and Rinaldi 1998). Thus, it is hypothesised that hypomethylation leads to neuropathologies, cognitive impairments, and affective or mood disturbances (Mischoulon 1996; Rosenberg and Miller 1992).
|
; Carney, Toone, and Reynolds 1987; Reynolds, Carney, and Toone 1984: Reynolds and Stramentinoli 1983; Young 1993). Low levels of folate lead to low levels of SAMe, which influence the metabolism of the monoamines, in particular serotonin (5-HT), a neurotransmitter thought to be implicated in the etiology of depressive illness and, perhaps, cognitive function (Alpert and Fava 1997; Bottiglieri 1996; Mischoulon 1996). Further support comes from the comparable effectiveness of other folate derivatives, such as 5-MTHF, as antidepressants (Godfrey et al. 1990). Moreover, this research provides a link between the hypomethylation hypothesis and the monoamine hypothesis of affective disorder (Crellin, Bottiglieri, and Reynolds 1993; Reynolds and Stramentinoli 1983) and, perhaps, cognitive function (Rosenberg and Miller 1992). Indeed, observations from these and other investigations have led to calls for the use of folate or its derivatives in the treatment of depressive disorders, either as an adjunct to standard antidepressant pharmacotherapy or even as a single agent in individuals with low or subclinical levels of folate (Alpert and Fava 1997; Mischoulon 1996; Young and Ghadirian 1989). In addition, Passeri and colleagues 1993 found that 5-MTHF was not only as effective as psychotropic medication in relieving depressive symptoms in patients with mild to moderate dementia, but they also found that those in the 5-MTHF group demonstrated significant improvement in a test of short-term memory compared with those in the psychotropic medication only group. Most recently, Fontanari, Di Plama, Giorgetti, Violante, and Voltolina 1994 found that administration of SAMe not only confirmed its efficacy as an antidepressant in their work with older adults, but, in a sample of older, nondepressed inpatients, also resulted in significant improvements in Mini-Mental State Examination (MMSE; Folstein, Folstein, and McHugh 1975) and in Sandoz Clinical Assessment Geriatric Scale (Venn 1983) scores. Taken together these results provide preliminary evidence for a link between folate, mood, cognitive performance, and the hypomethylation hypothesis.
The homocysteine hypothesis proposes an indirect and longer term effect of the B vitamins on the functioning of the brain. That is, neurocognitive changes may be mediated by cerebrovascular changes, linked to elevated plasma homocysteine concentrations, which are largely attributable to low levels of folate, B12, and B6 (Hankey and Eikelboom 1999; Lindeman et al. 2000; Selhub et al. 1995; Ueland and Refsum 1989). Studies have demonstrated that high levels of plasma homocysteine are associated with increased risk of vascular disease (Pancharuniti et al. 1994; Selhub et al. 1995; Ueland and Refsum 1989). Therefore, it is hypothesized that the B vitamins may function to preserve and protect the integrity of the CNS via its role in the prevention of vascular disease, which, in turn, is crucial to cognitive function (Bottiglieri et al., 1995; Clark et al., 1998; Hankey & Eikelboom, 1999; Homocysteine Lowering Trialists Collaboration [HLTC], 1998; Weir and Molloy 2000).
Folate is involved in homocysteine metabolism through methionine synthetase activity via a remethylation pathway requiring Vitamin B12 as an essential cofactor, or a trans-sulphuration pathway requiring B6 as an essential cofactor (see Fig. 1). Deficiencies of folate in particular, and to a lesser extent of B12, and/or B6, can lead to failure to methylate homocysteine effectively, leading to elevated levels in blood and urine (Bottiglieri 1996). Indeed, studies have demonstrated the reliability of using plasma homocysteine as a biochemical marker for folate, B12, and B6 deficiency (Ubbink, Vermaak, van der Merwe, and Becker 1993; Ueland and Refsum 1989). It has been argued that high plasma homocysteine has a toxic effect on vascular tissue, placing individuals at risk of cerebrovascular and other vascular diseases and this has been supported by several studies (Clark et al. 1998; Hankey and Eikelboom 1999, Selhub, Jacques, Wilson, Rush, and Rosenberg 1993; Selhub et al. 1995; Snowdon, Tully, Smith, Riley, and Markesbery 2000; Ubbink et al. 1993; Ueland and Refsum 1989). Parnetti and colleagues 1997 also suggest that increased homocysteine may lead to an excessive production of excitotoxic sulfur amino acids, homocysteic acid, and cystein sulphinic acid, that act as endogenous agonists of N-methyl-D-aspartate receptors. This results in neuronal injury and death through excessive glutamate receptor activation. Recent studies have shown that folate supplementation produced a 25% reduction, and B12 supplementation an additional 7% reduction, in plasma homocysteine (Brouwer et al. 1999; HLTC 1998). The HLTC (1998) argued that these findings have public health implications, and advocated supplementation with folate and B12 as a safe, cheap, and effective method of reducing blood homocysteine concentrations as a strategy to reduce the risk of vascular events.
