Increased pulse wave velocity is related to impaired working memory and executive function in older adults with metabolic syndrome

Abstract

Age-related vascular alterations promote the pathogenesis of vascular cognitive impairment (VCI). Cardiovascular risk factors that accelerate vascular aging exacerbate VCI. Metabolic syndrome (MetS) constitutes a cluster of critical cardiovascular risk factors (abdominal obesity, hypertension, elevated triglycerides, elevated fasting glucose, reduced HDL cholesterol), which affects nearly 37% of the adult US population. The present study was designed to test the hypotheses that MetS exacerbates cognitive impairment and that arterial stiffening moderates the association between cognitive dysfunction and MetS in older adults. MetS was defined by the NCEP ATP III guidelines. Cognitive function (digit span and trail-making tests) and brachial-ankle pulse wave velocity (baPWV; a non-invasive clinical measurement of arterial stiffness) were assessed in older adults with MetS and age- and sex-matched controls. Multiple linear regression models were applied to test for the main effects of MetS, baPWV, and their interaction on cognitive performance. Fifty-three participants with MetS (age: 68 ± 8 years) and 39 age-matched individuals without MetS (age: 66 ± 9 years) were enrolled into the study. In adjusted multivariable regression analyses of the digit span backward length score, both MetS (ß = 1.97, p = 0.048) and MetS by baPWV interaction (ß =  − 0.001, p = 0.026) were significant predictors. In participants with MetS, higher baPWV was associated with poorer performance on digit span backward length score, a test of working memory (R =  − 0.44, p = 0.0012), but there was no association in those without MetS (R = 0.035, p = 0.83). MetS was negatively associated with performance on the digit span backward length score, baPWV was negatively associated with multiple neuropsychological outcomes, and baPWV moderated the association between digit span backward length score and MetS, as individuals with both MetS and higher baPWV had the most impaired cognitive function. Our findings add to the growing body of evidence that individuals with MetS and higher baPWV may be prone to VCI.

Introduction

A growing body of evidence from epidemiological, clinical, and experimental studies indicates that age-related vascular dysfunction and structural alterations critically contribute to the pathogenesis of vascular cognitive impairment (VCI) and Alzheimer’s disease [1–4]. Pathophysiological conditions that promote accelerated vascular aging increase risk for VCI and Alzheimer’s disease [5–8], decreasing quality of life in older adults [9, 10]. Understanding and targeting cardiovascular risk factors the exacerbate age-related cognitive impairment is expected to have a major role in preserving brain health in older individuals.

Metabolic syndrome (MetS) constitutes a cluster of related cardiovascular risk factors defined by the presence of any 3 of the 5 following conditions: abdominal obesity, elevated blood pressure, elevated triglycerides, elevated fasting glucose, and reduced high-density lipoprotein (HDL) cholesterol [11]. Over the past years, the prevalence of MetS has steadily increased, reaching nearly 37% of the adult US population diagnosed with this debilitating condition today [12, 13]. Recent evidence demonstrates that MetS is linked to a threefold increased risk of cardiovascular and cerebrovascular diseases and cardiovascular-related mortality [14–17]. This increase in risk is above the risk attributable by the individual components [18], indicating that the accumulation of cardiometabolic risk factors represents a significant public health concern that is increasing in prevalence.

There is evidence that MetS promotes accelerated cognitive decline [19–25]. Specifically, MetS is associated with reduced performance on tasks related to executive function [19, 26–28], a domain of cognition that encapsulates multiple processes important for the regulation of thoughts, behaviors, and complex actions [29–31]. Although several cross-sectional and longitudinal studies indicate that MetS is associated with poorer cognitive functioning, there is no clear consensus if the cluster of components that defines MetS uniquely explains the decline in cognitive function beyond the contribution of the individual component(s) [21, 28, 32–34]. This may be because some previous reports used differing criteria for MetS, failed to control for important covariates (i.e., disease burden, education, etc.), and utilized single item assessments of cognitive function [35].

