Diabetes Mellitus,Elevated Hemoglobin A 1c,and Glycated Albumin Are Associated with the Presence of All-Cause Dementia and Alzheimer’s Disease: The JPSC-AD Study

Abstract

Background:

Glucose dysmetabolism is an important risk factor for dementia.

Objective:

We investigated the associations of diabetes mellitus, the levels of glycemic measures, and insulin resistance and secretion measures with dementia and its subtypes in a cross-sectional study.

Methods:

In this study, 10,214 community-dwelling participants were enrolled. Hemoglobin A1c (HbA_1c), the homeostasis model assessment (HOMA) for insulin resistance (HOMA-IR), the HOMA of percent β-cell function (HOMA-β), and the glycated albumin (GA) was evaluated. The associations of each measure with Alzheimer’s disease (AD) and vascular dementia (VaD) were investigated.

Results:

The multivariable-adjusted odds ratios (ORs) of AD were significantly higher in participants with diabetes mellitus than in those without diabetes (1.46 [95% CI: 1.08–1.97]). Higher HbA_1c levels were significantly associated with AD at diabetes (≥6.5%) and even at prediabetes (5.7 %–6.4 %) levels; multivariable-adjusted ORs for AD in participants at the diabetes level were 1.72 (95% CI: 1.19–2.49), and those in participants at the prediabetes level were 1.30 (95% CI: 1.00–1.68), compared with those in normal participants. Moreover, higher GA levels were associated with AD. No associations were observed between the diabetic status or the levels of glycemic measures and VaD. In addition, no significant relationships were observed between insulin resistance and secretion measurements and AD and VaD.

Conclusion:

Our findings indicate that diabetes mellitus and hyperglycemia are significantly associated with AD, even in individuals at the prediabetes level.

Keywords

Alzheimer’s disease diabetes mellitus glycated albumin hemoglobin A1c vascular dementia

INTRODUCTION

Glucose dysmetabolism is an important risk factor for Alzheimer’s disease (AD) and vascular dementia (VaD) [1, 2]. It has been observed that the prevalence of diabetes mellitus is increasing worldwide [3]. Therefore, identifying potentially modifiable risk factors for developing dementia is important for both the treatment strategy and reduction of the incidence of dementia in people with diabetes.

The etiology of developing dementia in people with glucose dysmetabolism is multifactorial; atherosclerosis, microvascular disease, glucose toxicity, and insulin resistance have been implicated [1]. Hemoglobin A1c (HbA_1c) is a widely used glycemic measure in clinical settings, and glycated albumin (GA) has been used as an alternative glycemic measure of postprandial glucose fluctuations [4]. As the glycation speed of albumin is faster than that of hemoglobin [5], the GA levels rise faster than the HbA_1c levels in response to a rapid increase in blood glucose levels [4, 6]. In addition, the homeostasis model assessment (HOMA) for insulin resistance (HOMA-IR) is utilized as a marker of insulin resistance [7], and the HOMA of percent β-cell function (HOMA-β) is considered to be a marker of insulin secretion [7].

This cross-sectional study was conducted to investigate the association of diabetic status, glycemic measures, and insulin resistance and secretion measures with dementia and its subtypes in an elderly Japanese population.

MATERIALS AND METHODS

Study population

The Japan Prospective Studies Collaboration for Aging and Dementia (JPSC-AD) study is a multisite, population-based prospective cohort study of dementia [8]. It includes eight sites in Japan [8]. From 2016 to 2018, 11,410 residents aged 65 years or older underwent health examination for this study. Of the 11,410 participants, 8,030 (65%) were recruited by full community surveys at 6 rural sites. The remaining 3,380 were selected at the 2 sites with larger populations by a simple random sampling (n = 1,099) and by a voluntary response sampling (n = 2,281) [8]. After the exclusion of 214 participants due to lack of drug information and 982 participants who did not have blood tests, a total of 10,214 participants (4,350 men and 5,864 women) were enrolled in this study (Fig. 1).

Fig.1

A flow chart of enrolled individuals for each analysis of the associations between presence of dementia and diabetic status, HbA_1c levels, GA levels, HOMA-IR and HOMA-β. GA, glycated albumin; HbA_1c, hemoglobin A1c; HOMA-β, homeostasis model assessment of percent β-cell function; HOMA-IR, homeostasis model assessment for insulin resistance.

Standard protocol approvals and participant consent

This study was approved by the Medical Ethics Review Board of Kanazawa University (approval number 2185) and Kyusyu University’s Institutional Review Board for clinical research (approval number 686-06). We obtained written informed consent from all participants.

Diagnosis of dementia and its subtypes

Dementia was diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd Revised Edition (DSM-III-R) [9]. The diagnosis of AD and VaD was made based on the following criteria: the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association criteria [10]. and the National Institute of Neurological Disorders and Stroke-Association International pour la Recherche et I’Enseignement en Neurosciences criteria [11], respectively. Clinical information, including neuroimaging studies, was utilized for the diagnosis of AD and VaD. Expert psychiatrists and neurologists in the study team adjudicated every case of dementia. The diagnostic procedure has been previously reported [8].

Diagnosis of diabetes mellitus and prediabetes

Diabetes mellitus was determined by plasma glucose and HbA_1c levels (fasting glucose level ≥7.0mmol/L, casual glucose level ≥11.1 mmol/L, or HbA_1c≥6.5%) using the 2010 American Diabetes Association (ADA) criteria [12] and/or the current use of glucose-lowering agents. Prediabetes was determined by plasma HbA_1c levels (HbA_1c 5.7%–6.4%) using standard clinical cutoff points.

Measurements of HbA_1c, GA, HOMA-IR, and HOMA-β

Plasma glucose and HbA_1c levels measured using the National Glycohemoglobin Standardization Program (NGSP), and serum GA levels were determined using the enzymatic method. Serum insulin levels were measured by chemiluminescent immunoassay. The detailed measurement methods of blood chemistry have been previously reported [8].