These hypotheses provide plausible mechanisms through which B vitamins can affect the brain and thus cognitive performance, although the mechanisms have not been tested directly. However, it could be argued that studies examining direct or acute effects of the B vitamins are consistent with the hypomethylation hypothesis and that those examining indirect or long term effects are consistent with the homocysteine hypothesis. So far, studies investigating links between folate, B12, and B6 fall into two broad categories: those examining the association between a range of nutrients, including the B vitamins, and measures of cognitive function, and those examining the combined and separate effects of folate and B12, or of folate, B12, and B6 alone. In addition, the extant literature has investigated the links between the B vitamins and cognitive impairment or diffuse effects on cognition in general. Very few studies have provided a systematic testing of the effects of B vitamins on specific cognitive constructs. Furthermore, most studies investigating the links between the B vitamins and cognitive performance have examined effects for older adults (see Table 1 ). Others have investigated effects among individuals with Alzheimer's disease and other forms of dementia and cognitive impairment (see Table 2 ). Very few studies have examined effects among healthy adults with ages from across the life span. Studies differ in the vitamins or combination of vitamins investigated and in their design, with most using a retrospective, cross-sectional design and others using longitudinal and experimental designs. Thus, the studies are reviewed according to the design used and the nutrients examined.
|
|
. Dietary intakes and nutritional status of several nutrients, including the B vitamins, was investigated among 260 community-dwelling older adults aged between 60 and 94 years. Cognitive function was assessed by the Russel revision of the Wechsler Memory Test (Russel 1975) and the Halstead-Reitan Categories Test (Reitan and Davidson 1974), a measure of abstract thinking and problem solving. Correlational analyses indicated that there were no significant relationships between nutrient intakes and cognitive function, and a significant, though weak, relationship between plasma levels of vitamins B12 and C and the memory test. However, when the mean scores on cognitive tests for participants in the bottom 5% to 10% of intake and plasma levels of individual nutrients were compared with the mean scores of the top 90% of the sample, a stronger association between poor nutritional status, in both intake and plasma levels, and poor cognitive performance was found. With regard to folate, those in the bottom 5% to 10% of dietary intake performed significantly more poorly on both cognitive measures than did those in the top 90%, and those in the bottom 5% to 10% of plasma folate performed significantly more poorly on the abstract thinking and problem-solving test. These differences remained significant after controlling for age, gender, education, and income. The authors concluded that although the association between low intake and blood levels of certain nutrients and cognitive performance was not indicative of a causal relationship, it did indicate that changes in cognitive performance in older adults might be secondary to nutrition rather than due solely to inevitable age-related decline.
These findings were supported by Ortega and colleagues 1996, Ortega and colleagues 1997, who examined the relationship between the B vitamins and cognitive status among community-dwelling healthy older adults. Weighed food records and serum and erythorocyte levels were used to assess dietary intake and nutrient status, respectively, and cognitive status was assessed using the MMSE (Folstein et al. 1975) and Pfeiffer's Mental Status Questionnaire (Pfeiffer 1975). They found that folate, but not B12 or B6, deficiency was common and that those with "adequate" MMSE scores had significantly higher serum and erythrocyte levels of folate concentrations than those with less adequate scores, independent of age, gender, education, and socioeconomic status. These authors advocated screening for folate deficiency in older age individuals to correct it and enhance performance and quality of life.
