Introduction

Dementia is a clinical syndrome characterized by progressive impairment in cognitive ability and capacity for independent living. An estimated 35.6 million people lived with dementia worldwide in 2010, with numbers expected to almost double every 20 years, to 65.7 million in 2020 and 115.4 million in 2050 [1••]. Dementia and cardiovascular diseases (CVD) have an increasing incidence and prevalence in the elderly population, causing a decline in quality of life, and are the leading causes of death. A growing literature implicates CVD as a risk factor for dementia. Studies of both clinical and subclinical CVD have consistently reported associations with impaired cognitive function [2, 3•]. In addition, a number of studies suggest that CVD risk factors (CVDRF) such as hypertension, diabetes, obesity, and smoking are independently associated with the development of dementia [4•]. Primary prevention that has a focus on improving CVDRF control could have an important impact on the future prevalence and incidence of dementia. Prior studies have shown that even small shifts in risk factor levels at the population level can explain up to two-thirds to three-fourths of the dramatic reductions in CVD mortality rates [5,6,7]. This population-level strategy can similarly be applied for reducing dementia rates. In a modeling study, it was estimated that a modest 10% reduction in risk exposure levels could reduce the prevalence of Alzheimer’s dementia by up to 1.1 million cases worldwide [8•]. Greater absolute reductions in dementia would come through public health measures that involve a modest lowering of risk factors among the larger proportion of the population with risk factors near or slightly above the mean [9•]. Further understanding the role of CVDRF in cognitive decline could have an important role in developing effective preventive strategies for dementia and improving quality of life in aging populations [10].

Prevention at the population level works best when programs to mitigate the risk factors are widely available throughout the whole population. In 2010, the American Heart Association adopted better cardiovascular health (CVH) as a population strategy to prevent heart disease and stroke epidemics by encouraging system approaches to help individuals identify and adopt healthier life choices [11••]. The construct of CVH is defined as a metric with the simultaneous presence of seven favorable health behaviors and factors (abstinence from smoking within the last year, ideal body mass index, physical activity at goal, consumption of a Mediterranean dietary pattern, untreated total cholesterol < 200 mg/dL, untreated blood pressure < 120/< 80 mm Hg, and absence of diabetes mellitus) [12•]. Each of the health behaviors and health factors have been consistently associated with CVD-free survival, quality of life, compression of morbidity, healthy aging, overall longevity, and reduction in healthcare costs [13]. Several recent studies have reinforced the relevance of this metric in reduction of CVD rates [14•, 15•].Given the close links between CVD and dementia pathophysiology, early identification and modification of common risk factors could have a major impact on reducing these epidemics, particularly among the elderly population. This article will review the current evidence linking CVD, CVDRF that are part of the CVH metric, and cognition. Even though there are many types of dementias, our review will focus on prevention of the two most common, Alzheimer’s dementia and vascular dementia.

Cardiovascular Disease and Cognitive Decline: Overlapping Pathophysiology

Several studies have highlighted the association of CVD in middle age with dementia in later life regardless, of race and gender [16,17,18,19]. Cognitive syndromes such as Alzheimer’s disease-type dementia and mild cognitive impairment (MCI) share several pathophysiological pathways with CVD, such as inflammation, increased oxidative stress, and changes to nitric oxide bioavailability [4•, 20]. The relationship between CVD and cognition suggests a primary role of metabolic and vascular damage in the etiology of dementia. Another postulated mechanism is that cognitive impairment could be due to changes in brain perfusion caused by CVD [21]. The association between the presence of CVD or CVDRF and late-life cognitive impairment has also been attributed to vascular changes and amyloid deposition [21]. A recent longitudinal study of community-dwelling adults strongly linked greater chronological age, symptoms of CVD, and processing speed decline to elevated white matter lesion burden [22]. Cerebral white matter lesions, which are prevalent in a majority of older adults, are thus also associated with both cognitive decline and CVD. Subclinical CVD has also been associated with poorer cognitive function. While associations in cross-sectional and prospective studies have been reported between prevalent CVD and cognition, there have also been some reports of important non-associations [16, 23]. These conflicting reports could be due to differences in the severity and definition of the disease, cognitive assessments utilized, duration of follow up, and sample size. Recent literature in the area of CVD and cognition is summarized in Table 1.