For analyses of the association between the HbA_1c category and the presence of all-cause dementia and its subtypes, HbA_1c was categorized using standard clinical cutoff points: normal, 5.7%; prediabetes, 5.7%–6.4%; and diabetes, 6.5%. For the analyses, the GA value was divided into quartiles. The HOMA-IR was calculated using the following formula: HOMA-IR = fasting serum insulin (μU/mL)/fasting plasma glucose (mmol/L)/22.5 [13]. Subsequently, the HOMA-β was calculated using the following equation: HOMA-β= fasting serum insulin (μU/mL)×20/(fasting plasma glucose (mmol/L) – 3.5) [13]. For analyses, the HOMA-IR and HOMA-β were categorized using standard clinical cutoff points: ≤1.6%, 1.7%–2.4%, ≥2.5%, ≤30%, and > 30%. For the HOMA-IR and HOMA-β analysis, we excluded individuals with taking insulin injection to remove effects of exogenous insulin [13]. Additionally, we excluded individuals with fasting plasma glucose level ≥7.8 mmol/L because we cannot assess insulin resistance with HOMA-IR in case of hyperglycemia [13].

Other risk factor measurements

Each participant completed a self-administered questionnaire that contains questions on sociodemographic data (age, sex, and educational levels), medical history (diabetes mellitus and hypertension), drug information, smoking and drinking habits, and regular exercise. Regular exercise was defined as physical activity specifically undertaken for sport or exercise performed for at least 30 min twice per week over the most recent year or longer. The completed questionnaires were reviewed by the researchers, who were trained to identify inconsistent or unanswered items. Depressive symptoms were evaluated using the Geriatric Depression Scale (GDS) (short form) [14], and a GDS score of ≥6 was used to denote depressive symptoms. Blood pressure was measured three times using a sphygmomanometer, with an interval of at least 5 min; the average of the three measurements was used for the analysis. Hypertension was determined by blood pressure levels of ≥140/90 mmHg or the current use of antihypertensive agents. Body mass index (BMI) (kg/m²) was measured as an indicator of obesity. Serum HDL and LDL cholesterols were enzymatically measured [8].

Statistical analyses

According to the diabetic status, clinical characteristics were evaluated using the t-test and chi-squared test for continuous and categorical variables, respectively. Simple linear correlations were calculated by determining the Spearman’s rank correlation coefficient r among HbA_1c, GA, HOMA-IR and HOMA-β variables. Generalized linear models (binominal distribution and logistic regression model) were utilized to analyze the interaction of sex and the independent effects of diabetes diagnosis, clinical categories of HbA_1c, HOMA-IR, HOMA-β, and quartiles of GA with the association of the presence of all-cause dementia and its subtypes. Model 1 was adjusted for age and sex, and Model 2 was further adjusted for research site, hypertension, BMI levels, serum LDL and HDL cholesterol levels, educational levels, smoking and drinking habits, depressive symptoms, and regular exercise. p < 0.05 was considered statistically significant. The SPSS software package (version 23; SPSS Inc., Chicago, IL, USA) was used to perform all statistical analyses.

RESULTS

Overall, 1,876 individuals (18.3%) had diabetes and 599 individuals (5.8%) were diagnosed as all-cause dementia. Of the 1,876 patients with diabetes mellitus, 276 had fasting glucose level ≥7.0 mmol/L, 184 had casual glucose level ≥11.1 mmol/L, 1,148 had HbA_1c≥6.5% and 1,585 currently used glucose-lowering agents. Regarding dementia subtypes, 410 (68.4%) were AD, 70 (11.6%) were VaD, 20 (3.3%) were dementia with Lewy bodies. The frequency of the mixed type of dementia was 7.6% (n = 46), the primary subtype of which was a combination of AD and VaD. Of the 10,214 enrolled individuals, 125 and 255 individuals without available measurement of HbA_1c and GA were excluded respectively. Therefore, we analyzed with 10,089 and 9,959 individuals for HbA_1c and GA, respectively (Fig. 1). Additionally, 104 individuals with fasting glucose level≥7.8 mmol/L and one individual of fasting glucose level of 3.5 mmol/L were also excluded in the analysis of the HOMA-IR and HOMA-β, and remaining 4,229 and 4,332 individuals were enrolled respectively (Fig. 1). The median values of HbA_1c, GA, HOMA-IR, and HOMA-β in the study population were 5.7% (interquartile range [IQR]: 5.4%–6.9%), 15.0% (IQR: 14.0%–16.3%), 1.07% (IQR: 0.76%–1.58%), and 65.1% (IQR: 45.2%–91.8%), respectively. A statistically significant correlations were observed between HbA_1c, GA, HOMA-IR and HOMA-β, GA and HOMA-β, and HOMA-IR and HOMA-β in both participants with and without diabetes mellitus groups (Supplementary Table 1A–C). The clinical characteristics of the study population are summarized according to the diabetic status, the levels of HbA_1c, GA, HOMA-IR, and HOMA-β (Table 1A–E). For the diabetes mellitus diagnosis, age, systolic blood pressure, BMI, depressive symptoms, the proportion of males, educational levels, smoking habit, all-cause dementia, AD, and VaD were significantly higher, and the levels of the diastolic blood pressure, serum LDL cholesterol, and serum HDL cholesterol were significantly lower in participants with diabetes than in those without diabetes (Table 1A). Similar findings were obtained for HbA_1c (Table 1B), except for age, educational levels, depressive symptoms, smoking and drinking habits, and all-cause dementia, AD, and VaD. In addition, similar findings were obtained for serum GA, except for sex, BMI, and smoking and drinking habits (Table 1C); for the HOMA-IR (Table 1D), except for age, sex, diastolic blood pressure, serum LDL cholesterol, educational levels, depressive symptoms, smoking and drinking habits, physical activities, and all-cause dementia, AD, and VaD; and for the HOMA-β (Table 1E), except for age, diastolic blood pressure, BMI, depressive symptoms, drinking habit, and all-cause dementia, AD, and VaD.

Table 1A

Clinical characteristics according to the 2010 American Diabetes Association diabetes mellitus criteria

Variables	Diabetes mellitus
	No	Yes
N	8,338	1,876
Age, y, median, IQR	72 (68–78)	73 (69–79)^*
Women, %	60.0	45.9^*
Hypertension, %	71.5	84.8^*
Mean sBP, mmHg, median, IQR	138.3 (127.0–151.3)	141.6 (130.0–154.0)^*
Mean dBP, mmHg, median, IQR	78.0 (70.6–85.6)	76.3 (69.0–84.0)^*
BMI, kg/m², median, IQR	22.9 (20.9–25.0)	24.1 (21.9–26.6)^*
Serum LDL-chol, mg/dL, median, IQR	116.0 (97.0–137.0)	107.0 (87.0–129.0)^*
Serum HDL-chol, mg/dL, median, IQR	61.0 (51.0–74.0)	55.0 (46.0–66.0)^*
Education≤9 y, %	31.4	37.6^*
Depressive symptoms, %	14.6	18.4^*
Smoking habit, %	7.7	9.9^*
Drinking habit, %	41.8	41.1
Regular exercise, %	41.8	41.4
All-cause dementia, n (%)	442 (5.3)	157 (8.3)^*
Alzheimer’s disease, n (%)	303 (3.6)	107 (5.7)^*
Vascular dementia, n (%)	48 (0.5)	22 (1.1)^*

BMI, body mass index; dBP, diastolic blood pressure; IQR, interquartile range; Serum HDL-chol, serum high-density lipoprotein cholesterol; Serum LDL-chol, serum low-density lipoprotein cholesterol; sBP, systolic blood pressure. ^*p < 0.05.

