What’s the Relationship Between Flavonoids and Cognition?

Written by Zachary Wenger

December 25, 2021

Yeh, Tian-Shin et al. “Long-term Dietary Flavonoid Intake and Subjective Cognitive Decline in US Men and Women.” Neurology vol. 97,10 (2021): e1041-e1056. doi:10.1212/WNL.0000000000012454

Introduction:

Subjective Cognitive Decline (SCD) is the self-perceived encounter of worsening or more frequent confusion or memory loss (1). In the US, SCD among adults ≥65 years of age is 11.7%, relative to 10.8% among adults 45-64 years of age (2). In addition, it is one of the earliest discernible symptoms of Alzheimer’s disease and related dementias (2).

Consequently, there has been a lot of interest in the role of delayment in cognitive decline. Specifically, there has been substantial interest with flavonoids, which are a class of polyphenols, which are compounds found abundantly in fruit, vegetables, tea, fruit juice, and cocoa. Mechanistically, flavonoids appear to elicit beneficial changes in cerebrovascular responses following consumption.

Specifically, one of the proposed mechanisms of action relates to changes in Brain-Derived Neurotrophic Factor (BDNF), as a function of polyphenol consumption. In a recent meta-analysis of randomized controlled trials (RCTs), acute (7 days) and “chronic” exposure (6 weeks) of polyphenols significantly increased BDNF, which is a protein that supports learning, memory functions, and declines often begin at 65 years old (3-4). The increase in BDNF as a function of the polyphenols, was accompanied by improvements in mental fatigue, simple reaction time, and the serial sevens subtraction task, which is a test used to assess cognitive function (3).

However, due to the nature of most RCTs, “chronic” exposure wasn’t particularly long, as was illustrated in the meta-analysis of 6 weeks. How these polyphenolic compounds impact cognition over the span of 6 weeks is one question, though how these compounds impact cognition over decades is an entirely different question, which is what we will be examining here. The goal of this study was to prospectively examine the associations between long-term dietary flavonoids and SCD (5).

The Study:

This prospective cohort study included 49,493 women from the Nurses’ Health Study (NHS) and 27,842 men from the Health Professionals Follow-Up Study (HPFS). Dietary intake by semiquantitative food frequency questionnaires (SFFQs) were assessed seven times in the NHS (follow-up = 22 years) and five times in the HPFS (follow-up = 16 years). 

The mean age of participants at the initial SCD assessment was 76.3 years for the NHS and 73 years for the HPFS. Additionally, the study participants evaluated their own perceived cognition twice during the duration of the study, using SCD assessments. The SCD scores for the HPFS and NHS were based on 6-7 yes/no questions regarding changes in memory, executive function, attention, and visuospatial skills.

Within the HPFS and NHS, top food contributors of total flavonoid consumption consisted of tea, apples, pears, oranges, orange juice, blueberries, and strawberries. The mean intake of total flavonoids was 345 mg/d in both men and women. 

Results:

  • In the fully adjusted model, examining the highest (699 mg/d) and lowest (143 mg/d) quintiles, significant inverse associations between total flavonoids (OR 0.81, 95% CI 0.74-0.88) and all the flavonoid subclasses with SCD were observed within the NHS cohort.
  • In the fully adjusted model, examining the highest (681 mg/d) and lowest quintiles (147 mg/d), significant inverse associations were observed between only the total flavonoids (OR 0.86, 95% CI 0.75-0.99), flavones (OR 0.68, 95% CI 0.58-0.79), and flavanones (OR 0.65, 95% CI 0.56-0.75) within the HPFS cohort.
  • In the fully adjusted model, the strongest associations were observed between the flavones and flavanones, which are a subclass of flavonoids. 

Analyzing:

Strengths: 

  1. The mean age of participants at the initial SCD assessment was 73-76.3 yrs old, which would be an appropriate age to assess cognition.
  2. The follow-up time of the publication was sufficient, given the characteristics of the study participants.
  3. The repeated dietary measures to minimize measurement errors. This is one of the aspects where this study stands out, as other cohorts examining this research question often administer simply one FFQ at baseline, without assessing diet multiple times throughout the follow-up.
  4. The large sample size of >77,000 study participants. This is the largest sample size to date with studies examining this research question. 

