If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Faculté de médecine et des sciences de la santé, Université de Sherbrooke, Sherbrooke, QC, CanadaCentre de Recherche sur le Vieillissement, CIUSSS de l'Estrie-CHUS, Sherbrooke, QC, Canada
Centre de Recherche sur le Vieillissement, CIUSSS de l'Estrie-CHUS, Sherbrooke, QC, CanadaDépartement de Psychologie, Université de Sherbrooke, Sherbrooke, QC, Canada
Faculté de médecine et des sciences de la santé, Université de Sherbrooke, Sherbrooke, QC, CanadaCentre de Recherche sur le Vieillissement, CIUSSS de l'Estrie-CHUS, Sherbrooke, QC, CanadaInstitut de la nutrition et des aliments fonctionnels, Université Laval, Québec, QC, Canada
Faculté de médecine et des sciences de la santé, Université de Sherbrooke, Sherbrooke, QC, CanadaCentre de Recherche sur le Vieillissement, CIUSSS de l'Estrie-CHUS, Sherbrooke, QC, CanadaInstitut de la nutrition et des aliments fonctionnels, Université Laval, Québec, QC, Canada
Faculté de médecine et des sciences de la santé, Université de Sherbrooke, Sherbrooke, QC, CanadaCentre de Recherche sur le Vieillissement, CIUSSS de l'Estrie-CHUS, Sherbrooke, QC, CanadaCentre de recherche, Institut universitaire de gériatrie de Montréal, CIUSSS du Centre-Sud de l'Ile-de-Montréal, Canada
Faculté de médecine et des sciences de la santé, Université de Sherbrooke, Sherbrooke, QC, CanadaCentre de Recherche sur le Vieillissement, CIUSSS de l'Estrie-CHUS, Sherbrooke, QC, CanadaInstitut de la nutrition et des aliments fonctionnels, Université Laval, Québec, QC, Canada
A 6-month n-3 PUFA supplementation did not improve cognitive performance in cognitively healthy adults.
•
APOE carriers and age were not effect modifiers for any outcomes.
•
Those with low baseline memory scores might benefit more from the n-3 PUFA supplement.
Abstract
Objective
To investigate whether omega-3 polyunsaturated fatty acids (n-3 PUFA) supplementation improve cognitive performance and if apolipoprotein E (APOE) genotype or age were effect modifiers.
Methods
Healthy adults of 20 to 80 years old (n = 193) were completed a 6-month double-blind randomized controlled trial with two groups: 2.5 g/day of n-3 PUFA or a placebo. Primary outcomes were visuospatial ability and working memory and secondary outcomes were episodic memory and executive function, measured at baseline and 6 months.
Results
Cognitive performances did not significantly differ between groups on primary or secondary outcomes after 6 months of treatment. APOE carriers and age were not effect modifiers for any outcomes. Those with low episodic memory scores and taking the n-3 PUFA supplement, significantly improved their scores (p = 0.043).
Conclusions
A 6-month n-3 PUFA supplementation did not improve cognitive performance in cognitively healthy adults and APOE status or age were not effect modifiers.
]. This is in line with other evidence from prospective studies acquired over two decades pointing towards a protective role of n-3 PUFA intake or in the blood on cognitive decline in aging [
]. There is however a disconnection between evidence stemming from prospective studies and those from randomized controlled trials (RCT): the latter concluding that n-3 PUFA supplementation has no or at best, modest benefits on cognitive performance or on progression of cognitive deficits in AD [
This discrepancy between the physiological functions of n-3 PUFA in the brain and RCT results on cognition might stem from the lack of proper consideration of participants’ age and apolipoprotein E (APOE) genotype, both being leading risk factors of cognitive decline and late-onset AD [
Higher plasma n-3 fatty acid status in the moderately healthy elderly in southern Québec: higher fish intake or aging-related change in n-3 fatty acid metabolism?.
Prostaglandins, Leukotrienes and Essential Fatty Acids (PLEFA).2010; 82: 277-280
]. Indeed, we and others have shown that during aging, eicosapentaenoic acid (EPA) levels are approximately two-fold higher in plasma lipids of older adults compared to younger individuals before any supplementation with fish oil [
Higher plasma n-3 fatty acid status in the moderately healthy elderly in southern Québec: higher fish intake or aging-related change in n-3 fatty acid metabolism?.
Prostaglandins, Leukotrienes and Essential Fatty Acids (PLEFA).2010; 82: 277-280
]. Hence, if plasma n-3 PUFA is higher in older adults, we would expect better cognition compared to young adults, which is counterintuitive. Our team also reported that after supplementation with 3.2 g of n-3 PUFA per day for 6 weeks, the rise in DHA level was 60% lower in plasma triglycerides (TG) of APOE4 carriers, the most important genetic risk factor of late-onset AD, compared to non-carriers [
]. Taken together, age and being a carrier of APOE4 genotype are potentially associated with dysregulation in the uptake of n-3 PUFA by tissues and organs, which might influence outcomes of RCTs testing the effects of n-3 PUFA supplementation on health.
