Skip to main content
NCBI home page
The Journal of Nutrition, Health & Aging logoLink to The Journal of Nutrition, Health & Aging
. 2018 Mar 6;22(6):718–725. doi: 10.1007/s12603-018-1010-z

Associations between Plasmatic Polyunsaturated Fatty Acids Concentrations and Cognitive Status and Decline in Neurocognitive Disorders

Marine Haution-Bitker 1,5, T Gilbert 1, A Vignoles 1,2, C Lecardonnel 1, S Watelet 1, E Blond 2,3, J Drai 2, M Bonnefoy 1,3,4
PMCID: PMC12876302  PMID: 29806861

Abstract

Objective

To examine the association of plasmatic and erythrocyte concentrations polyunsaturated fatty acids (PUFA) with both cognitive status and decline.

Design

Longitudinal observational cohort study. Setting: Memory Clinic of Lyon Sud university hospital.

Participants

140 patients, aged 60 and older, were referred to the memory clinic, and successively included in the cohort, between March 2010 and February 2014.

Measurements

Concentration of ω-3 PUFA (eicosapentaenoic: EPA and docosahexaenoic: DHA) and ω-6 PUFA (arachidonic: AA), were measured at baseline in plasma and in the erythrocytes membrane. Cognitive status was assessed using the mini mental state examination (MMSE), at baseline and every six months during follow-up. The median follow-up period was of 11,5 months.

Results

Compared to participants with minor neurocognitive disorders (MMSE≥24), participants with major neurocognitive disorders (NCD) had lower plasmatic concentrations of EPA and DHA (p<0.05) at baseline. Erythrocyte AA and DHA concentrations were significantly lower in patients with cognitive decline (defined as a ≥2 points loss of MMSE per year), while no difference in plasmatic concentrations was observed.

Conclusions

Our study suggests that ω-3 PUFA plasma concentrations (mainly EPA and DHA) could be associated with cognitive status in older people. Moreover, in this exploratory study, lower erythrocyte PUFA concentrations (AA and DHA) were associated with accelerated decline and could be proposed as a surrogate marker for prediction of cognitive decline.

Key words: Dementia, neurocognitive disorders, biomarkers, fatty acids, erythrocytes membrane

Abbreviations

AA

arachidonic acid

AAL

alpha-linolenic acid

AD

Alzheimer disease

ADL

Activities of daily living

AL

linolenic acid

CPG

choline phosphoglycerides

CSF

Cerebrospinal fluid

DHA

docosahexaenoic acid

EC

cholesteryl esters

EPA

eicosapentaenoic acid

EPG

ethanolamine phosphoglycerides

FA

fatty acids

IADL

Instrumental activities of daily living

MCI

mild cognitive impairment

MMSE

mini mental state examination

MRI

Magnetic resonance imaging

NCD

Neurocognitive disorders

Palm

palmitic acid

PL

phospholipid

PUFA

poly unsaturated fatty acids.

Introduction

There is growing evidence that polyunsaturated fatty acids (PUFA) of the omega-3 family are associated with cognitive function and neuro-development in humans. Omega-3 PUFA (ω-3 PUFA), including eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and alpha-linolenic acid (AAL), are all present in the Mediterranean diet (1). They are known to reduce cardiovascular morbidity and mortality and also have anti-inflammatory effects and neuronal protective functions (2). Many other roles are currently assumed to be present in physiological aging, as suggested by Vellas in The Multidomain Alzheimer's Preventive Trial (3) and in particular a possible association between Red blood cell membrane omega-3 fatty acid levels and physical performance (4).

Fatty acids (FA) are structural components of the brain and are essential for the synthesis of neuronal membrane phospholipids. They are strongly involved in the protective mechanisms against neuro-inflammation and neurodegeneration during normal aging, and may have protective effects on neurocognitive impairment (2, 5). Their neuro-protective effect may be related to their presence in the cellular membrane, by reinforcing its structure (2, 5, 6). They promote cellular survival by their antioxidant role, anti-inflammatory action, and antiapoptotic effect. Through those metabolic pathways, they may slow down the progression of cognitive decline in mild cognitive impairment (MCI) and mild Alzheimer's disease (AD), but not in patients with more advanced disease (7). These detrimental effects seems more important in patients presenting with mutations of the ApoE-6 allele (8, 9), and in those with genetic variants of the FADS gene cluster (10). In opposition, ω-6 PUFA daily intake could be associated with an increased risk of NCD, if not associated with increased ω-3 PUFA intake, expressed by a negative ratio of ω-3 PUFA over ω-6 PUFA. As the roles of ω-3 and ω-6 PUFA on inflammation and neuronal protection are intimately related and interdependent, it requires that both categories should be assessed simultaneously.

Nutritional elements, such as fish consumption, dietary habits, or frequency and quantity of intake, may play an important part in the intra-individual variability of cognitive decline. For example, higher adherence to the Mediterranean dietary is associated with a significant reduction in the risk of AD development (2). Plasma levels of PUFA and the balance between polyunsaturated and saturated FA, are reliable markers of dietary intake, and, as such, should be explored as potential surrogates of cognitive status and risk in older people (11).

To the best of our knowledge, the available epidemiological evidence is insufficient to recommend the use of any primary prevention therapy for AD. In previous studies, EPA and DHA intake (1, 2, 5), as well as higher plasmatic levels, was significantly associated with a lower risk of dementia. Eriksdotter et al. showed a significant increase in ω-3 PUFA plasmatic levels compared to baseline, after a six months oral supplementation, but failed to demonstrate its protective effect on the decline of the Mini Mental State Examination (MMSE) (12).

Hence, it is unclear whether low plasma levels of ω-3 PUFA are associated with cognitive impairment (13). The aim of our study was to evaluate the relationship of blood PUFA levels (both plasmatic and erythrocyte concentrations) with cognitive performance, assessed by the MMSE, using an observational longitudinal cohort.

