The Association of Vitamin D Status with Mild Cognitive Impairment and Dementia Subtypes: A Cross-Sectional Analysis in Dutch Geriatric Outpatients

Abstract

Background:

Vitamin D deficiency is associated with all-cause dementia and Alzheimer’s disease (AD). At the same time, this knowledge is limited specifically for vascular dementia (VaD), while data regarding other subtypes of dementia are even more limited.

Objective:

To investigate the association of 25-hydroxy vitamin D (25(OH)D) status with dementia subtypes in an outpatient geriatric population.

Methods:

In a cross-sectional design, we analyzed data from 1,758 patients of an outpatient memory clinic in The Netherlands. Cognitive disorders were diagnosed by a multidisciplinary team according to international clinical standards. At each first-visit 25(OH)D levels were measured. Data were analyzed using ANCOVA in four models with age, gender, BMI, education, alcohol, smoking, season, polypharmacy, calcium, eGFR, and glucose as co-variates. 25(OH)D was treated as a continuous square rooted (sqr) variable.

Results:

In the fully adjusted model, reduced 25(OH)D serum levels (sqr) were found in AD (estimated mean 7.77±0.11 CI95% 7.55-7.99): and in VaD (estimated mean 7.60±0.16 CI95% 7.28-7.92) patients compared to no-dementia (ND) patients (estimated mean 8.27±0.09 CI95% 8.10-8.45) (ND-AD: p = 0.006, CI95% 0.08-0.92.; ND-VaD p = 0.004 CI95% 0.13-1.22). We did not find differences in 25(OH)D levels of mild cognitive impairment (MCI) or other dementia patients compared to ND patients, nor differences in comparing dementia subtypes.

Conclusion:

We observed significantly lower 25(OH)D serum levels in both AD and VaD patients compared to no-dementia patients, but no significant differences between MCI and Lewy body and mixed dementia subtypes in this cross-sectional study of a geriatric outpatient clinic population.

Keywords

25(OH)D Alzheimer’s disease cognition dementia Lewy body dementia mild cognitive impairment vascular dementia vitamin D

INTRODUCTION

In 2019, the World Health Organization reported that 50 million people were living with dementia and this number is expected to be tripled in 2050 [1]. Dementia commonly refers to a collective term for a number of clinically and pathophysiologically different disorders. Alzheimer’s disease (AD) is the most common type of dementia corresponding to 60-70% of the cases, while vascular dementia (VaD) is the second with 10-15% of all cases. Other forms include Lewy body dementia (LBD), frontal lobe dementia (FLD), or are classified as mixed dementia (MD).

Dementia is commonly preceded by a period of increasing cognitive impairment. A number of risk factors for developing cognitive impairment and dementia are currently known and include age, genetic predisposition, pre-existing health conditions (in particular hypertension and other cardiometabolic risk factors), environmental and societal factors [2]. Within the field of nutrition, research focuses on dietary patterns, food groups, and individual nutritional components [3 –5]. With regards to food patterns, following a Mediterranean diet has been shown to improve cognitive function [6]. Within the group of individual nutritional components, monounsaturated fatty acids and n-3 polyunsaturated fatty acids were studied most with positive or neutral effects [3]. However, although in the past decade multiple observational studies have shown associations between diet-related modifiable risk factors and cognitive impairment and dementia, the possible relationships between nutrition, and in particular micronutrient status, and dementia are still under debate [7, 8]. One of the micronutrients that has been in the spotlight for some time is vitamin D, of which a lower status has been linked to the development of dementia and AD. However, several important questions remain, especially with regard to the definition of low status of the main circulating metabolite, 25(OH)D, and to what extent a distinction should be made between the different dementia subtypes.

According to the nowadays most commonly used cut-off values for 25(OH)D status (i.e., deficiency < 50 nmol/L plasma 25(OH)D and insufficiency < 75 nmol/L), about one billion people worldwide would have an insufficient 25(OH)D status [9 –11]. Older adults would be particularly at risk with a prevalence ranging from 50% to approximately 80% [12]. It should be noted, however, that the current consensus about optimal 25(OH)D status is primarily based on bone health outcomes. Little is still known about the roles of this prohormone in the brain and the associated optimal 25(OH)D status. Both the prevalence rates for dementia and deficiency of vitamin D increase with age [9 , 14]. However, 25(OH)D serum levels in elderly are dependent on the diminishing capacity of the aging skin to convert previtamin D3 into vitamin D under influence of sunlight [15]. Therefore, a causal relationship is not certain. However, in the past years several mendelian randomization studies have shown that genetically increased 25(OH)D levels are associated with reduced risk of developing AD [16 –20]. These studies make a causal effect more probable. Studies that apply other methodology have also found significantly lower serum 25(OH)D concentrations in persons with AD compared to healthy controls, but the results are conflicting [13 , 21–29]. More recently, a number of meta-analyses and systematic reviews on the association between 25(OH)D levels and cognitive impairment have been performed. Although their outcomes are not always consistent, they mostly point toward an association between lower 25(OH)D serum levels and dementia or AD [30 –34]. However, these meta-analyses also identify significant heterogeneity in study populations, methods of 25(OH)D assessment and dementia diagnosis among studies included. Besides this, many studies do not pay sufficient attention to over-the-counter preparations, the use of which is often under-reported. Finally, these meta-analyses fall back on a small number of cross-sectional and cohort studies [25 , 36]. As hypovitaminosis D often occurs before the first objective signs of AD and accompanies the onset of the first symptoms, it has been suggested that hypovitaminosis D may contribute to the initiation of dementia [37]. Up until now clinical trials and randomized controlled trials (RCTs) showed no or modest effects of supplementation on cognition in both healthy subjects and in mild cognitive impairment (MCI) and dementia patients [38 –46], but it should be noted that the intervention in many of these trials and RCTs was not only with vitamin D. Furthermore, there are other methodological issues like different dosage schemas in inclusion of patients without vitamin D deficiency [47].

Finally, only very few studies make a distinction between dementia subtypes in their analysis. Although one of the suggested mechanisms involves vascular etiology, evidence of an association between 25(OH)D status and VaD is scarce [48].

To address these knowledge gaps, the present study aims to further investigate the associations of 25(OH)D status with MCI and dementia subtypes. Our observational study used a well-defined patient population in which MCI and dementia subtypes were diagnosed according to the same internationally accepted procedures and criteria. Furthermore, both prescribed and non-prescribed use of vitamin D supplementation was included in the analysis, which is in contrast with most of the studies performed in the past.

