Ncpcardio_2005_031.indd

Large-artery stiffness, hypertension and
cardiovascular risk in older patients
Jacques Blacher and Michel E Safar*
INTRODUCTION
High blood pressure is a powerful cardiovascular Research studies on hypertension have, so far, generally focused on vascular
risk factor because of its actions on the arterial resistance and small arteries. The high prevalence of systolic hypertension
wall, and its role in events that affect the heart in patients older than 50 years and the development of noninvasive Doppler
and the brain. Drug treatment of hypertension and echotracking techniques have made it possible to determine large-artery
can prevent such complications. Two specific stiffness with a high degree of reproducibility. Increased arterial stiffness
points of the blood-pressure curve, the peak and disturbed wave reflections are the basis for understanding reduced
systolic blood pressure and the end-diastolic aortic elasticity and systolic hypertension, particularly in older people. This
blood pressure, are used to clinically define high hemodynamic pattern results from mechanical factors and other pressure-
blood pressure (Figure 1). Hemodynamically, independent risk factors, such as diabetes mellitus, renal failure, obesity and
however, the overall shape of the blood-pressure severe atherosclerosis. Distinct phenotypes of arterial stiffness and pulse
curve and the cross-sectional area under the curve, pressure result from specific gene polymorphisms, such as those related
rather than merely its maximum and minimum to the renin–angiotensin system. The roles of arterial stiffness and wave
values, should be taken into account to describe reflections in hypertension have been elucidated by modern interpretations
the mechanical effects on the arterial wall and to of the blood-pressure curve in relation to its propagation, mechanisms of
give an adequate description of hypertension.
systolic-blood-pressure amplification, and the pulse-pressure amplitude.
The blood-pressure curve might be consid- New predictors of cardiovascular risk have been identified, such as increased
ered as the superposition of a steady compo- pulse pressure and pulse-wave velocity, and disturbed wave reflections, all
nent—mean arterial pressure—and a pulsatile of which are independent predictors of cardiovascular risk that are more
component—pulse pressure (Figure 1).1 Mean powerful than either systolic or diastolic blood pressure alone. Therapeutic
arterial pressure is the steady pressure resulting trials are investigating ways to reduce stiffness, and thereby allow the selective
reduction of systolic and pulse pressure in hypertensive patients with or
reach the peripheral tissues and organs. Pulse without advanced renal failure. Modern pharmacologic agents need to be
pressure is a consequence of the intermittent identified, which could reduce systolic hypertension in patients older than
ventricular ejection from the heart. The role of 50 years. Here we discuss the structural and functional factors that influence
arterial stiffness, wave reflections and pulse pressure in hypertension, as well
large conduit arteries, particularly the thoracic as their related roles in cardiovascular risk.
aorta, is to dampen the pressure oscillations that result from intermittent ventricular ejection. KEYWORDS arterial stiffness, cardiovascular risk, hypertension,
wave reflections
In addition to cardiac factors, therefore, the main determinants of pulse pressure and peak REVIEW CRITERIA
systolic blood pressure are the cushioning We carried out a search for original articles that were focused on “arterial capacity of arteries, and the timing and inten- stiffness” and “pulse pressure”, and were published between 1966 and 2005, sity of WAVE REFLECTIONS.1 The cushioning in MEDLINE and PubMed. All the papers that were identified were English-language, full-text articles. We also searched the reference lists of identified capacity of arteries is influenced by the visco- elastic properties of the arterial wall and, thus, by arterial stiffness, which is usually expressed J Blacher and ME Safar are physicians and research workers in the Diagnosis in the quantitative terms of compliance and Center, Hôpital Hôtel-Dieu, Paris, France. distensibility. Wave reflections result from the cumulative action of a forward wave coming Correspondence
from the heart—which propagates at a given *Diagnosis Center, Hôpital Hôtel-Dieu, 1 Place du Parvis Notre-Dame, 75181 Paris, Cedex 04, France speed, the pulse-wave velocity (PWV)—and a backward wave returning towards the heart from particular reflection sites.1 Received 11 March 2005 Accepted 13 June 2005
During the past few years, arterial stiffness and www.nature.com/clinicalpracticedoi:10.1038/ncpcardio0307 wave reflections have been widely investigated in 450 NATURE CLINICAL PRACTICE CARDIOVASCULAR MEDICINE 2005 Nature Publishing Group
GLOSSARY
WAVE REFLECTION
Reflections of the pressure
wave generated by the
heart, which originate from
each discontinuity of the
Figure 1 Typical features of a blood-pressure curve. This curve is usually defined by the upper and lower
extremities, which show peak systolic blood pressure and end-diastolic blood pressure, respectively. A
better way to define blood pressure as shown by the curve is to describe the totality of the curve, which
is composed of a steady component (i.e. mean arterial pressure) and a pulsatile component (i.e. pulse
pressure); these features represent hemodynamic fluctuations of the curve around the mean arterial
pressure and describe the mechanical effects on the arterial wall, thereby giving an adequate definition
of hypertension.
