Heart failure (HF) is a chronic syndrome with an ever-increasing prevalence and high risk of mortality, hospitalizations and disability, particularly for patients not receiving optimal guideline-directed medical therapy (GDMT).1,2 Accurate risk stratification is a key step in HF assessment, and misperception of risk can lead to suboptimal care.3 Physicians managing HF patients are more easily prompted to modify the therapy in those with New York Heart Association (NYHA) class III or IV symptoms than in those who are asymptomatic or have a no more than mild limitation of functional capacity.3 Whereas the former are expected to have a progressively worsening condition, characterized by recurrent hospitalizations and high short to intermediate term mortality, the latter are perceived to have a low risk of adverse outcomes.3 Consequently, physicians will often refrain from optimizing drugs and doses, or implanting devices in HF with reduced ejection fraction (HFrEF) patients that are deemed clinically stable and in NYHA class II. This decision is taken to avoid the risks of side effects and the increased burden of multiple healthcare encounters.4

Yet, HFrEF should be viewed as a progressive disease in patients with a significant mass of at-risk viable myocardium, in whom myocardial dysfunction is amenable to improvement with optimal GDMT.5 Neurohormonal activation, which is initially compensatory, never switches off in HF.5 These systems are constantly on in an attempt to compensate for the failing heart's inability to maintain normal cardiovascular homeostasis. The chronic presence of these circulating neurohormones exacerbates hemodynamic abnormalities, which encourages further remodeling, neurohormone release and additional hemodynamic deterioration.

To illustrate the dire prognosis displayed in HFrEF patients, even if only mildly symptomatic, we assessed the rates of N-terminal pro b-type natriuretic peptide (NTproBNP) elevation and major adverse outcomes in an HFrEF population with good functional capacity, defined by a maximum oxygen consumption (pVO2) higher than 20 mL/kg/min.

In a real-world single-center HFrEF cohort, one in ten patients with good functional capacity diagnosed by CPET had major clinical events over a two-year period of follow up. More than 90% had persistent NT-proBNP elevations.

Heart failure remains a high risk disease with a grim short-term prognosis. Our study cohort consisted of patients receiving appropriate drug therapy (>90% BB and RAASi), who seemed clinically well (good functional capacity and half not requiring a loop diuretic). Nevertheless, 10% died or had at least one HF hospitalization within two years. Similar observations were previously reported in the PARADIGM-HF (Angiotensin–Neprilysin Inhibition versus Enalapril in Heart Failure) trial. Among the least symptomatic patients in NYHA class I or II, 22% either died from a cardiovascular cause or had at least one HF hospitalization during a median follow up of 27 months.10 Furthermore, among the more clinically stable patients without any history of HF hospitalization, 20% either died from a cardiovascular cause or had at least one HF hospitalization during the course of the trial. The primary event was CV death with no preceding HF hospitalization in 51% of these most clinically stable patients, therefore precluding patients and physicians from an alert that could have prompted treatment intensification and optimization to avoid this dismal outcome.11 Overall, 60% of those CV deaths were sudden cardiac deaths.11 In our study, death was preceded by a HF hospitalization in only one of three patients.

In HFrEF, evidence of subclinical disease progression can be found even in patients who seem to be doing well and to be receiving appropriate treatment. Neurohormonal activation persisted in the majority of our study patients, although exercise tolerance, as judged by NHYA class, was normal or only mildly impaired in >90%. Additional evidence of this illusion of clinical stability derives from studies using troponin measurements where the majority of stable HFrEF patients in NYHA class II have chronic cardiac troponin elevations, corroborating the subclinical continuous myocardial lesion.12 Several mechanisms can lead to troponin elevation in HF including supply-demand mismatch, which may occur in the absence of coronary obstruction due to increased oxygen demand related to increased wall tension, anemia, or other factors provoking subendocardial injury. Non-coronary triggers include cellular necrosis or apoptosis in the context of wall stress, as well as the toxic effects of circulating neurohormones, toxins, inflammation, and infiltrative processes. Regardless of the mechanism, troponin elevation in HF is associated with greater tendency toward LV remodeling, and a higher risk of death or hospitalization.12,13 Recent evidence indicates that changes in NT-proBNP levels are predictive of similar directed changes in troponin levels, which in turn are associated with left ventricular forward remodeling.14 Nevertheless, in our population, we had only access to baseline NT-proBNP levels, and serial assessments of cardiac biomarkers were beyond the scope of this study.

