asxan.ai Agents Hub

Cocoa Extract

Evidence Fact Sheet

Cocoa Flavanols

Cocoa extract (cocoa flavanols, e.g. (-)-epicatechin) is a flavanol-standardized Theobroma cacao extract studied for endothelial/eNOS-NO signaling, blood pressure within the normal range, cognition and cardiometabolic markers. Evidence includes meta-analyses and the large COSMOS RCT, whose primary cardiovascular endpoint was null. Research dose 200-900 mg flavanols/day; EU-authorized vascular claim ≥200 mg/day.

Also known as: Cocoa flavanols · Theobroma cacao extract · (-)-Epicatechin (representative flavanol)

Overview

Cocoa extract is a standardized extract of Theobroma cacao concentrating cocoa flavanols (notably (-)-epicatechin), the bioactives behind the eNOS/nitric-oxide pathway that supports endothelium-dependent vasodilation; theobromine and gut-microbiota and antioxidant mechanisms are also studied. Because flavanol content varies enormously with processing (Dutch/alkalized cocoa can lose much of its flavanols), research and labels rely on a standardized flavanol figure rather than cocoa mass. Typical research doses are 200-900 mg cocoa flavanols/day. It is a lawful dietary-supplement ingredient under DSHEA (US), carries one EFSA-authorized Article 13.5 claim that cocoa flavanols help maintain blood-vessel elasticity at a condition of >=200 mg flavanols/day (EU), and is a permitted supplement ingredient in Brazil (ANVISA) and China. This page reports research findings in studied populations and is not medical advice.

Mechanism of Action

eNOS/NO pathway activation supporting endothelium-dependent vasodilation (research context) · Flavanol antioxidant activity · BDNF / neuroplasticity signaling (research context) · Gut-microbiota modulation · Phosphodiesterase inhibition via theobromine

Body systems: Cardiovascular · Neurological & Cognitive · Mood & Stress Response · Digestive & Gut · Skin & Connective Tissue · Blood & Hematopoiesis

Evidence-Based Benefits

Each benefit below is anchored to a specific PubMed-indexed study. Effect sizes, sample sizes, and p-values are reported as published; no values are inferred. Honest negatives and null results are kept alongside the positive findings, and disease-research populations are described as such — Cocoa Extract is not characterized as a treatment for any disease.

Blood pressure within the normal range

Meta-analysis supported

In a Cochrane meta-analysis of mainly healthy adults, flavanol-rich cocoa products produced a small but statistically significant reduction in both systolic and diastolic blood pressure over the short term. The effect was larger in hypertensive participants (about -4 mmHg systolic) and was not significant in normotensive people, indicating the BP-lowering is concentrated in those with elevated pressure.

Effect size: not quantified on this page — see the linked study below for the reported figures.

Source: PMID 28439881 · Ried 2017 · Cochrane Database Syst Rev

Cardiovascular events (large RCT primary endpoint)

Null / no benefit RCT supported
  • HR 0.90total CVD · 95% CI 0.78-1.02 · P=0.11
  • 21,442participants · 3.6 y
  • HR 0.73CVD death · secondary · CI 0.54-0.98

COSMOS is the largest randomized trial of cocoa extract (500 mg flavanols/day). The prespecified primary outcome -- total cardiovascular events -- was not significantly reduced (HR 0.90, confidence interval crossing 1.0, P=0.11), an honest null result. A secondary cardiovascular-death endpoint was lower in the cocoa group, but secondary findings are hypothesis-generating only. This trial also established large-scale safety with no serious safety signal.

Reported effect: 410 participants taking cocoa extract and 456 taking placebo had confirmed total cardiovascular events (HR: 0.90; 95% CI: 0.78, 1.02; P = 0.11) among 21,442 adults over a median 3.6 years; CVD death HR 0.73 (95% CI: 0.54, 0.98).

“410 participants taking cocoa extract and 456 taking placebo had confirmed total cardiovascular events (HR: 0.90; 95% CI: 0.78, 1.02; P = 0.11) ... HRs were 0.73 (95% CI: 0.54, 0.98) for CVD death ... 21,442 US adults (12,666 women aged >=65 y and 8776 men aged >=60 y).”