METHODS

Study population

A total number of 2,368 participants with a cognitive diagnosis were enrolled in this cross-sectional study from January 2011 until June 2015 (Fig. 1). All participants were referred to the out-patient clinic of the geriatric department in a non-academic hospital in Ede, the Netherlands. Participants were consigned for cognitive problems, falls, or other reasons. When there were cognitive complaints in the latter two situations, patients underwent the standardized cognitive analysis as well. Participants were included if they were aged above 60 years. They were excluded when their 25(OH)D status was either unknown or determined more than 3 months prior to the out-patient clinic visit. Furthermore, patients with dementias not otherwise specified/of uncertain etiology were excluded. Because of the small number (n = 15), of patients with FLD in this group, these were excluded as well. The research protocol was ethically approved by the Assessment Committee of Scientific Research (BCWO) of the hospital (registration number 1612-657). This study was performed in line with the principles of the Declaration of Helsinki.

Fig. 1

Flowchart of the study population.

Diagnostic process

Data were collected by a standardized diagnostic process. Patients, in presence of their caregivers were examined by a geriatrician, a neurologist and a specialized nurse according to the Comprehensive Geriatric assessment. If it was necessary to come to a diagnosis, neuropsychological testing by a neuropsychologist and radiological investigations were performed during a second visit. Thereafter, clinical diagnoses were formulated during a multidisciplinary meeting of all team members and communicated by the geriatrician or neurologist to the patient finally.

At inclusion, global cognitive screening was done using the widely accepted Mini-Mental State Exam and Clock drawing tests [49, 50]. Assessment of cognitive functioning was based on the Dutch guidelines (based on international literature) for a comprehensive geriatric assessment [51, 52] and Dutch guidelines for diagnostics and treatment of dementia [53]. The diagnosis of MCI was based on the Petersen criteria [54] and comprised amnestic and non-amnestic MCI. The diagnosis of dementia was based on the DSM IV-TR and the criteria by McKahnn et al. [55]. The subtype of dementia was also diagnosed according to international criteria (NINCDS-ADRDA (AD), NINDS-AIREN (VaD) and 2017 revised DLB Consortium criteria (DLB), Rascovsky-criteria (FTLD)).

25(OH)D status

Blood samples were drawn directly after the visit to the out-patient-clinic. After clotting and centrifugation serum samples were frozen at -20°Celsius and once a week analyzed with a 25(OH) Vitamin D3/D2 Reagent Kit for HPLC analysis (Chromsystems Instruments & Chemicals, Gräfelfing, Germany).

Other measurements

Height was measured to the nearest 0.1 cm, with a telescopic height rod (Seca 220). Weight was measured in kilograms, to the nearest 0.1 kg, using a digital floor scale (Seca 770). A chair scale (Seca D94-09-033) was used if it was not possible or difficult for participants to stand. The day of blood collection was recorded into the variable season: winter (October– March) or summer (April– September) based on previous studies by Brouwer-Brolsma et al. [56, 57]. Education attainment was classified as either primary school (≤6 years) or post-primary education and higher (>6 years), smoking as never, ever, and current. Categorization of alcohol use was based on the Alcohol Consumption Index according to Garretsen: not/light, moderate, excessive/very excessive [58]. Drug use was assessed by (hetero)anamnesis and medication lists from pharmacies and patients were asked to bring their drugs and supplements to the clinic. Non-compliance of their drug use was structurally asked for. The variable drug use in this study reflects the total number of ATC-coded substances. ATC-codes of vitamin D supplements and (multi)vitamin supplements containing vitamin D were registered separately. Using more than five prescriptions was defined as polypharmacy and more than ten prescriptions was defined as severe polypharmacy [59].

Other blood parameters

Serum creatinine (eGFR, to estimate kidney function), serum glucose (glucose), and serum calcium (calcium) levels were all determined using colorimetric assays (Siemens Healthcare Diagnostics Inc. Erlangen, Germany). The ECREA-method was used for determination of serum creatinine levels. Serum creatinine levels were needed for calculation of the estimated GFR, which is a measure of kidney function. The GFR was estimated using the MDRD-formula [60]. Serum calcium levels were measured using a bichromatic (577, 540 nm) endpoint technique. Fasting glucose levels were also analyzed using a bichromatic (340, 383 nm) endpoint measurement.

Data entry

Data entry concerning clinical diagnosis was performed by two assessors. Raw patient data were extracted from the electronic patient file system NORMA/NeoZIS and then further processed. Clinical diagnoses were first checked independently and thereafter 10% was double checked to reduce errors. Inconsistent or unclear clinical diagnoses were triple checked by a geriatrician.

The cognitive diagnoses were coded into nine different categories: no dementia, MCI, dementia with etiological classification (AD, VaD, MD, LBD, and FLD), dementia not otherwise specified (DNOS), and other cognitive diagnoses (like cognitive disorders NOS, memory complaints, and participants suffering from Parkinson’s disease and other physical disorders). Participants were considered having no dementia when no cognitive disorders nor dementia were found during the analysis.

Statistical analysis

The statistical analysis was carried out using the statistical software program IBM SPSS Statistics version 27. First, data were checked for normality using Q-Q-plots, histograms and if needed followed by Shapiro-Wilk tests. A p-value<0.05 was considered significant. Baseline characteristics are presented using means and standard deviation if normally distributed. When data were not normally distributed, the median and interquartile range are displayed. All categorical variables are displayed in percentages. Serum 25(OH)D was square root-transformed because of non-normality. Differences in baseline characteristics were tested using one-way ANOVA for normally distributed continuous variables, Kruskal-Wallis for non-normally distributed continuous variables, and chi-square tests for categorical variables. A p-value<0.05 was considered significant. Interaction was tested between each of the covariates and 25(OH)D status. The primary outcome variable was clinical diagnosis of cognitive disorders, which was treated as a categorical variable and was split up into categories: ‘no dementia’, ‘MCI’, and four etiological categories of ‘dementia’. ANCOVA-analysis was used to examine significant differences in adjusted means of square rooted (sqr) serum 25(OH)D and the clinical diagnosis. To be sure ANCOVA-analysis was appropriate, significant interaction of vitamin D supplementation with cognitive diagnosis was excluded. Three models were used: 1) only age and gender adjusted, 2) additionally adjusted for BMI, education, smoking and alcohol use, season, and polypharmacy; and 3) additionally adjusted for calcium, eGFR, and glucose. The adjustments were based on literature and if a significant difference (p≤0.05) of a variable was found between the 6 cognitive categories in the baseline characteristics analysis.

