Cacao

Cacao drinks were being made before 1000 BC. Residues in pottery from Puerto Escondido, in present-day Honduras, date the practice at least 500 years earlier than it had been confirmed before ^{36Reference 36John S et al. · 2007Chemical and archaeological evidence for the earliest cacao beverages}.

The earliest of these drinks were probably made by fermenting the sweet pulp around the seeds. The later chocolate drink, served on special occasions and especially by elites, was made from the seeds ^{36Reference 36John S et al. · 2007Chemical and archaeological evidence for the earliest cacao beverages}.

The Maya and Aztecs used cacao as both food and medicine ^{35Reference 35Giovanni Scapagnini et al. · 2014Cocoa Bioactive Compounds: Significance and Potential for the Maintenance of Skin Health}. The drink was held to build resistance and fight fatigue ^{24Reference 24Becker K et al. · 2013In vitroImmunomodulatory properties of cacao extracts — potential consequences for medical applications — in vitro studyView study →}.

Leaves, seeds, bark, flowers, pulp and cocoa butter have all been used medicinally ^{25Reference 25Baharum Z et al. · 2014In vitroIn vitro antioxidant and antiproliferative activities of methanolic plant part extracts of Theobroma cacaoView study →}.

In Aztec Mexico the seeds served as small coins, and Grieve recorded the practice as still in use in parts of Mexico ^{31Reference 31A Modern Herbal | CacaoA Modern Herbal | Cacao. (n.d.). Retrieved from https://www.botanical.com/botanical/mgmh/c/cacao-02.htmlView study →}.

The seed is taken as a heart and kidney tonic. An infusion of the baked seed membranes is drunk for anaemia ^{38Reference 38Fern · 2014Theobroma cacao — habit, range, habitat, cultivation and usesView study →}.

In Amazonas State, Brazil, rural people rub cocoa butter on bruises ^{38Reference 38Fern · 2014Theobroma cacao — habit, range, habitat, cultivation and usesView study →}.

Cocoa butter was used to soften and protect chapped hands and lips, and as a base for suppositories ^{31Reference 31A Modern Herbal | CacaoA Modern Herbal | Cacao. (n.d.). Retrieved from https://www.botanical.com/botanical/mgmh/c/cacao-02.htmlView study →}.

Theobromine was given mainly as a diuretic, especially for fluid build-up caused by heart failure, often alongside digitalis. It was also used for high blood pressure, because it widens the blood vessels ^{31Reference 31A Modern Herbal | CacaoA Modern Herbal | Cacao. (n.d.). Retrieved from https://www.botanical.com/botanical/mgmh/c/cacao-02.htmlView study →}.

Flow-mediated dilation (FMD) is an ultrasound measure of how well an artery relaxes. A 2019 meta-analysis of 15 published studies with 18 intervention arms found cocoa flavanols raised FMD by 1.17%. The dose-response curve was an inverted U, with the largest effect at about 710 mg of total flavanols. Most of the studies carried a significant risk of bias ^{1Reference 1Sun Y et al. · 2019Meta-analysisDose-response relationship between cocoa flavanols and human endothelial function: a systematic review and meta-analysis of randomised trials — systematic review, meta-analysisView study →}.

In six people with smoking-related endothelial dysfunction, a high-flavanol cocoa drink (3 × 306 mg flavanols a day) raised fasting FMD from 3.7% on day 1 to 6.6% on day 8. It fell back to 3.3% after a week without cocoa. Plasma nitrite rose in step with FMD ^{2Reference 2Heiss C et al. · 2007RCTSustained increase in flow-mediated dilation after daily intake of high-flavanol cocoa drink over 1 week — randomised crossover trialView study →}.

In a randomised, double-masked trial, 100 healthy middle-aged adults took 450 mg of cocoa flavanols twice daily for a month. FMD rose by 1.2% over control, and blood pressure, total and LDL cholesterol and the Framingham Risk Score all fell ^{3Reference 3Sansone R et al. · 2015RCTCocoa flavanol intake improves endothelial function and Framingham Risk Score in healthy men and women: a randomised, controlled, double-masked trialView study →}.