Source: PMID 35294962 · Sesso 2022 · Am J Clin Nutr

Endothelial function (flow-mediated dilation)

Meta-analysis supported
  • 0.69FMD increase · dark chocolate · CI 0.22-1.16
  • 1.16FMD increase · flavan-3-ols · CI 0.52-1.80
  • 17 studies615 participants

In a meta-analysis of randomized trials, chronic consumption of dark chocolate and of flavan-3-ols both significantly increased flow-mediated dilation, a marker of endothelium-dependent vasodilation consistent with the eNOS/nitric-oxide mechanism. Acute consumption was also beneficial, and dose-response analysis suggested the largest FMD response around 40-60 g/day.

Reported effect: Chronic consumption of dark chocolate and flavan-3-ols both increased FMD (0.69, 95% CI 0.22-1.16, p < 0.001, and 1.16, 95% CI 0.52-1.80, p = 0.001, respectively); 17 studies, 615 participants.

“Seventeen studies involving 615 participants assessed the effect of dark chocolate ... Chronic consumption of dark chocolate and flavan-3-ols both increased FMD (0.69, 95% CI 0.22-1.16, p < 0.001, 1.16, 95% CI 0.52-1.80 p = 0.001, respectively).”

Source: PMID 32220363 · Ebaditabar 2020 · Clin Nutr ESPEN

Cardiometabolic biomarkers (insulin & lipids)

Meta-analysis supported
  • -0.93HOMA-IR · CI -1.31 to -0.55
  • -2.33 μIU/mLfasting insulin
  • 19 RCTs1131 participants

A meta-analysis of 19 RCTs found cocoa flavanol intake significantly improved insulin resistance (lower HOMA-IR and fasting insulin) and modestly improved the lipid profile (lower triglycerides, higher HDL). LDL cholesterol showed no significant change, and several other biomarkers were null, so the metabolic effect is selective rather than across-the-board.

Reported effect: 19 RCTs, 1131 participants: triglycerides WMD -0.10 mmol/L (95% CI -0.16, -0.04); HDL +0.06 mmol/L (95% CI 0.02, 0.09); fasting insulin -2.33 μIU/mL (95% CI -3.47, -1.19); HOMA-IR -0.93 (95% CI -1.31, -0.55). No significant LDL change.

“19 RCTs that comprised 1131 participants ... Triglycerides: WMDs between treatment and placebo were -0.10 mmol/L (95% CI: -0.16, -0.04 mmol/L); HDL cholesterol 0.06 mmol/L (95% CI: 0.02, 0.09 mmol/L); fasting insulin -2.33 μIU/mL (95% CI: -3.47, -1.19 μIU/mL); HOMA-IR -0.93 (95% CI: -1.31, -0.55). No significant associations were found for other biomarkers.”

Source: PMID 27683874 · Lin 2016 · J Nutr

Cognition / memory in older adults

Null / no benefit RCT supported
  • 3,562older adults · 3-y RCT
  • 500 mg/daycocoa flavanols
  • 1 yprimary endpoint not significant

In the COSMOS-Web cognition RCT (cocoa extract 500 mg flavanols/day), the prespecified primary memory endpoint at 1 year was not statistically significant -- an honest null. In a subgroup analysis, the intervention restored memory among participants with lower habitual diet quality or lower habitual flavanol consumption, but the abstract reports no effect-size number for that subgroup, so this is a hypothesis-generating signal rather than a confirmed benefit.

Effect size: this study reports the direction of the finding but does not state a specific numeric effect size, so none is given here rather than estimated.

“3,562 older adults, who were randomly assigned to a 3-y intervention of cocoa extract (500 mg of cocoa flavanols per day) or a placebo ... the prespecified primary end point testing for an intervention-related improvement in memory in all participants after 1 y was not statistically significant ... restored memory among participants in lower tertiles of habitual diet quality or habitual flavanol consumption.”

