Heart Failure: Can We Define, Assess, and Treat Diastolic Heart Failure?

This editorial refers to ‘Prevalence of preclinical and clinical heart failure in the elderly. A population-based study in Central Italy’, by G.F. Murredu et al., published in this issue on pages 718–729. As a rule, men worry more about what they can't see than about what they can. Attributed to Julius Caesar At the height of the Roman Empire, Romans were expected to live to <30 years of age. The Romans had many engineering advances including those of water transport that led to further improvements in public health and life expectancy by providing drinking water and sanitation to a larger public. Gains made by these public health initiatives had a lasting impact on human health. Fast-forward to modern Italy where the life expectancy is ~82 years of age1 and advances in cardiovascular care have contributed substantially to these gains of the last century. The major modifiable cardiac risk factors (such as tobacco use, elevated cholesterol, hypertension, and physical inactivity) have individually and collectively been addressed in epidemiological studies and clinical trials documenting the clinical gains of modifying either individual or global risk factors. Clarity in the definition of each of these risk factors relied on measurement of the risk factor (e.g. LDL or systolic blood pressure), which added precision by defining at what level the association of a clinical event (e.g. acute coronary syndrome or stroke) occurs. This is evident in heart failure (HF), where knowledge of the underlying structure and function of the heart and quantification of the left ventricular ejection fraction (LVEF) led to clinical trials enrolling those with a low LVEF. A tremendous reduction in the morbidity and mortality associated with HF with systolic dysfunction resulted.2 However, the limitations of this approach are now evident as further characterization of chronic HF has led to the identification of a ‘new’ entity: diastolic heart failure [also known as HF with preserved systolic function or ejection fraction (HF-PEF)]. Although potentially interesting findings for angiotensin-converting enzyme (ACE) inhibitors,3 angiotensin receptor blockers,4 and beta-blockers5 have been reported, none of these drugs has received a recommendation for HF-PEF in current guidelines.2 When exploring the link between this new entity and established data, three questions may be posed. (i) What is the prevalence of HF including HF-PEF? (ii) Can we adequately describe where normal healthy ageing stops and HF, as a disease, starts? (3) How well are the known risk factors for HF treated? The recent analysis by Mureddu et al. provides insight into the current state of affairs.6 Through standard screening methodology, they identified 2001 participants to examine cardiovascular health and the prevalence of heart failure or asymptomatic LV dysfunction. The study is compelling in identifying patients with any LV dysfunction: 42%. They go on to describe further the overall prevalence of clinical HF [requiring either signs or symptoms, New York Heart Association (NYHA) functional class >I, and echocardiographic findings of LV dysfunction]: 6.7%. Of these patients, nearly two-thirds had preserved systolic function, leaving an overall prevalence of symptomatic, low LVEF HF of 3.3% for males and 1.4% for females. These finindings are consistent with those of other population surveys in which adequate data exist for outpatient populations for the overall prevalence of HF—2.2% in Olmsted county7 and 2.5% in Canada8. High-quality outpatient HF prevalence data are lacking in Europe, so the findings of the Valutazione della PREvalenza di DIsfunzione Cardiaca asinTOmatica e di scompenso caRdiaco (PREDICTOR) study add to the body of literature supporting a continued burden of HF in the community as more patients survive longer with current therapies. However, there remains a significant problem with data such as these: what is ‘normal’ diastolic function? While most will accept that for defining systolic dysfunction, an LVEF cut-off of 40% or 50% is a reasonable fencepost from which to start calling a ‘low ejection fraction’, the relevance of this or other cut-off points remains artificially defined.9 The parameters for assessing HF-PEF used in PREDICTOR, those of Redfield et al.,7 are widely used by other studies and thus provide us with direct comparison between studies. In PREDICTOR, we see that patients with a normal LVEF in stage A, B, or C, regardless of symptoms, had a similar incidence of echocardiographic abnormalities in diastolic function. If one considers only those with mild diastolic dysfunction, this suggests that the measurement of these abnormalities is unlikely to provide insight into the underlying pathophysiology of ‘diastolic heart failure’. Furthermore, since most of the echocardiographic measures are a continuum, as is most human disease, establishing dichotomous criteria may be difficult when normal healthy ageing is also associated with the same abnormalities. Consistent with prior studies, elevated levels of natriuretic peptides in PREDICTOR are linked to worse NYHA class, lower LVEF, and worse diastolic