Rich chocolate: a small dip in blood pressure in cocoa trials

In short trials, flavanol-rich cocoa products lowered blood pressure modestly, mainly in people with high or borderline-high pressure. A meta-analysis of 10 randomized trials with 297 people found flavanol-rich cocoa products cut systolic pressure by 4.5 mmHg (95% CI 3.2 to 5.9) and diastolic by 2.5 mmHg over 2 to 18 weeks, with large differences between trials and no settled dose. A second meta-analysis, of 15 trial arms, found a 5.0 mmHg systolic drop in people with hypertension or prehypertension. In people with normal pressure the 1.6 mmHg fall did not reach significance. Those trials were small and short, and mostly set 50 to 70 percent cocoa chocolate against white chocolate. The biggest trial used a capsule instead of a bar. COSMOS gave 21,442 older adults a cocoa extract with 500 mg of flavanols for a median 3.6 years, and the standard analysis of cardiovascular events did not reach significance (HR 0.90, 95% CI 0.79 to 1.03). Only a secondary, severity-ranked analysis favored cocoa, with a win ratio of 0.84.

The numbers, per 3.5 oz (100 g)

What your body absorbs

• In a dose-response trial, plasma epicatechin 2 hours after eating 27, 53 or 2.8 oz (80 g) of procyanidin-rich chocolate rose by 133, 258 and 355 nmol/L. — healthy adults, single doses Evidence B source
  +133 +/- 27, +258 +/- 29, +355 +/- 49 nmol/L at 27, 53, 80 g
• After an 80 g bolus of semisweet chocolate, plasma epicatechin rose 12-fold within 2 hours and approached baseline by 6 hours. — healthy adults, single 80 g dose Evidence C source
  Plasma epicatechin 22 to 257 nmol/L at 2 h (P < 0.01); near baseline by 6 h
• In a 2-week crossover RCT in 14 overweight adults, 20 g dark chocolate with 500 mg polyphenols lowered fasting glucose and blood pressure as well as a 1000 mg version. — fourteen overweight and obese subjects, 2 weeks per arm (n = 14) Evidence B source
  500 mg equally effective as 1000 mg polyphenols on fasting glucose, SBP and DBP (size not in abstract)

What it does to your health

• A meta-analysis of 10 RCTs with 297 people found flavanol-rich cocoa products lowered systolic blood pressure by 4.5 mmHg over 2 to 18 weeks. — 10 RCTs, 297 healthy normotensive adults or patients with prehypertension/stage 1 hypertension, 2 to 18 weeks (n = 297) Evidence A source
  SBP -4.5 mmHg (95% CI -5.9 to -3.2); DBP -2.5 mmHg (95% CI -3.9 to -1.2)
• In the COSMOS trial of 21,442 older adults over 3.6 years, a cocoa flavanol extract cut a severity-ranked composite of cardiovascular events (win ratio 0.84), while the standard Cox analysis was not significant. — 21,442 adults (women aged 65 or older, men 60 or older), median 3.6 years (n = 21442) Evidence B source
  Reciprocal win ratio 0.84 (95% CI 0.72-0.99); Cox HR 0.90 (95% CI 0.79-1.03)

Safety, limits and interactions

• A German exposure assessment estimated worst-case cadmium intake from cocoa at 0.363 micrograms per kg body weight per week, with levels depending on cocoa origin. — German National Nutrition Survey II consumers, German Food Monitoring cocoa data Evidence C source
  Cadmium up to 0.363 ug/week/kg BW; aluminum up to 0.152 mg/week/kg BW (most conservative scenario)

What people believe that the data does not support

• In a 6-month RCT in 140 postmenopausal women, 10 g of 99% cocoa chocolate daily raised self-rated health by 6 points on a 100-point scale but did not change overall quality of life. — 140 postmenopausal women aged 50-64, 6 months (n = 140) Evidence B source
  EQ-VAS +6.0 points (95% CI 0.4, 11.7); EuroQoL-5D-3.2 quarts (3L) and Cervantes scores unchanged
• The EU did not authorize the claim that cocoa flavanols help maintain normal blood pressure, and no cocoa claim is authorized in the register snapshot. — — Evidence E source
  —

