Is Dark Chocolate Healthy? Cocoa, Flavanols and Metals
TL;DR
- A 2026 meta-analysis of 25 randomized trials in 1,167 people found cocoa flavanols lowered blood pressure by an average of 2.7 mmHg systolic and 1.8 mmHg diastolic, with measurable benefits in people with normal blood pressure as well.
- In the COSMOS randomized trial, 21,442 older adults taking a standardized 500 mg cocoa-flavanol extract daily had 27% lower cardiovascular deaths, while the trial's original main predefined outcome, which combined cardiovascular events, did not reach statistical significance.
- A 90–100% cocoa label combines cocoa butter with nonfat cocoa solids and carries no flavanol measurement: in a laboratory analysis, average flavanols fell from 34.6 mg/g in natural cocoa powder to 3.9 mg/g in heavily alkalized cocoa, roughly one-ninth of the natural average.
- Heavy metals are a separate measurement: in a 2024 US analysis, 43% of 72 cocoa and chocolate products exceeded California's precautionary daily lead threshold, and 35% of products with comparable data exceeded the cadmium threshold.
- Measured flavanols per serving plus separate cadmium and lead results describe a cocoa product better than its percentage or organic label, and a 5–10 g serving is a practical range for comparing product reports.
Key facts
- 25 trials, 2.7 mmHg systolic and 1.8 mmHg diastolic: A 2026 systematic review and meta-analysis of 25 randomized trials involving 1,167 people found average blood pressure reductions of 2.7 mmHg systolic and 1.8 mmHg diastolic from cocoa flavanols.
- COSMOS, 27% lower cardiovascular deaths: 21,442 older adults randomized to a standardized cocoa extract supplying 500 mg of cocoa flavanols daily had 27% lower cardiovascular deaths, while the original main predefined combined cardiovascular outcome did not reach statistical significance.
- 34.6 mg/g versus 3.9 mg/g: Heavy alkalization cut the average total flavanol concentration of cocoa powder from 34.6 mg/g (natural) to 3.9 mg/g (heavily alkalized), roughly one-ninth of the natural average.
- 43% lead, 35% cadmium: In a 2024 US investigation, 43% of 72 cocoa and chocolate products exceeded California's precautionary daily lead threshold and 35% of the 37 products with comparable cadmium data exceeded that threshold.
Cocoa flavanols, plant compounds in cocoa, can improve blood pressure and blood-vessel function. A 90–100% cocoa label tells you how much of a chocolate bar comes from cocoa ingredients, while processing and composition determine its flavanol content. For someone seeking those benefits, measured flavanols per serving and separate cadmium and lead results are more informative than cocoa percentage alone (Sakuda et al., 2026; PMID: 42679502; Langer et al., 2011; PMID: 21699218; Hands et al., 2024; PMID: 39144282).
Dark chocolate is often called heart-healthy, especially at 90–100% cocoa. The percentage can look like a simple way to choose. Yet it combines cocoa-derived ingredients with different compositions, and even organic cocoa can contain substantial heavy metals (Langer et al., 2011; PMID: 21699218; Hands et al., 2024; PMID: 39144282).
Is dark chocolate healthy for your heart?
Cocoa flavanols help blood vessels widen and respond to blood flow. Controlled trials support modest improvements in blood pressure, blood-vessel function and some metabolic markers from tested cocoa exposures. To apply those results to a product, the flavanol amount and the outcome measured matter (Sakuda et al., 2026; PMID: 42679502; Arisi et al., 2024; PMID: 38931273).
- Blood pressure. A 2026 systematic review and meta-analysis of 25 randomized trials involving 1,167 people found average reductions of 2.7 mmHg systolic and 1.8 mmHg diastolic. Systolic is the pressure as the heart contracts, and diastolic is the pressure between beats. Benefits were measurable in people with normal blood pressure as well (Sakuda et al., 2026; PMID: 42679502).
- Metabolic markers. A 2024 meta-analysis of 31 trials found average reductions of 9.5 mg/dL in LDL cholesterol and 4.9 mg/dL in fasting blood glucose. Body weight and long-term blood sugar control did not improve significantly (Arisi et al., 2024; PMID: 38931273).
