Is the recommended therapy for heart failure with mid-range ejection fraction adequate?
Beta-blockers for heart failure with reduced, mid-range, and preserved ejection fraction: an individual patient-level analysis of double-blind randomized trialsCleland JGF, Bunting KV, Flather MD, et al. - Eur Heart J 2018; 39: 26–35
ESC guidelines introduced the mid-range ejection fraction type of HF (HFmrEF), defined as HF with EF 40–49%, which should be treated like HFpEF, hence, without beta-blockers . Beta-blockers increase left ventricular ejection fraction (LVEF) and reduce morbidity and mortality for a broad range of patients with reduced LVEF in sinus rhythm. However, it is not known whether HFmrEF patients benefit from beta-blockers [2,3].
In this meta-analysis, the effect of beta-blockers on LVEF and prognosis was investigatedf, stratified according to the baseline LVEF and heart rhythm, in double-blind, randomized, placebo-controlled trials that recruited >300 patients, with a planned follow-up of >6 months and explicit reporting of mortality.
11 trials were eligible for this purpose, including 14,262 patients in sinus rhythm and 3050 AF patients, with a baseline LVEF measurement and a baseline ECG. The primary outcomes for this analysis were all-cause mortality and CV death.
- The mean follow-up was 1.5 years (SD: 1.1), and the median LVEF at baseline was 27% (21–33%). 721 patients had LVEF 40–49% and 317 had LVEF ≥50%.
- Baseline LVEF was inversely associated with all-cause mortality (adjHR for each 5% lower LVEF: 1.16; 95%CI: 1.26–1.19; P<0.0001) with a stronger association for patients in sinus rhythm than in AF.
- Beta-blockers were associated with reductions in all-cause and CV mortality compared to placebo for patients in sinus rhythm (interaction P>0.5 for LVEF as a continuous measure), except in the subgroup of patients with LVEF ≥50%.
- In the subgroup of LVEF 40-49% for patients in sinus rhythm, all-cause mortality occurred in 7.2% of patients randomized to beta-blockers compared to 12.4% of patients assigned to placebo (adjHR: 0.59; 95%CI: 0.34–1.03), and CV death occurred in 4.5% of patients on beta-blockers and in 9.2% of patients in the placebo group (adjHR: 0.48; 95% CI: 0.24–0.97); the authors conclude that there is no difference in benefit between the LVEF categories 40-49% and <40%. Patients with AF showed no clinical benefit with beta-blockers, regardless of LVEF.
- LVEF changes were measured at a follow-up visit in 4601 patients in sinus rhythm and 996 patients in AF (median 1.0 years after baseline). In sinus rhythm, LVEF increased more in patients with beta-blockers than placebo in all subgroups (+1.9% [1.1% SE] for LVEF 40-49%), except for the subgroup with LVEF ≥50%. This was also true for most LVEF categories of patients in AF (+4.8% [1.9%] for LVEF 40-49%), again not for LVEF ≥50%, nor for LVEF 35-39%.
The figure illustrates the changes in absolute mortality and LVEF in patients with sinus rhythm (beta-blockers vs. placebo), stratified by baseline LVEF.
For HF patients with sinus rhythm and LVEF <40%, beta-blockers improve left ventricular systolic function and reduce CV mortality. Based on the observations in patients with LVEF 40-49%, the authors conclude that they show similar CV mortality and change in LVEF as those with LVEF <40%. These findings suggest that HFmrEF patients might benefit from beta-blocker therapy similarly to HFrEF patients.
In their editorial article , Wilcox and Mann note that the group of patients with an LVEF of 40-49% with increased HF hospitalization and CV death created ‘…consternation, confusion, and intense academic debate…’, because they did not fit into the HFpEF and HFrEF paradigms. The authors explain simply that this group of patients, named HFmrEF, are mainly HFrEF patients whose LVEF has improved due to good clinical management and a smaller group of HFpEF patients whose condition is worsening. Therefore: ‘Given this understanding, the important observation in the report by Cleland and colleagues that beta-blockers decreased cardiovascular mortality in patients with an LVEF of 40–49% is both predictable and reassuring.’ However, based on current guidelines, a HFrEF patient on beta-blockers, whose LVEF has improved and reached 41%, would have been re-classified as a HFmrEF patient, and accordingly, the physician would have to discontinue the beta-blocker therapy. The authors conclude: ‘Hence, the assessment of LVEF is context dependent, and requires prior knowledge about the LVEF trajectory in order to understand patient prognosis fully. While it is premature to comment on whether the study by Cleland et al. will be sufficient to prompt a change in practice guidelines for the management of HFmrEF, this study does highlight the limitations of classifying HF based on LVEF alone and emphasizes the need for more detailed phenotyping in order to individualize targeted medical therapies to improve clinical outcomes in HF.’