Nutrition has become a staple of preventive medicine, so researchers are delving deeper into components of foods or supplements that might optimize different aspects of health^[1]. These so-called ‘bioactives’ are garnering more interest as life expectancy^[2] and the associated burden of age-related global disease increases^[3] alongside a lack of effective pharmacological treatments^[4].

Various plant bioactives, such as carotenoids, alkaloids, and polyphenols, appear to improve health markers^[1] or reduce risk of disease. Polyphenol compounds span a broad range of bioactives, such as flavonoids, phenolic acids, and other compounds that can be found in fruits, vegetables, spices, coffee, and wine. These compounds exhibit a broad range of benefits^[5], from accelerated exercise recovery^[6] to anticancer effects^[7].

Recently, the neuroprotective effects^[8] of polyphenols have been the focus of researchers attempting to prevent or even treat age-related neurodegenerative disease. A systematic review and meta-analysis^[9] covered in Study Deep Dives #67 reported that polyphenol-rich supplementation may improve some cognitive and brain functions in older (older than 55 years) adults. However, as a report from the World Health Organization suggests, the later in life the intervention begins, “the more costly and less effective the solutions are likely to be.”

Luckily, some studies have shown potential benefits^[10][11] of polyphenol-rich supplementation on cognition and brain-related health measures in healthy young and middle-aged adults. However, as is generally the case, some results have been conflicting^[12][13]. The various study designs, polyphenols used, doses, and durations of intervention, make the existing literature hard to analyze and interpret. Research is also further complicated by the lack of any diseases associated with polyphenol deficiency. Vitamin C is necessary to prevent scurvy, but it’s possible to live without resveratrol.

The authors of the study under review conducted a systematic review and meta-analysis to try and resolve the uncertainty behind polyphenol intake and its influence on cognitive and brain-related health measures.

Polyphenols are prominent plant bioactives that appear to have various health benefits, such as neuroprotection, but studies demonstrate mixed results and use different polyphenols, doses and durations of intervention. In addition, relatively few studies focus on people who would benefit most from their putative effects: young and middle-aged adults. This led researchers to conduct a systematic review and meta-analysis in an attempt to investigate the relationship between polyphenol intake and cognitive function in this age group.

This systematic review and random-effect meta-analysis identified 16 RCTs that evaluated the effects of acute and/or chronic polyphenol-rich supplementation on cognitive function and/or other markers of brain health (e.g., neuroplasticity, neuroimaging, etc.) in healthy adults younger than 55 years old.

Studies were selected according to the PICOS^[14] criteria by two independently working investigators. PICOS is an attempt to simplify the identification of key study information. It stands for: Population, Intervention, Control/Comparison, Outcome, Study type. Studies were excluded if they were not written in English or recruited participants with a disease. The characteristics of the studies that were included are summarized in Figure 1.

Figure 1: Summary of the included studies

The reported primary outcome was the change in a broad range of cognitive functions, like memory and attention. Secondary outcomes included other measures associated with cognitive function and brain health, like brain-derived neurotrophic factor (BDNF) levels, blood flow and oxygenation, and more.

Only simple reaction time, rapid visual information processing, 7s serial subtraction, mental fatigue, and BDNF levels were included in the random-effects meta-analysis because of the high level of variability in measurements and techniques used. Outcomes were standardized to one unit of measure to calculate the effect size (difference in measured value from baseline to endpoint of intervention) according to Cohen’s method (a type of standardized mean difference). One way to read these means is^[19]: trivial, <0.2; small, 0.2–0.6; moderate, 0.6–1.2; large 1.2–2.0; very large, >2.0; extremely large, >4.0). Studies with at least two different doses (low vs. high) and/or durations (acute vs chronic) were treated as separate studies. Sensitivity analysis and cumulative meta-analysis were also conducted to assess reliability and stability of results.