On the basis of the above-mentioned observations, GDMT uptitration to maximum tolerated doses or maximum tested doses is paramount and currently recommended in European Society of Cardiology HF guidelines.1 However, the aforementioned illusion of stability may be hazardous and lead to treatment inertia, particularly when dealing with chronic individuals not overwhelmingly symptomatic. Adherence to GDMT recommendations by physicians has traditionally been shown to be suboptimal and clinical stability may be one of the culprit reasons.15,16 Indeed, less than one in five patients admitted for acute HF and discharged on beta-blockers had these uptitrated in the following month.16 These observations account for a reduced number of HFrEF patients achieving GDMT-target doses as demonstrated in the IMPROVE-HF registry.5

Such inertia is particularly important in stable, pauci-symptomatic patients, where misperception of risk and concerns regarding side effects may deter clinicians and patients from appropriate treatment intensification.3 Yet, evidence shows that even less symptomatic, NYHA class II patients benefit from appropriate GDMT. Indeed, such cases often account for a significant number (if not the majority) of patients included in landmark HF trials.17–20 Moreover, the benefits of GDMT in less symptomatic patients are at least similar in magnitude to those observed in more symptomatic patients, as shown in pre-specified subgroup analyses of PARADIGM-HF and HEAAL (effects of high-dose versus low-dose losartan on clinical outcomes in patients with heart failure) trials.18,20

Our findings strengthen previous recommendations: GDMT initiation and uptitration is essential, since even clinically stable, ‘low risk’ patients, with good functional capacity have persistent neurohormonal activation and a significant rate of events. Accordingly, viewing chronic HF as a stable disease may be a misnomer and, therefore, must be avoided.

This study has some limitations. First, it is a single-center retrospective study with a limited sample size. Second, we did not assess other important adverse events related to HF, namely cardiovascular death, malignant arrythmias or implantable cardioverter defibrillator shocks. Third, we did not assess whether GDMT was being taken at maximum recommended dose. However, in order to estimate if our cohort was close to (or distant from) optimal GDMT we assessed patients’ SBP and HR as surrogates – assuming that those presenting a low SBP/HR would not tolerate further up-titration of renin-angiotensin-aldosterone inhibitors and/or beta-blockers. Only 36% of patients had a systolic blood pressure <110 mmHg. Although it may be possible that some patients with higher SBP and HR were still in the process of up-titration, already on maximal doses of disease-modifying drugs or unable to reach these because of intolerance or other limitations (e.g., kidney failure or hyperkalemia), it seems reasonable to consider that a significant proportion of the patients in our cohort would be indeed amenable to further drug up-titration. Fourth, included patients performed an echocardiographic assessment within six months of CPET. We cannot exclude a possible worsening or improvement of LGEF within this time frame.

In a clinically stable HFrEF population displaying preserved functional capacity, persistent neurohormonal activation was present in the majority and one in ten patients died or had one HF admission at two years post-CPET. These findings rightfully challenge the perception that treatment optimization may be less important for low-symptomatic patients. Uptitration of pharmacological therapy to maximal tolerated or tested doses is mandatory to slow disease progression as effectively as possible, even when the clinical scenario seems (deceptively) favorable. Therefore, strategies are urgently needed to motivate clinicians to pursue GDMT even in such patients.

SM wrote the first draft of this article; CB, GJL, BR, and FG incorporated feedback in subsequent drafts and revisions; CS, PF, AD, AT, AV, CA and MM contributed to revisions and reviewed the final draft; SM composed the figures and submitted the final version of this article, on behalf of all the authors. All named authors complied with the International Committee of Medical Journal Editors guidelines for authorship.

This investigation was conducted in accordance with the World Medical Association Declaration of Helsinki (seventh revision, Fortaleza, 2013) and the Declaration of Istanbul (2008). This study was exempt from ethical approval.

None declared.

The authors have no conflicts of interest to declare.