Several researchers have examined the relationship between folate and its catalyzing cofactor, Vitamin B12, to test for their separate and combined effects on cognitive performance. Bell and colleagues 1990 conducted a retrospective chart review study to examine the relationship between cognitive measures and serum folate and B12 status in 102 psychiatric inpatients whose ages ranged from 60 to 100 years. Even though participants' records indicated satisfactory nutritional status, correlational analyses indicated that those with below median values for both folate and B12 had significantly lower scores on the MMSE (Folstein et al. 1975) than those who were higher in folate and/or B12. The authors concluded that lower levels of folate and B12, even within the normal range, may interact to produce CNS metabolic abnormalities affecting cognitive function.
In a more recent study, Wahlin, Hill, Winblad, and Backman 1996 examined the separate and combined effects of folate and B12 on episodic memory functioning in a group of 250 community-dwelling healthy older adults aged 75 to 96 years. Consistent with Goodwin and colleagues 1983, they argued that there may be critical thresholds above or below which biological variables may exert a significant effect. Therefore, they examined the effects of low serum levels of folate, B12, or both, using a control group matched for age and level of education. Episodic-memory functioning was measured using immediate free recall and incidental recognition tests of common concrete nouns. Results indicated that those with low levels of folate alone or low levels of both folate and B12 performed significantly worse than did those with normal levels of folate and B12. Low levels of B12 alone had no impact on memory performance. The effects remained significant after controlling for MMSE scores, indicating that performance decrements were not accounted for solely by global age-related cognitive decline as measured by the MMSE. The authors suggested that because low levels of folate alone, rather than B12 alone, are associated with performance deficits in older age, it may be more critical to alter folate levels to enhance performance.
In an extension of the Wahlin and colleagues 1996 study, Hassing, Wahlin, Winblad, and Backman 1999 used the same statistical approach to investigate the separate and combined effects of serum folate and B12 on episodic memory functioning in a population-based sample of healthy very old adults ranging in age from 90 to 101. Episodic memory was assessed using face recognition, word recall and recognition, and object recall. Consistent with Wahlin and colleagues results indicated significant effects for folate alone on the episodic-memory tasks but none for B12 alone. However, in contrast to Wahlin and colleagues there were no combined effects for folate and B12, and the authors concluded that this provides further evidence that folate may be more critical than B12 to memory performance in old age.
The most recent study, by Lindeman and colleagues 2000, set out to examine the associations between serum folate, B12, and Vitamin C concentrations and measures of cognitive and affective functions in a group of 833 Hispanic and non-Hispanic Whites aged 65 years and older. Cognitive function was assessed using the MMSE (Folstein et al. 1975), Wechsler Adult Intelligence Scale–Revised (WAIS-R) Digit Span Forwards (Wechsler 1981), Fuld Object Memory Evaluation (Fuld 1981), Clock Drawing (Goodglass and Kaplan 1983), and two Color Trail Making Tests (D'Elia, Satz, Uchiyama, and White 1996). Using multivariate logistic regression models to adjust for gender, age, ethnicity, and years of education, they found significant associations between low and low-normal serum concentrations of folate and lower cognitive performance scores in all measures except the Clock Face Test and one of the two Color Trail Making Tests. Moreover, adjusting for the presence of depression and using lower cutoff points to define a deficiency state for folate produced the same results. No significant associations were found between low serum concentrations of B12 and C and cognitive performance. These results lend support to the argument that folate rather than B12 may be more critical to cognitive function.
Riggs, Spiro, Tucker, and Rush 1996 included measures of homocysteine as well as plasma levels of folate, B12, and B6 as predictors of cognitive performance among 70 men from the Boston Veterans' Affairs Normative Aging Study, aged 54 to 81 years. They argued that because homocysteine is considered a marker for folate, B12, and B6 levels, the measurement of homocysteine is necessary to identify individuals with low concentrations of those B vitamins who are actually deficient, especially in aging populations. In addition, a high homocysteine level may be implicated in cognitive impairment because of its association with vascular pathology, as outlined earlier. Cognitive performance was assessed using a battery of standardized tests measuring verbal ability, perceptual speed and attention, memory, spatial copying, and spatial reasoning. The authors found that, consistent with expectations, homocysteine was negatively associated with plasma concentrations of folate as well as B12, and there were significant associations between poor spatial copying performance and high plasma homocysteine, low plasma folate, and low plasma B12, whereas low plasma B6 was significantly associated with poor backward digit-span performance. Furthermore, after controlling for homocysteine, folate and B12 no longer significantly predicted spatial copying performance, but homocysteine continued to make a unique contribution. The authors proposed that, although homocysteine was not a significant predictor of vascular disease in this study, high plasma levels, largely attributable to low levels of folate, may be associated with subclinical metabolic or structural changes in the brain that in turn affect cognitive function.