elderly patients both with and without hyper- in the arterial tree. From these sites of discon- tension. Whereas end-diastolic blood pressure tinuity, which are mainly located at the origins was previously considered to be the better guide of arteriole branches, reflected waves are for determining disease severity, epidemiologic generated that travel backwards towards the studies have shown that peak systolic blood ascending aorta (Figure 2). Therefore, incident pressure is a more informative cardio vascular and reflected waves are in constant interaction risk factor, particularly in patients older than along the arterial circuit and are superimposed 50 years.2 Furthermore, it has been shown so that each becomes a part of the actual blood- that pulse pressure—calculated as the differ- pressure wave. The final amplitude and shape ence between the peak systolic blood pressure of the measured aortic blood pressure wave are and the end-diastolic blood pressure—is an determined by the timing, or phase relationship, independent predictor of cardiovascular risk, between the two component waves. This, in turn, particularly for myocardial infarction.3 Finally, depends on the PWV, the traveling distance of although drug control of end-diastolic blood the pressure waves and the duration of ventric- pressure to achieve values lower than 90 mmHg ular ejection. In young people, under physio- can be consistently obtained in treated hyper- logic conditions, the backward pressure wave tensive patients, it remains difficult to control returns from the distal arterial vessels during peak systolic blood pressure, and values lower diastole, which makes the pulse pressure higher than 140 mmHg are more difficult to achieve.4 in the peripheral arteries than in the central The purpose of this review, therefore, is to arteries. This physiologic phenomenon, which is describe the structural and functional factors called pulse-pressure amplification (Figure 2),1 that influence arterial stiffness, wave reflections is highly influenced by the PWV. With an and pulse pressure in people with hyper tension increased PWV, the reflecting sites of the distal and to determine the role of these variables arterial tree seem to be closer to the ascending in cardiovascular risk. On the basis of these aorta, and the reflected waves occur earlier and considerations, new approaches are proposed are more closely in phase with incident waves to further reduce cardiovascular risk in elderly in this region. The earlier return of wave reflec- tions means that the reflected wave impacts on the central arteries during systole and not BASIC CONCEPTS
during diastole. This disturbed signal results in The ejection of blood into the aorta generates an augmentation of aortic and ventricular pres- a pressure wave that is propagated at a given sures during systole, and reduces aortic pressure velocity to other arteries throughout the body during diastole. Hence, the altered mechanical (Figure 2). Under physiologic conditions, the properties of the aortic wall influence the levels aortic PWV is approximately 5–7 m/s. This of aortic peak systolic blood pressure and end- forward wave is reflected at all points of structural diastolic blood pressure, which are increased discontinuity, functional discontinuity or both and decreased, respectively, as a consequence of SEPTEMBER 2005 VOL 2 NO 9 BLACHER AND SAFAR NATURE CLINICAL PRACTICE CARDIOVASCULAR MEDICINE 451 2005 Nature Publishing Group
CARDIOVASCULAR RISK FACTORS
Physiologically, mean arterial pressure remains similar throughout the arterial tree, whereas vessel stiffness increases progressively from the proximal to the distal part of the vascular trajec- tion. These changes in elasticity result from a combination of factors: the continuous decrease of vessel cross-sectional area along the arterial tree, the progressive increased rigidity of the vascular wall material and the physiologic rise in pulse pressure from the central arteries to Propagation of the blood-pressure wave. After ventricular ejection, a forward wave propagates at a given pulse-wave velocity along the arterial the peripheral arteries. An increase in the stiff- tree (1). The wave is reflected at any discontinuity of the arterial wall (reflection ness gradient of the wall material is observed sites; 2), and superposition of the forward and reflected waves can be when the proximal arteries are compared with observed at any point of the arterial tree, but is particularly evident at the site of the distal arteries because vascular elasticity is the thoracic aorta (3). The effect of superposition is to make the peak systolic higher in the proximal arteries. In humans, the blood pressure and the pulse pressure significantly higher in the peripheral increase in stiffness between the carotid artery arteries than in the central arteries.