- Cardiovascular events. COSMOS randomized 21,442 older adults to a standardized cocoa extract or placebo. The extract supplied 500 mg of cocoa flavanols daily. Cardiovascular deaths were 27% lower in the extract group, while the original main predefined outcome, which combined cardiovascular events, did not reach statistical significance (Sesso et al., 2022; PMID: 35294962).
- Longer-term interpretation. A 2026 reanalysis of the same COSMOS participants found a significant cardiovascular benefit when the comparison prioritized more serious events. This method compares participants using a severity hierarchy, giving fatal events priority over less serious events. It asks a different statistical question from the original analysis of time to a first cardiovascular event. In a separate COSMOS analysis of 18,381 adults without diabetes at baseline, extract use did not significantly reduce new type 2 diabetes diagnoses over a median 3.5 years (Ogata et al., 2026; PMID: 41849087; Li et al., 2023; PMID: 37816167).
These results establish a case for the tested compounds and exposures. The COSMOS dose came from an extract, and a supermarket chocolate label gives no corresponding flavanol measurement.
What does 90–100% cocoa percentage tell you?
Cocoa percentage includes cocoa butter as well as the nonfat cocoa solids that contain most of the flavanols. A 99% or 100% bar can therefore contain substantial cocoa butter without telling you how many flavanols it supplies (Langer et al., 2011; PMID: 21699218; Barišić et al., 2019; PMID: 31480281).
The product forms have different uses and compositions:
- Unsweetened cocoa powder. Pressing cocoa mass removes much of its natural fat. The milled remainder retains the nonfat fraction containing most flavanols, minerals and theobromine. A reduced-fat label describes that pressing process (Barišić et al., 2019; PMID: 31480281; Risner, 2008; PMID: 19007497).
- Dark chocolate. The bar contains cocoa solids and cocoa butter. The percentage describes the amount of cocoa-derived ingredients, with no measurement of the flavanols that survived processing (Langer et al., 2011; PMID: 21699218).
- Instant cocoa mix. The cocoa proportion determines how much cocoa enters the prepared drink. For example, 20 g of a mix containing 20% cocoa provides 4 g of cocoa ingredients. Sugar does not itself destroy cocoa flavanols. The amount of actual cocoa and its processing history are more useful for assessing flavanol delivery (Belščak-Cvitanović et al., 2010; PMID: 20509612; Miller et al., 2008; PMID: 18710243).
Sugar, milk powder, flavorings and emulsifiers also change the composition of a mix. Their presence alone gives no flavanol dose.
How does Dutch processing change cocoa flavanols?
Alkalization, also called Dutch processing, reduces bitterness and darkens cocoa while substantially reducing its flavanols. In a 2008 laboratory analysis, the average concentrations varied with the degree of alkalization (Miller et al., 2008; PMID: 18710243).
- Natural cocoa powder: 34.6 mg/g average total flavanols.
- Lightly alkalized: 13.8 mg/g average total flavanols.
- Moderately alkalized: 7.8 mg/g average total flavanols.
- Heavily alkalized: 3.9 mg/g average total flavanols.
The heavily alkalized average was roughly one-ninth of the natural cocoa average. Based on those measurements, 5 g would supply about 170 mg versus 20 mg. Those are calculations from the tested averages (Miller et al., 2008; PMID: 18710243).
A 2024 investigation of 59 European cocoa powders also identified alkalization as an important factor in flavanols and related plant compounds, compared with organic status or roasting. Non-alkalized processing is useful information about flavanol preservation. A raw, organic or unfiltered label leaves the amount unmeasured, and darker cocoa may have lost more flavanols during processing (Esposito et al., 2024; PMID: 39123648).
What makes a high-flavanol claim informative?
A measured amount per serving, together with the analytical method, makes a flavanol claim assessable. Validated methods measure cocoa flavanols and procyanidins, related compounds made of linked flavanol units. Older total-flavanol results may cover a different set of compounds from modern standardized assays, so the method matters when comparing figures (Bussy et al., 2021; PMID: 33313842; Bussy et al., 2022; PMID: 35080238).