Table 1B

Clinical characteristics according to plasma HbA_1c levels

Variables	HbA_1c levels (%)
	< 5.7	5.7–6.4	≥6.5
N	4,851	4,090	1,148
Age, y, median, IQR	72 (68–78)	73 (68–78)	73 (69–78)
Women, %	58.6	59.2	47.3^*
Hypertension, %	70.2	75.2	82.7^*
Mean sBP, mmHg, median, IQR	138.0 (126.3–151.3)	138.6 (128.0–151.3)	143.0 (131.0–154.6) ^*
Mean dBP, mmHg, median, IQR	78.3 (70.6–86.0)	77.6 (70.3–85.3)	76.5 (69.6–84.0) ^*
BMI, kg/m², median, IQR	22.6 (20.6–24.6)	23.5 (21.4–25.7)	24.1 (22.1–26.7)^*
Serum LDL-chol, mg/dL, median, IQR	115.0 (97.0–136.0)	115.0 (96.0–137.0)	110.0 (90.0–132.0)^*
Serum HDL-chol, mg/dL, median, IQR	63.0 (52.0–75.0)	59.0 (49.0–71.0)	54.0 (45.0–64.0)^*
Education≤9 y, %	32.2	31.4	33.3
Depressive symptoms, %	15.3	14.4	16.5
Smoking habit, %	8.5	7.4	8.7
Drinking habit, %	44.5	39.0	41.1^*
Regular exercise, %	41.6	42.3	40.7
All-cause dementia, n (%)	262 (5.4)	236 (5.7)	75 (6.5)
Alzheimer’s disease, n (%)	179 (3.6)	166 (4.0)	53 (4.6)
Vascular dementia, n (%)	29 (0.5)	24 (0.5)	7 (0.6)

BMI, body mass index; dBP, diastolic blood pressure; HbA_1c, hemoglobin A_1c; IQR, interquartile range; Serum HDL-chol, serum high-density lipoprotein cholesterol; Serum LDL-chol, serum low-density lipoprotein cholesterol; sBP, systolic blood pressure. ^*p < 0.05 versus HbA_1c < 5.7%.

Table 1C

Clinical characteristics according to the serum GA levels

Variables	Serum GA levels (%)
	Q1 (≤14.0)	Q2 (14.1–15.0)	Q3 (15.1–16.3)	Q4 (≥16.4)
N	2,700	2,537	2,365	2,357
Age, y, median, IQR	70 (67–75)	72 (68–77)	74 (69–79)	75 (70–81)^*
Women, %	53.8	61.6	63.3	51.9
Hypertension, %	73.9	69.8	71.6	79.1^*
Mean sBP, mmHg, median, IQR	139.5 (128.3–152.3)	137.3 (126.3–150.6)	137.3 (126.0–150.6)	140.6 (128.3–153.0)^*
Mean dBP, mmHg, median, IQR	80.0 (73.0–87.6)	78.0 (71.0–85.6)	77.0 (69.3–84.3)	75.6 (68.3–83.0) ^*
BMI, kg/m², median, IQR	23.8 (21.9–25.9)	23.0 (21.0–25.1)	22.6 (20.5–24.8)	22.8 (20.6–25.2)^*
Serum LDL-chol, mg/dL, median, IQR	119.0 (99.0–139.0)	117.0 (99.0–138.0)	114.0 (96.0–135.0)	109.0 (89.0–130.0)^*
Serum HDL-chol, mg/dL, median, IQR	59.0 (49.0–70.0)	62.0 (52.0–74.0)	62.0 (52.0–74.0)	59.0 (48.0–72.0)^*
Education≤9 y, %	25.6	30.6	33.6	37.1^*
Depressive symptoms, %	12.8	14.5	15.4	17.9^*
Smoking habit, %	11.4	6.9	6.7	6.5^*
Drinking habit, %	49.4	42.7	37.5	36.7^*
Regular exercise, %	43.1	43.1	40.2	41.1
All-cause dementia, n (%)	74 (2.7)	87 (3.4)	146 (6.1)	263 (11.1)^*
Alzheimer’s disease, n (%)	45 (1.6)	69 (2.7)	110 (4.6)	171 (7.2)^*
Vascular dementia, n (%)	15 (0.5)	8 (0.3)	10 (0.4)	27 (1.1)^*

BMI, body mass index; dBP, diastolic blood pressure; GA, glycated albumin; IQR, interquartile range; Serum HDL-chol, serum high-density lipoprotein cholesterol; Serum LDL-chol, serum low-density lipoprotein cholesterol; sBP, systolic blood pressure. ^*p < 0.05 versus the first quartile (Q1) for GA.

Table 1D

Clinical characteristics according to HOMA-IR levels

Variables	HOMA-IR levels
	≤1.6	1.7–2.4	≥2.5
N	3,218	735	276
Age, y, median, IQR	71 (68–76)	71 (67–75)	71 (68–76)
Women, %	58.9	56.7	58.7
Hypertension, %	68.5	79.2	84.4^*
Mean sBP, mmHg, median, IQR	137.0 (125.3–151.3)	140.0 (126.3–153.0)	138.5 (127.6–152.5) ^*
Mean dBP, mmHg, median, IQR	78.3 (70.0–86.0)	78.0 (70.0–86.6)	76.6 (67.3–84.3)
BMI, kg/m², median, IQR	22.3 (20.4–24.1)	24.6 (22.7–26.7)	25.7 (23.7–28.0)^*
Serum LDL-chol, mg/dL, median, IQR	121.0 (100.0–141.0)	121.0 (99.0–140.0)	117.5 (95.2–135.0)
Serum HDL-chol, mg/dL, median, IQR	65.0 (55.0–77.0)	57.0 (48.0–66.0)	52.0 (44.2–63.0)^*
Education≤9 y, %	25.0	26.3	26.2
Depressive symptoms, %	16.0	15.5	15.9
Smoking habit, %	7.2	6.9	8.3
Drinking habit, %	45.3	43.7	34.5^*
Regular exercise, %	45.7	41.6	34.3^*
All-cause dementia, n (%)	130 (4.0)	22 (2.9)	13 (4.7)
Alzheimer’s disease, n (%)	103 (3.2)	15 (2.0)	11 (3.9)
Vascular dementia, n (%)	10 (0.3)	3 (0.4)	1 (0.3)