The Physiotherapy Evidence Database (PEDro) scale was used to assess for methodological quality of the included studies. The meta-analysis as a whole was conducted in accordance with PRISMA guidelines and it was not preregistered.

This systematic review and random-effects meta-analysis evaluated the effects of acute and/or chronic polyphenol-rich supplementation on cognitive function and/or other measures of brain health in healthy adults who were less than 55 years old. Studies included a wide range of supplements (consisting of different polyphenol types), doses, and durations.

Of the 13 studies assessing effects of polyphenol-rich supplementation on cognitive function in healthy adults, eight demonstrated significant improvements in at least one measure of cognitive function and were skewed toward an effect for smaller (about 250 milligrams) and acute compared to larger (about 900 milligrams) and chronic intake. Double the amount of studies investigated short-term intake in comparison to long-term intake.

Of the ten studies evaluating the effects of polyphenol-rich supplementation on brain parameters in adults, six demonstrated significant changes in markers of cerebral blood flow from acute supplementation, and three did not demonstrate changes in BDNF. One study demonstrated a significant improvement in cerebral blood flow from five days of cocoa supplementation, while 28 days of a six-fold higher dose showed no effect (when acute intervention did).

The random-effects meta-analysis demonstrated improvements in simple reaction time (moderate), 7s serial subtraction (moderate), mental fatigue (very large), and BDNF levels (very large), but not rapid visual information processing. Acute doses of 250 milligrams of polyphenols appear to be sufficient for improved 7s serial subtraction and mental fatigue, while higher acute dosages were needed to improve simple reaction time. More details are shown in Figure 2.

Significant and high levels of heterogeneity were found for all of the five outcomes included while publication bias was only found in the mental fatigue outcome. Methodological study quality was considered good to excellent according to the PEDro scale. Reliability and stability of results were confirmed with sensitivity analysis and cumulative meta-analysis.

Most studies demonstrated improvements in at least one measure of cognitive function tested and were skewed toward an effect for acute compared to chronic intake. Small acute doses (about 250 milligrams) appear to be sufficient to improve some indicators of cognitive function, but doses and types of polyphenols varied widely and most of the included studies investigated short-term effects, so general conclusions must be drawn carefully. The random-effects meta-analyses demonstrated moderate to very large effect sizes for four out of the five outcomes with enough data to be included (simple reaction time, 7 second serial subtraction, mental fatigue, and BDNF levels, but not rapid visual information processing), although each exhibited a significantly high level of heterogeneity.

Similar to the research reviewed in Study Deep Dives #67, which was conducted by the same authors, the large variety of supplements, doses, assessments, and biomarkers, along with small sample sizes and a lack of dietary control, make it difficult to draw any strong conclusions from the study at hand. There are several different kinds of polyphenols, as well as indicators of cognitive function, and some included studies didn’t even report the amount of polyphenols in the supplement being used. Moreover, most studies did not test the supplement themselves to ensure that the supplement label was accurate^[20]. Nor did they thoroughly consider the bioavailability of the polyphenols, which can range from 1–43%^[21] (thus impacting the extent of any potential effect) and depends on^[22] dosage, processing, interaction with other compounds, chemistry, and more. A greater level of standardization and consistency regarding measurements methods, supplements, and reporting is needed.

Double the amount of included studies were conducted with short-term polyphenol intervention, leaving the usefulness of longer-term intervention effects less clear. While the results from the study at hand suggest a larger effect from acute intervention, one would expect benefits to accumulate with a longer-term intervention. Then again, the underlying mechanism behind polyphenol intervention benefits is unclear—the effect might wear off after a while or there may be a threshold effect of some sort that has not been discovered.

Several pathways involving interactions with gut microbiota^[23], modulation of neuroinflammation^[8] and glucoregulation^[24], and increased neurogenesis have been proposed as potential mechanisms that appear to be linked to each other^[25] by the action of nitric oxide (NO). NO is mostly known^[26] for its role in vascular function and blood flow. The understanding of NO’s importance in the brain^[27] for neurotransmission and signal cascades is rapidly developing, yet still incomplete. To gain a better understanding of this relationship, future studies should monitor NO levels along with indicators of its activity, such as cerebral blood flow and cognitive testing.

