Dietary polyphenol intake and risk of hypertension in the Polish arm of the HAPIEE study

In this study, we investigated the association between total and individual classes of polyphenol intake and incidence of hypertension in urban population. We found that women with higher intake of polyphenols were less likely to develop hypertension compared to lower ones. Among specific classes and subclasses of polyphenols, higher consumption of flavonoids, phenolic acids, and “other” polyphenols was inversely associated with incident cases of hypertension. These results are in accordance with previous findings provided by a cross-sectional study conducted on the same cohort and testing the baseline relation between dietary polyphenol and presence of metabolic syndrome [21]; the study showed that among the main polyphenol classes, phenolic acids, flavonoids, and stilbenes were associated with lower odds of hypertension [21]. To date, only a limited number of epidemiological studies evaluated the association between total and individual classes of dietary polyphenols and cardiovascular outcomes, including hypertension [13, 14]. Despite no significant results were found in the analysis including total dietary polyphenol intake in a study conducted on Iranian adults, among individual classes sub-analyses revealed that phenolic acids and flavonoids were negatively associated with blood pressure, whereas flavonoids and stilbenes were negatively associated with hypertension [13]. Another study conducted in the context of the PREDIMED (PREvencion con DIeta MEDiterranea) cohort reported lower prevalence of hypertension among individuals with higher intake of dietary polyphenols [14] and further investigation in subgroup analysis revealed that also total polyphenols excretion was associated with blood pressure levels and prevalence of hypertension [22] as well as with plasma nitric oxide production, which is a well-known regulating factor of endothelial function [23]. Human diet contains a great variety of polyphenols but only some compounds and derivatives are bioactively relevant to endothelial function [24]. Moreover, differences in polyphenol food content databases and food items ascertained on dietary questionnaires may justify lack of significant findings in some of the aforementioned observational studies. In this study, we showed a wide picture of the possible association between polyphenol intake and hypertension, suggesting that independent associations of several polyphenol classes may exist, despite a concomitant and an overall synergic effect in preventing could be determinant in preventing hypertension. Up to date, this is one of the most comprehensive studies in terms of polyphenols investigated simultaneously considering hypertension as the outcome.

Results from this study may explain, at least in part, previous findings from meta-analyses of randomized controlled trials regarding the effect of flavonoids in lowering blood pressure, as they have been reported to be effective in hypertensive patients, but not in normotensive individuals [25, 26]. We reported that flavonoids were significantly associated with incident cases of hypertension when considered separately in the individual class analysis. Regarding subclasses of flavonoids, findings are contrasting and generally limited to some groups, such as anthocyanins [7] and isoflavone [8]. We did not found any significant result regarding these aforementioned subgroups of flavonoids, rather limited to flavanols. Literature on flavanol-rich food is wide, and findings from the few existing clinical trials show a potential blood pressure lowering effects of cocoa [27], while evidence on anthocyanins is contrasting [28, 29]. There is also evidence on the HAPIEE cohort that high adherence to dietary patterns rich in fruit and vegetable (among the main sources of flavonoids) were associated with better health outcomes compared to lower adherence [30‐32]. Flavonoids have been hypothesized to exert beneficial effects in cardiovascular health by ameliorating the inflammatory status at various levels, for instance decreasing circulating levels of TNF-α and IL-6 [33]. Regarding their potential effects on blood pressure, flavonoids can act as inhibitors of endothelial NAD(P)H oxidase [34, 35], an enzyme implicated in the regulation of NO metabolism in the vascular endothelium, which in turn regulate vasodilatory processes associated with blood flow. Indeed, ingestion of pure epicatechin in humans has been demonstrated to increase NO bioavailability and to acutely reduce plasma concentrations of endothelin-1, which is a potent endothelium-derived vasoconstrictor [36]. Among other potential mechanisms of actions, flavonoids lower the activity of arginase-2, which is an enzyme that competes with NO synthase for l-arginine [37] and inhibit activity on angiotensin-converting enzyme in vitro [38].