The US FDA allows narrowly qualified cardiovascular claims for certain high-flavanol cocoa powders and foods containing them. The specified powder contains at least 4% naturally conserved cocoa flavanols. At that concentration, 5 g supplies 200 mg. The FDA describes the evidence for cardiovascular disease risk reduction as very limited, and its claim applies to the qualifying powder and foods containing it. FDA qualified-claim notice
Europe's authorized wording concerns maintaining blood-vessel elasticity and normal blood flow, a different outcome from reducing cardiovascular disease risk. The wording and conditions of a claim tell you what is being asserted. European Commission regulation on cocoa flavanol claims
Generic antioxidant scores describe how a sample behaves in a laboratory assay. For comparing cocoa products, a report that identifies the flavanol amount, method and tested formulation or batch is more useful.
How much lead and cadmium can cocoa contain?
Concentrations vary by product. Cacao can take up cadmium from soil, and cocoa can acquire lead during production and handling. Surveys have measured both metals in cocoa powder and chocolate (Vanderschueren et al., 2023; PMID: 36708823; Rankin et al., 2005; PMID: 16203244; Abt et al., 2018; PMID: 29310543).
In a 2024 US investigation:
- 43% of 72 cocoa and chocolate products exceeded California's precautionary daily lead threshold.
- 35% of the 37 products with comparable daily cadmium data exceeded the corresponding cadmium threshold.
Those percentages use different denominators and concern daily exposure thresholds for the tested servings (Hands et al., 2024; PMID: 39144282).
Organic certification concerns production standards. It does not establish a product's cadmium or lead concentration, and neither does a raw label, premium price or high cocoa percentage (Hands et al., 2024; PMID: 39144282; Esposito et al., 2024; PMID: 39123648). The distinction also matters when interpreting organic and conventional food labels.
How do you calculate cadmium exposure from a serving?
Multiply the measured concentration by the serving size, then account for how often the product is consumed. One mg/kg equals one microgram per gram. For a powder containing 0.2 mg/kg cadmium, the arithmetic is:
- 5 g serving: 1 microgram of cadmium ingested.
- 10 g serving: 2 micrograms of cadmium ingested.
- 20 g serving: 4 micrograms of cadmium ingested.
At 10 g per day, that powder contributes 14 micrograms weekly. For a 70 kg adult, this is 8% of Europe's tolerable weekly cadmium intake of 175 micrograms from the entire diet (EFSA CONTAM Panel, 2011; PMID: 42006064).
EU rules set a maximum of 0.6 mg/kg cadmium for consumer cocoa powder. At that upper concentration, the same 10 g daily contributes 6 micrograms a day, or 42 micrograms weekly. That is 24% of the 70 kg adult's tolerable weekly intake, before cadmium from other foods is counted (EFSA CONTAM Panel, 2011; PMID: 42006064). European Commission Regulation 2023/915, Annex I, cocoa powder
The example describes cocoa at the regulatory maximum. Actual exposure depends on the measured product and the amount consumed. Cadmium accumulates slowly, particularly in the kidneys, so repeated intake over time is relevant. Lead needs a separate result, with particular importance for children and pregnancy because no harmless exposure level has been established (EFSA CONTAM Panel, 2011; PMID: 42006064; Lanphear et al., 2005; PMID: 16002379). WHO guidance on lead exposure
Can milk or mineral status change cadmium absorption?
The amount swallowed and the amount absorbed are different measurements. Nutritional status can influence cadmium uptake, with stronger human evidence for iron status than for calcium. Calcium deficiency may also increase uptake. Experiments in human intestinal cells found reduced uptake under some milk-exposure conditions (Peng et al., 2023; PMID: 37060803; Mata et al., 1996; PMID: 8653809).
A human feeding trial has yet to quantify the absorbed cadmium dose from ordinary cocoa mixed with milk. Milk remains a useful source of calcium and other nutrients. The cocoa's measured concentration tells you how much cadmium entered the drink, and the cell result leaves the human absorbed amount open.
