Do diabetes and hypertension interact to accelerate vascular ageing?

Type II diabetes and hypertension are inter-related disorders, each powerfully predisposing to the development of the other and to cardiovascular disease, and each showing an increasing incidence with the ageing of the population [1]. Diabetes is almost 2.5 times more likely to occur in the presence of hypertension. Both share common cardiovascular risk factors, including a propensity for obesity. At any level of blood pressure, diabetic individuals have a two- to three-fold greater chance of developing cardiovascular complications than do their non-diabetic counterparts [2]. Furthermore, both have been associated with accelerated vascular stiffness, which in turn has been equated with accelerated vascular ageing [3]. The article by Schram et al. [4] in this issue of the journal should be evaluated within this context. Using pulse pressure (a surrogate measure of arterial stiffness) adjusted for age as the independent variable and cardiovascular mortality as the dependent variable, do diabetics have accelerated vascular ageing compared to non-diabetics? Hypertension can produce arterial stiffness by both functional and structural mechanisms [5]. With the development of hypertension and increased intramural pressure, the load-bearing elastic lamellae stretch and become stiffer on a functional basis. In contrast, there are longitudinal structural changes with ageing that cause increased stiffness of the thoracic aorta and its branches, but spare the more peripheral muscular arteries. Elastin, with a half-life of 40 years, is one of the stable proteins in the body. Despite this stability, fatigue of elastin fibres and lamellae can occur by the sixth decade of life from the accumulated cyclic stress of more than two billion aortic expansions during ventricular contraction. Longstanding cyclic stresses in the media of elastic-containing arteries produce fatigue and eventual fracturing of elastin, along with structural changes of the extracellular matrix, including proliferation of collagen and deposition of calcium. This degenerative process, termed arterioscleroses, is the pathological process that results in increase central elastic arterial stiffness [5]. Similarly, diabetes through multiple biochemical mechanisms related to insulin resistance, hyperinsulinaemia and hyperglycaemia, accelerates the degenerative ageing process of arterioscleroses [6]. Proposed pathways include an alteration in oxidative stress, polyol formation, protein kinase C activation with resulting signal transduction for hormones, cytokines and multiple growth factors, and the increased formation of advanced glycosylation end-products (AGE) [6,7]. Perhaps the most promising mechanism for accelerated vascular ageing in diabetes is the cross-linking of extravascular proteins to matrix components through AGE formation [7]. AGE accumulate slowly with ageing and produce cross-linking of collagen and elastin proteins, resulting in an increased stiffness of both arteries and the heart. This natural process is accelerated by diabetes. There is now abundant evidence that indivuduals with diabetes display premature large artery stiffness. Increased artery stiffness was noted in type 1 diabetics free of vascular complications, and who were matched with normoglycaemic controls, using ultrasound Doppler to measure carotid artery wall diameter and distensibility [8,9]. Similar case–control studies in type 2 diabetics with micoalbuminuria showed increased carotid artery stiffness compared to diabetics with normoalbuminuria [10]. Cross-sectional studies in the ARIC cohort showed increased stiffness but not increased thickness of the common carotid artery in individuals with both glucose intolerance and frank type 2 diabetes [11]. Using applanation tomography of radial arteries with derived central waveforms, type 1 diabetics showed increased aortic stiffness associated with an elevated augmentation index compared to non-diabetic controls [12,13]. Most impressive was the finding that type 2 diabetics with end-stage renal disease showed greater increased aortic stiffness by pulse wave velocity and had a greater 5-year cardiovascular mortality than their non-diabetic end-stage renal disease counterparts [14]. Schram et al. [4] used pulse pressure as a surrogate measurement for arterial stiffness. Following the introduction of the sphygmomanometer at the beginning of the 20th century, the significance of diastolic blood pressure (DBP), systolic blood pressure (SBP) and pulse pressure as markers of cardiovascular risk, in the presence or absence of diabetes, has undergone considerable change. Initially, DBP was thought to be the best measure of risk, but thereafter the focus shifted to SBP and, more recently, to pulse pressure. These shifts in emphasis on different components of blood pressure measurement resulted from a better understanding of the underlying haemodynamic forces that confront the left ventricle during the alternating phases of systole and diastole. A rise in blood pressure can result from an increase in vascular resistance