Metabolic syndrome and arterial stiffness: evidence for gender disparity and early effects of non-traditional risk factors?

The concept of the entity currently known as the ‘metabolic syndrome’ (MetS) has a history almost as long as the history of its major components; as with many other ‘syndromes’, this history started from clinical observations early in the twentieth century [1] and continues up to the international efforts to define the MetS [2,3] and the well-known recent debate on the clinical utility of its use [4–6]. The two facts that all the sides agree upon are that the basic clinical components of the MetS (i.e. hyperglycemia, elevated blood pressure, central obesity, low high-density lipoprotein-cholesterol and high triglycerides) are risk factors for cardiovascular disease (CVD) and that these factors undoubtedly occur together more often than expected by chance alone [4–6]. With regard to ‘terminology’, only the latter fact of the ‘coexistence’ of these factors rather justifies the use of the term ‘syndrome’, which, by simple means, is just a group of signs/symptoms/disturbances that commonly cluster together. With regard to the underlying pathogenesis of the MetS, it is true that an important amount of information is missing [4], but it is also true that our knowledge in the field is more advanced than that for other common medical disorders, the recognition of which has never been debated [5,6]. With regard to the ‘predictive power’ of the MetS for cardiovascular disease, it can be argued that the critique mainly refers to the proposed definitions for it, which were not developed to this end, but rather as ‘easy to use’ tools to identify high-risk individuals in clinical practice; this is the reason why several typical members of the MetS clustering, such as insulin resistance, hyperinsulinemia, low-grade vascular inflammation, prothrombotic state and endothelial dysfunction, which are increasingly recognized as non-traditional CVD risk factors (but are less easy to measure), were simply not included [1–3,6,7]. Based on current evidence, it seems that the existence of the MetS clustering, as well as its high prevalence in western societies and its public health impact, cannot and should not be denied [2,3,6,7]. On the other hand, further epidemiological and background research to delineate the associations of the ‘total’ MetS clustering, the MetS under the current definitions and each of the ‘major’ and ‘hidden’ traits of it with CVD is awaited and improvement of the current definitional tools in relation to prognosis is welcome. Increase in the stiffness of the arterial tree with age, especially of central compartments, has long been implicated in the pathogenesis of cardiovascular disease [8]. Recent longitudinal data suggest that arterial stiffness of the aorta, measured with pulse wave velocity (PWV) is an independent predictor of coronary artery disease, stroke, and cardiovascular mortality [9,10], whereas peripheral arterial stiffness was associated with peripheral vascular disease [11]. To explore a potential mechanism connecting the MetS with increased cardiovascular risk, a considerable number of studies in the past decade examined the associations of the MetS clustering with arterial stiffness [12–20]. These studies provide important information, but, with few exceptions [14,15], examined only one specific arterial segment. Thus, whether the associations of the MetS with arterial stiffness are similar between central and peripheral compartments remains to be clarified. On the other hand, accumulating evidence suggests that the effect of diabetes and the MetS on cardiovascular morbidity and mortality is stronger in women than in men [21]. Again, the majority of studies on the relation of the MetS with arterial stiffness were conducted in single-sex cohorts or used sex-adjusted analyses, whereas the few studies that examined separately the two genders [13,18] supported the existence of such an association only in females, findings indicating the need for further investigation. As mentioned above, an important aspect of the relationship of the MetS with CVD risk is the fact that, for reasons of simplicity, several components of the clustering are not included in the major ‘definition’ approaches [1], whereas they are known to be associated with increased cardiovascular risk. To this end, these inter-related factors (i.e. insulin resistance, hyperinsulinemia, proinflammatory and prothrombotic state and endothelial dysfunction), which also have deleterious effects on arterial wall, could be potential mediators of the association between the MetS and arterial stiffness. Some studies have examined relations of some of these parameters with arterial stiffening [12,16,17]. However, their role as mediators of the association of the MetS defined on the basis of its basic traits and arterial stiffness has not been investigated. In this issue of the journal, Ferreira et al. [22] report on a cross-sectional analysis of data on 313 men and women aged 20–25 years who participated in the third screening phase of the Northern Ireland Young Hearts Project. The study adds important new information on the aforementioned associations because it attempts to simultaneously examine the above issues, namely the associations between the clustering of the MetS traits and both central and peripheral arterial stiffness, the possible differences of these associations between the two genders, as well as whether these relations are mediated by less commonly measured components of the MetS, such as insulin resistance and low-grade inflammation. The investigators observed that, after adjustment for several known confounders of arterial stiffness (sex, age, body weight and height and mean blood pressure) and other cardiovascular risk factors (low-density lipoprotein cholesterol, smoking and alcohol consumption status, and cardiorespiratory fitness levels), the PWV of all the arterial segments investigated (aorto-iliac, aorto-radial and aorto-dorsalis pedis segment) increased significantly with increasing traits of the MetS in women, whereas there were no such associations in men. Multiple linear regression analyses were further used to investigate whether the observed associations between the MetS score and arterial stiffness in women were confounded and/or mediated by several related risk factors. Although these relationships were not disturbed by traditional risk factors not included in the syndrome, the strength of the association between the MetS and the PWV of the aorto-iliac segment decreased when adjustments for heart rate, HOMA-IR and markers of inflammation (C-reactive protein and fibrinogen) were performed, that between the MetS and PWV of the aorto-radial segment attenuated after adjustments for heart rate and HOMA-IR, and that between the MetS and PWV of the aorto-dorsalis pedis segment attenuated after adjustment for heart rate. Apart from examining