Skip main navigation

Benefits in Cognitive Function, Blood Pressure, and Insulin Resistance Through Cocoa Flavanol Consumption in Elderly Subjects With Mild Cognitive Impairment

The Cocoa, Cognition, and Aging (CoCoA) Study
Originally published 2012;60:794–801


Flavanol consumption is favorably associated with cognitive function. We tested the hypothesis that dietary flavanols might improve cognitive function in subjects with mild cognitive impairment. We conducted a double-blind, parallel arm study in 90 elderly individuals with mild cognitive impairment randomized to consume once daily for 8 weeks a drink containing ≈990 mg (high flavanols), ≈520 mg (intermediate flavanols), or ≈45 mg (low flavanols) of cocoa flavanols per day. Cognitive function was assessed by Mini Mental State Examination, Trail Making Test A and B, and verbal fluency test. At the end of the follow-up period, Mini Mental State Examination was similar in the 3 treatment groups (P=0.13). The time required to complete Trail Making Test A and Trail Making Test B was significantly (P<0.05) lower in subjects assigned to high flavanols (38.10±10.94 and 104.10±28.73 seconds, respectively) and intermediate flavanols (40.20±11.35 and 115.97±28.35 seconds, respectively) in comparison with those assigned to low flavanols (52.60±17.97 and 139.23±43.02 seconds, respectively). Similarly, verbal fluency test score was significantly (P<0.05) better in subjects assigned to high flavanols in comparison with those assigned to low flavanols (27.50±6.75 versus 22.30±8.09 words per 60 seconds). Insulin resistance, blood pressure, and lipid peroxidation also decreased among subjects in the high-flavanol and intermediate-flavanol groups. Changes of insulin resistance explained ≈40% of composite z score variability through the study period (partial r2=0.4013; P<0.0001). To the best of our knowledge, this is the first dietary intervention study demonstrating that the regular consumption of cocoa flavanols might be effective in improving cognitive function in elderly subjects with mild cognitive impairment. This effect appears mediated in part by an improvement in insulin sensitivity.


Mild cognitive impairment (MCI) is a clinical state of individuals who are memory impaired but otherwise well functioning and without clinical demonstration of dementia.1,2 MCI represents an interesting field of research, because MCI increases the risk of later developing dementia, and, in this phase, an interventional therapy could have the greatest potential to improve cognitive performance and to slow down disease progression.3 A growing body of evidence suggests that specific dietary components may impact brain function through the regulation of neurotransmitter pathways, signal-transduction pathways, and synaptic transmission.4 Several dietary components, including omega-3 fatty acids, B vitamins, vitamins D and E, and choline, have been identified as having favorable effects on cognitive abilities.4

More recently, evidence suggests that the consumption of flavonoids, a diverse group of polyphenolic compounds widely present in plant-based foods, may be associated with a decreased risk of incident dementia5 and with a lower prevalence of cognitive impairment,6 a better cognitive evolution over a 10-year period,7 and better dose-dependent performance of several cognitive abilities in elderly subjects.8 Among the flavonoids, flavanols, a subclass abundant in tea, grapes, red wine, apples, and cocoa products, including chocolate, have been proposed to be highly effective in reversing age-related declines in neurocognitive performance by increasing the number of and strength of connections between neurons, reducing neuronal loss attributed to neurodegenerative processes, and through their ability to interact with the cellular and molecular architecture of the brain responsible for memory.9 These cellular effects of flavanols are proposed to be attributed to selective actions on different pathways, including binding to ATP sites on enzymes and modulation of proteins central to intracellular signaling cascades, such as the mitogen-activated protein kinase signaling pathway and the phosphoinositide 3-kinase/AKT signaling cascade, thus working both directly and indirectly by modulating transcription factor activation and binding to promoter sequences.9

Substantial evidence exists supporting the link between cognition and cardiovascular risk factors.10 In particular, high blood pressure has been linked with the pathophysiology of MCI other than with Alzheimer disease and vascular dementia.10,11 This relationship may be mediated by endothelial dysfunction and microvascular diseases, leading to an impairment of vascular reserve.10,11 Alterations in the insulin signaling pathway have also been suggested as potential contributors to cognitive dysfunction because of the pivotal role of this hormone in modulating brain structure and function.1214 Peripheral insulin resistance could promote cognitive dysfunction by reducing brain insulin uptake, raising brain levels of amyloid β and τ phosphorylation and promoting vascular damage through oxidative stress, proinflammatory cytokines, advanced glycation end products, and dyslipidemia,12,13 with chronic dysglycemia thought to contribute to cognitive dysfunction by promoting the development of cerebral microvascular disease and inflammation.15 Interestingly, it has been also proposed that the brain itself could become insulin resistant and that this promotes or even triggers key pathophysiological events leading to cognitive dysfunction.12,16 In this regard, dietary flavanol consumption has been shown to reduce blood pressure and improve insulin sensitivity,17,18 as well as to induce vasodilation of the peripheral17,18 and cerebral vascular system,19 increasing brain blood flow and perfusion.19,20 Taken together, these previously reported effects provide a strong foundation for the idea that the regular consumption of flavanols may have implications for cognitive function, particular among those at risk.