The results from these studies examining cross-sectional associations between B vitamin intake and status and cognitive performance suggest low folate intake and/or status emerges as the most reliable associate of cognitive performance, either alone or in combination with low B12. Many aspects of cognition appear to be related to B vitamin status, especially memory performance and measures of abstract reasoning. In addition, the study by Riggs and colleagues 1996 suggests that the relationship between the B vitamins and cognition may be mediated by homocysteine levels because homocysteine uniquely predicted cognitive performance after controlling for B vitamin status.
Longitudinal studies.
Very few studies have investigated links between the B vitamins and cognitive performance longitudinally. In a 6-year follow-up of a small, healthy subsample of the original Goodwin and colleagues 1983 sample, La Rue and colleagues 1997 assessed whether retest cognitive performance of 137 participants, aged between 66 and 90 years, was related to concurrent and/or past nutritional status. Cognition was assessed using standardized, age-sensitive tests of abstract reasoning, verbal and nonverbal memory, and visuospatial skills. Concurrent plasma, erythrocyte, and dietary intake levels of folate status were correlated significantly only with abstract reasoning. Past serum transferrin status and past intakes of vitamins B12, B6, A, and E were correlated with current cognitive performance. Past folate status did not correlate with current cognitive performance. It is worth noting that the abstract-reasoning task, a measure of fluid ability most often associated with measures of nutrition in this study, was the most difficult of the cognitive performance measures and the one in which participants' scores were lowest in relation to the other tasks when compared with normative data. Therefore it could be the case that this task was more sensitive to nutrition effects, which are likely to be subtle (Bellisle et al. 1998; Bryan 1998). Consistent with Goodwin and colleagues 1983, this study found modest associations between nutritional measures, including the B vitamins, and cognitive performance.
Ebly, Schaefer, Campbell, and Hogan 1998 conducted cross-sectional and longitudinal analyses among a large population-based sample of individuals, aged 65 years and over, to examine relationships between low serum folate levels and the prevalence of cognitive problems, stroke, vascular disease, and short-term mortality. Cognitive performance was assessed using measures of memory, abstract thinking, judgment, constructional praxis, language and object recognition, and a modified version of the MMSE. Results showed that individuals in the lowest serum folate quartile were more likely to have cognitive loss, to be demented, institutionalized, and depressed, and to have a higher mortality rate at a 2-year follow-up. They also scored lower on the MMSE and on short-term memory. However, the authors advised caution in generalizing their findings, as no central laboratory was used for blood assays and as ranges of normal folate varied between the many centers used for their sample. In addition, the majority of individuals for whom serum folate levels were available were more likely to be institutionalized and demented and to have lower global cognition scores, constituting a major bias in their sample.
In summary, results from longitudinal studies allow us to examine possible long-term effects of B vitamin dietary intake and status on cognition at a later date or the impact of B vitamins on cognitive change. The results of these studies suggest that prior intake of B vitamins is a predictor of cognitive status at a later date. The findings of Ebly and colleagues 1998 suggest that low folate status may be a predictor of cognitive change among older adults.
Experimental studies.
Only very few studies have experimentally manipulated B vitamin intake and assessed its affects on cognitive performance, and fewer still have used placebo-controlled designs. Dror, Stern, Nemesh, Hart, and Grinblat 1996 gave micronutrient supplements including folate (0.1 mg), B12 (0.003 mg), and B6 (1.8 mg) to 12 individuals aged between 65 and 87 years for 42 days. They found that supplementation had no effect on MMSE scores, but scores on the Geriatric Depression Scale (Rubenstein 1990) improved after supplementation. The results of this study cannot be readily interpreted because there was no control group and the small sample size may have compromised power to detect significant effects.