BMI, body mass index; dBP, diastolic blood pressure; HOMA-IR, homeostasis model assessment for insulin resistance; IQR, interquartile range; serum HDL-chol, serum high-density lipoprotein cholesterol; serum LDL-chol, serum low-density lipoprotein cholesterol; sBP, systolic blood pressure. ^*p < 0.05 versus HOMA-IR≤1.6.

Table 1E

Clinical characteristics according to the HOMA-β level

HOMA-β level (%)
	> 30.0	≤30.0
N	3,873	459
Age, y, median, IQR	71 (67–76)	72 (68–76)
Women, %	61.1	32.9^*
Hypertension, %	71.4	75.4
Mean sBP, mmHg, median, IQR	137.3 (125.3–151.3)	143.0 (128.0–154.6)^*
Mean dBP, mmHg, median, IQR	78.0 (70.0–86.3)	78.6 (70.6–86.0)
BMI, kg/m², median, IQR	22.9 (21.0–25.0)	21.7 (19.8–23.7)^*
Serum LDL-chol, mg/dL, median, IQR	121.0 (101.0–141.0)	112.0 (93.0–135.0)^*
Serum HDL-chol, mg/dL, median, IQR	62.0 (52.0–74.0)	66.0 (55.0–78.0)^*
Education ≤9 y, %	25.0	29.2^*
Depressive symptoms, %	15.8	17.3
Smoking habit, %	6.7	12.7^*
Drinking habit, %	42.5	59.0^*
Regular exercise, %	43.9	45.7
All-cause dementia, n (%)	155 (4.0)	15 (3.2)
Alzheimer’s disease, n (%)	121 (3.1)	13 (2.8)
Vascular dementia, n (%)	13 (0.3)	1 (0.2)

BMI, body mass index; dBP, diastolic blood pressure; HOMA-β, homeostasis model assessment of percent β-cell function; IQR, interquartile range; serum HDL-chol, serum high-density lipoprotein cholesterol; serum LDL-chol, serum low-density lipoprotein cholesterol; sBP, systolic blood pressure. ^*p < 0.05.

Table 2A

Age- and sex-adjusted and multivariable-adjusted ORs and 95% CIs for the presence of all-cause dementia and its subtypes according to the 2010 American Diabetes Association diabetes mellitus criteria

Diagnosis of diabetes	No. of cases/participants	Model 1 OR (95% CI)	p	Model 2 OR (95% CI)	p
All-cause dementia
No	442/8338	1.00 (reference)	1.00 (reference)
Yes	157/1876	1.68 (1.36–2.08)	< 0.001	1.43 (1.11–1.86)	0.006
Alzheimer’s disease
No	303/8199	1.00 (reference)	1.00 (reference)
Yes	107/1826	1.69 (1.32–2.17)	< 0.001	1.46 (1.08–1.97)	0.013
Vascular dementia
No	48/7944	1.00 (reference)	1.00 (reference)
Yes	22/1741	2.00 (1.19–3.34)	0.008	1.03 (0.47–2.21)	0.939

CI, confidence interval; OR, odds ratio. Model 1 was adjusted for age and sex. Model 2 was adjusted for age, sex, research site, hypertension, serum LDL cholesterol levels, serum HDL cholesterol levels, body mass index, educational levels, smoking habits, drinking habits, depressive symptoms, and regular exercise.

The interactions of sex were not significant for all the associations between the presence of all-cause dementia and its subtypes and diabetes diagnosis, clinical categories of HbA_1c, HOMA-IR, HOMA-β, and quartiles of GA (Supplementary Table 2). The adjusted ORs of all-cause dementia and its subtypes according to the diabetes diagnosis defined by the ADA criteria are presented in Table 2A. The age- and sex-adjusted OR of all-cause dementia was significantly higher in participants with diabetes than in those without diabetes. This association remained unchanged even after adjusting for age, sex, research site, hypertension, serum LDL cholesterol, serum HDL cholesterol, BMI, educational levels, smoking habits, drinking habits, depressive symptoms, and regular exercise. With regard to the subtypes of dementia, the age- and sex-adjusted and multivariable-adjusted ORs of AD were significantly higher in participants with diabetes than in those without diabetes. The age- and sex-adjusted OR of VaD was significantly increased in participants with diabetes compared with those without diabetes; however, the association was insignificant after multivariable adjustment.

Table 2B

Age- and sex-adjusted and multivariable-adjusted ORs and 95% CIs for the presence of all-cause dementia and its subtypes according to HbA_1c levels

HbA _1c, (%)	No. of cases/participants	Model 1 OR (95% CI)	p	Model 2 OR (95% CI)	p
All-cause dementia
< 5.7	262/4851	1.00 (reference)	1.00 (reference)
5.7–6.4	236/4090	1.14 (0.94–1.40)	0.175	1.29 (1.04–1.62)	0.021
≥6.5	75/1148	1.45 (1.08–1.94)	0.013	1.65 (1.20–2.27)	0.002
Alzheimer’s disease
< 5.7	179/4768	1.00 (reference)	1.00 (reference)
5.7–6.4	166/4020	1.20 (0.94–1.51)	0.129	1.30 (1.00–1.68)	0.044
≥6.5	53/1126	1.54 (1.09–2.17)	0.013	1.72 (1.19–2.49)	0.004
Vascular dementia
< 5.7	29/4618	1.00 (reference)	1.00 (reference)
5.7–6.4	24/3878	1.02 (0.59–1.77)	0.929	1.26 (0.66–2.39)	0.480
≥6.5	7/1080	1.08 (0.47–2.50)	0.850	1.09 (0.42–2.81)	0.855

CI, confidence interval; HbA_1c, hemoglobin A1c; OR, odds ratio. Model 1 was adjusted for age and sex. Model 2 was adjusted for age, sex, research site, hypertension, serum LDL cholesterol levels, serum HDL cholesterol levels, body mass index, educational levels, smoking habits, drinking habits, depressive symptoms, and regular exercise.