Polyphenols have been suggested^[25] to act indirectly by stimulating the production of NO, and directly by crossing the blood brain barrier and modulating neuronal receptors and/or signaling and increasing neurotrophins, such as BDNF. Improvements in BDNF levels are associated with increased synaptic plasticity^[17], neurogenesis^[17], learning and memory^[28]. This aligns with the improvements in various measures of cerebral blood flow, BDNF levels, and measures of cognitive function in the study under review, but specifics (such as whether NO is acting as a neurotransmitter or vasodilator) are unclear, especially given that different polyphenols could have different effects, and only a few studies used the same treatment.

Overall, the most important message of the study relates to the potential of preventive action for better brain health outcomes during aging. While aging research is always rather controversial, it appears that younger adults might actually have the choice^[29] to extend/improve their healthspan, and polyphenol intake might be one way to subtly do that. Moreover, the study under review suggests that as little as about 250 milligrams of polyphenols in one acute dose can make a difference in cognition!

The large variety across supplements and their polyphenolic content, doses and durations, polyphenol bioavailability, and measures of cognitive function and brain health make it difficult to draw strong conclusions from the study at hand. While nitric oxide and BDNF might be involved, the underlying mechanism behind improvements from small acute doses of polyphenols is unclear. Overall, the study serves as moderately convincing evidence that increased polyphenol intake, even at small doses, can offer brain-protective effects in young and middle-aged adults for a potentially improved healthspan.

Q. Do young adults consume adequate levels of polyphenols?

A systematic review^[5] of 45 studies that investigated polyphenol intake and health outcomes in samples of adults across most of the world reported an average intake of 900 milligrams of polyphenols per day. While most studies were conducted in older adults, what is clear is that higher intake was associated with reduced risk of various health outcomes, including cardiovascular disease (as little as 360 milligrams per day), bone fracture, macular degeneration, cognition, and more. An analysis^[30] of data from the National Health And Nutrition Examination Survey (NHANES) from 2013–2016 demonstrated that a sample of about 10,000 adults in the U.S. consume about 900 milligrams of dietary polyphenols per every daily 1000 kcal. Higher intakes were associated with a college education, female sex, non-hispanic Caucasians, and adults aged 40 and older.

Q. What are good sources of polyphenols?

Many herbs and spices^[31] have high levels of polyphenolic content, such as cloves, peppermint, and oregano, but their use is sparse in most diets and their intake is often not tracked. The main sources^[5] of polyphenols in the diet are coffee, tea, red wine, fruits, and vegetables. See Figure 3 for some rough estimates of polyphenol content.

Figure 3: Rough polyphenol content of select food and drink

References:Phenol Explorer. Accessed August 11, 2020 | Pérez-Jiménez et al. Eur J Clin Nutr. 2010 Nov.^[31] | Álvarez et al. Food Technol Biotechnol. 2016 Dec.^[32] | Fukushima et al. J Nutr Sci. 2014 Oct.^[33]

Note that food polyphenol content will depend on^[34] the source, variety, environmental factors (e.g., location of cultivation, season, etc.), processing, storage and more. Also, researchers are still developing methods^[32] to determine the most physiologically accurate ways to measure polyphenolic content and bioactivity.

This systematic review and random-effects meta-analysis identified 16 RCTs that found mild improvements in various measures of cognitive function and brain health following small acute doses of polyphenol-rich supplements in healthy adults younger than 55 years old. The meta-analyses demonstrated moderate to very large effect sizes for some specific measures of cognitive function, although each exhibited a significantly high level of heterogeneity. The wide variety of polyphenol supplements, their respective bioavailability, intervention dosing regimens, and cognitive function measurements used in the meta-analyses limit the reliability of these findings.