To our knowledge, previous RCTs in cognitively healthy participants investigating potential benefits of n-3 PUFA supplementation on cognition were conducted within well-defined age groups, either in young adults [
An 18-mo randomized, double-blind, placebo-controlled trial of DHA-rich fish oil to prevent age-related cognitive decline in cognitively normal older adults.
No effect of n-3 long-chain polyunsaturated fatty acid (EPA and DHA) supplementation on depressed mood and cognitive function: a randomised controlled trial.
Association between cognitive function and supplementation with omega-3 PUFAs and other nutrients in >/= 75 years old patients: a randomized multicenter study.
An 18-mo randomized, double-blind, placebo-controlled trial of DHA-rich fish oil to prevent age-related cognitive decline in cognitively normal older adults.
]. Therefore, the objectives of this study were to investigate 1) whether supplementation with n-3 PUFA during 6 months can improve cognitive performances in healthy adults aged between 20 and 80 years old; and 2) whether apolipoprotein E (APOE) genotype or age are effect modifiers of this cognitive improvement.
2. Patients and methods
2.1 Study design and participants
This study was a 6-month randomized, placebo-controlled trial conducted at the Research Center on Aging of the center Intégré Universitaire de Santé et des Services Sociaux de l'Estrie-center hospitalier universitaire de Sherbrooke (CIUSSSE-CHUS), Sherbrooke (QC), Canada. The protocol is registered in a public-access clinical trial database (www.clinicaltrials.gov) [NCT01625195]. The Research Ethics Board of the CIUSSSE-CHUS provided approval of this trial. All participants provided written informed consent.
Potential participants were invited to contact the research team through advertising in local pharmacies, social media (i.e. Facebook), existing database of past research participants, and word-of-mouth in Sherbrooke area. Recruitment extended from August 2012 to September 2016. During the screening phone call, participants were excluded if they had diabetes, cancer diagnosis in the past 6 months, uncontrolled hyperthyroidism or hypothyroidism, autoimmune disorders (i.e. rheumatoid arthritis, multiple sclerosis), as well as they were current smokers or had alcoholism, uncorrected visual problems, native language other than the French language in usage in the province of Quebec, any cognitive disorders either self-reported or as screened by the Mini-Mental State Examination (MMSE) [
], brain injuries (i.e. concussion) or any mental health disorders (i.e. severe depression) that might influence cognitive performances. Participants were also excluded if they consumed n-3 PUFA supplements within 6 months prior to recruitment, were eating regularly more than two servings of fatty fish per week, or were performing intense physical training (self-reported), since intense exercise modify lipid metabolism by increasing phospholipid concentrations in muscular tissue [
]. Medication had to be stable for at least 6 months prior to enrollment. Screened participants were then invited to the research center for a blood draw after fasting for at least 12 h. Blood TGs (reference value ≤1.7 mmol/L), HDL-cholesterol (≥1.3 mmol/L), glucose (<6 mmol/L), thyrotropin-releasing hormone (<3.5 mUI/L), C-reactive protein (<10 mg/L), and white blood cell count (4000–11,000 μL), were analyzed at the CHUS clinical laboratory using state-of-the-art protocols/methods. Individuals with values outside the reference range for any of these biomarkers were also excluded.
2.2 Randomization and blinding
Participants were randomly assigned to the n-3 PUFA group or the placebo group using a computerized, web-based, random number table with stratification by sex and age (20–39; 40–59; 60–80 years old) in a 1:1 ratio. Placebo and treatment capsules were identical in size, shape and flavor. Capsules were provided in identical opaque drug containers that were coded and distributed by the clinical research nurse according to the randomization scheme. Both research staff and participants were blinded to treatment allocation.
2.3 Procedures
All participants who met the eligibility criteria underwent a baseline visit for a detailed medical history and physical assessment including the analysis of a blood sample collected after a 12-hour overnight fast. Cognitive assessments, fasting blood samples, and anthropometric measurements were obtained at baseline and after 6 months. During the study, participants were asked to maintain their daily routine (eating patterns, physical activity, and alcohol consumption). The n-3 PUFA supplementation consisted of four capsules of fish oil per day (1.0 g of oil/capsule) providing 1.7 g EPA and 0.8 g DHA daily in ethyl ester form (Ocean Nutrition Canada; Supplemental Table 1). The rationale for this supplement was based on the daily dosage of an RCT with n-3 PUFA with positive cognitive outcomes [
Effects of supplementation with n-3 polyunsaturated fatty acids on cognitive performance and cardiometabolic risk markers in healthy 51 to 72 years old subjects: a randomized controlled cross-over study.