Source: PMID 37252983 · Brickman 2023 · Proc Natl Acad Sci U S A

Dosage (research context · not a recommendation)

200-900 mg cocoa flavanols/day (research context · EFSA endothelial-function condition ≥200 mg flavanols/day; COSMOS cardiovascular RCT used 500 mg flavanols/day over 3.6 years; cognition RCTs 500-900 mg/day). Flavanol content varies enormously with processing — Dutch (alkalized) cocoa can lose >60% of flavanols, so a standardized flavanol figure is required.

Regulatory Status · 4 Markets

US · FDA
Lawful dietary-supplement ingredient under DSHEA; cocoa flavanols carry an FDA qualified health claim (2023) on the relationship between high-flavanol cocoa powder and reduced cardiovascular-disease risk (qualified — mandatory limiting language). Structure/function claims permitted.
EU · EFSA
One authorized Art. 13.5 health claim (EU Reg 2014/462): 'Cocoa flavanols help maintain the elasticity of blood vessels, which contributes to normal blood flow' — condition ≥200 mg cocoa flavanols/day. Antioxidant and mood claims are non-authorized in the EU.
BR · ANVISA
Permitted dietary-supplement ingredient under RDC 243/2018 (Brazil is a major cocoa-producing country); no specific authorized functional claim — permitted as a constituent only.

Safety

Large-scale safety established in COSMOS (n=21,442 · 500 mg flavanols/day · 3.6 years · no serious safety signal · Sesso 2022 PMID 35294962); general tolerability comparable to placebo across RCTs. Theobromine at high total intake (>1000 mg/day) can cause palpitations, insomnia, or GI irritation, and extracts carry trace caffeine (1-4%). Cocoa is a natural accumulator of cadmium and lead — a California Prop 65 / heavy-metal sourcing consideration rather than an efficacy issue. Pregnancy/lactation should limit total theobromine; caution with MAOI medication (tyramine content). Not a substitute for medical care; people with cardiovascular disease, hypertension, or on medication should consult a healthcare professional before use.

References

PubMed-indexed citations anchoring the benefit findings above. Effect sizes are reported as published.

  1. PMID 28439881 · Ried 2017 · Cochrane Database Syst Rev — Blood pressure within the normal range
  2. PMID 35294962 · Sesso 2022 · Am J Clin Nutr — Cardiovascular events (large RCT primary endpoint)
  3. PMID 32220363 · Ebaditabar 2020 · Clin Nutr ESPEN — Endothelial function (flow-mediated dilation)
  4. PMID 27683874 · Lin 2016 · J Nutr — Cardiometabolic biomarkers (insulin & lipids)
  5. PMID 37252983 · Brickman 2023 · Proc Natl Acad Sci U S A — Cognition / memory in older adults

Frequently Asked Questions

1. Did the large COSMOS trial prove cocoa prevents heart disease?

No. In COSMOS (Sesso 2022, PMID 35294962), 21,442 adults took cocoa extract (500 mg flavanols/day) for a median 3.6 years, and the prespecified primary outcome -- total cardiovascular events -- was not significantly reduced (HR 0.90, 95% CI 0.78-1.02, P=0.11). A secondary cardiovascular-death endpoint (HR 0.73) was lower, but secondary results are hypothesis-generating only. The trial did establish large-scale safety with no serious safety signal.

2. What does the evidence show for cocoa flavanols and memory?

In the COSMOS-Web cognition RCT (Brickman 2023, PNAS, PMID 37252983), 3,562 older adults took cocoa extract (500 mg flavanols/day), and the primary memory endpoint at 1 year was not statistically significant. A subgroup with lower habitual diet quality or flavanol intake showed restored memory, but the abstract reports no effect-size number for that subgroup, so it remains a signal rather than a confirmed benefit.

3. How much cocoa flavanol is used in research, and is processing relevant?

Research doses typically range from 200-900 mg cocoa flavanols/day; the EFSA-authorized EU vascular claim uses a condition of >=200 mg/day. Flavanol content varies enormously with processing -- Dutch (alkalized) cocoa can lose much of its flavanols -- so a standardized flavanol figure is required rather than relying on cocoa mass. This is educational information, not dosing guidance.

Last evidence review: 2026-06-27

← Evidence library

Ask Agent Axor