RESULTS

Characteristics

Of 2,368 patients who underwent cognition analysis, 1,758 were enrolled in the study (Fig. 1).

The characteristics of the study population are shown in Table 1. After analysis, 31% of the patients had no objectivated cognitive disorder. Of the patients in this study, 20% were diagnosed with MCI and 49% with dementia. Of these dementia patients, the vast majority had AD (49%). The other dementia subtypes were VaD (19%), MD (28%), and LBD (5%). There was no significant difference between the groups regarding serum calcium, serum glucose, and the season in which the serum levels of 25(OH)D were measured. All other parameters differed significantly between the groups. Within the group diagnosed with a form of dementia, there were fewer women represented in the VaD (43%) and LBD (40%) group than in the AD (66%) and MD group (61%).

Table 1

Characteristics of a population of 1,758 Dutch geriatric outpatients, aged 60 years and older

	Total	No cognitive disorder (n = 542)	MCI (n = 358)	AD (n = 420)	VaD (n = 159)	MD (n = 236)	LBD (n = 43)	p
Gender (% women)	58	64	50	66	43.4	61.4	39.5	<0.001
Age (y)	79.2±7.2	78.4±7.8	76.6±7.2	80.9±6.3	80.1±6.7	81.3±6.4	80.1±6.4	<0.001
BMI (kg/m²)¹	26.2 (23.6-29.4)	27.1 (24.1-30.1)	26.6 (23.6-29.7)	25.8 (23.2-28.5)	26.1 (22.8-29.4)	25.8 (23.3-28.5)	28.1 (25.8-30.4)	<0.001
Alcohol use (%)² unknown	11.3	14.0	4.5	11.7	13.8	12.3	14.0	<0.001
Not/light	75.6	73.2	76.3	80.7	71.1	74.2	74.4
moderate	10.9	9.2	15.9	6.9	13.8	11.9	11.6
(very) excessive	2.3	3.5	3.4	0.7	1.3	1.7	0
Smoking (%) unknown	11.2	13.8	4.5	11.4	14.5	12.3	14.0	<0.001
Never	60.3	59.6	57.5	66.2	55.3	58.9	60.5
Former	19.	16.2	27.9	16.7	20.8	20.3	25.6
Current	8.6	10.3	10.1	5.7	9.4	8.5	0
Education (%) unknown	13.9	22.3	5.0	12.1	12.6	12.7	11.6	<0.001
≤primary school	26.9	23.4	19.3	32.6	28.3	35.2	27.9
>primary school	59.2	54.2	75.7	55.2	59.1	52.1	60.5
Glucose (mmol/l)	5.7 (5.3-6.6)	5.8 (5.1-6.7)	5.6 (5.0-6.3)	5.7 (5.0-6.4)	5.7 (5.0-6.4)	5.8 (5.2-6.5)	5.9 (5.2-6.6)	0.567
eGFR (ml/kg/min) % <60⁴	27.4	25.3	25.5	26.9	38.5	29.2	23.3	0.027
Calcium (mmol/l)	2.40±0.12	2.40±0.14	2.39±0.11	2.39±0.11	2.39±0.14	2.39±0.11	2.38±0.13	0.446
Polypharmacy (%) ³								<0.001
0-4	35.2	28.4	38.9	45.6	24.5	35.7	23.3
5-10	43.9	45.4	41.2	40.8	48.4	43.4	65.1
>10	20.9	26.2	19.9	13.6	27.0	20.9	11.6
Vit D supplement (% use≥1)	30.3	35.8	27.9	22.6	40.3	28.0	30.2	<0.001
Serum 25(OH)D (nmol/L)	57.0 (35.0-82.0)	66.0 (42.0-90.0)	61.0 (40.5-81.5)	49.0 (28.5-69.5)	51.0 (25.5-76.5)	53.0 (28.0-78.0)	50.0 (28.5-72.0)	<0.001
Season (% summer)	55.4	57.2	55.6	52.4	52.8	57.2	58.1	0.655

The interaction effect of clinical cognitive diagnosis and vitamin D supplementation was analyzed with boxplots, scatterplots, and Q-Q-plots. No difference in variance of sqr 25(OH)D level between the cognitive diagnosis groups was observed in the vitamin D using and non-using groups. This was confirmed by Levene’s tests which were not significant for all three models (Model 1: F (11, 1746)=0.98, p = 0.46; Model 2: F (11,1718)=0.88, p = 0.56; Model 3: F (11,1478)=1.45, p = 0,15). This indicates homogeneity of variance between the diagnostic categories.

A one-way ANCOVA showed a significant difference between sqr 25(OH)D and cognitive status for all three models (Model 1: F (5,1744)=3.62, p = 0.003; Model 2: F (5,1710)=4.22, p = 0.001; Model 3: F(5, 1467)=4.92, p < 0.001). The interaction between vitamin D supplementation use and cognitive categories was not significant in all models (model 1. F (5,1744)=1.16, p = 0.18; Model 2: F (5,1710)=1.53, p = 0.18; Model 3: F (5,1467)=1.51, p = 0.19).

Table 2 presents the estimated means of square rooted 25(OH)D by cognitive category and vitamin D (non-)use for the crude and adjusted models. Vitamin D supplementation was used as a co-factor in this analysis. Pairwise comparisons were performed with correction for multiple testing by Bonferroni adjustment.

