Hypertension and Heart Failure: a Dysfunction of Systole, Diastole Or Both?

Home , Systole

Journal of Human Hypertension (2009) 23, 295–306 & 2009 Macmillan Publishers Limited All rights reserved 0950-9240/09 $32.00 www.nature.com/jhh REVIEW Hypertension and heart failure: a dysfunction of systole, diastole or both?

GW Yip1, JWH Fung1, Y-T Tan2 and JE Sanderson2 1Division of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, People’s Republic of China and 2Department of Cardiovascular Medicine, The Medical School, University of Birmingham, Edgbaston, Birmingham, UK

The pathological myocardial hypertrophy associated and volume changes. Furthermore, dichotomizing heart with hypertension contains the seed for further mala- failure into systolic and diastolic clinical entities has led daptive development. Increased myocardial oxygen to a paucity of clinical trials of therapies for heart failure consumption, impaired epicardial coronary perfusion, with a normal ejection fraction. Therapies aimed at ventricular fibrosis and remodelling, abnormalities in reversing myocardial fibrosis, and targets outside the long-axis function and torsion, cause, to a varying heart such as enhancing vasodilator reserve and degree, a mixture of systolic and diastolic abnormalities. improving chronotropic incompetence deserve further In addition, chronotropic incompetence and peripheral study and may improve the exercise capacity of factors such as lack of vasodilator reserve and reduced hypertensive heart failure patients. Hypertension heart arterial compliance further affect cardiac output parti- disease with heart failure is simply not a dysfunction of cularly on exercise. Many of these factors are common systole and diastole. Other peripheral factors including to hypertensive heart failure with a normal ejection heart rate and vasodilator response with exercise may fraction as well as systolic heart failure. There is deserve equal attention in an attempt to develop more increasing evidence that these apparently separate effective treatments for this disorder. phenotypes are part of a spectrum of heart failure Journal of Human Hypertension (2009) 23, 295–306; differing only in the degree of ventricular remodelling doi:10.1038/jhh.2008.141; published online 27 November 2008

Keywords: heart failure; remodelling; pressure overload; volume overload

Introduction heart failure (SHF) and contributes significant morbidity and mortality. It represents a chain of High blood pressure, including pre-hypertension, is pathological events starting with ageing and athero- a major global health economic issue, accounting for sclerosis, left ventricular hypertrophy (LVH), cor- about half of the total cases of stroke and ischaemic onary artery disease with or without myocardial heart disease. It is responsible for about 7.6 million infarction, to finally heart failure and end-stage deaths each year and costs over 92 million dis- heart disease. This cardiovascular disease conti- ability-adjusted life years, both of which represent nuum, by which common and overlapping patho- 13.5 and 6.0% of the respective global total in the physiological mechanisms are involved in multiple year 2001. Low-income and middle-income coun- steps in disease development across the entire tries, predominantly in eastern Europe, east and spectrum of cardiovascular disease as well as south Asia (China and India) and the Pacific regions, diseases of other target organs, including the brain shoulder about 80% of the cardiovascular disease 1 and kidneys, has received experimental and clinical burden, half of which is in people of working age. validation, with an increasing recognition of the The failing heart in hypertensive heart disease important roles of oxidative stress and nitric oxide (HHD) may present as predominant heart failure synthase, metalloproteinases and their inhibitors, with normal ejection fraction (HFNEF) or systolic renin–angiotensin–aldosterone system and sympa- thetic nervous system, in the development of heart failure.2–5 Correspondence: Professor JE Sanderson, Department of Cardio- A number of primary prevention trials demon- vascular Medicine, The Medical School, University of Birming- strated that treating hypertension can reduce the ham, Edgbaston, Birmingham B15 2TT, UK. development of heart failure up to 50%. In patients E-mail: [email protected] 6 Received 24 June 2008; revised 12 September 2008; accepted with left ventricular systolic dysfunction or LVH, 14 September 2008; published online 27 November 2008 control of blood pressure prevents or retards Hypertensive heart failure GW Yip et al 296 ventricular remodelling and decreases the incidence perfusion.31,32 The increased myocardial oxygen of overt heart failure.7 In patients with established demand from a combination of raised systolic heart failure, further decreases of blood pressure blood pressure, LVH and prolonged ejection time with therapy may lower the mortality rate, slow the in addition to compromised coronary perfusion progression of disease, reduce the number of predisposes hypertensive patients to coronary hospitalizations and exacerbations, and enhance ischaemia, which may be further jeopardized in quality of life and functional capacity.8 those with a pre-existing coronary artery disease.33 About half of the patients with symptoms of heart LVH is a strong predictor of cardiovascular morbid- failure have a normal ejection fraction. HFNEF is ity and mortality, regardless of blood pressure common among female elderly patients with a values or the presence of other cardiovascular risk history of hypertension or atrial fibrillation and factors.34–37 has an adverse prognosis similar to that of SHF.9–12 There is currently no evidence of the superiority of specific antihypertensive drugs in HHD presenting Ventricular fibrosis with HFNEF.13,14 Though a simple diuretic may be as efficacious as a combination with angiotensino- Biomechanical stretch and neurohumoral factors gen-converting enzyme inhibitor or angiotensin induce changes in intracellular signalling pathways, receptor antagonist in relieving symptoms and resulting in an increased protein synthesis and improving submaximal exercise capacity, there is activation of specific genes promoting cardiomyo- still a lack of convincing evidence of modern cyte growth, eventually leading to left ventricular pharmacological heart failure therapy that improves remodelling and cardiac dysfunction.38–40 Ventricu- prognosis in patients with HFNEF, and further lar remodelling describes structural changes in the clinical trial results are pending.15,16 The inefficacy LV in response to chronic alterations in volume of modern heart failure therapy on survival in overload or pressure loads. The remodelling process patients with HFNEF was also emphasized by large results from alterations in cardiac myocytes as well epidemiological surveys, which observed an im- as the extracellular matrix, including fibroblasts and proved prognosis over the last two decades in several different types of macromolecules, for patients with SHF, but not in patients with HFNEF.9 example, type I and III collagen, elastin and Furthermore, there is still no general consensus that fibronectin. Unlike in athletic training, there is a HFNEF and SHF are two distinct heart failure disproportionate growth of the extracellular matrix phenotypes,17 or just a precursor stage of SHF,18–20 and increased modification and enhancement of and that the role of systolic function in HFNEF is perivascular and interstitial (perimysial) collagen still a matter of debate.17–19,21–26 We will review the mass, in addition to the development of endothelial pathophysiology underlying the progression of HHD dysfunction, increasing vulnerability of the ventri- to congestive heart failure, in particular the func- cle to myocardial ischaemia, dysfunction, heart tional and morphological changes in both types of failure, arrhythmia and sudden death.41 Abnormal- heart failure. ities in the extent and distribution of collagen within the remodelling heart have been recognized as playing an important role in the pathogenesis of Cardiac hypertrophy heart failure. The collagen homoeostasis is determined by the rate and extent of synthesis and The current view of HHD is that it first leads to degradation. increased cardiac hypertrophy, a maladaptive pro- Carboxy-terminal propeptide procollagen type I cess required to normalize wall stress and to (PICP), a breakdown product in a 1:1 stoichio- maintain cardiac output.27 It occurs in response to metric ratio to fibrillar collagen type I that accounts an increase in systolic blood pressure and, therefore, for 85–90% of myocardial collagen, can be measured ventricular loading throughout the ejection period from serum and reflect fibrogenesis.42 Type III due to a combination of increased proximal aortic collagen accounts for the majority of the remaining stiffening and augmentation of late systolic pressure myocardial collagen. Amino-terminal propeptide from early return of pulse wave reflection that is procollagen type III (PIIINP), an extension peptide commonly associated with ageing.28,29 A high base- that cleaved off from the procollagen type III in line left ventricular mass value in initially normo- the conversion to collagen type III, can also be tensive patients predicts subsequent increases in measured in the serum. However, in contrast blood pressure and the development of hyperten- to PICP, amino-terminal propeptide procollagen sion, independent of other risk factors.30 The LVH type III is not completely removed from its results in prolonged ejection period at the expense procollagen precursor during the extracellular of reduction of diastole at any given heart rate and processing and as a result, serum PICP may reflect causes delayed myocardial relaxation. The de- fibrogenesis better than serum amino-terminal pro- creased diastolic blood pressure from aortic stiffen- peptide procollagen type III. Collagen degradation is ing and a reduced LV filling time with or without under the control of matrix metalloproteinases rise in LV filling pressure may compromise coronary (MMPs), a family of zinc-dependent interstitial