Table 1 Summary of recent relevant studies of association between clinical or subclinical cardiovascular disease (CVD) and cognition
Full size table

Association of Cardiovascular Health Factors and Behaviors with Cognition

More than one in three US adults have at least one CVDRF and the prevalence of these factors increases with age. Several of these CVDRF are modifiable and part of the CVH metric [36•]. Growing evidence indicates that CVDRF interfere with normal cognitive functioning and could be directly related to the pathogenesis of dementia via overlapping vascular and cellular mechanisms [37]. The CVDRF result in subtle structural changes in the brain at first, accelerating decline in cognitive abilities and eventually causing dementia [2]. Morphological changes to brain structures seem to occur with the presence of even a single untreated risk factor, causing cognitive impairment and dementia [2]. In a study of MCI patients, 60% of them developed dementia and subgroups with CVDRF had a higher conversion rate to Alzheimer’s dementia [38]. Estimates suggest that up to a third of Alzheimer’s cases are potentially attributed to CVDRF and, thus, could be prevented [39]. All known CVDRF continue to be the focus of studies to further identify modifiable risk factors of dementia. Below, we review individual health factors and behaviors that are part of the CVH metric and their relationship with cognitive function.

Hypertension

Blood pressure (BP) is the most studied CVDRF in cognition literature [40••]. Hypertension is an established risk factor for stroke and silent infarcts and is associated with both vascular and Alzheimer’s dementia [41]. Sustained exposure to high pressure flow has multiple neuropathological effects including cerebrovascular atherosclerosis, vascular remodeling, hypoperfusion, and increased frequency of white matter lesions in the brain [42•, 43]. Reduced brain perfusion leads to ischemic lesions, lacunar, cortical infarcts [44••, 45]. Some studies have also shown that hypertension can be directly involved in amyloid beta deposits and neurofibrillary tangle formation [46, 47]. This can adversely affect cognitive function particularly relating to memory, attention, and executive function [48, 49•]. The Honolulu-Asia Aging study reported that elevated levels of BP are associated with lower gray matter volumes in the hippocampus and lateral temporal lobe [49•]. Studies in the elderly have similarly shown that higher systolic BP is associated with higher rates of cognitive decline [50•]. Higher diastolic BP by itself is also associated with poor cognition and could impact executive function [51•]. Further evidence of the BP-cognition relationship is illustrated with the Atherosclerotic Risk in Communities cohort (N~11,000), which showed the relationship between CVDRF (particularly hypertension and diabetes) and decline in cognitive functioning in an older population [52••]. On the other hand, a reduction of blood pressure has a protective effect on cognition [53]. For example, patients who received anti-hypertensive treatment had lower rates of neuritic plaques and neurofibrillary tangles than controls [54].

Despite the strong BP-cognition data, there has been some controversy about the target Systolic BP for preventing cognitive decline [55••, 56]. The Systolic-Hypertension in Europe study reported that BP-lowering therapy reduced the risk of dementia by 55% [57]. In the Systolic Blood Pressure Intervention Trial - Memory and Cognition in Decreased Hypertension (SPRINT MIND), intensive lowering of BP to a goal of <120mm Hg reduced the risk of MCI and dementia risk [58••].This is the first randomized clinical trial demonstrating that an intervention can reduce the incidence of MCI/dementia, highlighting the significance of BP as a risk factor for cognitive decline.

Some observational studies have also found an association of low BP with cognitive impairment and that hypertension could be a protective response to cerebral hypoperfusion [59•]. Possible reasons for these conflicting findings could be use of varying cognitive instruments in different studies and heterogeneity in the impact of hypertension on specific cognitive domains. Further research into the effects of hypertension, particularly in midlife, could point to interventions for prevention of MCI and later dementia.

Diabetes

Diabetes is positively associated with a decline in cognitive function, including MCI, and dementia [60, 61, 62•, 63]. In a pooled analysis of 14 studies, individuals with type 2 diabetes were at 60% greater risk for developing dementia [64]. In a prospective study, an increase in the number of metabolic syndrome components, particularly diabetes, was associated with a 23% age-adjusted increase in the risk of dementia [65•]. Diabetes has an impact on overall brain volume and cognition particularly on measures of attention and working memory [66]. Chronic exposure to hyperglycemia results in improper cellular utilization of glucose, impacting most organs in the body, and is particularly damaging to the central nervous system [67]. There is often a convergence of physiological factors that result in comorbidity, such as increased oxidative stress in individuals having comorbid diabetes mellitus (DM) and Alzheimer’s disease or the commonality of infarcts and atrophy in the brains of individuals with diabetes mellitus [68].