].. Placebo capsules contained high–oleic acid soybean/corn oil (50/50) providing no DHA or EPA (Supplemental Table 1). In both groups, participants were asked to consume two capsules at breakfast and two capsules at dinner.
Participants kept weekly diaries to record their consumption of the n-3 PUFA or placebo capsules, fishes and seafoods (type and portion size), and monitor any deviations from the study protocol (e.g. use of nutritional supplements, changes in dietary habits). No notable deviations were reported. Treatment adherence was determined by using pill-counting of leftover capsules and analysis of plasma phospholipid fatty acids at baseline and after 6 months.
2.4 Blood biochemistry and plasma phospholipid analysis
Plasma lipids were extracted from 200 μL of plasma using the method described by Folch et al. [
]. 1,2-dipentadecanoyl-sn‑glycero-3-phosphocholine (Avanti Polar Lipids Inc) was added to each sample as an internal standard, as previously described [
Interaction between BMI and APOE genotype is associated with changes in the plasma long-chain-PUFA response to a fish-oil supplement in healthy participants.
]. Elutions with the different solvent was conducted under normal gravity (no vacuum). The purity of the fractions were validated by thin-layer chromatography as previously described [
]. FA methyl esters were then generated by adding boron trifluoride in methanol (12%; Sigma-Aldrich) to the phospholipids and heating the solution at 90 °C for 30 min. The FA methyl esters were analyzed by gas chromatography as previously described [
Apolipoprotein E genotyping was performed at the Plateforme de Séquençage et Génotypage des Génomes, center de Recherche du CHUS. Genomic DNA was extracted from 200 μL of unfrozen whole blood using the QIAamp DNA extraction kit (Qiagen Hilden, Germany) according to the manufacturer's instructions. DNA was genotyped with TaqMan assays using the predesigned TaqMan SNP Genotyping Assay (Applied Biosystems, Foster City, CA) for rs429358 (Assay ID C__3,084,793_20) and rs7412 (C__904,973_10). A Stratagene Mx3005P multiplex quantitative PCR system (Agilent Technologies, La Jolla, CA) was used to perform real-time PCR amplification. Reactions were carried out in a total volume of 10 µl containing 10 ng of DNA (2.5 µl at 4 ng/µl)), 0.5 µl of 20X Primer/probe mix, 5 µl 2X TaqMan genotyping mix (Applied Biosystem) and 2 µl of H2O. Samples underwent 40 cycles of amplification at 95 °C. Each cycle consisted of 30 s at 95 °C, and 1 min at 60 °C. Participants were classified into two genetic groups based upon the presence of at least one APOE4 allele (carriers of the APOE4 allele vs. non-carriers).
2.6 Cognitive testing
At baseline and after 6 months of supplementation, participants underwent cognitive assessment performed by neuropsychologists blinded from participants’ treatment. A large battery of cognitive tests were (pre)selected to assess performances in visuospatial ability and working memory as primary outcomes, and episodic memory and executive function as secondary outcomes. Primary and secondary outcomes were selected based on preliminary (unpublished) data collected in a pre-post quasi-experimental study performed by our group [
Chouinard-Watkins R., Lorrain D., Viens I., Normand M.P., Perron P., Rioux-Perreault C., Cunnane S., P. M., Performances cognitives avant et après une supplémentation en huile de poisson., in: none (Ed.) 10e journée scientifique du Réseau québécois de recherche sur le vieillissement, none, Montreal, Canada, 2012.
]. For each test, scores were examined by a neuropsychologist for ceiling effects and reproducibility in this sample of healthy adults with a wide age range [
].. In the end, the most sensitive test score was selected for each cognitive domain. In that respect, visuospatial ability was measured using the spatial Cued Letter Discrimination task adapted from Greenwood et al [
Effects of apolipoprotein E genotype on spatial attention, working memory, and their interaction in healthy, middle-aged adults: results From the National Institute of Mental Health's BIOCARD study.
].. More specifically, the reaction time (in milliseconds) of the incongruent arrow with a cue time of 200 ms was used since it is the most difficult condition of this test, so the most discriminant score. Working memory was measured using the Brown-Peterson task [
]. This version of the test is a shorter French version of the Category Cued Recall portion of the Double Memory Test developed initially by Buschke and colleagues [
]. A detailed description of each of these cognitive tests as well as their scoring procedure are provided in Supplemental Material.
2.7 Descriptive variables
Age, sex, education, and ethnicity were determined at baseline. Body weight (kg) and height (m) were measured and the body mass index (BMI; kg/m2) calculated. The Montreal Cognitive Assessment test (MoCA) [
] were administered to describe global cognitive performance.