To the best of our knowledge, no dietary intervention study has yet specifically investigated the topic of neurocognitive performance in response to dietary flavanols in subjects with MCI. In the current report, we tested the hypothesis that the regular dietary inclusion of a beverage containing cocoa flavanols would be effective in improving cognitive performance in subjects with MCI. Furthermore, the impact of this daily dietary modification on blood pressure, glucose metabolism, and oxidative stress was also studied in this population because of their potential influential role on cognitive function.



Participants were recruited among those referred to the Alzheimer unit of the University of L'Aquila Geriatric Division for MCI, with MCI diagnosed according to the revised Petersen criteria.21 These criteria included a decline in memory, objectively verified by neuropsychological testing in combination with a precise history from the participant, proxy, or both, as suggested by Petersen,22 and adjusted for age and education, or a decline in other cognitive domains, with normal functional of activities. Among 184 subjects screened, 61 individuals did not meet eligibility criteria, and 33 individuals refused to participate. The remaining 90 individuals were enrolled in the study (Figure S1, available in the online-only Data Supplement). The study was approved by the local ethics committee, and written informed consent to participate was obtained by each participant. Please see the online-only Data Supplement for a detailed Methods section.

Study Design and Outcomes

To investigate the impact of regular cocoa flavanol consumption on cognitive function in individuals with MCI, an 8-week double-blind, randomized, parallel arm study was conducted. After a 1-week run-in period, participants were randomly assigned to consume once daily a dairy-based cocoa drink containing cocoa flavanols either at a high (HF; ≈990 mg of flavanols per serving), intermediate (IF; ≈520 mg of flavanols per serving), or low level (LF; ≈45 mg of flavanols per serving) for 8 weeks. The dairy-based cocoa drinks used in this study were specially designed so that they were indistinguishable in taste and appearance, calorically balanced, and contained similar macronutrient, mineral, theobromine, and caffeine content, varying significantly only in the content of cocoa flavanols (Table S1, available in the online-only Data Supplement). The HF and IF drink mixes were made with a flavanol-rich cocoa powder (Cocoapro processed cocoa powder; Mars Inc), whereas the LF drink was made with a highly processed, alkalized cocoa powder.

Main outcome measures examined were changes in cognitive function after 8 weeks of regular cocoa flavanol consumption. Secondary outcome measures examined included changes in blood pressure, metabolic parameters, and plasma markers of lipid peroxidation.

Cognitive Function Assessment

Cognitive testing was performed at baseline and after 8 weeks (±2 days) using a combination of 4 well-validated standardized tests: the Mini Mental State Examination, Trail Making Test (TMT) A, TMT B, and a verbal fluency test. As a predefined procedure, an integrated measure of overall cognitive function, composite cognitive z score, was also constructed for each participant by converting the log-transformed raw scores from the individual tests to standardized scores (z score) that were based on the means and pooled SDs of the whole population at baseline.

Statistical Analysis

The statistical analysis was conducted according to the intention-to-treat principle. Accurate prestudy sample-size calculation was hindered by the lack of available data on the effect of cocoa flavanol administration on cognitive function in subjects with MCI. Thus, the study was based on an estimated sample size of 90 subjects, with a ratio of 1:1:1 for the 3 treatment groups, which has been calculated to be adequate to achieve 90% power to detect a moderate effect size (Cohen d=0.40) with 2 degrees of freedom and an α of 0.05 on the global cognitive z score between HF and LF. SAS version 9.1 was used to perform a 2-way mixed-design ANOVA using the repeated statement with the general linear model procedure. Post hoc analysis between treatment groups was performed by Tukey studentized range − honestly significant difference. χ2 was used to compare categorical variables. Spearman nonparametric correlation was used to evaluate correlations between variables. Change of composite cognitive z score was entered as the dependent variable in a hierarchical multiple regression analysis in which homeostasis model assessment–insulin resistance (HOMA-IR), isoprostanes, and blood pressure were entered sequentially. Analysis was performed on variables logarithmically transformed to enhance symmetry of measures. If it is not otherwise specified, data are presented as mean±SD.


Study Population

General characteristics of the study population are shown in Table S2. According to the selection criteria, none of participants was obese (body mass index >30 kg/m2), a current smoker, or under statin treatment. Hypertension was the most prevalent cardiovascular risk factor. The 3 study groups were comparable with regard to sex, age, anthropometric characteristics, cardiovascular risk factor prevalence, and pharmacological treatments (Table S2). Dietary total flavanol intake was also similar in the 3 study groups at enrollment and slightly decreased during the run-in period to a similar degree in all of the groups, a consequence of the dietary prescriptions (Table S3).