and the radial artery is approximately 25% in healthy people.1,5 With increasing age, the gradient decreases significantly, because the age- GLOSSARY
early wave reflections. Finally, a disturbed pres- mediated increase in arterial stiffness progresses YOUNG’S MODULUS (E)
sure signal arising from the distal arteries might more rapidly in the central arteries than in the alter the heart–vessel coupling through early peripheral arteries.1,5 This faster increase of represented by the slope of the stress–strain relationship wave reflections and lead to increased cardio- stiffness in the central arteries is independent vascular risk. This signal, through an increase of the mean arterial pressure and involves FOOT-TO-FOOT METHOD
of peak systolic blood pressure, favors cardiac several changes: enlargement of the mean hypertrophy and congestive heart failure, and, diameter, reduction of the pulsatile diameter, an between successive arrivals of the foot of a pressure through the decrease of end-diastolic blood age-mediated reduction in endothelial function, pressure, reduces coronary perfusion and causes huge development of the connective tissue; and the reduction of pulse-pressure amplification APPLANATION
TONOMETRY
Aortic PWV is influenced by many con founding resulting from the increased stiffness and altered factors, including age, mean arterial pressure and reflectance properties of the arterial wall.1 sex, which need to be taken into account during Aging is the dominant process that alters of a superficial (radial or carotid) artery, to record the statistical evaluations. PWV is related to YOUNG’S vascular stiffness, wave reflections and pulse MODULUS (E) by the Moens–Korteweg equation, pressure. There is, however, extreme variability in these age-mediated changes.5 This variation is influenced by the histopathologic muscular or musculoelastic peculiarities of arterial tissue, along with the presence of other cardiovascular where ρ is the blood density, E is Young’s risk factors. In middle-aged individuals with modulus, r is the internal radius of the vessel, hypertension, the dominant contribution to and h is the arterial thickness.
changes in arterial stiffness is made by high mean arterial pressure, whereas in older people changes in the vascular wall that are independent to measure aortic PWV by calculating the of the mean arterial pressure are more impor- ratio of distance to time.1 In addition, local tant.1,5 This factor is of particular importance in arterial stiffness might be determined from hypertensive patients with endothelial dysfunc- pulsatile changes in the diameter and blood tion, including those with diabetes mellitus, end- pressure of superficial large arteries, such as the stage renal disease or numerous atherosclerotic carotid artery. The measurement of changes in lesions.1,5 Under such conditions, the increases arterial dia meter requires the use of high- in stiffness involve not only endothelial dysfunc- resolution ultrasound techniques, whereas tion but also increased binding and cross-linking changes in pulsatile pressure are measured of molecules within the connective tissue, such as collagen, proteoglycans and fibronectin.
