Cocoa Drink Improves Walking Distance in Those with Peripheral Artery Disease

A review of cocoa research and findings

By Paul Richard Saunders, PhD, ND, DHANP, CCH

Printer Friendly PagePrinter Friendly Page


McDermott MM, Criqui MH, Domanchuk K, et al. Cocoa to improve walking performance in older people with peripheral artery disease: the COCOA-PAD pilot randomized clinical trial. Circ Res. 2020;126(5):589-599.

Study Objectives

To test the hypothesis that daily cocoa consumption for 6 months would prevent a decline in 6-minute walking distance of older adults with peripheral artery disease (PAD)


Parallel-design, double-blind, randomized clinical trial. Randomization was separate if baseline 6-minute walk was less than 365.75 m (1,200 feet) versus a greater distance.


All participants (N=118) were volunteers aged more than 60 years and from the Chicago area. They were either participants in prior research with the principal investigator (McDermott MM) and had expressed interest in future research or they replied to Chicago Transit Authority advertisements.

Of the 118 individuals screened, 44 were randomized to either the cocoa arm (n=23) or placebo arm (n=21). Of the 44 randomized, 40 completed the 6-month follow-up, with 36 completing the 6-minute walk at the 6-month follow-up.

The researchers performed a muscle biopsy on 22 participants, but only 21 had adequate tissue for testing, and 17 of those completed a follow-up biopsy. Among the 4 who did not undergo a follow-up biopsy, 1 had left the state, 1 had a stroke, 1 yielded no muscle tissue on biopsy, and 1 canceled their biopsy.

The placebo and cocoa arms were similar except that the body mass index (BMI) was lower in the cocoa group and there were more smokers and blacks in the cocoa group. Treatment adherence was 68% with the cocoa intervention and 80% with the placebo, but 7 participants discontinued their given beverage, yielding an adherence rate of 80% and 87%, respectively.

Inclusion Criteria

Adults aged more than 60 years with PAD defined by ankle brachial index (ABI) <0.90 in either leg or vascular laboratory or angiographic evidence of PAD. The absence of intermittent claudication was not an exclusion criterion because most adults with PAD do not have intermittent claudication symptoms.

Exclusion Criteria

Adults who were unwilling to give up dietary sources of chocolate, allergic to chocolate, or unable to consume products manufactured on equipment processing nuts, eggs, wheat, soy, or milk were excluded. The researchers also excluded adults with major leg amputation, critical limb ischemia, wheelchair confinement, walking-aid use, walking impairment for reasons other than PAD, or significant visual and hearing impairment. Also excluded were individuals who were on dialysis, required oxygen, or had a major cardiovascular event, revascularization, or major surgery in the past 3 months, and those planning major surgery or revascularization in the next 6 months. Additional exclusion criteria included adults treated for cancer in the last 2 years unless prognosis was excellent; adults with a baseline 6-minute walk of less than 500 feet or more than 1,600 feet; and those with a Mini-Mental Status Examination score of less than 23 at baseline.


Flavanol-rich cocoa and matching placebo were manufactured by the Hershey Company and dispensed in packets. The researchers instructed the participants to mix the packet contents with water or milk and consume 3 packets per day. To prevent weight gain, participants were individually counseled on similar caloric foods that could be eliminated during the study. Participants brought unopened packets to monthly study visits for counting. Hershey notified the data management team at the end as to which packet was cocoa and which was placebo. The researchers measured flavanol metabolites and theobromine.


Researchers used a handheld Doppler probe to measure systolic pressure twice in each brachial, dorsalis pedis, and posterior tibial artery. They calculated the ankle brachial index (ABI) by dividing the mean of the dorsalis pedis and posterior tibial pressures of each leg by the mean of the 4 brachial pressures. Researchers also collected medical history, race, and demographics and measured height and weight.

Leg symptoms were characterized using the San Diego Claudication Questionnaire (SDCQ). Intermittent claudication was defined as exertional calf pain that did not begin at rest, caused the subject to stop walking, and resolved within 10 minutes of rest.

Researchers measured 6-month change in 6-minute walk distance at 2.5 hours and 24 hours after study-beverage consumption. They measured 6-month change in brachial artery flow-mediated dilation 2.5 hours and 24 hours after participants consumed the study beverage. They also measured change in 6-month maximal pain-free treadmill walking distance 48 hours after consumption of the study beverage.

Muscle biopsy from the medial head of the gastrocnemius muscle was performed at baseline and 6-month follow-up and measured PGC-1α (peroxisome proliferator-activated receptor-gamma coactivator 1-alpha), follistatin, myostatin, citrate synthase, and COX (cytochrome C oxidase) activity.