Methods

Participants and setting

The GERIOX monocentric observational cohort prospectively included all patients, aged 60 years and over, referred to the memory clinic of Lyon Sud academic hospital (Pierre-Bénite, France), with the aim to evaluate the influence of oxidative stress and nutritional biomarkers on cognitive decline (clinicaltrials.gov reference number NCT02800395) (14). After a baseline consultation, all patients for whom a neurodegenerative disease was suspected were admitted to the outpatient department as part of routine practice (2). These baseline clinical, radiological (brain MRI) and biological assessments included neuropsychological testing, as well as a set of blood assays (PUFA, Leptin, Adiponectine, Acylated Ghrelin and unacylated Ghrelin), in accordance with practice guidelines of the French Health Authorities (16).

Demographic and cognitive status assessment

Main demographic variables were collected at baseline, including age, weight and the Mini Nutritional Assessment screening tool. The cognitive status of patients was assessed by the MMSE, which was performed at baseline and every 6 months at each visit during follow-up (16). A diagnosis of NCD was confirmed by a screening procedure performed by an independent panel of experts in cognitive care, who did not take part in the present study. NCD diagnosis, based on available medical notes, and following the DSM-V criteria (17, 18) including neuropsychological testing (categorical verbal fluency test, MMSE, RL/RI 16, the clock test and Dubois's five-word-test). The diagnoses of Alzheimer's disease, vascular dementia, Lewy body dementia, were similarly performed. Cerebrospinal fluid (CSF) biomarkers were not systematically screened, as they are not standard-of-care in our centre, and are exceptionally useful except in atypical cases, early NCD or fastchanging cognitive patterns.

For analytical purposes, we dichotomized the population into cognitive subsets, based on 1) MMSE at baseline and 2) rate of decline of the MMSE during follow-up. Patients with a MMSE ≥ 24 but with a disability attributable to cognitive complaint (as assessed by the Instrumental Activities of Daily Living score – IADL) were categorized as having minor NCD, according to DSM-V (18, 19, 20). Patients with a MMSE < 24 associated with abnormal results in the Activities of Daily Living (ADL) and IADL scores were diagnosed with major NCD

The rate of decline was computed by performing, for each individual, a univariate linear regression using MMSE as the dependent variable, and “time of assessment” as the independent variable. Cognitive decline was defined by a MMSE point loss of ≥ 2 points per year (as opposed to those without cognitive decline). Rapid cognitive decline was defined as having a MMSE point loss of ≥ 4 per year (21).

Polyunsaturated fatty acids assays

Fatty acids assays were measured at baseline (first visit), both in the plasma and in the erythrocyte membranes. A peripheral venous puncture was carried out at in the morning between 08:00 and 09:00 AM, after a 12-hour overnight fast. Samples for nutritional biomarkers were taken on an EDTA blood collection tube, and analysed in the biochemistry laboratory of Lyon Sud university hospital. To quantify fatty acids, lipids were extracted from plasma or erythrocyte pellet using a chloroform/methanol mix (1:1, v/v), following the Folch method (22). The lipid classes were separated by thin-layer chromatography on silicagel plates (Merck 5721, Germany), using petroleum of ether-ethylic and ether-acetic acids as developing solvent (23). The plates were sprayed with bromophenol blue and individual bands of phospholipids and cholesteryl-esters were scraped off in separated tubes. The phospholipid (PL) fraction was saponified and transmethylated with sodium methylate, and the cholesteryl ester (EC) fraction was methylated with sulphuric acid and dehydrated methanol. The methyl esters of each fraction were removed by hexane, and analysed by gas-liquid chromatography, using a FOCUS, Trace/ULTRA GC gas chromatograph (Thermo Separation Products, Les Ulis, France) equipped with a CP-SIL fused silica capillary column (25 m x 0.25 mm internal diameter) coated with 100% cyanopropyl siloxane (22, 24).

Identification of individual methyl ester components was performed by frequent comparison of the retention times with those of standard values. The results were given based on the relative abundance of individual FA (in mg/l) in each of both lipid fractions (plasma and erythrocyte membrane). For the plasma fraction, we measured two components (or portions): phospholipids (PL) and cholesterol esters (EC); in the erythrocyte fraction, we measured two components: choline phosphoglycerides (CPG) and ethanolamine phosphoglycerides (EPG).

Six FA were selected for analysis, of which three were ω-3 PUFA (EPA, DHA and AAL); two ω-6 PUFA (linoleic acid [AL], arachidonic acid [AA]), and one saturated FA (palmitic acid [palm]) (25). We also collected leptin and albumin serum concentrations.

Statistical analysis

Data was analysed using the R software (27). A p value < 0.05 was considered statistically significant. Continuous variables were presented as mean ± standard deviations (SD). Gender and treatment introduction were presented as qualitative variables, with frequency and percentage.

At baseline, we first compared variables (demographic and FA concentrations) between patients with minor NCD or MCI (MMSE ≥ 24) and those with major NCD (MMSE < 24), using Welch t test for quantitative variables with unequal variances, and Chi2 test for qualitative variables.

Then, a univariate linear regression model was used to investigate the associations between FA concentrations and MMSE, both measured at baseline. A multivariate analysis was finally including relevant demographic variables and those plasmatic FA concentrations with a p value < 0.2 in univariate analysis. The multivariate analysis only incorporated FA plasmatic concentrations, as the small sample size of erythrocyte membrane assays precluded a similar methodology.

The association of cognitive decline during follow-up with demographic variables and FA levels at baseline was explored by comparing patients with and without cognitive decline (as defined above), using Welch's t test for quantitative variables, with unequal variances, and the Chi2 test for qualitative variables. Then, we performed a univariate logistic regression of baseline variables associated with cognitive decline status.

The association of rapid cognitive decline with demographic variables and baseline FA levels was explored by comparing patients with and without rapid cognitive decline (as defined above), using Welch's t test for quantitative variables with unequal variances, and the Chi2 test for qualitative variables. Finally, we assessed the association of baseline variables with rapid cognitive decline, by performing a univariate logistic regression.

Ethics

The cohort was approved by the ethics committee of the “Hospices civiles de Lyon” and declared to the French commission for data protection “Comité de Protection des Personnes (CPP)” (Favourable opinion on 18/04/2014, EudraCT Number: 2014-A00939-38). This was an observational study performed in the context of current care.