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 297 enzymes, and their tissue inhibitors (TIMPs). serve to monitor interventions designed to alter the Commonly measured serum markers of myocardial remodelling process.48 collagen turnover include gelatinases (MMP-2 and -9), collagenase (MMP-13), tissue inhibitors of MMPs (TIMP-1 and -2) and carboxy-terminal telo- Patterns of hypertrophic remodelling peptide of collagen type I, a breakdown product of collagen type I.43 In addition to the above, the pattern of hypertrophic Several reports have provided important new remodelling is relevant. Three major patterns of LV insights into patients with HHD who present with remodelling are recognized: concentric remodelling, SHF or heart failure with normal ejection fraction eccentric hypertrophy and that due to a myocardial (HFNEF). Querejeta et al.44 demonstrated a positive infarction, which is a combination of the other two correlation between coronary sinus and peripheral patterns, when the distended and dilated infarct serum levels of PICP and these correlated well with tissue increases the LV volume with a combined myocardial accumulation of fibrillar collagen type I volume and pressure overload on the non-infarct assessed histologically. Moreover, there was a areas.5 Under pressure overload, as in hypertension, progressive significant increase in PICP levels from remodelling is essentially concentric by adding normal control individuals, HHD patients without sarcomeres in parallel to achieve an increased heart failure, to hypertensive heart failure patients cardiomyocyte thickness, and is characterized by with variable ejection fractions. In addition to nondilated, thick-walled LV.49 Thus, the LV wall elevated PICP and amino-terminal propeptide pro- thickness increases more than the LV cavity volume; collagen type III, patients with HFNEF may also as a result, the relative wall thickness (normally show serologic evidence of an active fibrotic process p0.42) and the ratio of the LV mass to the end- in terms of collagen turnover or MMP activities diastolic volume increase. Earlier prospective stu- (carboxy-terminal telopeptide of collagen type I, dies have demonstrated that hypertension-related MMP-2 and -9), which corresponds with the severity concentric remodelling is associated with adverse of diastolic dysfunction.45 Ahmed et al.46 have cardiovascular outcomes, including myocardial in- demonstrated normal MMP/TIMP profiles in hyper- farction, heart failure and increased mortality.35,50–53 tensive patients with normal LV structure and Despite normal LV ejection fraction and unchanged ejection fraction, and elevated serum levels of endocardial stress-shortening relationship in con- TIMP-1 and MMP-9 in those with LVH and HFNEF. centric LVH, a subtle systolic dysfunction can be A TIMP-1 level 41200 ng mlÀ1 predicted the pre- demonstrated by depressed LV midwall fractional sence of LVH with HFNEF (specificity 88%; positive shortening on traditional two-dimensional echocar- predictive value of 94% with area under the receiver diography 51–54 and tagged cardiac magnetic reso- operator curve of 0.71). Lopez et al.47 found that nance55 and, more recently, abnormalities of LV long mysial collagen that is related to muscle compart- and short axes by tissue Doppler and strain ments was reduced in SHF patients but that echocardiography.56–58 perivascular and scar-related collagen volumes were In contrast, eccentric remodelling is associated higher in both SHF and HFNEF hypertensive with volume overload and involves the addition of patients than in normotensive individuals and in sarcomeres in series to achieve myocyte lengthen- SHF compared with HFNEF patients. Thus, in HHD, ing. There is a proportional increase in LV mass an increase in interstitial collagen is associated with and volume, so that the ventricle becomes HFNEF, whereas a degradation of endo- and peri- dilated, with normal relative wall thickness but mysial components of the collagen scaffolding is reduced LV systolic function. Eccentric hypertrophy accompanied by muscle fibre slippage, thinning of has been found to be associated with worse myocardium, spherical dilation of ventricular cham- systolic function in a subgroup of patients with ber and the development of SHF with a reduced LV hypertension.59 ejection fraction. These morphological findings in The difference in the hypertrophic remodelling the SHF hypertensive group were associated with an between HFNEF and SHF hypertensive patients increased protein expression of MMP-1 relative to correlates with distinct patterns of peptide growth TIMP-1 in tissue and blood when compared with factor induction in both conditions.60 In addition, control individuals and HFNEF patients, which also HFNEF is characterized by cardiomyocyte hyper- correlated directly with LV end-diastolic diameter trophy and preserved myofilamentary density, and inversely with LV ejection fraction.47 Thus, whereas SHF by minimal cardiomyocyte hypertro- patterns of collagen deposition differ between SHF phy and loss of myofilaments.60 In addition, the and HFNEF. An imbalance in the MMP-1/TIMP-1 resting tension of isolated cardiomyocytes is higher ratio may herald LV dilation and reduced EF in SHF. in HFNEF.61,62 This may be related to an increased In future, serum measurements of MMP and TIMP interstitial collagen with lack of change in the MMP- levels may be used to detect abnormalities in the 1/TIMP-1 ratio and a higher ratio of shorter (stiffer) extracellular matrix of the heart that could account N2B isoform to longer (more compliant) N2BA titin for ventricular remodelling and dysfunction in isoform, a giant sarcomeric cytoskeletal protein that hypertensive patients, and these markers might also extends from the Z-lines to the centre of the thick

