Omega-3 Versus Placebo

Seven RCTs with low to medium risk of bias enrolling a total of 21,027 adults compared some form of omega-3 fatty acids versus placebo in adults.^{101, 103, 107, 115, 117, 119, 120} Total sample sizes ranged from 65 to 11,685. Yurko-Mauro et al. used only docosahexaenoic acid (DHA),¹¹⁹ all others used some combination of eicosapentaenoic acid (EPA) plus DHA. Geleijnse et al. also used alpha-linolenic acid (ALA) as another omega-3 study arm.¹⁰⁷ Only the ORIGIN study (n=15,077) allowed adults already using omega-3 supplementation to participate in the study.¹²⁰ All studies assessed baseline cognition; six reported baseline Mini-Mental State Examination (MMSE) score of at least 28^{103, 107, 115, 117, 119, 120} while one study used the Isaacs Set Test (35.8).¹⁰¹ However, only three studies specified a baseline cognition inclusion criterion.^{103, 115, 119} Populations studied included adults with diabetes or impaired glucose tolerance,¹²⁰ a history or ischemic heart disease,¹⁰¹ coronary patients,¹⁰⁷ or healthy adults.^{103, 115, 117, 119}

No study reported incident diagnosis of dementia or MCI as determined solely by clinical diagnosis. The ORIGIN study, a large multinational study of adults with diabetes or impaired glucose tolerance, used a combination of clinical diagnosis or an MMSE score less than 24 and found no difference in probable dementia incidence between EPA+DHA or placebo groups for the median duration of 6.2 years (HR 0.93 [0.86 to 1.0]).¹²⁰

Overall, the studies provide low-strength evidence suggesting that omega-3 fatty acids do not improve cognitive performance between adults with normal cognition as compared to placebo. None of four studies (n=16,431) found a statistical improvement in brief cognitive test performance, such as the MMSE;^{101, 107, 119, 120} likewise, one study that assessed multidomain neuropsychological performance using a global composite also found no statistical difference between groups.¹⁰³ Of 32 tests to assess executive function in five studies (n=5,079), 29 tests did not find a significant difference between groups, with a maximum followup of 6 years.^{103, 115, 117, 119, 120} The two tests with significant differences that favored the omega-3 fatty acid group were based on 548 participants and for only a 6 month followup.^{117, 119} Similarly, of 25 tests to assess memory in five studies (n=3,428),^{101, 103, 115, 117, 119} 22 did not find a significant difference between groups, with a maximum followup of 4 years. The three tests with the omega-3 fatty acid group performing better than the placebo group were from a single 6-month study that used six memory tests (n=483).¹¹⁹

No studies found significant differences in adverse events for omega-3 supplementation.

Four studies examined the effects of the omega-3 fatty acid interventions versus placebo on several subgoups. No significant differences in effect were found for age,^{101, 107, 115, 120} sex,^{107, 115, 120} or inclusion criteria disease condition.^{107, 120}

Andreeva et al. used a 2X2 factorial design, assigning adults with a history of ischemic heart disease to four groups: placebo, omega-3, B vitamins (folate, B₆, B₁₂), or omega-3 plus B vitamins.¹⁰¹ Results noted above collapsed the four arms into one group with any omega-3 assignment versus one group without omega-3 assignment. Results when comparing the omega-3 alone group with the B vitamins alone group also found no significant differences between groups for any outcome. Likewise, the omega-3 plus B vitamins versus B vitamins alone did not result in significant differences between groups.

Ginkgo Biloba Extract

Three randomized controlled trials (RCTs) (four publications) with low to medium risk of bias enrolling a total of 5,559 older adults with presumed normal cognition compared 240 mg/day of ginkgo biloba versus placebo in adults.^{104, 105, 113, 116} Total sample sizes ranged from 118 to 3,069. All studies assessed baseline cognition, two reporting baseline MMSE scores of at least 27.6^{105, 116} while one reported baseline Modified Mini-Mental State Examination (3MS) of 93 and Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) of 6.5.^{104, 113} All studies specified a baseline cognition inclusion criterion.^{103, 115, 119} Age inclusion criterion were ≥70,¹¹⁶ ≥75,^{104, 113} and ≥85.¹⁰⁵

Two studies provide low-strength evidence suggesting that ginkgo biloba does not affect incidence of probable CATD compared to placebo.^{104, 113, 116} Both studies assessed probable CATD according to Diagnostic Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) criteria by adjudication panels of clinical experts.

