The Relationships Between Components of Metabolic Syndrome and Mild Cognitive Impairment Subtypes: A Cross-Sectional Study of Japanese Older Adults

Abstract

Background:

The associations between components of metabolic syndrome (MetS) and mild cognitive impairment (MCI) subtypes remain unclear.

Objective:

The study aim was to identify the prevalence of MetS for MCI subtypes and to investigate sex differences in the association between MetS and MCI subtypes in older Japanese adults.

Methods:

The study analyzed data from 3,312 men and women aged 70 years or more. MetS was diagnosed according to International Diabetes Federation criteria. Participants completed cognitive tests and were categorized into normal cognition, amnestic MCI (aMCI), and non-amnestic MCI (naMCI). The associations between MetS and its components and MCI subtypes were analyzed using multiple logistic regression.

Results:

MetS prevalence was greater in participants with naMCI (men: p = 0.030; women: p = 0.040). Participants with naMCI showed higher odds ratios (OR) of MetS (men: 2.45, 95% confidence intervals (CI): 1.13–5.32; women: OR: 1.94, 95% CI: 1.12–3.39) compared with participants with normal cognition. MetS was not associated with aMCI. Analysis of MetS components showed that raised glucose (OR: 1.62, 95% CI: 1.19–2.22) and reduced high-density lipoprotein cholesterol (OR: 1.97, 95% CI: 1.25–3.12) were associated with naMCI in men. In women, raised blood pressure (OR: 1.42, 95% CI: 1.03–1.94) and raised glucose (OR: 1.32, 95% CI: 1.02–1.71) were associated with naMCI.

Conclusion:

MetS was associated only with naMCI regardless of sex, which suggests etiologic differences in MCI subtypes. We also found sex differences in the relationship between naMCI risk and MetS and its components.

Keywords

Elderly population metabolic syndrome mild cognitive impairment sex differences

INTRODUCTION

Metabolic syndrome (MetS) is a complex disorder that is associated with increased risk of cardiovascular disease and diabetes mellitus. MetS characteristics include 1) central obesity, 2) elevated blood pressure, 3) elevated triglycerides (TG), 4) lowered high-density lipoprotein cholesterol (HDL-C), and 5) insulin resistance [1, 2]. Recent studies have shown that MetS is associated with mild cognitive impairment (MCI) [3], development of vascular dementia (VD) [4], and Alzheimer’s disease (AD) [5]. Because MCI is a transitional state between normal aging and dementia, the identification of modifiable risk factors for MCI at an early stage is particularly important for dementia prevention strategies.

MCI has been divided into two main subtypes: amnestic MCI (aMCI) and non-amnestic MCI (naMCI). aMCI is a symptomatic predementia phase of AD and is characterized by memory impairment, whereas naMCI most likely reflects neurodegenerative or nondegenerative conditions, such as vascular or psychiatric disorders. Thus, naMCI is characterized by more preservation of memory function with greater degrees of impairment in cognitive domains, such as attention/executive and visuospatial functions [6, 7].

A few studies have investigated the relationship between MetS and MCI subtypes [8, 9], but their conclusions are conflicting. One cross-sectional, population-based study found no significant association between MetS and MCI or aMCI, and only a combination of MetS and high levels of inflammation (i.e., C-reactive protein) were significantly associated with naMCI [9]. However, another study found that MetS was associated with aMCI, especially in individuals who showed APOE ɛ4 (a genetic risk factor for AD) at a younger age in this middle-aged and older cohort [8]. Possible reasons for such conflicting results are differences in age, diagnostic criteria for MCI and MetS, and lifestyles of different study populations. To our knowledge, there are no reports of the role of MetS in different MCI subtypes in the Japanese population.

Previous studies have reported sex differences in incidence and prevalence of MCI, with higher estimates for men [10, 11]. Some studies have reported sex differences in brain aging and inflammatory markers that may partly explain the sex difference in the prevalence of MCI [12, 13]. These suggest sex differences in the risk factors for overall MCI or its subtypes and highlight the need to identify modifiable factors that have a differential effect on MCI risk in men and women. However, sexual dimorphism in the association between MCI subtypes and MetS has not been widely investigated. As older Japanese people have different genetic backgrounds and lifestyle characteristics to older Western people, the effect of sex on the association between MetS and MCI needs to be investigated in the older Japanese population.

Therefore, the aim of this study was to estimate the prevalence of MetS and its components in MCI subtypes, to investigate the different associations between MetS and MCI subtypes, and to examine sex differences in these associations in the Japanese population, using a community-based survey.

MATERIALS AND METHODS

Study population

The present study included 5,278 community-dwelling older adults who were enrolled in the National Center for Geriatrics and Gerontology Study of Geriatric Syndrome [14]. All participants were aged 70 years or more at January 1, 2013, lived in Midori-Ward of Nagoya city or Obu city, Aichi prefecture, Japan, and were without long-term care needs or support. We excluded participants with a history of dementia (n = 26) or Parkinson’s disease (n = 13) and those with Mini-Mental State Examination (MMSE) [15] scores <24 (n = 682). We also excluded participants who had missing data for MCI assessment (n = 177) or missing blood samples (n = 786) or missing data for MetS assessment (n = 282). The remaining 3,312 participants of average age 75.7±4.2 years (women 55.8%) were included in the final analysis. The study protocol was developed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the National Center for Geriatrics and Gerontology. Informed consent was obtained from all participants prior to their participation in the study.

