Advancing age can be associated with an increase in cognitive dysfunction, a spectrum of disability that ranges in severity from normal age-related changes through mild cognitive impairment (MCI) to dementia; with the latter defined as a progressive decline in memory, thinking, language and judgment that is sufficient to impair activities of daily living [1
]. An estimated 50% of those diagnosed with MCI will go onto develop dementia within 5 years of diagnosis [2
]. Globally, it is estimated that 48 million people are currently suffering from dementia and the figures are predicted to triple by 2050 [3
]. Dementia is a leading cause of disability, dependency and decreased quality of life among older people [3
] and presents many social, economic and health care challenges that will continue to increase with an ageing population. Therefore, the identification of strategies to prevent or delay the onset of dementia has become a major global public health priority.
A number of nutritional and lifestyle factors have emerged as potential modifiable risk factors for cognitive decline in ageing [4
]. In particular, there is considerable epidemiological evidence to suggest that sub-optimal status of folate, the related B-vitamins, and/or elevated concentrations of the metabolite homocysteine, contribute to cognitive dysfunction [5
] and to a greater rate of cognitive decline in ageing [11
]. Elevated plasma homocysteine and lower folate have been most consistently associated with cognitive dysfunction in ageing [6
]. There is also evidence to support a role for vitamin B12 [15
] and to a lesser extent vitamin B6, although the latter has been far less extensively investigated [9
]. There is also some evidence in the form of randomised controlled trials to show beneficial effects of B-vitamin supplementation on cognition in ageing [18
]. A number of other trials have failed to detect significant benefits [21
] with recent meta-analyses concluding that there was no beneficial effect of B-vitamin supplementation on cognition [24
]. However, a number of these trials may have been too short in duration; conducted in healthy individuals, patients with severe dementia; or in those with optimal B-vitamin status and so unlikely to benefit from vitamin supplementation [26
]. The strongest evidence to date of a causal relationship between B-vitamins and cognition comes from the Homocysteine and B-vitamin in Cognitive Impairment (VITACOG) study. This study showed that combined B-vitamin supplementation for two years had beneficial effects on cognitive performance in participants with MCI and elevated plasma homocysteine concentrations [20
]. More importantly, it also demonstrated that B-vitamin supplementation reduced the rate of brain atrophy by 30% as measured using MRI [27
]. A subsequent report from the VITACOG investigators reported that the atrophy occurred in grey matter areas of the brain which are particularly vulnerable to Alzheimer’s disease [28
The intervention doses administered in VITACOG were well in excess of recommended dietary intakes and therefore whilst the VITACOG papers provide powerful evidence of a role for folate, vitamin B12, and/or vitamin B6 in cognition, the relevance of these results to nutrition, and thus prevention of cognitive dysfunction in ageing is unclear. Furthermore, epidemiological research generally in this area has predominantly focused on plasma homocysteine, folate and vitamin B12; most studies have overlooked vitamin B6 and all have ignored the role of vitamin B2. Consequently, the influence of all the relevant B-vitamins involved in one-carbon metabolism on cognition is not fully understood. Therefore, the aim of this study was to investigate whether lower dietary intake or biomarker status of B-vitamins (folate, vitamin B12, vitamin B6 or riboflavin) at baseline was associated with a greater rate of cognitive decline over a 4-year follow-up period in healthy older adults.
Of the 662 healthy volunteers, 255 ≥ 60 years were identified as potential participants and of these 155 were available to participate in the follow-up assessment (Figure 1
). Only those that participated at both timepoints are presented in this paper; those lost to follow-up were older and had significantly lower vitamin B12 status (Appendix A
The characteristics of participants at initial screening are shown in Table 1
. Participants had a mean age of 70 years, were predominantly female, well-educated and had a low rate of depression. The majority of participants were regular consumers of foods fortified with B-vitamins (75%). Dietary intakes compared favourably with current UK dietary recommendations [39
] as reflected in good overall B-vitamin biomarker status. As a result of the exclusion criteria, no participant was deficient in vitamin B12, however some 3% were identified as deficient in folate and 11% deficient in vitamin B6. Gastric function was assessed by using pepsinogen I:II ratio, 12% had evidence of atrophic gastritis (pepsinogen I:II ratio < 3; data not shown).
The relationship between dietary intakes and corresponding blood biomarker concentrations were examined for each B-vitamin of interest (Figure 2
). Dietary intakes for total folate, vitamin B6 and riboflavin were each significantly correlated with the corresponding blood biomarker concentration. Of note, vitamin B12 intake was significantly correlated with serum holoTC but not serum total vitamin B12, the more typically measured biomarker (r
= 0.134, p
= 0.104; data not shown).
The influence of several lifestyle factors, B-vitamin dietary intake and B-vitamin biomarker status, as determinants of rate of cognitive decline are shown in Table 3
. Of the general health and lifestyle factors examined only age and baseline MMSE score were predictive of cognitive decline. In addition, after adjustment for age and baseline MMSE score, no associations were observed between disease history (CVD, diabetes and gastrointestinal; data not shown) or medication use with the exception of use of analgesic medication (p
= 0.035; data not shown). Vitamin B6 was found to be the only B-vitamin that was predictive of cognitive decline. After adjustment for age and baseline MMSE score, individuals with lower vitamin B6 biomarker status (PLP range ≤ 43.3 nmol/L; p
= 0.002) or lower dietary B6 intakes (0.9–1.4 mg/day; p
= 0.018) were at a 3.5–4 fold greater risk of cognitive decline. None of the other B vitamins or plasma homocysteine concentrations were associated with the risk of cognitive decline in this cohort.