Table 2C

Age- and sex-adjusted and multivariable-adjusted ORs and 95% CIs for all-cause dementia and its subtypes according to GA levels

GA, %	No. of cases/participants	Model 1 OR (95% CI)	p	Model 2 OR (95% CI)	p
All-cause dementia
≤14.0	74/2699	1.00 (reference)	1.00 (reference)
14.1–15.0	87/2532	0.92 (0.66–1.28)	0.622	0.81 (0.57–1.16)	0.268
15.1–16.3	146/2363	1.17 (0.86–1.59)	0.311	0.95 (0.68–1.32)	0.771
≥16.4	263/2353	1.69 (1.26–2.25)	< 0.001	1.38 (1.01–1.90)	0.042
Alzheimer’s disease
≤14.0	45/2670	1.00 (reference)	1.00 (reference)
14.1–15.0	69/2514	1.16 (0.78–1.73)	0.441	1.04 (0.69–1.58)	0.830
15.1–16.3	110/2327	1.37 (0.94–1.99)	0.097	1.11 (0.74–1.66)	0.584
≥16.4	171/2261	1.69 (1.18–2.42)	0.004	1.46 (0.99–2.14)	0.054
Vascular dementia
≤14.0	15/2640	1.00 (reference)	1.00 (reference)
14.1–15.0	8/2453	0.46 (0.19–1.09)	0.081	0.48 (0.18–1.28)	0.146
15.1–16.3	10/2227	0.48 (0.21–1.09)	0.082	0.31 (0.11–0.91)	0.034
≥16.4	27/2117	1.10 (0.56–2.16)	0.773	1.08 (0.49–2.38)	0.838

CI, confidence interval; GA, glycated albumin; OR, odds ratio. Model 1 was adjusted for age and sex. Model 2 was adjusted for age, sex, research site, hypertension, serum LDL cholesterol levels, serum HDL cholesterol levels, body mass index, educational levels, smoking habits, drinking habits, depressive symptoms, and regular exercise.

Next, we estimated the adjusted ORs of all-cause dementia and its subtypes according to the levels of HbA_1c, GA, HOMA-IR, and HOMA-β (Table 2B–E). The associations of HbA_1c levels with all-cause dementia and AD were significant even after adjusting for potential confounding factors (Table 2B). The multivariable-adjusted ORs for all-cause dementia and AD in participants at the prediabetes level were 1.29 (95% CI: 1.04–1.62) and 1.30 (95% CI: 1.00–1.68), respectively, whereas those in participants at the diabetes level were 1.65 (95% CI: 1.20–2.27) and 1.72 (95% CI: 1.19–2.49), respectively, compared with those of normal participants (Table 2B). Similarly, we found a significant association between the GA levels and all-cause dementia even after adjusting for potential confounding factors (Table 2C). Compared with the first quartile, the multivariable-adjusted OR for all-cause dementia in the fourth quartile was 1.38 (95% CI: 1.01–1.90) (Table 2C). In addition, a positive but insignificant association was observed between higher GA levels and AD (Table 2C). Alternatively, no significant associations were observed between HbA_1c or GA level and VaD (Table 2B, C). We could not find any significant associations between HOMA-IR or HOMA-β level and all-cause dementia and its subtypes (Tables 2D, E).

Table 2D

Age- and sex-adjusted and multivariable-adjusted ORs and 95% CIs for all-cause dementia and its subtypes according to HOMA-IR levels

HOMA -IR	No. of cases/participants	Model 1 OR (95% CI)	p	Model 2 OR (95% CI)	p
All-cause dementia
≤1.6	130/3218	1.00 (reference)	1.00 (reference)
1.7–2.4	22/735	0.91 (0.55–1.50)	0.736	1.02 (0.59–1.77)	0.917
≥2.5	13/276	1.03 (0.54–1.96)	0.916	1.14 (0.57–2.27)	0.708
Alzheimer’sdisease
≤1.6	103/3191	1.00 (reference)	1.00 (reference)
1.7–2.4	15/728	0.82 (0.45–1.48)	0.518	1.00 (0.52–1.88)	0.999
≥2.5	11/274	1.11 (0.55–2.23)	0.761	1.46 (0.68–3.11)	0.328
Vascular dementia
≤1.6	10/3098	1.00 (reference)	1.00 (reference)
1.7–2.4	3/716	1.45 (0.39–5.37)	0.571	1.23 (0.30–5.04)	0.766
≥2.5	1/264	1.11 (0.14–8.80)	0.919	0.70 (0.07–6.79)	0.762

CI, confidence interval; HOMA-IR, homeostasis model assessment for insulin resistance; OR, odds ratio. Model 1 was adjusted for age and sex. Model 2 was adjusted for age, sex, research site, hypertension, serum LDL cholesterol levels, serum HDL cholesterol levels, body mass index, educational levels, smoking habits, drinking habits, depressive symptoms, and regular exercise.

Table 2E

Age- and sex-adjusted and multivariable-adjusted ORs and 95% CIs for all-cause dementia and its subtypes according to HOMA-β levels

HOMA -β, %	No. of cases/participants	Model 1 OR (95% CI)	p	Model 2 OR (95% CI)	p
All-cause dementia
≥30	155/3873	1.00 (reference)	1.00 (reference)
< 30	15/459	0.89 (0.50–1.59)	0.705	0.93 (0.51–1.70)	0.831
Alzheimer’s disease
≥30	121/3839	1.00 (reference)	1.00 (reference)
< 30	13/457	1.07 (0.57–2.00)	0.833	1.12 (0.58–2.15)	0.723
Vascular dementia
≥30	13/3731	1.00 (reference)	1.00 (reference)
< 30	1/445	0.49 (0.06–3.87)	0.505	0.64 (0.08–5.24)	0.685

CI, confidence interval; HOMA-β, homeostasis model assessment of percent β-cell function; OR, odds ratio. Model 1 was adjusted for age and sex. Model 2 was adjusted for age, sex, research site, hypertension, serum LDL cholesterol levels, serum HDL cholesterol levels, body mass index, educational levels, smoking habits, drinking habits, depressive symptoms, and regular exercise.