Results

Patients and cohort characteristics

One hundred and forty patients were consecutively included in the GERIOX cohort between March 2010 and February 2014, of which 91 were women (65,0%). The median followup period was 11,5 months. Their demographic baseline characteristics are reported in Table 1. Forty-four patients presented typical Alzheimer's disease (31.4%), 29 were diagnosed with vascular dementia (20.7%), 4 with Lewy body dementia (2.9%), and 63 had unclassified disorders (45.0%). Unclassified disorders correspond to uncertain, mixed or pending diagnoses.

Table 1.

Population characteristics at baseline

Mean ± SD Min Max
Age (years) 80.16 ± 6.48 61 92
Weight (kg) 61.95 ± 13.12 37 100
Albumine (g/l) 39.89 ± 4.2 21,5 47,3
MNA 23.28 ± 4.39 2 29
MMSE at baseline 23.71 ± 3.96 10 30
Open in a new tab

MMSE: mini mental state examination; MNA: mini nutritional assessment; SD: standard deviation

Regarding FA measurements at baseline, 140 plasmatic concentrations were available at baseline (100%), and 70 for erythrocyte membrane quantifications (50%). Baseline FA concentrations are available in Supplementary Table S1.

Association between FA concentrations and MMSE at baseline

The mean MMSE was 23.7 ± 4.0 at baseline, and 58 patients had a MMSE < 24 (major NCD) at baseline. Patients with major NCD at baseline had lower plasmatic concentrations of ω-3 PUFA (namely EC and PL portions of EPA and DHA) compared with patients with mild cognitive impairment or normal cognition. No difference was observed between groups regarding ω-6 PUFA, saturated FA, and other nutritional or demographic variables (Table 2). Erythrocyte membrane FA concentrations did not differ between groups (Supplementary Table 2. S2).

Table 2.

Comparison of variables between MMSE groups at baseline

MMSE < 24 (n=58) MMSE ≥ 24 (n=82) P
Demographic variables
Age (years) 81.05 ± 5.66 79.52 ± 6.97 0.16
Female gender 42 (72.4%) 49 (59.7%) 0.17
Weight (kg) 62.98 ± 15.49 61.39 ± 11.78 0.68
Plasma ω-3 PUFA (in mg/l)
α linolenic acid.PL 2.09 ± 1.26 2.03 ± 0.98 0.75
α linolenic acid.EC 4.3 ± 1.97 4.22 ± 1.89 0.81
EPA.PL 12.1 ± 9 15.66 ± 10.01 0.03
EPA.EC 6.91 ± 4.13 10.38 ± 9.57 <0,01
DHA.PL 45.34 ± 14.95 54.13 ± 21.45 <0,01
DHA.EC 4.93 ± 2.76 6.02 ± 3.52 0.04
Plasma ω-6 PUFA (in mg/l)
linoleic acid.PL 195.71 ± 43.98 194.46 ± 52.03 0.88
linoleic acid.EC 358.35 ± 112.21 363.8 ± 100.23 0.77
arachidonic acid.PL 129.69 ± 27.17 125.74 ± 34.84 0.45
arachidonic acid.EC 58.3 ± 17.11 60.09 ± 21.11 0.58
Plasma saturated FA (in mg/l)
palmitic acid.PL 314.71 ± 52.56 315.4 ± 56.74 0.94
palmitic acid.EC 103.26 ± 26.64 105.43 ± 23.46 0.62
Other nutritional variables
Leptin (mg/l) 19.08 ± 23.03 16.74 ± 22.51 0.55
Albumin (g/l) 40.58 ± 3.65 39.6 ± 4.43 0.4
MNA 22.77 ± 3.79 23.52 ± 4.67 0.56
Open in a new tab

Data are reported as mean ± SD or count (percentage); DHA: docosahexaenoic acid; EC: cholesterol ester portion; EPA: eicosapentaenoic acid; FA: fatty acids; MMSE: mini mental state examination; MNA: mini nutritional assessment; ω-3 PUFA: omega-3 polyunsaturated fatty acid; ω-6 PUFA: omega-6 polyunsaturated fatty acid; PL: phospholipid portion; SD: standard deviation.

In the univariate analysis, MMSE at baseline was significantly associated with the EC and PL portions of plasmatic EPA, as well as the PL portion of plasmatic DHA, but not with its EC portion (Table 3). The multivariate analysis, using age, gender, EPA (both EC and PL portions), and DHA (EC portion) as independent variables, showed no significant association with MMSE at baseline (Table 3).

Table 3.

Association of variables measured at baseline with initial MMSE

Univariate analysis Multivariate analysis
Estimate ± SE p Estimate ± SE P
Demographic variables
Age (per 1 year) -0.02 ± 0.05 0.70 -0.02 ± 0.05 0.75
Female gender 0.01 ± 0.7 0.99 0.12 ± 0.72 0.87
Weight (per 1 kg) -0.02 ± 0.03 0.55 - -
MNA (per 1 point) 0.04 ± 0.1 0.73 - -
Plasma ω-3 PUFA (per 1 mg/l)
Alpha Linolenic.PL 0.08 ± 0.31 0.8 - -
Alpha Linolenic.EC 0.07 ± 0.18 0.69 - -
EPA.PL 0.08 ± 0.03 0.01 0.01 ± 0.06 0.83
EPA.EC 0.12 ± 0.04 <0.01 0.1 ± 0.07 0.13
DHA.PL 0.04 ± 0.02 0.04 0.01 ± 0.02 0.83
DHA.EC 0.12 ± 0.1 0.25 - -
Plasma ω-6 PUFA (per 1 mg/l)
Linoleic.PL 0 ± 0.01 0.83 - -
Linoleic.EC 0 ± 0 0.64 - -
Arachidonic.PL -0.01 ± 0.01 0.46 - -
Arachidonic.EC 0.01 ± 0.02 0.62 - -
Plasma Saturated FA (per 1 mg/l)
Palmitic.PL 0 ± 0.01 0.46 - -
Palmitic.EC 0.01 ± 0.01 0.43 - -
Other nutritonal variables
Leptine (per 1 mg/l) 0 ± 0.01 0.74 - -
Albumine (per 1 g/l) -0.06 ± 0.09 0.54 - -
Open in a new tab

DHA: docosahexaenoic; EC: cholesterol ester portion; EPA: eicosapentaenoic; FA: fatty acids; MMSE: mini mental state examination; MNA: mini nutritional assessment; ω-3 PUFA: omega-3 polyunsaturated fatty acid; ω-6 PUFA: omega-6 polyunsaturated fatty acid; PL: phospholipid portion; SE: standard error.