Table 2

Estimated mean values and standard errors of sqr 25(OH)D by cognitive diagnosis*vitamin D supplement use (CI95%)

	Crude model	Model 1	Model 2	Model 3
Category
ND	8.20±0.08 (8.04-8.36)	8.18±0.08 (8.02-8.33)	8.18±0.08 (8.02-8.34)	8.27±0.09 (8.10-8.45)
MCI	8.19±0.11 (7.98-8.40)	8.11±0.11 (7.90-8.32)	8.09±0.11 (7.88-8.30)	8.17±0.11 (7.95-8.40)
AD	7.74±0.11 (7.54-7.95)	7.80±0.11 (7.60-8.01)	7.79±0.11 (7.58-7.99)	7.77±0.11 (7.55-7.99)
VaD	7.57±0.15 (7.28-7.85)	7.61±0.15 (7.33-7.90)	7.55±0.15 (7.26-7.84)	7.60±0.16 (7.28-7.92)
MD	7.90±0.13 (7.64-8.16)	7.97±0.13 (7.72-8.23)	7.93±0.13 (7.67-8.19)	8.11±0.14 (7.83-8.39)
LBD	7.52±0.30 (6.93-8.10)	7.58±0.30 (6.99-8.16)	7.55±0.30 (6.95-8.15)	7.52±0.30 (6.93-8.12)
Adj p (CI 95%)^a
ND versus AD	p=0.008 (0.071-0.851)	p=0.07 (-0.017-0.763)	p=0.04 (0.006-0.785)	p=0.006 (0.084-0.918)
ND versus VaD	p=0.002 (0.145-1.126)	p=0.01 (0.071-1.053)	p=0.003 (0.136-1.124)	p=0.004 (0.127-1.215)
MCI versus AD	p=0.04 (0.011-0.890)	p=0.63 (-0.137-0.753)	p=0.66 (-0.140-0.746)	p=0.18 (-0.068-0.875)
MCI versus VaD	p=0.008 (0.095-1.156)	p=0.09 (-0.0834-1.027)	p=0.046 (0.005-1.071)	p=0.06 (-0.010-1.156)

Data are presented as adjusted mean±standard error. Model 1 was adjusted for age and gender; model 2 was adjusted for age, gender, BMI, education, alcohol, smoking, season, polypharmacy, model 3 was adjusted for age, gender, BMI, education, alcohol, smoking, season, polypharmacy, calcium, eGFR, glucose. ND, no-dementia; MCI, mild cognitive impairment; AD, Alzheimer’s disease; VaD, vascular dementia; MD, mixed dementia; LBD, Lewy body dementia. ^aAdjusted p value using Bonferroni correction. The covariates gender, education, alcohol intake, polypharmacy, calcium, glucose, and GFR were not statistically significant related, indicating these covariates had no significant effect on square rooted 25(OH)D. The co-variates age (F (1,1467)=38.03, p < 0.001), BMI (F (1,1467)=15.12, p < 0.001), season (F (1,1467)=21.79, p < 0.001), and smoking (F (1,1467)=9.52, p = 0.002) were significant related, indicating that these covariates had a significant effect on square rooted 25(OH)D. The effect was moderate. There was a negative relationship between on the one hand age, BMI, season, and smoking and on the other hand sqr 25(OH)D.

A significantly higher 25(OH)D estimated mean was found in all models for patients in the no-dementia group compared to the AD (crude model: p = 0.008, CI95% 0.07-0.85; full model: p = 0.006, CI95% 0.08-0.92) and VaD group (crude model: p = 0.002, CI95% 0.15-1.13; full model: p = 0.004, CI95% 0.13-1.22). Also, a significantly higher 25(OH)D estimated mean was found for MCI-patients compared to the AD patients (crude model: p < 0.04, CI95% 0.01-0.89, but this effect did not remain in the three adjusted models. The significantly higher 25(OH)D estimated mean of the MCI-patients compared to VaD patients in de crude model (p = 0.008, CI95% 0.10-1.16), disappeared in the fully adjusted model. No other significant differences were found in the crude and adjusted models.

DISCUSSION

In our study, we compared serum 25(OH)D levels in a large population of Dutch geriatric outpatients without dementia or with a diagnosis of MCI and AD, VaD, LBD, and MD. We observed significantly lower mean 25(OH)D serum levels in AD patients compared to no-dementia patients. A similar observation was made at comparing VaD patients to no-dementia patients. By contrast we did not find significant associations between 25(OH)D serum levels in MCI, LBD, nor MD patients compared to no-dementia patients. No further significant differences were observed comparing sub-groups.

Our results in the AD group of patients are in line with many other observational studies and a recent meta-analysis [61] that analyzed five prospective cohort studies. A meta-analysis of Chai et al. including cross sectional studies and prospective cohort studies, showed the same association between lower 25(OH)D levels and general dementia and AD [31].

In contrast to studies in AD patients, data regarding VaD patients are only scarce. In our study we also observed significantly lower 25(OH)D levels in VaD patients compared to no-dementia patients. We found one prospective cohort study of Afzal et al. which did not find an increased risk to develop VaD in patients (n = 92) with lower 25(OH)D serum levels [48]. Although Soysal et al. recently reported an increased risk of malnutrition in VaD patients, they did not find a significant difference in 25(OH)D levels in VaD patients [62]. The number of VaD patients in that study (n = 51) was small compared to our study (n = 159). Vitamin D deficiency is associated with vascular damage, due to a presumed endothelium modification [63]. Furthermore, neurovascular degeneration processes in the brain could be positively influenced by vitamin D, through preserving calcium homeostasis and modulating immune system and inflammatory processes. [64 –67]. This would support a role of vitamin D deficiency and development of VaD.

We did not find significant differences between 25(OH)D serum levels in LBD patients and no-dementia patients, nor in MD patients compared to no-dementia patients. Soysal et al. found a higher risk of malnutrition in LBD patients compared to other dementias. However, although elderly people are more dependent on oral intake of vitamin D due to a lower endogenous synthesis, they did not find a difference in 25(OH)D serum level in LBD patients compared to healthy persons [62]. Overall, literature data on vitamin D status specifically for LBD and MD patients are very scarce.

Our study did not find significant differences between MCI and all other cognitive categories including no-dementia patients. This is in contrast with a review of Etgen et al. who reported a significantly lower 25(OH)D serum level in MCI patients compared to healthy subjects [68]. A plausible explanation is that the groups may differ between studies. MCI is a heterogeneous etiologic category, and, in our study, we did not differentiate between amnestic and non-amnestic MCI’s or MCI patients with other causes (e.g., tumors, psychological causes, depressions).

To compare our study to other studies on all-cause dementia, we performed an analysis without subdividing the dementia group into etiological categories. In this analysis, we also found a significant difference in 25(OH)D serum level between no-dementia patients and all-cause dementia patients. Therefore, as the composition of an all-cause dementia group may differ, our study results show the importance of distinguishing etiological categories of dementia. This may explain why some observational and prospective studies did not find a difference between 25(OH)D levels of no-dementia and dementia groups.