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 298 filament and functions as a molecular spring, individuals without a known cardiovascular dis- contributing to the high resting passive stiffness of ease, Rosen et al.69 demonstrated that LV systolic the myocardium63,64 and perhaps to the higher function, evaluated by midwall circumferential restoring forces and preservation of untwisting of strain or LV ejection fraction by magnetic resonance the LV (see below).65 imaging, decreased with increasing concentric remodelling (higher LV mass/volume ratio) in men. In women, circumferential strain was also declined in Does HHD evolve as concentric LVH with those with the highest quintile of LV mass/volume HFNEF and transition to SHF with LV ratio, but there was no such association with LV dilation and reduced EF? ejection fraction. In both genders, the inverse association between regional strain and the LV Although it has been recognized clinically that mass/volume ratio was more pronounced in the left occasional patients who present with typical SHF anterior descending artery territory, though this have only hypertension as a possible aetiology and observation appeared inconsistent in women in this is especially common where coronary artery multivariable analysis, and there was no significant disease is uncommon and treatment of hypertension association between septal or posterior LV thickness is poor such as in Africa and parts of Asia, there is and regional LV strain. Nevertheless, the study little information on whether patients with hyper- suggests that there may be a subtle regional systolic tension and LVH can progress to heart failure with dysfunction in response to concentric remodel- LV dilatation and it has been disputed whether this ling.69 It might be possible that those in the highest does occur in the absence of ischaemic heart disease quintile of the LV mass/volume ratio with reduced (see Figure 1). Recently, additional clinical insights circumferential strain have been imaged during the into the transition from concentric LVH to failure transition from compensatory remodelling to overt have been provided by two recently completed SHF with dilated LV. Contrary to this notion, longitudinal studies using serial echocardiography however, the LV volumes in this study were and one cross-sectional multiethnic study using decreased in the highest quintile of LV mass/volume 66 cardiac magnetic resonance imaging with tagging. ratio, whereas LV volume is increased with LV In a retrospective cohort of 159 predominantly failure. This finding was corroborated by Olmsted middle-aged hypertensive African Americans with County database that observed an opposite evolu- concentric LVH and a normal LV ejection fraction, tion with progressive reduction in LV end-diastolic only 18% developed a reduced LV ejection fraction volumes in HHD as HFNEF develops. In addition, after a follow-up of about 4 years. Moreover, nearly variable LV geometric patterns were observed and, half of those who did develop a low LVEF had an despite traditional teaching, concentric LVH or 67 interval myocardial infarction. In the Cardiovas- remodelling is not invariably present in HFNEF.70 cular Health Study, it has been shown that increased Moreover, in hypertrophic cardiomyopathy, a dis- baseline LV mass was an independent risk factor for ease also characterized by diastolic LV dysfunction, the development of depressed ejection fraction 5 the evolution towards a dilated LV is only 3.5%.71 years later in elderly Caucasian individuals, inde- Nevertheless, the time course of remodelling does pendent of coronary artery disease or myocardial depend to a certain extent on the aetiology. infarction. However, the majority of cases of LVH Sufficient myocardial necrosis from infarction or were of eccentric remodelling. Furthermore, con- myocarditis is a potent stimulus for the initiation of centric LVH was not an independent risk factor in the post-infarct remodelling process, in which multivariate analysis for the development of a low stretched and dilated infarcted tissue increases LV ejection fraction, possibly due to the small number volume with a combined volume and pressure load 68 of patients (n ¼ 26) with concentric LVH. In 441 on non-infarcted areas, leading to LV dilation and a reduced ejection fraction, irrespective of whether there is pre-existing concentric remodelling. The time course and pattern of development of heart failure are therefore quicker after myocardial infarction, which is the commonest cause of SHF. In HHD, remodelling in the absence of myocardial necrosis is a slower process.19 Initially, LV hypertrophy by itself leads to reduced systolic and diastolic function particularly in the long axis. Compensatory circumferential contraction and rotation normalize the ejection fraction (see below). However, at later stages, a further remodelling will occur, with resultant LV enlargement and development from HFNEF to more obvious SHF.19 Serial longitudinal Figure 1 Hypertension and heart failure (adapted from imaging studies are therefore much needed in future Drazner66). to clarify the progression of HHD to heart failure

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 299 and whether concentric LVH in the absence of Although a global LV systolic performance is myocardial infarction is a common pathway leading preserved,17 HFNEF has reduced LV long-axis short- to SHF.5,66 ening21 and depressed tissue Doppler myocardial systolic velocity.21–26 The success of cardiac resynchronization therapy has underscored the impor- Insights from traditional to translational tance of dys-synchrony in the pathophysiology of echocardiography SHF, independent of the QRS duration.81 Interest- ingly, about 33–39% of HFNEF patients showed There is still debate whether HFNEF and SHF are evidence of systolic dys-synchrony and 56–58% of indeed distinct HF phenotypes.18,72 Community- them diastolic dys-synchrony at rest.82–83 Moreover, based epidemiology data suggest that compared 25% of these patients demonstrated isolated systolic with hypertensive or healthy control individuals, dys-synchrony, and the correlation between systolic patients with HFNEF have a more impaired renal and diastolic dys-synchrony was poor.82 This was in function, despite smaller end-diastolic volumes, contrast to the close curvilinear relationship be- lower cardiac output, higher end-diastolic pres- tween the absolute annulus systolic and early sures, and more vascular and ventricular systolic diastolic velocities seen in a range of cardiac stiffening with increased diastolic stiffness.70 diseases and normal individuals.84 Systolic dys- A careful evaluation of echocardiographic data synchrony in HFNEF is associated with reduced reveals that patients with HFNEF have accentuated peak basal myocardial velocities, lower stroke work LVH, more severe diastolic dysfunction and left and even lower ‘normal’ ejection fraction than in atrial dilation or failure.70,73 Furthermore, HFNEF is those without it.83 This suggests that systolic dys- frequently referred to as diastolic heart failure as synchrony contributes towards a wider spectrum of slow LV relaxation and high LV stiffness were systolic dysfunction in HFNEF, casting doubt about further confirmed from invasive haemodynamic the orthodox view that HFNEF is truly a clinical studies.74–77 entity of diastolic dysfunction. On the basis of the above data acquired at rest, More recently, with the advent of strain echocar- three different pathophysiologic theories have been diography based on tissue Doppler85 or speckled- postulated for HFNEF. First, an increase in intrinsic tracking technology,86,87 myocardial deformation in muscle stiffness and impaired relaxation in diastole the longitudinal, radial and circumferential direc- result in higher left atrial pressures to fill ade- tions as well as torsion or twist can now be quately, predisposing these patients to pulmonary quantified, which was feasible earlier only in venous congestion and dyspnoea especially on magnetic resonance imaging using tagging88 or exertion.74,75,77 Second, increased systolic ventricu- tissue phase-contrast velocity mapping.89 Strain lar and arterial stiffening, that is, deranged ventri- measures the extent of myocardial deformation and culoarterial coupling, may contribute to the is free from effects of tethering and translation of pathophysiology of HFNEF by exaggerating hyper- adjacent myocardial segments that are associated tensive response with increased systolic load and with tissue Doppler velocities.90 Speckle-tracking inducing load-dependent diastolic dysfunction imaging based on two-dimensional automated especially during exercise or other stresses.78 Third, frame-to-frame tracking (between 40 and 100 frames an enhanced sensitivity to volume overload from per s) of unique patterns of natural acoustic markers increased LV remodelling and dilation with volume- (speckles) on standard gray-scale B-mode images is dependent elevation of filling pressures was ob- independent of myocardial translation and angle served by Maurer et al.79 in a subgroup of hyperten- dependency inherent to tissue Doppler-derived sive HFNEF patients from the Cardiovascular Health strain parameters.91 Thus, strain echocardiography Study who had renal impairment and larger LV may have a clinical advantage over tissue Doppler volumes but normal systolic ventricular and vascu- velocities of better assessment of segmental and lar stiffness. Thus, impaired renal function and global myocardial contractility.92 renal arterial atherosclerosis in the elderly may also Interest in ventricular torsion has increased dur- be involved in causing rapid rises in blood pressure ing the last two decades. Studies have suggested that and excessive fluid retention due to chronic activa- the twist during the cardiac cycle is important for tion of the renin–angiotensin–aldosterone system. normal function, possibly contributing to the sub- These elevations, together with underlying diastolic sequent untwisting or suction that aids early dysfunction, are the causes of recurrent acute diastolic filling, and the delay of which is associated pulmonary oedema in these patients. Renal revas- with diastolic dysfunction and raised LV filling cularization has been shown to decrease incidence pressures in aortic stenosis.93 Torsion helps normal- of pulmonary oedema due to renal artery stenosis80 ize relatively uniform stresses, sarcomere length and is likely the result of reduction of fluctuating changes, and myocardial workload across the LV hypertension and subsequent exacerbation of dia- wall, and contributes to energy-efficient ejec- stolic dysfunction. These different mechanisms tion.94,95 Speckle-tracking echocardiography offers proposed for HFNEF may have important implica- torsional or ‘twisting’ motion assessment, derived tions for differing therapeutic strategies. from circumferential strain at different short-axis