DISCUSSION

This cross-sectional study indicated significant associations of diabetic status and the levels of HbA_1c with the presence of all-cause dementia and AD in a general elderly population of Japanese who participated in this large-scale dementia cohort study. Particularly, the multivariable-adjusted ORs for all-cause dementia and AD were significant in the diabetes and even prediabetes levels of HbA_1c. Moreover, we found that increased GA levels were significantly associated with the presence of all-cause dementia and positively but not significantly associated with AD. On the other hand, we could not observe associations between diabetic status and the levels of glycemic measures and the presence of VaD. Furthermore, no relationships were observed between insulin resistance or secretion and all-cause dementia and its subtypes.

This is the first population-based study that revealed the significant associations between prediabetes and all-cause dementia and AD in elderly individuals. These data suggest that the primary prevention of diabetes or sufficient glucose control are essential to prevent dementia or AD. According to the ADA recommendations, maintaining HbA_1c levels of less than 7.0% could help prevent diabetes-related microvascular complications in elderly individuals [12]. Therefore, in several studies, poor glycemic control was defined as HbA_1c of 7.0%, and a significantly higher risk of cognitive impairment was observed in elderly persons with poorly controlled diabetes, but not in persons with well-controlled diabetes (HbA_1c of < 7.0%) [3, 15]. However, in addition to microvascular diseases, glucose toxicity, which is induced by hyperglycemia, is believed to have other pathways to induce cognitive decline, such as the production of advanced glycation end products (AGEs) [16] and inflammation [17]. The levels of blood receptor of AGEs (i.e., RAGE levels) were changed in prediabetic patients associated with higher oxidative stress levels [18]. In an animal study, prediabetic rat models with cognitive decline exhibited neuroinflammation and neuronal apoptosis in the hippocampus [19]. Thus, glucose toxicity might occur even in prediabetes levels. With regard to the relationship between prediabetes and cognitive decline, the results are inconsistent among studies. It was reported that cognitive decline was significantly faster among middle-aged adults with prediabetes levels of HbA_1c than among those with normal HbA_1c levels during a 20-year follow-up period [20], which is in agreement with the findings of this study.

Conversely, two studies, one conducted on a middle-aged population and the other on an elderly population, have reported that adults with prediabetes had similar rates of cognitive decline to those with normoglycemia [3, 21]. Moreover, no clear association was reported between HbA_1c levels and the risk of AD and cognitive decline [22, 23]. Further studies are required to elucidate the pathogenesis of mild hyperglycemia, such as prediabetes or mild diabetes (i.e., HbA_1c levels of 6.5%–6.9%) in the development of dementia or cognitive decline. We demonstrated that elevated levels of GA were associated with all-cause dementia. It was reported that higher blood GA levels were significantly associated with global brain or hippocampal atrophy [24] as well as higher levels of oxidative stress [25]. GA has been found to reflect postprandial glucose fluctuations [4], and postprandial hyperglycemia has a more specific triggering effect on oxidative stress compared with chronic sustained hyperglycemia [26, 27]. Increased blood GA levels were significantly associated with a higher risk of developing dementia [20] and a trend of developing AD [22]; this is in agreement with the findings of this study. On the other hand, no association was observed between blood GA levels and the risk of all-cause dementia or AD [28]. Further epidemiological and basic studies are required to elucidate the relationships and underlying mechanisms between the levels of glycemic measures, such as HbA_1c and GA, and the risk of dementia.

In this study, no significant association was observed between the HOMA-IR and HOMA-β levels and all-cause dementia and its subtypes. Although animal studies have provided significant evidence that insulin resistance can promote the accumulation of amyloid-β and tau [29] and might be directly involved in the acceleration of amyloid pathology in the brain [30], the associations between insulin resistance and cognition or AD pathology in humans remain unclear. Insulin resistance was reported to be associated with cognitive decline among non-demented older adults [31]; however, no association was observed between the measures of peripheral insulin resistance and pathological features of AD in the postmortem brain [32]. These findings suggest that insulin resistance might contribute to developing mild cognitive impairment (MCI) or early stage of dementia of AD, however, there was no evidence of the associations between insulin resistance and late stage of dementia of AD. We speculate that insulin resistance might be associated with MCI or mild dementia due to AD, but not moderate or severe dementia due to AD. The present study was cross-sectional design, and considerable number of patients with dementia had moderate to severe dementia. The lack of association between insulin resistance and dementia might reflect the severity of dementia of the participants in this study. In addition, no study could elucidate the relationships between insulin secretion and AD pathology in humans.

In this study, no apparent relationships between diabetic status, levels of glycemic measures, and insulin resistance and secretion measures, and the presence of VaD were observed. The lack of association might reflect the small number of patients with VaD in this study; the number of patients with VaD was one-sixth of that of patients with AD [8]. The associations between diabetes and VaD remain controversial. Some studies reported that diabetic status and glycemic measures were significantly associated with VaD [33-36], while others reported non-significant associations [37, 38]. It was reported that diabetes was associated with pre-stroke dementia, but not associated with post-stroke dementia [37]. We diagnosed a participant as having mixed dementia or non-vascular dementia, who had a clinical history of dementia before the onset of stroke. This might be another reason for the lack of association between diabetic status and VaD. The strengths of this study are the large dataset used and the opportunity to adjust for possible confounding factors. However, this study also has several limitations. First, as the findings of this study were derived from cross-sectional data, estimation of a causal association among diabetic status, glycemic measures, and dementia was difficult. Second, when AD and VaD were diagnosed, we have evaluated neuroimaging studies, such as head MRI or CT, in almost every case but not all cases. Additionally, neuropathology was not examined in this study. Third, healthy screenee bias might exist because 2,281 of the total participants (n = 11,410) were selected by a voluntary response sampling.

In conclusion, the data in this study revealed that diabetes mellitus, higher plasma HbA_1c levels, even in prediabetes levels, and higher serum GA levels were related to the higher prevalence of all-cause dementia and AD. Our findings indicate that the primary prevention of diabetes or sufficient glucose control is important for the prevention of dementia or AD and that postprandial glucose fluctuations and postprandial hyperglycemia have a more specific effect on the development of cognitive decline. Further prospective longitudinal studies and basic research are required to verify the findings of this study.