Who this works differently for

• A meta-analysis of 15 trial arms found cocoa chocolate lowered systolic blood pressure by 5.0 mmHg in people with hypertension or prehypertension but not significantly in normotensive people. — 15 trial arms from 13 RCTs, hypertensive and normotensive adults, 2 to 18 weeks (n = 15 trial arms) Evidence A source
  SBP -5.0 +/- 3.0 mmHg, DBP -2.7 +/- 2.2 mmHg in hypertensive/prehypertensive; SBP -1.6 mmHg (P=0.17) in normotensive
• In a 6-month RCT in 132 postmenopausal women, adding 10 g of 99% cocoa chocolate daily cut body fat mass by 1.4 lb (0.63 kg) without changing weight. — postmenopausal women, 57.2 (sd 3.6) years, 10 g/day of 99% cocoa chocolate for 6 months (n = 132) Evidence B source
  Fat mass -1.4 lb (95% CI -2.5 to -0.24; metric: -0.63 kg, -1.15 to -0.11); body fat -0.79% (95% CI -1.31, -0.26); BMI -0.20 (NS)
• In a 6-month RCT in 140 postmenopausal women, 10 g of 99% cocoa chocolate daily shortened Trail Making Test B time by 12 seconds, with no change in memory or attention. — 140 postmenopausal women aged 50-64, 10 g/day 99% cocoa chocolate for 6 months (n = 140) Evidence B source
  TMT-B -12.08 s (95% CI -23.99, -0.18); other cognitive domains unchanged
• In 8,905 COSMOS participants followed 3.4 years, cocoa extract did not reduce new hypertension overall (HR 0.96, 95% CI 0.88-1.05). — 8905 older adults free from hypertension, mean age 71.1 years, median 3.4 years (n = 8905) Evidence B source
  Overall HR 0.96 (95% CI 0.88-1.05); subgroup analysis: SBP <120 HR 0.76 (0.64-0.90), SBP 120-139 HR 1.05
• In a pilot crossover RCT in 10 people with chronic fatigue syndrome, 8 weeks of high-polyphenol chocolate improved the Chalder Fatigue score from 33 to 21.5, while a polyphenol-free chocolate did not. — ten subjects with chronic fatigue syndrome, 8 weeks per arm (n = 10) Evidence B source
  Chalder Fatigue Scale median 33 to 21.5 (P = 0.01) on high cocoa chocolate; worsened 28.5 to 34.5 on control

The bottom line

Your body takes the flavanols up quickly and in proportion to the dose. Two hours after 27, 53 or 2.8 oz (80 g) of procyanidin-rich chocolate, plasma epicatechin rose by 133, 258 and 355 nmol/L, in a trial whose abstract does not say how many people took part. In 14 overweight adults over two weeks, 20 g of dark chocolate with 500 mg of polyphenols did as well on fasting glucose and blood pressure as a 1000 mg version. That trial had no cocoa-free arm and reported no sizes. The everyday-dose trials come from one Spanish group of postmenopausal women eating 10 g of 99 percent cocoa chocolate a day for 6 months, against a control group given nothing. Fat mass fell by 1.4 lb (95% CI 0.24 to 2.5; metric: 0.63 kg, 0.11 to 1.15) with no change in weight. Time on one thinking test, Trail Making B, shortened by 12 seconds, with an interval that nearly touched zero (0.18 to 23.99) and no change in memory or attention. Self-rated health rose 6 points on a 100-point scale, while measured quality of life did not move. In COSMOS, cocoa extract did not reduce new hypertension overall in 8,905 older adults (HR 0.96, 95% CI 0.88 to 1.05). A pilot in 10 people with chronic fatigue syndrome reported less fatigue after 8 weeks of high-polyphenol chocolate, too small to lean on. The EU has not authorized the claim that cocoa flavanols maintain normal blood pressure. Who should be careful. A German assessment put worst-case cadmium intake from cocoa at 0.363 micrograms per kg of body weight per week, varying with where the cocoa grew, and its abstract does not compare that with a safe limit. We have no card on chocolate with blood pressure drugs, in pregnancy or for children, including whether that cadmium matters more for them. The one USDA row under this name is a fortified drink powder, a different product from the chocolate these trials used.