452 NATURE CLINICAL PRACTICE CARDIOVASCULAR MEDICINE BLACHER AND SAFAR SEPTEMBER 2005 VOL 2 NO 9 2005 Nature Publishing Group
Environmental (sodium) and genetic factors was 5.4 (95% CI 2.4–11.9) for all-cause mortality are now considered to have a large influence and 5.9 (95% CI 2.3–15.5) for cardiovascular on arterial mechanics. Gene polymorphisms mortality. For each increase in PWV of 1 m/s might be related either to the pathophysiology the all-cause mortality-adjusted OR was 1.39 of hypertension, through effects on the renin– (95% CI 1.19–1.62). Therefore, aortic PWV was angiotensin–aldosterone system, or to cardio- shown to be a strong independent predictor of vascular aging.1 Combinations of two or three specific polymorphisms can affect the proper- Brachial and upper-limb or lower-limb PWVs ties of the vessel wall more consistently than do not predict this risk.10 Arterial calcifications, one polymorphism alone. In elderly individuals which are simple markers of cardiovascular with systolic hypertension, the DD genotype risk,11 do not contribute directly to mortality. (homozygous deletion) of the angiotensin- Another index of arterial stiffness, the carotid converting-enzyme gene (ACE) insertion/ artery incremental elastic modulus, is a strong deletion (I/D) polymorphism, in association and independent predictor of cardiovascular with specific genotypes of the aldosynthase mortality.12 Finally, the principal predictor of and α-adducin 1 gene (ADD1) polymorphisms, cardiovascular risk independent of aortic PWV involves a consistent increase of arterial stiffness is the presence of early carotid wave reflections.13 together with an increase of peak systolic blood Similar predictive results have been reported in diabetic patients with end-stage renal disease, renal transplant recipients and individuals with Arterial stiffness and cardiovascular risk
ischemic heart disease,14–16 but such studies are In people older than 60 years, pulse pressure is the most powerful mechanical factor that In a cross-sectional study, 710 patients with predicts myocardial infarction.3 The predictive essential hypertension were investigated,17 and value of this factor applies even in individuals the presence or absence of atherosclerotic alter- with hypertension that is successfully main- ations was detected on the basis of the presence tained at close to normal values by drug therapy.8 of clinical cardiovascular events. The calculation Because pulse pressure is influenced by both of cardiovascular risk using the Framingham cardiac and arterial factors, increased pulse pres- data18 was first done for those patients without sure means greater arterial stiffness only in older atherosclerotic alterations: in this population, populations with preserved ejection fractions. all the calculated cardiovascular risks increased In young subjects, increased systolic blood pres- progressively with PWV. In patients of any age sure and pulse pressure are mainly due to cardiac with atherosclerotic alterations, aortic PWV was factors (i.e. increased ventricular ejection).1,2 the best predictor of cardiovascular mortality. Therefore, aortic PWV should be used as a main Meaume and colleagues19 reported similar predictor of cardiovascular risk only in individ- findings in the elderly after long-term follow- uals who are older than 50 years of age.1–3 The up. In middle-aged hypertensive patients and three populations that have predominantly been in the elderly, another longitudinal study gave studied are those with end-stage renal disease, essential hypertension and diabetes mellitus.
Cardiovascular risk is easy to assess in stiffness or both, several groups have shown that patients with end-stage renal disease because arterial stiffness is significantly elevated in indi- of the related severe complications, high cardio- viduals with type 1 or type 2 diabetes, compared vascular mortality and access to operational with nondiabetic individuals matched for age facilities in relation to their regular dialysis. In and mean arterial pressure. A similar finding a cohort of 241 patients studied between 1987 has been reported in patients with metabolic and 1998, 73 deaths occurred, including 48 due syndrome.21–23 Of the usual functional and to cardio vascular causes.9 From Cox regression structural changes of the media and the endo- analyses, two factors emerged as predictors of thelium, the most important was the formation all-cause and cardiovascular mortality—age of advanced glycation end-products, which are and aortic PWV. After adjustment for all responsible for the increased collagen cross- potentially confounding variables, for values linkage that results in increased stiffness.22 Data of PWV higher than 12.0 m/s compared with from studies in hypertensive patients have shown PWV lower than 9.4 m/s the odds ratio (OR) that an advanced glycation end-product breaker, SEPTEMBER 2005 VOL 2 NO 9 BLACHER AND SAFAR NATURE CLINICAL PRACTICE CARDIOVASCULAR MEDICINE 453 2005 Nature Publishing Group
ALT-711 (alagebrium chloride, Alteon Inc., changes, which were observed in patients who Parsippany, NJ) is able to decrease the amount were treated with perindopril and indapamide of collagen cross links,21 thereby improving but not with atenolol, point to parallel features in carotid distensibility and reducing pulse pressure the reduction of peak systolic blood pressure and without changing the mean arterial pressure. In pulse pressure, and in the regression of structural patients with diabetes, both pulse pressure and arteriolar changes.28 This finding is consistent aortic PWV have been shown to be markers of with previous results reported with ACE inhibi- tion, including changes in reflection coefficients that are usually located at arteriolar branching NEW STRATEGIES FOR CARDIOVASCULAR
points29 and delays in the return of wave reflec- tions.28,29 Therefore, complex interactions are A long-term (>1 year) therapeutic trial that required between small and large arteries to investigated the role of arterial stiffness, peak achieve reductions of peak systolic blood pres- systolic blood pressure and the control of pulse sure and pulse pressure through the use of ACE pressure was performed in patients with end- inhibitors. Rizzoni and colleagues30 reported that, stage renal disease.26 The objective was to reduce in hypertensive patients, structural alterations of their cardiovascular risk by use of an initial the small arteries predicted cardiovascular risk treatment that included salt and water deple- in association with increased pulse pressure, the tion. After randomization, the patients were traditional surrogate of aortic stiffness.