Adherence to the Study Protocol

Two participants assigned to IF discontinued beverage consumption after 4 weeks because of personal reasons, whereas 1 participant assigned to LF treatment discontinued beverage consumption after 2 weeks because of reported gastric discomfort. All 3 of these participants were followed up during the entire study period, and the data were included in the database for statistical analysis according to an intention-to-treat procedure. The overall compliance was 99.6% at week 4 and 99.4% at week 8, without difference among the LF, IF, and HF interventions. Adherence to the study protocol was evaluated at each visit by a checklist questionnaire of specified food items and by monitoring body weight. Compliance with the dietary restrictions was good, because none of the subjects reported regular consumption of the restricted flavanol/procyanidin-containing foods and beverages. Furthermore, caloric intake was controlled throughout the study, because no significant differences in body weight (HF, 70.67±8.39 kg; IF, 68.59±8.19 kg; LF, 71.17±8.65 kg; P=0.486) or body mass index (HF, 27.20±2.7 kg/m2; IF, 26.27±2.50 kg/m2; LF, 26.90±2.75 kg/m2; P=0.427) among the 3 study groups were observed at the end of follow-up.

Cognitive Function Measures

Baseline performance on the cognitive function tests was similar for the 3 treatment groups, indicating an adequate randomization procedure (Table 1). Mini Mental State Examination score did not significantly change in relation to the 3 different treatments during the study period (P=0.13; Table 1).

Table 1. Changes in Neuropsychological Test Score During the Study Period in the 3 Treatment Groups

Neuropsychological TestsHigh CFIntermediate CFLow CFANOVATukey HSD
MMSETime×treatment interaction: F=2.10; P=0.13
    Week 027.43±1.2827.17±2.0927.60±1.45F=0.53NS
    Week 828.07±1.0127.43±1.9427.93±1.20F=1.61NS
TMT A, sTime×treatment interaction: F=27.62; P<0.0001
    Week 052.37±14.8749.00±14.3551.50±18.16F=0.36NS
    Week 838.10±10.9440.20±11.3552.60±17.97F=9.67<0.05*
TMT B, sTime×treatment interaction: F=78.19; P<0.0001
    Week 0133.33±36.52138.80±33.41135.40±44.04F=0.16NS
    Week 8104.10±28.73115.97±28.35139.23±43.02F=8.26<0.05*
VFT, words per 60 sTime×treatment interaction: F=22.79; P<0.0001
    Week 019.53±6.0719.04±5.4221.01±8.06F=0.69NS
    Week 827.50±6.7524.20±6.1022.30±8.09F=4.20<0.05
Z scoreTime×treatment interaction: F=62.13; P<0.0001
    Week 0−0.007±0.523−0.064±0.7040.072±0.912F=0.26NS
    Week 80.687±0.4820.340±0.6630.000±0.803F=8.07<0.05

CF indicates cocoa flavanol; MMSE, Mini Mental State Examination; TMT, trail making test; VFT, verbal fluency test; NS, not significant. P<0.05 is considered a statistically significant difference.

*Data are low vs high and intermediate CF.

Data are low vs high CF.

The time required to complete TMT A significantly changed throughout the study period (P<0.0001), with significant reductions observed in subjects assigned to HF (−14.3±4.2 seconds; P<0.0001) and IF (−8.8±3.4 seconds; P<0.0001) interventions but not in those assigned to the LF intervention (+1.1±13.0 seconds; P=0.65; Table 1). Similarly, the time required to complete TMT B significantly changed during the study period (P<0.0001), with significant reductions observed among HF (−29.2±8.0 seconds; P<0.0001) and IF (−22.8±5.1 second; P<0.0001) subjects but not in those assigned to the LF intervention (+3.8±16.3 seconds; P=0.21; Table 1). Thus, TMT A and TMT B scores at the end of follow-up were significantly (P<0.05) better in subjects assigned to HF and IF treatments in comparison with those assigned to the LF group (Table 1).

Verbal fluency test scores significantly improved (P<0.0001) in the study, again with improvements demonstrated in the HF (+8.0±5.3 words per 60 seconds; P<0.0001) and IF (+5.1±3.1 words per 60 seconds; P<0.0001) groups and, to a lesser extent, in those assigned to the LF group (+1.2±2.7 words per 60 seconds; P=0.014). The improvement of verbal fluency test score was significantly (P<0.05) greater in HF subjects in comparison with those assigned to the LF group (Table 1).

According to the above results, the composite cognitive z score significantly changed during the study period (P<0.0001), with HF (+0.693±0.223; P<0.0001) and IF (+0.404±0.141; P<0.0001) groups demonstrating significant improvement; no change was observed in the LF group (−0.072±0.383; P=0.31; Table 1 and Figure S2). Composite cognitive z score at the end of follow-up was significantly (P<0.05) better in subjects assigned to HF in comparison with those assigned to LF (Table 1 and Figure S2).

Blood Pressure and Metabolic Parameters

Baseline blood pressure levels and metabolic parameters were similar for the 3 treatment groups (Table 2).

