Heart Failure at the Crossroads: Moving Beyond Blaming Stakeholders to Targeting the Heart

Unlike myocardial infarction, improvement in outcomes for patients with heart failure (HF) by comparison has lagged behind. Despite the discovery of effective therapies for HF and reduced ejection fraction (HFrEF) in the chronic setting, these patients remain at high risk. The mortality risk for patients with HF and preserved EF (HFpEF), or for patients hospitalized for HF irrespective of EF, remains remarkably high,1 and no approved therapies exist so far for either of these two groups of patients. Many efforts over the last two decades to improve outcomes for HF patients, especially in the latter two subgroups, have failed, leading to considerable contemplation by the scientific community.1-4 This viewpoint summarizes the overarching positions from the meeting ‘Heart Failure at Crossroads’ held in Bergamo, Italy on 27–29 November 2014 sponsored by the Clinical Research Foundation of Bergamo Hospital and endorsed by the Italian Federation of Cardiology. Many efforts to reduce HF hospitalizations have centred on the assumption that these are related to patient non-compliance, leading to multiple telemanagement and self-care intervention trials, and quality improvement efforts targeting non-compliance.5, 6 Unfortunately, they failed to improve outcomes. Data regarding non-compliance as a main reason for hospitalization in HF are mainly observational and fraught with recall, ascertainment, and documentation biases. Poor self-care behaviours are also highly prevalent among stable outpatients.7 There is no doubt that compliance and patient education are important and there should be further studies in greater detail. However, the lack of outcome improvement by interventions targeting non-compliance thus far highlights a need to better understand the primary biological reasons for worsening HF as targets of future intervention development. There remains a gap between best practices and practices implemented in the community. However, the evidence that lack of provision of guideline-derived medical care largely determines poor outcomes in HF is modest. A high proportion of hospitalized patients receive evidence-based therapies, risk factor control, symptomatic improvement, and resolution of HF precipitants when present, e.g. ischaemia, arrhythmias, or infections.8 Despite this, these patients remain at a high risk post-discharge. Increasing provision of guideline-derived therapies and yet persistent poor outcomes suggest that alternative biological explanations should be sought, while continued attempts targeting wider use of guideline-derived medical therapies should also be encouraged. In the USA, the decision of the Centers for Medicare and Medicaid to impose financial penalties for a high readmission rate for HF was rooted in financial considerations, since some hospitals benefited from readmissions,which may in part be related to suboptimal hospital, transitional, and post-discharge care. To avoid incentivizing such a paradigm, these penalties were implemented. There is no other indication by payers to suggest that the 30-day readmission rate should be a biological target. Many HF trials have failed to improve mortality risk. Potential explanations for this lack of success include an incomplete understanding of the pathophysiology, the heterogeneity of the population, inadequate matching of therapeutic mechanisms and primary pathophysiological processes, or suboptimal study designs and conduct of HF clinical trials.2, 3 However, this lack of success over the last over two decades is not a wasted effort as many insights were gained that will guide future efforts. Some have argued that hospitalized HF patients have end-stage disease or that HFpEF patients have a high competing risk of mortality, explaining the observed poor outcomes; however, the data do not support these assertions. In one study, at discharge the systolic blood pressure was 124 ± 24 mmHg, heart rate 76 ± 15 b.p.m., serum sodium 137 ± 11 mEq/L, haemoglobin 12.1 ± 3.4 g/dL, and creatinine 1.4 ± 1.2 mg/dL, and there were 84% patients with no or minimal congestion; characteristics that are not suggestive of end-stage HF.9 In another study among patients with HFrEF, of the 1759 dysfunctional myocardial segments assessed with magnetic resonance imaging, 581 showed no evidence of hyperenhancement. Of the other 1178 segments that did have late gadolinium enhancement, 623 segments had the scar limited to <50% of the wall thickness, showing that in these segments there is potential to improve.10 Thus not further exploring options to improve outcomes of these patients is not prudent. Much of the research focus in HF has been on its secondary biological or clinical manifestation, e.g. neurohormonal activation, arrhythmias, congestion, haemodynamics, and renal function, but not targeting the primary cardiac pathology. Indeed, many of these interventions improve