Epidemiological studies showed that the consumption of flavonoid-rich foods such as cocoa and tea is associated with a lower risk of cardiovascular disease (CVD). Randomised controlled trials (RCTs) showed that cocoa and tea improved markers of cardiometabolic health including blood pressure, endothelial function, insulin resistance, arterial stiffness and inflammation.
Cocoa is particularly rich in the flavan-3-ol epicatechin and tea is the main dietary source of epicatechin and of the major flavonol quercetin. However, evidence on the individual roles of epicatechin and quercetin in the health effects of cocoa and tea is still scarce. Therefore, we estimated the strength of the association between epicatechin intake and CVD mortality in a prospective cohort study. Furthermore, we also investigated the effects of epicatechin and quercetin on markers of cardiometabolic health and gene expression, by means of two RCTs.
In Chapter 2, the association between epicatechin intake and CVD mortality was studied using data from the Zutphen Elderly Study, a cohort of 744 elderly Dutch men. During 25 years of follow-up, 329 men died from CVD and 148 from coronary heart disease (CHD). Results from this study showed that men in the highest tertile of epicatechin intake had a 38% lower risk of CHD mortality compared to men in the lowest tertile. For men with prevalent CVD, the risk of CVD mortality was 46% lower for men in the highest tertile of intake, compared to men in the lowest tertile. This is the first epidemiological study to have investigated the association between epicatechin intake and CVD mortality. Hence, more and larger cohort studies are required to confirm this association, possibly with a focus on populations with a high risk of CVD.
In Chapter 3, the chronic effects of pure epicatechin and quercetin on markers of cardiometabolic health were investigated by means of a RCT. Thirty-seven apparently healthy men and women aged 40–80 years consumed (-)-epicatechin (100 mg/d), quercetin-3-glucoside (160 mg/d), or placebo capsules for 4 weeks, in random order. Markers of cardiometabolic health were measured before and after each 4-week intervention. The results of this study showed that epicatechin improved insulin resistance and had a borderline significant effect on endothelial function. This suggests that epicatechin contributes to the cardioprotective effects of cocoa and tea, however, larger long-term RCTs are required to confirm these effects. Pure quercetin supplementation did not affect any of these markers of cardiometabolic health.
Using data from the same study, we investigated the effects of supplementation of pure epicatechin and quercetin on a comprehensive set of biomarkers of endothelial dysfunction and inflammation (Chapter 4). With the exception of sE-selectin (a biomarker of endothelial dysfunction), epicatechin supplementation did not beneficially influence any of the biomarkers, suggesting a lack of evidence for a role of epicatechin in inflammation. Quercetin also lowered sE-selectin as well as the inflammatory biomarker IL-1β and the overall z-score for inflammation. This suggests that quercetin may contribute to the cardioprotective effects of tea by reducing inflammation and possibly by improving endothelial function.
In the same study, the effects of pure epicatechin supplementation on whole genome gene expression profiles of circulating immune cells were also assessed (Chapter 5). Pure epicatechin supplementation modestly reduced gene expression related to inflammation signalling routes in circulating immune cells – routes which are known to play a role in cardiovascular health. However, there was no evidence that epicatechin affected pathways related to insulin resistance or endothelial function.
To directly compare the acute effects of pure epicatechin and epicatechin from dark chocolate on vascular function, we carried out an acute RCT in 20 apparently healthy men aged 40-80 years (Chapter 6). On three separate occasions, subjects consumed: 1) 70g dark chocolate (150 mg epicatechin) with two placebo capsules; 2) two pure epicatechin capsules (100 mg epicatechin) with 75g white chocolate and 3) two placebo capsules with 75g white chocolate (0 mg epicatechin). Endothelial function and arterial stiffness were measured before and two hours after each intervention. To determine epicatechin bioavailability, epicatechin metabolites were measured in blood samples taken at repeated intervals over a period of 8 hours. There was no significant difference in improvement in endothelial function or arterial stiffness between pure epicatechin and dark chocolate. There was also no difference in bioavailability of pure epicatechin and epicatechin from dark chocolate, when standardised per 100 mg of epicatechin. This suggests that epicatechin may contribute to the vascular effects of cocoa and that the bioavailability of pure epicatechin and epicatechin from dark chocolate is similar.
In the general discussion, the main findings of this thesis were first summarised. Methodological considerations related to cohort studies, such as the assessment of flavonoid intake and the possibility of residual confounding were also discussed. Issues related to the relevance of cardiometabolic markers in RCTs and the effect of cocoa flavan-3-ol bioavailability were addressed. Finally, suggestions for future research were put forward.
In conclusion, the results of this thesis suggest that epicatechin contributes to the cardioprotective effects of cocoa and tea. Epicatechin intake was inversely related to CHD mortality in elderly men, and to CVD mortality in men with prevalent CVD. The cardioprotective effects of epicatechin are likely mediated through improvements in insulin resistance and possibly endothelial function. In contrast, quercetin is unlikely to play a major role in the cardioprotective effects of tea. Results for quercetin from cohort studies are inconclusive, and based on the results of our chronic RCT, quercetin did not affect vascular function or insulin resistance, but may help to lower inflammation. Evidence of the role that individual flavonoids play in the aetiology of CVD is still limited. More studies with pure flavonoids are required to elucidate their role.