Antihypertensive treatment-induced changes in arterial stiffness

From a physiology point of view and by its definition, the arterial tree should be one of the main – if not the ultimate – target organ of arterial hypertension. Indeed, arterial hypertension is associated with both functional and structural changes in small and large arteries, as well as, in the capillary bed [1]. These alterations include endothelial dysfunction, capillary rarefaction, arterial remodelling and stiffening [1]. Hypertension is therefore considered as a major player in the ‘early vascular ageing’ process as well as in advanced arterial disease [1]. In the last decades, the noninvasive assessment of arterial stiffness and the use of various biomarkers have led to better stratification of the cardiovascular risk in hypertensive patients and in the general population [2,3]. Therefore, individualization of antihypertensive drug treatment initiation by taking into account the level of arterial stiffness is feasible. Antihypertensive drug treatment leads to regression of arterial stiffness through functional [i.e. reduction of the distending blood pressure (BP)] and/or structural (i.e. remodelling) effects on the arterial wall [4–8]. However, evidence on the effect of long-term drug-induced BP control on the progression/regression of arterial stiffness remains very limited. Moreover, the role of the arterial stiffness in guiding long-term antihypertensive treatment has not been well studied. Significantly, recent evidence from large prospective epidemiological studies suggests that arterial stiffness, not only is a consequence of arterial hypertension, but may also antedate and predict the incidence of arterial hypertension – contributing thereby in its pathogenesis [9,10]. In this issue of the Journal, Tedla et al.[11] studied 906 hypertensive individuals and showed that effective long-term BP control was associated with a slower progression of carotid stiffness in a 10-year follow-up. However, this benefit was not achieved in diabetic or elderly patients, implying that BP control alone may be insufficient to slow the progression of established arterial stiffness, as in the case of people with diabetes and the elderly with hypertension [11]. These findings raise important questions regarding the association of BP with arterial stiffness: (i) which arterial biomarker and/or arterial site is more accurate for the assessment of the arterial stiffness to be used for risk stratification in hypertension, and (ii) which arterial stiffness index is more sensitive to track antihypertensive treatment-induced changes of the arterial wall in hypertensive patients and whether this change offers independent prognostic information. MAPPING THE ARTERIAL TREE AS A TARGET-ORGAN OF HYPERTENSION The ideal vascular biomarker should ensure proof of concept, independent predictive ability for cardiovascular risk with incremental value, slowing of progression or better regression with treatment that improves prognosis, good reproducibility and ease in clinical application [2,3]. Several vascular biomarkers have been introduced, which provide useful information for the burden of cardiovascular risk factors on the arterial wall, including carotid intima–media thickness along with plaque assessment, ankle-brachial index, arterial stiffness, central haemodynamics, and endothelial function (flow-mediated dilatation) [2]. Indeed, these biomarkers carry significant cardiovascular prognostic value and their clinical utility is focused on cardiovascular risk reclassification of individuals at ‘low’ or ‘intermediate cardiovascular risk’ [2,3]. Among these indices, arterial stiffening refers to reduced capability of an artery to expand and contract in response to pressure changes and is linked to the process of arteriosclerosis rather than atherosclerosis, being the hallmark of normal ‘vascular ageing’ [2]. In terms of physiology, age and hypertension are the main determinants of arterial stiffening, with numerous studies supporting this robust association [12–14]. Thus, the quantification of the arterial stiffness emerges as the ‘ideal’ tool for the evaluation of hypertension-induced arterial damage and, most importantly, for tracking its changes in response to the antihypertensive treatment. However, its clinical application faces important challenges. First, the elastic wall properties vary along the arterial tree, with the proximal arteries (e.g. aorta and carotid arteries) being more elastic and the distal (brachial) more muscular and stiff, whereas other arteries are rather of mixed nature (e.g. femoral) [15]. Thus, in the process of vascular ageing, it is expected that the arterial tree exhibits distinct segment-specific changes in terms of atherosclerosis and arteriosclerosis. Second, arterial wall properties exhibit diurnal variation, which is missed with single static assessments in the office. The recent introduction of 24-h ambulatory monitoring of arterial stiffness indices is expected to provide new insight and research challenges in this field [16]. Third, there is considerable heterogeneity in the technology and methodology for noninvasive assessment of the arterial stiffness [17]. In contrast to the systemic arterial stiffness, which can only be estimated from models of the circulation, regional and local arterial stiffness can be measured directly and noninvasively, using different methodologies (applanation tonometry, oscillometry, ultrasound or MRI-based approaches) at various sites along the arterial tree [17]. Table 1 presents the main arterial stiffness indices regarding the available evidence on their characteristics, cardiovascular predictive value and response to antihypertensive treatment [4–8,16–50]. At present, among all these indices, carotid–femoral pulse wave velocity appears to gather most of the necessary features of the ‘ideal’ index. Thus, it has been endorsed by the European Society of Hypertension as an arterial stiffness index for cardiovascular risk evaluation in hypertension [3].TABLE 1: Comparison of the main arterial stiffness indicesIn the study by Tedla et al.