Many cross sectional and cohort studies used the Mini-Mental State Exam or Montreal Cognitive Assessment screening test, with or without the Cambridge Cognition Examination to confirm diagnosis of dementia or AD. In other studies, it is often unclear how the diagnosis was made. These differences and lack of information limit comparison of studies, an issue which is also mentioned by many authors of reviews and meta-analyses. In our present study the diagnosis of cognitive status was made by an internationally accepted standardized diagnostic process using multiple (neuropsychological) tests and confirmation by a multidisciplinary team. A recent meta-analysis stressed the heterogeneity of the studies concerning the lack of neuropsychological test results as a source of bias [69]. Chai et al. found a significant association between 25(OH)D status and cognitive status when neuropsychological tests were added to their analysis [31]. Our study reduces this possible bias in diagnosis.

Moreover, only a minority of studies report on vitamin supplement use, especially when it comes to over-the-counter use of vitamin D. Although we do not have data on duration of consumption nor prescription dosage, 30% of the participants in our study used vitamin D. As older people are more dependent of dietary intake of vitamin D and international guidelines advice on the use of vitamin D supplementation to reduce fracture and osteoporosis risk, it is important to correct for all forms of supplementation use in the statistical analysis. Our patients were asked for over-the-counter supplementation as well as for prescriptions. Therefore, we could reliably take use of vitamin D supplementation into account.

Another problem with interpretation of the conflicting results in literature could be caused by lack of standardization of 25(OH)D analysis. In many studies, the process of handling the samples is not clear and the analytical techniques used are different. Recently, Sempos et al. clearly showed that some analytical techniques are more vulnerable for introducing errors due to technique, sample handling, and transport than others [70]. The meta-analysis of Chai et al. showed that none of the studies included the methods of collection, preservation, or duration of storage of samples prior to 25(OH)D analysis was described and analytical techniques differed [31]. In our study, samples were drawn at the same moment as the primary anamnesis of patients and handled the same day. Once a week the assays were done. In the meantime, the samples were deep-frozen to optimally maintain 25(OH)D stability.

Furthermore, many studies use internationally accepted cut-off values for 25(OH)D deficiency, insufficiency, and sufficiency that are used in studies after osteoporosis risk instead of a continuous variable like our study. Vitamin D receptors to which its active form, 1,25(OH)₂D can bind are present in different brain areas. Brain tissue also contains the enzyme to convert 25(OH))D into active 1,25 (OH)₂D [71 –73]. This supports the idea that vitamin D may have neurosteroid actions in specific brain regions which are essential for cognition [23]. Processes important to normal brain homeostasis and development could be influenced by active vitamin D [64 –67]. Animal studies have shown that vitamin D deficiency could promote forming of amyloid-β plaques [74]. Another explanation might be the effect on neuronal repair by influencing calcium channels that are damaged by Aβ peptides or by an antioxidant effect [74, 75]. These factors may contribute to the development of AD. Although this evidence suggests that vitamin D might have an important function in the brain, it is not known which serum levels do correspond with sufficient or insufficient levels in the brain itself. In our study all 25(OH)D levels are on average in the insufficient (osteoporosis fracture risk) range.

Besides the strengths of our study that we mentioned above, there are some weaknesses in our study. Firstly, the cross-sectional design itself does not allow to conclude to a causal relationship between lower 25(OH)D levels and AD and VaD. To investigate causal relationships, long term prospective studies or randomized controlled trials are needed.

Secondly, sunlight exposure is a main source of vitamin D in humans. Therefore, exposure could be important to correct for in the analysis. Because of apathy caused by dementia, there might be a significant difference between 25(OH)D status in no-dementia patients and dementia patients. Unfortunately, in our study we did not have data on sunlight exposure. However, we did not find a significant difference between 25(OH)D levels in summer and winter periods in our population. Therefore, we assume the sun-exposure effect is small in our study. A recent study from a group of colleagues [76] showed a significant increase in 25(OH)D levels during the summer season in a group of very active Dutch elderly people training for and participating in a 4-day walking event. Because our controls are patients referred to the out-patients clinic, they are likely to be more frail than this very healthy aged population mentioned above.

Thirdly, although a clear significantly lower 25(OH)D status in AD and VaD patients was observed, clinical relevance remains uncertain. When we back transform sqr 25(OH)D levels, there is a small (up to 9 nmol/L) difference in 25(OH)D levels. However, both group means are already below serum target levels of 25(OH)D as specified by the Dutch National Health Council and other advisory bodies as well as the scientific literature. Up until now prospective studies showed inconsistent result in effects and trials report mostly negative results [38 –43].

Conclusion

In conclusion, we found significantly lower 25(OH)D levels in patients with AD and VaD compared to no-dementia patients, but not in MCI, LBD, and MD patients. This is consistent with prior cross-sectional and some prospective studies regarding AD patients. Regarding the scarce evidence on 25(OH)D serum levels in VaD patients our study adds new information and finally this study provides new evidence in LBD patients. A thorough diagnostic process including neuropsychological testing and transparency about handling 25(OH)D samples is important for reliable results in future trials. At last, more fundamentally, research is needed to determine which 25(OH)D levels in the brain are sufficient or insufficient.

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgments to report.

FUNDING

The authors have no funding to report.

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

DATA AVAILABILITY

The data supporting the findings of this study are available on request from the corresponding author.

References

WHO, Factsheet on Dementia, https://www.who.int/news-room/fact-sheets/detail/dementia

Alzheimer’s Association, Dementia in Europe Yearbook 2019 Estimatingthe prevalence of dementia in Europe, https://www.alzheimer-europe.org/sites/default/files/alzheimer_europe_dementia_in_europe_yearbook_2019.pdf

Scarmeas

, Anastasiou

, Yannakoulia

(2018) Nutrition and prevention of cognitive impairment. Lancet Neurol 17,1006–1015.

Vauzour

, Camprubi-Robles

, Miquel-Kergoat

, Andres-Lacueva

, Banati

, Barberger-Gateau

, Bowman

, Caberlotto

, Clarke

, Hogervorst

, Kiliaan

, Lucca

, Manach

, Minihane

, Mitchell

, Perneczky

, Perry

, Roussel

, Schuermans

, Sijben

, Spencer

, Thuret

, van de Rest

, Vandewoude

, Wesnes

, Williams

, Ramirez

(2017) Nutrition for the ageing brain: Towards evidence for an otimal diet. Ageing Res Rev 35,222–240

Jennings

, Cunnane

, Minihane

(2020) Can nutrition support healthy cognitive ageing and reduce dementia risk? BMJ 369,m2269.

