Since 1990, the number of people globally living with dementia has more than doubled. This is mainly due to increased population growth and ageing [1
]. Importantly, independent of life expectancy, dementia mortality rates in women in 2016 were almost twice that of men. This may be partly attributable to the abrupt decline of estrogen at menopause and the associated loss of its protective effects on cardiovascular [2
] and neural functions [3
Estrogen activates estrogen α and β receptors (ER) on endothelial cells to facilitate vasodilatation by increasing endothelial nitric oxide (NO). Thus, estrogen deprivation can accelerate age-related arterial stiffening and impair tissue perfusion by reducing endothelium-dependent vasodilatation. This not only increases the risk of cardiovascular disease postmenopausally [4
] but also reduces cerebrovascular responsiveness (CVR) in postmenopausal women compared to pre-menopausal women and men [5
]. Reduced cerebral blood flow (CBF) and CVR are associated with cognitive impairment [7
]. We have also reported that reduced CVR during mental task activation (neurovascular coupling) is predictive of poor cognitive performance in postmenopausal women [8
]. A meta-analysis has shown that postmenopausal women perform worse on verbal memory and executive function tests compared to peri-menopausal women [9
]. Therefore, maintaining the health of the cerebral vasculature may slow cognitive decline in postmenopausal women.
Resveratrol (3,5,4′-trihydroxy-trans-stilbene), a phytoestrogen present in a variety of foods such as grapes, berries and nuts, has been shown to improve endothelial vasodilator function in humans [10
]. Resveratrol can act through multiple mechanisms including activation of endothelial ER to increase NO production and thereby facilitate endothelium-dependent vasodilatation necessary for adequate cerebral perfusion [11
]. Resveratrol has been shown to improve verbal memory in older adults after six months of 200 mg of supplementation per day compared to placebo [12
]. Using functional magnetic resonance imaging, Witte et al. found that resting-state functional connectivity in the hippocampus (a region critical for memory functions) and other brain regions was significantly increased in the resveratrol group and was correlated with improvement in verbal memory [12
]. Kennedy et al. reported that compared to placebo, acute doses of 250 mg and 500 mg of resveratrol resulted in dose-dependent increases of resting CBF in younger adults, although cognitive function was not affected [13
]. In a dose-response evaluation, Wong et al. found that 75 mg resveratrol, the lowest dose tested, was the most efficacious dose to acutely elicit global cerebral vasodilatation [14
] and improve performance of a sustained attention task in type 2 diabetes patients [15
]. In a subsequent pilot study of chronic resveratrol treatment (2 × 75 mg/day for 14-weeks) in 80 postmenopausal women, we observed an improvement of cognitive performance, which was accompanied by enhanced neurovascular coupling, suggesting that regular resveratrol supplementation might attenuate accelerated cognitive ageing in a vulnerable population by helping to maintain normal circulatory function [16
]. We now intend to investigate whether these benefits can be sustained long-term as well as examining its effects on a wide range of cardiometabolic markers.
In this 12-month trial, we sought to confirm the unique findings of our 14 week pilot study in postmenopausal women [16
] and ascertain whether the benefits of low-dose resveratrol on cerebrovascular and cognitive functions could be sustained with long-term supplementation.
We observed an improvement in overall cognitive performance which appeared to be due to improvements in processing speed and cognitive flexibility. Both processing speed and cognitive flexibility are part of executive function that requires speed, perceptual reasoning and accuracy to accomplish tasks [24
]. After peaking in the third decade of life, executive function declines at an estimated annual rate of -0.02 standard deviations [25
]. In fact, slowing of processing speed and mental flexibility are among the first cognitive changes reported in healthy older adults [24
]. This “slowing” can negatively affect performance on other neuropsychological tests such as verbal fluency [24
]. Our observed improvement in overall cognitive performance with resveratrol (d
= 0.18) could potentially reverse cognitive ageing by up to 10 years. Therefore, optimising executive function in healthy older adults may delay subsequent impairment across other cognitive domains.
Apart from our pilot study [16
], there are only two clinical trials on cognitive effects of chronic resveratrol supplementation [12
]. A six-month study by Witte et al. showed improvement of verbal memory in healthy older adults with a similarly low dose (200 mg/day) [12
]. On the other hand, a study in young adults [26
] reported a lack of interpretable cognitive effects following resveratrol supplementation (500 mg/day for a month), despite showing improvement of cerebrovascular function. This may be due to the high-performing cognitive status of this cohort, the shorter study duration or the higher dose of resveratrol given, bearing in mind that we have previously reported lower neurovascular coupling efficacy of resveratrol at higher doses [14
We hypothesized that the observed cognitive benefits of resveratrol might be partly mediated by sustained improvement of endothelium-dependent vasodilator function, which modulates CBF during times of demand [11
]. Supporting this, we found improved resting BFV, PI and attenuation of decline in neurovascular coupling following resveratrol supplementation. Allowing for normal ageing processes, cognitive decline and ultimately, dementia are linked to accelerated decline in resting CBF and CVR due to a decreased delivery of oxygen and nutrients in vulnerable brain regions such as the hippocampus [27
]. In fact, a large population-based study in Rotterdam reported that healthy older adults with lower CBF measured by MRI performed significantly worse on tasks of information processing speed, executive and global cognitive function compared to those with higher CBF [28
]. In addition, cerebral artery stiffness, marked by increased PI, is also associated with cognitive impairment [29
] and predicts the progression from mild cognitive impairment to dementia [30
]. Combination of low BFV and high PI can result in chronic hypoperfusion, which may cause progressive loss of neuronal function [31
]. Given this evidence, our observation of improvements in resting BFV and PI with resveratrol highlights the ability of regular resveratrol supplementation to sustain cerebrovascular function, which may, in turn, preserve cognitive function in elderly women.