Table 2. Changes of Blood Pressure and Metabolic Parameters During the Study Period in the 3 Treatment Groups

VariablesHigh CFIntermediate CFLow CFANOVA
SBP, mm HgTime×treatment interaction: P<0.0001
    Week 0141.1±9.9142.4±9.8141.4±8.6P=0.85
    Week 8131.0±9.2134.2±8.5140.0±10.9P=0.0018*
DBP, mm HgTime×treatment interaction: P<0.0001
    Week 084.5±5.686.4±7.386.0±6.4P=0.51
    Week 879.7±5.383.0±6.985.1±7.1P=0.007*
Glucose, mmol/LTime×treatment interaction: P<0.0001
    Week 05.45±0.605.41±0.755.45±0.80P=0.98
    Week 84.81±0.484.95±0.685.33±0.74P=0.0065*
Insulin, mU/LTime×treatment interaction: P<0.0001
    Week 015.63±8.2812.52±5.6613.80±9.22P=0.31
    Week 810.54±6.789.65±5.3713.67±9.12P=0.08
HOMA-IRTime×treatment interaction: P<0.0001
    Week 03.90±2.373.07±1.553.44±2.49P=0.35
    Week 82.33±1.612.19±1.333.28±2.35P=0.023
TC, mmol/LTime×treatment interaction: P<0.0001
    Week 05.31±1.055.37±0.995.12±0.89P=0.60
    Week 84.85±0.984.94±0.955.06±0.86P=0.70
LDL-C, mmol/LTime×treatment interaction: P<0.0001
    Week 03.38±0.803.25±0.813.22±0.74P=0.70
    Week 82.92±0.742.85±0.743.16±0.71P=0.25
HDL-C, mmol/LTime×treatment interaction: P=0.29
    Week 01.25±0.381.41±0.451.25±0.35P=0.20
    Week 81.24±0.381.43±0.461.25±0.36P=0.14
TG, mmol/LTime×treatment interaction: P<0.0001
    Week 01.53±0.321.55±0.331.42±0.37P=0.32
    Week 81.49±0.321.45±0.311.42±0.38P=0.74
8-iso-PGF2α, pg/LTime×treatment interaction: P<0.0001
    Week 0377.3±112.8402.5±125.01412.2±147.5P=0.56
    Week 8277.4±112.1337.3±114.9408.7±140.6P=0.0004

CF indicates cocoa flavanol; SBP, systolic blood pressure; DBP, diastolic blood pressure; HOMA-IR, homeostasis model assessment of insulin resistance; TC, total cholesterol: LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglycerides; 8-iso-PGF2α, 8-iso-prostaglandin F2α. Plus-minus values are mean±SD. Differences within groups were analyzed by ANOVA. Differences between groups were analyzed by ANOVA followed by Tukey HSD.

*P<0.05 low vs high and intermediate CF.

P<0.05 low vs high CF.

Blood Pressure

Systolic and diastolic blood pressures significantly (P<0.0001) changed through the study period with significant pressure reductions among HF (systolic: −10.0±3.1 mm Hg, P<0.0001; diastolic: −4.8±1.8 mm Hg, P<0.0001) and IF (systolic: −8.2±3.5 mm Hg, P<0.0001; diastolic: −3.4±2.0 mm Hg, P<0.0001) subjects but not in those assigned to the LF group (systolic: −1.4±5.4 mm Hg, P=0.16; diastolic: −0.9±3.4 mm Hg, P=0.14; Table 2). Thus, systolic and diastolic blood pressure levels at the end of follow-up were significantly (P<0.05) lower in subjects assigned to HF and IF in comparison with those assigned to LF (Table 2).

Metabolic Parameters

Plasma glucose levels significantly changed during the study period (P<0.0001), with significant reductions observed in subjects assigned to HF (−0.6±0.3 mmol/L; P<0.0001) and IF (−0.5±0.1 mmol/L; P<0.0001) but not in those assigned to LF (−0.1±0.5 mmol/L; P=0.19; Table 2). Thus, plasma glucose levels at the end of follow-up were significantly (P<0.05) lower in subjects assigned to HF and IF in comparison with those assigned to LF (Table 2). With regard to plasma insulin levels, no significant differences were found among the 3 treatment groups at the end of follow-up (Table 2). HOMA-IR significantly improved through the study period (P<0.0001), with significant reductions in subjects assigned to HF (−1.6±1.0; P<0.0001) and IF (−0.9±0.2; P<0.0001) but not in those assigned to LF (−0.1±0.5; P=0.29; Table 2). HOMA-IR at the end of follow-up was significantly (P<0.05) better in subjects assigned to HF in comparison with those assigned to LF (Table 2). Spearman nonparametric correlation revealed significant relationships between changes of plasma glucose and HOMA-IR during the study period and their respective values at baseline (r=−0.325, P=0.0018 and r=−0.421 and P<0.0001, respectively) in the study population considered as a whole.

With regard to total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglycerides concentrations, no significant differences were found among the 3 treatment groups at the end of follow-up (Table 2). Plasma total 8-iso-PGF levels significantly decreased during the study period (P<0.0001), with significant reductions observed among subjects assigned to HF (−99.8±60.3 pg/L; P<0.0001) and IF (−65.2±87.2 pg/L; P=0.0003), but not in those assigned to LF (−3.6±51.4 pg/L; P=0.71) drinks (Table 2). Thus, plasma total 8-iso-PGF levels at the end of follow-up were significantly (P<0.05) lower in subjects assigned to HF in comparison with those assigned to LF (Table 2).