A single 1,350 mg dose of cocoa flavanols raised FMD in the leg (femoral) artery as well as the arm, in both healthy people and people with type 2 diabetes ^{4Reference 4Bapir M et al. · 2022RCTCocoa flavanol consumption improves lower extremity endothelial function in healthy individuals and people with type 2 diabetes — randomised controlled trialView study →}.

Gap: FMD is a surrogate measure rather than a clinical outcome, and the consistent FMD signal has not translated into fewer cardiovascular events ^{15Reference 15Sesso HD et al. · 2022RCTEffect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomised controlled trialView study →}.

The 2017 Cochrane review pooled 35 trials (40 comparisons, 1,804 mostly healthy participants) lasting 2–18 weeks, with 30–1,218 mg of flavanols a day. Systolic and diastolic pressure each fell by 1.76 mmHg against flavanol-free or low-flavanol control ^{5Reference 5Ried K et al. · 2017Meta-analysisEffect of cocoa on blood pressure — Cochrane systematic review and meta-analysisView study →}.

The effect depended on baseline pressure. Systolic pressure fell significantly, by about 4 mmHg, in people with hypertension, but not in people with normal blood pressure ^{5Reference 5Ried K et al. · 2017Meta-analysisEffect of cocoa on blood pressure — Cochrane systematic review and meta-analysisView study →}. An earlier meta-analysis likewise found the fall significant only in hypertensive and prehypertensive groups ^{6Reference 6Ried K et al. · 2010Meta-analysisDoes chocolate reduce blood pressure? A meta-analysis — systematic review, meta-analysisView study →}, and a third, of 10 trials in 297 people, found a mean fall of 4.5/2.5 mmHg ^{7Reference 7Desch S et al. · 2010Meta-analysisEffect of cocoa products on blood pressure: systematic review and meta-analysisView study →}.

Cocoa flavanols are thought to act by increasing nitric oxide formation in the blood-vessel lining, which widens the vessels ^{5Reference 5Ried K et al. · 2017Meta-analysisEffect of cocoa on blood pressure — Cochrane systematic review and meta-analysisView study →}.

Gap: the average fall is small, and outcome trials have not shown that it prevents cardiovascular events ^{15Reference 15Sesso HD et al. · 2022RCTEffect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomised controlled trialView study →}.

A 2016 meta-analysis of 19 randomised trials (1,131 participants, 166–2,110 mg flavanols a day for 2–52 weeks) found cocoa flavanols improved several cardiometabolic markers. Fasting insulin, HOMA-IR, triglycerides and C-reactive protein fell, and HDL cholesterol rose slightly ^{8Reference 8Lin X et al. · 2016Meta-analysisCocoa flavanol intake and biomarkers for cardiometabolic health: a systematic review and meta-analysis of randomised controlled trialsView study →}.

Standard markers of oxidative stress and antioxidant status in the blood did not change after cocoa flavanols in one detailed study ^{2Reference 2Heiss C et al. · 2007RCTSustained increase in flow-mediated dilation after daily intake of high-flavanol cocoa drink over 1 week — randomised crossover trialView study →}.

In the laboratory, bean polyphenol extracts scavenge free radicals ^{26Reference 26Oboh G et al. · 2014In vitroIn vitro studies on the antioxidant property and inhibition of alpha-amylase, alpha-glucosidase, and angiotensin-I-converting enzyme by polyphenol-rich extracts from cocoa (Theobroma cacao) beanView study →}. Among the other plant parts, root extract had the strongest antioxidant activity ^{25Reference 25Baharum Z et al. · 2014In vitroIn vitro antioxidant and antiproliferative activities of methanolic plant part extracts of Theobroma cacaoView study →}.

Gap: the human data are blood biomarkers rather than disease outcomes, and the direct oxidative-stress markers measured in people did not change ^{2Reference 2Heiss C et al. · 2007RCTSustained increase in flow-mediated dilation after daily intake of high-flavanol cocoa drink over 1 week — randomised crossover trialView study →}.