Association of FA concentrations with cognitive decline

The median follow-up period was 11,5 months. The average MMSE annual decline rate was -1.5 ± 4.9 per year, and 62 patients presented with cognitive decline (44.3%).

When comparing patients with and without cognitive decline, we observed a lower erythrocyte membrane concentration of DHA (both CPG and EPG portions), and of AA (both CPG and EPG portions) in patients with cognitive decline (Table 4). No difference in plasmatic FA concentrations was observed between groups (Supplementary Table 3. S3) or in univariate analysis.

Table 4.

Comparison of variables according to cognitive decline status during follow-up

Cognitive decline (n =27) No cognitive decline (n = 43) p
Demographic variables
Age (years) 80.31 ± 6.85 80.04 ± 6.22 0.81
Female gender 36 (58.1%) 55 (70.5%) 0.18
Weight (kg) 60.19 ± 11.21 63.40 ± 14.50 0.33
MNA 23.33 ± 3.47 23.24 ± 5.15 0.95
Erythrocyte ω-3 PUFA (in mg/l)
α linolenic acid.CPG 0.76 ± 0.72 0.81 ± 0.55 0.77
α linolenic acid.EPG 0.47 ± 0.45 0.39 ± 0.43 0.51
EPA.CPG 1.68 ± 2.7 1.61 ± 1.77 0.9
EPA.EPG 1.15 ± 1.28 1.78 ± 2.07 0.14
DHA.CPG 2.83 ± 2.35 6.9 ± 9.83 0.02
DHA.EPG 2.64 ± 1.93 6.57 ± 9.83 0.02
Erythrocyte ω-6 PUFA (in mg/l)
linoleic acid.CPG 42.52 ± 20.77 47.86 ± 31.57 0.42
linoleic acid.EPG 5.11 ± 2.53 6.4 ± 4.89 0.18
arachidonic acid.CPG 16.11 ± 9.58 25.27 ± 24.21 0.04
arachidonic acid.EPG 14.78 ± 8.69 24.84 ± 24.56 0.03
Erythrocytes Saturated FA (in mg/l)
palmitic acid.CPG 142.26 ± 57.03 144.86 ± 63.5 0.86
palmitic acid.EPG 22.59 ± 9.3 27.27 ± 17.05 0.16
Other nutritional variables
Leptin (mg/l) 17.77 ± 23.21 17.66 ± 22.39 0.98
Albumin (g/l) 40.56 ± 5.32 39.35 ± 3.03 0.33
Open in a new tab

Data are reported as mean ± SD or count (percentage); CPG: choline phosphoglycerides; DHA: docosahexaenoic acid; EPA: eicosapentaenoic; EPG: ethanolamine phosphoglycerides; FA: fatty acids; MMSE: mini mental state examination; MNA: mini nutritional assessment; ω-3 PUFA: omega-3 polyunsaturated fatty acid; ω-6 PUFA: omega-6 polyunsaturated fatty acid; PL: phospholipid portion; SD: standard deviation.

Association of FA levels with rapid cognitive decline

Rapid cognitive decline (i.e. decline rate of MMSE ≥ 4 points per year) was present in 39 patients (27.9%). There was no significant difference in comparing plasmatic and erythrocyte membrane concentrations of FA in rapid decliners, compared with patients with mild or no cognitive decline (Table 5).

Table 5.

Comparison FA concentrations (mg/l) according rapid decline

Rapid No rapid p Rapid No rapid p
cognitive cognitive cognitive cognitive
decline decline decline decline
(n = 39) (n = 101) (n = 15) (n = 55)
PLASMA ERYTHROCYTES
ω-3 PUFA
α linolenic acid.PL 1.86 ± 1.11 2.13 ± 1.11 0.21 α linolenic acid.CPG 0.81 ± 0.83 0.79 ± 0.56 0.93
α linolenic acid.EC 3.95 ± 1.76 4.37 ± 1.97 0.22 α linolenic acid.EPG 0.52 ± 0.47 0.39 ± 0.43 0.36
EPA.PL 12.41 ± 5.98 14.87 ± 10.79 0.09 EPA.CPG 1.39 ± 1.24 1.71 ± 2.41 0.51
EPA.EC 8.13 ± 4.16 9.28 ± 9.04 0.31 EPA.EPG 1.43 ± 1.56 1.54 ± 1.87 0.84
DHA.PL 50.87 ± 17.25 50.35 ± 20.33 0.88 DHA.CPG 3.22 ± 2.65 5.78 ± 8.79 0.08
DHA.EC 5.56 ± 2.73 5.58 ± 3.46 0.98 DHA.EPG 2.88 ± 2.05 5.54 ± 8.75 0.06
ω-6 PUFA
linoleic.PL 183.28 ± 46.27 199.5 ± 49.08 0.07 linoleic acid.CPG 43 ± 19.5 46.41 ± 29.61 0.61
linoleic.EC 335.21 ± 93.36 371.85 ± 107.82 0.05 linoleic. acid.EPG 5.4 ± 2.5 5.99 ± 4.47 0.52
arachidonic acid.PL 123.92 ± 30.76 128.71 ± 32.3 0.42 arachidonic acid.CPG 17.47 ± 11.04 22.61 ± 21.79 0.23
arachidonic acid.EC 57.72 ± 17.26 59.99 ± 20.38 0.51 arachidonic acid.EPG 15.87 ± 9.19 22.04 ± 22.19 0.13
Saturated FA
palmitic acid.PL 306.67 ± 57.8 318.38 ± 53.62 0.28 palmitic acid.CPG 145.73 ± 56.93 143.16 ± 61.98 0.88
palmitic acid.EC 100.92 ± 20.46 105.95 ± 26.18 0.23 palmitic acid.EPG 22.93 ± 8.64 26.02 ± 15.74 0.33
Open in a new tab