MCI criteria and evaluation of cognitive function

The following MCI criteria were used in this study [16 –19]: 1) objective cognitive impairment (indicated by an age- and education-adjusted score at least 1.5 standard deviations (SD) below the reference threshold in one or more specific cognitive domains including memory, attention and executive function, and processing speed, all of which are commonly used for detailed neuropsychological assessments); 2) normal general cognitive functioning (scored ≥24/30 on the MMSE); 3) no evidence of functional dependency (no need for supervision or external assistance in performing basic activities of daily life); and 4) no dementia. Based on cognitive test results, participants who suffered from cognitive decline in the memory domain were classified as having aMCI, whereas those who had no memory impairment but cognitive decline in the nonmemory domain were classified as having naMCI.

The MCI criterion of objective cognitive impairment was assessed for multiple domains using the National Center for Geriatrics and Gerontology-Functional Assessment Tool [20]. This tool consists of four cognitive domains, including to memory (word list memory-I (immediate recognition); word list memory-II (delayed recall)), attention and executive function (an electronic tablet version of the Trail Making Test, TMT-part A and B), and processing speed (an electric tablet version of the symbol digit substitution test). All tests had established standardized thresholds for the definition of MCI in the corresponding domain (score <1.5 SDs below the age and education-specific means based on our own algorithm from a database including more than 10,000 community-dwelling older Japanese individuals) derived from a population-based cohort consisting of community-dwelling older adults [19].

Definition of metabolic syndrome

We defined MetS according to the International Diabetes Federation criteria [21] as consisting of central obesity (waist circumference for Japanese population ≥90 cm in men and ≥80 cm in women) plus any two or more of these four components: serum TG ≥150 mg/dL or specific treatment for this lipid abnormality; HDL-C <40 mg/dL in men and <50 mg/dL in women or specific treatment for this lipid abnormality; systolic blood pressure ≥130 mmHg and/or diastolic blood pressure ≥85 mmHg or treatment for previously diagnosed hypertension; and fasting plasma glucose ≥100 mg/dL or treatment for previously diagnosed type 2 diabetes mellitus. Glucose, TG, and HDL-C were measured from blood samples, which were collected between 10 a.m. and 4 p.m. in the fasted state. The blood samples were kept at room temperature for 30 min to allow for clotting, then the samples were centrifuged at 1,000× g for 15 min. Samples were harvested and stored at –25°C until analysis, and then were analyzed for glucose, TG, and HDL-C levels using standard laboratory techniques. Systolic and diastolic blood pressures were measured using a standard sphygmomanometer in the sitting position after a 5-min rest. Waist circumference was measured in centimeters at the midpoint between the lowest rib margin and the top of the iliac crest at minimal respiration to the closest 0.1 cm.

Covariates

Participants were interviewed about their medical history and medication use. They were also asked about their age, sex, educational history, and smoking status (never and previous versus current). In addition, their scores on the Geriatric Depression Scale (GDS) [22] were recorded. A total score of ≥6 points was considered to indicate depressive symptoms. To assess frequency of physical exercise, participants were asked, “How many days per week do you do physical exercise?” Participants were categorized into those who exercised ≤1 day, 2–4 days, and ≥5 days a week. Body mass index (BMI) was calculated as body weight (kg) divided by the square of body height (m²). BMI data were categorized into underweight (<18.5 kg/m²), normal (18.5–24.9 kg/m²), or overweight or obese (≥25 kg/m²) [23]. All potential covariates were adjusted as categorical variables in multivariate logistic regression analysis. Age (70–74 y, 75–79 y, ≥80 y), BMI (<18.5 kg/m², 18.5–24.9 kg/m², ≥25 kg/m²), education year (≤9 y, 10–12 y, ≥13 y), physical exercise (≤1 d/week, 2–4 d/week, ≥5 d/week), current smoking (yes or no), GDS (≤5 or ≥6), depression (yes or no), stroke (yes or no), and heart disease (yes or no).

Statistical analysis

All of the following analyses were conducted stratified by sex. Two-sample t tests and the chi-square (χ ²) test were used to examine differences in continuous and categorical variables and to compare sociodemographic and clinical characteristics between participants with and without MetS. The relationship between the prevalence of MetS or its components and cognitive status (normal cognition, aMCI, naMCI) were analyzed using χ ² tests, and the statistical significance of cells in the tables was examined using residual analysis. The cells were considered to contain significantly more people than expected when the adjusted standardized residual values were greater than 1.96, whereas the cells were considered to contain significantly fewer people than expected when the values were lower than –1.96. We performed multivariate logistic regression analysis to estimate the odds ratios (OR) and 95% confidence intervals (CI) of MCI subtypes for MetS and its components. We then constructed models to adjust for potential confounding factors (including age, education, GDS scores, BMI, physical exercise, current smoking, stroke, heart disease, depression) that have shown associations with cognitive function and MetS in previous studies [8, 24]. Analyses were conducted using the IBM SPSS Statistics software package (version 23.0; SPSS Inc., Chicago, IL, USA). Statistical significance was set a priori at p < 0.05.