Limitations:

  1. The lack of an SCD assessment at baseline. If utilized, there would be the option to interpret change scores (baseline score vs follow-up score ~20 yrs later).
  2. The lack of temporality between administering the two SCD assessments. The two SCD assessments were administered closely in time. As stated in the previous limitation, spacing out the SCD assessments would allow us to examine the rate of change in cognition.
  3. The lack of an objective cognitive assessment. For example, in a previous prospective cohort investigating a similar question, three objective psychometric tests were utilized, Mini-Mental State Examination, Benton’s Visual Retention Test, and ‘Isaacs’ Set Test (6). 

Interpretation:

To date, this is the largest prospective cohort published on flavonoid intake and cognition in the literature. Not only did the researchers provide us with the odd ratios for the different flavonoid subclasses, they additionally provided us with the associations between the individual food components that made up total flavonoids, per 3 servings/wk. The researchers state: “blueberries, strawberries, apples, orange juice, grapefruit juice, bananas, onions, tea, peaches, cauliflower, brussels sprouts, lettuce, and potatoes were selected as independent predictors of subsequent SCD status.” In other words, these foods were found to be independently protective against subjective cognitive decline.

Additionally, these associations persisted with an inclusive multivariate adjustment model. The adjustment model included dietary covariates, such as sugar-sweetened beverages, sweets/desserts, whole grains, refined grains, and animal fat consumption. The adjustment model also incorporated non-dietary covariates, including smoking status, education, depression, physical activity, and income.

However, it’s important to mention that there are other studies examining this relationship that don’t find statistically significant improvements in cognition. Firstly, in The Rotterdam Study, there was no statistically significant association on Alzheimer’s disease incidence between the highest tertile (>32.7 mg/d) and the lowest tertile (<22.6 mg/d) of flavonoids (RR 1.03, 95% CI 0.68-1.55) (7). However, in current smokers, there was a massive reduction in Alzheimer’s disease risk, when comparing the same high and low tertile (RR 0.54, 95% CI 0.31-0.96). In the Honolulu-Asia Aging Study, with an enormous 30.2 years of follow-up, there was still no statistically significant association within the highest quartile (8.2 mg/d) and the lowest quartile (2.0 mg/d) of flavonoids on the incidence of dementia (RR 1.36, 95% CI 0.95-1.96), Alzheimer’s disease (RR 1.56, 95% CI 0.92-2.67), and a collection of other outcomes (8). 

Why may there be discordance between the findings? These discrepancies can likely be explained by the drastic differences in flavonoid intake amounts between the different cohorts. Notice that the mean flavonoid intake amounts vary significantly between The Rotterdam Study (mean 28.5 mg/d), Honolulu-Asia Aging Study (mean 4.1 mg/d), and the current study (mean 345 mg/d). Evidently, these mean values are in completely different ballparks. Moreover, notice the lack of contrast between the differing “low vs high” tertiles and quartiles of flavonoid consumption. 2.0 mg/d vs 8.2 mg/d in the Honolulu-Asia Aging Study and <22.6 mg/d vs >32.7 mg/d in The Rotterdam Study, compared to 143-147 mg/d vs 681-699 mg/d in the current study. This is likely an artifact of a limited amount of flavonoid-containing foods included in the food frequency questionnaires (FFQ) and variation within intakes of flavonoids in different populations.

Fortunately, there are other cohorts that have considerably higher amounts of flavonoid consumption that we can examine. Within the SU.VI.MAX (Supplémentation en Vitamines et Minéraux Antioxydants) study, mean flavonoid intake was 571 mg/d for men and 499 mg/d for women, which is considerably higher than The Rotterdam Study and the Honolulu-Asia Aging Study (9). Consequently, total flavonoids were beneficially associated with language and verbal memory (P = 0.03). However, there was individual variation between the different types of flavonoids, though total flavonoids were still positively associated with improvements in long-term language and verbal memory capacities.