Valls-Pedret

, Sala-Vila

, Serra-Mir

, Corella

, de la Torre

, Martinez-Gonzalez

, Martinez-Lapiscina

, Fito

, Perez-Heras

, Salas-Salvado

, Estruch

, Ros

(2015) Mediterranean diet and age-related cognitive decline: A randomized clinical trial. JAMA Intern Med 175,1094–1103.

Livingston

, Sommerlad

, Orgeta

, Costafreda

, Huntley

, Ames

, Ballard

, Banerjee

, Burns

, Cohen-Mansfield

, Cooper

, Fox

, Gitlin

, Howard

, Kales

, Larson

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2017) Dementia prevention, intervention, and care. Lancet 390,2673–2734.

Livingston

, Huntley

, Sommerlad

, Ames

, Ballard

, Banerjee

, Brayne

, Burns

, Cohen-Mansfield

, Cooper

, Costafreda

, Dias

, Fox

, Gitlin

, Howard

, Kales

, Kivimäki

, Larson

, Ogunniyi

, Orgeta

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selb

, Teri

, Mukadam

(2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396,413–446.

Annweiler

, Souberbielle

, Schott

, de Decker

, Berrut

, Beauchet

(2011) [Vitamin D in the elderly: 5 points to remember]. Geriatr Psychol Neuropsychiatr Vieil 9,259–267.

10.

Holick

(2007) Vitamin D deficiency. N Engl J Med 357,266–281.

11.

Mithal

, Wahl

, Bonjour

, Burckhardt

, Dawson-Hughes

, Eisman

, El-Hajj Fuleihan

, Josse

, Lips

, Morales-Torres

(2009) Global vitamin D status and determinants of hypovitaminosis D. Osteoporos Int 20,1807–1820.

12.

Mayeux

, Stern

(2012) Epidemiology of Alzheimer disease. Cold Spring Harb Perspect Med 2,a006239.

13.

Annweiler

, Llewellyn

, Beauchet

(2013) Low serum vitamin D concentrations in Alzheimer’s disease: A systematic review and meta-analysis. J Alzheimers Dis 33,659–674.

14.

Annweiler

, Milea

, Whitson

, Cheng

, Wong

, Ikram

, Lamoureux

, Sabanayagam

(2016) Vitamin D insufficiency and cognitive impairment in Asians: A multi-ethnic population-based study and meta-analysis. J Intern Med 280,300–311.

15.

MacLaughlin

, Holick

(1985) Aging decreases the capacity of human skin to produce vitamin D3. J Clin Invest 76,1536–1538.

16.

Maddock

, Zhou

, Cavadino

, Kuzma

, Bao

, Smart

, Saum

, Schottker

, Engmann

, Kjaergaard

, Karhunen

, Zhan

, Lehtimaki

, Rovio

, Byberg

, Lahti

, Marques-Vidal

, Sen

, Perna

, Schirmer

, Singh-Manoux

, Auvinen

, Hutri-Kahonen

, Kahonen

, Kilander

, Raikkonen

, Melhus

, Ingelsson

, Guessous

, Petrovic

, Schmidt

, Vollenweider

, Lind

, Eriksson

, Michaelsson

, Raitakari

, Hagg

, Pedersen

, Herzig

, Jarvelin

, Veijola

, Kivimaki

, Jorde

, Brenner

, Kumari

, Power

, Llewellyn

, Hypponen

(2017) Vitamin D and cognitive function: A Mendelian randomisation study. Sci Rep 7,13230.

17.

Larsson

, Traylor

, Markus

, Michaelsson

(2018) Serum parathyroid hormone, 25-hydroxyvitamin D, and risk of Alzheimer’s disease: A Mendelian randomization study. Nutrients 10,1243.

18.

Wang

, Qiao

, Zhang

, Hua

, Jin

, Liu

(2020) Circulating vitamin D levels and Alzheimer’s disease: A Mendelian randomization study in the IGAP and UK Biobank. J Alzheimers Dis 73,609–618.

19.

Zhang

, Wang

, Han

, Wang

, Zhang

, Wang

, Liu

(2020) Impact of vitamin D binding protein levels on Alzheimer’s disease: A Mendelian randomization study. J Alzheimers Dis 74,991–998.

20.

Mokry

, Ross

, Morris

, Manousaki

, Forgetta

, Richards

(2016) Genetically decreased vitamin D and risk of Alzheimer disease. Neurology 87,2567–2574.

21.

Buell

, Dawson-Hughes

, Scott

, Weiner

, Dallal

, Qui

, Bergethon

, Rosenberg

, Folstein

, Patz

, Bhadelia

, Tucker

(2010) 25-Hydroxyvitamin D, dementia, and cerebrovascular pathology in elders receiving home services. Neurology 74,18–26.

22.

Chei

, Raman

, Yin

, Shi

, Zeng

, Matchar

(2014) Vitamin D levels and cognition in elderly adults in China. J Am Geriatr Soc 62,2125–2129.

23.

Landel

, Annweiler

, Millet

, Morello

, Féron

(2016) Vitamin D, cognition and Alzheimer’s disease: The therapeutic benefit is in the D-tails. J Alzheimers Dis 53,419–444.

24.

Lopes da Silva

, Vellas

, Elemans

, Luchsinger

, Kamphuis

, Yaffe

, Sijben

, Groenendijk

, Stijnen

(2014) Plasma nutrient status of patients with Alzheimer’s disease: Systematic review and meta-analysis. Alzheimers Dement 10,485–502.

25.

Miller

, Harvey

, Beckett

, Green

, Farias

, Reed

, Olichney

, Mungas

, DeCarli

(2015) Vitamin D status and rates of cognitive decline in a multiethnic cohort of older adults. JAMA Neurol 72,1295–1303.

26.

Nagel

, Herbolsheimer

, Riepe

, Nikolaus

, Denkinger

, Peter

, Weinmayr

, Rothenbacher

, Koenig

, Ludolph

, von Arnim

(2015) Serum vitamin D concentrations and cognitive function in a population-based study among older adults in South Germany. J Alzheimers Dis 45,1119–1126.

27.