RESULTS

Table 1 shows the demographic and clinical characteristics of participants with or without MetS by sex. The prevalence of MetS was 24.1% among men and 48.9% among women. Men and women with MetS had higher BMI (both p < 0.001), higher frequency of hypertension (both p < 0.001), a higher frequency of diabetes mellitus (both p < 0.001), and a higher frequency of hyperlipidemia (both p < 0.001) than those without MetS. Women with a lower educational level (p < 0.001) and who engaged in less frequent physical exercise (p = 0.033) had a higher rate of MetS than those with a higher education level and those who engaged in more frequent physical exercise.

Table 1

Characteristics of participants with MetS and those without MetS by sex

	Men (n = 1,463)			Women (n = 1,849)
	With MetS	Without MetS	p	With MetS	Without MetS	p
	(n = 353)	(n = 1,110)		(n = 904)	(n = 945)
Age, y	76.0 (4.2)	75.8 (4.4)	0.374^a	75.8 (4.0)	75.5 (4.2)	0.147^a
BMI, kg/m²	25.9 (2.1)	22.4 (2.3)	<0.001^a	24.4 (2.7)	20.9 (2.6)	<0.001^a
Education, n (%)			0.525^b			0.001^b
≤9 y	71 (20.1)	221 (19.9)		279 (30.9)^a	219 (23.2)
10–12 y	147 (41.6)	429 (38.6)		425 (47.0)	488 (51.6)^a
≥13 y	135 (38.2)	460 (41.4)		200 (22.1)	238 (25.2)
Physical exercise, n (%)			0.099^b			0.033^b
≤1 d/week	249 (70.5)	736 (66.4)		671 (74.3)^a	659 (69.7)
2–4 d/week	80 (22.7)	256 (23.1)		178 (19.7)	234 (24.8)^a
≥5 d/week	24 (6.8)	117 (10.6)		54 (6.0)	52 (5.5)
Current smoking, n = yes (%)	42 (11.9)	125 (11.3)	0.743^b	17 (1.9)	14 (1.5)	0.504^b
GDS ≥6, n = yes (%)	47 (13.4)	160 (14.6)	0.595^b	134 (14.9)	131 (13.9)	0.555^b
Systolic blood pressure, mmHg	142.7 (18.4)	137.0 (20.3)	<0.001^a	139.1 (19.8)	134.0 (22.0)	<0.001^a
Diastolic blood pressure, mmHg	80.0 (9.2)	77.0 (9.9)	<0.001^a	76.4 (9.5)	74.8 (10.0)	<0.001^a
Abdominal circumference, cm	94.4 (5.3)	82.8 (6.3)	<0.001^a	88.2 (6.7)	76.6 (7.7)	<0.001^a
Glucose, mg/dL	110.2 (40.6)	103.4 (35.0)	<0.001^a	102.9 (29.0)	95.8 (26.4)	<0.001^a
Triglyceride, mg/dL	196.3 (105.2)	137.4 (79.4)	<0.001^a	177.7 (99.9)	117.7 (56.8)	<0.001^a
High-density lipoprotein, mg/dL	50.0 (12.7)	58.6 (15.3)	<0.001^a	60.3 (14.8)	71.3 (16.7)	<0.001^a
Depression, n = yes (%)	11 (3.1)	24 (2.2)	0.303^b	41 (4.5)	48 (5.1)	0.585^b
Stroke, n = yes (%)	34 (9.7)	85 (7.7)	0.239^b	42 (4.7)	46 (4.9)	0.831^b
Heart disease, n = yes (%)	72 (20.4)	184 (16.6)	0.104^b	122 (13.5)	123 (13.0)	0.754^b
Hypertension, n = yes (%)	223 (63.2)^c	490 (44.1)	<0.001^b	514 (56.9)^c	319 (33.8)	<0.001^b
Diabetes mellitus, n = yes (%)	80 (22.7)^c	159 (14.3)	<0.001^b	103 (11.4)^c	46 (4.9)	<0.001^b
Hyperlipidemia, n = yes (%)	189 (53.7)^c	309 (27.9)	<0.001^b	603 (66.7)^c	292 (30.9)	<0.001^b

Values are presented as mean (SD) or n (%). MetS, metabolic syndrome; BMI, body mass index; GDS, Geriatric Depression Scale. ^a p-value obtained by Student’s t test. ^b p-value obtained by Pearson’s chi-square test. ^cStatistically significant association by adjusted standardized residual >1.96 (p < 0.05).