In healthy adults aged 50–69, three months of a high-flavanol diet improved dentate gyrus function on MRI and performance on a memory task linked to it ^{10Reference 10Brickman AM et al. · 2014RCTEnhancing dentate gyrus function with dietary flavanols improves cognition in older adults — randomised controlled trialView study →}.

In 90 cognitively healthy older people, eight weeks of cocoa drinks with 993 or 520 mg of flavanols improved scores on the Trail Making and verbal fluency tests, though not the Mini-Mental State Examination ^{11Reference 11Mastroiacovo D et al. · 2015RCTCocoa flavanol consumption improves cognitive function, blood pressure control, and metabolic profile in elderly subjects: the Cocoa, Cognition, and Aging (CoCoA) randomised trialView study →}.

In 211 adults aged 50–75, twelve weeks of 260–770 mg of flavanols a day did not improve the trial’s primary memory task. A list-learning task did improve, most clearly in people whose baseline diet quality was poorest ^{13Reference 13Sloan RP et al. · 2021RCTInsights into the role of diet and dietary flavanols in cognitive aging: results of a randomised controlled trialView study →}.

The two large COSMOS studies found no overall benefit. In COSMOS-Mind, 2,262 older adults took 500 mg of flavanols a day for three years with no effect on global cognition ^{9Reference 9Baker LD et al. · 2023RCTEffects of cocoa extract and a multivitamin on cognitive function: a randomised clinical trial (COSMOS-Mind)View study →}. In the COSMOS clinic subcohort, 573 people tested in person over two years also showed no benefit, although an uncorrected subgroup analysis suggested one for those with poorer diets ^{28Reference 28Vyas CM et al. · 2024RCTEffect of cocoa extract supplementation on cognitive function: results from the clinic subcohort of the COSMOS randomised trialView study →}.

A systematic review of 12 trials found effects on memory and executive function at 500–750 mg a day, and noted how varied the studies were ^{12Reference 12Barrera-Reyes PK et al. · 2020Systematic reviewEffects of cocoa-derived polyphenols on cognitive function in humans — systematic reviewView study →}.

Gap: the largest trials were null. The positive results cluster in smaller studies and in subgroups with poor baseline diets.

A 2011 meta-analysis of 10 trials (320 participants, 2–12 weeks) found dark chocolate and cocoa products lowered LDL cholesterol by 5.9 mg/dL and total cholesterol by 6.2 mg/dL. HDL cholesterol and triglycerides did not change ^{14Reference 14Tokede OA et al. · 2011Meta-analysisEffects of cocoa products/dark chocolate on serum lipids: a meta-analysisView study →}.

A later meta-analysis of cocoa flavanol trials found a small rise in HDL and a small fall in triglycerides ^{8Reference 8Lin X et al. · 2016Meta-analysisCocoa flavanol intake and biomarkers for cardiometabolic health: a systematic review and meta-analysis of randomised controlled trialsView study →}, and one month of flavanols raised HDL in healthy adults ^{3Reference 3Sansone R et al. · 2015RCTCocoa flavanol intake improves endothelial function and Framingham Risk Score in healthy men and women: a randomised, controlled, double-masked trialView study →}.

Gap: the effects are small, and the trials were short.

COSMOS is the largest trial: a randomised, double-blind, placebo-controlled study of 21,442 older US adults given 500 mg/day cocoa flavanols (including 80 mg epicatechin) or placebo ^{15Reference 15Sesso HD et al. · 2022RCTEffect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomised controlled trialView study →}.

Over a median 3.6 years, the primary composite of total cardiovascular events was not significantly reduced (HR 0.90; 95% CI 0.78–1.02; P = 0.11). A secondary endpoint, cardiovascular death, was 27% lower (HR 0.73; 95% CI 0.54–0.98). A per-protocol analysis that stopped counting at non-adherence found fewer total events (HR 0.85; 95% CI 0.72–0.99) ^{15Reference 15Sesso HD et al. · 2022RCTEffect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomised controlled trialView study →}.