Results from several studies have shown that among people who have DM, those with longer disease duration and higher levels of glycosylated hemoglobin A1C have faster rates of cognitive decline and decreased cognitive function [60].

Studies that also explored the pathogenesis at the cellular level have concluded that physiological links between DM and Alzheimer’s disease exist, often resulting in the exacerbation of one another [63, 69]. On the other hand, autopsy studies have pointed that DM is associated mostly with non-Alzheimer’s type dementia pathology [70•]. Thus, the exact mechanisms and temporality informing this relationship between DM and cognitive functioning are still not fully understood [67, 71]. In the elderly population, DM has been linked with MCI and those with DM have greater baseline deficits in domains of memory and language [72•, 73]. Even in young adults, higher intra-individual fasting glucose variability was associated with worse processing speed, memory, and language fluency [74•]. Despite the growing evidence of diabetes and risk of cognitive impairment, there is no clear evidence that treating diabetes with intensive control is beneficial for cognitive outcomes. Action to Control Cardiovascular Risk in Diabetes - Memory in Diabetes Study (ACCORD MIND) is the first randomized trial in older persons with DM to test the effect of intensive compared with standard glycemic treatment strategies on multiple cognitive domains. The study failed to show a long-term benefit in cognitive outcomes (except brain volume) with intensive glycemic, BP, or lipid intervention [75••, 76•]. Cognitive dysfunction also influences the ability of patients to follow complex chronic disease management, affecting medication adherence and increasing DM complications such as hypoglycemia [77]. Due to the high prevalence of DM in the US population, future studies must be done to understand the impact of DM interventions on cognition.

Dyslipidemia

Dyslipidemia is also a highly prevalent condition in the US and is a widely recognized CVDRF. Proteins required for cholesterol distribution such as low-density lipoprotein (LDL) receptor and LDL receptor-related protein 1 (LRP1), may play a role in amyloid β homeostasis [78]. The APOE gene, a strong genetic risk factor for sporadic Alzheimer’s disease, is also involved in cholesterol transport [79]. Studies have shown that higher levels of LDL, very low-density lipoprotein, and triglycerides are associated with poorer performance in attention, working memory, category fluency, and delayed recall [80]. Other prospective studies have shown that midlife dyslipidemia is associated with vascular dementia and Alzheimer’s disease [81•, 82•, 83]. Animal studies have also shown that dietary cholesterol can accelerate amyloid β deposition in the brain and that it may indirectly promote production of neurofibrillary tangles [84]. Some epidemiological studies have, however, shown no association between dyslipidemia and dementia risk [85•, 86]. For example, in the Framingham heart study, cholesterol levels were not associated with the risk of Alzheimer’s dementia [87•]. In a systematic review, cholesterol levels in midlife but not later in life were associated with dementia [88•]. Despite observational studies suggesting that statins could be protective for dementia, several randomized clinical trials involving treatment of dyslipidemia with statins have shown no clear beneficial effect on cognition [89, 90•]. The role of cholesterol in cognitive decline and as a cause of dementia is thus unclear. More well-designed and larger studies are needed to further clarify the link between dyslipidemia and cognition.

Smoking

Tobacco use is one of many modifiable lifestyle factors associated with several negative health outcomes [91]. This behavioral risk factor has been associated with increased risks of accelerated cognitive decline, incidence of MCI, and Alzheimer’s disease [92, 93]. A meta-analysis has shown that current smokers, relative to non-smokers, have a 79% increased risk of Alzheimer’s dementia [94•]. Smoking is associated with increased oxidative stress, low-grade inflammation which leads to more white matter lesions, cerebral hypoperfusion, and accelerated cerebral atrophy, ultimately resulting in cognitive decline [95]. The mechanisms and degree of association are not completely understood and often debated [96]. Many studies that have not been successful in assessing the relationship between smoking and cognition are studies that included individuals over the age of 60 [97•]. The finding that smokers have been shown to have lower rates of cognitive impairment than non-smokers, could be attributable to survival bias. That is, older participants are generally more likely to experience issues with cognition and are less likely to be smokers because smokers have a decreased likelihood of surviving into old age [98].