The prevalence estimates of MetS and its components by MCI subtypes are shown in Table 2. Of the 3,312 participants included in the analyses, 79.5% of men were classified as having normal cognition, 5.5% classified as having aMCI, and 15.0% classified as having naMCI. For women, 78.5% were classified as having normal cognition, 3.7% classified as having aMCI, and 17.8% classified as having naMCI. For men and women, the prevalence of MetS was high in participants with naMCI and lower in those with normal cognition (men: p = 0.030; women: p = 0.040). Although the frequency of raised glucose was higher for men with naMCI and lower for men with normal cognition (p < 0.001), the frequency of raised TG was lower for those with naMCI and higher for those with normal cognition (p = 0.013). For women, the frequency of raised blood pressure (p = 0.027) and raised glucose (p = 0.031) was higher for those with naMCI and lower for those with normal cognition. No significant difference was observed between the frequency of MetS components and aMCI.

Table 2

Prevalence estimates of MetS and its components by MCI subtypes

	Men (n = 1,463)				Women (n = 1,849)
	Normal	aMCI	naMCI	p ^a	Normal	aMCI	naMCI	p ^a
	(n = 1,163)	(n = 80)	(n = 220)		(n = 1,451)	(n = 69)	(n = 329)
Metabolic syndrome, n = yes (%)	264 (22.7)^c	21 (26.3)	68 (30.9)^b	0.030	687 (47.3)^c	38 (55.1)	179 (54.4)^b	0.040
Number of components				0.153				0.439
0	78 (6.7)	7 (8.8)	13 (5.9)		73 (5.0)	2 (2.9)	15 (4.6)
1	223 (19.2)	20 (25.0)	50 (22.7)		203 (14.0)	4 (5.8)	33 (10.0)
2	280 (24.1)	16 (20.0)	51 (23.2)		274 (18.9)	13 (18.8)	60 (18.2)
3	282 (24.2)	17 (21.3)	38 (17.3)		356 (24.5)	18 (26.1)	83 (25.2)
4	232 (19.9)	13 (16.3)	45 (20.5)		365 (25.2)	22 (31.9)	88 (26.7)
5	68 (5.8)	7 (8.8)	23 (10.5)		180 (12.4)	10 (14.5)	50 (15.2)
Central obesity, n = yes (%)	313 (26.9)	27 (33.8)	72 (32.7)	0.111	867 (59.8)	47 (68.1)	201 (61.1)	0.362
Raised BP, n = yes (%)	902 (77.6)	59 (73.8)	173 (78.6)	0.667	1069 (73.7)^c	57 (82.6)	262 (79.6)^b	0.026
Raised Glu, n = yes (%)	530 (45.6)^c	31 (38.8)	129 (58.6)^b	<0.001	531 (36.6)^c	29 (42.0)	148 (45.0)^b	0.015
Raised TG, n = yes (%)	673 (57.9)^b	42 (52.5)	103 (46.8)^c	0.008	913 (62.9)	47 (68.1)	217 (66.0)	0.431
Reduced HDL-C, n = yes (%)	479 (41.2)	34 (42.5)	94 (42.7)	0.897	799 (55.1)	41 (59.4)	188 (57.1)	0.640

Values are presented as n (%). MCI, mild cognitive impairment; BP, blood pressure; Glu, glucose; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol. ^a p-value obtained by Pearson’s chi-square test. ^bStatistically significant association by adjusted standardized residual >1.96 (p < 0.05). ^cStatistically significant association by adjusted standardized residual < –1.96 (p < 0.05).

Table 3 shows the results of cognitive function tests and number of impaired cognitive domain by the presence of MetS. Men and women with MetS had significantly lower scores in the attention factor compared with participants without MetS. Men with MetS had lower scores in executive function, while women with MetS had lower scores in processing speed compared with participants without MetS. There were no statistically significant differences between participants with MetS and without MetS in the number of impairment cognitive domains.

Table 3

Cognitive testing and number of impaired cognitive domain by the presence of the MetS

	Men			Women
	With MetS	Without MetS	p	With MetS	Without MetS	p
	(n = 353)	(n = 1,110)		(n = 904)	(n = 945)
Memory, score	10.5 (2.6)	10.4 (2.7)	0.716^a	10.9 (2.8)	11.1 (2.8)	0.314^a
Attention, s	23.9 (5.2)	21.8 (6.5)	0.038^a	23.4 (6.0)	21.3 (5.6)	0.034^a
Executive function, s	45.4 (16.6)	42.8 (15.4)	0.044^a	43.9 (16.5)	43.2 (16.7)	0.354^a
Processing speed, score	38.0 (6.8)	38.0 (6.8)	0.978^a	36.8 (6.8)	37.4 (6.9)	0.042^a
Number of impaired cognitive domains, n (%)			0.364^b			0.318^b
0	276 (78.2)	899 (81.0)		707 (78.2)	764 (80.8)
1	61 (17.3)	147 (13.2)		161 (17.8)	142 (15.0)
2	12 (3.4)	44 (4.0)		27 (3.0)	26 (2.8)
3	3 (0.8)	16 (1.4)		9 (1.0)	11 (1.2)
4	1 (0.3)	4 (0.4)		0 (0.0)	2 (0.2)

Values are presented as mean (SD) or n (%). ^a p-value obtained by Student’s t test. ^b p-value obtained by Pearson’s chi-square test.