This study in healthy older adults, initially with normal cognitive performance, indicates that vitamin B6 is an important predictor of cognitive decline in ageing. Lower dietary intake and biomarker status of vitamin B6 were associated with a greater rate of cognitive decline over a subsequent 4 years follow-up period. No significant association of dietary intake or biomarker status with cognitive decline were observed for the other B-vitamins (folate, vitamin B12 and riboflavin). To our knowledge, this is the first longitudinal study to consider the impact of both dietary intake and biomarker status of all four relevant B-vitamins involved in one-carbon metabolism on cognitive health in ageing.
Whilst the influence of vitamin B6 on cognition has not been as fully investigated as folate and vitamin B12 a number of studies have reported observations consistent with the current study. Our results showed that participants with lower status of vitamin B6 (PLP; the measure of active vitamin B6) at baseline were 3.5 times more likely to have a greater rate of cognitive decline over a 4 years follow-up period. Furthermore, the association between vitamin B6 and cognitive decline was not confined to those with clinical deficiency, lower status included individuals in both the deficient (PLP < 30 nmol/L) and sufficient range (PLP 30–43 nmol/L) which would suggest that optimal vitamin B6 may be important for cognitive health in ageing. Consistent with the biomarker data, those with lower dietary intakes of vitamin B6 at baseline were 4 times more likely to have a greater rate of cognitive decline over the 4 years time period. Our results are in good agreement with findings from other studies, low vitamin B6 status (PLP < 46 nmol/L) and corresponding dietary intakes were previously associated with cognitive decline over a 3 years period in the Veteran Affairs Normative Ageing Study [40
]. There is also evidence from several cross-sectional studies to support an association between low vitamin B6 and cognitive dysfunction [9
] and Alzheimer’s disease [43
]. Furthermore, vitamin B6 status was associated with cognitive performance in high functioning older adults at baseline, though not with cognitive decline over the 7 years follow-up period in the MacArthur study of Successful Ageing [8
]. Certain other studies have failed to detect any significant association between vitamin B6 and cognitive function, however, these studies have relied on dietary intake measures alone with no corresponding measurement of blood biomarker status [45
]. Few RCTs have investigated the independent effect of vitamin B6 on cognitive function and only one very early study reported beneficial effects of vitamin B6 supplementation on memory [19
]. Subsequent RCTs have investigated the effect of vitamin B6 in combination with folate and vitamin B12, with some studies reporting beneficial effects on cognitive function however the independent effect of vitamin B6 cannot be determined [18
Whilst elevated plasma homocysteine, low folate and, to a lesser extent, vitamin B12 status have been frequently associated with cognitive decline [11
] there was no evidence of significant associations for these biomarkers in the current study. A number of other studies have reported similar findings [49
]. The findings in the current study may be explained to some degree by the fact that vitamin B6 seemed to be the limiting nutrient within the cohort. There was a greater incidence of deficiency of vitamin B6 (11% clinical deficiency) compared with folate (3%) or vitamin B12 (0%). Also, the lack of a significant association between cognitive decline and plasma homocysteine concentration is almost certainly is a reflection of the low prevalence of folate deficiency. Furthermore, the concept that the association between B-vitamin status and cognition is determined by the limiting nutrient within that population group is further supported by evidence from published RCTs. One trial of healthy older adults in New Zealand reported no benefit of combined B-vitamin supplementation on cognitive function [21
], whereas another similar study from the Netherlands showed that supplementation with folic acid significantly improved cognitive performance [18
]. A notable difference between these two studies was that baseline folate status tended to be far lower in the Dutch trial, suggesting that the cognitive benefit related to the correction of sub-optimal B-vitamin status whereas additional B-vitamins to an already optimal population may have no beneficial effect.
The mechanism linking vitamin B6 with cognitive health in this and other studies in ageing is not clear however, it is biologically plausible given the widespread functions of vitamin B6 within the brain and nervous system [53
]. Vitamin B6 has a crucial role in the synthesis of a variety of neurotransmitters including dopamine and serotonin [55
] and can act as a potent antioxidant [56
]. In addition, higher vitamin B6 intakes have been associated with greater grey matter volume [58
] and combined B-vitamin supplementation (including vitamin B6) has been shown to slow brain atrophy, an important feature cognitive dysfunction [27
The current study has a number of strengths and limitations that merit comment. To our knowledge, this is the first longitudinal study to investigate the association between cognitive decline and all relevant B-vitamins along with their corresponding dietary intakes. The MMSE is the most widely used screening tools for cognitive dysfunction and although it has been criticised for lacking sensitivity, few previous studies have used it to measure cognitive change in a healthy older population. However, a meta-analysis reported a mean decline in MMSE of between 0.16 and 0.56 points per year in cognitively healthy people which compares favourably to the overall rate of decline observed in this study (mean 0.39 points per year) [38
]. In addition, the rate of decline observed in this study was identical to that observed in the Rotterdam Study of community dwelling older adults free from cognitive impairment [59
]. While the use of the MMSE may be perceived as a limiting factor in the current study, it could be argued that its use would only attenuate the associations observed and that the use of more sensitive tools would have, if anything, detected more subtle differences thus strengthening the results. Another well-recognised limitation of longitudinal studies of this kind is that individuals with the greatest decline in cognitive function are more likely to be lost to follow-up [60
]. Indeed, in this study the non-participants were more likely to be older but any non-response bias would ultimately underestimate the associations between baseline B-vitamin status and cognitive decline and this could not have influenced the current findings.