Beta-Blockers and Heart Failure with Mildly Reduced and Preserved Ejection Fraction. A Personalized Approach Beyond the Heart Failure Syndrome Phenotypes is Needed

This article refers to ‘Beta-blocker use and outcomes in patients with heart failure and mildly reduced and preserved ejection fraction’ by S. Matsumoto et al., published in this issue on pages xxx. Phenotypic characterization of the heart failure (HF) syndrome based merely on left ventricular ejection fraction (LVEF) may have several pathophysiological and clinical limitations. HF with reduced ejection fraction (HFrEF) is a more homogeneous clinical condition where a reduced contractile performance of the heart, congestion and neurohormonal activation defined by an unbalance between protective and deleterious mechanisms are characteristics providing well defined therapeutic targets. HF with mildly reduced ejection fraction (HFmrEF) and particularly HF with preserved ejection fraction (HFpEF) are more heterogeneous and probably need a better understanding based on pathophysiological and clinical determinants for the clinical syndrome presentation.1 A number of publications were aimed to identify phenotypic (clusters) groups of patients with common clinical characteristics associated with (clusters of patients) prognostic differences and have been suggested that may require a more personalized treatment. The heart's functional and structural alterations accounted for the HFpEF diagnosis may vary according with the underlying clinical conditions of the patients; and have been suggested different clusters of patients which may orientate treatment selection.2 Based on those considerations we would speculate that in a near future the concept of HFpEF may be abandoned and may be substituted by a more useful clinically-based patient profile, such as obese-diabetic, cardio-renal, elderly-atrial fibrillation, which is characterized by normal ejection fraction.3 However, while phenotypic clustering offers a promising framework for personalizing HF management, its real-world application is often challenged by the overlap of clinical characteristics across clusters, limiting its utility for straightforward point-of-care decision-making. Future research should focus on refining these clusters and developing tools to integrate overlapping phenotypes into prospective clinical management strategies. These considerations may influence the clinical trial results in HFpEF and also should be considered for the design of innovative clinical trials on highly prevalent and relevant clusters of patients with HFpEF. Beta-blocker (BB) therapy is a major cornerstone for HFrEF, despite limited evidence due to the absence of randomized clinical trials in patients with HFmrEF (Figure 1). The latest European Society of Cardiology and American Heart Association (AHA)/American College of Cardiology (ACC)/Heart Failure Society of America (HFSA) guidelines recommend the use of BB with the same low class of recommendation (IIb) but with differing levels of evidence (C and B-NR, respectively)4, 5 based on the results from a meta-analysis of 11 randomized trials suggesting that the benefits of BB are observed in this group of patients.6 This analysis suggests that for patients with HF in sinus rhythm, the effect of BB on mortality in patients with LVEF 40–49% is similar to that observed with LVEF <40% without any significant effect in patients in atrial fibrillation despite an observed improvement in LVEF associated with BB therapy. Most clinicians agree with this recommendation; nevertheless, more clinical research should be promoted including dedicated clinical trials for a better knowledge, and to improve the level of evidence, on the effects of BB in HFmrEF. It is also relevant to highlight findings from the REDUCE-AMI trial,7 which reported no significant benefit of BB in post-myocardial infarction (MI) patients with preserved ejection fraction (LVEF >50%). This is particularly important as many patients with HFmrEF fall into the post-MI category and exhibit mildly reduced ejection fraction or HF with improved ejection fraction (HFimpEF). These subgroups represent a unique population where the benefit of BB therapy remains uncertain and may differ from those with chronic HFrEF. Such observations further underscore the need for dedicated trials to clarify the role of BB in these specific patient subsets and determine whether current recommendations are applicable to post-MI and HFimpEF patients within the HFmrEF spectrum. These guidelines do not recommend the BB use in patients with HFpEF and several registries and meta-analysis described conflicting results ranging from harm, particularly increasing the hospitalization risk or exercise limitation due to chronotropic incompetence, to benefit.8 Modern registries and clinical trials describe that BB are frequently prescribed in this prevalent group of patients with HF probably based on the presence of comorbidities, for example hypertension, with some degree of recommendation for BB use.9 Peikert et al.10 have recently published a subgroup analysis from the DELIVER trial on the contemporary use and implications of BB in patients with HFmrEF and HFpEF. The vast majority of the patients included in the trial (four out of five participants) were treated with BB probably based on a high prevalence of potential non-HF indications; 89% hypertension, 57% atrial fibrillation, 53% ischaemic heart disease or 19% previous LVEF >40% that has since improved. The described benefit of BB treatment on the primary trial outcome (composite of cardiovascular death or worsening HF), as well as on the individual outcomes need to be taken with some caution due to the intrinsic limitation in this type of analysis