Figure 1 shows the multivariate associations of MetS with aMCI and naMCI. After adjusting for demographic and clinical covariates, participants with MetS showed significantly greater odds of developing naMCI than participants with normal cognition in men and women (respectively: OR: 2.45; 95% CI: 1.13–5.32 and OR: 1.94; 95% CI: 1.12–3.39). For MetS individual components, men with raised glucose (OR: 1.62; 95% CI: 1.19–2.22) and reduced HDL-C (OR: 1.97; 95% CI: 1.25–3.12) had a higher risk of naMCI. However, raised TG was associated with lower odds of naMCI for men (OR: 0.35; 95% CI: 0.22–0.55). Women with raised blood pressure (OR: 1.42; 95% CI: 1.03–1.94) and raised glucose (OR: 1.32; 95% CI: 1.02–1.71) had a higher risk of naMCI. No association was found between MetS and/or its components and aMCI in either sex.

Fig.1

Multivariate logistic regression results showing associations between MCI subtypes and MetS and its components. All analyses were adjusted for age, education, GDS scores, BMI, physical exercise, current smoking, stroke, heart disease, depression. Participants with normal cognition were used as a reference group. aMCI, amnestic mild cognitive impairment; naMCI, non-amnestic mild cognitive impairment; MetS, metabolic syndrome; BP, blood pressure; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; CI, confidence interval.

DISCUSSION

In our community-based elderly cohort, the prevalence of MetS was higher in both men and women with naMCI than in those with normal cognition. MetS was nearly twice as likely to be associated with naMCI in both men and women after adjusting for demographic and clinical covariates; however, there was no association between MetS and aMCI. These findings suggest that the association between MetS and MCI may reflect etiologic differences in MCI subtype. However, the sample size of aMCI participants was relatively small. This limitation should be taken into consideration when interpreting the current findings. These results are consistent with previous findings that have demonstrated a positive relationship between MetS and naMCI. Roberts et al. found that a combination of MetS and high inflammation levels was significantly associated with naMCI but not with aMCI in older adults [9]. They suggest that MetS might affect cognitive impairment because of cerebral atherosclerosis and endothelial dysfunction, which are associated with the inflammatory response.

Moreover, we found that raised glucose was significantly associated with a higher prevalence of naMCI in both men and women. In addition, for men, raised glucose and reduced HDL-C were significantly associated with greater odds of naMCI. In women, raised glucose and raised blood pressure were associated with greater odds of naMCI. Ma et al. investigated the association between MCI subtypes and vascular risk factors. They found that the naMCI single domain subtype was associated with a higher prevalence of history of type 2 diabetes mellitus, stroke, and hypertension [25]. These diseases involve vascular abnormalities, such as pathological changes in the long penetrating arteries of the brain [26]. Findings from another study showed that, compared with cognitively normal subjects, subjects with naMCI had a higher frequency of elevated blood pressure [9]. A recent study has reported associations between cardiac disease (i.e., diagnosis of atrial fibrillation, coronary heart disease, congestive heart failure) and higher incidence of naMCI in women, but not in men [27]. Our results support these previous findings, which suggest that vascular risk factors and vascular diseases may be associated with naMCI. Therefore, naMCI subtypes meet the criteria for vascular MCI, which results from cerebrovascular lesions in relevant brain regions and is characterized by cognitive features, such as dysexecutive syndrome with relative sparing of memory [28, 29]. As previously mentioned, aMCI is considered a symptomatic predementia phase of AD and naMCI is related to the early stages of subcortical vascular and other non-AD type dementias [30].

However, this study revealed that raised TG was associated with lower odds of naMCI in men. These findings may reflect age and race differences. Our study participants were aged 70 years and above at recruitment, and the mean age was 75 years. Previous studies have concluded that TG levels are lower in patients with dementia compared with cognitively healthy older adults [31, 32]. Yin et al. have suggested that although TG is negatively associated with cognitive impairment among oldest-old Chinese adults, those with high normal levels of TG have good cognitive function [33]. High TG has been considered a risk factor for cognitive impairment, but this is not always the case for elderly people. TG may improve the transport of insulin and ghrelin across the blood brain barrier [34, 35]; these are peripherally derived peptides that may improve cognitive function. Elevated TG levels indicate the expression of orexigenic hypothalamic peptides [36], which could play a role in improving cognitive function [37]. Our results support these previous findings and suggest that raised TG levels are beneficial for cognitive function in older adults.