Data are reported as mean ± SD; CPG: choline phosphoglycerides; DHA: docosahexaenoic; EC: cholesterol esters; EPA: eicosapentaenoic; EPG: ethanolamine phosphoglycerides; FA: fatty acids; MNA: mini nutritional assessment; PL: phospholipids; ω-3 PUFA: omega 3 poly unsaturated fatty acid; ω-6: omega 6; SD: standard deviation;

Discussion

Fatty acid profile might be a valuable risk assessment tool, evaluating nutritional status and its potential association with the development of cognitive impairment. In the present study, we compared the plasma and erythrocyte membrane FA profile between levels of NCD and cognitive decline. Our data suggests that patients with major NCD might have lower concentrations of plasma ω-3 PUFA compared to those of participants with minor NCD. The main findings reveal that cognitive function, as measured by the MMSE, was significantly associated with lower plasmatic EPA and DHA (ω-3 PUFA) concentrations. In addition, patients with cognitive decline over a 12 month-period had lower erythrocyte membrane DHA (ω-3 PUFA) and AA (ω-6 PUFA) levels at baseline.

Blood levels of FA are generally considered to be relevant markers of dietary intake, usually reflecting short-term intake (28). However, recent studies suggest that they may also reflect long-term intake and that their levels are consistent with changes in FA composition in the adipose tissue (29). Beside dietary intake, their levels also reflect processes involved in their use, namely absorption and metabolism (30). For study diet impact, subjective food frequency questionnaires are common. Such dietary assessments were not possible in our study, given the cognitive impairment of included patients, with potential bias in their reliability. Finally, circulating levels might be a better measure of exposure than dietary intake assessment, as they are less liable to imprecision.

The role of the PL and EC portions of plasmatic FA may be equivalent on the cognitive function, but a dietary deficiency in EC could be more precociously visible and later corrected by supplementation than a deficit in PL (31). Hence, the analysis of both portions is mandatory to assess the real effects of FA dietary intake.

The assessment of erythrocyte FA concentrations also appears of interest. Indeed, the concentrations of PUFA in the membranes of erythrocytes are higher than those of other lipid fractions (32). The literature also indicates the existence of a better correlation between PUFA levels in the brain tissue and erythrocyte membranes PUFA concentrations, as compared to plasmatic levels, suggesting that brain tissue and CSF FA levels may be indirectly estimated via the measurement of erythrocyte membrane FA concentrations (33). A similar relationship may exist between plasmatic FA concentrations and hippocampic cortex volume (12,34). Using the Folch method, we systematically separated each erythrocyte membrane FA, making it easier, and more reliable, to interpret their specific effect (21, 33). Moreover, erythrocyte concentrations in FA may reflect with higher precision dietary intake over the longer term (estimated range of 60 to 90 days) (35). To date, few studies have explored their specific impact on cognitive impairment, with non-significant results, a matter related to the complexity of the technique (29, 34).

Although there exists evidence of the relationship between PUFA and cognitive performance in adults without dementia, few studies have investigated the relationship between FA concentrations and the evolution of MMSE over time in patients with NCD (36, 37). Of note, associations of FA levels with the ADAS COG score, frontal lobe performances, semantic and working memory, or verbal fluency test were previously demonstrated (12, 31, 38). However, number of studies exploring the effect of FA levels on the commonly used MMSE are scarcer.

Our results are consistent with previously published studies. In a study exploring the relationship between plasmatic FA concentrations and cognitive function, Tully et al. showed, with 119 patients, that serum EPA and DHA levels were significantly lower in all MMSE quartiles of patients with AD compared with control values (39). In a recent study, Yin concluded that several nutrient biomarkers, such as plasma levels of EPA, DHA, and total ω-3PUFA were lower in MCI patients compared to healthy controls, suggesting their potential role in development of MCI (40). Finally, Conquer and al. with 84 patients, found that plasma levels of EPA, DHA, total ω-3 PUFA, and the ω-3 PUFA/ω-6 PUFA ratio were lower in AD, MCI and others dementia aetiologies (41). Our work demonstrates that lower DHA and EPA plasma concentrations are associated with a higher degree of cognitive impairment (39).

Our results also suggests that patients with cognitive decline during follow-up have lower erythrocyte membrane ω-3 and ω-6 PUFA concentrations (namely DHA and AA) at baseline. They complete the findings of the retrospective analysis of the large WHISCA cohort, which showed that women participants in the highest DHA and EPA concentration tertile exhibited better fine motor speed, verbal knowledge, and verbal fluency. However, after full covariate adjustment, these associations were attenuated and no longer statistically significant (34). Interestingly, Yuan et al. found that erythrocyte membranes levels of ω-3 and ω-6 PUFA were predictive of cognitive decline (23). Compared with control subjects, the erythrocyte FA profile of the MCI subjects was characterized as having a higher portion of AA (ω-6 PUFA) and EPA (ω-3 PUFA). In the EVA study, higher erythrocyte proportions of ω-6 PUFA were associated with a greater risk of cognitive decline, whereas a higher proportion of total ω-3 PUFA was associated with a protective effect (42). Our study similarly suggests the protective role of ω-3 PUFA on cognitive decline. Nevertheless, the observation of lower erythrocyte AA concentrations in patients with cognitive decline wasn't expected. Rather, our initial hypothesis was that the risk of cognitive decline would be associated with erythrocyte ω-6 PUFA higher rates, reflecting the deleterious imbalance of the ω-3 PUFA/ ω-6 PUFA ratio. However, Amman study's showed comparable results, with lower erythrocyte AA concentrations in patients with MCI, without any significant association with cognitive decline over time (34). The ω-6 PUFA erythrocyte assays may be more specifically related to cerebral structures and amyloid plaques as Hooper finds in his recent study (43). Even if after adjustment the results are not significant, the associations closest to significance were those between Aβ and erythrocyte membrane arachidonic and linoleic acid. Finally, we also observed a trend of lower DHA erythrocyte levels at baseline in rapid decliners, although results were not significant. This may be the result of a lack of statistical power due to the limited number of erythrocyte samples in this subgroup. Hence, our study emphasizes the importance of nutritional evaluation and, more specifically, FA concentrations assessment, as they are associated with cognitive impairment, and potentially cognitive decline, in older patients.