despite an adequate statistical approach mainly due to the reduced number of patients not treated with BB. A key limitation of the DELIVER analysis is the potential for confounding by indication, as patients who tolerate more guideline-directed medical therapy, including BB, are often intrinsically healthier. This may partially explain the observed associations with improved outcomes, making it difficult to discern a true treatment effect from the inherent selection bias present in observational studies. The patient characteristics included in the DELIVER trial may influence these results, particularly the younger age and the high prevalence of comorbidities with possible BB benefit compared with other meta-analyses and registries, which fail to observe a positive influence on the outcomes. Uijl et al.3 published a cluster model from 6909 HFpEF patients from the Swedish HF Registry (SwedeHF) with an external validation in 2153 patients from the Chronic Heart Failure ESC-guideline based Cardiology practice Quality project (CHECK-HF) registry. Five distinct clusters were identified with different clinical characteristics, HF drug treatment and prognosis. They described an association between BB treatment and a reduced risk of all-cause mortality in the younger patients and the cluster defined as atrial fibrillation and hypertensive without an association with the HF hospitalization risk. In the other three clusters (elderly-atrial fibrillation, obese-diabetic and cardio-renal) the BB use was not associated with any signal of benefit. The possible worsening BB effects on chronotropic incompetence, insulin resistance and response to hypoglycaemia, and congestive patients with the cardio-renal syndrome may help to explain these discrepancies. Nevertheless, a recent published systematic review and meta-analysis of randomized or observational cohort studies on the outcome effects of BB therapy in HFpEF suggested an association with a reduced all-cause mortality without any effect on hospitalization for worsening HF. The magnitude of the all-cause mortality reduction in patients older than 75 years of age was the same as that observed in the overall group.11 In this issue of the Journal, Matsumoto et al.12 performed an observational analysis of the association between BB use and clinical outcomes in patients with HFmrEF and HFpEF. They pooled individual patient data from four trials; two historical trials (I-Preserve and TOPCAT) and other two contemporary (PARAGON-HF and DELIVER). The authors conclude with a cautious sentence: ‘there was no suggestion that BB use was associated with worse HF outcomes in HFmrEF/HFpEF’ since the results suggest a significant association with a reduced risk of cardiovascular death or hospitalization for worsening HF across the range of LVEF studied. Patients with atrial fibrillation appeared to have a better outcome with BB treatment compared to those not in atrial fibrillation. The investigators are to be congratulated for performing a high-quality methodologic assessment on the role of BB therapy in the outcomes of patients with HFmrEF and HFpEF. These results extend and are almost the same as those recently published from the DELIVER trial.10 BB use was not associated with adverse HF outcomes and mortality with a relative risk reduction of the primary outcome identical in both analyses and absence of a benefit interaction across the LVEF. Taken together with the previous analysis, the present results may reinforce the week guideline recommendation for BB use in HFmrEF (IIbC), although the positive and significant interaction with atrial fibrillation, contrary to what was described in other publications, limits the clinical implication of the finding. So, we suggest that the guideline recommendation for BB use in HFmrEF should be maintained and encourage the need of dedicated clinical trials in this group of patients. Some of the BB benefit observed in this group of patients, as well as in HFpEF, may be driven by the patients with improved ejection fraction, another issue which needs to be clarified in this kind of analysis, particularly in patients with HFpEF. Recent survey results suggest that a great majority of physicians use BB in patients with HFpEF despite the absence of evidence-based indications, probably based on patient's comorbidities.13 As previously commented, the outcome effects of BB in HFpEF remain inconsistent. The results presented here describe an association of BB use and better outcome (cardiovascular death and HF hospitalization) in HFpEF, without any significant interaction according to the presence of atrial fibrillation or other comorbidities. However, beyond traditional clinical outcomes such as cardiovascular mortality and hospitalization, the real question remains whether BB impact patient-centred outcomes, including physical and mental health. These aspects are particularly important in HFpEF, where improving quality of life and functional status is often more relevant to patients than survival alone. Future studies should prioritize these outcomes to provide a more comprehensive understanding of the benefits and limitations of BB therapy in this population. We mentioned that the cohort patient's characteristics, particularly younger age and a high prevalence of atrial fibrillation and hypertension, may influence the results despite the absence of a significant interaction. From our point of view, these new results may contribute to support the BB safety in HFpEF patients with comorbidities with a clear indication for these drugs despite recommendations for some of these clinical conditions have weakened over time and the need of large scale prospective randomized controlled trials. Conflict of interest: none declared.