first assigned either ACE inhibition or calcium blockade, followed by administration of the CONCLUSION
two agents combined, either with or without a In conclusion, whereas pulse pressure might β-blocker. During the follow-up of 51 months, provide effective semiotic information in the mean arterial pressure, pulse pressure and clinical practice, the measurement of arterial aortic PWV were all reduced in the survivors. stiffness is clinically relevant not only for risk In the individuals who died from cardiovascular assessment but also to quantify reductions in events, the mean arterial pressure was lowered cardio vascular risk. New therapeutic strate- to the same extent as that observed in the survi- gies that selectively reduce pulse pressure and vors, but the pulse pressure and PWV remained arterial stiffness should be developed. On the unmodified by drug therapy. The adjusted basis of the results of short-term and medium- relative risk of all-cause death and cardio- term controlled trials,31 such a regimen should vascular death in response to blood-pressure involve a combination of ACE inhibitors and changes were 2.59 (95% CI 1.51–4.43) and 2.35 diuretics, with the use of new vasopeptidase (95% CI 1.23–4.51), respectively (P <0.01). In inhibitors or spironolactone, and might include this trial, prolonged survival seemed to be more newly identified or as yet unidentified agents as closely related to the use of ACE inhibitors than to β-blockers or dihydropyridine derivatives.
In the Preterax® in Regression of Arterial References
Stiffness in Controlled Double-Blind (REASON) et al. (2003) Current perspectives on arterial study27 of essential hypertension in middle- stiffness and pulse pressure in hypertension and aged patients, combined treatment with the cardiovascular diseases. Circulation 107: 2864–2869
et al. (1997) Hemodynamic patterns ACE inhibitor perindopril and the diuretic of age-related changes in blood pressure. The indapamide for 1 year decreased the brachial Framingham Heart Study. Circulation 96: 308–315
3 Safar ME (2001) Systolic blood pressure, pulse systolic blood pressure and pulse pressure to pressure and arterial stiffness as cardiovascular risk a greater extent than did treatment with the factors. Curr Opin Nephrol Hypertens 10: 257–261
β-blocker atenolol, after achieving similar 4 Black HR (1999) The paradigm has shifted to systolic blood pressure. Hypertension 34: 386–387
reductions in diastolic blood pressure and mean et al. (2000) Stiffness of carotid artery wall arterial pressure. This finding was obvious material and blood pressure in humans. Stroke 31:
when blood pressure was measured centrally et al. (2001) Carotid and femoral artery within the carotid artery. The results included stiffness in relation to three candidate genes in a white a delayed return of the carotid backward wave population. Hypertension 38: 1190–1197
that was observed after 6 months of therapy and 7 Safar ME (2004) Angiotensin-converting enzyme D/I gene polymorphism and age-related changes in pulse a pressure-independent reduction in PWV that pressure in subjects with hypertension. Arterioscler was observed after 1 year of treatment. These Thromb Vasc Biol 24: 782–786
454 NATURE CLINICAL PRACTICE CARDIOVASCULAR MEDICINE BLACHER AND SAFAR SEPTEMBER 2005 VOL 2 NO 9 2005 Nature Publishing Group
21 Creager MA et al. (2003) Diabetes and vascular Competing interests
determinants of cardiovascular events during 20 years disease: pathophysiology, clinical consequences, and of successful antihypertensive treatment. J Hypertens medical therapy: Part I. Circulation 108: 1527–1532
16: 761–769