Changes in calf muscle perfusion and calf biopsy measures of satellite cell abundance, capillary density, centrally nucleated and eMyHC (embryonic myosin heavy chain)–expressing fiber abundance, and oxidative stress measures (nitrotyrosine, 4-hydroxynonenal) were also taken at baseline and at 6 months. Participants performed the 6-minute walk distance in a 100-foot hallway for 6 minutes with instructions to walk as far as possible. They did free-living physical activity over 7 days with an ActiGraph accelerometer worn on the right hip except during bathing and sleep.


Sample size was based on a 10% dropout rate. Baseline characteristics were summarized by simple statistics; 1-tailed t-tests were used to compare changes at 6 months between cocoa and placebo. A P<0.10 was considered significant. Analyses were repeated using ANCOVA adjusting for baseline differences in characteristics between the 2 groups. Statistical analyses were performed using SAS, version 9.4.

Key Findings

The primary outcome of the 6-minute walk distance at 6 months significantly improved in the cocoa group by 42.6 m (P=0.005) at the walk 2.5 hours post beverage. However, at 24 hours post beverage, improvement was an insignificant 18.0 m (P=0.12). In comparison, the placebo-group distance decreased by 24.2 m for the 2.5-hour-post-beverage walk, and decreased by 2.9 m in the 24-hour-post-beverage walk. When the missing data were imputed, the cocoa group improved by 32.0 m (P=0.04) at the walk 2.5 hours post beverage, and improved by a statistically significant 21.5 m (P=0.084) at the walk 24 hours post beverage. In the placebo group, the imputed values decreased by 22.2 m at the 2.5-hour-post-beverage walk, and decreased by 4.3 m in the 24-hour-post-beverage walk. In PAD a small, meaningful change in the 6-minute walk is 12.0 m (39.3 feet) while a large change is 34 m (111.5 feet).1

The changes in epicatechin metabolites and theobromine were significantly greater in the cocoa versus the placebo group, P=0.001 and P=0.010, respectively.

The 6-month treadmill walking time at 48 hours post beverage; the change in brachial artery flow-mediated dilation; change in physical activity; and calf muscle abundance of PGC-1α, myostatin, follistatin, citrate synthase activity, and COX activity were not statistically significant. The cocoa group had increased calf muscle COX activity compared to the placebo arm, but this was not statistically significant. The cocoa group had increased capillary density and increased calf muscle perfusion, but the values were not statistically significant.

There was no significant change in BMI of cocoa versus placebo groups. Adverse events in the cocoa group included 1 death to myocardial infarction, 2 hospitalizations due to lower-extremity revascularizations, and 1 hospitalization each from ischemic stroke, foot infection, urinary difficulty, and depression. In the placebo group there was 1 hospitalization for atrial fibrillation and 1 for cough with chest pain. None of the adverse events were considered related to the study.

Practice Implications

Theobroma cacao, Malvaceae family, is a small tree native to South America that spread into Central America where it was used by the Mayan and Aztec civilizations since at least 460 AD.2 Aztecs called the drink xocoatl, after the Aztec god Quetzalcoatl, who brought heavenly cacao down to earth.3 The beverage was a currency, reserved for the highest social classes, and a medicinal to fight fatigue and gastrointestinal distress.3 It was brought to Spain in 1528 following the conquest of modern Mexico by Hernán Cortés and for the next 100 years swept Europe as a fad, with people adding cane sugar, vanilla, cinnamon, and anise seed to hide its bitterness. These additions contrasted with the pepper and native herbs added by the Aztecs.3 In France it was drunk only in royal courts. In 1657 the first chocolate house opened in London. It was not available in the United States until the mid-1800s due to high duties on imported cocoa beans and sugar. By the time the United States was participating in World War I, it was a part of the rations to American soldiers because of its high resistance to spoilage, a byproduct of its potent antioxidants.3 Medically doctors gave it to induce weight gain in emaciated patients, stimulate the nervous system, and to improve digestion and elimination.2 Today its per capita annual consumption ranges from 0.12 kg in China to 11.85 kg in Ireland, with the United States at 5.18 kg (11.42 pounds). The leading producer is Côte d’Ivoire, producing a half million tons in 2006-2007.2