Among the most studied hydroxycinnamic acids, chlorogenic acids (CGAs) (compounds richly contained in coffee) contained in coffee have been reported to exert beneficial effects towards CVD risk factors and metabolic disorders [39]. Meta-analyses of experimental studies CGAs reported significant reduced blood pressure compared placebo treatment [40], whereas results of observational studies on coffee consumption and hypertension showed a U-shaped association [41] and overall lower risk of metabolic syndrome [42]. Previous studies conducted on the HAPIEE cohort showed an inverse association between coffee consumption and metabolic syndrome and risk of hypertension [43, 44]. Together to their anti-oxidant properties, CGAs have been found to exert direct effects in regulation of blood pressure, as well as glucose and lipids metabolism [45]. CGAs have been hypothesized (i) to exert antihypertensive effects attenuating oxidative stress (reactive oxygen species) by reducing NAD(P)H-dependent super-oxide production and ameliorating endothelial dysfunction and (ii) to interact with the renin–angiotensin aldosterone system by inhibiting angiotensin-converting enzyme activity both in vitro and in vivo [46, 47].

Studies on stilbenes have been mostly focused on the effects of resveratrol on cardio-metabolic health [48, 49], revealing that resveratrol consumption significantly decreases systolic blood pressure at high dose [50]. We found inconclusive results on the risk analysis regarding the association between dietary stilbenes intake and hypertension risk, despite a linear association with individual intake and blood pressure measurements was found in men. Previous inconclusive prospective investigations on resveratrol efficacy on CVD outcomes suggested that the main limitation of studying stilbenes intake relied on the very small amount consumed [51]. The present study may be affected by similar limitation, as contribution of “normal” diets in stilbenes is generally negligible and, thus difficult to accurately estimate and associate with health outcomes [52].

Besides the most studied polyphenol classes, in this study, we opened the debate on other classes of polyphenols that may exert beneficial effects toward uncontrolled blood pressure and have not been previously explored in epidemiological studies. This study adds evidence that other polyphenols, such as tyrosol, may play an important role in decreasing the risk of hypertension. Studies on less studied polyphenols are highly warranted because they could explain the favorable effects of foods high in such compounds that have been inversely associated with hypertension risk. For instance, tyrosol and hydroxytyrosol contained in olive oil [53, 54] and some alcoholic beverages [55, 56] have been reported to reduce blood pressure, and may explain the beneficial effects related to cardiovascular prevention by consumption of the aforementioned food sources (moderate for alcohol). More research is needed on this topic, and further evidence will help to overcome potential limitation of observational studies on dietary polyphenol intake and health outcomes.

Strengths of the present study are prospective design, large size, and reliable assessment of individual diet. However, some methodological issues should be addressed when considering results from this study. First, we cannot conclude causation due to observational study design. Second, baseline evaluation of food intake may have introduced misclassification, because diet may have changed over the follow-up period. Third, certain polyphenol-rich foods, such as herbs and spices, may not have been entirely captured by the FFQ. Forth, use of table content databases would have inevitably led to some misclassification of polyphenol intake. However, such issues are common to all previous studies using the same methodology, and since polyphenol exposure was ascertained before diagnosis of disease, misclassification would tend to bias estimates toward the null and underestimate true associations. Finally, we found significant results only among women. A possible explanation is the natural differences occurring in men and women due to hormonal protection in the latter. However, potential differences need to be further investigated.

In conclusions, we reported that higher intake of some classes of polyphenols are associated with lower risk of hypertension. Results from the present study underline the importance of investigating all classes of polyphenols as possible determinants of health outcomes. Further studies are needed to clarify whether polyphenol classes consumed in small amounts may exert beneficial effects on health if consumed in higher concentrations and to establish the effects specifically attributed to each polyphenol class.