2.1. Coffee–caffeine

Surprisingly enough, the effect of coffee consumption on cardiovascular risk has not been defined yet. Results of studies vary from a positive to a neutral or to even a J- or a U-shape association of coffee consumption with cardiovascular risk. Genetic predisposition for rapid caffeine metabolism may account for these discrepancies.

The main vasoactive substance in coffee is caffeine which, as we [2–4] and others [5,6] have shown, exerts an acute detrimental effect on aortic stiffness and wave reflections, both in normal subjects and in hypertensives. The ingestion of caffeine results in a sustained impairment of arterial stiffness, which lasts for at least three hours. Interestingly, the unfavourable effect of caffeine is more evident in central hemodynamics than in peripheral (brachial) pressure [2]. Furthermore in cross-sectional studies we have shown a positive association of chronic coffee consumption with increased aortic stiffness and wave reflections in healthy subjects [7] and with increased wave re- flections only in hypertensives (C. Vlachopoulos et al., unpublished data). The latter denotes that the stiffening effect of coffee in the aorta is probably less prominent in the already stiff aortas of hypertensives than in normal aortas. In other words, it appears that coffee does not have a stiffening effect on the aorta additive to that of hypertension.

Caffeine exerts its main cardiovascular effects through the antagonism of adenosine and the release of catecholamines, both resulting in vasoconstriction. Beyond this, chronic coffee consumption is associated with increased inflammatory markers, another possible way by which coffee may induce vascular dysfunction [1].

An important issue is whether substances other than caffeine contribute to the effect. In preliminary studies we have shown that coffee increases arterial stiffness more than its contained caffeine and that de-caffeinated coffee also increases arterial stiffness to a certain extent (C. Vlachopoulos et al., unpublished data). In line are studies that show that de-caffeinated coffee has cardiovascular effects. More specifically, it has been shown that it augments blood pressure and muscle sympathetic nervous activity, denoting that substances other than caffeine are responsible for cardiovascular activation [8].

Coffee consumption is very frequently combined with smoking. Interestingly enough, the combination of these two habits has a synergistic detrimental effect on aortic stiffness and wave reflections, both on an acute and on a chronic basis (Figure 1) [4].

Figure 1.

Pulse wave velocity response during the combined caffeine and smoking study. Each line represents response defined as net caffeine and smoking effect minus placebo and sham smoking effect (diamonds) or net smoking effect minus sham smoking effect (rectangles) at each time point. When smoking followed caffeine intake, pulse wave velocity reached a total increase of 0.85 m/s, 0.17 m/s in excess of the mere sum of caffeine effect (0.33 m/s) alone and smoking effect alone (0.35 m/s). P values on the graphs refer to the interaction between caffeine and smoking. Error bars: SEM. Reprinted with permission from Vlachopoulos et al. (2004) [4].

2.2. Tea

Tea consumption has been associated with decreased cardiovascular risk, although there is no consistency in the findings of all studies. Tea consumption has an acute and short-term (i.e. one month) beneficial effect on endothelial function, mainly attributed to its flavonoid content. To date, only acute studies on arterial stiffness and wave reflections are available. Contrary to studies on endothelial function, tea has an acute unfavourable effect on aortic stiffness and wave reflections. As we have shown, both black and green tea increase acutely wave reflections in normal subjects [9]. This effect becomes evident promptly after tea consumption (30 minutes), which is actually the time that it takes for caffeine to absorb. On the other hand, the effect of tea on aortic stiffness is less straightforward. Green tea has no significant effect, whereas black tea increases aortic stiffness for less than 2 hours. For comparison, the same amount of caffeine contained in black tea, when ingested alone, has a pronounced effect on aortic stiffness and wave reflections that lasts for 3 hours (Figure 2). A unifying explanation is that caffeine contained in tea increases arterial stiffness. Flavonoids, which notably take more time than caffeine to absorb (90–120 min) have a counterbalancing beneficial effect resulting in a total neutral effect in arterial stiffness with green tea (high concentration in flavonoids) and a delayed neutral effect with black tea (lower concentration in flavonoids).

Fig. 2.

Pulse wave velocity and augmentation index response during the black tea and the green tea study. Each line represents response defined as net tea or caffeine effect minus hot water effect at each time point. P values refer to the response of tea or caffeine compared to hot water during the whole study duration. Error bars: SEM. Reprinted with permission from Vlachopoulos et al. (2006) [9].

