With the widespread availability of non-invasive bedside assessment of ventricular function by echocardiography in the last two decades the paradigm of heart failure (HF), presumed to be usually due to an impaired left ventricular (LV) systolic function, has been challenged by the finding that about half of patients suffering from HF syndrome exhibit a normal LV ejection fraction (LVEF), and the established notion that HF is mostly associated with a poorly contracting and dilated ventricle revealed as a misconception. At the present time the terms 'diastolic heart failure' or HF with normal/preserved ejection fraction (HFnEF/HFpHF) are often used interchangeably. Because it is well known that significant LV systolic dysfunction is present in patients with chronic regurgitant valvular heart disease despite a near normal EF or in volume overload conditions (iatrogenic or not), or high cardiac-output states, it follows that these terms are not equivalents and, according to Kennet and Zile et al., patients with DHF must be viewed as a particular great subgroup (characterised by normal EF and primary diastolic dysfunction, usually due to hypertension and concentric LV hypertrophy (LVH) of patients with HFpEF.
DHF has been observed worldwide in different racial groups and appears to be increasingly important in both North America and Europe. Although the prevalence of all types of congestive heart failure (CHF) increases with age, DHF appears more than systolic HF a disease of aged individuals, mainly women. African-Americans appear to be affected at a relatively younger age because of greater prevalence of hypertension and LVH. It has become increasingly evident that HF patients enrolled in therapeutic clinical trials do not represent real-world patients with HF in that these are older, more often women, with preserved LV systolic function, with significantly lower prevalence of previous myocardial infarction (MI) and a higher prevalence of concomitant morbid conditions such as hypertension, diabetes, obesity and renal dysfunction than clinical trial patients. Better preservation of cardiac myocytes number and greater life expectancy of cardiac myocytes in women than in men account for the greater LVEF in women; a finding confirmed by the recent Dallas Heart Study suggesting greater propensity of women to develop concentric LVH and diastolic dysfunction with normal-supranormal EF in response to hypertension, whereas the cardiac phenotype of secondary heart field (SHF), characterised by LV dilation and reduced LVEF, is more frequently observed in men who exhibit progressive loss of cardiac myocytes by apoptosis with ageing and higher prevalence of myocardial necrosis by myocardial ischaemia (MI).
When valvular cause of heart failure is considered, aortic stenosis appears as a frequent underlying condition associated with DHF in the elderly. Less common causes include restrictive cardiomyopathy, pericardial disease and hypertrophic cardiomypathy. On echocardiography, compared with the clinical phenotype of SHF, which is characterised by LV dilatation, and reduced LVEF, the phenotype of DHF exhibits concentric LV hypertrophy, smaller LV dimension and mass, but similar LV diastolic dysfunction grade and similar increase in left atrial size as a consequence of similar long-standing increase in LV filling pressures and accounting for the same prevalence of atrial fibrillation. Accordingly, mitral inflow Doppler indexes are not significantly different in patients with either diastolic or systolic CHF because diastolic dysfunction is common to both phenotypes.
Diastolic dysfunction results from a heterogeneous group of disease states (myocardial, pericardial or valvular) of different aetiology and pathophysiology. A large body of evidence has been accumulated from experimental and clinical studies, suggesting that an isolated diastolic dysfunction manifesting with failure of the Frank-Starling mechanism, with a reduced stroke volume response and an up- and leftward shift of the LV end-diastolic pressure-volume relation during exercise, is central to the pathophysiology of the most common form of DHF, i.e. that associated with hypertension, obesity and diabetes. Myocytes hypertrophy and varying degree of interstitial fibrosis are typical pathological features. The following mechanisms have been identified to have responsibility in DHF syndrome: extramyocardial factors (haemodynamic load, heterogeneity, pericardium), myocardial factors (cardiomyocytes, myofilaments, energetics, cytoskeleton), extracellular matrix (fibrillar collagen, basement membrane proteins, proteoglycans, matrix metallo-proteinases (MMP)/tissue inhibitor of MMP (TIMP)) and neurohormonal activation. Evidence from invasive assessment that DHF patients have abnormal diastolic properties (incomplete relaxation and increased passive stiffness) of the left ventricle that 'are sufficient to explain the patients' haemodynamic abnormalities and occurrence of heart failure' has been provided.
