Unmasking early heart failure with preserved ejection fraction: the role of exercise stress testing

This editorial refers to ‘Prognostic benefit of early diagnosis with exercise stress testing in heart failure with preserved ejection fraction’, by Y. Saito et al., https://doi.org/10.1093/eurjpc/zwad127. Heart failure (HF) with preserved ejection fraction (HFpEF) is a complex heterogeneous systemic syndrome that accounts for approximately half of the patients with HF. Increased recognition, together with ageing of the population and growing burden of comorbidities, especially diabetes, hypertension, and obesity, are among the main reasons for the rising prevalence of this condition.1 Initially, the term diastolic HF was used to address this disorder,2,3 drawing the attention to diastolic dysfunction as the necessary and sufficient abnormality to cause this syndrome. Heart failure with preserved ejection fraction is a broader term, which basically encompasses what is not (even mildly) reduced ejection fraction and includes a heterogeneous population that, besides and beyond diastolic dysfunction, may present subclinical systolic dysfunction, pericardial disease, valve diseases, right heart abnormalities, and endothelial and vascular dysfunction,4 but also specific treatable ‘masqueraders’ that require specialized diagnostic pathways often based on multi-modality imaging.5 Comorbidities and associated conditions, frequently present and variably combined in each patient, may be both confounders and contributing pathogenetic determinants, challenging phenotyping and classification of patients in homogeneous phenogroups.1 Poor understanding of the pathophysiology and, until recently, perceived dearth of treatments able to modify outcomes in HFpEF, have contributed to the scarce interest in early HFpEF diagnosis in daily clinical practice. As a result, the disease is often detected when it has already progressed to an advanced stage and when irreversible derangements have occurred. Heart failure with preserved ejection fraction diagnosis requires the presence of signs or symptoms of HF, an ejection fraction ≥50%, and a documentation of increased cardiac filling pressures. While the diagnosis of HFpEF is unequivocal in the presence of overt congestion and in advanced stages, it is challenging to determine whether filling pressures are normal at rest but become pathologically elevated only during exertion. This implies the need to examine the patient with a dynamic test. Invasive heart catheterization at rest or during stress testing allows adequate measurement of left ventricular filling pressures and is considered the gold standard for the diagnosis of HFpEF6 and has shown prognostic utility in dyspnoeic patients with HFpEF.7 However, catheterization is not universally available nor cost-effective, especially considering the extensive and rising prevalence of HFpEF in the community. Furthermore, it lacks standardization across laboratories and carries the potential risks of an invasive examination.8 For these reasons, there is growing interest to accurately diagnose HFpEF by non-invasive methods in the largest possible number of patients so that invasive catheterization could be reserved for a few selected equivocal cases. In the absence of universally accepted diagnostic criteria, multi-parametric algorithms based on clinical, laboratory, and echocardiography data have been proposed by current guidelines to estimate the probability of HFpEF in patients with unexplained dyspnoea.9,10 In this issue of European Journal of Preventive Cardiology, Saito et al.11 conducted a retrospective study on 368 patients with exertional dyspnoea and 182 were diagnosed with HFpEF applying the Heart Failure Association Pretest Assessment, Echocardiographic and Natriuretic Peptide Score, Functional Testing in Case of Uncertainty, and Final Aetiology (HFA-PEFF) algorithm. The HFA-PEFF algorithm proposed by the European Society of Cardiology (ESC) consists of a stepwise approach that starts with an assessment of the pre-test likelihood based on history, a second morpho-functional step, a third step to search for concealed increased filling pressures by diastolic exercise echocardiography or catheterization, and finally, aetiology characterization.9 The present study has the merit of the application of Step 3 (diastolic exercise stress echocardiography or exercise right heart catheterization), which is omitted in most studies and underused in clinical practice, and which allowed an identification of 60 patients (33%) with HFpEF that would have been missed without a provocative testing. The authors demonstrated that patients diagnosed with HFpEF had higher rates of all-cause death and decompensated HF requiring hospitalization or IV diuretics compared with those without confirmed diagnosis. The stress testing was combined with expired gas analysis, which did not add useful information. It is well known that when HFpEF is symptomatic for congestion and requires hospitalization, the prognosis is poor,12 at least as bad as HFrEF.1 Of note, in the present study, only 5.6% of patients with confirmed HFpEF had a history of previous HF hospitalization. This study demonstrates that early detection of HFpEF, before it becomes clinically manifest, is possible using exercise stress testing, and this is important because these patients are at high risk of adverse outcomes. Stress echocardiogram was performed in all patients regardless of Step 2 score results. Conversely, catheterization was reserved to patients with equivocal non-invasive tests, which may reflect clinical practice. This study was not aimed to validate the algorithm. Interestingly, there was no gradient of risk between patients who were diagnosed with HFpEF at Step 2 (with echocardiographic and biomarker data) vs. patients who qualified as HFpEF only after applying Step 3 (only after provocative echocardiogram or catheterization). This indicates that these patients likely have concealed diastolic dysfunction at rest, but they experience a similar poor prognosis and thus should receive appropriate treatment for their HFpEF. It is conceivable that an awareness of HFpEF status might have improved patient management, including but not limited to intensification of medical therapy, similar to what was recently observed in a dedicated HFpEF clinic.13 The authors concluded that the early therapy initiation improved the prognosis in these HFpEF patients. However, caution should be exercised in deriving this conclusion as this is not a randomized controlled trial and therapy was initiated by referring physicians only in half of the patients after the HFpEF diagnosis with no information on other non-pharmacological interventions nor on whether any changes occurred in the control group. The authors acknowledged that the sample size and number of events were limited, as shown by the wide confidence intervals, but there was also imbalance between groups, with HFpEF subjects being older. Finally, as far as external validity was concerned, the study was conducted at a tertiary referral centre and involved patients of Asian origin, a population in whom cardiometabolic abnormalities are frequent determinants of HFpEF and can occur with normal weight and in whom sodium glucose cotransporter-2 inhibitors, the only Class IIa HFpEF therapy recommended currently, seem to be more beneficial than in other populations.14 Well-designed randomized trials are needed to determine whether (and which) therapy is associated with improved clinical outcomes in patients with early-stage HFpEF, yet the study by Saito et al. interestingly highlights that the focus of HFpEF research should shift to the early stages of the disease. The study by Saito et al. is notable because it underscores the importance of dynamic assessment in HFpEF and it corroborates the clinical usefulness of the ESC-proposed HFA-PEFF score comprehensive of exercise testing as a diagnostic algorithm, but especially as a risk-stratification tool.