2.8 Statistical analysis
The sample size for this trial was calculated using data from a previous study which revealed a significant difference between the intervention group and the control group on scores from cognitive tests assessing visuospatial ability [
], the total number of participants required has been estimated to be 64 per group. Taking into account (i) 15% loss to follow up at 6 months was estimated and (ii) the stratification by APOE4 genotype (20% carriers in Canada, [
]), the sample size was set to 100 participants per group, providing at least 20 APOE4 carriers per group.
Primary analyses were on an intention-to-treat (ITT) basis and included all participants randomized and who completed the trial (i.e. the cognitive assessment at baseline and at 6 months). Individuals were analyzed in the group to which they were randomized regardless of their level of adherence to treatment. Baseline characteristics and raw scores of each cognitive test were compared between groups with independent-sample T tests for continuous variables and Χ2 tests, as appropriate. Comparisons of baseline characteristics and raw scores of cognitive tests between those who completed the trial and those who dropped out was performed to document the extent and nature of any attrition bias.
Based on pill count, adherence to treatment (n-3 PUFA supplementation or placebo) was calculated for each participant as percentage. Adherence was compared between groups using an independent-sample T test. Only participants with ≥80% adherence to treatment over the 6 months established by pill counts were included in the per-protocol analysis. Plasma phospholipid concentrations of EPA and DHA were compared within groups between baseline and after 6 months of treatment using paired-sample T tests, and differences (concentration at 6-month minus that at baseline) were compared between groups using independent-sample T tests. Also, in order to determine whether the response to the n-3 PUFA supplementation was indeed influenced by age and APOE4 status (as postulated above), an analysis of covariance (ANCOVA) was performed. Specifically, log-transformed plasma phospholipid concentrations of EPA+DHA after 6 months of n-3 PUFA supplementation, adjusted for the concentrations at baseline and sex, were analyzed in function of the APOE4 status (carriers vs. non-carriers) and age group (20–39, 40–59, 60–80 years old).
Cognitive scores measured as “time” (i.e. Trail Making Test B and spatial Cued Letter Discrimination task) were log-transformed because of departure from normality and multiplied by −1 so that increasing scores always indicate better performance. For each cognitive test, z scores at baseline and 6 months were calculated using the mean and standard deviation at baseline for all participants altogether.
The ITT and per-protocol analyses were performed using an ANCOVA comparing post-intervention z scores between groups, adjusted for baseline z scores, for the primary and secondary outcomes. The ITT ANCOVA analysis was also performed on unadjusted scores in each cognitive domain.
To test whether APOE genotype was an effect modifier of treatment, an ANCOVA adding APOE genotype (carriers vs non-carriers) to the model as well as the interaction term with treatment was computed. As well, to determine if age was an effect modifier of treatment, an ANCOVA adding three age groups (20–39, 40–59, 60–80 years old) to the model as well as the interaction term with treatment was computed.
Exploratory ITT and per-protocol analyses were also performed to examine whether n-3 PUFA supplementation was more effective in those with low baseline cognitive scores or in those with low baseline phospholipid n-3 PUFA plasma concentration. For the latter, tertile of phospholipid n-3 PUFA were generated as previously described [
Homocysteine status modifies the treatment effect of Omega-3 fatty acids on cognition in a randomized clinical trial in mild to moderate Alzheimer's disease: the OmegAD study.
]. Then, this grouping variable was added as the interaction term with treatment group to the ANCOVA model. Significance was considered when p <0.05. All analyses were performed using SPSS software version 23 (IBM Corp).
2.9 Data availability statement
The anonymized data will be shared upon request from any qualified investigator.
3. Results
3.1 Baseline characteristics
Of the 272 community volunteers screened for eligibility, 243 were randomly assigned to the n-3 PUFA supplementation or placebo groups, from which seven participants were later found to be non-eligible to the present trial, leaving 118 participants per group (Fig. 1). Of these 236 participants, 193 (79%) completed the 6-month trial and were included in ITT analyses (Fig. 1). Reasons for dropping out included minor adverse events, health issues (not related to the trial), and other unspecified personal reasons (Fig. 1). The 43 participants who dropped out did not differ significantly from those who completed the trial on baseline characteristics including cognitive scores (Supplemental Table 2).
Fig. 1Participant flow through the study. Per-protocol analysis was based on treatment adherence of ≥80%.
In the ITT analysis, participants allocated to the n-3 PUFA (n = 96) and placebo (n = 97) groups did not differ on most baseline characteristics. Also, plasma phospholipid DHA concentrations were significantly higher at baseline in the placebo group (p = 0.009). Similar differences, although attenuated, were observed between the original groups, i.e. before exclusion of dropouts (Supplemental Table 3). In the per-protocol analysis, a total of 172 participants were included (n-3 PUFA, n = 89 and placebo, n = 83, Fig. 1). Their baseline characteristics are presented in Supplemental Table 4 and their profiles was those of the ITT analysis.