A consideration of the effects of biological mechanisms of MetS individual components on MCI may help to explain our results. Hyperglycemia refers to higher concentrations of blood glucose and induces oxidative stress [38]. Oxidative stress contributes to vascular and endothelial dysfunction, and this microvascular damage leads to cognitive decline and risk of cognitive impairment [39, 40]. Previous studies have reported that diabetes is strongly associated with prevalent VD, but only marginally associated with AD [41, 42]. Yoshitake et al. [43] examined longitudinal data, reporting that diabetes was a significant risk factor for VD but not AD. These previous studies are consistent with the current finding that raised glucose was associated with greater odds of naMCI, which is related to the early stages VD in both men and women [43]. Thus, hyperglycemia might play an important role in the association between MetS and MCI, regardless of sex. Chronic high blood pressure is linked to brain amyloid-β accumulation and induces cerebral arterial stiffness and microvascular dysfunction, thus contributing to dementia pathophysiology [44 –46]. Ogbera et al. have reported that women show a significantly higher incidence of hypertension [47]. We found that women were significantly more likely than men to have raised blood pressure and that this was associated with naMCI for women, but not for men. Dyslipidemia is the presence of elevated TG and low levels of HDL-C. Low HDL-C is considered a risk factor for cardiovascular disease, which may be involved in β-amyloid metabolism and deposition in the brain [48]. Thus, low HDL-C can lead to cognitive impairment and dementia [49]. We found that reduced HDL-C was associated with greater odds of naMCI in men. The mechanism underlying this sex difference remains speculative; we suggest that the overall prevalence of MetS in MCI or risk of MCI does not differ by sex, but that MetS individual components show a different prevalence or risk of MCI between men and women.

To our knowledge, the present study is the first to report an association between MetS and MCI subtypes in a large sample of community-dwelling elderly Japanese people. We found that MetS is associated with naMCI, regardless of sex. However, this study has some limitations. First, we used cross-sectional data. Therefore, we were unable to determine whether there was a causal relationship between MetS and MCI subtypes. A recent prospective study indicated that MetS, as well as being associated with cardiovascular risk factors and diabetes mellitus, is a significant risk factor for MCI progression to dementia, although this study did not address MCI subtypes [50]. Longitudinal and/or interventional studies are needed to address this issue. Second, we did not measure dietary factors or genetic risk factors (e.g., APOE ɛ4), which may be important covariates in the relationship between MetS and MCI. However, we adjusted for BMI in the multivariate model, which may have partially cancelled out nutritional influences. Third, because our participants were older Japanese people, these findings may not generalize to other populations of different nationalities or ages. Finally, the sample size of aMCI was relatively small. This limitation may have affected the current results. A larger sample would provide more power to observe the relationships between MetS and MCI.

In conclusion, the results clearly indicate that MetS is associated with naMCI among elderly Japanese people, regardless of sex, and suggest etiologic differences in MCI subtypes. However, we found sex differences in individual MetS components, which were differentially associated with risk of naMCI by sex. Managing MetS and its components may reduce the risk of MCI and reduce dementia and related medical care costs. Further research on the causal role of MetS and MCI subtypes using longitudinal designs is needed.

Footnotes

ACKNOWLEDGMENTS

This study was supported by Strategic Basic Research Programs (RISTEX Redesigning Communities for Aged Society), Japan Science and Technology Agency, and by Health and Labour Sciences Research grants (Comprehensive Research on Aging and Health). We would like to thank the Midori-Ward of Nagoya city and Obu city office for assistance with participant recruitment.

Authors’ disclosures available online ().

References

Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (2001) Executive summary of the third report of The National Cholesterol Education Program (NCEP).

Alberti

, Eckel

, Grundy

, Zimmet

, Cleeman

, Donato

, Fruchart

, James

, Loria

, Smith

Jr ; International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; International Association for the Study of Obesity (2009) Harmonizing the metabolic syndrome: A joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 120, 1640–1645.

Liu

, He

, Jiang

, Wu

, Wang

, Yang

, Wang

(2015) Association between metabolic syndrome and mild cognitive impairment and its age difference in a Chinese community elderly population. Clin Endocrinol (Oxf) 82, 844–853.

Frisardi

, Solfrizzi

, Capurso

, Imbimbo

, Vendemiale

, Seripa

, Pilotto

, Panza

(2010) Is insulin resistant brain state a central feature of the metabolic-cognitive syndrome? J Alzheimers Dis 21, 57–63.

Vanhanen

, Koivisto

, Moilanen

, Helkala

, Hanninen

, Soininen

, Kervinen

, Kesaniemi

, Laakso

, Kuusisto

(2006) Association of metabolic syndrome with Alzheimer disease: A population-based study. Neurology 67, 843–847.

Petersen

(2011) Clinical practice. Mild cognitive impairment. N Engl J Med 364, 2227–2234.

Winblad

, Palmer

, Kivipelto

, Jelic

, Fratiglioni

, Wahlund

, Nordberg

, Backman

, Albert

, Almkvist

, Arai

, Basun

, Blennow

, de Leon

, DeCarli

, Erkinjuntti

, Giacobini

, Graff

, Hardy

, Jack

, Jorm

, Ritchie

, can Duijn

, Visser

, Petersen

(2004) Mild cognitive impairment–beyond controversies, towards a consensus: Report of the International Working Group on Mild Cognitive Impairment. J Intern Med 256, 240–246.

Feng

, Chong

, Lim

, Lee

, Collinson

, Yap

, Ng

(2013) Metabolic syndrome and amnestic mild cognitive impairment: Singapore Longitudinal Ageing Study-2 findings. J Alzheimers Dis 34, 649–657.

