[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
August 27, 2003

Vascular Effects of Cocoa Rich in Flavan-3-ols

Author Affiliations

Letters Section Editor: Stephen J. Lurie, MD, PhD, Senior Editor.

JAMA. 2003;290(8):1030-1031. doi:10.1001/jama.290.8.1030

To the Editor: In vitro studies have suggested that flavonoids may have specific vascular effects, but their mechanism of action has not been clarified.1 A subclass of flavonoids—flavan-3-ols and their oligomers (procyanidins)—are constituents of cocoa beans, which can be detected in human plasma after ingestion of cocoa.2 In turn, plant extracts rich in flavan-3-ols can increase the activity of nitric oxide synthase (NOS) in endothelial cells.3 Nitric oxide is an essential signaling molecule in vascular physiology. Nitric oxide bioactivity can be preserved in human plasma in a circulating pool via increases in a number of nitrosated compounds.4,5 Thus, it is possible that cocoa rich in flavan-3-ols may lead to improved endothelium-dependent dilation via an increase of nitric oxide bioactivity.

However, commercially available cocoa drinks contain only small amounts of flavan-3-ols due to roasting and alkalization of cocoa beans, which are known to degrade flavan-3-ols. We tested the hypothesis that ingestion of flavan-3-ol rich cocoa can increase the circulating pool of nitric oxide in human plasma, thus increasing endothelium-dependent dilation.


Participants were 26 outpatients with at least 1 cardiovascular risk factor, including history of coronary artery disease, hypertension, hyperlipidemia, diabetes, or current tobacco use. Individuals were excluded if they had C-reactive protein levels greater than 0.5 mg/dL, atrial fibrillation, acute coronary syndrome, or New York Heart Association class III or IV heart failure. Individuals were studied in the morning after a 12-hour fasting period.

In an initial study involving the first 6 participants, we assessed the time course of flavan-3-ol effects on flow-mediated dilation (FMD). This was measured at 0, 2, 4, and 6 hours after ingestion of 100 mL of cocoa drink containing 176 mg of flavan-3-ols (70 mg of epicatechin plus catechin, 106 mg of procyanidins [The Positive Food Co, Wokingham, England]) (n = 6) or control (100 mL cocoa drink with <10 mg of flavan-3-ols [Dovedrink, Mars Inc, Hackettstown, NJ] or water) (n = 3).

We then used these results to guide the timing of a double-blind crossover study. Twenty participants received 100 mL of cocoa drinks with high or low levels of flavan-3-ols, in random order, on 2 consecutive days. The sum of nitrosylated and nitrosated species (collectively referred to as RNO) was measured by reductive chemiluminescence assay 2 hours after ingestion on both days.4 Nitrate and nitrite levels were measured as previously described.6 Endothelium-dependent dilation was assessed by measuring FMD of the brachial artery. In addition, we measured a number of other vascular parameters that would not be expected to change as a result of flavan-3-ol, including blood pressure, heart rate, and plasma levels of nitrite and nitrate. Similarly, we measured endothelium-independent dilation of the brachial artery following sublingual application of 400 µg of glyceroltrinitrate, diameter of the brachial artery, and forearm blood-flow at rest and during reactive hyperemia, as assessed by venous occlusion plethysmography. (Technical details are available from the authors.)

All variables except endothelium-independent dilation were measured both before and after ingestion of the cocoa. Endothelium-independent dilation was measured only after ingestion of each drink, as nitroglycerine could have interfered with measurement of the other variables. Differences were assessed by paired t tests, with P values for multiple comparisons adjusted by the Bonferroni criterion. Our study was approved by the ethics board of the Medical Faculty of the Heinrich Heine-University, and all participants gave written informed consent.


The sample had a mean (SD) age of 41 (14) years and body mass index (calculated as weight in kilograms divided by the square of height in meters) of 25 (4). Sixty-two percent were male, 19% had coronary heart disease, 27% had arterial hypertension, 31% had hyperlipidemia, 8% had diabetes mellitus, and 77% were smokers.

In the initial study (Figure 1), ingestion of 100 mL of cocoa drink rich in flavan-3-ols increased FMD maximally at 2 hours, whereas the cocoa drink low in flavan-3-ols did not affect this parameter. There was no significant difference in FMD in the 3 individuals receiving either water or the cocoa drink low in flavan-3-ols.