3.2 Adherence, safety and response to n-3 PUFA supplementation
Based on the capsule count, median (IQR) adherence to treatment was 96.9% (93.6%−99.1%) for participants receiving the n-3 PUFA supplementation and 97.7% (95.3%−98.8%) for those receiving the placebo; no difference between groups was detected (p = 0.930). During the study, participants consumed less than two fish servings/week, did not consume other supplements than those provided during the trial, and did not modify their lifestyle. In total, 10 participants (n-3 PUFA n = 4, placebo n = 6) dropped out because of side effects thought to be related to treatment, notably gastrointestinal complaints, and nausea (Fig. 1).
Mean plasma concentrations of phospholipid EPA and DHA increased significantly (p<0.001) after 6 months in both the n-3 PUFA (EPA: from 21.4 to 100.4 mg/L, DHA: from 65.5 to 116.2 mg/L) and placebo groups (EPA: from 23.1 to 29.3 mg/L, DHA: from 74.7 to 93.1 mg/L). As expected, the increase was found to be significantly higher in the n-3 PUFA group (p < 0.001). Results of the ANCOVA revealed no significant effect of the APOE4 status on plasma phospholipid concentrations of EPA+DHA after 6 months of n-3 PUFA supplementation (adjusted geometric means [95%CI]; APOE4 carriers: 219.8 [196.3–246.6] mg/L; non-carriers: 196.8 [183.2–211.8] mg/L; p = 0.101). However, the post-supplementation concentrations increased with age (<40 years: 190.5 [17.6–210.4] mg/L; 40–59 years: 212.8 [190.1–238.2] mg/L; ≥60 years: 222.3 [200.0–247.7] mg/L), with a significant difference between those aged 60 years or older and those <40 years (p = 0.048).
3.3 Overall effect of the n-3 PUFA supplementation on cognition
Performances in all four cognitive domains did not significantly differ between groups after the 6-month n-3 PUFA supplementation in comparison with a placebo, according to the ITT analyses (Table 2). Similar results were obtained when post-intervention scores were not adjusted for baseline scores (Supplementary Table 5) and in the per protocol analyses (Supplementary Table 6).
Table 1Baseline characteristics of the participants.
n-3 PUFA(n = 96)
Placebo(n = 97)
p value
Age (years), mean (SD)
49.4 (16.4)
50.5 (16.6)
0.645
Women, n (%)
66 (68.8)
63 (64.9)
0.575
Age groups, n (%) 20–39 40–59 ≥60
30 (31.3) 31 (32.3) 35 (36.5)
29 (29.9) 27 (27.8) 41 (42.3)
0.683
Ethnicity, n (%) Caucasian Others
95 (99.0) 1 (1.0)
96 (99.0) 1 (1.0)
0.994
Education level (years), mean (SD)
16.1 (3.6)
15.2 (2.7)
0.068
APOE4 carriers, n (%)
26 (27.1)
16 (16.5)
0.075
BMI (kg/m2), mean (SD)
25.7 (4.4)
26.5 (5.4)
0.314
Mini-Mental State Evaluation (/30), mean (SD)
29.2 (1.0)
29.2 (1.1)
0.937
Modified Mini-Mental State evaluation (/100), mean (SD)
97.7 (3.0)
97.5 (3.0)
0.671
Montreal Cognitive Assessment score (/30), mean (SD)
26.5 (2.5)
26.4 (2.4)
0.873
EPA (mg/L), mean (SD)
21.4 (11.2)
23.1 (9.9)
0.251
DHA (mg/L), mean (SD)
65.5 (22.8)
74.7 (25.7)
0.009
EPA + DHA (mg/L), mean (SD)
88.6 (32.7)
97.8 (32.0)
0.051
Visuospatial ability
Cued Letter, reaction time (ms), mean (SD)
800.6 (169.6)
812.5 (157.3)
0.617
Working memory
Brown-Peterson, score /60, mean (SD)
46.4 (7.9)
46.9 (7.6)
0.626
Episodic memory
Delayed Cued Recall score, /48, mean (SD)
23.7 (6.6)
24.5 (7.2)
0.389
Executive function
Trail Making Test part B, seconds, mean (SD)
71.2 (30.7)
73.5 (30.6)
0.608
BMI: body mass index, APOE4: apolipoprotein E epsilon 4. Chi square and independent Student's T tests were performed to compare groups.
Visuospatial ability and executive function: higher score mean worse performance.
Working and episodic memory: higher score means better performance.
Table 2Standardized adjusted post-intervention cognitive scores according to the treatment groups.