The mechanisms by which MetS impairs cognitive function are not completely understood. The cardiovascular risk factors constituting MetS associate with multifaceted structural and functional vascular abnormalities, which were shown to exacerbate arterial stiffness. Increased arterial stiffness is known to impose significant hemodynamic burden on the cerebral circulation, contributing to the pathogenesis of various forms of dementia. However, the role of increased arterial stiffness in the genesis of cognitive impairment associated with MetS is yet to be fully understood [36].

The present study was designed to test the hypothesis that older individuals with MetS exhibit decreased cognitive performance when compared with those without MetS, and that arterial stiffening moderates the association between cognitive dysfunction and MetS. To test this hypothesis, we enrolled older adults with MetS and age-matched controls and assessed cognitive function (digit span and trail-making tests) and brachial-ankle pulse wave velocity (baPWV; a proxy measurement for arterial stiffness). Multiple linear regression modeling was applied to test the main effects of MetS, baPWV, and their interaction on cognitive performance.

Methods

Participants

Approval and informed consent

The procedures of this study were approved by the Penn State College of Medicine institutional review board, and written informed consent was obtained from each patient prior to beginning the investigation.

Recruitment

Participants were recruited from flyers posted throughout The Milton S. Hershey Medical Center and from local newspaper advertisements for possible enrollment into the study. Participants were evaluated in the Clinical Research Center and were screened on inclusion and exclusion criteria that have been used previously [37], and which are listed below.

Inclusion and exclusion criteria

Participants were included in this study if they met the following criteria: (a) age ≥ 50 years and (b) ambulatory without the need of an assistive device. Participants were excluded for the following conditions: (a) age < 50 years, (b) non-ambulatory, (c) neurological diseases (including but not limited to Parkinson’s disease, Alzheimer’s disease, multiple sclerosis, amyotrophic lateral sclerosis), (d) active cancer, and (e) stage 5 chronic kidney disease (end stage), as defined by an estimated glomerular filtration rate < 15 ml/min per 1.73 m [38].

Medical screening

Participants arrived at the Clinical Research Center in the morning fasted but were permitted to take their usual medications. Vital signs, demographic information, education level, height, weight, body mass index, waist circumference, and ankle/brachial index were obtained according to standard guidelines [39]. Participants had blood samples drawn, which were then sent to a central laboratory for analyses for a fasting complete metabolic panel, a lipid panel, and insulin.

Participants then underwent a medical history and physical examination by study physicians, in which comorbid conditions, cardiovascular risk factors, and current medications were recorded. Based on this battery of baseline assessments, participants were coded on cardiovascular risk factors according to standard definitions for hypertension, dyslipidemia, diabetes mellitus, obesity, abdominal obesity, and MetS [11]. Additionally, coronary artery disease, cerebrovascular disease, peripheral artery disease, and chronic kidney disease [37] were coded according to standard definitions, as previously described [40]. Participants were characterized on the history and symptoms of arthritis, and chronic obstructive pulmonary disease. The physical examination concluded with the assessment of peripheral neuropathy, as a 10 g Semmes–Weinstein monofilament was applied to 10 sites on each foot [41, 42], and a C 128-Hz tuning fork was applied to the big toe, first metatarsal joint, ankle, and knee [43] to assess vibration perception. Peripheral neuropathy was recorded if the participants did not correctly perceive the application of the monofilament or the disappearance of the vibration of the tuning fork at the various locations.

Assessment of body composition

Body fat percentage was obtained using a model D1000-3 eight-electrode bio-electric impedance device (Rice Lake Weighing Systems, Rice Lake, WI) while subjects stood barefoot on two stainless-steel rectangular foot-pad electrodes on the base of the machine and held hand grip electrodes [44, 45].

MetS group classification

According to the National Cholesterol Education Program (NCEP) Adult Treatment Panel (ATP) III [46–48], MetS is defined as having three or more of the following components: (1) abdominal obesity (waist circumference > 102 cm in men and > 88 cm in women), (2) elevated triglycerides (≥ 150 mg/dl), (3) reduced HDL cholesterol (< 40 mg/dl in men and < 50 mg/dl in women), (4) elevated blood pressure (≥ 130/85 mmHg), and (5) elevated fasting glucose (≥ 100 mg/dl) as well as those with diabetes mellitus. Although several definitions of MetS exist, the NCEP ATP III definition was used in this investigation because it was specifically established on a population from the USA. In the total group of 92 patients, 53 patients (58%) screened positive for MetS, whereas the remaining 39 patients (42%) screened negative.