Comparing exercise haemodynamics in healthy elderly with age-matched patients with either SHF or DHF, it has been shown that patients with DHF show a similar reduction in exercise workload, VO2, cardiac output and stroke volume, with similar increase in exercise pulmonary capillary wedge pressure and similar degree of haemodynamic derangement to that exhibited by patients with SHF. This is with a completely different mechanism associated with an increased exercise LV end-systolic volume in SHF, but with a reduced exercise LV end-diastolic volume in patients with DHF, reflecting the very different basic myocardial impairment of these two HF phenotypes.
It has been shown that patients with DHF exhibit exercise tolerance intermediate between SHF patients and controls, but with an excessive breathlessness for a given minute ventilation. Arguably, in relation to the less severe neurohormonal activation, a less grim prognosis appears associated with DHF.
In the last few years evidence has been provided that a number of nondiastolic mechanisms exist in patients with diastolic CHF, potentially influencing exercise capacity, mainly in older patients. The first one is represented by dynamic LV outflow tract obstruction in absence of hypertrophic cardiomyopathy, which has been show to occur in a subset of patients with exertional dyspnoea with a reduced LV cavity size, mild LV hypertrophy, supranormal systolic function and moderate diastolic abnormalities at rest, therefore suggesting a potential causative link between pressure gradient and effort-induced dyspnoea and its contribution to the clinical picture of diastolic CHF in these patients. A 6% prevalence of this subgroup of patients with dynamic LV outflow tract obstruction induced by low-dose dobutamine infusion was demonstrated.
Another systolic mechanism substantially contributing to exercise intolerance has been identified by Kass et al. in another subset of hypertensive patients with severe LV diastolic dysfunction. This subset of patients has been previously described by Topol et al. as carriers of the so-called 'hypertrophic hypertensive cardiomyopathy of the elderly' and are characterised by marked LV concentric hypertrophy with supranormal LVEF and distal cavity obliteration during systole. In these patients, in addition to the inability to utilise the Frank-Starling mechanism, the exercise performance is reduced by a systolic mechanism represented by a reduced systolic reserve owing to resting LV cavity obliteration and consequent inability to further improve stroke volume during exercise. It has been proved that reducing cavity obliteration by VDD pacing can restore the reduced systolic reserve and can improve exercise capacity in these patients.
A third nondiastolic mechanism contributing to exercise intolerance in elderly patients with hypertensive DHF is a reduced cardiac-cycle dependent on changes in aortic area and distensibility, as well as an increased arterial stiffness. Limitation of both systolic and diastolic reserve function by these mechanisms, and in particular by systolic-ventricular and arterial stiffening beyond that associated with ageing and/or hypertension, has been demonstrated by Kawaguchi et al., with LV pressure-volume relations measurement in ten patients (nine female) with recent pulmonary oedema and EF >50%.
These patients exhibited higher LV end-diastolic pressure (EDP) than controls with significant upward shift of EDP volume relations in both early and late diastole, yet similar isovolumic relaxation rate. As a consequence of combined ventricular-arterial stiffening, these patients had greater hypertensive response than control subjects after isometric handgrip exercise with adverse consequences on LV diastolic function as represented by a significantly longer relaxation time and a marked increase in LVEDP, ultimately supporting a mechanism whereby ventricular arterial stiffening could couple to diastolic dysfunction.
Based on these findings and in line with previous studies showing improved exercise tolerance with pharmacological intervention aimed at reducing ventricular-vascular stiffening in aged subjects, it has been suggested that similar intervention targeted to reverse the abnormal cardiac-arterial coupling would be a novel interesting therapeutic strategy for reducing exercise intolerance either in DHF and SHF patients.
Three obligatory conditions to be simultaneously satisfied for the diagnosis of DHF have been established by the European Working Group in 1998:
- presence of signs or symptoms of congestive heart failure;
- presence of normal or mildly abnormal LV systolic function; and
- evidence of diastolic LV dysfunction.
Updating of these guidelines will be published shortly in a consensus statement of the Heart Failure and Echocardiographic Associations of the European Society of Cardiology.