Determinant of Cognitive Changes After Flavanol Consumption

Because of the collinearity between systolic and diastolic blood pressures, only changes in systolic blood pressure were entered in the hierarchical regression analysis, because they were found to be more strictly correlated with changes of composite cognitive z score than those of diastolic blood pressure (r=−0.4967, P<0.0001 versus r=−0.4036, P<0.0001). Changes of HOMA-IR were found to be the main determinants of changes in cognitive function, explaining ≈40% of composite z score variability through the study period (r2=0.4013, β=−0.2910; P<0.0001; Table S4). Changes in systolic blood pressure levels and plasma isoprostane concentrations explained ≈2% and ≈7% of cognitive improvement throughout the study period, respectively (Table S4).


In this randomized, double-blind, parallel arm study, we sought to evaluate the effect of daily consumption of flavanol-rich cocoa drinks for 8 weeks on cognitive function in older adults with early memory decline. We found that cognitive performance was improved with regular cocoa flavanol consumption without evidence of any relevant adverse effects. In addition to these cognitive improvements, the regular dietary inclusion of cocoa flavanols was associated with significant reductions in blood pressure, as well as in improvements in some metabolic markers. Interestingly, the improvement of cognitive performance was associated with a reduction in insulin resistance, suggesting a possible influential role of glucose metabolism in modulating cognitive function in these subjects.

Epidemiological studies have previously reported a favorable association between increased flavonoid consumption and cognitive function in elderly subjects.68,23 Moreover, it has been demonstrated recently that 12-week supplementation with polyphenol-rich concord grape juice may enhance cognitive function in a small group (n=12) of older adults with early memory decline.24 The current study sheds new light on this relevant topic, providing the first evidence that regular cocoa flavanol consumption can positively affect cognitive function in older adults with early memory decline. To the best of our knowledge, this is the first well-controlled dietary intervention study of this size and duration examining neurocognitive performance in response to multiple levels of dietary flavanols. We found significant improvements in performance on a battery of neuropsychological tests mainly exploring executive function but also aspects of working memory, short-term memory, long-term episodic memory, processing speed, and global cognition, after 8 weeks of regular cocoa flavanol consumption. In particular, the benefits were remarkably evident for processing speed, executive function, and working memory, as indicated by the improvement in TMT A and TMT B scores. Interestingly, these improvements were already evident at an intermediate cocoa flavanol content, suggesting a particular sensitivity of this subset of neuropsychological functions to the benefits of cocoa flavanols. According to the above findings, a significant improvement in performance in the verbal fluency test, a test commonly used as a measure of executive function and language, was also observed in subjects assigned to HF consumption. On the other hand, the lack of an effect of flavanols on the Mini Mental State Examination score is likely a reflection of the low sensitivity of this test to detect small changes at the upper end of cognitive performance over time. Taken together, these data are suggestive of a possible clinical benefit derived from the regular dietary inclusion of cocoa flavanol-containing foods in subjects with MCI. In this regard, the level of flavanols used in this current study is comparable to the level of cocoa flavanols fed in previous studies where improvements in brain blood flow,19 vascular function, and blood pressure were observed.25 However, our study is the first well-controlled trial that addresses the relationship between cocoa flavanols and cognitive function, and from this study alone, we cannot say what is the effective flavanol amount that is required to improve or maintain cognition.

Our data do not provide definite information as to possible mechanisms underlying the observed beneficial effects, although some putative pathways can be considered. In this regard, we found a significant improvement of insulin resistance after cocoa flavanol consumption. This evidence fully fits with evidence from the recent meta-analysis by Hooper et al25 indicating consistent acute and chronic benefits of chocolate- and cocoa-derived flavanols on insulin resistance. Interestingly, flavonoids have been reported recently to exert additional benefits to established hypoglycemic therapy in postmenopausal type 2 diabetic patients.26 Although it is unclear the mechanism underlying this effect, these previously published studies provide consistent evidence that the consumption of flavanols, either directly or indirectly, can mediate important improvements in insulin sensitivity. We found an independent negative relationship between changes in composite cognitive z score and change in HOMA-IR, which explained ≈40% of composite cognitive z score variation throughout the study period. These data suggest a possible influential role of insulin-resistance in modulating cognitive function in subjects with early memory decline. Supportive of this concept, it has been demonstrated recently that the insulin sensitizer rosiglitazone could protect against cognitive decline in diabetic subjects with MCI.27 An improvement in brain perfusion could have also played a contributory role in the ameliorating cognitive performance in our study population. This hypothesis is supported by the findings of another study, which demonstrated that 2 weeks of regular flavanol-rich cocoa consumption, providing an intake of cocoa flavanols close to that of HF drinks used in our study, was associated with a significant increase of cerebral blood in healthy elderly.19 In this regard, although we did not specifically evaluate the effect of cocoa flavanols on endothelial function, a number of studies, some also from our group, clearly demonstrated the ability of flavonoids to improve NO bioavailability in vasculature,17,18,25 which, in turn, plays a pivotal role in modulating cerebrovascular tone.28