Most of the recent literature shows that there is a dose-response association—those who smoke—and for a lengthy time period, have declining cognitive abilities compared with those who never smoked. To fully understand the relationship between smoking and cognitive abilities, several studies have been conducted to include younger (under the age of 60) populations [97•]. A cross-sectional study assessed midlife changes in cognitive abilities of 3035 individuals until they reached the age of 53 [97•]. Individuals in this cohort who smoked more than 20 cigarettes per day experienced significantly lower scores in domains of cognitive flexibility and psychomotor speeds. Because smoking tobacco has been previously identified as a strong risk factor for vascular disease, and vascular disease is often associated with dementia and other cognitive diseases, this potential mechanism continues to draw interest from researchers [99•].

Diet

Studies have shown positive effects of dietary patterns such as the Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND), and antiinflammatory diets on cognitive health [100]. These dietary patterns generally emphasize plant-based, rich in poly-unsaturated fatty acids and lower consumption of processed foods. The most well-known is the MIND, which has been associated with a reduced risk of memory complaints and dementia [101•]. The MIND diet has ten brain healthy food groups (green leafy vegetables, other vegetables, nuts, berries, beans, whole grains, seafood, poultry, olive oil, and wine) and five unhealthy food groups (red meats, butter and stick margarine, cheese, pastries and sweets, and fried/fast food). The Mediterranean part of the MIND diet has been associated with increased total brain volume, total grey matter volume, total white matter volume, and mean cortical thickness [102••]. Morris et al. compared the Mediterranean, DASH, and MIND diets and the onset of Alzheimer’s disease and found all diets slowed the rate of Alzheimer’s disease onset over the course of an average of 4.5 years, but the MIND diet most effectively reduced this rate [103]. On the other hand a “western diet” pattern of red meat and sausages has been associated with a smaller left hippocampal volume [104].

Many of these healthy diets for cognition contain high levels of nutrients such as vitamin C, flavonoids, tyrosine, and unsaturated lipids like omega-3 polyunsaturated fatty acids (n-3 PUFA) that have been positively associated with cognition by promoting hippocampal neurogenesis, cognitive function and plasticity during aging via the gut-brain axis [102••]. More recently, the Maine-Syracuse Longitudinal Study found an association between sugar-sweetened soft drink consumption in DM patients and decreased cognitive function but found no significant decrease with artificially sweetened beverages [105]. Research has moved to understanding the effect of whole diets rather than specific nutrients and supplements as they may interact in unclear ways [106•]. More randomized trials and longitudinal studies will help explain further the association of different diets with cognition.

Physical Activity

Aerobic and non-aerobic, acute and long-term physical activities are among the lifestyle factors that are positively correlated with connectivity of different parts of the brain [107,108,109]. Increasing physical activity is known to have favorable implications on cognitive abilities and functionality for all individuals, including children, healthy mid-aged adults, and older adults [108, 110•]. Exercise and physical activity has also been associated with cognitive improvement for individuals with MCI [111]. The domains of cognitive improvement associated with physical activities include attention, processing speed, and working memory [112, 113•].

Though brain atrophy is often considered the norm among aging populations, preservation of brain functionality while aging is more likely for those who have maintained a consistent physically active lifestyle. However, several studies have shown that even acute physical activity and exercises have positive short-term alterations on cognition [114]. In a randomized controlled trial evaluating differing intensities of physical activity and cognitive response, those who performed moderately intense exercise had higher scores on the cognition tests than those who did not exercise [115]. In another study, individuals’ cognitive performance improved after cycling, however, more fit individuals had longer cognitive improvement post-workout [114]. In another RCT, individuals with 378 MCI who participated in aerobic exercises showed more improvement in their cognitive assessment compared with those who participated in a health education seminar [111]. Though the neurophysiological mechanisms of acute and long-term physical activities’ relationship with brain health and cognition have been researched, the degree and moderators of these mechanisms should be further explored [116]. Currently, most individuals in US are not meeting the scientifically recommended levels of physical activity. Therefore, public health measures are needed to motivate individuals to improve their levels of physical activity, and thus their neural efficiency and brain health.

Obesity

A higher body mass index (BMI) negatively affects brain structure and cognition [117]. Obesity is independently associated with reduced gray matter volumes and poor performance in measures of executive function that may subsequently affect cognition [118]. Obesity is also related to several other health conditions such as DM and hypertension and could exert a negative effect on cognition. Although studies demonstrate an association between obesity and cognitive decline, there remains inconsistency in the findings, likely from differences in methodology, type of cognitive measures used, and whether important covariates such as other CVDRF were considered. In the Cardiovascular Health Study, participants with high BMI in midlife had a significantly higher risk of dementia, but obesity in later life was protective against dementia [119•]. These findings suggest that a low BMI in elderly life is related to cognitive decline likely from frailty and poor nutrition [120•]. Thus, the association of obesity and cognition appears to be age-dependent and likely non-linear. More studies are needed to understand optimal weight and biological mechanisms such as the role of circulating leptins [121•].