Does milk block the absorption of cocoa flavanols?
Small human experiments found broadly similar circulating concentrations in their tested settings, alongside some differences in how the body handled flavanol metabolites, the compounds formed as it processes flavanols. Measuring what appears in blood and what leaves in urine gives different information.
- Cocoa in whole milk. In a randomized crossover experiment in 21 adults, researchers compared 40 g of cocoa powder in whole milk with cocoa in water. The blood concentration of a compound formed when the body processes epicatechin, a cocoa flavanol, was numerically lower with milk. The difference was not statistically significant (Roura et al., 2007; PMID: 18032884).
- Blood and urine. A nine-person experiment found little change in plasma kinetics, the pattern of concentrations in blood over time. Urinary recovery of four flavanol metabolites fell from 18.3% to 10.5% of the ingested dose (Mullen et al., 2009; PMID: 19403635).
- Added milk protein. Another human experiment found comparable average circulating cocoa polyphenol concentrations with added milk protein (Keogh et al., 2007; PMID: 17995820).
Some experimental portions were much larger than an ordinary cup of hot cocoa.
What does milk add to the drink?
In an everyday drink, milk also contributes its own fluid, complete protein, calcium, potassium and carbohydrate. Its hydration properties have been studied separately. Controlled beverage research found greater short-term fluid retention with milk than water, while tea hydrated comparably to water with a different mineral profile (Maughan et al., 2016; PMID: 26702122). A 2023 randomized crossover trial in 12 younger and 11 older men compared beverages consumed in 1 L amounts. Over three hours after drinking, low-fat milk produced greater fluid retention than water in the younger men. That advantage was not statistically significant in the older men (Rodriguez-Sanchez & Galloway, 2023; PMID: 37294517).
What does cocoa butter contribute?
Cocoa butter supplies a fat profile that is predominantly saturated and monounsaturated. It is a separate fraction from the cocoa solids that concentrate most flavanols and methylxanthines, including caffeine and theobromine (Barišić et al., 2019; PMID: 31480281; Risner, 2008; PMID: 19007497).
- Stearic acid: 33–37% approximate share, saturated.
- Palmitic acid: 24–27% approximate share, saturated.
- Oleic acid: 33–38% approximate share, monounsaturated.
- Linoleic acid: 2–4% approximate share, omega-6 polyunsaturated.
Overall, cocoa butter is about 60% saturated fat, one-third monounsaturated fat and around 3% polyunsaturated fat (PUFA). That low-PUFA composition helps this culinary fat resist oxidative breakdown (Barišić et al., 2019; PMID: 31480281).
Different saturated fats can produce different cholesterol responses. In a randomized crossover experiment in 10 men, each liquid-formula diet featuring a different fat lasted three weeks. Cocoa butter produced lower LDL cholesterol than dairy butter, and olive oil produced the lowest LDL among the tested fats. These were responses to the compared diets and the measured LDL endpoint (Denke & Grundy, 1991; PMID: 1845606).
What else is in cocoa powder?
Calculated from the USDA composition entry for unsweetened cocoa powder, a 5 g serving supplies about 100 mg theobromine, 12 mg caffeine and 25 mg magnesium. USDA food-composition table, entry 19165
Analytical studies also measured methylxanthines and minerals in cocoa samples, with concentrations varying across the tested products (Risner, 2008; PMID: 19007497; Silva et al., 2026; PMID: 42178899).
Theobromine and caffeine are methylxanthines. Four controlled human experiments involving 47 volunteers found greater short-term widening of blood vessels when flavanols were combined with methylxanthines than when flavanols were given alone. The mixture may be relevant to cocoa's distinctive cardiovascular activity (Sansone et al., 2017; PMID: 28003203).
The lower esophageal sphincter is the muscular valve between the esophagus and stomach. A 1975 experiment in nine healthy volunteers found reduced pressure in that valve after 120 mL of chocolate syrup. A 1988 experiment in patients with esophagitis found greater esophageal acid exposure after chocolate than after a matched control drink (Wright & Castell, 1975; PMID: 239592; Murphy & Castell, 1988; PMID: 3376917). For someone whose symptoms follow an evening serving, the timing provides individual context. It also matters when considering the caffeine in a late drink.