[11], carotid stiffness was used for long-term monitoring of arterial stiffness. In contrast to pulse wave velocity which is an index of segmental stiffness, carotid stiffness is an index of local stiffness and has independent predictive value particularly for stroke risk [18,40,41]. It is closely related to aortic stiffness evaluated by carotid–femoral pulse wave velocity, yet this association has been shown to become weaker in the presence of cardiovascular risk factors, such as hypertension or diabetes [51]. In addition, accumulating evidence suggests that carotid artery mechanics are influenced by the degree of atherosclerosis and the presence of carotid plaques, which lead to decreased carotid distensibility [52]. Thus, the behaviour of this index in response to antihypertensive treatment might not be predicted in terms of optimal BP control only and independent of other cardiovascular risk factors, especially diabetes. In conclusion, the effect of BP control on a specific arterial bed might differ due to anatomical factors, local pathologies such as atherosclerosis (plaques) and artificial factors specific to the applied methodology. These issues support the hypothesis that, at least in a research setting, a holistic arterial stiffness mapping might be required for the accurate assessment of the hypertensive individual [15]. ANTIHYPERTENSIVE TREATMENT-INDUCED CHANGE IN ARTERIAL STIFFNESS AND PROGNOSTIC VALUE Accumulating evidence from long-term observational studies and randomized controlled trials suggests that antihypertensive treatment reduces arterial stiffness, and this effect seems to be attributed not only to BP lowering but also to arterial remodelling [4–8]. Among all the available arterial stiffness indices, carotid–femoral pulse wave velocity has the strongest evidence regarding the effect of antihypertensive treatment on arterial ‘de-stiffening’ [4–8]. However, the evidence on the prognostic value of treatment-induced regression in arterial stiffness is very limited and has been obtained mainly using carotid–femoral pulse wave velocity (Table 1). More specifically, two prospective studies in patients with end-stage renal failure showed that the change in carotid–femoral pulse wave velocity was associated with the incidence of cardiovascular events [34] and that the absence of its improvement in response to BP decrease predicted all-cause and cardiovascular mortality [35]. The study by Tedla et al.[11] showed slower progression of carotid stiffness with optimal BP control, which is in line with the concept of the reciprocal causal association between elevated BP and arterial stiffening. However, elderly and diabetic hypertensive patients were exempt from this observation. Significantly, data from two recent community-based prospective studies questioned the direction and timing of the relationship between BP levels and arterial stiffness by showing that increased arterial stiffness predicted incident hypertension and initial BP did not predict arterial stiffening [9,10]. Prospective studies also showed baseline arterial stiffness to be an independent predictor of BP response to antihypertensive treatment [53,54]. Another explanation for the lack of an effect of optimal BP control on the progression of arterial stiffness in the elderly and patients with diabetes might be that, after a certain degree of arterial derangement, reversal of abnormalities is not possible. Indeed, advanced age and diabetes are the most powerful cardiovascular risk factors, and progression of arterial stiffening certainly is more pronounced in older adults, independent of the BP status [12]. A recent cross-sectional study in 351 patients showed disproportionately increased aortic stiffness in patients with diabetes compared with nondiabetic hypertensive patients [55]. Most importantly, age and glycated haemoglobin were the sole determinants of increased aortic stiffness [55]. Thus, in the presence of diabetes, several mechanisms such as medial calcification, oxidative stress, inflammation and advanced glycation end products accumulation might result in irreversible arterial alterations [55]. Because BP is a quantitative rather than qualitative variable presenting a linear relationship with the cardiovascular risk, the finding that the relationship between BP control and slower progression of carotid stiffness was more evident in patients with optimal BP control (SBP < 120 mmHg) compared with those with borderline control (SBP < 140 mmHg) [11] is not a surprise. This relationship almost reached statistical significance even in elderly hypertensive patients [11]. This observation is in line with recent studies showing that a more strict BP control is associated with greater cardiovascular benefit [56]. It should be mentioned that in the study of Tedla et al.[11], the number of elderly or diabetic hypertensive patients with controlled BP was rather small, which does not allow solid conclusions to be made on the impact of BP control on carotid stiffness progression in these subgroups. CONCLUSIONS Hypertension is closely associated with arterial stiffening and, although this association is robust and well established, the time sequence and the causal nature remain in question. The assessment of arterial damage and especially arterial stiffness in hypertension should take into consideration particularities between different arterial sites and different methodologies. A multisite arterial mapping appears to be a rational research strategy in terms of a more accurate assessment. The evaluation of the long-term antihypertensive treatment effect on arterial stiffness is influenced by several factors such as the index used, the level of BP control and the presence of risk factors including old age and diabetes. The study by Tedla et al. puts forward the notion that to reduce arterial stiffness, action is needed ‘not too late and not too little’. Further large-scale prospective studies should elucidate these important research gaps. ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.