Roberts

, Geda

, Knopman

, Cha

, Boeve

, Ivnik

, Pankratz

, Tangalos

, Petersen

(2010) Metabolic syndrome, inflammation, and non-amnestic mild cognitive impairment in older persons: A population-based study. Alzheimer Dis Assoc Disord 24, 11–18.

10.

Roberts

, Geda

, Knopman

, Cha

, Pankratz

, Boeve

, Tangalos

, Ivnik

, Rocca

, Petersen

(2012) The incidence of MCI differs by subtype and is higher in men: The Mayo Clinic Study of Aging. Neurology 78, 342–351.

11.

Petersen

, Roberts

, Knopman

, Geda

, Cha

, Pankratz

, Boeve

, Tangalos

, Ivnik

, Rocca

(2010) Prevalence of mild cognitive impairment is higher in men. The Mayo Clinic Study of Aging. Neurology 75, 889–897.

12.

Trollor

, Smith

, Baune

, Kochan

, Campbell

, Samaras

, Crawford

, Brodaty

, Sachdev

(2010) Systemic inflammation is associated with MCI and its subtypes: The Sydney Memory and Aging Study. Dement Geriatr Cogn Disord 30, 569–578.

13.

Skup

, Zhu

, Wang

, Giovanello

, Lin

, Shen

, Shi

, Gao

, Lin

, Fan

, Zhang

; Alzheimer’s Disease Neuroimaging Study (2011) Sex differences in grey matter atrophy patterns among AD and aMCI patients: Results from ADNI. Neuroimage 56, 890–906.

14.

Shimada

, Makizako

, Lee

, Doi

, Lee

, Tsutsumimoto

, Harada

, Hotta

, Bae

, Nakakubo

, Harada

, Suzuki

(2016) Impact of cognitive frailty on daily activities in older persons. J Nutr Health Aging 20, 729–735.

15.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

16.

Hänninen

, Hallikainen

, Tuomainen

, Vanhanen

, Soininen

(2002) Prevalence of mild cognitive impairment: A population-based study in elderly subjects. Acta Neurol Scand 106, 148–154.

17.

Jungwirth

, Weissgram

, Zehetmayer

, Tragl

, Fischer

(2005) VITA: Subtypes of mild cognitive impairment in a community-based cohort at the age of 75 years. Int J Geriatr Psychiatry 20, 452–458.

18.

Yaffe

, Middleton

, Lui

, Spira

, Stone

, Racine

, Ensrud

, Kramer

(2011) Mild cognitive impairment, dementia, and their subtypes in oldest old women. Arch Neurol 68, 631–636.

19.

Makizako

, Shimada

, Doi

, Tsutsumimoto

, Hotta

, Nakakubo

, Makino

, Suzuki

(2016), Comorbid mild cognitive impairment anddepressive symptoms predict future dementia in community olderadults: A 24-month follow-up longitudinal study. J Alzheimers Dis 54, 1473–1482.

20.

Makizako

, Shimada

, Park

, Doi

, Yoshida

, Uemura

, Tsutsumimoto

, Suzuki

(2013) Evaluation of multidimensional neurocognitive function using a tablet personal computer: Test-retest reliability and validity in community-dwelling older adults. Geriatr Gerontol Int 13, 860–866.

21.

Alberti

, Zimmet

, Shaw

(2006) Metabolic syndrome–a newworld-wide definition. A consensus statement from theInternational Diabetes Federation. Diabet Med 23, 469–480.

22.

Yesavage

(1988) Geriatric depression scale. Psychopharmacol Bull 24, 709–711.

23.

World Health Organization (1995) Physical status. The use and interpretation of anthropometry. Report of a WHO expert consultation. WHO Technical Report Series Number 854. Geneva. World Health Organization.

24.

Yaffe

, Kanaya

, Lindquist

, Simonsick

, Harris

, Shorr

, Tylavsky

, Newman

(2004) The metabolic syndrome, inflammation, and risk of cognitive decline. JAMA 292, 2237–2242.

25.

, Wu

, Zhao

, Ji

, Song

, Zhang

, Huang

(2016) Prevalence of mild cognitive impairment and its subtypes among Chinese older adults: Role of vascular risk factors. Dement Geriatr Cogn Disord 41, 261–272.

26.

Wallin

, Milos

, Sjögren

, Pantoni

, Erkinjuntti

(2003) Classification and subtypes of vascular dementia. Int Psychogeriatr 15, 27–37.

27.

Roberts

, Geda

, Knopman

, Cha

, Pankratz

, Boeve

, Tangalos

, Ivnik

, Mielke

, Petersen

(2013) Cardiac disease increases risk of non-amnestic mild cognitive impairment: Stronger impact in women. JAMA Neurol 70, 374–382.

28.

Frisoni

, Galluzzi

, Bresciani

, Zanetti

, Geroldi

(2002) Mild cognitive impairment with subcortical vascular features: Clinical characteristics and outcome. J Neurol 249, 1423–1432.

29.

Galluzzi

, Sheu

, Zanetti

, Frisoni

(2005) Disstinctive clinical features of mild cognitive impairment with subcortical cerebrovascular disease. Dement Geriatr Cogn Disord 19, 196–203.