multiple questions, an important strength of the study is the choice of a young population, who were apparently free of major cardiovascular complications. This cohort seems suitable for studying associations that correspond to pathophysiological alterations occurring early in the long course of atherosclerotic and arteriosclerotic vascular damage (i.e. the relationships of insulin resistance and low-grade inflammation with early steps of arterial stiffening). The researchers should be also credited with the fact that they went through a detailed protocol for subject evaluation, as well as a thorough analysis. In addition, instead of getting involved in the recent debate on the relationship of MetS definitions with CVD risk [4,6], they appear to focus on the deep nature of the clustering by carefully handling the issue of MetS definition with two approaches; (i) to divide their population in categories, they have used cut-off points for the basic traits of the MetS that, although less strict, seem reasonable for their young subjects and (ii) they calculated a continuous MetS score with the use of sex-specific z-scores for each trait. The major limitation of this study relates to the cross-sectional form of the analysis, which precludes the establishment of longitudinal associations between the traits of the MetS and arterial stiffness and, thus, drawing conclusions in terms of causality. Another point of critique relates to the fact that the cut-off points used for the dichotomization of each risk factor of the MetS were based on the specific population distribution for these variables and were therefore unrelated with the thresholds included in the National Cholesterol Education Panel [2], or the International Diabetes Federation [3] definitions. As the authors acknowledge, the use of any of those definitions would result in a very low prevalence of the MetS and thus a limited power to address the relevant research questions. However, because all the major components of the MetS show a graded association with CVD risk [4], and the level where the risk increases may be less clear in young adults [23], this approach seems to be reasonable for examining ‘epidemiological’ associations in an attempt to uncover underlying ‘etiological’ connections. Finally, although the choice of a young healthy population is largely justified, it is not known to what extent the findings of this study apply to populations of older age, different risk profile, as well as those of non-Caucasian origin. The observations of Ferreira et al. [22] add to the existing evidence supporting an adverse association of the MetS with arterial stiffness and indicating a possible pathophysiologic link of the clustering with CVD [12–20]. Of note, in the subgroups where the significant associations were noted (i.e. in female subjects), the adverse effects of the MetS were similar throughout the arterial tree. This finding contradicts data from previous studies that examined more than one arterial segment, where the MetS was associated with increased stiffness of central but not peripheral arteries [14,15]. The latter observations can relate to the markedly different structure and function between central arteries, which are characterized by low stiffness and high sensitivity to ageing and blood pressure, and peripheral arteries, which have higher stiffness and are sensitive mainly to vasoactive substances of endothelial origin [8]. They can also underlie the different associations of central and peripheral arterial stiffness with cardiovascular outcomes [9–11], as mentioned above. However, Ferreira et al. [22] noted another important difference between central and peripheral parts of the arterial tree. In multiple linear regression analyses, performed to investigate possible mediators of the observed associations, insulin resistance and low-grade inflammation affected the association of the MetS with central arterial stiffness, only insulin resistance affected the association with arterial stiffness of the upper limb and none of the two affected the respective association with arterial stiffness of the lower limb. These observations again support differences between the two parts of the arterial tree with regard to the possible mechanisms that could mediate the effect of the MetS traits on arterial stiffness, as the authors nicely discuss [22]. In addition, these analyses, performed for the first time in this particular field, have another, perhaps more important, implication. None of the observed associations was influenced by other traditional risk factors, but that between the MetS and central arterial stiffness was in part attenuated by markers of insulin resistance and inflammation. Therefore, this study indirectly supports the hypothesis that members of the clustering other than those included in the major definitions play important roles in the MetS-associated vascular damage [6,7]. In addition to the above, perhaps the most intriguing finding of this study is that the MetS clustering was significantly associated with arterial stiffness only in women [22]. As recently discussed elsewhere [21], several lines of evidence support that the impact of both diabetes and the MetS on cardiovascular outcomes is greater in women than in men. This study adds to current knowledge in the field by reporting differences between sexes in the association of MetS with an intermediate end-point (i.e. stiffening of the arterial tree). The two previous studies on this relationship that performed separate analyses for the two genders obtained similar findings [13,18]. Among the plausible explanations for this gender disparity in cardiovascular outcomes are a heavier burden of CVD risk factors in the context of MetS (a higher number or a more severe degree of these abnormalities), a greater direct impact of a given combination of the MetS clustering on CVD, or the suboptimal management of the various risk factors in women compared to men [21]. The study by Ferreira et al. [22] is in support of the first explanation, as the authors noted a different risk profile across increasing traits of the MetS with regard to inflammatory markers, which were higher in women. The possibility that a heavier burden of ‘hidden’ risk factors is responsible for the more pronounced effect of the MetS and diabetes on CVD in women is certainly tantalizing and calls for further examination. Overall, Ferreira et al. [22] illustrate the increase in arterial stiffening in women with moderate elevations of factors included in the MetS and the contribution in this process of parameters such as insulin resistance and low-grade inflammation. These results highlight the need for better primary prevention strategies of all the MetS components in the female population. In addition, they emphasize the need for further research to elucidate this gender disparity in MetS outcomes, as well as the relevant contribution of each of the MetS traits in CVD risk.