To date, it appears that only three studies have used a placebo-controlled design to investigate effects of B vitamin supplementation on cognitive performance among older adults. Tolonen and colleagues 1988 investigated B6 status among older Finnish and Dutch adults aged between 64 and 96 years and younger Dutch adults aged from 22 to 55 years. In addition, they gave daily oral doses of 2 mg of B6 for 1 year to 20 Finnish older adults, while 24 Finnish older adults, matched for age, gender, and general health, received a placebo. Biochemical results clearly indicated that both Finnish and Dutch older adults had lower B6 status than did the younger adults. B6 supplementation improved B6 status among those receiving supplementation. Clock drawing performance was improved by supplementation relative to controls, but there were no significant effects of supplementation on memory or Digit Span (Wechsler 1974) performance.
Deijen, van der Beek, Orlebeke, and van den Berg 1992 also investigated the effects of B6 supplementation on cognitive performance and mood among 76 older men aged between 70 and 79 years. They gave the men 20 mg of B6 or placebo daily for 3 months. Significant positive effects of B6 supplementation, compared with placebo, were found on measures of the amount of information retained in long-term memory, but there were no effects for iconic or short-term memory. The authors concluded that B6 supplementation might have a modest but significant effect in improving the storage of information, thereby reducing age-related memory loss.
Fioravanti and colleagues 1997 used a double-blind, placebo-controlled study to assess the effects of folate supplementation (15 mg daily for 60 days) on the memory performance of 30 community- or aged-care-dwelling older participants, ranging in age from 70 to 90 years. Participants were selected for low plasma folate levels (<3 ng/ml) and mild to moderate memory complaints. They were tested pre- and postsupplementation using parallel forms of the Randt Memory Test (Randt and Brown 1983), a test that measures several aspects of acquisition and recall of verbal and nonverbal material. Fioravanti and colleagues found significant differences between the treatment and placebo groups on measures of attention and memory. Moreover, although there was no significant relationship between folate and cognitive performance at baseline, the sensitivity of the cognitive measures to treatment with folate was related to the initial level of the folate deficiency, such that the greater the deficiency at the start of the treatment the greater the cognitive improvement at the end of the treatment. Although these investigators administered daily doses of folate well above the tolerable upper intake levels, the results from their pilot study provide preliminary evidence for the efficacy of folate supplementation, but replication with larger and more diverse samples is needed.
Results from well-conducted placebo-controlled experiments would provide the most compelling evidence of the effects of B vitamins on cognition. So far, very few have been conducted, but results for the few studies that have been done suggest that B6 and folate supplementation appear to have positive effects on the memory performance of older adults. These findings require replication with an investigation of dose–response relationships as two of these studies (Deijen et al. 1992; Fioravanti et al. 1997) used dosages of supplements well above the recommended daily intake (RDI) levels.
B Vitamins and Dementia
Research into the effects of B vitamins on cognitive impairment associated with dementia rests on the assumption that nutrition may be one factor that plays a protective role in dementia of the Alzheimer's type (Nourhashemi et al. 2000). Research to date, although equivocal (Piccini, Bracco, and Amaducci 1998), has resulted in recommendations by the United States National Institute of Aging to conduct a blood chemistry profile for folate and B12 as part of the diagnostic process of dementia to rule out preventable causes (Pary, Tobias, and Lippmann 1990). The links between the B vitamins, particularly folate and B12, and dementia are largely based on findings that individuals with dementia, especially those with Alzheimer's disease, have lower serum folate and B12 concentrations (Selhub et al. 2000). Furthermore, serum levels of folate and homocysteine have been found to be related to severity of cognitive impairment.
Several studies have investigated the link between B12 deficiency and the incidence of dementia. For example, Karnaze and Carmel 1987 analyzed serum B12 levels in 17 people with primary dementia and 11 with secondary dementia and found that those with primary dementia had a higher incidence of B12 deficiency (20%) than did those with secondary dementia (0%). Ikeda, Furukawa, Mashimoto, Takahashi, and Yamada 1990 also found evidence of B12 deficiency among those with Alzheimer's disease compared with individuals with other dementias. In this study the deficiency was evident in the cerebrospinal fluid but not in plasma concentrations of B12. In contrast, Crystal and colleagues 1994 found that the incidence of dementia did not appear to be related to B12 deficiency and that treatment with B12 did not benefit performance on measures of cognitive impairment. However, in this study it was likely that B12 deficient participants were present in both the groups with and without dementia, making differences between the groups difficult to detect (Stabler et al. 1997).