Neuropsychological testing

Participants were assessed using neuropsychological test battery as previously described [49]. Tests were selected for previously demonstrated sensitivity to vascular diseases including hypertension, PAD, and/or stroke, and to briefly assess the following domains of cognitive function: attention, working memory, perceptuo-motor speed, and executive function.

Attention and working memory were assessed by standard administration of the Digits Forward and Digits Backward portions of the Wechsler Adult Intelligence Scale – Revised, respectively [49]. Digits Forward and Digits Backward test performance were measured both as total score (total trials correct out of 14) and as best length score (highest string of numbers reached, i.e., 7 numbers in length). Trail making test parts A and B were performed to assess perceptuo-motor speed and executive function. Part A of the trail making test requires participants to draw a line connecting randomly arrayed, consecutively numbered circles as quickly as possible. In part B, participants draw a line connecting consecutively numbered and lettered circles as quickly as possible by alternating between numbers and letters (i.e., 1-A-2-B-3). In addition to perceptuo-motor speed, this test assesses mental flexibility which is a key dimension of executive function [50].

The Digit Span tests and the Trail Making tests are part of the neuropsychological tests from the Uniform Data Set (UDS) and were completed by participants in this study. The obtained scores were converted into normative z-scores using a calculator based on linear regression models to better interpret performance at a population level, by taking into consideration participant age, sex, and education level [51]. Use of this calculator has been widely adopted to calculate normative z-scores across various and diverse settings.

Assessment of systemic arterial pulse wave velocity (baPWV)

Measurements of baPWV were completed using a computerized vascular profiling machine (VP-1000, Omron Colin Medical). The VP-1000 device simultaneously measures an electrocardiogram, phonogram, and oscillometric bilateral brachial and ankle blood pressures [52]. Electrocardiographic electrodes were placed on the wrists, and a microphone for the phonocardiogram was placed over the left chest to provide timing markers. Distances between measurement sites were derived from participant data entered into the machine and on-board algorithms for body segment lengths [53]. Automated baPWV was measured in participants after 10 min of supine rest. The measurement of baPWV is highly reliable with repeat testing (coefficient of variation = 3.8%) [54].

Statistical analyses

Summary statistics including mean and standard deviations for continuous variables, and frequency with percentage for categorical variables are reported. The normality assumption for continuous variables was assessed using Shapiro–Wilk tests. P values for two group comparisons were obtained based on two-sample t-tests or Wilcoxon rank sum tests for continuous variables, and Pearson’s chi-square tests or Fisher’s exact tests for categorical variables, as appropriate. Correlation analysis between baPWV and neuropsychological outcomes for overall and stratified by group were performed with both Pearson correlation and Spearman’s rank correlation coefficients. Further, to address our hypothesis that arterial stiffening, as measured by baPWV, moderates the association between cognitive dysfunction and MetS in older adults, multivariable regressions were fitted to evaluate their relationship, after adjusting for standard demographic data (age, sex, race, and education), chronic obstructive pulmonary disease because of its association with cognitive function [55, 56], and potential confounding variables that were shown to have significant group differences in Table 1 (ankle/brachial index and prevalence of cardiovascular disease). Other variables in Table 1 that were significantly different between the two groups were not considered as covariates because they are either factors involved in defining MetS (blood pressure and prevalence of hypertension and dyslipidemia), or they are highly correlated with MetS (weight, body mass index, body fat percentage, and prevalence of obesity). Consequently, to avoid the collinearity issue in regression analysis, we did not consider these variables for adjustment. All hypothesis tests were two-sided with the significance level of 0.05. All data analyses were performed using R and SAS version 9.4.

Table 1.