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 300 levels of the heart, which is promoted by the virtue mechanics may provide pathophysiological insights of a single myofibre band forming a double helix in into the mechanism of LV dysfunction in HHD. The the LV, changing gradually from a right-handed helix transmural gradient between subendocardial (long- in the subendocardium to a left-handed helix in the itudinal function) and epicardial (twisting function) subepicardium.96 Accordingly, LV torsion is defined fibres may provide an early detection of subclinical as apical rotation relative to the base. The larger myocardial disease.106 radius of subepicardial fibres generates higher torque that dominates the direction of the twist, causing rotation of the base and apex in clockwise Non-diastolic and peripheral factors: the missing link and counterclockwise directions, respectively, when between hypertension and HFNEF viewed from the apex. The LV untwisting predomi- There is good evidence that arterial compliance is nantly occurs during isovolumetric relaxation and also associated with diastolic dysfunction. Mottram begins just slightly before aortic valve closure, et al.107 demonstrated that in hypertensive patients followed in time by the peak intraventricular with exertional breathlessness, progressively abnor- gradient, and then by long-axis lengthening and mal diastolic function is associated with reduced short-axis expansion that may precede or coincide arterial compliance and arterial stiffness was an with peak early diastolic filling.97,98 However, like independent predictor of diastolic dysfunction. many diastolic parameters, there are age-related Arterial stiffness is also linked to reduced coronary degenerative changes in the LV torsion, with flow velocity reserve, which may be relevant to the increased peak systolic twisting but reduction and development of subendocardial ischaemia and re- delay in untwisting velocities. No gender differ- duced longitudinal ventricular function.108 How- ences were noted in peak twist or untwisting ever, most studies in this area are undertaken under velocities.99,100 Another source of the restoring force resting conditions that demonstrate normal or near- that drives diastolic torsional recoil is systolic normal LV systolic performance but abnormal deformation within the sheet structure, the so-called diastolic properties in hypertension or HFNEF. shear strain, due to significant fibre shearing within A reduced aortic distensibility has been reported the sheet planes during systole.101 Measuring long- in HFNEF and it correlated with exercise limita- itudinal or circumferential shear is another promis- tion.109 Haemodynamic data obtained during ing approach to the assessment of torsion with exercise and other stresses may be key in differ- speckle tracking, allowing full assessment of overall entiating HFNEF and hypertensive patients, though deformation in addition to three-dimensional very few studies on the response to aerobic exercise strains and investigation of regional differences in have been reported in HFNEF and most have torsion.102 exclusively focused on LV and vascular function. Hypertension, like other progressive myocardial An individual’s exercise capacity depends not only diseases for example, coronary artery disease and on the cardiovascular function but also on the hypertrophic cardiomyopathy, preferentially affects coordinated efforts and integrity of other major subendocardial function and the longitudinal fibres systems including pulmonary, neural, endocrine, running along it.103 The epicardial function may haematological and skeletal muscle systems. In SHF, remain relatively unaffected. Thus, in HFNEF, LV each system has major abnormalities, for example, longitudinal and radial strains are reduced but the skeletal muscles, which exert significant impact circumferential strain and torsion, which reflect on exercise intolerance independent of LV func- subepicardial function, are normal at least at rest. tion.110 If all components of the exercise response This transmural heterogeneity in myocardial con- are taken into account, the LV abnormalities found tractile function with normal LV torsion and in SHF, including diastolic dysfunction, increased circumferential strain may be a compensatory passive stiffness and impaired systolic function, mechanism for preserving the normal LV ejection have only a modest role in determining exercise fraction in these patients.104 However, the untwist- capacity, less than thought earlier.111 Several inves- ing, which is determined by preceding peak systolic tigators have reported that a significant number of twist and LV end-systolic volume, is delayed and HFNEF patients have attenuated heart rate response reduced in parallel to the degree of LV hypertrophy to exercise, similar to SHF patients.77,112–114 Further- in hypertension,65 whereas untwisting is preserved more, those with such chronotropic incompetence but its onset is delayed in HFNEF.105 These variable have an even more severe exercise intolerance than results should be clarified in further studies. those without it. The slower heart rate rise, lower As the disease progresses, midwall and subepi- peak heart rate and impaired recovery suggest cardial myocardium are affected by pathological autonomic (parasympathetic) dysfunction. Contrac- changes, including reduction of mysial collagen, tile reserve is also impaired at peak exercise. It is eccentric remodelling and LV dilation. The LV worth noting that systemic vasodilation with ex- circumferential strain, twisting and untwisting ercise has been reported to be impaired in some, but rates, in addition to longitudinal and radial strains, not all, studies of HFNEF.113,115 Borlaug et al.113 may be reduced in the later stages and in SHF. Thus, noted an impaired systemic vasodilation in 17 the relative differences in longitudinal and twisting HFNEF patients during exercise, which accounted

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 301 for the depressed cardiac output increase in addi- dysfunction and ventricular–arterial stiffening as tion to lower peak heart rate. There were no mentioned above. differences in exercise-induced cardiac preload rise, estimated lung congestion on radionuclide ventri- culography or plasma catecholamines. In contrast, Clinical implications Hundley et al.115 reported relatively preserved endothelial function by cardiac magnetic resonance From the maladaptive development of cardiac assessment of flow-mediated dilation in the super- hypertrophy to detrimental consequences of in- ficial femoral artery followed within 48 h by symp- creased myocardial oxygen consumption, impaired tom-limited exercise in nine HFNEF elderly patients epicardial coronary perfusion, ventricular fibrosis in the absence of advanced arteriosclerosis, diabetes and remodelling, abnormalities in long-axis func- or hypercholesterolaemia. tion and torsion, there is clearly a mixture of varying These data suggest that other non-diastolic per- degrees of systolic and diastolic abnormalities ipheral factors, such as heart rate and vasodilator representing the same disease processes that evolve reserve, may also play a role in exercise intolerance gradually as a continuum in HHD. Patients with in patients with HFNEF and challenge the notion HFNEF have similar, though not as severe, patho- that diastolic dysfunction alone primarily explains physiologic characteristics compared with patients exercise-related symptoms in this disorder by limit- with typical SHF, including severely reduced ex- ing left ventricular end-diastolic volume and thus ercise capacity (peak oxygen consumption and cardiac reserve.112 It appears that the development ventilatory anaerobic threshold), neuroendocrine of HFNEF depends on loss of compensatory sys- activation (norepinephrine and B-type natriuretic temic vasodilator and chronotropic reserve, in peptide levels) and impaired quality of life.116 addition to other abnormalities, such as diastolic In addition, chronotropic incompetence and lack

Table 1 Morphological, microscopic, resting and dynamic functional changes in HFNEF and SHF

Parameters HFNEF SHF

Microscopic and neuroendocrine features Cardiac cell hypertrophy Increased Minimal Resting tension of cardiomyocytes Increased Decreased Myofilament density Preserved Decreased Titin N2B/N2BA ratio Increased Decreased Interstitial collagen Increased Decreased MMP-1/TIMP-1 ratio Little changed Decreased Patterns of peptide growth factor induction Different Different b-Receptor downregulation and/or myocardial b-adrenergic Present Present receptor desensitization (post-synaptic) Norepinephrine Increased Increased B-type natriuretic peptide Increased More increased

Resting echocardiographic parameters LV cavity size Normal or decreased Increased LV shape and geometry Little changed Spherical LV mass index Increased Increased LV mass to cavity ratio Increased Normal or decreased Relative wall thickness Increased Normal End-diastolic volume/wall stress Normal or decreased Increased End-systolic volume/wall stress Normal Increased LV ejection fraction Normal Decreased Longitudinal velocity/strain Decreased More decreased Radial strain Decreased More decreased Circumferential strain Normal Decreased LV twist (torsion) Normal or decreased Decreased LV twisting rate Normal or decreased Decreased LV untwisting rate Normal or decreased Decreased

Exercise response Peak oxygen consumption Reduced Reduced Ventilatory anaerobic thresholds Reduced Reduced Heart rate response Blunted Blunted Stroke volume/cardiac output Decreased More decreased Contractile reserve Decreased Decreased Systemic vascular resistance Normal or increased Increased

Abbreviations: HFNEF, heart failure with normal ejection fraction; LV, left ventricular; MMP-1, matrix metalloproteinase; SHF, systolic heart failure; TIMP-1, tissue inhibitor of matrix metalloproteinase.