Folate status has also been linked with the incidence of dementia and cognitive impairment. Sneath, Chanarin, Hodkinson, McPherson, and Reynolds 1973 examined serum folate levels among 113 patients of a geriatric ward and found that the 14 with dementia had levels lower than those of the group as a whole. In addition, they found a positive correlation between red-cell folate levels and cognitive performance scores among those with low folate levels. In support, Sommer and Wolkowitz 1988 reported a positive correlation between red-cell folate and scores on the MMSE among 13 patients with dementia, 10 of whom had a diagnosis of Alzheimer's disease.
Most recently, Snowdon and colleagues 2000 conducted neuropathological examinations of the brains of 30 participants in the Nun Study who had died and for whom blood measures were also available. On the basis of findings that high homocysteine levels are associated with progressive atrophy of the medial temporal lobe in those with Alzheimer's disease (Clark et al. 1998), they set out to determine whether serum folate is inversely related to the severity of atrophy of the neocortex. They found that serum folate was significantly related to atrophy of the cerebral cortex, but only among those participants with a significant number of Alzheimer's disease lesions. However, because folate was negatively correlated with atrophy, even in participants without brain infarcts and minimal atherosclerosis, these authors suggested that folate may exert an influence in maintaining CNS integrity in older age through nonvascular mechanisms too.
As discussed earlier, folate and B12 status may also be more accurately marked by homocysteine levels, and some studies have found associations between homocysteine levels and cognitive impairment. For example, Nilsson, Gustafson, Faldt, Andersson, and Hultberg 1994 found increased plasma homocysteine levels in cognitively impaired older adults with normal blood levels of folate and serum B12. McCaddon, Davies, Hudson, Tandy, and Cattell 1998 conducted a prospective case-controlled study of 30 individuals with Alzheimer's disease and found that they had significantly elevated total serum homocysteine levels compared with a cognitively intact control group. In addition, homocysteine levels and serum B12 were correlated with cognitive scores for the Alzheimer's group but not for the control group.
Levitt and Karlinsky 1992 propose two hypotheses to account for the relationship between B vitamin deficiency and cognitive impairment associated with dementia. They labeled the first the "low intake" hypothesis, which proposes that cognitively impaired individuals have a reduced capacity for self-care and nutrition and as a consequence develop vitamin deficiency. If this is the case, then cognitive impairment should be associated with deficiencies in a wide range of nutritional indexes. The second hypothesis is the "etiologic" hypothesis, which proposes that the B vitamins play a specific role in the development and severity of cognitive deficits. If this is the case, then the relationship between B vitamins and cognitive performance should exist for all types of dementias.
Levitt and Karlinsky 1992 aimed to test these hypotheses by examining the relationship between indexes of nutrition status and the severity of cognitive impairment among people with Alzheimer's disease and those with other forms of cognitive impairment. Serum B12 concentration correlated with MMSE scores only for the group with Alzheimer's disease, and this relationship remained after controlling for age, education, and duration of illness. The components of the MMSE that were most highly correlated were those of attention/calculation and sequencing. There were no relationships between serum and red-cell folate and cognition scores. The authors concluded that the results do not support the low intake hypothesis because B12 was related to cognitive impairment, but none of the other nutritional indexes (folate, magnesium, calcium, protein, globulin, glucose) were. However, the etiologic hypothesis was not supported either as the relationship between B12 and cognition was found only for those with Alzheimer's disease. The authors suggest a third hypothesis that may account for the association; that is, the disease process in Alzheimer's, which results in neuronal death, may also affect cellular function resulting in a decrease in the absorption, storage, utilization and/or excretion of B12. B12 levels may therefore be a marker for the progress of Alzheimer's disease. If this is the case, then we might expect that even though B12 supplementation may improve the nutritional status of deficient individuals, it may not affect cognitive performance. Indeed, Levitt and Karlinsky 1992 reported that two participants in their study with low B12 and possible Alzheimer's dementia who subsequently received B12 replacement therapy continued to show decline in cognitive function despite a return to normal blood levels of B12.