A 2026 reanalysis weighting the events by severity (“win ratio”) found a significant reduction in cardiovascular disease ^{16Reference 16Ogata S et al. · 2026RCTCocoa flavanol supplementation and prevention of cardiovascular disease: a novel analysis of the COSMOS randomised trialView study →}.

Gap: the primary endpoint missed significance. The cardiovascular-death benefit is a secondary finding and needs independent confirmation.

In women drinking high-flavanol cocoa (326 mg flavanols a day) for 12 weeks, skin reddening after a fixed UV dose fell by 15% at six weeks and 25% at twelve. Skin blood flow, density and hydration rose, and roughness fell ^{17Reference 17Heinrich U et al. · 2006RCTLong-term ingestion of high-flavanol cocoa provides photoprotection against UV-induced erythema and improves skin condition in women — randomised controlled trialView study →}.

In 30 people eating 20 g of high-flavanol chocolate daily, the UV dose needed to redden the skin more than doubled in 12 weeks, while conventional chocolate had no effect ^{18Reference 18Williams S et al. · 2009RCTEating chocolate can significantly protect the skin from UV light — randomised controlled trialView study →}.

A larger trial in 74 women eating 30 g of high-flavanol chocolate daily for 12 weeks found no difference from low-flavanol chocolate in UV sensitivity ^{19Reference 19Mogollon JA et al. · 2014RCTChocolate flavanols and skin photoprotection: a parallel, double-blind, randomised clinical trialView study →}.

Gap: the trials are small and their results disagree, with the largest finding no effect.

Pooled trials show cocoa flavanols lower fasting insulin and HOMA-IR ^{8Reference 8Lin X et al. · 2016Meta-analysisCocoa flavanol intake and biomarkers for cardiometabolic health: a systematic review and meta-analysis of randomised controlled trialsView study →}. In the laboratory, bean polyphenol extracts inhibit the starch-digesting enzymes α-amylase and α-glucosidase ^{26Reference 26Oboh G et al. · 2014In vitroIn vitro studies on the antioxidant property and inhibition of alpha-amylase, alpha-glucosidase, and angiotensin-I-converting enzyme by polyphenol-rich extracts from cocoa (Theobroma cacao) beanView study →}.

In COSMOS, among 18,381 people without diabetes at the start, cocoa extract did not reduce the incidence of type 2 diabetes over a median 3.5 years (HR 1.04; 95% CI 0.91–1.20) ^{20Reference 20Li J et al. · 2023RCTCocoa extract supplementation and risk of type 2 diabetes: the Cocoa Supplement and Multivitamin Outcomes Study (COSMOS) randomised controlled trialView study →}.

Gap: the markers improve in short trials, but the one large incidence trial was null.

In 40 young adults, a single 250 mg dose of cocoa flavanols eased self-rated mental fatigue and improved one serial-subtraction task. Four weeks of daily use showed no effect ^{21Reference 21Massee LA et al. · 2015RCTThe acute and sub-chronic effects of cocoa flavanols on mood, cognitive and cardiovascular health in young healthy adults — randomised controlled trialView study →}.

In a pilot trial in 60 women aged 40–60 who reported fatigue, 240 mg of cacao flavanols a day for eight weeks lowered negative mood scores and raised vigour compared with placebo. The fatigue scale did not change ^{22Reference 22Murakami R et al. · 2023RCTEffect of flavanol-rich cacao extract on the profile of mood state in healthy middle-aged Japanese women: a randomised controlled trialView study →}.

Gap: the trials are few and small, and the clearest effects are acute.

Theobromine suppressed citric-acid-induced cough in guinea pigs. In a randomised, double-blind, placebo-controlled human study, it suppressed capsaicin-induced cough with no adverse effects ^{23Reference 23Usmani OS et al. · 2005RCTTheobromine inhibits sensory nerve activation and cough — randomised controlled trialView study →}.