The current demonstration of significant reductions in both systolic and diastolic blood pressure levels after regular cocoa flavanol consumption is consistent with previous findings obtained both in normotensives and in hypertensive subjects, as well as medicated individuals.17,18,29,30 Our findings seem to exclude any relevant contribution of blood pressure reduction to the variations of cognitive performance after 8 weeks of cocoa flavanol consumption, because blood pressure changes explained ≤2% of cognitive improvement throughout the study period. However, we cannot exclude that these reductions in blood pressure, in addition to improving cardiovascular health, may contribute to supporting improvements in cognitive performance and even help in the prevention of blood pressure-related cognitive declines across a longer time course.31 Indeed, evidence is mounting that the management of high blood pressure and other conventional cardiovascular risk factors could favorably affect the clinical course of MCI.3,32

The last interesting finding of our study was the significant reduction of circulating levels of plasma F2-isoprostanes after flavanol-rich cocoa consumption. F2-isoprostanes are generated from arachidonic acid through a process of nonenzymatic free radical-catalyzed lipid peroxidation, thus representing an established marker of oxidative stress.33 These effects probably represent the consequence of the global effects of these nutrients on the cardiovascular system rather then the expression of a direct effect of cocoa flavanol. Indeed, a direct antioxidant effect of polyphenols in vivo has been questioned recently, because concentrations in blood are low compared with other antioxidants, and extensive metabolism after ingestion lowers their antioxidant activity.34 However, the contribution of oxidative stress, at least as assessed throughout lipid peroxidation products, to the variations of cognitive performance observed after cocoa flavanol consumption was quite modest, because changes of plasma isoprostane level concentrations explained ≈7% of cognitive improvement throughout the study period.

In summary, the results of the current study provide encouraging evidence that the regular inclusion of flavanol-containing foods may be an effective dietary approach for improving some aspects of cognitive dysfunction in adults with MCI. Other than the possible direct effects of flavanols on cognitive function, general improvements in cardiovascular function and specific metabolic parameters could have, alone or in combination, played a role in improving cognitive performance in this study population.


Although results of our study are remarkably good, their potential clinical relevance requires some considerations. First, because standard MCI criteria do not define optimal tests to establish cognitive performance in subjects with MCI,22 cognitive testing was performed using a battery of well-validated, standardized, and widely used neuropsychological tests exploring only some aspects of the complex cognitive dysfunction in subjects with MCI.1,2 Second, because our study consisted of a 2-month intervention, the extent of the cognitive benefits and their duration, as well as their impact on a clinical course of MCI, remain to be established. Third, from pathophysiological perspective, it is yet unclear whether the observed benefits in neurocognition are a direct consequence of cocoa flavanol themselves or a secondary effect related to general improvements in cardiovascular function or health. Finally, subjects enrolled in our study were generally in good health and without known cardiovascular disease. Thus, our study population could be not completely representative of all subjects with MCI. On the other hand, the exclusion of habitual consumers of dietary supplements with antioxidant properties likely did not affect the representativeness of our study population, because the consumption of these supplements among our elderly population is quite low, generally <10%, similar to that observed in other cohorts from our country.35


Our data seem to support the hypothesis that the regular consumption of cocoa flavanols may be able to improve cognitive performance in MCI adults in a relatively short period of time. Although additional confirmatory studies are warranted, the findings reported herein suggest that the regular dietary inclusion of flavanols could be one element of a dietary approach to the maintaining and improving not only cardiovascular health but also specifically brain health.

Sources of Funding

This study was supported by a grant from Mars Inc, which also supplied the standardized powdered cocoa drinks used in this investigation.


C.K.-U. is employed by Mars Inc.


The online-only Data Supplement is available with this article at

Correspondence to Giovambattista Desideri,
University of L'Aquila, Department of Life, Health, and Environmental Sciences, Viale S Salvatore, Delta 6 Medicina, 67100 Coppito, L'Aquila, Italy.