Our study is one of the largest prospective longitudinal cohort which specifically addresses the question of the role of FA concentrations on cognitive performance. Our results are consistent with those observed in the general population of older age (21). Furthermore, we were able to collect and analyse a large panel of fatty acids: ω-3 PUFA, ω-6 PUFA and saturated FA, measuring their concentrations both in the plasma and in the erythrocyte membranes, in all their fractions and portions. This biochemical rigor in FA assays allowed a precise interpretation of the role and the implication of each one. In addition, our study included patients referred for a cognitive complaint, for which early intervention may be of interest.

Yet some limitations must be acknowledged. First, this study was monocentric, which may affect the generalizability of the results. Second, the median follow-up time was limited to one year. An extended follow-up period would be needed to confirm the suspected associations of cognitive decline with certain FA profiles. In particular, a longer-term analysis may confirm the trend of lower erythrocyte membrane DHA concentrations in rapid decliners. Third, we were unable to clarify whether major NCD is the cause or the consequence of an abnormal FA profile, as the statistical associations do not prove causality (44). Indeed, while it could be hypothesized that low levels of plasma FAs preceded the clinical diagnosis of NCD, it is also possible that these levels decrease due to a change in eating habits secondary to the onset of NCD. Fourth, we were not able to classify the NCD in 45% of patients. An unknown proportion of these patients probably presented with a subtype of multifactorial NCD, formerly known as mixed dementia (which associates neurodegenerative and vascular mechanisms). However, we were unable to precisely evaluate its prevalence, mainly because diagnosis of certainty relies on necropsy, and still requires frequent clinical monitoring to confirm it. Under these circumstances, our data and results should not be analyzed according to the type of NCD but only according to the severity of cognitive impairment, as estimated by the MMSE.

Conclusion

Our findings provide evidence that patients with major NCD have a lower plasma concentration of ω-3 PUFA, namely EPA and DHA, while erythrocyte membrane concentrations of FA may have an interesting potential in the assessment of cognitive prognosis. Further longitudinal analyses based on the GERIOX cohort, including various fatty acids assays, seem relevant to better understand the role of ω-3 and ω-6 fatty acids in cognitive impairment, as well as to assess their potential use as biomarkers of cognitive prognosis, and may offer future therapeutic perspectives in reducing the risk of developing cognitive impairment. This study may open the field for further interventional studies, in which oral supplementations of both ω-3 and ω-6 FA may be better targeted and tested prospectively and in accordance to baseline measures.

Acknowledgdgments

The authors thank all participants in the GERIOX study. We gratefully acknowledge the important contribution of the medical and biochemistry staff research, nurses, biologist and doctors: Dr Jocelyne Drai, Dr Emilie Blond, Dr Pascale Rebaudet, Dr Claire Geodon, Dr Philippe Selva, Dr Jean-Philippe Gallat, Muriel Lemoine. We are particularly grateful to Dr Laurent Bitker for his considerable work in the realization of the statistical analysis and his assistance in the completion of this article.

Disclosures

The authors declare having no conflicts of interest related to this work. Authors' disclosures available online. No funder was requested. The “Hospices Civils de Lyon” had only committed as promoter.

Electronic supplementary material

Supplementary material is available for this article at https://doi.org/10.1007/s12603-018-1010-z and is accessible for authorized users.

SUPPLEMENTARY TABLES

mmc1.pdf (159.5KB, pdf)