22 Lakatta EG (2003) Arterial and cardiac aging: major et al. (1999) Impact of aortic stiffness on shareholders in cardiovascular disease enterprises. survival in end-stage renal failure. Circulation 99:
Part III: cellular and molecular clues to heart and arterial aging. Circulation 107: 490–497
10 Pannier B et al. (2005) Stiffness of capacitive and 23 Schram MT et al. (2004) Increased central artery conduit arteries: prognostic significance for end-stage stiffness in impaired glucose metabolism and type 2 renal disease patients. Hypertension 45: 592–596
diabetes: the Hoorn Study. Hypertension 43: 176–181
11 Blacher J et al. (2001) Arterial calcifications, arterial 24 Schram MT et al. (2003) Pulse pressure is associated stiffness, and cardiovascular risk in end-stage renal with age and cardiovascular disease in type 1 diabetes: disease. Hypertension 38: 938–942
the Eurodiab Prospective Complications Study. J 12 Blacher J et al. (1998) Carotid arterial stiffness as a Hypertens 21: 2035–2044
predictor of cardiovascular and all-cause mortality in 25 Cruickshank K et al. (2002) Aortic pulse-wave velocity end-stage renal disease. Hypertension 32: 570–574
and its relationship to mortality in diabetes and glucose 13 London GM et al. (2001) Arterial wave reflections and intolerance: an integrated index of vascular function? survival in end-stage renal failure. Hypertension 38:
Circulation 106: 2085–2090
26 Guerin AP et al. (2001) Impact of aortic stiffness 14 Shoji T et al. (2001) Diabetes mellitus, aortic stiffness, attenuation on survival of patients in end-stage renal and cardiovascular mortality in end-stage renal failure. Circulation 103: 987–992
disease. J Am Soc Nephrol 12: 2117–2124
27 London GM et al. (2004) Mechanism(s) of selective 15 Barenbrock M et al. (2002) Reduced arterial systolic blood pressure reduction after a low-dose distensibility is a predictor of cardiovascular disease combination of perindopril/indapamide in hypertensive in patients after renal transplantation. J Hypertens 20:
subjects: comparison with atenolol. J Am Coll Cardiol 43: 92–99
16 Weber T et al. (2004) Arterial stiffness, wave 28 Intengan HD et al. (1999) Resistance, artery reflections, and the risk of coronary artery disease. mechanics, structure, and extracellular components in Circulation 109: 184–189
spontaneously hypertensive rats: effects of angiotensin 17 Blacher J et al. (1999) Aortic pulse wave velocity as a receptor antagonism and converting enzyme inhibition. marker of cardiovascular risk in hypertensive patients. Circulation 100: 2267–2275
Hypertension 33: 1111–1117
29 Ting CT et al. (1995) Short- and long-term effects 18 Anderson KM et al. (1991) Cardiovascular disease risk of antihypertensive drugs on arterial reflections, profiles. Am Heart J 121: 293–298
compliance and impedance. Hypertension 26: 524–530
19 Meaume S et al. (2001) Aortic pulse wave velocity 30 Rizzoni D et al. (2003) Prognostic significance of predicts cardiovascular mortality in subjects >70 years small-artery structure in hypertension. Circulation 108:
of age. Arterioscler Thromb Vasc Biol 21: 2046–2050
20 Laurent S et al. (2001) Aortic stiffness is an 31 Van Bortel LM et al. (2001) Pulse pressure, arterial independent predictor of all-cause mortality in stiffness, and drug treatment of hypertension. hypertensive patients. Hypertension 37: 1236–1241
Hypertension 38: 914–921
SEPTEMBER 2005 VOL 2 NO 9 BLACHER AND SAFAR NATURE CLINICAL PRACTICE CARDIOVASCULAR MEDICINE 455 2005 Nature Publishing Group

Source: https://fusion.dns-systems.net/secure/mtswbout/client_root/15077172/files/large-artery_stiffness__hypertension_and_cardiovascular_risk_in_older_patients.pdf