Cocoa beans, endothelial function and aging: an unexpected friendship?

In the current issue of the journal, Fisher and Hollenberg [1] elegantly demonstrate a significant increase in flow-mediated dilation, as measured by finger tonometry, in 15 healthy individuals aged < 50 years and 19 healthy individuals aged ≥ 50 years, after 4–6 days of cocoa supplementation. Of note, cocoa-related changes in flow-mediated dilation were age-dependent, being evident in both groups but more striking among older individuals. Similar findings were found for pulse wave velocity, an index of arterial stiffness rather than of endothelial function [2], that remained unchanged after cocoa supplementation in both groups, but increased after an additional cocoa drink on the final study day. Because the vascular effects of cocoa were markedly attenuated by Nω-nitro-L-arginine-methyl-ester, the intriguing findings provided by Fisher and Hollenberg [1] suggest that cocoa can significantly improve nitric oxide bioavailability in healthy individuals, particularly after 50 years of age. More generally, because the risk for developing atherosclerosis in industrialized countries has been suggested to be linked to the endothelial dysfunction, their study also suggests that changes in dietary habitudes could be of help in preserving nitric oxide bioavailability, thereby blunting the onset of age-related atherosclerotic lesions. Concordant with the above hypothesis, Chauhan et al. [3] have already reported an improvement in the coronary artery response to acetylcholine after L-arginine infusion in healthy individuals aged more than 50 years. Because L-arginine is the natural substrate of nitric oxide synthase and its infusion is ineffective in younger people [3], it could be argued that an impaired nitric oxide biovailability with advancing age can be restored in older populations by foods containing L-arginine. However, the issue appears to be controversial because, although some studies have reported an improvement in vascular function following oral arginine supplementation [4], others were unable to document such favourable effects [5,6]. At variance from the conflicting results obtained with L-arginine, the positive effects of cocoa on endothelium-dependent vasorelaxation are well documented. First, as anticipated, the improvement in cocoa-induced vasodilation described by Fisher and Hollenberg [1] was markedly attenuated after Nω-nitro-L-arginine-methyl-ester administration, thereby indicating that cocoa ingestion promotes vasodilation through the nitric oxide pathway. Second, it has previously been demonstrated that dark chocolate administration is capable of improving endothelium-dependent vasorelaxation but not the vascular response to 25 μg sublingual glyceryl trinitrate in adult grade I essential hypertensive individuals [7]. In addition, previous reports in healthy adults [8,9], and in patients with either a history of coronary artery disease and/or hypertension and/or diabetes [10], also indicate a marked increase in nitric oxide-dependent vasorelaxation with different chocolate interventions. In a study by Heiss et al. [10], chocolate intake significantly increased plasma nitrite and nitrate, an indirect biomarker of nitric oxide production [11]. Therefore, the finding that cocoa improves endothelium-dependent vasorelaxation is almost univocally confirmed by different studies in humans. This supports the intriguing hypothesis that changes in dietary habitude could be followed by comparatively positive modifications in vascular reactivity. More generally, the results of the study by Fisher and Hollenberg [1] provide a pathophysiological basis for a ‘Mediterranean-like’ diet being capable of reducing the onset of both cardiovascular and cerebrovascular events [12,13]. Should the interesting findings obtained by Fisher and Hollenberg [1] be confirmed in large-scale trials, this would suggest that cocoa is a healthy food capable of improving endothelial cell function and protecting the vasculature at all ages, but particularly in individuals aged ≥ 50 years. Cocoa and vascular function: the role of flavanols Several substances are naturally contained in the cocoa bean and may be responsible for the observed potentiation in nitric oxide-dependent vasorelaxation. Interestingly, all controlled studies on cocoa performed to date have been conducted with chocolate bars or beverages that are particularly rich in a subclass of flavonoids known as flavanols (i.e. the wine, fruit, vegetable, and tea components that have often been suggested as the main substances responsible for the cardiovascular benefits deriving from their assumption). Fisher and Hollenberg [1] also agree with this hypothesis, and suggest that it was the flavanols in the cocoa to which the observed vasodilatory responses should be attributed. Indeed, drinking tea, a flavonoid-rich beverage, also improves flow-mediated dilation in the brachial arteries of both hypertensive and normotensive patients with coronary artery disease [14]. Similarly, flavonoid-rich purple grape juice has been reported to increase flow-mediated dilation in association with a decrease in oxidative stress in patients with coronary artery disease [15]. In vitro, cocoa procyanidins (i.e. oligomers comprised of epicatechin and catechin subunits that are particularly represented in fresh cacao beans) have been reported to significantly increase nitric oxide synthase expression and activity [16], to induce rabbit aortic ring relaxation [16] and to prevent nitric oxide reaction with the superoxide anion to form peroxynitrite [17]. Taken together, the above data support the hypothesis that flavanols, and not other substances present in the cocoa bean (magnesium, theobromine, fats, etc.), comprise the main factors responsible for the observed improvement in flow-mediated vasorelaxation [1]. Obviously, this does not mean that other potential health benefits from cocoa may not derive from either theobromine [18] and/or magnesium [19] (i.e. two constituents of cocoa beans suggested