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 302 of vasodilator and contractile reserves in addition to College of Cardiology, Heart and European Heart diastolic LV dysfunction, such as increase in Journal for similar and related articles. Reference intrinsic passive diastolic stiffness and evidence of lists in key articles were searched to identify older raised LV filling pressure, are not unique to patients publications. More than 200 articles were analysed. with HFNEF but also occur in SHF. Thus, classifying heart failure into systolic and diastolic heart failure on the basis of ejection fraction is somewhat References misleading as it implies a primary or dominating 1 Lawes CMM, Vander Hoorn S, Rodgers A, for the 113 mechanism to which therapy should be targeted. International Society of Hypertension. Global burden It appears more useful to classify heart failure of blood-pressure-related disease, 2001. Lancet 2008; according to the aetiology and the mechanisms 371: 1513–1518. involved in the individual patient, which may be 2 Dzau VJ, Antman EM, Black HR, Hayes DL, Manson different.19 Table 1 summarizes the similarities and JE, Plutzky J et al. The cardiovascular disease differences between HFNEF and SHF and illustrates continuum validated: clinical evidence of improved that increased ventricular volume, reflecting remo- patient outcomes: part I: pathophysiology and clinical delling, is the main discrepancy. trial evidence (risk factors through stable coronary artery disease). Circulation 2006; 114: 2850–2870. Multimodal cardiovascular imaging, including 3 Dzau VJ, Antman EM, Black HR, Hayes DL, Manson resting and dynamic echocardiography, vascular JE, Plutzky J et al. The cardiovascular disease assessments for arterial stiffness, carotid intima– continuum validated: clinical evidence of improved media thickness and endothelial dysfunction, may patient outcomes: part II: Clinical trial evidence have a vital role in analysing these processes and (acute coronary syndromes through renal disease) uncovering the subtle morphological and physiolo- and future directions. Circulation 2006; 114: gical characteristics of this heterogeneous condition, 2871–2891. and to guide active intervention for treatment and 4 Dzau V. The cardiovascular continuum and renin– prevention. An emerging concept in the treatment of angiotensin–aldosterone system blockade. JHypertens hypertension recognizes that progressive vascular 2005; 23(Suppl 1): S9–S17. 5 Opie LH, Commerford PJ, Gersh BJ, Pfeffer MA. damage can continue to occur even when blood Controversies in ventricular remodelling. Lancet pressure is controlled. It is conceivable that anti- 2006; 367: 356–367. hypertensive drugs with vascular modifying proper- 6 The SOLVD Investigators. Effect of enalapril on ties of attenuating or reversing age-associated survival in patients with reduced left ventricular vascular wall remodelling and arterial stiffening ejection fractions and congestive heart failure. N Engl will be more preferable to those without it.117,118 J Med 1991; 325: 293–302. Furthermore, dichotomizing heart failure into sys- 7 Mathew J, Sleight P, Lonn E, Johnstone D, Pogue J, tolic and diastolic clinical entities potentially Yi Q et al. Heart Outcomes Prevention Evaluation narrows our therapeutic focus. Given the disap- (HOPE) Investigators. Reduction of cardiovascular pointing results of recent clinical trials, these kind risk by regression of electrocardiographic markers of left ventricular hypertrophy by the angiotensin-con- of approaches that target extra-cardiac factors may verting enzyme inhibitor ramipril. Circulation 2001; be more fruitful such as chronotropic incompetence 104: 1615–1621. 119–120 with exercise training, pharmacologic inter- 8 Packer M, Coats AJ, Fowler MB, Katus HA, Krum H, ventions and/or selective, physiological pacing Mohacsi P et al. Carvedilol Prospective Randomized during exercise (cardiac resynchronization therapy Cumulative Survival Study Group. Effect of carvedilol may be one of the possible pacing modes).111 Further on survival in severe chronic heart failure. N Engl investigation of treatments to enhance vasodilator J Med 2001; 344: 1651–1658. reserve and their effectiveness in this population is 9 Owan TE, Hodge DO, Herges RM, Jacobsen SJ, Roger also warranted.121–125 VL, Redfield MM. Trends in prevalence and outcome In summary, HHD with heart failure is simply not of heart failure with preserved ejection fraction. N Engl J Med 2006; 355: 251–259. a dysfunction of systole or diastole alone but to 10 Bhatia RS, Tu JV, Lee DS, Austin PC, Fang J, Haouzi A some extent involves both. In addition, other et al. Outcome of heart failure with preserved ejection peripheral factors including heart rate, arterial fraction in a population-based study. N Engl J Med compliance and vasodilator response with exercise 2006; 355: 260–269. deserve equal attention in an attempt to develop 11 Tribouilloy C, Rusinaru D, Mahjoub H, Souliere V, more effective treatments and prevention strategies. Levy F, Peltier M et al. Prognosis of heart failure with preserved ejection fraction: a 5 year prospective Search strategy and selection criteria population-based study. Eur Heart J 2008; 29: We searched MEDLINE using ‘hypertension’ as the 339–347. key word in combination with, ‘heart failure’, 12 Fung JW, Sanderson JE, Yip GW, Zhang Q, Yu CM. ‘remodelling’, ‘pressure overload’, ‘volume over- Impact of Atrial Fibrillation in Heart Failure with Normal Ejection Fraction: a Clinical and Echocardio- load’ and ‘echocardiography’. We searched all major graphic Study. J Card Fail 2007; 13: 649–655. cardiovascular journals: New England Journal of 13 Yusuf S, Pfeffer MA, Swedberg K, Granger CB, Held P, Medicine, The Lancet, Journal of the American McMurray JJ et al. CHARM Investigators and Medical Association, Circulation, Journal American Committees. Effects of candesartan in patients with