In recent years, a routine widespread incorporation of diastolic function assessment into echo-labs has occurred and different patterns and grading of LV diastolic function impairment have been established by the combined study of transmitral and pulmonary vein flow or tissue Doppler imaging (by E/E' ratio), ultimately providing a refined evaluation of LV diastolic function and obviating the need of invasive measurements. Moreover, the left atrial volume by 2-D echocardiography in absence of mitral regurgitation represents the so-long-expected, easy-to-obtain, load-independent and reliable index of chronic diastolic dysfunction regardless of the actual loading conditions. In addition, serum markers of increased LV diastolic pressure, such as BNP, simplified the diagnostic approach, ultimately making the problem of an accurate diagnosis of diastolic CHF less challenging.
Furthermore, it has been shown that virtually all or the great majority of patients with the clinical diagnosis of DHF do exhibit an abnormal LV diastolic function at cardiac catheterisation or at Doppler investigation when assessed by standard criteria also including the E/E' ratio or the combined transmitral and pulmonary vein flow study, ultimately confirming the original Gaasch suggestion that CHF in patients without valvular or pulmonary disease and normal LV systolic function is a manifestation of diastolic dysfunction.
Few clinical trials have been carried out in these patients mainly because of difficulties in recruiting the older, predominantly female patients with several co-morbidities. This explains also why trials in these patients also have high drop-out rates and follow-up may be limited. Unlike SHF there is no evidence-based treatment for DHF, which improves clinical outcomes. The only evidence for any treatment effect comes from small studies with verapamil where this drug increased exercise capacity and reduced a heart failure score. Three randomised trials have been completed in the last three years, including a percentage of patients with near normal LVEF: the CHARM-Preserved, the SENIORS and PEP-CHF trials. Three trials that are examining the effect of treatment with the β-blocker carvedilol, the AT1 receptor antagonist irbesartan and the aldosterone antagonist in patients with DHF are currently under way.
The first landmark therapeutic study in patients with diastolic CHF, the CHARM-Preserved trial, was published at the end of 2003. In this study, investigating the affect of the AT1 receptor antagonist candesartan, some effect on hospitalisation rate has been observed but no therapeutic effect of candesartan on cardiovascular mortality emerged during the three-year follow-up, suggesting that this group of patients is difficult to treat. The SENIORS study suggests that nebivolol, a β-blocking agent with NO donor properties, may be beneficial in patients with a relatively preserved LVEF, but the cut point of 0.35 for the 'preserved' LVEF subgroup used in this analysis is misguiding and a further analysis using a higher partition value of LVEF would be valuable. The multicenter double-blind, placebo-controlled PEP-CHF trial planned to assess the usefulness of the ACE inhibitor perindopril 4mg per day in a group of 1,000 aged patients (ÔëÑ70 years old) with DHF has recently concluded. As also shown by the CHARM-Preserved trial, the PEP-CHF failed to meet its primary end-points, and this has been attributed to the short follow-up duration.
A large study in DHF funded by the US National Heart, Lung and Blood Institute (NHLBI), the Treatment Of Preserved Cardiac function heart failure with an Aldosterone anTagonist (TOPCAT) trial is planned to examine the effects of aldosterone antagonist spironolactone (15mg) vs placebo in 4,500 adult patients with HF. Patients will be enrolled on the basis of an echocardiographically assessed EF and will be recruited over a 2.5-year period and followed for a minimum of two years. Approximately 250 clinical centres are expected to participate and the study is expected to be completed in 2011.
The limited evidence currently available suggests that, while waiting for new evidence and new treatment options, a major therapeutic potential in reducing the human, social, and economic burden associated with the DHF epidemic is that of treating the affected patients following the results of the above mentioned trials, as well as to adopt a set of therapeutic principles directed at reversing or reducing the consequences of the underlying conditions and the precipitating factors. However, at present time, superior results may be achieved by increasing our aggressiveness in treating DHF precursors, i.e. hypertension and all other risk factors for atherosclerosis and LV hypertrophy, such as diabetes, obesity and metabolic syndrome, with the aim of preventing/reversing LV hypertrophy and interrupting the downstream pathway leading to increased vascular and myocardial stiffening and accelerated cardiovascular ageing.
Future treatment options should be targeted to key factors responsible for LVH and myocardial fibrosis, as well as to subcellular determinants of the increased resting tension of the cardiac myocyte in DHF.