Overall the studies also provide low-strength evidence that ginkgo biloba does not improve cognitive performance as compared to placebo. One study that assessed multidomain neuropsychological performance using the 3MS and the ADAS-Cog found no statistical difference between groups.¹¹³ Likewise, no differences between groups were found in either executive function¹¹³ or memory.^{105, 113}

All studies reported adverse events. No studies found significant differences in adverse events for omega-3 supplementation. The two larger studies found no differences in adverse events between groups (n=5,437).^{104, 113, 116} Dodge et al., who recruited 122 adults 85 years and older with normal cognition, reported a larger number of strokes and transient ischemic attacks (TIA) in the gingko biloba group over 3.5 years (7 vs. 0, p=.01).¹⁰⁵ However, the larger study by Vellas et al. (n=2,820) found no significant differences between groups in stroke, hemorrhagic events, and cardiac disorders over 5 years.

Two studies explored the effects of the ginkgo biloba interventions versus placebo on several subgoups. Vellas et al. found differences in effect in men, people who consumed alcohol at baseline, and adults who continued the intervention for at least four years.¹¹⁶ The authors also advised caution in interpreting the results since they assessed 13 planned subgroups (including age, APOE-4, MMSE ≤27 at baseline, hypertension, diabetes, hypercholesterolemia, body mass index (BMI) ≥27, and failing leg balance test) and did not adjust for multiple testing (all 3 groups showing differences would have been nonsignificant with a Bonferroni correction).¹¹⁶ In contrast, the GEM study did not find significant effect modification for sex. They also did not find differences for age, sex, race, APOE-E4 status, education, or MCI at baseline. However, CVD at baseline did show a significant treatment by group interaction (p=.02).

Other Nutraceuticals

Three additional RCTs examined the effects of nutraceuticals on cognition. Resveratrol, a member of a group of plant compounds called polyphenols with possible antioxidant properties, was examined in one study. In this 6-month study on the use of resveratrol in 46 healthy overweight people aged 50–80 years, people assigned to resveratrol performed better on 2 of 6 memory tests and showed significant increases in functional connectivity of the hippocampus to frontal, parietal, and occipital areas of the brain when compared to placebo.¹¹⁸ No significant changes between groups in total gray matter volume or in the volume or microstructure of the hippocampus were noted.

Schiepers et al. (n=57) compared cognition in 57 adults assigned to consume margarines enriched with plant sterol or stanol esters with those using a control margarine and found after 85 weeks no differences between groups.¹¹¹

Strike et al. (n=27) examined a commercial supplement containing 1 g DHA, 160 mg EPA, 240 mg ginkgo biloba, 60 mg phosphatidylserine, 20 mg vitamin E, 1 mg folic acid, and 20 mcg vitamin B₁₂ per day versus placebo.¹²¹ The authors hypothesized the combination would provide a synergistic effect. After 6 months, the intervention group improved compared to the control group in one out of three executive function/attention/processing speed outcomes and one out of three memory tests.

No adverse effects were reported in any study. Due to the evidence base of single studies with small sample sizes (n<500), strength of evidence was not assessed for these three interventions.

Nutraceuticals Versus Inactive Control

Three RCTs compared nutraceuticals to inactive controls in older adults with MCI.^{104, 106, 108} Summaries of study results are detailed in Table 4C.3.

Table 4C.3

Results overview: Nutraceutical interventions in adults with MCI.

Lee et al. (n=36) examined the effects of daily omega-3 fatty acids (fish oil supplementation, DHA 430 mg and EPA 150 mg) on cognitive function in people aged 60 and older with MCI.¹⁰⁸ After 1 year, no significant change in MMSE scores was observed. However, people taking omega-3 performed better than those on placebo on one of three tests of executive function/attention/processing speed, and better on three of five memory tests. No serious adverse effects were reported. Evidence to draw conclusions was insufficient due to limited data (single study with n<500) or no data.

Two (2) studies compared the effects of ginkgo biloba to placebo in people with MCI.^{104, 106} Follow-up periods in the studies varied, with Gavrilova’s study lasting 6 months⁸¹ and median follow-up in DeKosky et al. lasting 6.1 years.⁷⁹

DeKosky et al. examined diagnostic outcomes.¹⁰⁴ Of five categories of dementia, no significant differences were found between ginkgo and placebo groups. Gavrilova et al. included two objective measures of cognition, both related to the executive function/attention/processing speed domain. In both tests, participants taking ginkgo performed significantly better than those taking placebo.¹⁰⁶

Gavrilova et al. reported no serious adverse effects.¹⁰⁶ DeKosky et al. found no significant differences between ginkgo and placebo groups in rates of serious adverse effects, including death, bleeding, coronary heart disease (CHD), and stroke.¹⁰⁴ Evidence to draw conclusions was insufficient due to limited data (single study with n<500) or no data.