References

  • 1.Scarmeas N, Stern Y, Tang M-X, Mayeux R, Luchsinger JA. Mediterranean diet and risk for Alzheimer’s disease. 2006;59(6):912–921. doi: 10.1002/ana.20854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Cederholm T, Salem N, Palmblad J. ω-3. Fatty Acids in the Prevention of Cognitive Decline in Humans. 2013;123.4(6):672–676. doi: 10.3945/an.113.004556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Vellas B, Carrie I, Gillette-Guyonnet S, Touchon J, Dantoine T, Dartigues JF, et al. MAPT Study: A multidomain approch for preventing Alzheimer’s disease: design and baseline data. J Prev Alzheimers Dis. 2014;1(1):13–22. PubMed PMID: 26594639, PMCID 4652787. [PMC free article] [PubMed] [Google Scholar]
  • 4.Fougère B, de Souto Barreto p, Goisser S, Soriano G, Guyonnet S, Andrieu S, et al, Red blood cell membrane omega-3 fatty acid levels and physical performance: Crosssectional data from the MAPT study, Clin Nutr Edinb Scotl., 2017 [DOI] [PubMed]
  • 5.Issa AM, Mojica WA, Morton SC, Traina S, Newberry SJ, Hilton LG, et al. The efficacy of omega-3 fatty acids on cognitive function in aging and dementia: a systematic review. Dement Geriatr Cogn Disord. 2006;21(2):88–96. doi: 10.1159/000090224. 10.1159/000090224 PubMed PMID: 16340205. [DOI] [PubMed] [Google Scholar]
  • 6.Yanai H. Effects of N-3 Polyunsaturated Fatty Acids on Dementia. J Clin Med Res. 2017;9(1):1–9. doi: 10.14740/jocmr2815w. 10.14740/jocmr2815w PubMed PMID: 27924168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Solfrizzi V, Capurso C, D’Introno A, Colacicco AM, Frisardi V, Santamato A, et al. Dietary fatty acids, age-related cognitive decline, and mild cognitive impairment. J Nutr Health Aging. 2008;12(6):382–386. doi: 10.1007/BF02982670. 10.1007/BF02982670 PubMed PMID: 18548175. [DOI] [PubMed] [Google Scholar]
  • 8.Samieri C, Féart C, Letenneur L, Dartigues J-F, Pérès K, Auriacombe S, et al. Low plasma eicosapentaenoic acid and depressive symptomatology are independent predictors of dementia risk. Am J Clin Nutr [Internet]. 2008;88(3):714–721. doi: 10.1093/ajcn/88.3.714. 10.1093/ajcn/88.3.714 PubMed PMID: 18779288. [DOI] [PubMed] [Google Scholar]
  • 9.Whalley LJ, Deary IJ, Starr JM, Wahle KW, Rance KA, Bourne VJ, et al. n–3 Fatty acid erythrocyte membrane content, APOE ε4, and cognitive variation: an observational follow-up study in late adulthood. Am J Clin Nutr [Internet]. 2008;87(2):449–454. doi: 10.1093/ajcn/87.2.449. 10.1093/ajcn/87.2.449 PubMed PMID: 18258638. [DOI] [PubMed] [Google Scholar]
  • 10.Schuchardt Jp, Köbe T, Witte V, Willers J, Gingrich A, Tesky V, et al. Genetic Variants of the FADS Gene Cluster Are Associated with Erythrocyte Membrane LC PUFA Levels in Patients with Mild Cognitive Impairment. J Nutr Health Aging. 2016;20(6):611–620. doi: 10.1007/s12603-016-0720-3. 10.1007/s12603-016-0720-3 PubMed PMID: 27273350. [DOI] [PubMed] [Google Scholar]
  • 11.El Zoghbi M, Boulos C, Amal AH, Saleh N, Awada S, Rachidi S, et al. Association between cognitive function and nutritional status in elderly: A cross-sectional study in three institutions of Beirut—Lebanon. Geriatr Ment Health Care. 2013;1(4):73–81. 10.1016/j.gmhc.2013.04.007 [Google Scholar]
  • 12.Eriksdotter M, Vedin I, Falahati F, Freund-Levi Y, Hjorth E, Faxen-Irving G, et al. Plasma Fatty Acid Profiles in Relation to Cognition and Gender in Alzheimer’s Disease Patients During Oral Omega-3 Fatty Acid Supplementation: The OmegAD Study. J Alzheimers Dis JAD. 2015;48(3):805–812. doi: 10.3233/JAD-150102. 10.3233/JAD-150102 PubMed PMID: 26402079. [DOI] [PubMed] [Google Scholar]
  • 13.MacLean CH, Issa AM, Newberry SJ, Mojica WA, Morton SC, Garland RH, et al, Effects of Omega-3 Fatty Acids on Cognitive Function with Aging, Dementia, and Neurological Diseases, Agency for Healthcare Research and Quality (US), 2005 [PMC free article] [PubMed]
  • 14.Revel F, Gilbert T, Roche S, Drai J, Blond E, Ecochard R, et al. Influence of oxidative stress biomarkers on cognitive decline. J Alzheimers Dis JAD. 2015;45(2):553–560. doi: 10.3233/JAD-141797. 10.3233/JAD-141797 PubMed PMID: 25589716. [DOI] [PubMed] [Google Scholar]
  • 16.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–198. doi: 10.1016/0022-3956(75)90026-6. 10.1016/0022-3956(75)90026-6 PubMed PMID: 1202204. [DOI] [PubMed] [Google Scholar]
  • 17.American Psychiatric Association. Diagnostic and statistical manual of mental disorders. (5th ed). Washington, DC; 2013.
  • 18.Sachdev PS, Mohan A, Taylor L, Jeste DV. DSM-5 and mental disorders in older individuals: an overview. Harv Rev Psychiatry [Internet]. 2015;23(5):320. doi: 10.1097/HRP.0000000000000090. 10.1097/HRP.0000000000000090 PubMed PMID: 26332215, PMCID 4562208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, et al. The diagnosis of mild cognitive impairment 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 J Alzheimers Assoc. 2011;7(3):270–279. doi: 10.1016/j.jalz.2011.03.008. 10.1016/j.jalz.2011.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Portet F, Ousset PJ, Visser PJ, Frisoni GB, Nobili F, Scheltens p, et al, Mild cognitive impairment (MCI) in medical practice: a critical review of the concept and new diagnostic procedure. Report of the MCI Working Group of the European Consortium on Alzheimer’s Disease, J Neurol Neurosurg Psychiatry, 2006, 714–8 [DOI] [PMC free article] [PubMed]
  • 21.Dullemeijer C, Durga J, Brouwer IA, Rest O, van de Kok FJ, Brummer R-J, et al. n−3 Fatty acid proportions in plasma and cognitive performance in older adults. Am J Clin Nutr. 2007;86(5):1479–1485. doi: 10.1093/ajcn/86.5.1479. 10.1093/ajcn/86.5.1479 PubMed PMID: 17991662. [DOI] [PubMed] [Google Scholar]
  • 22.Folch J, Ascoli I, Lees M, Meath JA, LeBARON N. Preparation of lipide extracts from brain tissue. J Biol Chem. 1951;191(2):833–841. PubMed PMID: 14861228. [PubMed] [Google Scholar]