The few intervention studies that have been conducted provide some support for Levitt and Karlinsky 1992 third hypothesis. Teunisse, Bollen, von Gool, and Walstra 1996 gave B12 supplementation for 6 months to 26 individuals with dementia (all but one had a diagnosis of probable Alzheimer's disease) and subnormal B12 serum levels. The results indicated no effect of supplementation on the severity of cognitive decline among the treated group compared with an untreated group of individuals with Alzheimer's disease. Carmel and colleagues 1995 evaluated B12, neuropsychologic, and electrophysiologic indexes among 13 older adults with dementia and with low serum B12 levels before and after treatment with B12 supplementation. Improvements were found for homocysteine and hemoglobin levels, neuropathological symptoms, electroencephalographic abnormalities, and visual evoked and somatosensory abnormalities, but none were found on the performance of neuropsychological tests of cognitive performance.
Martin, Francis, Protetch, and Huff 1992 set out to examine the effects of B12 supplementation on cognitive performance in a group of 18 older-age participants who had low serum cobalamin and evidence of cognitive dysfunction. The participants received 1,000 micrograms of cyanocobalamin intramuscularly daily for 1 week, weekly for 1 month, then monthly for 6 months. Post supplementation scores from the Mattis Dementia Rating Scale (Mattis 1970) for 11 of the 18 participants showed improvement. However, only those participants whose impairment on mental-status testing had been in the mild range and who had been symptomatic for less than 1 year showed improvement. Most notably, those who had been symptomatic for less than 6 months responded best to supplementation. The authors concluded that age-related cognitive losses in early B12 deficiency might be reversible if the supplementation is initiated early in the process.
The findings from these intervention studies generally support Levitt and Karlinsky 1992 suggestion that Alzheimer's disease may be associated with a decrease in the ability to use Vitamin B12. Studies in which B12 was given as a supplement to those with cognitive impairment resulted in no improvement in the performance of cognitive tasks, with the exception of the Martin and colleagues 1992 study, in which those who were in the earliest stages of cognitive decline and whose impairment was in the mild category improved. Thus, it could be argued that supplementation might be critical in the early stages of Alzheimer's disease and that the window of opportunity for effective intervention with supplementation might be time limited, as the structural changes in the CNS occurring later in the disease process will preclude amelioration of cognition (Martin et al. 1992). Alternatively, studies in which SAMe, a metabolite of folate and B12, was given as a supplement (e.g., Fontanari et al. 1994) resulted in improvements in cognition. This may indicate that the use of SAMe bypasses some crucial metabolic step that is compromised during the disease. Clearly, more intervention studies are required to determine if and how B12 metabolism may be compromised among those with Alzheimer's disease.
|
General Summary |
|---|
|
TOPAbstractB Vitamins and Brain...General SummaryReferences |
|---|
The first consideration has to do with the selection of suitable outcome measures of cognitive performance. Research that investigates factors influencing cognitive function generally focuses on two distinct aspects of cognition: fluid and crystallised abilities (Horn 1982). Crystallised abilities are commonly demonstrated by the performance of tasks tapping verbal ability and are thought to reflect the capacity to apply information learned over the life span. They depend on stored knowledge and educational and cultural experiences and rely essentially on automatic information processing that uses past learning. In contrast, fluid abilities, demonstrated in the performance of tasks tapping speed of information processing, abstract reasoning, and episodic memory, reflect innate information-processing capacities that are determined, in part, by the integrity of the CNS. They depend on the ability to apply mental processes to situations requiring no previous knowledge, thus necessitating more conscious and effortful processing (Christensen et al. 1999; Salthouse 1996). Hence, this aspect of cognitive function may be more vulnerable or sensitive to the influence of nutritional factors known to enhance or impair the integrity and functioning of the CNS (Bryan 1998). Support for this argument comes from the findings of studies conducted to date indicating that B vitamin intake or status and intervention effects are found for abstract reasoning, memory performance, and spatial abilities, all of which are measures of fluid abilities. Because the impact of micronutrients on the brain and on cognitive performance is still not fully understood, it is important to use a comprehensive range of cognitive tasks that tap diverse aspects of cognitive performance to capture diffuse effects (i.e., effects that affect a wide range of cognitive abilities) and/or specific effects (i.e., effects that have an impact on specific cognitive abilities, such as memory encoding or strategic retrieval search). Moreover, because any nutritional effects are likely to be subtle, outcome measures should be sufficiently sensitive to capture variability in performance to facilitate the detection of subtle effects. Many previous studies investigating the effect of B vitamins on cognition among older adults have used tests of cognitive impairment such as the MMSE (Folstein et al. 1975) as outcome measures. Although relationships have been found between B vitamin intake and status and MMSE performance, these tests of impairment may not be suitable for use among healthy, nonimpaired samples of older adults. They may not be sufficiently sensitive, when used in unimpaired samples, to capture variability in performance or to discriminate between those aspects of cognitive function that are resistant to the effects of inadequate nutrition and those that are more vulnerable, and may therefore result in only weak effects. In other words, for the nonimpaired, the variability in performance scores on measures of cognitive impairment such as the MMSE may simply not be adequate to detect subtle effects.