30.

Luchsinger

, Reitz

, Patel

, Tang

, Manly

, Mayeux

(2007) Relation of diabetes to mild cognitive impairment. Arch Neurol 64, 570–575.

31.

Dimopoulos

, Piperi

, Salonicioti

, Psarra

, Mitsonis

, Liappas

, Lea

, Kalofoutis

(2007) Characterization of the lipid profile in dementia and depression in the elderly. J Geriatr Psychiatry Neurol 20, 138–144.

32.

Lepara

, Valjevac

, Alajbegović

, Zaciragić

, Nakas-Ićindić

(2009) Decreased serum lipids in patientswith probable Alzheimer’s disease. Bosn J Basic Med Sci 9, 215–220.

33.

Yin

, Shi

, Kraus

, Fitzgerald

, Qian

, Xu

, Zhai

, Sereny

, Zeng

(2012) High normal plasma triglycerides are associated with preserved cognitive function in Chinese oldest-old. Age Ageing 41, 600–606.

34.

Banks

, Burney

, Robinson

(2008) Effects of triglycerides, obesity, and starvation on ghrelin transport across the blood-brain barrier. Peptides 29, 2061–2065.

35.

Urayama

, Banks

(2008) Starvation and triglycerides reverse the obesity-induced impairment of insulin transport at the blood-brain barrier. Endocrinology 149, 3592–3597.

36.

Chang

, Karatayev

, Davydova

, Leibowitz

(2004) Circulating triglycerides impact on orexigenic peptides and neuronal activity in hypothalamus. Endocrinology 145, 3904–3912.

37.

Jaeger

, Farr

, Banks

, Morley

(2002) Effects of orexin-A on memory processing. Peptides 23, 1683–1688.

38.

Capellini

, Baldo

, Celotto

, Batalhão

, Cárnio

, Rodrigues

, Evora

(2010) Oxidative stress is notassociated with vascular dysfunction in a model of alloxan-induceddiabetic rats. Arq Bras Endocrinol Metabol 54, 530–539.

39.

Liu

, Zhou

, Xu

, Zhang

, Li

, Zhao

(2009) Metabolic syndrome and cognitive impairment amongst elderly people in Chinese population: A cross-sectional study. Eur J Neurol 16, 1022–1027.

40.

Yates

, Sweat

, Yau

, Turchiano

, Convit

(2012) Impact of metabolic syndrome on cognition and brain: A selected review of the literature. Arterioscler Thromb Vasc Biol 32, 2060–2067

41.

Hassing

, Johansson

, Nilsson

, Berg

, Pedersen

, Gatz

, McClearn

(2002) Diabetes mellitus is a risk factor for vascular dementia, but not for Alzheimer’s disease: A population-based study of the oldest old. Int Psychogeriatr 14, 239–248.

42.

Cheng

, Huang

, Deng

, Wang

(2012) Diabetes as a risk factor for dementia and mild cognitive impairment: A meta-analysis of longitudinal studies. Intern Med J 42, 484–491.

43.

Yoshitake

, Kiyohara

, Kato

, Ohmura

, Iwamoto

, Nakayama

, Ohmori

, Nomiyama

, Kawano

, Ueda

, Sueishi

, Tsuneyoshi

, Fujishima

(1995) Incidence and risk factors of vascular dementia and Alzheimer’s disease in a defined elderly Japanese population: The Hisayama Study. Neurology 45, 1161–1168.

44.

Perrotta

, Lembo

, Carnevale

(2016) Hypertension and dementia: Epidemiological and experimental evidence revealing a detrimental relationship. Int J Mol Sci 17, 347.

45.

Hughes

, Craft

, Lopez

(2015) Review of ‘the potential role of arterial stiffness in the pathogenesis of Alzheimer’s disease’. Neurodegener Dis Manag 5, 121–135.

46.

Langbaum

, Chen

, Launer

, Fleisher

, Lee

, Liu

, Protas

, Reeder

, Bandy

, Yu

, Caselli

, Reiman

(2012) Blood pressure is associated with higher brain amyloid burden and lower glucose metabolism in healthy late middle-age persons. Neurobiol Aging 33, 827.e11–e19.

47.

Ogbera

(2010) Prevalence and gender distribution of the metabolic syndrome. Diabetol Metab Syndr 2, 1.

48.

Olesen

, Dagø

(2000) High density lipoprotein inhibits assembly of amyloid beta-peptides into fibrils. BiochemBiophys Res Commun 270, 62–66.

49.

van

Exel E

, de Craen

, Gussekloo

, Houx

, Bootsma-van der Wiel

, Macfarlane

, Blauw

, Westendorp

(2002) Association between high-density lipoprotein and cognitive impairment in the oldest old. Ann Neurol 51, 716–721.

50.

, Feng

, Nyunt

, Feng

, Gao

, Lim

, Collinson

, Chong

, Lim

, Lee

, Yap

(2016) Metabolic syndrome and the risk of mild cognitive impairment and progression to dementia: Follow-up of the Singapore Longitudinal Ageing Study Cohort. JAMA Neurol 73, 456–463.