Figure 1. Time Course of Flow-Mediated Dilation After Ingestion of 100 mL of Cocoa Drink Containing High (176 mg mL; n = 6) or Low (<10 mg mL; n = 3) Amounts of Flavan-3-ols
Image description not available.
Data given as mean (SEM). *Indicates significant difference from baseline (P<.001).

In the crossover study (Figure 2), ingestion of cocoa rich in flavan-3-ols increased plasma levels of RNO from 22 to 36 nmol/L and FMD from 3.4% to 6.3% (P<.001 for both comparisons). Changes in plasma levels of RNO and in FMD were correlated (r = 0.42; P = .02). As expected, there were no significant differences in other vascular measures before vs after ingestion of the cocoa drink rich in flavan-3-ol (diameter of the brachial artery, 4.54 vs 4.53 mm; forearm blood-flow at rest and during reactive hyperemia, 1.8 vs 2.1 and 11.4 vs 11.6 mL/min per 100 mL of tissue, respectively; mean blood pressure, 91 vs 90 mm Hg; heart rate, 66 vs 62 bpm; levels of plasma nitrite [128 vs 119 nmol/L] and nitrate [32 vs 28.5 µmol/L]). Flow-mediated dilation, levels of RNO, and all other vascular variables were unchanged before and 2 hours after ingestion of cocoa drink low in flavan-3-ols. Endothelium-independent dilation of the brachial artery was not significantly different after ingestion of either cocoa drink (12.3% vs 12.9%; P = .82).

Figure 2. Effects of Cocoa Drink Containing Low (<10 mg/100 mL) or High (176 mg/100 mL) Amounts of Flavan-3-ols on Plasma Nitric Oxide Pool and Flow-Mediated Dilation
Image description not available.
Nitric oxide pool was measured as the nitrosylated and nitrosated species collectively referred to as RNO (S-nitrosothiols, N-nitrosoproteins). Ingestion of 100 mL of the cocoa drink rich in flavan-3-ols increased RNO concentrations and flow-mediated dilation significantly over the controls. Squares indicate means; error bars, SEM. *Indicates significant difference from baseline (P<.001).

We found that a single dose of a cocoa drink rich in flavan-3-ols transiently increased nitric oxide bioactivity in human plasma and significantly reversed endothelial dysfunction. The correlation between FMD and levels of RNO suggests that flavan-3-ols induce arterial dilation via their effects on nitric oxide availability, a conclusion that is supported by the negative results for the other vascular variables. The long-term clinical effect of flavan-3-ols, however, remains to be established.

Funding/Support: This work was supported by grants from the Deutsche Forschungsgemeinschaft (SFB 575 and 612), Biomedizinisches Forschungszentrum of the University of Düsseldorf, and Mars, Inc (Slough, England). Dr Sies is a Fellow of the National Foundation for Cancer Research.

Acknowledgment: We gratefully acknowledge the technical assistance of S. Matern and G. Doemer. We thank Drs J. Wills, H. Schmitz, and J. Hammerstone, of Mars Inc, for supplying and analyzing the cocoa drinks.

Ross JA, Kasum CM. Dietary flavonoids: bioavailability, metabolic effects, and safety.  Annu Rev Nutr.2002;22:19-34.PubMed
Holt RR, Schramm DD, Keen CL, Lazarus SA, Schmitz HH. Chocolate consumption and platelet function.  JAMA.2002;287:2212-2213.PubMed
Leikert JF, Räthel TR, Wohlfart P, Cheynier V, Vollmar AM, Dirsch VM. Red wine polyphenols enhance endothelial nitric oxide synthase expression and subsequent nitric oxide release from endothelial cells.  Circulation.2002;106:1614-1617.PubMed
Feelisch M, Rassaf T, Mnaimneh S.  et al.  Concomitant S-, N-, and heme-nitros(yl)ation in biological tissues and fluids: implications for the fate of NO in vivo.  FASEB J.2002;16:1775-1785.PubMed
Cannon III RO, Schechter AN, Panza JA.  et al.  Effects of inhaled nitric oxide on regional blood flow are consistent with intravascular nitric oxide delivery.  J Clin Invest.2001;108:279-287.PubMed
Rassaf T, Preik M, Kleinbongard P.  et al.  Evidence for in vivo transport of bioactive nitric oxide in human plasma.  J Clin Invest.2002;109:1241-1248.PubMed