Secondly, the nature or etiology of B vitamin deficiency might have implications for the form of supplementation in intervention studies. Risks of clinical and subclinical deficiencies of Vitamins B12, B6, and folate are associated with age-related changes in the gastrointestinal tract and/or vitamin metabolism (Rosenberg and Miller 1992). Therefore, information about the bioavailability of the vitamins from food or dietary supplements would be useful to guide research. Such information would be helpful in making decisions about whether to use foods or dietary supplementation to administer dosages of B vitamins in intervention studies. The length of time that dietary supplementation should be administered and the dose–response relationships should also be investigated. Dose–response relationships provide important information about the optimum dosage necessary to affect cognitive performance. An important initial question is whether to use the RDI, or higher dosages that fall within the tolerable upper limits (TUL). For example, researchers using folate supplementation for optimum physiological effects have recommended dosages greater than the RDI but within the TUL (Fenech et al. 1998). Investigations into the length of administration of dietary supplementation required to affect cognitive outcomes would be informative about whether effects of supplementation on the brain are acute and direct or longer term and perhaps indirect. This would give greater insight into biological mechanisms underlying the impact of nutritional factors on cognition. Longitudinal studies suggest that long-term effects of continuous, adequate dietary intake of B vitamins may be important in the maintenance of cognitive function with increasing age. However, short-term supplementation studies among those with clinical or subclinical vitamin deficiencies suggest that short-term interventions may be effective in enhancing or maintaining cognitive function. Because studies to date have not systematically investigated the consequences of varied lengths of administration or intake of the B vitamins, the significance of time-lag effects is unknown. In addition, given that absorption rates in older age are variable, the use of objective measures of B vitamin status, such as red-blood cell levels, which are more reliable measures of stable tissue levels (Bottiglieri 1996), or homocysteine levels, which are considered reliable markers of folate and B12 in particular (Ubbink et al. 1993), is crucial. Furthermore, objective measures would allow for the examination of threshold effects in the absence of linear relationships.
Finally, future studies could benefit from the use of samples using community-dwelling older age participants as well as clinical populations that are large enough to provide adequate power, given that nutritional effects are likely to be subtle. It is also important to control for extraneous variables that might affect the absorption of the B vitamins (e.g., smoking), and to identify other background and demographic factors that should be used as covariates. The methodological issues addressed here could be applied to the investigation of other nutrients thought to have an impact on cognitive performance and other psychological factors, such as mood and well-being, to enhance researchers' understanding of the links between nutrition and psychological function in older age.
Received for publication September 5, 2000. Accepted for publication May 22, 2001.
|
References |
|---|
|
TOPAbstractB Vitamins and Brain...General SummaryReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Wegkamp, W. van Oorschot, W. M. de Vos, and E. J. Smid Characterization of the Role of para-Aminobenzoic Acid Biosynthesis in Folate Production by Lactococcus lactis Appl. Envir. Microbiol., April 15, 2007; 73(8): 2673 - 2681. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Smith, S. A. Mathews Oliver, S. R. Zwart, G. Kala, P. A. Kelly, J. S. Goodwin, and C. B. Dyer Nutritional Status Is Altered in the Self-Neglecting Elderly J. Nutr., October 1, 2006; 136(10): 2534 - 2541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L Tucker, N. Qiao, T. Scott, I. Rosenberg, and A. Spiro III High homocysteine and low B vitamins predict cognitive decline in aging men: the Veterans Affairs Normative Aging Study Am. J. Clinical Nutrition, September 1, 2005; 82(3): 627 - 635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bryan, E. Calvaresi, and D. Hughes Short-Term Folate, Vitamin B-12 or Vitamin B-6 Supplementation Slightly Affects Memory Performance But Not Mood in Women of Various Ages J. Nutr., June 1, 2002; 132(6): 1345 - 1356. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