Clinical characteristics of participants with and without metabolic syndrome (MetS)

Variables	Entire group (n = 92)	Control group (n = 39)	MetS group (n = 53)	P value
Age (years)	67 ± 8	66 ± 9	68 ± 8	0.36
Weight (kg)	75.5 ± 14.8	68.6 ± 10.7	80.5 ± 15.5	< 0.001
Height (cm)	165.8 ± 9.5	164.7 ± 6.8	166.7 ± 11.0	0.28
Body mass index (kg/m2)	27.4 ± 4.3	25.3 ± 3.6	28.9 ± 4.2	< 0.001
Body fat percentage (%)	33.8 ± 8.6	31.7 ± 7.0	35.2 ± 9.4	0.045
Ankle/brachial index	1.10 ± 0.17	1.15 ± 0.11	1.06 ± 0.20	0.007
Pulse wave velocity (cm/s)	1690 ± 414	1607 ± 399	1753 ± 417	0.093
Systolic pressure (mm Hg)	134 ± 17	128 ± 18	139 ± 15	0.005
Diastolic pressure (mm Hg)	76 ± 9	75 ± 10	77 ± 8	0.198
Pulse pressure (mm Hg)	58 ± 14	54 ± 13	61 ± 14	0.007
Sex (men)	33 (36)	8 (21)	25 (47)	0.016
Race (Caucasian)	84 (91)	36 (92)	48 (91)	1.0
Education (college graduate)	57 (62)	25 (64)	32 (60)	0.88
Coronary artery disease	13 (14)	1 (3)	12 (23)	0.015
Cerebrovascular disease	4 (4)	0 (0)	4 (8)	0.21
Peripheral artery disease	11 (12)	2 (5)	9 (17)	0.16
Chronic kidney disease	9 (10)	4 (10)	5 (9)	1.0
Peripheral neuropathy	36 (39)	11 (28)	25 (47)	0.10
Current/previous smoking	31 (34)	10 (26)	21 (40)	0.26
Hypertension	71 (77)	23 (59)	48 (91)	< 0.001
Dyslipidemia	73 (79)	21 (54)	52 (98)	< 0.001
Diabetes mellitus	5 (5)	0 (0)	5 (9)	0.13
Obesity	25 (27)	5 (13)	20 (38)	0.016
Arthritis	33 (36)	11 (28)	22 (42)	0.24
Chronic obstructive pulmonary disease	9 (10)	1 (3)	8 (15)	0.10

Data are presented as means ± SD or n (%)

Results

Descriptive study participant characteristics are shown in Table 1. In summary, the MetS group had a higher body mass index (p < 0.001), higher body fat percentage (p = 0.045), lower ankle/brachial index (p = 0.007), and greater percentage of men (p = 0.016), and prevalence of coronary artery disease (p = 0.015), hypertension (p < 0.001), dyslipidemia (p < 0.001), and obesity (p = 0.016), when compared to the control group. With respect to the prevalence of MetS components, the MetS group had significantly greater prevalence of abdominal obesity (p < 0.001), elevated triglycerides (p < 0.001), decreased HDL cholesterol (p < 0.001), elevated blood pressure (p = 0.002), and elevated plasma glucose (p = 0.005), compared to the control group (Table 2).

Table 2.

Prevalence of each metabolic syndrome (MetS) component in participants with and without MetS

Variables	Control group (n = 39)	MetS group (n = 53)	P valuea
Abdominal obesity	12 (31)	42 (79)	< 0.001
Elevated triglycerides	2 (5)	47 (89)	< 0.001
Decreased high-density lipoprotein	1 (3)	50 (94)	< 0.001
Elevated blood pressure	24 (62)	48 (91)	0.002
Elevated plasma glucose	6 (15)	24 (45)	0.005

Data are presented as n (%)

^aP value based on chi-square tests or Fisher’s exact tests, as appropriate

The mean unadjusted memory and executive function task performances for each group are presented in Table 3. There were no significant differences between groups in unadjusted neuropsychological performance tests, except for the z-score analysis of the digit span forward length score (p = 0.02).

Table 3.