PARTICIPATING INSTITUTES AND PRINCIPAL COLLABORATORS FOR THE JPSC-AD STUDY GROUP

Kyushu University ([Epidemiology and Public Health] Toshiharu Ninomiya, Jun Hata, Mao Shibata, Takanori Honda, [Neuropsychiatry] Tomoyuki Ohara, Shigenobu Kanba, [Ocular Pathology and Imaging Science] Masato Akiyama); Hirosaki University (Shigeyuki Nakaji, Kazushige Ihara, Koichi Murashita, Kaori Sawada, Songee Jung); Iwate Medical University (Tetsuya Maeda, Yasuo Terayama, Hisashi Yonezawa, Junko Takahashi, Hiroshi Akasaka); Kanazawa University (Masahito Yamada, Moeko Noguchi-Shinohara, Kazuo Iwasa, Tsuyoshi Hamaguchi, Kenji Sakai); Keio University School of Medicine (Masaru Mimura, Hidehito Niimura, Ryo Shikimoto, Hisashi Kida, Shogyoku Bun); National Hospital Organization Matsue Medical Center (Kenji Nakashima, Yasuyo Fukada and Hisanori Kowa); Kawasaki Medical School (Kenji Wada-Isoe); Tottori Prefectural Kousei Hospital (Masafumi Kishi); Ehime University (Takaaki Mori, Taku Yoshida, Hideaki Shimizu, Ayumi Tachibana, Shu-ichi Ueno); Kumamoto University (Minoru Takebayashi, Ryuji Fukuhara, Tomohisa Ishikawa, Asuka Koyama); Osaka University Graduate School of Medicine (Manabu Ikeda); Kindai University (Mamoru Hashimoto); National Cerebral and Cardiovascular Center (Yoshihiro Kokubo); Nakamura-Gakuen University (Kazuhiro Uchida, Midori Esaki); Tohoku University (Yasuyuki Taki, Yuji Takano); Kumagai Institute of Health Policy (Shuzo Kumagai); University of the Ryukyus (Koji Yonemoto); Osaka City University Graduate School of Medicine (Hisako Yoshida); University of Tokyo (Kaori Muto, Yusuke Inoue); RIKEN Center for Integrative Medical Sciences (Yukihide Momozawa, Chikashi Terao); Hisayama Research Institute for Lifestyle Diseases (Michiaki Kubo, Yutaka Kiyohara)

DATA AVAILABILITY

The datasets used in the current study are not publicly available, because they contain confidential clinical data on the study participants. However, the data are available on reasonable request and with the permission of the principal investigator, Toshiharu Ninomiya (Department of Epidemiology and Public Health, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan) and the steering committee of the JPSC-AD Study.

Footnotes

ACKNOWLEDGMENTS

We wish to thank all residents for their participation in this study.

This study was supported by the Japan Agency for Medical Research and Development (dk0207025) and Suntory Holdings Limited (Osaka, Japan). The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Biessels

, Staekenborg

, Brunner

, Brayne

, Scheltens

(2006) Risk of dementia in diabetes mellitus: A systematic review. Lancet Neurol 5, 64–74.

Ninomiya

(2014) Diabetes mellitus and dementia. . Curr Diab Rep 14, 487.

Zheng

, Ley

, Hu

(2018) Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. . Nat Rev Endocrinol 14, 88–98.

Kim

, Lee

(2012) The roles of glycated albumin as intermediate glycation index and pathogenic protein. . Diabetes Metab J 36, 98–107.

Iberg

, Fluckiger

(1986) Nonenzymatic glycosylation of albumin in vivo. Identification of multiple glycosylated sites. J Biol Chem 261, 13542–13545.

Koga

(2014) Glycated albumin; clinical usefulness. . Clin Chim Acta 433, 96–104.

Conwell

, Trost

, Brown

, Batch

(2004) Indexes of insulin resistance and secretion in obese children and adolescents: A validation study. Diabetes Care 27, 314–319.

Ninomiya

, Nakaji

, Maeda

, Yamada

, Mimura

, Nakashima

, Mori

, Takebayashi

, Ohara

, Hata

, Kokubo

, Uchida

, Taki

, Kumagai

, Yonemoto

, Yoshida

, Muto

, Momozawa

, Akiyama

, Kubo

, Ikeda

, Kanba

, Kiyohara

, JPSFC-AD Study Group (2020) Study design and baseline characteristics of a population-based prospective cohort study of dementia in Japan: The Japan Prospective Studies Collaboration for Aging and Dementia (JPSC-AD). Environ Health Prev Med 25, 64.

American Psychiatric Association (1987) Diagnostic and statistical manual of mental disorders, 3rd ed, revised. American Psychiatric Association Washington, DC.

10.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. . Neurology 34, 939–944.

11.

Roman

, Tatemichi

, Erkinjuntti

, Cummings

, Masdeu

, Garcia

, Amaducci

, Orgogozo

, Brun

, Hofman

, et al. (1993) Vascular dementia: Diagnostic criteria for research studies. Report of the NINDS-AIREN International Workshop. . Neurology 43, 250–260.

12.

American Diabetes A (2010) Diagnosis and classification of diabetes mellitus. Diabetes Care 33 Suppl 1, S62–69.

13.

Matthews

, Hosker

, Rudenski

, Naylor

, Treacher

, Turner

(1985) Homeostasis model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. . Diabetologia 28, 412–419.

14.

Sheik

, Yesavage

(1986) Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. . Clin Gerontol 5, 165–173.

15.

Rawlings

, Sharrett

, Albert

, Coresh

, Windham

, Power

, Knopman

, Walker

, Burgard

, Mosley

, Gottesman

, Selvin

(2019) The association of late-life diabetes status and hyperglycemia with incident mild cognitive impairment and dementia: The ARIC Study. . Diabetes Care 42, 1248–1254.

16.

Vlassara

, Palace

(2002) Diabetes and advanced glycation endproducts. . J Intern Med 251, 87–101.

17.

Yaffe

, Kanaya

, Lindquist

, Simonsick

, Harris

, Shorr

, Tylavsky

, Newman

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

18.

Huang

, Que

, Shen

(2015) Correlation of the plasma levels of soluble RAGE and endogenous secretory RAGE with oxidative stress in pre-diabetic patients. . J Diabetes Complications 29, 422–426.

19.

Fakih

, Mroueh

, Salah

, Eid

, Obeid

, Kobeissy

, Darwish

, El-Yazbi

(2020) Dysfunctional cerebrovascular tone contributes to cognitive impairment in a non-obese rat model of prediabetic challenge: Role of suppression of autophagy and modulation by anti-diabetic drugs. Biochem Pharmacol 178, 114041.

20.