  • 1. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol. 1999; 56:303–308.CrossrefMedlineGoogle Scholar
  • 2. Petersen RC, Stevens JC, Ganguli M, Tangalos EG, Cummings JL, DeKosky ST. Practice parameter: early detection of dementia–mild cognitive impairment (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2001; 56:1133–1142.CrossrefMedlineGoogle Scholar
  • 3. Li J, Wang YJ, Zhang M, Xu ZQ, Gao CY, Fang CQ, Yan JC, Zhou HD. Vascular risk factors promote conversion from mild cognitive impairment to Alzheimer disease. Neurology. 2011; 76:1485–1491.CrossrefMedlineGoogle Scholar
  • 4. Gómez-Pinilla F. Brain foods: the effects of nutrients on brain function. Nat Rev Neurosci. 2008; 9:568–578.CrossrefMedlineGoogle Scholar
  • 5. Commenges D, Scotet V, Renaud S, Jacqmin-Gadda H, Barberger-Gateau P, Dartigues JF. Intake of flavonoids and risk of dementia. Eur J Epidemiol. 2000; 16:357–363.CrossrefMedlineGoogle Scholar
  • 6. Kuriyama S, Hozaka A, Ohmori K, Shimazu T, Matsui T, Ebihara S, Awata S, Nagatomi R, Arai H, Tsuji I. Green tea consumption and cognitive function: a cross-sectional study from the Tsurugaya Project. Am J Clin Nutr. 2006; 83:355–361.CrossrefMedlineGoogle Scholar
  • 7. Letenneur L, Proust-Lima C, Le Gouge A, Dartigues JF, Barberger-Gateau P. Flavonoid intake and cognitive decline over a 10-year period. Am J Epidemiol. 2007; 165:1364–1371.CrossrefMedlineGoogle Scholar
  • 8. Nurk E, Refsum H, Drevon CA, Tell GS, Nygaard HA, Engedal K, Smith AD. Intake of flavonoid-rich wine, tea, and chocolate by elderly men and women is associated with better cognitive test performance. J Nutr. 2009; 139:120–127.CrossrefMedlineGoogle Scholar
  • 9. Williams RJ, Spencer JPE. Flavonoids, cognition, and dementia: actions, mechanisms, and potential therapeutic utility for Alzheimer disease. Free Radic Biol Med. 2012; 52:35–45.CrossrefMedlineGoogle Scholar
  • 10. Gorelick PB, Scuteri A, Black SE, DeCarli C, Greenberg SM, Iadecola C, Launer LJ, Laurent S, Lopez OL, Nyenhuis D, Petersen RC, Schneider JA, Tzourio C, Arnett DK, Bennett D, Chui HC, Higashida RT, Lindquist R, Nilsson PR, Roman GC, Selke FW, Seshadri S. Vascular contributions to cognitive impairment and dementia: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2011; 42:2672–2713.LinkGoogle Scholar
  • 11. Novak V, Hajjar J. The relationship between blood pressure and cognitive function. Nat Rev Cardiol. 2010; 7:686–698.CrossrefMedlineGoogle Scholar
  • 12. De la Monte S. Insulin resistance and Alzheimer's disease. BMB Reports. 2009; 42:475–481.CrossrefMedlineGoogle Scholar
  • 13. Neumann KF, Rojo L, Navarrete LP, Navarrete G, Reyes P, Maccioni RB. Insulin resistance and Alzheimer's disease: molecular links & clinical implications. Curr Alzheimer Res. 2008; 5:438–447.CrossrefMedlineGoogle Scholar
  • 14. Li L, Holscher C. Common pathological processes in Alzheimer disease and type 2 diabetes: a review. Brain Res Rev. 2007; 56:384–402.CrossrefMedlineGoogle Scholar
  • 15. Strachan MW, Reynolds RM, Marioni RE, Price JF. Cognitive function, dementia and type 2 diabetes mellitus in the elderly. Nat Rev Endocrinol. 2011; 7:108–114.CrossrefMedlineGoogle Scholar
  • 16. Talbot K, Wang HY, Kazi H, Han LY, Bakshi KP, Stucky A, Fuino RL, Kawaguchi KR, Samoyedny AJ, Wilson RS, Arvanitakis Z, Schneider JA, Wolf BA, Bennett DA, Trojanowski JQ, Arnold SE. Demonstrated brain insulin resistance in Alzheimer's disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. J Clin Invest. 2012; 122:1316–1338.CrossrefMedlineGoogle Scholar
  • 17. Grassi D, Desideri G, Necozione S, Lippi C, Casale R, Properzi G, Blumberg JB, Ferri C. Blood pressure is reduced and insulin sensitivity increased in glucose-intolerant, hypertensive subjects after 15 days of consuming high-polyphenol dark chocolate. J Nutr. 2008; 138:1671–1676.CrossrefMedlineGoogle Scholar
  • 18. Grassi D, Necozione S, Lippi C, Croce G, Valeri L, Pasqualetti P, Desideri G, Blumberg JB, Ferri C. Cocoa reduces blood pressure and insulin resistance and improves endothelium-dependent vasodilation in hypertensives. Hypertension. 2005; 46:398–405.LinkGoogle Scholar
  • 19. Sorond FA, Lipsitz LA, Hollenberg NK, Fisher NDL. Cerebral blood flow response to flavanol-rich cocoa in healthy elderly humans. Neuropsychiatr Dis Treat. 2008; 4:433–440.MedlineGoogle Scholar
  • 20. Francis ST, Head K, Morris PG, MAcDonald IA. The effect of flavanol-rich cocoa on the fMRI response to a cognitive task in healthy young people. J Cardiovasc Pharmacol. 2006; 47:S215–S220.CrossrefMedlineGoogle Scholar
  • 21. Winblad B, Palmer K, Palmer K, Jelic V, Fratiglioni L, Wahlund LO, Nordberg A, Bäckman L, Albert M, Almkvist O, Arai H, Basun H, Blennow K, de Leon M, DeCarli C, Erkinjuntti T, Giacobini E, Graff C, Hardy J, Jack C, Jorm A, Ritchie K, van Duijn C, Visser P, Petersen RC. Mild cognitive impairment–beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med. 2004; 256:240–246.CrossrefMedlineGoogle Scholar