Unadjusted neuropsychological performance of participants with and without metabolic syndrome (MetS)

	Raw scores			Normative z-scores (unadjusted)
Variables	Control group (n = 39)	MetS group (n = 53)	P valuea	Control group (n = 39)	MetS group (n = 53)	P valueb
Digit span forward score	10.4 ± 2.0	9.7 ± 2.2	0.16	0.85 ± 0.94	0.54 ± 1.07	0.16
Digit span backward score	8.2 ± 2.7	7.5 ± 2.1	0.19	0.59 ± 1.21	0.33 ± 1.05	0.28
Digit span forward length score	8.0 ± 1.0	7.5 ± 1.3	0.14	1.16 ± 0.86	0.62 ± 1.14	0.02
Digit span backward length score	5.9 ± 1.4	5.5 ± 1.2	0.19	0.68 ± 1.11	0.40 ± 0.97	0.20
Trail making A (s)a	27.8 ± 7.9	31.6 ± 14.2	0.21	0.44 ± 0.51	0.20 ± 0.94	0.15
Trail making B (s)a	89.9 ± 56.1	94.7 ± 51.3	0.66	0.01 ± 1.11	− 0.11 ± 0.98	0.59

Data are presented as means ± SD

^aP value based on Wilcoxon rank sum tests

^bP value based on Mann–Whitney tests

The scatterplot relations between baPWV and unadjusted neuropsychological outcome z-scores in all participants (Fig. 1) demonstrate a significant correlation between baPWV and digit span forward score (R =  − 0.28, p = 0.008; Fig. 1A), digit span backward score (R =  − 0.29, p = 0.005; Fig. 1C), digit span backward length score (R =  − 0.24, p = 0.021; Fig. 1D), trail making test A (R =  − 0.23, p = 0.029; Fig. 1E), and trail making test B performance (R =  − 0.40, p = 0.0001; Fig. 1F). The same significance results were achieved based on Spearman’s correlation. Examination of the relations between pulse pressure and unadjusted neuropsychological outcome z-scores in all participants combined demonstrated similar significant associations as those found for baPWV (Table 4). Within the MetS group pulse pressure was associated with digit span backward score (− 0.35, p = 0.011) and digit span backward length score (− 0.42, p = 0.002), whereas pulse pressure was not significantly associated with any of the variables in the control group (Table 4).

Fig. 1.

Correlations between brachial-to-ankle pulse wave velocity (baPWV) and neuropsychological outcomes z-scores^a in older adults. Digit span forward score (A); digit span forward length score (B); digit span backward score (C); digit span backward length score (D); trail making test A (E); trail making test B (F). ^aR(Pearson) and rho(spearman) are both displayed

Table 4.

Association between pulse pressure and unadjusted normative z-scores of neuropsychological performance

Variables	Entire group (n = 92)	Control group (n = 39)	MetS group (n = 53)
Digit span forward score	− 0.26*	− 0.17	− 0.26
Digit span backward score	− 0.30**	− 0.19	− 0.35*
Digit span forward length score	− 0.18	0.08	− 0.21
Digit span backward length score	− 0.31**	− 0.12	− 0.42**
Trail making A	− 0.15	− 0.2	− 0.08
Trail making B	− 0.28**	− 0.31	− 0.25

Data are Pearson correlation coefficients

^*p < 0.05; **p < 0.01

Table 5 shows the result of the multivariable regression model for digit span backward length score. Both MetS (ß = 1.97, p = 0.048) and MetS by baPWV interaction (ß =  − 0.001, p = 0.026) were significant predictors. The significant interaction indicates that baPWV moderated the relation between digit span backward length score and MetS (Table 5). Thus, only for the MetS group, higher baPWV was related to poorer scores for the digit span backward length score (R =  − 0.44, p = 0.0012) (Fig. 2). In contrast, in the control group, baPWV was not related to the digit span backward length score (R = 0.035, p = 0.83). No other multivariable regression model showed significant interaction effects between baPWV and MetS (data not shown) for the remaining neuropsychological outcomes, consisting of digit span forward score, digit span forward length score, digit span backward score, trail making test A, and trail making test B.

Table 5.