Sources

1. rchoc-daily-r4-06 · randomized controlled trial
   Herein, we demonstrate that 2 h after the ingestion of a procyanidin-rich chocolate containing 5.3 mg total procyanidin/g, of which 1.3 mg/g was (-)-epicatechin (epicatechin), plasma levels of epicatechin increased 133 +/- 27, 258 +/- 29 and 355 +/- 49 nmol/L in individuals who consumed 27, 53 and 80 g of chocolate, respectively.
   J Nutr, 2000 · checked 2026-10-02
2. rchoc-daily-r4-07 · clinical trial
   By 2 h after ingestion, there was a 12-fold increase in plasma epicatechin, from 22 to 257 nmol/L (P < 0.01).
   J Nutr, 2000 · checked 2026-10-02
3. rchoc-daily-r4-08 · randomized controlled trial
   It was observed that the 500 mg polyphenol dose was equally effective in reducing fasting blood glucose levels, systolic BP (SBP) and diastolic BP (DBP) as the 1000 mg polyphenol dose suggesting that a saturation effect might occur with increasing dose of polyphenols.
   Br J Nutr, 2010 · checked 2026-10-02
4. rchoc-daily-r4-02 · meta-analysis
   The mean BP change in the active treatment arms across all trials was -4.5 mm Hg (95% confidence interval (CI), -5.9 to -3.2, P < 0.001) for systolic BP and -2.5 mm Hg (95% CI, -3.9 to -1.2, P < 0.001) for diastolic BP.
   Am J Hypertens, 2010 · checked 2026-10-02
5. rchoc-daily-r4-09 · randomized controlled trial
   GPC analyses showed cocoa flavanol wins of 3.41% and placebo wins of 2.87%, yielding a reciprocal WR of 0.84 (95% CI 0.72-0.99) and negative NTBs of - 0.54% (- 1.04 to - 0.03), p = 0.037.
   Eur J Epidemiol, 2026 · checked 2026-10-02
6. rchoc-daily-r4-12 · exposure assessment
   The most conservative standard scenario resulted in the highest intake estimates for aluminium and cadmium with 0.152 mg/week/kg BW and 0.363 μg/week/kg BW and covered the origin influence calculated in origin-related scenarios.
   PLoS One, 2019 · checked 2026-10-02
7. rchoc-daily-r4-05 · randomized controlled trial
   The intervention group showed an increase of 6.0 points (95% confidence interval (CI): 0.4, 11.7) in the EQ-VAS compared to the control group (p = 0.036).
   Nutrients, 2020 · checked 2026-10-02
8. rchoc-daily-r4-13 · EU claims register entry
   POL-HC-6949 Cocoa flavanols Maintenance and promotion of a normal blood pressure Non-authorised
   EU Register on nutrition and health claims, claim POL-HC-6949, snapshot 2026-09-21 · checked 2026-10-02
9. rchoc-daily-r4-01 · meta-analysis
   However, subgroup meta-analysis was significant only for the hypertensive or prehypertensive subgroups (SBP: -5.0 +/- 3.0 mmHg; P = 0.0009; DBP: -2.7 +/- 2.2 mm Hg, P = 0.01), while BP was not significantly reduced in the normotensive subgroups (SBP: -1.6 +/- 2.3 mmHg, P = 0.17; DBP: -1.3 +/- 1.6 mmHg, P = 0.12).
   BMC Med, 2010 · checked 2026-10-02
10. rchoc-daily-r4-03 · randomized controlled trial
    The main effect of the intervention showed a favourable reduction in the IG with respect to the CG in body fat mass (-0·63 kg (95 % CI -1·15, -0·11), P = 0·019; Cohen's d = -0·450) and body fat percentage (-0·79 % (95 % CI -1·31, -0·26), P = 0·004; Cohen's d = -0·539).
    Br J Nutr, 2021 · checked 2026-10-02
11. rchoc-daily-r4-04 · randomized controlled trial
    Trail Making Test B execution time showed a decreased of -12.08 s (95% CI: -23.99, -0.18; p = 0.047) in the intervention group compared to the control group, after adjusting for age, educational level, time elapsed from the beginning of menopause and daily energy consumption (Cohen's d = -0.343).
    Nutr Neurosci, 2022 · checked 2026-10-02
12. rchoc-daily-r4-10 · randomized controlled trial
    Over a median follow-up of 3.4 years, cocoa extract supplementation had no significant effect on incident hypertension in an intention-to-treat analysis, with incidence rates of 7.1 and 7.4 per 100 person-years in cocoa and placebo groups, respectively (hazard ratio, 0.96 [95% CI, 0.88-1.05]).
    Hypertension, 2025 · checked 2026-10-02
13. rchoc-daily-r4-11 · randomized controlled trial
    The Chalder Fatigue Scale score improved significantly after 8 weeks of the HCL/PR chocolate arm [median (range) Exact Sig. (2-tailed)] [33 (25 - 38) vs. 21.5 (6 - 35) 0.01], but that deteriorated significantly when subjects were given simulated iso-calorific chocolate (CLF/CP) [ 28.5 (17 - 20) vs. 34.5 (13-26) 0.03].
    Nutr J, 2010 · checked 2026-10-02