What is useful to check before choosing cocoa?
For someone assessing a cocoa product, flavanols per serving describe its cocoa-compound content, while lead and cadmium concentrations describe metal exposure. The measurements answer separate parts of the decision.
A useful independent laboratory report identifies:
- The finished product. Powder, a drink mix or chocolate tested in the form being assessed.
- The batch. A batch-linked result gives information about that production run.
- Numerical results. Actual concentrations allow serving-size calculations.
- Separate assays. Flavanols, lead and cadmium require their own measurements.
- The flavanol method. A validated assay makes the reported compound amount interpretable (Bussy et al., 2021; PMID: 33313842; Bussy et al., 2022; PMID: 35080238; Hands et al., 2024; PMID: 39144282).
A cocoa can be high in flavanols and also contain substantial cadmium. Non-alkalized processing helps preserve flavanols, while a metals result answers a separate question (Miller et al., 2008; PMID: 18710243; Hands et al., 2024; PMID: 39144282).
For example, if a powder report showed 4% flavanols and 0.2 mg/kg cadmium, a 5 g serving would supply 200 mg flavanols and 1 microgram cadmium. Lead would still require a separate numerical result. The same serving has a separate calculation for each measured compound.
For someone who enjoys hot cocoa, a modest amount of well-characterized powder in warm milk can add flavor and cocoa compounds to a nutritious drink. A 5–10 g portion is a useful range for comparing product reports and serving calculations. The clinical studies used particular formulations and exposures, so that range describes a culinary comparison.
Black and green tea also supply flavanols. Citrus fruits and orange juice provide different polyphenols, including flavanones, alongside vitamin C, potassium and carbohydrate (Tao et al., 2016; PMID: 27589031; Morand et al., 2011; PMID: 21068346).
Summary: cocoa quality at a glance
- What did the cardiovascular studies test? Specific cocoa exposures, including standardized extract in COSMOS. Blood pressure, metabolic markers and clinical events are different outcomes.
- What does cocoa percentage show? The share of cocoa-derived ingredients, including cocoa butter.
- What predicts flavanol preservation? Processing, especially alkalization. Measurement establishes the amount.
- What determines metal exposure? Product concentration, serving size and frequency. Lead and cadmium need separate results.
- How does cocoa fit into a meal? Modest culinary cocoa in milk combines cocoa compounds with milk's fluid, protein and nutrients.
Frequently asked questions
Is 100% dark chocolate better for your heart?
A 100% cocoa label does not quantify flavanols or contaminant levels. It includes cocoa butter as well as nonfat cocoa solids, and processing affects the amount of flavanols that remains (Langer et al., 2011; PMID: 21699218; Miller et al., 2008; PMID: 18710243).
Does sugar in a cocoa mix destroy flavanols?
Sugar does not itself destroy them. The cocoa proportion and whether the cocoa was alkalized tell you more about the mix's likely flavanol delivery (Belščak-Cvitanović et al., 2010; PMID: 20509612; Miller et al., 2008; PMID: 18710243).
Does organic cocoa have less lead or cadmium?
Organic certification does not establish the amount of either metal. Numerical testing of the finished product gives the concentration needed to assess exposure (Hands et al., 2024; PMID: 39144282).
Is hot cocoa with milk a way to obtain cocoa compounds?
Milk supplies fluid, protein, minerals and carbohydrate alongside cocoa's compounds. Small human studies found broadly similar circulating flavanol measurements in their tested settings, with some differences in metabolite handling (Roura et al., 2007; PMID: 18032884; Mullen et al., 2009; PMID: 19403635; Keogh et al., 2007; PMID: 17995820).
How much cocoa powder should a serving contain?
The studies establish no required everyday serving. Five to 10 g is a practical comparison range for culinary use, with the measured flavanol and metal concentrations determining what that portion supplies.