The seeds in the pod are dried and fermented; from the ground-fermented beans, a paste is made that contains nonfat cocoa solids and cocoa butter. Cocoa powder is made by removing some of the cocoa butter; sugar is added, and the portion of cocoa liquor determines how dark a chocolate it is. Milk chocolate in the United States is typically 10% to 12% cocoa liquor, semisweet or bittersweet is at least 35% cocoa liquor by weight, and white chocolate is only cocoa butter, at least 20% by weight, with sweeteners and dairy added.2 Cocoa powder was developed by Coenraad van Houten of Holland in 1828, and solid chocolate followed shortly after. Milk chocolate was created around 1875 by Daniel Peter of Vevey, Switzerland, a neighbor of Henri Nestlé, then a manufacturer of baby food.4

Like most plant extracts, chocolate is a complex food with multiple bioactive compounds. Cocoa butter contains fatty acids including monounsaturated oleic acid (34%), stearic acid (34%), palmitic acid (25%), and polyunsaturated linoleic acid (2%).4 Much of the fiber or bran (26% to 40%) is removed in processing; protein is 15%, while the methylxanthines are theobromine (2%), caffeine (0.5%), and theophylline.4 The important polyphenols can constitute up to 50 mg per gram, including epicatechin, catechins, and procyanidins.2

Cocoa also contains several important minerals. A 100-kcal serving of dark chocolate (70% to 85% cacao) provides 36 mg of magnesium, 31% of the US Recommended Dietary Allowance (RDA) of copper, 114 mg of potassium, and 25% of the RDA of iron for men and postmenopausal women.2 Overall, the most significant components for cardiovascular health are the fatty acids, polyphenols, potassium, and magnesium.

The bulk of the data suggest that chocolate is beneficial to the human cardiovascular system. It contains beneficial fatty acids, and while the fiber is removed during processing, important minerals and polyphenols remain.

One of the earliest reports of the benefit of chocolate was from a comparison of hypertension rates in 2 geographic populations that share an ethnic heritage. The Kuna Indians on the San Blas Islands near Panama were compared with the Kuna Indians in Panama City, Panama.5,6 Island dwellers had a very low incidence of hypertension versus those who resided on the mainland. The Island Kuna diet included 5 cups of cocoa per day, yielding 900 mg/d of flavonoids.7 Diet, smoking, and other factors were not significant compared to chocolate intake.

The Dutch Zutphen Elderly Study of nondiabetic men found an inverse relationship between systolic (P=0.06) and diastolic (P=0.03) blood pressures, lower cardiovascular disease, and an all-cause mortality in the highest tertile of cocoa intake that suggested risk was cut by up to 50%.8 These men consumed 2.3 grams of chocolate per day.

A Swedish study of nondiabetic hospital patients after their first myocardial infarction, followed over 8.6 years, found that those who reported eating chocolate at least twice per week were 66% less likely to suffer a cardiac death compared to those who never ate chocolate.9 A Japanese study of men found that the risk of diabetes was reduced 35% among those who consumed chocolate snack pieces once or more per week compared to those who never consumed it; the risk reduction for women did not reach statistical significance.10 An Italian study of men and women found that those consuming 20 grams per day of dark chocolate had the lowest C-reactive protein.11 Among 2,217 participants in the National Heart, Lung, and Blood Institute (NHLBI) Family Heart Study, there was an inverse relationship between higher chocolate consumption and calcified atherosclerotic plaque in the coronary arteries.12 Those consuming chocolate of any type twice or more per week were 32% less likely to have coronary atherosclerosis.

Several studies have shown benefit for women as well. The Iowa Women’s Health Study followed 34,489 women for 16 years. Consumption of chocolate was measured by completion of a food-frequency questionnaire. Following multivariate adjustments, there was a significant inverse relationship between chocolate intake and cardiovascular disease mortality.13,14 The Swedish Mammography Cohort found an inverse relationship between heart failure hospitalization or death and moderate chocolate intake, defined as 1 to 2 servings (average portion 19-30 g) per week.15 In a study of 57,502 Danish women and men who were enrolled when aged 50 to 64 years in 1993 to 1997, the incidence of atrial fibrillation was inverse to chocolate consumption ranging from once per month to once per day for both women and men in a significant manner (P<0.0001).16

Short-term consumption effects appear to be less apparent than long-term consumption effects. A double-blind, placebo-controlled trial in 101 healthy adults (41 men and 60 women, aged 60 years or more) assessed neuropsychological, hematological, and blood pressure variables over 6 weeks.17 The chocolate dose was a 37-g dark chocolate bar (397.30 mg proanthocyanidins/g) and a 237-mL artificially sweetened cocoa beverage (357.42 mg proanthocyanidins/g) daily or similar placebo (0.20 mg/g and 40.87 mg/g proanthocyanidins, respectively). At the conclusion of the trial, there was no effect on BMI, diverse mental function tests (5), memory, thinking processes, mood, energy, overall health, cholesterol values, C-reactive protein, systolic blood pressure, or diastolic blood pressure. The only significant finding was a significantly higher pulse rate by 5 beats per minute (P=0.007) in the chocolate group after 6 weeks.17