Outcome
n-3 PUFA (n = 96)
Placebo (n = 97)
p value
Adjusted mean (SEM)
Adjusted mean (SEM)
Visuospatial ability
0.087 (0.068)
0.132 (0.067)
0.639
Working memory
0.139 (0.078)
0.012 (0.078)
0.251
Episodic memory
0.286 (0.059)
0.227 (0.059)
0.483
Executive function
0.199 (0.069)
0.195 (0.069)
0.967
Visuospatial ability: Cued Letter Discrimination incongruent task, Working memory: Brown-Peterson score, Episodic memory: RI-48 delayed cued recall score, Executive function: Trail Making Test part B, ANCOVA adjusted for baseline standardized score.
Note: missing values were found for visuospatial ability (n = 5 in n-3 PUFA group and n = 1 in placebo group) and for executive function (n = 1 in placebo group).
3.4 Effect modification by APOE4 and age of the n-3 PUFA supplementation on cognition
In ITT and per-protocol analyses, APOE4 status as well as age were not significant effect modifiers of treatment (Table 3 and Supplemental Table 7). However, using the ITT and the per-protocol analysis, APOE4 carriers on either supplement for 6 months significantly improved episodic memory scores compared to non-carriers (p = 0.037 for ITT and p = 0.033 in per-protocol). Besides, adults aged between 60 and 80 years old on either treatment for 6 months had a significantly decreased performance on visuospatial ability (p = 0.006 on ITT analysis and p = 0.008 on per-protocol) and executive function (p < 0.001 on ITT and per-protocol (Table 3 and supplemental Table 7). In the per-protocol analysis, there was also lower cognitive score on episodic memory in 60–80 years old on either treatment for 6 months (Supplemental Table 7).
Table 3Subgroup analyses (APOE4 status and age groups) for primary and secondary outcomes.
APOE4 status (carrier vs non carrier)
Treatment
p
APOE4 status
p
Treatment* APOE4 status
p
Outcome
Visuospatial ability
0.03 (-0.18–0.25)
0.755
0.20 (-0.15–0.55)
0.269
-0.38 (-0.85–0.09)
0.113
Working memory
0.10 (-0.14–0.35)
0.408
-0.28 (-0.70–0.13)
0.180
0.20 (-0.34–0.74)
0.471
Episodic memory
0.09 (-0.10–0.27)
0.356
0.33 (0.02–0.64)
0.037
-0.23 (-0.64–0.18)
0.261
Executive function
0.02 (-0.20–0.24)
0.842
-0.02 (-0.40–0.36)
0.925
-0.06 (-0.55–0.43)
0.810
Age group (60–80 and 40–59.9 vs 20–39.9 years old)
Treatment
p
Age group
p
Treatment*age group
p
Visuospatial ability
0.04 (-0.29–0.38)
0.797
-0.44 (-0.76- −0.13)
0.006
-0.15 (-0.59–0.30)
0.517
Age group 40–59.9
−0.11 (−0.44–0.23)
0.534
-0.12 (-0.59–0.35)
0.620
Working memory
0.21 (−0.18–0.61)
0.287
-0.29 (-0.67–0.08)
0.127
−0.08 (−0.61–0.44)
0.754
Age group 40–59.9
-0.04 (-0.44–0.36)
0.840
-0.23 (-0.79–0.33)
0.423
Episodic memory
0.14 (-0.16–0.44)
0.363
-0.20 (-0.50–0.09)
0.175
-0.03 (-0.43–0.37)
0.875
Age group 40–59.9
0.10 (-0.21–0.41)
0.534
-0.27 (-0.70–0.15)
0.204
Executive function
0.02 (-0.32–0.35)
0.918
-0.66 (-0.99- -0.31)
<0.001
0.13 (-0.31–0.58)
0.557
Age group 40–59.9
-0.10 (-0.44–0.25)
0.588
-0.25 (-0.73–0.22)
0.299
Values are β (95% CI).
Positive values indicate an improvement in cognitive score at post intervention.
Negative values indicate decrease in cognitive score at post intervention.
ANCOVA adjusted for baseline standardized score and adding status (APOE4 carriage or age groups) with allocation group as an interaction term. Visuospatial ability: Cued Letter Discrimination incongruent task; Working memory: Brown-Peterson score.
Episodic memory: RI-48 delayed recall score; Executive function: Trail-Making Test, part B.
Note: missing values were found for visuospatial ability (n = 5 in n-3 PUFA group and n = 1 in placebo group) and for executive function (n = 1 in placebo group).
In the per-protocol analysis, there was a significant interaction for episodic memory in those with the lowest scores at baseline and their improvement when supplemented with n-3 PUFA (Supplemental Table 6, p = 0.043).