downstream at the level of the arterioles and small arteries, by an increase in large elastic arteries stiffness upstream, or by a combination of both processes [5]. Both increases in resistance and stiffness raise SBP. By contrast, DBP rises with increased resistance but falls with increased stiffness; the relative contribution of each determines the level of DBP and, ultimately, the pulse pressure. The rise in pulse pressure is largely driven by ageing, with increases beginning after the age of 50 years and accelerating after age 60 years [15]. Recent studies have shown that pulse pressure, after adjusting for age, sex and multiple risk factors (including diabetes), is superior to SBP as a predictor of cardiovascular events [16–18]. Because SBP and pulse pressure rise with both increasing resistance and stiffness, these two blood pressure components are highly correlated. However, when assessed individually in a Framingham Study [19], increments in pulse pressure at a fixed SBP (cyclic stress) were associated with a greater risk of coronary heart disease than were parallel increments in SBP and DBP at a fixed pulse pressure (static stress). Thus, despite the high correlation of pulse pressure with SBP, pulse pressure is the dominant predictor of coronary heart disease risk in individuals with discordantly low DBP, and this form of blood pressure elevation, termed isolated systolic hypertension (ISH), is predominantly found in the elderly. The Hoorn Study [4] is a Dutch population-based cohort of individuals, aged 50–74 years at baseline, with normal glucose tolerance, impaired glucose tolerance and frank type 2 diabetes, who were followed for incidence of cardiovascular death from 1989 to 2000. The authors concluded that this was the first cohort study to show that pulse pressure, independent of age, was a predictor of cardiovascular mortality in individuals with type 2 diabetes. First, there was both a greater increase in SBP and a greater decrease in DBP in the diabetic population, suggesting that pulse pressure was a better predictor of cardiovascular risk than SBP. Second, there was a stronger association of increased pulse pressure per year of age in the diabetic compared to the non-diabetic population, suggesting that diabetes is a state of accelerated ageing. The success of the authors in proving their hypotheses depended in part on having an appropriate reference group of non-diabetics in their population-based cohort study. There are certain caveats to consider when using pulse pressure as a surrogate measure of aortic stiffness and as a predictor of cardiovascular risk. First, pulse pressure can be affected by changes in heart rate, cardiac contractility, stroke volume or vasomotor tone [5]. Second, and most importantly, pulse pressure measured at the brachial artery is greatly influenced by the phenomenon of peripheral amplification [5]. Pulse wave propagation and reflection varies considerably according to age and the extent of elastic artery stiffness. The summation of the forward and reflective wave results in progressive amplification of the SBP and pulse pressure from the aorta to the brachial artery where blood pressure is customarily measured. With increasing age or hypertension, there is progressive stiffening of central elastic arteries and a resulting decrease in peripheral amplification of the arterial pressure wave [20]. As arteries stiffen and pulse wave amplification decreases with ageing, there is a gradual shift from DBP to SBP, and eventually to pulse pressure, as predictors of cardiovascular risk [21]. In middle-aged, healthy populations or older individuals with both systolic and diastolic hypertension, SBP and mean arterial pressure may be equal or superior to pulse pressure as a predictor of cardiovascular risk [22]. Indeed, the Hoorn study [4] did not find that pulse pressure was superior to SBP in predicting cardiovascular mortality in the non-diabetic population; this could be explained by a ‘healthy cohort effect', as demonstrated by an overall low mortality rate. In contrast, wide pulse pressure isolated systolic hypertension, after adjusting for appropriate risk factors, was shown to be an independent predictor of coronary heart disease risk in individuals aged over 60 years [16–18]. Moreover, pulse pressure becomes an even stronger predictor of risk at a younger chronological age in individuals with prevalent cardiovascular disease [23] or target organ damage [24], or in the presence of multiple associated cardiovascular risk factors [25]. As noted by Schram et al. [4], the study was limited by having only 34 cardiovascular deaths in the diabetic population, which diminishes the precision of the risk estimates and suggests that a longer follow-up may be desirable to strengthen the conclusions of the current study. Furthermore, Schram et al. [4] reasoned that pulse pressure must be adjusted for mean arterial pressure in order to correctly predict cardiovascular risk. This adjustment is necessary for the evaluation of pulse pressure as a predictor of cardiovascular risk when SBP is not used as a reference