As further evidence, a systemic review and meta-analysis published in 2012 included 42 acute and short-term chronic (fewer than 18 weeks), randomized, controlled trials.18 Insulin resistance was improved as suggested by significant reductions in serum insulin. Flow-mediated dilatation after acute and chronic intake also improved, as did diastolic blood pressure, mean arterial pressure, and low-density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol. Doses above 50 mg epicatechin/day resulted in a greater effect.

A subsequent meta-analysis in 2017 assessed chocolate intake on risk of coronary heart disease, stroke, and diabetes. Fourteen prospective studies of primary prevention involving 508,705 patients were included.19 The dose-response for coronary heart disease and stroke was at least 3 servings (30 g of chocolate) per week. In those with diabetes the benefit occurred at 2 servings per week.19 The authors concluded that less than 6 servings per week may be optimal for prevention of these disorders.

The bulk of the data suggest that chocolate is beneficial to the human cardiovascular system. It contains beneficial fatty acids, and while the fiber is removed during processing, important minerals and polyphenols remain. The flavanols increase plasma antioxidant capacity and improve endothelial wall and platelet function even in young smokers.20 There is a reduction in nitrogen reactive species and protection of LDL against oxidation. The procyanidins inhibit the activation and expression of matrix metalloproteinases-2 (MMP-2) that can cause the rupture of the unstable plaques, and the flavanols may interfere with the NF-kB (nuclear factor kappa-light-chain-enhancer of activated B cells) pathway.13 There may also be an inhibition of the angiotensin-converting enzyme after consumption of 75 grams of dark chocolate (72% cocoa).21 Like aspirin, chocolate can be inhibitory to epinephrine-stimulated platelet activation and function, but to a lesser degree. As a food, chocolate inhibits platelet effects in vivo as seen in numerous trials, thus it can be eaten and does not have to be in a capsule.13

Some of the unanswered questions include: Does everyone benefit? Is it only helpful if there is endothelial or platelet dysfunction? Do any medications reduce or potentiate its clinical effects? What is the ideal dose and how often should it be taken? How long does chocolate need to be consumed to be of cardiovascular benefit? It does seem that more polyphenols are better, but how many? Dark chocolate has less dairy and sugar and, therefore, fewer calories than white or milk chocolate, but what percentage of dark chocolate is optimal? Is chocolate potentiated or mitigated by any dietary foods?

The article under review here was well written. The Hershey Company provided the chocolate treatment, while Mars, Inc., performed the detailed analysis. The detailed analysis of the chocolate’s contents, which the article stated would be in the supplemental data, was absent, and there was no response to my email inquiry requesting this information. Funding was by the National Institute on Aging and the National Institutes of Health’s Office of Dietary Supplements. The calf muscle biopsy and analysis was a unique feature of the study. Chocolate intake seemed to be well controlled in the placebo group, but no mention was made of exercise recommendations or restrictions to either group, although the assumption, because of the PAD, is that it was self-limited.


In this study of adults aged more than 60 years with confirmed PAD who consumed a >85% cacao beverage 3 times per day for 6 months, there was a significant improvement in the 6-minute walking test distance at 2.5 hours post last cocoa beverage, compared to the placebo group, but not 24 hours after the last cocoa beverage. There also was no statistically significant effect on change in the treadmill walking time 48 hours post last cocoa beverage. There were several positive changes in the calf muscles of the treatment group, but none of these were statistically significant. There was no change in BMI between the 2 groups.

About the Author

Paul Richard Saunders, PhD, ND, DHANP, CCH, completed his PhD in forest ecology at Duke University, his naturopathic degree at Canadian College of Naturopathic Medicine, and his homeopathic residency at National University of Naturopathic Medicine, where he also earned a second naturopathic degree. He is professor of materia medica and clinical medicine at the Canadian College of Naturopathic Medicine; senior naturopathic doctor, Beaumont Health System, Troy Hospital, Michigan; and adjunct professor of integrative medicine, Oakland University William Beaumont Medical School and has a private practice in Dundas, Ontario. Saunders was a member of the transition team that formed the Office of Natural Health Products, served as a natural health expert to the Directorate, and has served on several expert panels for Health Canada. He has conducted clinical research, supervised students and residents, and published widely.