Baseline phospholipid concentrations of n-3 PUFA was not an effect modifier of treatment when using the ITT (Supplemental Table 8) or the per-protocol analysis (Supplemental Table 9).
4. Discussion
The present study evaluated whether a 6-month supplementation with n-3 PUFA improve cognitive performance in healthy adults aged from 20 to 80 years. Our study showed that those randomly allocated to receive n-3 PUFA supplementation daily did not significantly improved their performances in visuospatial ability, working memory, episodic memory, and executive function, compared to those allocated to placebo. Furthermore, age and APOE4 status were not significant effect modifiers of treatment.
Our results are in line with recent meta-analysis supporting that n-3 PUFA supplementation have no or very little effect on new neurocognitive illness, new cognitive impairment or global cognition assessed with the mini mental state exam [
An 18-mo randomized, double-blind, placebo-controlled trial of DHA-rich fish oil to prevent age-related cognitive decline in cognitively normal older adults.
Effect of 2-y n-3 long-chain polyunsaturated fatty acid supplementation on cognitive function in older people: a randomized, double-blind, controlled trial.
]. conducted a study on 302 cognitively healthy participants aged 65 years or older, with a maximum dose of 1.8 g EPA + DHA per day during 6 months and they reported no overall cognitive benefits in those consuming the n-3 PUFA supplement compared to placebo [
]. Similar results were obtained by Dangour et al. evaluating a 2-y intervention with 700 mg/d n-3 PUFA compared to an olive oil placebo in 867 cognitively healthy adults [
Effect of 2-y n-3 long-chain polyunsaturated fatty acid supplementation on cognitive function in older people: a randomized, double-blind, controlled trial.
The absence of overall cognitive benefits repeatedly found in RCTs may be because participants were overall cognitively healthy at baseline hence, room for improvement was limited. Our exploratory analyses support this hypothesis since those with low scores on episodic memory improved their scores when supplemented with n-3 PUFA. Another study also suggested that future clinical trials with n-3 PUFA supplementation should perhaps focus on participants with low-normal cognitive scores [
Effects of a 3-year multi-domain intervention with or without Omega-3 supplementation on cognitive functions in older subjects with increased CAIDE dementia scores.
] although our results does not support the latter. Another hypothesis for the lack of cognitive benefits with the consumption of n-3 PUFA is perhaps related to the type of cognitive tests used to evaluate cognition. In the current study, we used cognitive tests without ceiling effects according to age and tests that are not specifically designed for the diagnosis of cognitive decline rather than testing global cognition in specific cognitive domains. On the other hand, one recent meta-analysis reported a neutral effect on global cognition with the consumption of n-3 PUFA and they used the scores on the Mini Mental State Exam which is a questionnaire used for the diagnosis of dementia, with ceiling effects and that does not evaluate global cognition [
]. Therefore, cognitive tests used in meta-analysis should be considered more carefully and cautiously when concluding about a cognitive benefit or not with diagnostic tests which is considered not appropriate. Indeed, this test is good for screening participants before inclusion into an RCT with cognitively healthy participants, not to detect any cognitive benefits from a treatment.
Our team also investigated whether the APOE4 allele and age were significant effect modifiers of treatment since we previously suggested that plasma response to an n-3-PUFA supplementation might differ between APOE4 carriers and non-carriers [
]. However, in the present study, the response to the n-3 PUFA supplementation did not differed according to the APOE4 status. In the same way, there was no group by APOE genotype interactions on cognitive scores either. This result is not in line with the study of van de Rest et al., [
], where they reported cognitive benefit only in APOE4 carriers supplemented with n-3 PUFA compared to non-carriers. The main difference between the two studies is the participants age: 20–80 years in this study vs. > 65 years in that of van de Rest et al [
].. It is also important to emphasize that the number of APOE4 carriers in our study was low (< 25 in each group) and this result should hence be taken with caution. However, we found significant improvement in episodic memory after 6 months in APOE4 carriers compared to non-carriers and this improvement was independent of treatment. This result was surprising but one hypothesis explaining this is related to the idea of a partial metabolic shift towards overall increased fatty acid beta-oxidation in APOE4 carriers as we reported [
]. This metabolic shift would hence partially support some of the brain energy requirements, but this needs to be further validated especially since our results are based on a small sample size.
], we found that the plasma response to n-3 PUFA supplementation varied with age, with higher concentrations in those aged 60 years or older compared to those <40 years. Thus, we would expect that n-3 PUFA supplementation would lead to higher improvement of cognitive scores in older adults compared to younger ones. However, it did not translate in better cognitive scores in older adults. Instead, we found that adults >60 years old had lower performance in visuospatial ability and executive functions after 6 months, regardless of treatment groups, which illustrates age-related decline in these specific cognitive domains despite higher plasma phospholipid n-3 PUFA concentrations.