Rawlings

, Sharrett

, Schneider

, Coresh

, Albert

, Couper

, Griswold

, Gottesman

, Wagenknecht

, Windham

, Selvin

(2014) Diabetes in midlife and cognitive change over 20 years: A cohort study. . Ann Intern Med 161, 785–793.

21.

Tuligenga

, Dugravot

, Tabak

, Elbaz

, Brunner

, Kivimaki

, Singh-Manoux

(2014) Midlife type 2 diabetes and poor glycaemic control as risk factors for cognitive decline in early old age: A post-hoc analysis of the Whitehall II cohort study. Lancet Diabetes Endocrinol 2, 228–235.

22.

Mukai

, Ohara

, Hata

, Hirakawa

, Yoshida

, Kishimoto

, Koga

, Nakamura

, Kitazono

, Kiyohara

, Ninomiya

(2017) Alternative measures of hyperglycemia and risk of Alzheimer’s disease in the community: The Hisayama Study. J Clin Endocrinol Metab 102, 3002–3010.

23.

Christman

, Matsushita

, Gottesman

, Mosley

, Alonso

, Coresh

, Hill-Briggs

, Sharrett

, Selvin

(2011) Glycated haemoglobin and cognitive decline: The Atherosclerosis Risk in Communities (ARIC) study. . Diabetologia 54, 1645–1652.

24.

Ohara

, Furuta

, Hirabayashi

, Hata

, Hirakawa

, Honda

, Yoshida

, Shibata

, Kitazono

, Ninomiya

(2020)J Diabetes Investig 11, 971–979 Elevated serum glycated albumin and glycated album, hemoglobin A 1c ratio were associated with hiocampal atrophy in a general elderly population of Japanese: The Hisayama Study. .

25.

Kohata

, Ohara

, Nagaike

, Fujikawa

, Osaka

, Goto

, Fukase

, Kushima

, Hiromura

, Terasaki

, Mori

, Fukui

, Ouchi

, Suzuki

, Hirano

, Yamagishi

(2020) Association of hemoglobin A1c, 1,5-anhydro-D-glucitol and glycated albumin with oxidative stress in type 2 diabetes mellitus patients: A cross-sectional study. . Diabetes Ther 11, 655–665.

26.

Monnier

, Mas

, Ginet

, Michel

, Villon

, Cristol

, Colette

(2006) Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. . JAMA 295, 1681–1687.

27.

Ceriello

, Esposito

, Piconi

, Ihnat

, Thorpe

, Testa

, Boemi

, Giugliano

(2008) Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. . Diabetes 57, 1349–1354.

28.

Van Himbergen

(2012) Biomarkers for insulin resistance and inflammation and the risk for all-cause dementia and Alzheimer disease. Arch Neurol 69, 594.

29.

Verdile

, Fuller

, Martins

(2015) The role of type 2 diabetes in neurodegeneration. . Neurobiol Dis 84, 22–38.

30.

Wakabayashi

, Yamaguchi

, Matsui

, Sano

, Kubota

, Hashimoto

, Mano

, Yamada

, Matsuo

, Kubota

, Kadowaki

, Iwatsubo

(2019) Differential effects of diet- and genetically-induced brain insulin resistance on amyloid pathology in a mouse model of Alzheimer’s disease. Mol Neurodegener 14, 15.

31.

Laws

, Gaskin

, Woodfield

, Srikanth

, Bruce

, Fraser

, Porter

, Newsholme

, Wijesekara

, Burnham

, Dore

, Li

, Maruff

, Masters

, Rainey-Smith

, Rowe

, Salvado

, Villemagne

, Martins

, Verdile

(2017) Insulin resistance is associated with reductions in specific cognitive domains and increases in CSF tau in cognitively normal adults. Sci Rep 7, 9766.

32.

Thambisetty

, Metter

, Yang

, Dolan

, Marano

, Zonderman

, Troncoso

, Zhou

, Wong

, Ferrucci

, Egan

, Resnick

, O’Brien

(2013) Glucose intolerance, insulin resistance, and pathological features of Alzheimer disease in the Baltimore Longitudinal Study of Aging, 70, 1167.

33.

, Qiu

, Wahlin

, Winblad

, Fratiglioni

(2004) Diabetes mellitus and risk of dementia in the Kungsholmen project: A 6-year follow-up study. . Neurology 63, 1181–1186.

34.

Peila

, Rodriguez

, Launer

, Honolulu-Asia Aging

(2002) Type 2 diabetes, APOE gene, and the risk for dementia and related pathologies: The Honolulu-Asia Aging Study. . Diabetes 51, 1256–1262.

35.

Shang

, Fratiglioni

, Marseglia

, Plym

, Welmer

, Wang

, Xu

(2020) Association of diabetes with stroke and post-stroke dementia: A population-based cohort study. . Alzheimers Dement 16, 1003–1012.

36.

Wium-Andersen

, Rungby

, Jorgensen

, Sandbaek

, Osler

, Wium-Andersen

(2019) Risk of dementia and cognitive dysfunction in individuals with diabetes or elevated blood glucose. Epidemiol Psychiatr Sci 29, e43.

37.

Pendlebury

, Rothwell

(2009) Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: A systematic review and meta-analysis. . Lancet Neurol 8, 1006–1018.

38.

Allan

, Rowan

, Firbank

, Thomas

, Parry

, Polvikoski

, O’Brien

, Kalaria

(2011) Long term incidence of dementia, predictors of mortality and pathological diagnosis in older stroke survivors. Brain 134, 3716–3727.

Diabetes Mellitus,Elevated Hemoglobin A 1c,and Glycated Albumin Are Associated with the Presence of All-Cause Dementia and Alzheimer’s Disease: The JPSC-AD Study

Abstract

Background:

Objective:

Methods:

Results:

Conclusion:

Keywords

INTRODUCTION

MATERIALS AND METHODS

Study population

Standard protocol approvals and participant consent

Diagnosis of dementia and its subtypes

Diagnosis of diabetes mellitus and prediabetes

Measurements of HbA1c, GA, HOMA-IR, and HOMA-β

Other risk factor measurements

Statistical analyses

RESULTS

DISCUSSION

PARTICIPATING INSTITUTES AND PRINCIPAL COLLABORATORS FOR THE JPSC-AD STUDY GROUP

DATA AVAILABILITY

Footnotes

ACKNOWLEDGMENTS

References

Measurements of HbA_1c, GA, HOMA-IR, and HOMA-β