It also blocked capsaicin-induced activation of guinea-pig and human vagus nerve, which points to an action on the sensory nerves rather than the brain ^{23Reference 23Usmani OS et al. · 2005RCTTheobromine inhibits sensory nerve activation and cough — randomised controlled trialView study →}.

Gap: the study used isolated theobromine, not cacao, and no trial has tested whether cocoa or chocolate reduces cough.

In stimulated human blood immune cells, aqueous and ethanolic cacao extracts suppressed tryptophan breakdown by indoleamine-2,3-dioxygenase (IDO) and the formation of neopterin and interferon-γ, in a dose-dependent way. The effect tracked the extracts’ antioxidant capacity ^{24Reference 24Becker K et al. · 2013In vitroImmunomodulatory properties of cacao extracts — potential consequences for medical applications — in vitro studyView study →}.

Gap: this is cell-based work; no human immune outcome has been measured.

In a single cell study, methanolic extracts of eight cacao plant parts were tested against cancer cell lines. Leaf extract was the most active against MCF-7 breast-cancer cells and had little effect on normal liver cells. The leaf extract contained flavonoids, saponins, condensed tannins and triterpenes, and no alkaloids ^{25Reference 25Baharum Z et al. · 2014In vitroIn vitro antioxidant and antiproliferative activities of methanolic plant part extracts of Theobroma cacaoView study →}.

Gap: this is one cell study of the leaf, not the seed, with no animal or human work.

Side effects. Theobromine and a smaller amount of caffeine can cause insomnia, restlessness, headache or palpitations when cacao is taken late in the day or in large amounts.

Stimulant and heart-rhythm drugs. Theobromine stimulates heart muscle ^{35Reference 35Giovanni Scapagnini et al. · 2014Cocoa Bioactive Compounds: Significance and Potential for the Maintenance of Skin Health}, so people on stimulant or heart-rhythm medication should use cacao with care. The interaction is inferred from the pharmacology of theobromine and caffeine, not measured.

Blood-pressure medicines. Cocoa flavanols lower blood pressure by a small margin, mainly in people with hypertension ^{5Reference 5Ried K et al. · 2017Meta-analysisEffect of cocoa on blood pressure — Cochrane systematic review and meta-analysisView study →}, so a small added effect with antihypertensive drugs is plausible. It has not been studied.

Drug metabolism. No herb–drug interaction study of cocoa exists. Its theobromine and caffeine are cleared mainly by the liver, and their clearance changes with drugs that induce or inhibit CYP1A2 and with smoking. Pregnancy and oral contraceptives slow it ^{44Reference 44Arnaud MJ · 2011ReviewPharmacokinetics and metabolism of natural methylxanthines in animal and man — reviewView study →}.

The cocoa flavanols are handled mainly by conjugation and by gut bacteria ^{41Reference 41Di Pede G et al. · 2023Systematic reviewRevisiting the bioavailability of flavan-3-ols in humans: A systematic review and comprehensive data analysis — systematic reviewView study →}. In one small study, a chocolate matrix lowered the absorption of the galloylated catechins in green tea taken with it ^{45Reference 45Mukai R et al. · 2021Clinical trialChocolate as a food matrix reduces the bioavailability of galloylated catechins from green tea in healthy women — clinical trialView study →}.

Allergy and migraine. Chocolate can cause allergies and migraine in some people ^{38Reference 38Fern · 2014Theobroma cacao — habit, range, habitat, cultivation and usesView study →}. Anyone with a known sensitivity to cocoa or chocolate should avoid it.

Cadmium and lead. Cocoa and chocolate can contain cadmium and lead from the environment. In US retail products, cadmium ranged from 0.004 to 3.15 mg/kg and lead up to 0.38 mg/kg. Both rose with the percentage of cocoa solids, and cadmium was higher in products from Latin America than from Africa ^{39Reference 39Abt et al. · 2018Cadmium and lead in cocoa powder and chocolate products in the US market — analytical surveyView study →}. For people who eat a lot of high-percentage products, it is worth choosing brands that test for heavy metals.