References
- Abt E et al. (2018). Cadmium and lead in cocoa powder and chocolate products in the US Market. Food Additives & Contaminants: Part B, Surveillance, 11(2), 92–102. PMID: 29310543.
- Arisi TOP et al. (2024). Effects of Cocoa Consumption on Cardiometabolic Risk Markers: Meta-Analysis of Randomized Controlled Trials. Nutrients, 16(12), 1919. PMID: 38931273.
- Barišić V et al. (2019). The Chemistry behind Chocolate Production. Molecules, 24(17), 3163. PMID: 31480281.
- Belščak-Cvitanović A et al. (2010). Physical properties and bioactive constituents of powdered mixtures and drinks prepared with cocoa and various sweeteners. Journal of Agricultural and Food Chemistry, 58(12), 7187–7195. PMID: 20509612.
- Bussy U et al. (2021). Single-Laboratory Validation for the Determination of Cocoa Flavanols and Procyanidins (by Degree of Polymerization DP1-7) in Cocoa-Based Products by Hydrophilic Interaction Chromatography Coupled with Fluorescence Detection: First Action 2020.05. Journal of AOAC International, 104(2), 413–421. PMID: 33313842.
- Bussy U et al. (2022). Determination of Cocoa Flavanols and Procyanidins (by Degree of Polymerization DP1-7) in Cocoa-Based Products by Hydrophilic Interaction Chromatography Coupled With Fluorescence Detection: Collaborative Study. Journal of AOAC International, 105(4), 1060–1068. PMID: 35080238.
- Denke MA, Grundy SM. (1991). Effects of fats high in stearic acid on lipid and lipoprotein concentrations in men. The American Journal of Clinical Nutrition, 54(6), 1036–1040. PMID: 1845606.
- EFSA Panel on Contaminants in the Food Chain (CONTAM). (2011). Statement on tolerable weekly intake for cadmium. EFSA Journal, 9(2), 1975. PMID: 42006064.
- Esposito L et al. (2024). A Survey on Potentially Beneficial and Hazardous Bioactive Compounds in Cocoa Powder Samples Sourced from the European Market. Foods, 13(15), 2457. PMID: 39123648.
- Hands JM et al. (2024). A multi-year heavy metal analysis of 72 dark chocolate and cocoa products in the USA. Frontiers in Nutrition, 11, 1366231. PMID: 39144282.
- Keogh JB et al. (2007). The effect of milk protein on the bioavailability of cocoa polyphenols. Journal of Food Science, 72(3), S230–S233. PMID: 17995820.
- Langer S et al. (2011). Flavanols and methylxanthines in commercially available dark chocolate: a study of the correlation with nonfat cocoa solids. Journal of Agricultural and Food Chemistry, 59(15), 8435–8441. PMID: 21699218.
- Lanphear BP et al. (2005). Low-level environmental lead exposure and children's intellectual function: an international pooled analysis. Environmental Health Perspectives, 113(7), 894–899. PMID: 16002379.
- Li J et al. (2023). Cocoa Extract Supplementation and Risk of Type 2 Diabetes: The Cocoa Supplement and Multivitamin Outcomes Study (COSMOS) Randomized Clinical Trial. Diabetes Care, 46(12), 2278–2284. PMID: 37816167.
- Mata L et al. (1996). Cadmium uptake by Caco-2 cells. Effect of some milk components. Chemico-Biological Interactions, 100(3), 277–288. PMID: 8653809.
- Maughan RJ et al. (2016). A randomized trial to assess the potential of different beverages to affect hydration status: development of a beverage hydration index. The American Journal of Clinical Nutrition, 103(3), 717–723. PMID: 26702122.
- Miller KB et al. (2008). Impact of alkalization on the antioxidant and flavanol content of commercial cocoa powders. Journal of Agricultural and Food Chemistry, 56(18), 8527–8533. PMID: 18710243.
- Morand C et al. (2011). Hesperidin contributes to the vascular protective effects of orange juice: a randomized crossover study in healthy volunteers. The American Journal of Clinical Nutrition, 93(1), 73–80. PMID: 21068346.