to exert positive vascular and metabolic effects). Should the correction of endothelial dysfunction be targeted in clinical practice? Over the last three decades, a consistent volume of data indicated that the vascular endothelium resembles a widely distributed organ rather than a silent cobblestone barrier. Thus, endothelial cells are currently known to act as the main modulators of vascular tone, inter-relation with platelets and clotting factors, vascular cell migratory and proliferative abilities [20]. In addition, it is generally accepted that the resting vascular endothelium also actively blocks the trafficking of white cells from the bloodstream onto the vascular wall [20–22]. In the presence of an inflammatory milieu, such as that triggered by common modifiable risk factors, endothelial cells lose these protective properties and allow the penetration of neutrophils, monocytes/macrophages and lymphocytes into the sub-endothelial environment, and thereby initially put in motion and then maintain the ultrachronic, low-degree intravascular inflammation leading to atherosclerotic plaque development [20–22]. A decrease in nitric oxide bioavailability has frequently been claimed to represent the common feature of different endothelial dysfunctions [23]. Concordant with this, an abnormal endothelium-dependent vasorelaxation has been described in human conditions characterized by an increased risk for developing atherosclerosis and its clinical sequelae (i.e. essential hypertension, hypercholesterolemia, type 2 diabetes, obesity and aging) [24–27]. In turn, because the increase in risk for developing atherosclerosis in industrialized countries is suggested to be linked to the endothelial dysfunction, it has been speculated that positive changes in nitric oxide bioavailability (i.e. correction of endothelial dysfunction) might protect against atherosclerosis in the general, as well as the older, population [27]. By contrast to this hypothesis, the role of endothelial dysfunction as a cardiovascular risk factor in the whole population, as well as in elderly patients, is still unclear. In a study by Ghiadoni et al. [28], flow-mediated brachial artery responses were inversely related to carotid intima–media thickness, but only very few patients were > 50 years old. Although changes in the brachial artery diameter in response to reactive hyperemia were modestly but significantly correlated with age in individuals aged 25–79 years who underwent coronary arteriography [29], no significant relationship was found between age-related endothelial dysfunction and the severity of coronary atherosclerosis. With regard to follow-up studies, a consistent volume of data has been obtained with disparate methodologies in different vascular districts [30–32]. Again, data obtained from either the coronary or the upper arm circulations show that endothelial dysfunction predicts both coronary and cerebrovascular events. However, no reports have focused on elderly patients. Thus, several studies conducted in both animals [33] and humans [34] have demonstrated that endothelium-dependent vascular relaxation is impaired in elderly individuals. However, the amelioration in vascular endothelial function described in healthy individuals aged ≥ 50 years by Fisher and Hollenberg [1] after flavanol-rich cocoa supplementation, although of pathophysiological interest, cannot reflect an effective vascular protection against the onset of atherosclerosis. Other potential vascular benefits from cocoa A renal hemodynamic state consistent with nitric oxide system activation, a low prevalence of atherosclerotic diseases, and no age-dependent blood pressure increase has been described in Kuna Indians [35], an amerind population living in the San Blas island chain off the Coast of Panama who drink large amounts of cocoa that is rich in flavanols (an average of 3–4 cups of cocoa bean-derived beverage per day) [35]. In accordance with this, a decrement of blood pressure has been observed with chocolate intervention in grade I essential hypertensives, elderly patients with isolated systolic hypertension and healthy young soccer players [36,37]. In addition, dark chocolate has also been reported to reduce insulin resistance [38], platelet aggregability [39] and circulating low-density lipoprotein cholesterol levels [37], and to increase serum high-density lipoprotein concentrations [40] in humans. Although recent epidemiological data have indicated an association between chocolate assumption and blood pressure elevation [41], consumers of chocolate and other candies have been reported to have reduced mortality compared to ‘non-candy’ eaters [42]. Together with the observed changes in endothelial function, the above findings suggest that modifications in dietary habitude by the introduction of cocoa could result in cardiovascular prevention. Thus, the cocoa bean, probably due to its high flavanol content, restores normal endothelial function and may promote vascular protection. In addition, cocoa-derived foods are known to be very tasty, pleasant, and their use may often be followed by mood elevation [43]. However, based on the growing pandemic explosion of obesity in children, as well as in young, middle-aged and older adults [44,45], we should be very careful in recommending cocoa ingestion to our patients, at least until, as Fisher and Hollenberg have correctly pointed out [1], large-scale trials and measures in both younger and older populations become available. This suggestion is particularly reinforced by the fact that the common food-processing practices used to prepare commercially available cocoa and cocoa-derived products can often significantly reduce or eliminate flavanols. Thus, the flavanol-rich cocoa products used in experimental studies, and even present in some commercially available flavanol-rich chocolate bars that have been tested in controlled short-lasting studies, should not be confused with a number of commercially available snacks that contain many calories but are low in natural cocoa and flavanols.