Multivariable regression model for digit span backward length score

Measure	Coefficients	Estimate	Std error	P value
Digit span backward length score	Intercept	1.424	1.520	0.35
	MetS	1.966	0.976	0.048
	baPWV	0.001	0.001	0.37
	Age	− 0.027	0.017	0.11
	ABI	0.717	0.780	0.36
	Sex (= male)	− 0.016	0.252	0.95
	Race (= Caucasian)	− 0.025	0.403	0.54
	Education(= college grad or higher)	− 0.023	0.251	0.36
	CAD	− 0.166	0.353	0.64
	COPD	0.165	0.404	0.68
	MetS × baPWV	− 0.001	0.001	0.026

MetS metabolic syndrome, baPWV brachial-ankle pulse wave velocity, ABI ankle-brachial index, CAD coronary artery disease, COPD chronic obstructive pulmonary disease

Fig. 2.

Correlations between brachial-to-ankle pulse wave velocity (baPWV) and digit span backward length score^a in older adults with or without metabolic syndrome (MetS). A Correlation for control participants without MetS. B Correlation for participants with MetS. ^aR(Pearson) and rho(spearman) are both displayed

Discussion

The main findings of this study were (1) MetS was negatively associated with working memory performance (assessed by the digit span backwards test), (2) baPWV was negatively associated with multiple neuropsychological outcomes, and (3) baPWV moderated the association between working memory performance and MetS in older adults.

To our knowledge, this is the first study to examine the association between MetS and cognition while accounting for changes in baPWV. We found that MetS was a significant predictor of working memory performance (assessed by the digit span backwards test) in a multivariate model, with poorer performance seen in those with MetS. The digit span backward length score is a sub score of the digit span backward test [57], and its changes are interpreted to indicate impaired working memory and/or executive function [58]. Individuals with MetS exhibit worse digit span backward length score, extending findings of previous reports [26, 28, 33]. For example, a study performed in 197 midlife adults showed that MetS was associated with lower performance in tests of executive function and memory from the Wechsler Abbreviated Tests of Intelligence (WASI-II), including the digit span tests and trail making tests [26]. Additionally, in older individuals, MetS was associated with poor performance across different domains of executive function [33]. Lastly, individuals with MetS performed worse on tests of executive function, including trail making test B and clock drawing test [28]. We confirmed that the negative impact of MetS on the aforementioned cognitive domains is independent of age, demographic variables, and presence of cardiovascular or pulmonary diseases.

Short-term working memory relies significantly on the prefrontal cortex, which focuses attention on the relevant sensory representation, selects information and performs executive function, which are needed for regulation of cognitive processing of the information [59–61]. The anterior cingulate cortex is connected to the prefrontal cortex and is implicated in sustained attention. There is growing evidence that function of the aforementioned brain regions is critically sensitive to microvascular pathophysiological alterations decreasing regional cerebral blood supply. Importantly, aging was shown to impair blood flow both to the prefrontal cortex [62, 63] and the anterior cingulate cortex [64, 65] which negatively impacts cognitive function regulated by these brain regions. It is generally considered that components of the MetS induce accelerated (micro)vascular aging [66–72]. Thus, future studies should determine how MetS in older adults affect microvascular regulation of cerebral blood supply, including the dynamics and magnitude of neurovascular coupling responses.

The results from the current study showed that higher baPWV was related to poorer performance across the digits span tests and trail making tests, which agrees with previous findings across different populations [73, 74]. In cognitively healthy adults, increases in baPWV by 1 m/s was inversely related to tasks of scanning and tracking, which fall under the domain of executive functioning [1]. Additionally, the combination of old age and high baPWV was associated with the lowest level of cognitive performance in this cohort. In the Baltimore longitudinal Study of Aging, individuals with higher baseline baPWV exhibited a greater decline in performance on tests of verbal and working memory [73]. In the current study, after adjusting for several known risk factors for cognitive decline, individuals with MetS who had higher baPWV had the worst cognitive function. Increased baPWV is clinically interpreted as a measure of increased arterial stiffness. The mechanisms by which increased arterial stiffness exacerbates the deleterious effects of MetS on the cerebral circulation are likely multifaceted. Arterial stiffness results in increased pulse pressure and increased penetration of the systolic pressure wave into the vulnerable, distal portion of the microcirculation, promoting microvascular damage. Due to this consideration, increased baPWV can be used for assessment of end-organ damage in hypertensive patients [75]. In the aged brain, penetration of increased pulsatile pressure to the cerebral microcirculation is thought to contribute to blood–brain barrier disruption, genesis of cerebral microhemorrhages and microvascular rarefaction, all of which may promote cognitive dysfunction [3, 71, 76]. The myogenic response of proximal cerebral arteries has a critical role in neuroprotection by preventing the penetration of pulsatile pressure to the thin-walled distal portion of the microcirculation, thereby protecting the microcirculation from high pressure-induced damage [66, 77, 78]. Importantly, experimental studies show that aging [66, 79] and components of MetS, including obesity [66, 67] impair these autoregulatory protective mechanisms, which likely exacerbate microvascular damage induced by transient increases in arterial pulsatile pressure (e.g., during Valsalva maneuvers). Future studies should determine how synergistic interaction of old age, MetS and arterial stiffening promote microvascular damage, including blood brain barrier disruption, microhemorrhages, and other radiological signs of small vessel disease.