  1. Gardner AW, Montgomery PS, Wang M. Minimal clinically important differences in treadmill, 6-minute walk, and patient-based outcomes following supervised and home-based exercise in peripheral artery disease. Vasc Med. 2018;23(4):349-357.
  2. Katz DL, Doughty K, Ali A. Cocoa and chocolate in human health and disease. Antioxid Redox Signal. 2011;15(10):2779-2811.
  3. Engler MB, Engler MM. The emerging role of flavonoid-rich cocoa and chocolate in cardiovascular health and disease. Nutr Reviews. 2006;64(3):109-118.
  4. Borches AT, Keen CL, Hannum SH, Gershwin ME. Cocoa and chocolate: composition, bioavailability, and health implications. J Med Foods. 2000;3(2):77-105.
  5. Pucciarelli DL. Cocoa and hearth health: a historical review of the science. Nutrients. 2013;5(10):3854-3870.
  6. McCullough M, Chevaux K, Jackson L, et al. Hypertension, the Kuna, and the epidemiology of flavanols. J Cardiovasc Pharmacol. 2006;47(Supple 2):S103-S109;discussion 119-121.
  7. Hollenberg N. Vascular action of cocoa flavanols in humans: the roots of the story. J Cardiovasc Pharmacol. 2006;47(Suppl 2):S99-S102;discussion S119-121.
  8. Buijsse B, Feskens EJ, Kok FJ, Kromhout D. Cocoa intake, blood pressure, and cardiovascular mortality: the Zutphen Elderly Study. Arch Intern Med. 2006;166(4):411-417.
  9. Janszky I, Mukamal KJ, Ljung R, Ahnve S, Ahlbom A, Hallqvist J. Chocolate consumption and mortality following a first myocardial infarction: the Stockholm Heart Epidemiology Program. J Intern Med. 2009;266(3):248-257.
  10. Obs S, Nagata C, Nakamura K, et al.Composition of coffee, green tea, oolong tea, black tea, chocolate snacks, and the caffeine content in relation to risk of diabetes in Japanese men and women. Br J Nutr. 2010;103(3):453-459.
  11. Di Giuseppe R, Di Castelnuovo A, Centritto F, et al. Regular consumption of dark chocolate is associated with low serum concentration of C-reactive protein in a healthy Italian population. J Nutr. 2008;138(10):1939-1945.
  12. Djousse L, Hopkins PN, Arnett DK, et al. Chocolate consumption is inversely associated with calcified atherosclerotic plaque in the coronary arteries: the NHLBI family heart study. Clin Nutr. 2011;30(1):38-43.
  13. Fernandez-Murga L, Tarin JJ, Garcia-Perez MA, Cano A. The impact of chocolate on cardiovascular health. Maturitas. 2011;69(4):312-321.
  14. Mink PJ, Scrafford CG, Barraj LM, et al. Flavonoid intake and cardiovascular disease mortality: a prospective study in postmenopausal women. Am J Clin Nutr. 2007;85(3):895-909.
  15. Mostofsky D, Levitan EB, Wolk A, Mittleman MA. Chocolate intake and incidence of heart failure: a populations-based prospective study of middle-aged and elderly women. Circ Heart Fail. 2010;3(5):612-616.
  16. Mostofsky E, Johansen MB, Tjonneland A, Chahal HS, Mittleman MA, Overvad K. Chocolate intake and risk of clinically apparent atrial fibrillation: the Danish diet, cancer and health study. Heart. 2017;103(15):1163-1167.
  17. Crews WD, Harrison DW, Wright JW. A double-blind, placebo-controlled, randomized trial of the effects of dark chocolate on variables associated with neuropsychological functioning and cardiovascular health: clinical findings from a sample of healthy, cognitively intact older adults. Am J Clin Nutr. 2008;87(4):872-880.
  18. Hooper L, Kay C, Abdelhamid A, et al. 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(3):740-751.
  19. Yuan S, Li X, Jin Y, Lu J. Chocolate consumption and risk of coronary heart disease, stroke, and diabetes: meta-analysis of prospective studies. Nutrients. 2017;9(7).688. pii:E688.
  20. Hermann F, Spieker LE, Ruschitzka F, et al. Dark chocolate improves endothelial and platelet function. Heart. 2006;92(1):119-120.
  21. Persson IA, Persson K, Hagg S, Andersson RG. Effects of cocoa extract and dark chocolate on angiotensin-converting enzyme and nitric oxide in human endothelial cells and healthy volunteers—a nutrigenomics perspective. J Cardiovasc Pharmacol. 2011;57(1):44-50.