Olde Rikkert

, Verhey

, Sijben

, Bouwman

, Dautzenberg

, Lansink

, Sipers

, van Asselt

, van Hees

, Stevens

, Vellas

, Scheltens

(2014) Differences in nutritional status between very mild Alzheimer’s disease patients and healthy controls. J Alzheimers Dis 41,261–271.

28.

Ulstein

, Bøhmer

(2017) Normal vitamin levels and nutritional indices in Alzheimer’s disease patients with mild cognitive impairment or dementia with normal body mass indexes. J Alzheimers Dis 55,717–725.

29.

Zhao

, Sun

, Ji

, Shen

(2013) Vitamin D levels in Alzheimer’s and Parkinson’s diseases: A meta-analysis. Nutrition 29,828–832.

30.

Balion

, Griffith

, Strifler

, Henderson

, Patterson

, Heckman

, Llewellyn

, Raina

(2012) Vitamin D, cognition, and dementia: A systematic review and meta-analysis. Neurology 79,1397–1405.

31.

Chai

, Gao

, Wu

, Dong

, Gu

, Lin

, Zhang

(2019) Vitamin D deficiency as a risk factor for dementia and Alzheimer’s disease: An updated meta-analysis. BMC Neurol 19,284.

32.

Jayedi

, Rashidy-Pour

, Shab-Bidar

(2019) Vitamin D status and risk of dementia and Alzheimer’s disease: A meta-analysis of dose-response (†). Nutr Neurosci 22,750–759.

33.

Shen

, Ji

(2015) Vitamin D deficiency is associated with increased risk of Alzheimer’s disease and dementia: Evidence from meta-analysis. Nutr J 14,76.

34.

Sommer

, Griebler

, Kien

, Auer

, Klerings

, Hammer

, Holzer

, Gartlehner

(2017) Vitamin D deficiency as a risk factor for dementia: A systematic review and meta-analysis. BMC Geriatr 17,16.

35.

Jorde

, Mathiesen

, Rogne

, Wilsgaard

, Kjærgaard

, Grimnes

, Schirmer

(2015) Vitamin D and cognitive function: The Tromsø Study. J Neurol Sci 355,155–161.

36.

Wilson

, Houston

, Kilpatrick

, Lovato

, Yaffe

, Cauley

, Harris

, Simonsick

, Ayonayon

, Kritchevsky

, Sink

(2014) Relationship between 25-hydroxyvitamin D and cognitive function in older adults: The Health, Aging and Body Composition Study. J Am Geriatr Soc 62,636–641.

37.

Tot Babberich Ede

, Gourdeau

, Pointel

, Lemarchant

, Beauchet

, Annweiler

(2015) Biology of subjective cognitive complaint amongst geriatric patients: Vitamin D involvement. Curr Alzheimer Res 12,173–178.

38.

Goodwill

, Szoeke

(2017) A systematic review and meta-analysis of the effect of low vitamin D on cognition. J Am Geriatr Soc 65,2161–2168.

39.

Rutjes

, Denton

, Di Nisio

, Chong

, Abraham

, Al-Assaf

, Anderson

, Malik

, Vernooij

, Martínez

, Tabet

, McCleery

(2018) Vitamin and mineral supplementation for maintaining cognitive function in cognitively healthy people in mid and late life. Cochrane Database Syst Rev 12,CD011906.

40.

Corless

, Dawson

, Fraser

, Ellis

, Evans

, Perry

, Reisner

, Silver

, Beer

, Boucher

, Cohen

(1985) Do vitamin D supplements improve the physical capabilities of elderly hospital patients? Age Ageing 14,76–84.

41.

Dhesi

, Jackson

, Bearne

, Moniz

, Hurley

, Swift

, Allain

(2004) Vitamin D supplementation improves neuromuscular function in older people who fall. Age Ageing 33,589–595.

42.

Rossom

, Espeland

, Manson

, Dysken

, Johnson

, Lane

, LeBlanc

, Lederle

, Masaki

, Margolis

(2012) Calcium and vitamin D supplementation and cognitive impairment in the women’s health initiative. J Am Geriatr Soc 60,2197–2205.

43.

Stein

, Scherer

, Ladd

, Harrison

(2011) A randomized controlled trial of high-dose vitamin D2 followed by intranasal insulin in Alzheimer’s disease. J Alzheimers Dis 26,477–484.

44.

Pettersen

(2017) Does high dose vitamin D supplementation enhance cognition?: A randomized trial in healthy adults. Exp Gerontol 90,90–97.

45.

Aspell

, Healy

, Mc Partlin

, Lawlor

, O’Sullivan

(2017) Effects of vitamin D supplementation on cognitive function in healthy, community dwelling older adults: Results from a randomised double-blind placebo-controlled pilot trial. Proc Nutr Soc 76,E59.

46.

Owusu

, Islam

, Katumuluwa

, Stolberg

, Usera

, Anwarullah

, Shieh

, Dhaliwal

, Ragolia

, Mikhail

, Aloia

(2019) Cognition and vitamin D in older African-American women- Physical performance and osteoporosis prevention with vitamin D in older African Americans Trial and Dementia. J Am Geriatr Soc 67,81–86.

47.

Pilz

, Trummer

, Theiler-Schwetz

, Grubler

, Verheyen

, Odler

, Karras

, Zittermann

, Marz

(2022) Critical appraisal of large vitamin D randomized controlled trials. Nutrients 14,303.

48.

Afzal

, Bojesen

, Nordestgaard

(2014) Reduced 25-hydroxyvitamin D and risk of Alzheimer’s disease and vascular dementia. Alzheimers Dement 10,296–302.

49.

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.

50.

O’Bryant

, Humphreys

, Smith

, Ivnik

, Graff-Radford

, Petersen

, Lucas

(2008) Detecting dementia with the mini-mental state examination in highly educated individuals. Arch Neurol 65,963–967.

51.

(1989) Comprehensive geriatric assessment. AGS Public Policy Committee. J Am Geriatr Soc 37,473–474.

52.

Marjal. , Molag GAMG-W, Arend J. Arends en Teus A. van Barneveld (2011) De richtlijn ‘Comprehensive geriatric assessment’ (in Dutch). Ned Tijdschr Geneeskd 155,A3815.

53.

(2014) Richtlijn Diagnostiek en Behandeling van dementie. (in Dutch). Nederlandse Vereniging voor Klinische Geriatrie.

54.

Petersen

, Doody

, Kurz

, Mohs

, Morris

, Rabins

, Ritchie

, Rossor

, Thal

, Winblad

(2001) Current concepts in mild cognitive impairment. Arch Neurol 58,1985–1992.