Methodological considerations

Carotid-femoral PWV (cfPWV) is the gold standard for assessing PWV and has strong prognostic value in CVD development and mortality [80]. In this study, baPWV was determined, which is an emerging and well-accepted non-invasive technique of PWV that is easily obtained in the clinical research setting. BaPWV has also shown predictive value for cardiovascular mortality and morbidity [81], shown strong positive correlations against measurement of cfPWV (r = 0.73 [82], and is largely determined by the stiffness of the elastic central arteries similar to cfPWV [83]. For these reasons, baPWV is a clinical measurement that is suitable for the assessment of systemic arterial stiffness.

Limitations

There are several limitations to this study. The MetS group consistently scored lower on all cognitive outcomes compared to non-MetS group, but only the digit span backward length score was significantly different. Lack of group differences might have been due to the non-MetS group also having clinically significant cardiovascular risk factor burden, including hypertension. It is likely that the presence of these conditions in the non-MetS group also impact cognitive function. However, this observation increases the generalizability of our study because the prevalence of cardiovascular risk factors is relatively high in older adults. Additionally, a larger sample may have provided increased power to determine the effects of baPWV and MetS on cognitive function. Another important limitation is that age-related vascular changes were only determined by systemic baPWV. Additional measurements such as aortic baPWV, carotid compliance, or microvascular functional measurements may have provided different insight into the relation between MetS and cognitive function. Another limitation is that because only two neurocognitive assessments were utilized (digits span tests and trail making tests), it is unclear whether performance on other neurocognitive tests that assess different domains of cognition (learning, subdomains of executive function, verbal memory) might be affected by MetS. However, the use of these two cognitive tests provided a broad scope of cognitive function, as they assessed short-term memory, working memory, attention, and executive function which are regulated in two general regions of the brain. Lastly, due to the cross-sectional nature of this study, causal relations cannot be established, and future studies should utilize randomized controlled trials to determine the mechanisms and directionality regarding the associations observed in this study.

Conclusion

In conclusion, increased brachial-ankle pulse wave velocity is related to impaired working memory and executive function in older adults with MetS. Our findings add to the growing evidence that MetS promotes the pathogenesis of vascular cognitive impairment, by impairing the function of the cerebral microcirculation [2, 84–86]. Since MetS is a condition comprised of modifiable risk factors, interventions designed to improve cardiovascular risk factors and promote cerebrovascular health may also be a clinical approach to improve cognitive functioning in individuals with MetS.

Author contribution

All authors have read and approved the manuscript.

Funding

This work was supported by grants from the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1 TR002014, the American Heart Association, the National Institute on Aging (RF1AG072295, R01AG055395, R01AG068295; R01AG070915, K01AG073614), the National Institute of Neurological Disorders and Stroke (R01NS100782), the National Cancer Institute (R01CA255840), and the Cellular and Molecular GeroScience CoBRE (P20GM125528). The funding sources had no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, the American Heart Association, or the Presbyterian Health Foundation.

Declarations

Conflict of interest

Dr. Anna Csiszar and Dr. Andriy Yabluchanskiy serve as Associate Editors for GeroScience. Dr. Zoltan Ungvari serves as Editor-in-Chief for GeroScience and as Consulting Editor for The American Journal of Physiology-Heart and Circulatory Physiology.