  • 23.Cui Y, Chen X, Liu L, Xie W, Wu Y, Wu Q, et al. Gas chromatography-mass spectrometry analysis of the free fatty acids in serum obtained from patients with Alzheimer’s disease. Biomed Mater Eng. 2015;26(1):S2165–2177. doi: 10.3233/BME-151522. PubMed PMID: 26405996. [DOI] [PubMed] [Google Scholar]
  • 24.Nutrients. 2016;8(7) [Google Scholar]
  • 25.Dupuy A, Le Faouder p, Vigor C, Oger C, Galano J-M, Dray C, et al. Simultaneous quantitative profiling of 20 isoprostanoids from omega-3 and omega-6 polyunsaturated fatty acids by LC-MS/MS in various biological samples. Anal Chim Acta. 2016;921:46–58. doi: 10.1016/j.aca.2016.03.024. 10.1016/j.aca.2016.03.024 PubMed PMID: 27126789, PMCID 5517622. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.; 2014.
  • 28.Ma J, Folsom AR, Shahar E, Eckfeldt JH. Plasma fatty acid composition as an indicator of habitual dietary fat intake in middle-aged adults. The Atherosclerosis Risk in Communities (ARIC) Study Investigators. Am J Clin Nutr. 1995;62(3):564–571. doi: 10.1093/ajcn/62.3.564. 10.1093/ajcn/62.3.564 PubMed PMID: 7661118. [DOI] [PubMed] [Google Scholar]
  • 29.Baylin A, Kim MK, Donovan-Palmer A, Siles X, Dougherty L, Tocco p, et al. Fasting whole blood as a biomarker of essential fatty acid intake in epidemiologic studies: comparison with adipose tissue and plasma. Am J Epidemiol. 2005;162(4):373–381. doi: 10.1093/aje/kwi213. 10.1093/aje/kwi213 PubMed PMID: 16014782. [DOI] [PubMed] [Google Scholar]
  • 30.Arab L. Biomarkers of fat and fatty acid intake. J Nutr. 2003;133(3):925S–932S. doi: 10.1093/jn/133.3.925S. 10.1093/jn/133.3.925S PubMed PMID: 12612178. [DOI] [PubMed] [Google Scholar]
  • 31.Beydoun MA, Kaufman JS, Satia JA, Rosamond W, Folsom AR. Plasma n-3 fatty acids and the risk of cognitive decline in older adults: the Atherosclerosis Risk in Communities Study. Am J Clin Nutr. 2007;85(4):1103–1111. doi: 10.1093/ajcn/85.4.1103. 10.1093/ajcn/85.4.1103 PubMed PMID: 17413112. [DOI] [PubMed] [Google Scholar]
  • 32.Huang M-C, Brenna JT, Chao AC, Tschanz C, Diersen-Schade DA, Hung H-C. Differential Tissue Dose Responses of (n-3) and (n-6) PUFA in Neonatal Piglets Fed Docosahexaenoate and Arachidonoate. J Nutr. 2007;137(9):2049–2055. doi: 10.1093/jn/137.9.2049. 10.1093/jn/137.9.2049 PubMed PMID: 17709441. [DOI] [PubMed] [Google Scholar]
  • 33.Brenner SR. Red Blood Cell Omega-3 Fatty Acid Levels and Markers of Accelerated Brain Aging. Neurology. 2012;79(1):106–107. doi: 10.1212/WNL.0b013e31825e41b2. 10.1212/WNL.0b013e31825e41b2 PubMed PMID: 22753447. [DOI] [PubMed] [Google Scholar]
  • 34.Ammann EM, Pottala JV, Harris WS, Espeland MA, Wallace R, Denburg NL, et al. ω-3 fatty acids and domain-specific cognitive aging: secondary analyses of data from WHISCA. Neurology. 2013;81(17):1484–1491. doi: 10.1212/WNL.0b013e3182a9584c. 10.1212/WNL.0b013e3182a9584c PubMed PMID: 24068783, PMCID 3888166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hodson L, Skeaff CM, Fielding BA. Fatty acid composition of adipose tissue and blood in humans and its use as a biomarker of dietary intake. Prog Lipid Res. 2008;47(5):348–380. doi: 10.1016/j.plipres.2008.03.003. 10.1016/j.plipres.2008.03.003 PubMed PMID: 18435934. [DOI] [PubMed] [Google Scholar]
  • 36.PLoS ONE [Internet]. 2015;10(3) [Google Scholar]
  • 37.Mazereeuw G, Lanctôt KL, Chau SA, Swardfager W, Herrmann N. Effects of ω-3 fatty acids on cognitive performance: a meta-analysis. Neurobiol Aging. 2012;33(7):29–1482.e17. doi: 10.1016/j.neurobiolaging.2011.12.014. 10.1016/j.neurobiolaging.2011.12.014 [DOI] [PubMed] [Google Scholar]
  • 38.D’Ascoli TA, Mursu J, Voutilainen S, Kauhanen J, Tuomainen T-p, Virtanen JK. Association between serum long-chain omega-3 polyunsaturated fatty acids and cognitive performance in elderly men and women: The Kuopio Ischaemic Heart Disease Risk Factor Study. Eur J Clin Nutr. 2016;70(8):970–975. doi: 10.1038/ejcn.2016.59. 10.1038/ejcn.2016.59 PubMed PMID: 27071510. [DOI] [PubMed] [Google Scholar]
  • 39.Tully AM, Roche HM, Doyle R, Fallon C, Bruce I, Lawlor B, et al. Low serum cholesteryl ester-docosahexaenoic acid levels in Alzheimer’s disease: a case-control study. Br J Nutr. 2003;89(4):483–489. doi: 10.1079/BJN2002804. 10.1079/BJN2002804 PubMed PMID: 12654166. [DOI] [PubMed] [Google Scholar]
  • 40.Yin Y, Fan Y, Lin F, Xu Y, Zhang J. Nutrient biomarkers and vascular risk factors in subtypes of mild cognitive impairment: a cross-sectional study. J Nutr Health Aging. 2015;19(1):39–47. doi: 10.1007/s12603-014-0510-8. 10.1007/s12603-014-0510-8 PubMed PMID: 25560815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Conquer JA, Tierney MC, Zecevic J, Bettger WJ, Fisher RH. Fatty acid analysis of blood plasma of patients with alzheimer’s disease, other types of dementia, and cognitive impairment. Lipids. 2000;35(12):1305–1312. doi: 10.1007/s11745-000-0646-3. 10.1007/s11745-000-0646-3 PubMed PMID: 11201991. [DOI] [PubMed] [Google Scholar]
  • 42.Heude B, Ducimetière p, Berr C, Cognitive decline and fatty acid composition of erythrocyte membranes—The EVA Study, Am J Clin Nutr, 2003, 803–8 [DOI] [PubMed]
  • 43.Hooper C D S, Barreto p, Payoux p, Salabert AS, Guyonnet S, Andrieu S, et al. Cross-sectional associations of cortical β-amyloid with erythrocyte membrane longchain polyunsaturated fatty acids in older adults with subjective memory complaints. J Neurochem. 2017;142(4):589–596. doi: 10.1111/jnc.14062. 10.1111/jnc.14062 PubMed PMID: 28467602. [DOI] [PubMed] [Google Scholar]
  • 43.Lopes da Silva S, Vellas B, Elemans S, Luchsinger J, Kamphuis p, Yaffe K, et al, Plasma nutrient status of patients with Alzheimer’s disease: Systematic review and meta-analysis, Alzheimers Dement, 2013, S1552526013024643 [DOI] [PubMed]

Uncited references

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SUPPLEMENTARY TABLES

mmc1.pdf (159.5KB, pdf)

Articles from The Journal of Nutrition, Health & Aging are provided here courtesy of Elsevier

RESOURCES