Evidence has shown that resveratrol can modulate CBF through multiple mechanisms including activation of Sirtuin-1, adenosine-monophosphate protein kinase and ER α and β to increase endothelial NO synthase activity. This, in turn, increases NO production and bioavailability to facilitate vasodilator responsiveness and arterial smooth muscle relaxation during demand [11
]. To assess the cerebral vasodilator response, hypercapnia provocation is commonly used to increase of blood CO2
concentration. The increase in BFV from resting values reflects dilation in the downstream microvasculature and thus is a good measure of global cerebral vasodilatation, independent of specific neuronal activation [32
]. We did not observe enhancement of CVR to hypercapnia with resveratrol in this study, perhaps because cerebral vasodilator capacity was still optimal in our cohort of healthy elderly women. Older adults with established cardiovascular risk factors such as hypertension, dyslipidaemia or diabetes had lower CVR to hypercapnia compared to healthy older and young adults [32
]. Moreover, impaired CVR to hypercapnia has been observed in individuals with mild cognitive impairment and Alzheimer’s disease [33
], implicating cerebral hypoperfusion in the pathogenesis of cognitive impairment.
Although we did not see any enhancement of CVR to hypercapnia, we found that resveratrol attenuated the decline of neurovascular coupling. Neurovascular coupling differs from CVR to hypercapnia as it represents localised changes in cerebral BFV in response to specific neuronal events. When neurons fire, the endothelium is activated to release NO, resulting in dilatation of local arterioles, which is detected as increased blood flow in the arteries supplying the activated brain region. Deficits in this synchronised action can lead to inadequate perfusion of critical brain regions, resulting in poor performance of cognitive tasks [34
]. Therefore, impaired neurovascular coupling may be considered the beginning of a chain of events leading to a progressive decline in brain metabolism and cognition that characterizes dementia [35
]. Identification of early cerebrovascular dysfunction has important diagnostic implications for future cognitive decline [36
]. We have previously shown that poor cerebrovascular function is predictive of cognitive decline in healthy postmenopausal women [8
We have previously shown that the beneficial effects of resveratrol on systemic vascular function are greater in those individuals with poorer vascular function at baseline [37
]. This may account for the greater magnitude of improvement in neurovascular coupling seen in our pilot study of resveratrol supplementation in postmenopausal women [16
] (d = 0.71), which elicited a greater improvement in overall cognitive performance (d = 0.69) than observed in the present study, as the participants in the pilot study had much lower overall neurovascular coupling at baseline. In this study, we have once again found that lower cognitive function at baseline is associated with greater improvements in overall cognitive performance by resveratrol. Taken together, regular resveratrol supplementation may be more beneficial for improving cognitive function in older adults with higher level of baseline endothelial dysfunction or cognitive impairments. The present study not only confirms the significant benefits of resveratrol seen in the pilot study but, most importantly, it shows that these benefits are not short-lived but can be sustained with ongoing supplementation for at least 12 months. Indeed, data from the placebo-treated group revealed a further decline of neurovascular coupling over 12 months, even in healthy older women. The present study demonstrates the potential of resveratrol to attenuate this decline, thereby protecting higher brain function in the elderly.
We did not find any significant difference in systemic vascular function (BP, AC) or fasting glucose, insulin and lipids between the two treatments, although Timmers et al. had shown that resveratrol (150 mg/day for 30 days) can reduce plasma triglycerides and improve insulin sensitivity in obese male subjects (mean age: 52 ± 2 years, BMI: 31.6 ± 0.7 kg/m2
) with low-grade chronic inflammation [38
]. They also reported reduced inflammation; however, we did not observe an effect of resveratrol on hs-CRP, indicating that resveratrol is unlikely to alter cardiometabolic markers in elderly participants who are otherwise healthy (BMI: 25.6 ± 0.3 kg/m2
) and without overt inflammation or other metabolic disturbances.
Interestingly, we found a significant inverse relationship between the supplementation-induced changes in fasting glucose and overall neurovascular coupling. Our finding is consistent with Witte et al., who observed a reduction in HbA1C, the long-term biomarker of glucose control, which correlated with increased functional connectivity in the hippocampus assessed by functional MRI [12
]. This is consistent with our previously published hypothesis that resveratrol might improve glucose uptake and/or reduce insulin demand by enhancing vasodilator function in skeletal muscle [39
]. A recent meta-analysis has also shown that resveratrol can improve insulin sensitivity and insulin secretion in pancreatic β-cells and increase glucose uptake via Sirtuin-1 or adenosine monophosphate protein kinase mediated pathways [40
]. Resveratrol could also facilitate expression of glucose transporter type-4 by acting on estrogen receptors on the endothelium and stimulate skeletal muscle glucose uptake [41
], which could be an added metabolic benefit for postmenopausal women. Although we cannot rule out the exact mechanism of action, this finding could imply that resveratrol may induce improvements in energy metabolism through improvements of microcirculatory function, which could, in turn, protect neuronal function and counteract cognitive decline.
This is the first long-term study of effects of resveratrol supplementation in postmenopausal women. The 12-month duration not only increases confidence in the sustainability of any benefits but it also eliminates any influences of seasonality on the outcomes. Whilst generating further evidence on the benefits of resveratrol for healthy ageing, there are some obvious limitations to its application. Individuals with higher health consciousness and cognitive awareness tend to volunteer for such intervention trials, which may influence their compliance and performance and ultimately, the interpretation of the results. Moreover, as our sample included healthy postmenopausal women only, the results cannot be generalized to the entire population. Further trials are necessary to determine the extent of potential benefit for men and for specific risk groups, e.g., hypertensives and diabetics, of both sexes.