outcomes in chronic HFrEF patients; however, unless therapies are targeted at the abnormalities that lead to worsening cardiac function, it is unlikely that in most cases the therapy will lead to recovery of cardiac function. Currently there is a lack of understanding of the biology of worsening HF; however, most patients have dysfunctional but viable myocardium.10 These biological abnormalities are related to the myocytes, e.g. the signalling pathway, myofibrillar function, mitochondrial energetics, calcium handling etc., microvascular dysfunction, or the interstitium.11 Omics technology provides an unprecedented opportunity to understand disease biology and selectively target interventions. However, currently it is unclear which abnormalities in which groups of patients and in what combination should be targeted. One reason why previous research focused on peripheral HF manifestations is that it produced positive results, whereas some therapies targeted at improving cardiac output were associated with poor outcomes, e.g. inotropes. It is important to realize that with the exception of beta-blockers and cardiac resynchronization, which alter both primary and secondary HF abnormalities, other therapies such as defibrillators or renin–angiotensin–aldosterone system modulation do not revert cardiac function back to normal.12 Thus, despite improved outcomes, patients with persistent abnormal cardiac structure and function need restoration therapies. Inotropes improve cardiac output but also cause myocyte damage. Cardiac reverse remodelling is currently defined as improvement in EF or ventricular volumes.13 However, neither is an accurate assessment of cardiac function, as both are affected by preload and afterload. Also, with so many medications affecting haemodynamics, we may have reached a plateau of benefit from haemodynamically active agents, and further modulation may lead to hypotension and adverse outcomes. Also, many of the drugs that may impact primary cardiac abnormalities may be haemodynamically inert. Thus to better elucidate effective therapies for HF, cardiac reverse remodelling needs to be redefined as improvement in myocyte contractility without safety concerns with parallel improvement in cellular and interstitial characteristics. In the case of myocardial infarction and stroke, the outcomes improvements were related to a sequence of understanding disease biology first, then developing therapies targeting it, followed by quality improvement effort. Unfortunately, among patients with worsening HF, simple quality improvement measures such as discharge instruction, smoking cessation counselling, and growing emphasis on transitions of care planning and telemanagement have prevailed in the absence of proof that these interventions either target the reasons for worsening HF or indeed that they even work. The problem with HF is the failing heart! In the early 1980s the heart was the centre of our thoughts and efforts, but with the poor results of inotropic therapy such as ibopamine we moved from the heart to the periphery, with the subsequent successes obtained in chronic HFrEF. Now is the time to return to the heart (‘swinging pendulum’) as a possible therapeutic target for improving the outcome in HFpEF and in hospitalized patients (Figure 1). Unless we understand the biological determinants of the failing heart better, it will be impossible to segment the patient population appropriately and develop optimal taxonomy, target relevant abnormalities, and, in turn, reverse the abnormal cardiac function. Bypassing this phase of understanding the disease process and developing effective therapies and implementing unproven short-term interventions has not worked over the last two decades and is unlikely to work in the future. Blaming patients, providers, payers, researchers, and the disease will not improve HF outcomes. More research into understanding the fundamental science behind worsening HF is needed. This manuscript was generated from discussions held during an international workshop (Bergamo, Italy, 27–29 November 2015) organized by the Cardiovascular Department of Hospital Papa Giovanni XXIII Bergamo, and the FROM Research Foundation. The workshop was supported by an unrestricted grant from Fondazione Internazionale Menarini, Milan, Italy. Conflict of interest: M.S. reports relationships with Novartis and Abbott Vascular. A.G. has no conflicts to declare. M.G. reports relationships with Abbott, Astellas, AstraZeneca, Bayer, Cardiorentis, CorThera, Cytokinetics, CytoPherx, DebioPharm, Errekappa Terapeutici, GlaxoSmithKline, Ikaria, Intersection Medical, INC, Johnson & Johnson, Medtronic, Merck, Novartis, Ono Pharma, Otsuka, Palatin Technologies, Pericor Therapeutics, Protein Design, Sanofi-Aventis, Sigma Tau, Solvay, Sticares InterACT, Takeda, and Trevena Therapeutics. J.B. is a consultant to Amgen, Bayer, Cardiocell, Celladon, Novartis, Stealth Peptide, Relypsa, Z Pharma, Trevena, and Zensun.