point. However, brachial artery pulse pressure cannot replace SBP as a single measure of cardiovascular risk. Pulse pressure, a good measure of arterial stiffness, may significantly underestimate peripheral vascular resistance [26]. Overall, SBP is the single best measure of cardiovascular risk because it rises with increases in both resistance and arterial stiffness. Thus, the best strategy for assessing risk in the elderly (or in the diabetic with accelerated ageing) begins with determining the level of SBP elevation, and is followed by the adjustment of risk upward when wide pulse pressure (i.e. discordantly low DBP) is present. This approach is identical to jointly placing SBP and DBP in the same Cox regression model to assess cardiovascular risk. Adjusting pulse pressure to the level of mean arterial pressure is an alternative approach, although this may be less accurate in the elderly because the mean arterial pressure equation underestimates vascular resistance as DBP decreases with ageing and with diseases that accelerate arterial stiffness [26]. The observation that pulse pressure rose in direct proportion to the elevation of mean arterial pressure in the diabetic group suggests that hypertension and diabetes interact to accelerate vascular ageing. As noted by Schram et al. [4], the findings from the Framingham study [15] also support an interaction between ageing and hypertension in the progressive rise in pulse pressure that begins with the onset of middle-age. Subjects with mean baseline blood pressures of 111/70 mmHg had no rise in pulse pressure and only a minimal increase in mean arterial pressure at 30–49 years. Nevertheless, this group with optimal blood pressure showed a significant rise in pulse pressure and fall in DBP after 50–59 years, presumably caused by an increase in elastic artery stiffness secondary to ageing. In contrast, individuals with baseline mean blood pressures of 173/90 mmHg showed a steeper rise in SBP and a steeper fall in DBP after 50–59 years than was observed in the optimal blood pressure group. This divergent rather than parallel tracking pattern has been referred to as the horse-racing effect, there being a close correlation between the speed of the horse and its position in the race. These findings suggest a link between hypertension left untreated and the subsequent acceleration of large artery stiffness. This, in turn, can perpetuate a vicious cycle of accelerated hypertension and further increases in large artery stiffness, leading to accelerated ageing of conduit arteries. On the basis of the findings of Schram et al. [4], the coexistence of untreated hypertension and diabetes further contributes to this vicious cycle. Finally, what are the therapeutic implications involving individuals with accelerated arterial stiffness and impaired glucose tolerance or type 2 diabetes? Drugs that block the renin–angiotensin system have been shown to improve arterial distensibility [27,28] and have been proven to be useful in: (i) preventing the progression of impaired glucose tolerance to frank diabetes in high-risk coronary heart disease patients [29] or in those with advanced left ventricular hypertrophy [30]; (ii) preventing progression of diabetic microalbuminuria to clinical proteinuria [31]; (iii) and reducing cardiovascular events in both normotensive and hypertensive patients with diabetes [29]. Furthermore, early instituted therapy may prevent the age-related increases in SBP and pulse pressure and decreases in DBP that lead to ISH in the elderly. Again, agents that inhibit the renin–angiotensin–aldosterone system may be useful in preventing age-related deterioration in elastin protein and collagen deposition that occurs in conduit arteries. Finally, agents that maximize the reduction in SBP and pulse pressure while minimizing the reduction in DBP may be especially effective in treating diabetic patients with ISH. Long-acting nitrates [32–34], by delaying the return of reflected pressure waves and thereby reducing central SBP and cardiac afterload without lowering DBP, have proven useful in the long-term treatment of elderly patients with ISH. Whether nitrate therapy, in conjunction with other antihypertensive therapy in the diabetic patient, further lowers cardiovascular morbidity and mortality remains to be determined. In addition, an experimental advanced glycation end-product crosslink breaker (ALT-711) has been shown to improve total arterial stiffness in older individuals with vascular stiffening [35]. This may provide a novel approach to treating premature diabetic arterial stiffening if and when this agent can be shown safe and effective in reducing cardiovascular events. Finally, polypharmacy involving the use of two or more drugs is often necessary to reach the diabetic target goal of < 130/80 mmHg, as advocated by national and international guidelines [36,37]. Clearly, the findings of the Hoorn Study [4], showing that pulse pressure best predicts increased cardiovascular mortality in type 2 diabetes, support an early, aggressive and specific therapeutic approach towards correcting accelerated arterial stiffness.