2.3. Cocoa–Chocolate

Originally cultivated in Mesoamerica, and considered a medicine and a “food of the Gods” by the Aztecs, cocoa acquires a scientific basis for its beneficial effects in our era. Habitual chocolate consumption is inversely associated with blood pressure and, importantly, cardiovascular risk. Its blood-lowering effect was originally noted in Kuna Indians who live in isolation, drink large amounts of cocoa, and have very low prevalence of hypertension. When they migrate to an urban area their blood pressure rises, denoting a protective effect of environmental/nutritional factors (most likely cocoa) in their island of origin. We have recently shown in a cross-sectional study that chocolate consumption is inversely associated with aortic pulse wave velocity and wave reflections in healthy subjects. Consumption of at least 12 grams of chocolate per day was associated with a significant decrease in carotid-femoral pulse wave velocity and in augmentation index. Furthermore, increasing intake of chocolate is associated with a decrease of central (aortic) systolic pressure and of central pulse pressure, but not of peripheral (brachial) pressure, denoting that its beneficial effect is not always evident when only peripheral pressures are measured (C. Vlachopoulos et al., unpublished data).

Most intervention studies have used cocoa and dark chocolate rich in cocoa flavonoids, because their beneficial effect is mainly attributed to their high flavonoid content. Both of them exert a beneficial acute and short-term effect on endothelial function, as it has been shown both in healthy subjects and in subjects with cardiovascular risk factors, or cardiovascular disease. Dark chocolate results acutely in a decrease in wave reflections in normal individuals [10]. Aortic stiffness is not largely affected by acute chocolate consumption, while a tendency to decrease at a later stage (3 hours or more) begins to show [10]. These taken together imply that cocoa exerts a vasodilatory effect on small and medium-size arteries possibly due to increased flavonoid-related NO bioavailability and/or increased prostacyclin production. The effect of the more largely consumed milk chocolate is not known. Effects are anticipated to be similar, although not necessarily identical, because milk chocolate contains much less flavonoids than dark chocolate, and also because flavonoids are absorbed to a lesser degree when milk is present.

2.4. Alcohol and alcoholic drinks

Moderate alcohol consumption is associated with decreased cardiovascular risk. Similarly to studies regarding cardiovascular risk, most cross-sectional studies on the effect of alcohol on arterial function do not discriminate between alcoholic drinks. In keeping again with these general studies, a J- or U-shaped curve has been demonstrated between alcohol consumption and aortic stiffness and wave reflections both in men and in women [11–14]. Possible mechanisms involved include increase in high density lipoprotein cholesterol, while, interestingly enough, the inflammatory effect of alcohol shows a U-shaped behaviour. The effect of alcohol on endothelial function is not so clear. Experimental data suggest that moderate alcohol consumption increases NO production, while heavy alcohol drinkers have impaired endothelial function. We have shown that the acute effect of pure alcohol is neutral; the vasodilation induced both in resting and hyperemic diameter of the brachial artery results in a non-significant change in flow-mediated dilatation [15].

Red Wine: Studies, particularly acute intervention ones, have focused specifically on red wine given the possible preponderance of this alcoholic drink in reducing cardiovascular risk (French paradox). Red wine decreases aortic stiffness and wave reflections acutely. Studies with de-alcoholized wine have shown that part of the effect is due to alcohol-induced vasodilation [14,16] and part to the high flavonoid content [17]. The short-term (6 weeks) effect of red wine has been recently studied in postmenopausal women. Neither red wine, nor de-alcoholized red wine induced any changes in wave reflections and central hemodynamics [18].

As regards endothelial function, the high flavonoid content of red wine, high in antioxidant capacity, has been pointed as the substance responsible for the beneficial effect of this drink on endothelial function.

2.5. Other foods and vitamins

Isoflavones – another type of flavonoids, found in soy and in red clover – have been shown to reduce systemic arterial compliance and aortic pulse wave velocity when given for a period of 5–10 weeks [19]. The n-3 fatty acids eicosapentaenoic acid and docosahexaenoic acid, found in fish, have a beneficial effect on arterial stiffness [20].

Furthermore, vitamin C (ascorbic acid) and vitamin E seem to favourably affect arterial elastic properties [21].

2.6. Salt

High salt intake has been associated with increased aortic stiffness in the general population [22] and in normotensive adults. Furthermore, salt-sensitive hypertensives have increased arterial stiffness compared to salt-resistant subjects with the same blood pressure levels [23].

High-salt diet results in a decrease in femoral distensibility in diabetic subjects compared to non-diabetics, which is reversed with angiotensin converting enzyme inhibition [24]. More importantly, a relatively brief period of sodium restriction decreases arterial stiffness. The same effect has been observed promptly, just one week after sodium restriction, in older adults with systolic hypertension [25].

The detrimental effect of high sodium consumption on arterial stiffness may be attributed, at least partly, to endothelial impairment. Furthermore, and probably as a consequence of the endothelial dysfunction, structural changes in the arterial wall further deteriorate elastic properties.

4.1. Obesity–weight loss

Obesity is not strictly a lifestyle component, but it is rather a disease. However, it is discussed in this review because it is largely the consequence of unhealthy diet and lack of exercise.