Furthermore, in the Framingham Heart Study Offspring Cohort, by investigating high (421.1 mg/d) vs low intakes (62.1 mg/d) of total flavonoids, while they didn’t find a statistically significant association in developing Alzheimer disease and related dementias (ADRD), within the flavonoid subclasses, flavonols, anthocyanins, and flavonoid polymers, were associated with statistically significant lower risks in developing ADRD (10). Interestingly, within their sensitivity analysis, when extending the cutoffs for ADRD from 5 years to 10 years, there was a massive observed 42% risk reduction, when comparing the highest vs lowest intake range of total flavonoids, which still didn’t quite reach statistical significance, though is overwhelmingly clinically relevant (HR 0.58 95% CI 0.32-1.08). Additionally, the researchers found even stronger associations between intakes of total flavonoids and the flavonoid subclasses, and risk of ADRD amongst those diagnosed with ADRD at ≥80 y compared to those diagnosed at <80 y. 

Lastly, within a subset of the NHS cohort, greater consumption of dietary flavonoids (684.1 mg/d) were associated with significantly slower rates of cognitive decline in women aged ≥70 years old, relative to lower consumption (145.4 mg/d) (11). The researchers discovered that women in the highest tertile of berries (≥1 serving/wk) and total flavonoids appeared to have delayed cognitive aging by as much as 2.5 years. 

Take Home Message:

When examining large enough ranges and intakes of flavonoid consumption, cognitive improvements in older populations is a consistent finding. In the current study, under the total flavonoid umbrella, flavones, flavanones, and anthocyanins had the strongest protective associations with SCD. This would translate into a dietary pattern including tea, as well as a wide array of colorful fruits and vegetables. The findings have considerable public health implications, as cognitive impairment resulting from aging is a detriment to families across the globe. While there are a collection of modifiable habits for delaying cognitive decline, including regular physical activity and not smoking, higher consumption of flavonoids is certainly an additional consideration that appears to play critical roles in protecting our brains as we age.

References:

  1. Jessen F, Amarigliod RE, et.al, Subjective Cognitive Decline Initiative (SCD-I) Working Group, A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease, Alzheimers Dement. 2014 November; 10(6): 844–852.
  2. Subjective Cognitive Decline — A Public Health Issue. https://www.cdc.gov/aging/data/subjective-cognitive-decline-brief.html.
  3. Ammar, Achraf et al. “Effects of Polyphenol-Rich Interventions on Cognition and Brain Health in Healthy Young and Middle-Aged Adults: Systematic Review and Meta-Analysis.” Journal of clinical medicine vol. 9,5 1598. 25 May. 2020, doi:10.3390/jcm9051598
  4. Neshatdoust, Sara et al. “High-flavonoid intake induces cognitive improvements linked to changes in serum brain-derived neurotrophic factor: Two randomised, controlled trials.” Nutrition and healthy aging vol. 4,1 81-93. 27 Oct. 2016, doi:10.3233/NHA-1615
  5. Yeh, Tian-Shin et al. “Long-term Dietary Flavonoid Intake and Subjective Cognitive Decline in US Men and Women.” Neurology vol. 97,10 (2021): e1041-e1056. doi:10.1212/WNL.0000000000012454
  6. Letenneur, L et al. “Flavonoid intake and cognitive decline over a 10-year period.” American journal of epidemiology vol. 165,12 (2007): 1364-71. doi:10.1093/aje/kwm036
  7. Engelhart, Marianne J et al. “Dietary intake of antioxidants and risk of Alzheimer disease.” JAMA vol. 287,24 (2002): 3223-9. doi:10.1001/jama.287.24.3223
  8. Laurin, Danielle et al. “Midlife dietary intake of antioxidants and risk of late-life incident dementia: the Honolulu-Asia Aging Study.” American journal of epidemiology vol. 159,10 (2004): 959-67. doi:10.1093/aje/kwh124
  9. Kesse-Guyot, Emmanuelle et al. “Total and specific polyphenol intakes in midlife are associated with cognitive function measured 13 years later.” The Journal of nutrition vol. 142,1 (2012): 76-83. doi:10.3945/jn.111.144428
  10. Shishtar, Esra et al. “Long-term dietary flavonoid intake and risk of Alzheimer disease and related dementias in the Framingham Offspring Cohort.” The American journal of clinical nutrition vol. 112,2 (2020): 343-353. doi:10.1093/ajcn/nqaa079
  11. Devore, Elizabeth E et al. “Dietary intakes of berries and flavonoids in relation to cognitive decline.” Annals of neurology vol. 72,1 (2012): 135-43. doi:10.1002/ana.23594