Association of Composite Measures of CVDRF and Cognition

Several CVDRF are independently associated with cognitive decline as discussed above (Table 2). Studies have also investigated the combined effect of these risk factors on cognition. A favorable CVH profile has ample evidence to prevent CVD but could have secondary benefits for protection against cognitive decline and dementia [122, 123•]. Prior studies that have examined composite measures of CVDRF such as the Framingham risk score have found higher risk scores to be associated with worse cognitive function, markers of cognitive aging such as smaller brain volume, and a predictor of progression of MCI to Alzheimer’s [124••, 125•]. The Cardiovascular Risk factors, Aging, and Incidence of Dementia (CAIDE) risk score which was specifically developed to assess dementia risk, shares many of the same CVDRF as the other composite scores [126••]. The CVH profile particularly emphasizes modifiable CVDRF. It has been used in the Framingham Heart Study Offspring cohort where ideal levels of the metric had significant association with stroke, vascular dementia, frontal brain atrophy, and cognitive decline on tasks measuring visual memory and reasoning [127•]. This is the first study to demonstrate an association between ideal CVH and incident dementia. The Maine-Syracuse Longitudinal study also showed that a better CVH profile (particularly ideal levels of smoking, diet, and physical activity) was associated with superior neuropsychological performance across multiple cognitive domains such as visual-spatial memory, working memory, executive function, and a global composite score [128].

Table 2 Cardiovascular health (CVH) factors and association with cognition
Full size table

The association of overall CVH profile and cognition has been replicated in several ongoing cohort studies. In the Reasons for Geographic and Racial Differences in Stroke (REGARDS), ideal levels of CVH were associated with lower incident neurocognitive impairment in both Blacks and Whites regardless of US region [129•]. The Hispanic Community Health Study (HCHS/SOL) extended this finding to a population of Hispanic/Latino adults [130] and also showed that the benefits appear to be consistent across multiple domains of neurocognitive health, including episodic learning and memory, verbal fluency, and psychomotor speed. In the Coronary Artery Risk Development in Young Adults (CARDIA) study, a favorable CVH profile was similarly associated with better neurocognitive function in midlife [28•]. Similarly, in the ARIC study, higher midlife CVH scores were associated with better midlife cognition and reduced 20-year cognitive decline [31••]. In the Northern Manhattan Study, the number of ideal CVH factors was associated with less decline in the domains of processing speed, but had a weaker association with executive function and episodic memory [131•]. More recent data from a large population-based study has shown that a favorable CVH profile at a younger age is associated with lower risk of dementia in older ages [132]. Even in the more elderly population, CVH was clearly associated with incident dementia [34••]. However, some studies have provided inconsistent results with incident dementia likely from the complexity of the interactions and measurements of various CVDRF that are part of the CVH metric [133•]. Despite this, data on overall CVH and cognition remains robust.

Conclusions

The remarkable heterogeneity of Alzheimer’s disease and other forms of dementia, lack of curative treatments, and high cost of managing the disease burden, highlight the urgent need for scaling up preventive measures [134•]. The relationship of CVD and dementia is complex but accumulating evidence implicates an important role of CVDRF in the pathogenesis of cognitive decline. Timely detection and control of CVDRF in primary care can thus have a significant public health impact in the control of both CVD and dementia epidemics. In this context, the AHA concept of ideal CVH could be critical for promoting not just heart health but brain health as well. The focus of CVH is on modifiable health factors and behaviors; increasing education and improving motivation in different population subgroups can decrease the risk of cognitive decline later in life. Studies discussed in this review highlight that incorporating cardio-protective strategies particularly in early and midlife can improve patient’s CVH profile long-term and help safeguard cognitive health. Further understanding the role of CVH and cognitive decline could have an important role in developing effective population strategies and addressing health disparities in aging populations. Few such measures include further taxation on tobacco products, education about salt reduction, emphasis on MIND diet, and more opportunities for young and old to increase physical activity. The CVH-cognition link also highlights the need for effective multimodal interventions that can target multiple pathophysiological pathways at the same time. More research is required in identifying common and disparate relationships of CVDRF and cognition, understanding genetics and biological pathways, which would in turn lead to targeted interventions in reducing cognitive decline in aging populations.