55.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr., , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7,263–269.

56.

Brouwer-Brolsma

, Vaes

AMM

, van der Zwaluw

, van Wijngaarden

, Swart

KMA

, Ham

, van Dijk

, Enneman

, Sohl

, van Schoor

, van der Velde

, Uitterlinden

, Lips

, Feskens

EJM

, Dhonukshe-Rutten

RAM

, de Groot

(2016) Relative importance of summer sun exposure, vitamin D intake, and genes to vitamin D status in Dutch older adults: The B-PROOF study. J Steroid Biochem Mol Biol 164,168–176.

57.

Brouwer-Brolsma

, van de Rest

, Tieland

, van der Zwaluw

, Steegenga

, Adam

, van Loon

, Feskens

, de Groot

(2013) Serum 25-hydroxyvitamin D is associated with cognitive executive function in Dutch prefrail and frail elderly: A cross-sectional study exploring the associations of 25-hydroxyvitamin D with glucose metabolism, cognitive performance and depression. J Am Med Dir Assoc 14,852.e859–817.

58.

Garretsen

HFL

, Knibble

(1983) Alkohol prevalentie onderzoek, Rotterdam/Limburg (in Dutch). Landelijk Eindrapport. Ministerie vanWelzijn, Volksgezondheid en Cultuur, Leidschendam.

59.

Masnoon

, Shakib

, Kalisch-Ellett

, Caughey

(2017) What is polypharmacy? A systematic review of definitions. BMC Geriatr 17,230.

60.

Levey

, Bosch

, Lewis

, Greene

, Rogers

, Roth

(1999) A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 130,461–470.

61.

Kalra

, Teixeira

, Diniz

(2020) Association of vitamin D levels with incident all-cause dementia in longitudinal observational studies: A systematic review and meta-analysis. J Prev Alzheimers Dis 7,14–20.

62.

Soysal

, Dokuzlar

, Erken

, Dost Gunay

, Isik

(2020) The relationship between dementia subtypes and nutritional parameters in older adults. J Am Med Dir Assoc 21,1430–1435.

63.

Chung

, Park

, Kim

, Shin

, Park

, Chung

, Ha

, Ahn

, Shin

, Kim

, Moon

(2015) 25-hydroxyvitamin D status is associated with chronic cerebral small vessel disease. Stroke 46,248–251.

64.

Almeras

, Eyles

, Benech

, Laffite

, Villard

, Patatian

, Boucraut

, Mackay-Sim

, McGrath

, Féron

(2007) Developmental vitamin D deficiency alters brain protein expression in the adult rat: Implications for neuropsychiatric disorders. Proteomics 7,769–780.

65.

Eyles

, Almeras

, Benech

, Patatian

, Mackay-Sim

, McGrath

, Féron

(2007) Developmental vitamin D deficiency alters the expression of genes encoding mitochondrial, cytoskeletal and synaptic proteins in the adult rat brain. J Steroid Biochem Mol Biol 103,538–545.

66.

Eyles

, Burne

, McGrath

(2013) Vitamin D, effects on brain development, adult brain function and the links between low levels of vitamin D and neuropsychiatric disease. Front Neuroendocrinol 34,47–64.

67.

Groves

, McGrath

, Burne

(2014) Vitamin D as a neurosteroid affecting the developing and adult brain. Annu Rev Nutr 34,117–141.

68.

Etgen

, Sander

, Bickel

, Sander

, Forstl

(2012) Vitamin D deficiency, cognitive impairment and dementia: A systematic review and meta-analysis. Dement Geriatr Cogn Disord 33,297–305.

69.

Aghajafari

, Pond

, Catzikiris

, Cameron

(2018) Quality assessment of systematic reviews of vitamin D, cognition and dementia. BJPsych Open 4,238–249.

70.

Sempos

, Williams

, Carter

, Jones

, Camara

, Burdette

, Hahm

, Nalin

, Duewer

, Kuszak

, Merkel

, Hoofnagle

, Lukas

, Cavalier

, Durazo-Arvizu

, Crump

, Popp

, Beckert

, Schultess

, Van Slooten

, Tourneur

, Pease

, Kaul

, Villarreal

, Ivison

, Fischer

, van den Ouweland

JMW

, Ho

, Law

EWK

, Simard

, Gonthier

, Holmquist

, Batista

, Meadows

, Cox

, Jansen

, Khan

, Robyak

, Creer

, Kilbane

, Twomey

, Freeman

, Parker

, Yuan

, Fitzgerald

, Mushtaq

, Clarke

, Breen

, Simpson

, Wise

(2022) Assessment of serum total 25-hydroxyvitamin D assays for Vitamin D External Quality Assessment Scheme (DEQAS) materials distributed at ambient and frozen conditions. Anal Bioanal Chem 414,1015–1028.

71.

Eyles

, Smith

, Kinobe

, Hewison

, McGrath

(2005) Distribution of the vitamin D receptor and 1 alpha-hydroxylase in human brain. J Chem Neuroanat 29,21–30.

72.

Stumpf

, O’Brien

(1987) 1,25 (OH)2 vitamin D3 sites of action in the brain. An autoradiographic study. Histochemistry 87,393–406.

73.

Wion

, MacGrogan

, Neveu

, Jehan

, Houlgatte

, Brachet

(1991) 1,25-Dihydroxyvitamin D3 is a potent inducer of nerve growth factor synthesis. J Neurosci Res 28,110–114.

74.

Durk

, Han

, Chow

, Ahrens

, Henderson

, Fraser

, Pang

(2014) 1alpha,25-Dihydroxyvitamin D3 reduces cerebral amyloid-beta accumulation and improves cognition in mouse models of Alzheimer’s disease. J Neurosci 34,7091–7101.

75.

Supriya

, Chandra

, Prabhakar

, Prasad

, Christopher

(2018) Vitamin D receptor (VDR) gene polymorphism and vascular dementia due to cerebral small vessel disease in an Asian Indian cohort. J Neurol Sci 391,84–89.

76.

Ten Haaf

DSM

, Balvers

MGJ

, Timmers

, Eijsvogels

TMH

, Hopman

MTE

, Klein Gunnewiek

JMT

(2019) Determinants of vitamin D status in physically active elderly in the Netherlands. Eur J Nutr 58,3121–3128.