  • 22. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med. 2004; 256:183–194.CrossrefMedlineGoogle Scholar
  • 23. Macready AL, Kennedy OB, Ellis JA, Williams CM, Spencer JPE, Butler LT. Flavonoids and cognitive function: a review of human randomized controlled trial studies and recommendations for future studies. Genes Nutr. 2009; 4:227–2242.CrossrefMedlineGoogle Scholar
  • 24. Krikorian R, Nash TA, Shidler MD, Shukitt-Hale B, Joseph JA. Concord grape juice supplementation improves memory function in older adults with mild cognitive impairment. Br J Nutr. 2010; 103:730–734.CrossrefMedlineGoogle Scholar
  • 25. Hooper L, Kay C, Abdelhamid A, Kroon PA, Cohn JS, Rimm EB, Cassidy A. Effects of chocolate, cocoa, and flavan-3-ols on cardiovascular health: a systematic review and meta-analysis of randomized trials. Am J Clin Nutr. 2012; 95:740–751.CrossrefMedlineGoogle Scholar
  • 26. Curtis PJ, Sampson M, Potter J, Dhatariya K, Kroon PA, Cassidy A. Chronic ingestion of flavan-3-ols and isoflavones improves insulin sensitivity and lipoprotein status and attenuates estimated 10-year CVD risk in medicated postmenopausal women with type 2 diabetes: a 1-year, double-blind, randomized, controlled trial. Diabetes Care. 2012; 35:226–232.CrossrefMedlineGoogle Scholar
  • 27. Abbatecola AM, Lattanzio F, Molinari AM, Cioffi M, Mansi L, Rambaldi P, DiCioccio L, Cacciapuoti F, Canonico R, Paolisso R. Rosiglitazone and cognitive stability in older persons with type 2 diabetes and mild cognitive impairment. Diabetes Care. 2010; 33:8:1706–1711.CrossrefGoogle Scholar
  • 28. Andresen J, Shafi NI, Bryan RM. Endothelial influences on cerebrovascular tone. J Appl Physiol. 2006; 100:318–327.CrossrefMedlineGoogle Scholar
  • 29. Egan BM, Laken MA, Donovan JL, Woolson RF. Does dark chocolate have a role in the prevention and management of hypertension?: commentary on the evidence. Hypertension. 2010; 55:1289–1295.LinkGoogle Scholar
  • 30. Heiss C, Jahn S, Taylor M, Real WM, Angeli FS, Wong ML, Amabile N, Prasad M, Rassaf T, Ottaviani JI, Mihardja S, Keen CL, Springer ML, Boyle A, Grossman W, Glantz SA, Schroeter H, Yeghiazarians Y. Improvement of endothelial function with dietary flavonoids is associated with mobilization of circulating angiogenic cells in patients with coronary artery disease. J Am Coll Cardiol. 2010; 56:218–224.CrossrefMedlineGoogle Scholar
  • 31. Kilander L, Nyman H, Boberg M, Hansson L, Lithell H. Hypertension is related to cognitive impairment: a 20-year follow-up of 999 men. Hypertension. 1998; 31:780–786.LinkGoogle Scholar
  • 32. Solfrizzi V, Panza F, Colacicco AM, D'Introno A, Capurso C, Torres F, Grigoletto F, Maggi S, Del Parigi A, Reiman EM, Caselli RJ, Scafato E, Farchi G, Capurso A, for the Italian Longitudinal Study on Aging Working Group. Vascular risk factors, incidence of MCI, and rates of progression to dementia. Neurology. 2004; 63:1882–1891.CrossrefMedlineGoogle Scholar
  • 33. Patrono C, Fitzgerald GA. Isoprostanes: potential markers of oxidant stress in atherothrombotic disease. Arterioscler Thromb Vasc Biol. 1997; 17:2309–2315.LinkGoogle Scholar
  • 34. Hollman PC, Cassidy A, Comte B, Heinonen M, Richelle M, Richling E, Serafini M, Scalbert A, Sies H, Vidry S. The biological relevance of direct antioxidant effects of polyphenols for cardiovascular health in humans is not established. J Nutr. 2011; 141:989S–1009S.CrossrefMedlineGoogle Scholar
  • 35. Skeie G, Braaten T, Hjartåker A, Lentjes M, Amiano P, Jakszyn P, Pala V, Palanca A, Niekerk EM, Verhagen H, Avloniti K, Psaltopoulou T, Niravong M, Touvier M, Nimptsch K, Haubrock J, Walker L, Spencer EA, Roswall N, Olsen A, Wallström P, Nilsson S, Casagrande C, Deharveng G, Hellström V, Boutron-Ruault MC, Tjønneland A, Joensen AM, Clavel-Chapelon F, Trichopoulou A, Martinez C, Rodríguez L, Frasca G, Sacerdote C, Peeters PH, Linseisen J, Schienkiewitz A, Welch AA, Manjer J, Ferrari P, Riboli E, Bingham S, Engeset D, Lund E, Slimani N. Use of dietary supplements in the European Prospective Investigation into Cancer and Nutrition calibration study. Eur J Clin Nutr. 2009; 63:S226–S238.CrossrefMedlineGoogle Scholar

Novelty and Significance

What Is New?

  • For the first time, regular cocoa flavanol consumption has been shown to positively affect cognitive function in older adults with early memory decline.

What Is Relevant?

  • The cognitive improvements were demonstrated along with significant improvements in blood pressure and insulin sensitivity.


Regular consumption of cocoa flavanols might be effective in improving cognitive function in elderly subjects with MCI. This effect appears mediated in part by an improvement in insulin sensitivity.


eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.