This study has strengths and limitations. The adherence was high (≥93%), suggesting that n-3 PUFA supplementation might be well accepted across all age groups. Also, n-3 PUFA absolute concentrations were quantified in plasma total phospholipids, which provides a better picture of blood EPA and DHA than relative percentages [
]. Because of our selection criteria, participants were in relatively good mental and physical health and these inclusion criteria could have reduced the likelihood to improve cognition. We also had a potential attrition bias in APOE4 carriers in the placebo group (n = 16) compared to the n-3 PUFA group (n = 26), which may have decreased statistical power. Another limitation is the age range of our study population.
5. Conclusion
This trial reported that a 6-month n-3 PUFA supplementation in cognitively and overall healthy adults aged from 20 to 80 years did not improve cognitive performance. Age and carrying the APOE4 genotype were not effect modifiers of treatment. Future RCT with n-3 PUFA might consider focusing on participants with baseline low-normal cognitive scores.
Authors’ Contributions
PL recruited the participants and was highly involved in the day-to-day trial; AV, PL and AH performed the lipids analysis at the Research Center on Aging. NP, MM and MP planned the statistical analysis. MM performed the statistical analysis. MM, PL, DL, NP and MP interpreted the data. MM, PL, DL, NP and MP drafted the manuscript. DL and IV were responsible for the cognitive tests. All authors have read and approved this manuscript
Funding source
This work was supported by the Canadian Institutes of Health Research between 2012 and 2017 (Grant number: MOP119454).
Disclosure statement
MP has received funding from Neptune Wellness Solutions for conducting clinical trials unrelated to the current project. No other financial disclosures are reported. All the remaining authors do not have any disclosures.
Clinical Trial Registration: This study was registered at clinicaltrials.gov as NCT01625195 https://clinicaltrials.gov/ct2/show/NCT01625195
Acknowledgments
Mathieu Maltais held a fellowship from the “Nutrition, Exercice and Lifestyle Team of the Canadian Consortium on Neurodegeneration and Aging”. Pauline Léveillé held a post-doctoral fellowship from the Fonds de la Recherche du Québec - Santé (FRQ-S); Anita Houeto held a scholarship from the Research Center on Aging. Nancy Presse is a Junior 1 Scholar of the FRQ-S. Mélanie Plourde holds a salary award from the Canadian Institutes of Health Research and FRQ-S and holds a Chair on lipid metabolism during aging funded by the Medical Research Center of Université de Sherbrooke. Nancy Presse and Mélanie Plourde are active members of the Canadian Consortium on Neurodegeneration and Aging.The authors thank Christine Brodeur-Dubreuil, Afiwa N'Bouke, and Audrey Perreault for their valuable assistance.
Higher plasma n-3 fatty acid status in the moderately healthy elderly in southern Québec: higher fish intake or aging-related change in n-3 fatty acid metabolism?.
Prostaglandins, Leukotrienes and Essential Fatty Acids (PLEFA).2010; 82: 277-280
An 18-mo randomized, double-blind, placebo-controlled trial of DHA-rich fish oil to prevent age-related cognitive decline in cognitively normal older adults.
No effect of n-3 long-chain polyunsaturated fatty acid (EPA and DHA) supplementation on depressed mood and cognitive function: a randomised controlled trial.
Association between cognitive function and supplementation with omega-3 PUFAs and other nutrients in >/= 75 years old patients: a randomized multicenter study.
Effects of supplementation with n-3 polyunsaturated fatty acids on cognitive performance and cardiometabolic risk markers in healthy 51 to 72 years old subjects: a randomized controlled cross-over study.
Interaction between BMI and APOE genotype is associated with changes in the plasma long-chain-PUFA response to a fish-oil supplement in healthy participants.
Chouinard-Watkins R., Lorrain D., Viens I., Normand M.P., Perron P., Rioux-Perreault C., Cunnane S., P. M., Performances cognitives avant et après une supplémentation en huile de poisson., in: none (Ed.) 10e journée scientifique du Réseau québécois de recherche sur le vieillissement, none, Montreal, Canada, 2012.
Effects of apolipoprotein E genotype on spatial attention, working memory, and their interaction in healthy, middle-aged adults: results From the National Institute of Mental Health's BIOCARD study.
Homocysteine status modifies the treatment effect of Omega-3 fatty acids on cognition in a randomized clinical trial in mild to moderate Alzheimer's disease: the OmegAD study.
Effect of 2-y n-3 long-chain polyunsaturated fatty acid supplementation on cognitive function in older people: a randomized, double-blind, controlled trial.
Effects of a 3-year multi-domain intervention with or without Omega-3 supplementation on cognitive functions in older subjects with increased CAIDE dementia scores.
00027-8&id=gr1.jpg){kind=link}