Obesity and especially central (abdominal) obesity is associated with increased cardiovascular risk. The effect of obesity on arterial stiffness has been extensively studied. Most studies have shown that obesity is associated with increased arterial stiffness in various arterial beds [32–34], although there are a few studies with opposite results [35]. Multiple explanations for this discrepancy have been proposed. It seems that different vascular beds are affected in different ways by obesity. Furthermore, the decreased stiffness found in some studies is related to an increase in arterial diameter and to a hyperdynamic circulation with decreased systemic vascular resistance. Finally, alterations in perivascular structures possibly offer less resistance to vessel distension [33].

Interestingly, the negative impact of obesity on arterial stiffness has also been investigated in children and adolescents [36,37]. Obese children have increased arterial stiffness and increased carotid intima-media thickness, denoting a faster progression to atherosclerosis [37].

Central fat accumulation has gained specific attention. In a cross-sectional study with prospective analysis it was found that truncus subcutaneous fat accumulation during adolescence was associated with increased central arterial stiffness at the end of the follow-up (age 36) [33]. On the other hand, peripheral fat, i.e. in the limbs, is associated with decreased arterial stiffness, possibly involving mechanisms such as those mentioned above. The main pathophysiological mechanisms linking central obesity and vascular dysfunction include the production of adipocytokines, with detrimental metabolic effects and an inflammatory stimulation. Also, high leptin level is independently associated with decreased radial artery distensibility in obese subjects [38].

The effect of obesity on arterial function is further stressed when changes in weight occur. As demonstrated recently, changes in body weight in a 2-year follow-up were positively associated with changes in aortic pulse wave velocity. While weight gain increased large-artery stiffness, weight loss had the opposite effect [39]. The beneficial effect of weight loss has been demonstrated also in various populations [32,40–42] and is independent of the diet used (meat- or plant-based) [41]. The improvement in arterial elastic properties observed with weight loss is largely explained by the concomitant reduction in blood pressure [41]; however the reduction in aortic pulse wave velocity observed with weight loss was independent of blood-pressure changes in one study [39].

4.2. Exercise

Exact definition of the association between exercise and arterial stiffness is often cumbersome because it may differ depending on the type (aerobic, resistance, combined) and the amount of exercise (leisure, athletic). Furthermore, it should be kept in mind that the initiation point of the series of events is difficult to define. In other words, while exercise may in itself lead to an improvement in elastic properties, it may also well be that those with already elastic arteries can more easily exercise for this very reason; thus, they are more often engaged in such activities, setting a positive feedback mechanism in operation.

Aerobic (endurance) exercise has a beneficial effect on arterial function. Increased aerobic capacity has been associated with reduced aortic stiffness and wave reflections [43]. Age-related increases in central arterial stiffness are blunted in postmenopausal women who are physically active [44]. The effect of regular exercise is evident mainly in central arterial stiffness [44], while peripheral arterial stiffness may not be altered [44]. An interesting finding was that the level of fitness is associated with large-artery stiffness even in children aged 9–11 years, denoting a possible protective effect of aerobic exercise even from this age [45]. The beneficial effect of physical activity has also been demonstrated in a study of female twins, in which subjects at high genetic risk of high augmentation index with leisure-time physical activity had lower augmentation index than non-participants [46].

The beneficial effect of aerobic exercise on arterial elastic properties has been elucidated using intervention studies. Aerobic exercise increases acutely total arterial compliance [47]. And this is probably related to vasodilation. Regular aerobic exercise has the potential to ameliorate central arterial stiffness in middle-aged sedentary men [48]. However, the same effect was not observed when patients with isolated systolic hypertension were trained for 8 weeks [49].

The mechanisms through which habitual exercise exerts its favourable effects on arterial elastic properties are not fully understood. Structural changes may ensue. In an interesting study in athletes performing asymmetric training of the two arms (hammer throwers and baseball players) there was a difference in radial artery structure between the two arms, which was accompanied by a significant increase in distensibility of the artery (muscular type) of the dominant arm [50]. The role of endothelium is unequivocal and a possible anti-inflammatory effect may also play a part.

Resistance (strength) exercise does not have the same effect on arterial stiffness as aerobic exercise. In fact, resistance exercise actually increases arterial stiffness acutely [51] and resistance-trained athletes have increased arterial stiffness [52]. Furthermore, the detrimental effect of resistance exercise on arterial stiffness was shown in an intervention study, in which four months of resistance training resulted in a significant increase in arterial stiffness; this increase was completely reversed after four months of detraining [53].

Finally, athletes performing combined aerobic and strength training, such as rowing, exhibit decreased aortic stiffness, denoting that endurance training is able to negate the stiffening effects of strength training [54].