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 303 chronic heart failure and preserved left-ventricular 30 Devereux RB, Roman MJ. Left ventricular hypertro- ejection fraction: the CHARM-Preserved Trial. Lancet phy in hypertension: stimuli, patterns, and conse- 2003; 362: 777–781. quences. Hypertens Res 1999; 22: 1–9. 14 Cleland JG, Tendera M, Adamus J, Freemantle N, 31 Hoffman J, Buckberg G. The myocardial supply:de- Polonski L, Taylor J. The perindopril in elderly mand ratio—a critical review. Am J Cardiol 1978; 41: people with chronic heart failure (PEP-CHF) study. 327–332. Eur Heart J 2006; 27: 2338–2345. 32 Hoffman JI. A critical review of coronary reserve. 15 Yip GW, Wang M, Wang T, Chan S, Fung JW, Yeung L Circulation 1987; 75: I6–I11. et al. The Hong Kong diastolic heart failure study: a 33 Ferro G, Duilio C, Spinelli L, Liucci GA, Mazza F, randomised controlled trial of diuretics, irbesartan Indolfi C. Relationship between diastolic perfusion and ramipril on quality of life, exercise capacity, left pressure time and coronary artery stenosis during ventricular global and regional function in heart stress-induced myocardial ischemia. Circulation failure with a normal ejection fraction. Heart 2008; 1995; 92: 342–347. 94: 573–580. 34 Levy D, Garrison RJ, Savage DD, Kannel WB, 16 Carson P, Massie BM, McKelvie R, McMurray J, Castelli WP. Prognostic implications of echocardio- Komajda M, Zile M et al. The irbesartan in heart graphically determined left ventricular mass in the failure with preserved systolic function (I-PRE- Framingham Heart Study. N Engl J Med 1990; 322: SERVE) trial: rationale and design. J Card Fail 2005; 1561–1566. 11: 576–585. 35 Koren MJ, Devereux RB, Casale PN, Savage DD, 17 Baicu CF, Zile MR, Aurigemma GP, Gaasch WH. Left Laragh JH. Relation of left ventricular mass and ventricular systolic performance, function, and con- geometry to morbidity and mortality in uncompli- tractility in patients with diastolic heart failure. cated essential hypertension. Ann Intern Med 1991; Circulation 2005; 111: 2306–2312. 114: 345–352. 18 Brutsaert DL. Cardiac dysfunction in heart failure: the 36 Vasan RS, Larson MG, Benjamin EJ, Evans JC, Levy D. cardiologist’s love affair with time. Prog Cardiovasc Left ventricular dilatation and the risk of congestive Dis 2006; 49: 157–181. heart failure in people without myocardial infarction. 19 Sanderson JE. Heart failure with a normal ejection N Engl J Med 1997; 336: 1350–1355. fraction. Heart 2007; 93: 155–158. 37 Haider AW, Larson MG, Benjamin EJ, Levy D. 20 Ouzounian M, Lee DS, Liu PP. Diastolic heart failure: Increased left ventricular mass and hypertrophy are mechanisms and controversies. Nat Clin Pract Cardi- associated with increased risk for sudden death. JAm ovasc Med 2008; 5(7): 375–386. Coll Cardiol 1998; 32: 1454–1459. 21 Yip G, Wang M, Zhang Y, Fung JW, Ho PY, 38 Susic D, Frohlich ED. The aging hypertensive heart: a Sanderson JE. Left ventricular long axis function in brief update. Nat Clin Pract Cardiovasc Med 2008; 5: diastolic heart failure is reduced in both diastole 104–110. and systole: time for a redefinition? Heart 2002; 87: 39 Mudd JO, Kass DA. Tackling heart failure in the 121–125. twenty-first century. Nature 2008; 45: 919–928. 22 Yu CM, Lin H, Yang H, Kong SL, Zhang Q, Lee SW. 40 Hill JA, Olson EN. Cardiac plasticity. N Engl J Med Progression of systolic abnormalities in patients with 2008; 358: 1370–1380. ‘isolated’ diastolic heart failure and diastolic dysfunc- 41 Weber KT. Fibrosis and hypertensive heart disease. tion. Circulation 2002; 105: 1195–1201. Curr Opin Cardiol 2000; 15: 264–272. 23 Petrie MC, Caruana L, Berry C, McMurray JJV. 42 Lopez B, Gonzalez A, Varo N, Laviades C, Querejeta ‘Diastolic heart failure’ or heart failure caused by R, Diez J. Biochemical assessment of myocardial subtle left ventricular systolic dysfunction? Heart fibrosis in hypertensive heart disease. Hypertension 2002; 87: 29–31. 2001; 38: 1222–1226. 24 Nikitin NP, Witte KK, Clark AL, Cleland JGF. Color 43 Spinale FG. Myocardial matrix remodeling and the tissue Doppler-derived long-axis left ventricular matrix metalloproteinases: influence on cardiac form function in heart failure with preserved global and function. Physiol Rev 2007; 87: 1285–1342. systolic function. Am J Cardiol 2002; 90: 1174–1177. 44 Querejeta R, Lopez B, Gonzalez A, Sanchez E, Larman 25 Bruch C, Gradaus R, Gunia S, Breithardt G, Wichter T. M, Martinez Ubago JL et al. Increased collagen type I Doppler tissue analysis of mitral annular velocities: synthesis in patients with heart failure of hyperten- evidence for systolic abnormalities in patients with sive origin: relation to myocardial fibrosis. Circula- diastolic heart failure. J Am Soc Echocardiogr 2003; tion 2004; 110: 1263–1268. 16: 1031–1036. 45 Martos R, Baugh J, Ledwidge M, O’Loughlin C, 26 Vinereanu D, Nicolaides E, Tweddel AC, Fraser AG. Conlon C, Patle A et al. Diastolic heart failure: ‘Pure’ diastolic dysfunction is associated with long- evidence of increased myocardial collagen turnover axis systolic dysfunction. Implications for the diag- linked to diastolic dysfunction. Circulation 2007; 115: nosis and classification of heart failure. Eur J Heart 888–895. Fail 2005; 7: 820–828. 46 Ahmed SH, Clark LL, Pennington WR, Webb CS, 27 Levy D, Larson MG, Vasan RS, Kannel WB, Ho KK. Bonnema DD, Leonardi AH et al. Matrix metallopro- The progression from hypertension to congestive teinases/tissue inhibitors of metalloproteinases: rela- heart failure. JAMA 1996; 275: 1557–1562. tionship between changes in proteolytic determinants 28 Lakatta EG, Levy D. Arterial and cardiac aging: major of matrix composition and structural, functional, and shareholders in cardiovascular disease enterprises: clinical manifestations of hypertensive heart disease. part I: aging arteries: a ‘set up’ for vascular disease. Circulation 2006; 113: 2089–2096. Circulation 2003; 107: 139–146. 47 Lopez B, Gonzalez A, Querejeta R, Larman M, Diez J. 29 O’Rourke MF, Hashimoto J. Mechanical factors in Alterations in the pattern of collagen deposition may arterial aging. J Am Coll Cardiol 2007; 50: 1–13. contribute to the deterioration of systolic function in