- Mullen W et al. (2009). Milk decreases urinary excretion but not plasma pharmacokinetics of cocoa flavan-3-ol metabolites in humans. The American Journal of Clinical Nutrition, 89(6), 1784–1791. PMID: 19403635.
- Murphy DW, Castell DO. (1988). Chocolate and heartburn: evidence of increased esophageal acid exposure after chocolate ingestion. The American Journal of Gastroenterology, 83(6), 633–636. PMID: 3376917.
- Ogata S et al. (2026). Cocoa flavanol supplementation and prevention of cardiovascular disease: a novel analysis of the COSMOS randomized trial using "win ratio". European Journal of Epidemiology, 41(5), 599–609. PMID: 41849087.
- Peng X et al. (2023). Associations of micronutrients exposure with cadmium body burden among population: A systematic review. Ecotoxicology and Environmental Safety, 256, 114878. PMID: 37060803.
- Rankin CW et al. (2005). Lead contamination in cocoa and cocoa products: isotopic evidence of global contamination. Environmental Health Perspectives, 113(10), 1344–1348. PMID: 16203244.
- Risner CH. (2008). Simultaneous determination of theobromine, (+)-catechin, caffeine, and (-)-epicatechin in standard reference material baking chocolate 2384, cocoa, cocoa beans, and cocoa butter. Journal of Chromatographic Science, 46(10), 892–899. PMID: 19007497.
- Rodriguez-Sanchez N, Galloway SDR. (2023). A randomised trial to assess fluid and electrolyte balance responses following ingestion of different beverages in young and older men. European Journal of Applied Physiology, 123(10), 2331–2340. PMID: 37294517.
- Roura E et al. (2007). Milk does not affect the bioavailability of cocoa powder flavonoid in healthy human. Annals of Nutrition & Metabolism, 51(6), 493–498. PMID: 18032884.
- Sakuda M et al. (2026). Impact of cocoa flavanols on blood pressure and vascular function: an updated systematic review and meta-analysis of randomized controlled trials. The Journal of Nutrition, Health & Aging, 30(11), 100966. PMID: 42679502.
- Sansone R et al. (2017). Methylxanthines enhance the effects of cocoa flavanols on cardiovascular function: randomized, double-masked controlled studies. The American Journal of Clinical Nutrition, 105(2), 352–360. PMID: 28003203.
- Sesso HD et al. (2022). Effect of cocoa flavanol supplementation for the prevention of cardiovascular disease events: the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomized clinical trial. The American Journal of Clinical Nutrition, 115(6), 1490–1500. PMID: 35294962.
- Silva MB et al. (2026). Mineral profile of cocoa powder: Analytical approach in commercial samples. Journal of the Science of Food and Agriculture, 106(12), 7056–7065. PMID: 42178899.
- Tao W et al. (2016). Simultaneous determination of eight catechins and four theaflavins in green, black and oolong tea using new HPLC-MS-MS method. Journal of Pharmaceutical and Biomedical Analysis, 131, 140–145. PMID: 27589031.
- Vanderschueren R et al. (2023). Revealing the pathways of cadmium uptake and translocation in cacao trees (Theobroma cacao L.): A 108Cd pulse-chase experiment. Science of the Total Environment, 869, 161816. PMID: 36708823.
- Wright LE, Castell DO. (1975). The adverse effect of chocolate on lower esophageal sphincter pressure. The American Journal of Digestive Diseases, 20(8), 703–707. PMID: 239592.
- European Commission. (2015). Commission Regulation (EU) 2015/539, amending the cocoa-flavanol health claim in Regulation (EU) No 432/2012. Official text.
- European Commission. (2023). Commission Regulation (EU) 2023/915 on maximum levels for certain contaminants in food. Annex I, section 3.2.15.4. Official text.
- US Food and Drug Administration. (2023). Qualified health claim for cocoa flavanols and cardiovascular disease. Official notice.
- US Department of Agriculture, Agricultural Research Service. (2015). National Nutrient Database for Standard Reference, Release 28. Cocoa, dry powder, unsweetened, NDB 19165. Official composition table.
- World Health Organization. (2026). Lead poisoning. Official fact sheet.