dysfunction (with a three-fold increase from mild to moderate–severe diastolic dysfunction). Yet caution must be exercised as the differing distribution of ages and gender and co-morbid conditions are not accounted for in these levels and therefore may simply reflect these other variables given the close relationship between age, gender, and N-terminal pro brain natriuretic peptide (NT-proBNP). No measures of functional changes during exercise, skeletal muscle function, or other biomarkers are provided, leading to speculation that perhaps these would enhance our comprehension of how to define the clinical phenotype(s). Even utilizing the European Society of Cardiology guidelines on how to diagnose HF-PEF would be useful, although these remain a work in progress given the lack of validation of the overall diagnostic schema.10 Further research must focus on the pathophysiological abnormalities of a clinical phenotype that probably encompasses the full spectrum of disease and multiple subtypes—a ‘one size fits all’ approach is unlikely to work. Much like the recognition of different subtypes of chronic HF by aetiology (e.g. ischaemic, alcoholic, peripartum, and myocarditis), HF-PEF will require further dissection in order to understand and either prevent or treat this prevalent disease (see Figure 1). Given the differing underlying pathophysiology that may exist in the subtypes of HF-PEF, it will be important to utilize the diagnostic testing representing the underlying mechanism. For example, exercise testing may highlight a differential exercise reserve or capacity, biomarkers linked to metabolic function may identify those in whom diabetes is the ‘cause’, and tissue level imaging may highlight extracellular matrix remodelling as is found in inflammatory or infiltrative diseases. Regardless, imaging remains only one aspect of identifying patients or their underlying pathophysiological cause which will direct prevention and treatment. Until we establish ‘normal’ with respect to these imaging or other biomarkers in the modern era of testing, clarity on what is within the spectrum of normal, how to define HF-PEF, and how to treat HF-PEF remain open questions.11 While the question of ‘does diastolic dysfunction represents a normal ageing phenomenon’ remains open, the additional risk factors for cardiovascular events identified are unchallenged: hypertension and elevated cholesterol. Whilst the risk, benefits, and patient preferences of any therapy may change with age, there are also choices available for the treatment of asymptomatic LV dysfunction and systolic HF. By way of example, in patients with a known diagnosis of HF enrolled in a study testing for tolerability of carvedilol, it was found that patients across all ages had equal tolerability to a first-line beta-blocker, achieving doses of 30 mg per day, even in those patients over 80 years of age.12 Denying such life-saving therapy is evident from PREDICTOR: This would appear to be in contradiction to the evidence which supports a relative risk reduction equal to, or greater than, that for younger patients enrolled in landmark clinical trials. Denying patients adequate therapy solely due to age and unfounded perceptions of lack of benefit or reduced tolerability should be avoided in medical practice. This important clinical question has led to further tolerability trials which have identified that not only can beta-blockers be tolerated when titrated in a slow and methodical fashion in elderly patients, they do so even in patients with relative contraindications.13 A similar tale can be told for hypertension to prevent HF even for older patients—germane to the findings of Mureddu et al. As an example, in the HYVET trial, systolic blood pressure control in patients >80 years of age led to a 64% relative risk reduction of HF as an endpoint.14 Given the overall prevalence of hypertension (~58%), current undertreatment in the stage A and B patients in PREDICTOR, one could hypothesize that there is opportunity to intervene and reduce the prevalence of HF by simply tackling this issue. Hypertension is a visible risk factor, it is measurable, and interventions are well tolerated. Significant population gains are measureable such as those in Canada, and occur when public health agencies, industry, and non-governmental organizations focus their attention on what we can see—such as elevated systolic blood pressure.15 While the path before the scientific community is challenging for HF-PEF, proven therapy remains to be deployed in patients with risk factors in order to prevent HF. The PREDICTOR study lays the ground work for understanding the prevalence and undertreatment of HF in an ageing population. Getting therapy to patients at risk is critical for the prevention of HF, and understanding how HF-PEF develops, progresses, and can be treated will be critical challenges for this decade. As the quote implies, we need to spend time considering what we can see as well as what we cannot. J.A.E. is funded by the Canadian Institutes of Health Research and Alberta Innovates–Health Solutions and by the team grant Alberta HEART. Conflict of interest: none declared.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.