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 304 hypertensive patients with heart failure. J Am Coll 61 van Heerebeck L, Borbe´ly A, Niessen HWM, Bronz- Cardiol 2006; 48: 89–96. waer JGF, van der Velden J, Stienen GJM et al. 48 Shirwany A, Weber KT. Extracellular matrix remodel- Myocardial structure and function differ in systolic ing in hypertensive heart disease. J Am Coll Cardiol and diastolic heart failure. Circulation 2006; 113: 2006; 48: 97–98. 1966–1973. 49 Lorell BH, Carabello BA. Left ventricular hypertro- 62 Borbe´ly A, van der Velden J, Papp Z, Bronzwaer JGF, phy: pathogenesis, detection, and prognosis. Circula- Edes I, Stienen GJ et al. Cardiomyocyte stiffness tion 2000; 102: 470–479. in diastolic heart failure. Circulation 2005; 111: 50 Verdecchia P, Schillaci G, Borgioni C, Ciucci A, 774–781. Battistelli M, Bartoccini C et al. Adverse prognostic 63 Aurigemma GP, Zile MR, Gaasch WH. Contractile significance of concentric remodeling of the left behavior of the left ventricle in diastolic heart failure: ventricle in hypertensive patients with normal with emphasis on regional systolic function. Circula- left ventricular mass. J Am Coll Cardiol 1995; 25: tion 2006; 113: 296–304. 871–878. 64 Katz AM, Zile MR. New molecular mechanism 51 Sadler DB, Aurigemma GP, Williams DW, Reda DJ, in diastolic heart failure. Circulation 2006; 113: Materson BJ, Gottdiener JS. Systolic function in 1922–1925. hypertensive men with concentric remodeling. 65 Wang J, Khoury DS, Yue Y, Torre-Amione G, Nagueh Hypertension 1997; 30: 777–781. SF. Left ventricular untwisting rate by speckle 52 de Simone G, Devereux RB, Koren MJ, Mensah GA, tracking echocardiography. Circulation 2007; 116: Casale PN, Laragh JH. Midwall left ventricular 2580–2586. mechanics: an independent predictor of cardiovas- 66 Drazner MH. The transition from hypertrophy to cular risk in arterial hypertension. Circulation 1996; failure: how certain are we? Circulation 2005; 112: 93: 259–265. 936–938. 53 Aurigemma GP, Gottdiener JS, Shemanski L, Gardin J, 67 Rame JE, Ramilo M, Spencer N, Blewett C, Mehta SK, Kitzman D. Predictive value of systolic and diastolic Dries DL et al. Development of a depressed left function for incident congestive eart failure in the ventricular ejection fraction in patients with left elderly: the Cardiovascular Health Study. J Am Coll ventricular hypertrophy and a normal ejection frac- Cardiol 2001; 37: 1042–1048. tion. Am J Cardiol 2004; 93: 234–237. 54 Aurigemma GP, Silver KH, Priest MA, Gaasch WH. 68 Drazner MH, Rame JE, Marino EK, Gottdiener JS, Geometric changes allow normal ejection fraction Kitzman DW, Gardin JM et al. Increased left ventri- despite depressed myocardial shortening in hyper- cular mass is a risk factor for the development of a tensive left ventricular hypertrophy. J Am Coll depressed left ventricular ejection fraction within five Cardiol 1995; 26: 195–202. years: the Cardiovascular Health Study. J Am Coll 55 Palmon LC, Reichek N, Yeon SB, Clark NR, Brownson Cardiol 2004; 43: 2207–2215. D, Hoffman E et al. Intramural myocardial shortening 69 Rosen BD, Edvarsen T, Lai S, Castillo E, Pan L, in hypertensive left ventricular hypertrophy Jerosch-Herold M et al. Left ventricular concentric with normal pump function. Circulation 1994; 89: remodeling is associated with decreased global and 122–131. regional systolic function: the Multi-Ethnic Study of 56 Palka P, Lange A, Fleming AD, Donnelly JE, Atherosclerosis. Circulation 2005; 112: 984–991. Dutka DP, Starkey IR et al. Differences in myocardial 70 Lam CS, Roger VL, Rodeheffer RJ, Bursi F, Borlaug velocity gradient measured throughout the cardiac BA, Ommen SR et al. Cardiac structure and ventri- cycle in patients with hypertrophic cardiomyopathy, cular–vascular function in persons with heart failure athletes and patients with left ventricular hypertro- and preserved ejection fraction from Olmsted County, phy due to hypertension. J Am Coll Cardiol 1997; 30: Minnesota. Circulation 2007; 115: 1982–1990. 760–768. 71 Harris KM, Spirito P, Maron MS, Zenovich AG, 57 Yuda S, Short L, Leano R, Marwick TH. Myocardial Formisano F, Lesser JR et al. Prevalence, clinical abnormalities in hypertensive patients with normal profile and significance of left ventricular remodeling and abnormal left ventricular filling: a study of in the end-stage phase of hypertrophic cardiomyo- ultrasound tissue characterization and strain. Clin pathy. Circulation 2006; 114: 216–225. Sci (Lond) 2002; 103: 283–293. 72 Burkhoff D, Maurer SM, Packer M. Heart failure with 58 Wang M, Yip GW, Wang AY, Zhang Y, Ho PY, Tse MK a normal ejection fraction, is it really a disorder of et al. Tissue Doppler imaging provides incremental diastolic function? Circulation 2003; 107: 656–658. prognostic value in patients with systemic hyperten- 73 Melenovsky V, Borlaug B, Rosen B, Hay I, Ferrucci L, sion and left ventricular hypertrophy. J Hypertens Morell C et al. Cardiovascular features of heart failure 2005; 23: 183–191. with pre-served ejection fraction versus non-failing 59 Wachtell K, Rokkedal J, Bella JN, Aalto T, Dahlo¨fB, hypertensive left ventricular hypertrophy in the Smith G et al. Effect of electrocardiographic left urban Baltimore community. J Am Coll Cardiol ventricular hypertrophy on left ventricular systolic 2007; 49: 198–207. function in systemic hypertension (the LIFE Study). 74 Liu CP, Ting CT, Lawrence W, Maughan WL, Chang Losartan Intervention For Endpoint. Am J Cardiol MS, Kass DA. Diminished contractile response to 2001; 87: 54–60. increased heart rate in intact human left ventricular 60 Calderone A, Takahashi N, Izzo NJ, Thaik CM, hypertrophy: systolic versus diastolic determinants. Colucci WS. Pressure and volume-induced left Circulation 1993; 88: 1893–1906. ventricular hypertrophies are associated with distinct 75 Zile MR, Baicu CF, Gaasch WH. Diastolic heart myocyte phenotypes and differential induction of failure: abnormalities in active relaxation and passive peptide growth factor mRNAs. Circulation 1995; 92: stiffness of the left ventricle. N Engl J Med 2004; 350: 2385–2390. 1953–1959.

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 305 76 Yamanaka T, Onishi K, Tanabe M, Dohi K, 92 Abraham TP, Dimaano VL, Liang HY. Role of tissue Funabiki-Yamanaka K, Fujimoto N et al. Force- and Doppler and strain echocardiography in current relaxation-frequency relations in patients with clinical practice. Circulation 2007; 116: 2597–2609. diastolic heart failure. Am Heart J 2006; 152: 93 Rademakers FE, Buchalter MB, Rogers WJ, Zerhouni 966e1–966e7. EA, Weisfeldt ML, Weiss JL et al. Dissociation 77 Westermann D, Kasner M, Steendijk P, Spillmann F, between left ventricular untwisting and filling. Riad A, Weitmann K et al. Role of left ventricular Accentuation by catecholamines. Circulation 1992; stiffness in heart failure with normal ejection fraction. 85: 1572–1581. Circulation 2008; 117: 2051–2060. 94 Arts T, Reneman RS, Veenstra PC. A model of the 78 Kawaguchi M, Hay I, Fetics B, Kass DA. Combined mechanics of the left ventricle. Ann Biomed Eng ventricular systolic and arterial stiffening in patients 1979; 7: 299–318. with heart failure and preserved ejection fraction: 95 Beyar R, Sideman S. Left ventricular mechanics implications for systolic and diastolic reserve limita- related to the local distribution of oxygen demand tions. Circulation 2003; 107: 714–720. throughout the wall. Circ Res 1986; 58: 664–677. 79 Maurer MS, King DL, El-Khoury Rumbarger L, 96 Torrent-Guasp F, Ballester M, Buckberg GD, Carreras Packer M, Burkhoff D. Left heart failure with a F, Flotats A, Carrio I et al. Spatial orientation of the normal ejection fraction: identification of different ventricular muscle band: physiologic contribution pathophysiologic mechanisms. J Card Fail 2005; 11: and surgical implications. J Thorac Cardiovasc Surg 177–187. 2001; 122: 389–392. 80 Missouris CG, Belli AM, MacGregor GA. ‘Apparent’ 97 Sengupta PP, Khandheria BK, Korinek J, Wang J, heart failure: a syndrome caused by renal artery Jahangir A, Seward JB et al. Apex-to-base dispersion stenosis. Heart 2000; 83: 152–155. in regional timing of left ventricular shortening and 81 Yu CM, Chau E, Sanderson JE, Fan K, Tang MO, Fung lengthening. J Am Coll Cardiol 2006; 47: 163–172. WH et al. Tissue Doppler echocardiographic evidence 98 Notomi Y, Popovic ZB, Yamada H, Wallick DW, of reverse remodeling and improved synchronicity by Martin MG, Oryszak SJ et al. Ventricular untwisting: simultaneously delaying regional contraction after a temporal link between left ventricular relaxation biventricular pacing therapy in heart failure. Circula- and suction. Am J Physiol Heart Circ Physiol 2008; tion 2002; 105: 438–445. 294: H505–H513. 82 Yu CM, Zhang Q, Yip GWK, Lee PW, Kum LC, Lam 99 Notomi Y, Srinath G, Shiota T, Martin-Miklovic MG, YY et al. Diastolic and systolic asynchrony in patients Beachler L, Howell K et al. Maturational and with diastolic heart failure: a common but ignored adaptive modulation of left ventricular torsional condition. J Am Coll Cardiol 2007; 49: 97–105. biomechanics: Doppler tissue imaging observation 83 Wang J, Kurrelmeyer KM, Torre-Amione G, Nagueh from infancy to adulthood. Circulation 2006; 113: SF. Systolic and diastolic dyssynchrony in patients 2534–2541. with diastolic heart failure and the effect of medical 100 Takeuchi M, Nakai H, Kokumai M, Nishikage T, Otani therapy. J Am Coll Cardiol 2007; 49: 88–96. S, Lang RM. Age-related changes in left ventricular 84 Yip GW, Zhang Y, Wang M, Ho PY, Sanderson JE. Left twist assessed by two-dimensional speckle-tracking ventricular long axis changes in early diastole and imaging. J Am Soc Echocardiogr 2006; 19: 1077–1084. systole are closely coupled to systolic function. Clin 101 Waldman LK, Fung YC, Covell JW. Transmural Sci (Lond) 2002; 102: 515–522. myocardial deformation in the canine left ventricle. 85 Notomi Y, Setser RM, Shiota T, Martin-Miklovic MG, Normal in vivo three-dimensional finite strains. Circ Weaver JA, Popovic ZB et al. Assessment of left Res 1985; 57: 152–163. ventricular torsional deformation by Doppler tissue 102 Ashikaga H, Criscione JC, Omens JH, Covell JW, imaging: validation study with tagged magnetic Ingels Jr NB. Transmural left ventricular mechanics resonance imaging. Circulation 2005; 111: 1141–1147. underlying torsional recoil during relaxation. Am J 86 Notomi Y, Lysyansky P, Setser RM, Shiota T, Popovic Physiol Heart Circ Physiol 2004; 286: H640–H647. ZB, Martin-Miklovic MG et al. Measurement of 103 Henein MY, Gibson DG. Long axis function in ventricular torsion by two-dimensional ultrasound disease. Heart 1999; 81: 229–231. speckle tracking imaging. J Am Coll Cardiol 2005; 45: 104 Wang J, Khoury DS, Yue Y, Torre-Amione G, Nagueh 2034–2041. SF. Preserved left ventricular twist and circumferen- 87 Helle-Valle T, Crosby J, Edvardsen T, Lyseggen E, tial deformation, but depressed longitudinal and Amundsen BH, Smith HJ et al. New noninvasive radial deformation in patients with diastolic heart method for assessment of left ventricular rotation: failure. Eur Heart J 2008; 29: 1283–1289. speckle tracking echocardiography. Circulation 2005; 105 Takeuchi M, Borden WB, Nakai H, Nishikage T, 112: 3149–3156. Kokumai M, Nagakura T et al. Reduced and delayed 88 Epstein FH. MRI of left ventricular function. J Nucl untwisting of the left ventricle in patients with Cardiol 2007; 14: 729–744. hypertension and left ventricular hypertrophy: a 89 Jung B, Markl M, Foll D, Hennig J. Investigating study using two-dimensional speckle tracking ima- myocardial motion by MRI using tissue phase ging. Eur Heart J 2007; 28: 2756–2762. mapping. Eur J Cardiothorac Surg 2006; 29 106 Sengupta PP, Tajik AJ, Chandrasekaran K, Khandheria (Suppl 1): S150–S157. BK. Twist mechanics of the left ventricle: principles 90 Yip G, Abraham TP, Belohlavek M, Khandheria BK. and application. J Am Coll Cardiol Img 2008; 1: Clinical application of strain echocardiography. JAm 366–376. Soc Echocardiogr 2003; 16: 1334–1342. 107 Mottram PM, Haluska BA, Leano R, Carlier S, Case C, 91 Marwick TH. Measurement of strain and strain rate by Marwick T. Relation of arterial stiffness to diastolic echocardiography: ready for prime time? J Am Coll dysfunction in hypertensive heart disease. Heart Cardiol 2006; 47: 1313–1327. 2005; 91: 1551–1556.

Journal of Human Hypertension Hypertensive heart failure GW Yip et al 306 108 Saito M, Okayama H, Nishimura K, Ogimoto A, comparison to systolic heart failure. JAMA 2002; 288: Ohtsuka T, Inoue K et al. Possible link between large 2144–2150. artery stiffness and coronary flow velocity reserve. 117 Lakatta EG, Levy D. Arterial and cardiac aging: major Heart 2008; 94: e20. shareholders in cardiovascular disease enterprises: 109 Hundley WEG, Kitzman DW, Morgan TM, Hamilton part I: aging arteries: a ‘set up’ for vascular disease. CA, Darty SN, Tewart KP et al. Cardiac cycle- Circulation 2003; 107: 139–146. dependent changes in aortic area an distensibility 118 Lakatta EG, Levy D. Arterial and cardiac aging: major are reduced in older patients with isolated diastolic shareholders in cardiovascular disease Enterprises: heart failure and correlate with exercise intolerance. part II: the aging heart in health: links to heart J Am Coll Cardiol 2001; 38: 796–802. disease. Circulation 2003; 107: 346–354. 110 Higginbotham MB. Cardiac performance during sub- 119 Keteyian SJ, Levine AB, Brawner CA, Kataoka T, maximal and maximal exercise in healthy persons. Rogers FJ, Schairer JR et al. Exercise training in Heart Fail 1988; 4: 68–76. patients with heart failure. A randomized, controlled 111 Kitzman DW. Diastolic dysfunction: one piece of the trial. Ann Intern Med 1996; 124: 1051–1057. heart failure with normal ejection fraction puzzle. 120 Keteyian SJ, Brawner CA, Schairer JR, Levine TB, Circulation 2008; 117: 2044–2046. Levine AB, Rogers FJ et al. Effects of exercise training 112 Kitzman DW, Higginbotham MB, Cobb FR, Sheikh on chronotropic incompetence in patients with heart KH, Sullivan M. Exercise intolerance in patients with failure. Am Heart J 1999; 138(2 Part 1): 233–240. heart failure and preserved left ventricular systolic 121 Hambrecht R, Fiehn E, Weigl C, Gielen S, Hamann C, function: failure of the Frank-Starling mechanism. Kaiser R et al. Regular physical exercise corrects J Am Coll Cardiol 1991; 17: 1065–1072. endothelial dysfunction and improves exercise capa- 113 Borlaug BA, Melenovsky V, Russell SD, Kessler K, city in patients with chronic heart failure. Circulation Pacak K, Becker LC et al. Impaired chronotropic and 1998; 98: 2709–2715. vasodilator reserves limit exercise capacity in pa- 122 Hambrecht R, Gielen S, Linke A, Fiehn E, Yu J, tients with heart failure and a preserved ejection Walther C et al. Effects of exercise training on left fraction. Circulation 2006; 114: 2138–2147. ventricular function and peripheral resistance in 114 Brubaker PH, Kee-Chan J, Stewart KP, Fray B, Moore patients with chronic heart failure: a randomized B, Kitzman DW. Chronotropic incompetence and its trail. JAMA 2000; 283: 3095–3101. contribution to exercise intolerance in older 123 Guazzi M, Samaja M, Arena R, Vicenzi M, Guazzi heart failure patients. J Cardiopulm Rehabil 2006; MD. Long-term use of sildenafil in the therapeutic 26: 86–89. management of heart failure. J Am Coll Cardiol 2007; 115 Hundley WG, Bayram E, Hamilton CA, Hamilton EA, 50: 2136–2144. Morgan TM, Darty SN et al. Leg flow-mediated 124 Lewis GD, Lachmann J, Camuso J, Lepore JJ, Shin J, arterial dilation in elderly patients with heart failure Martinovic ME et al. Sildenafil improves exercise and normal left ventricular ejection fraction. Am J hemodynamics and oxygen uptake in patients with Physiol Heart Circ Physiol 2007; 292: H1427–H1434. systolic heart failure. Circulation 2007; 115: 59–66. 116 Kitzman DW, Little WC, Brubaker PH, Anderson RT, 125 Goldsmith SR. Type 5 phosphodiesterase inhibition Hundley WG, Marburger CT et al. Pathophysiological in heart failure: the next step. J Am Coll Cardiol 2007; characterization of isolated diastolic heart failure in 50: 2145–2147.

Journal of Human Hypertension