It’s a new age for heart failure (HF) management, with growing appreciation for its “four pillars” of drug therapy, but also, it appears, the waning status of left ventricular ejection fraction (LVEF) as the foremost guide to which patients should be treated with them.
The metric, usually obtained by echocardiography, is viewed by some as a relic of an earlier age when it was important for tagging patients most likely to respond to a growing arsenal of drugs that target neurohormonal activation: those with HF with reduced ejection fraction (HFrEF), usually defined as an LVEF of 40% or less.
But the field’s shifting dynamic between LVEF cut points, disease phenotypes, and drug efficacy is blurring the boundary between HFrEF and HF with mid-range ejection fraction (HFmrEF). At the same time, some drug classes approved for HFrEF have earned new respect and, in some cases, regulatory approval for use in patients with LVEF above 40%, including those with HF with preserved ejection fraction (HFpEF).
“We would encourage you to forget about ejection fraction. We certainly would encourage you to forget about an ejection fraction of 40%,” said Milton Packer, MD, Baylor Heart and Vascular Institute, Dallas. His comment was in response the question: “Are we entering a new therapeutic era where ejection fraction doesn’t matter?”
“It’s an artificial metric,” Packer said. “It doesn’t have any physiological significance.”
There is an increasingly fuzzy line between HFrEF and HFmrEF, in that the two “may have a lot more in common” than they have differences, including in their responses to some drug therapies, agreed Javed Butler, MD, MPH, University of Mississippi, Jackson.
Both Packer and Butler spoke during a panel presentation on HFpEF at the Heart Failure Society of America (HFSA) 2021 Annual Scientific Meeting, conducted virtually and also live in Denver.
The ideas that LVEF in HF patients should be seen as a continuum more than the basis for distinct categories, and so can have diminishing value as a guide for HF therapy, and even that it should be abolished as a basis for classifying HF, have been aired for several years at least. But they were energized recently after the full unveiling of the EMPEROR-Preserved trial results in August at the European Society of Cardiology Congress 2021.
For the first time in a major HFpEF trial, patients on a test medication — in this case a sodium-glucose transporter 2 (SGLT2) inhibitor — showed a clear-cut significant benefit for the primary endpoint. The agent was empagliflozin (Jardiance, Boehringer Ingelheim/Eli Lilly), and the endpoint was cardiovascular (CV) death or HF hospitalization.
Given those results and earlier evidence suggesting that the benefits of some other HF meds transcend the traditional LVEF categories, there is debate over whether LVEF should continue as an important guide to HF therapy.
“There are high emotions on both sides, and probably, like most things in life, the reality is somewhere in the middle,” Butler told theheart.org | Medscape Cardiology in an interview.
“Part of the Nomenclature”
Randall C. Starling, MD, MPH, said he sees the HF distinctions by LVEF “as more of a convenience” when, for example, communicating with other physicians. “If I say this patient has HFrEF, then in general my colleague is going to think they have a reduced ejection fraction. But they’re not going to necessarily want to know, is it 10% or 20% or 30%.”
The term is “part of the nomenclature now,” Starling, from the Cleveland Clinic, told theheart.org | Medscape Cardiology. He comoderated a panel presentation on the universal definition of HF at the HFSA sessions.
“But I think that those of us in the heart failure field are beginning to recognize that we have to be more open-minded about this measurement. And we don’t want to necessarily restrict a patient based on a specific number, knowing its vagaries and limitations.”
Starling said he doesn’t think LVEF as a guide to HF therapy “is ever going to be eliminated, but I think more and more, we’re going to see treatments that will not be restricted based on a specific ejection fraction.”
The emergence of HF meds “that are agnostic to the ejection fraction I think is a great thing. And we may wind up seeing more patients treated.”
Speaking at the same HFpEF panel session as Packer and Butler, Clyde W. Yancy, MD, MSc, Northwestern University, Chicago, proposed that advances in imaging, for example, might sharpen understanding of ventricular physiology enough in HF to develop new therapies based more on the disease process than on a metric like LVEF.
And he joined in questioning the dogma of a 40% threshold separating HFrEF from HFmrEF. “What we’ve been calling HFrEF probably needs to be reconsidered,” Yancy said. The “real take-away” from their discussion that day “may be that we need to revisit what we call HFrEF and let that be the new standard.”
“Where HFrEF ends and where HFpEF begins may be different than we’ve been thinking about,” agreed Douglas L. Mann, MD, Washington University School of Medicine, St. Louis. For example, patients with LVEF in the 40% to 50% range can show abnormal ventricular strain on echocardiography, “which is a systolic problem,” he told theheart.org | Medscape Cardiology. That supports a definition of HFrEF with an LVEF up to perhaps 50%.
LVEF-Agnostic Quartet Therapy?
In his HFSA presentation, Butler called up results from several clinical trials that had helped reveal — for practitioners and for regulators — the therapeutic potential of some of the four pillars of HFrEF therapy in patients with an LVEF up to at least 50%.
The quartet of drug classes making up the four pillars are angiotensin-receptor neprilysin-inhibitors (ARNIs), mineralocorticoid receptor antagonists (MRAs), beta blockers, and SGLT2 inhibitors. Rising consensus and even newly minted guidelines urge expeditious initiation of a drug from all four classes soon after a new HFrEF diagnosis, although European recommendations prefer ACE inhibitors over the ARNIs as a first-line approach.
Analyses of trials like PARADIGM-HF and PARAGON-HF, which explored the ARNI sacubitril–valsartan (Entresto, Novartis) in HFrEF and HFpEF, respectively, convincingly suggest, experts have said, that the drug can improve the risk for HF hospitalizations in both groups.
“We see that the benefit is extended to patients beyond the traditional HFrEF category. Sort of until the mid-50% range, you’re seeing the benefit, and at that point, you see some attenuation of the benefit,” said Butler. That is, the higher the LVEF, the less sacubitril–valsartan seems to improve outcomes.
A similar effect had been previously seen for the MRA spironolactone in the HFpEF TOPCAT trial, he noted. That trial was negative, in that the MRA showed no benefit for a clinical primary endpoint that included CV mortality and HF hospitalization.
But in a post hoc analysis, the rate of HF hospitalization by itself fell by a modestly significant 17%, which helped convince a US Food and Drug Administration (FDA) advisory panel to favor approval of the MRA for HF beyond its existing HFrEF indication.
And in February, the agency approved an expanded indication for sacubitril–valsartan “to reduce the risk of cardiovascular death and hospitalization for heart failure in adult patients with chronic heart failure.” That wording “pretty much opened the door” to use the drug in anyone considered to have heart failure, Starling noted.
Most recently, hopes for an expeditious HFpEF approval for empagliflozin climbed on the recent announcement of the EMPEROR-Preserved trial’s milestone finding of a significant 21% drop in risk for the primary endpoint of CV death or HF hospitalization.
Based on current recommendations and clinical trial evidence, said Butler from the panel, “if you have ejection fraction less than 55%, you should use ‘quad’ therapy. And if you have a heart failure diagnosis, you should use empagliflozin irrespective of ejection fraction.”
Limitations of LVEF
Most everyone, whether favoring a rethink of the issue or not, agrees that LVEF-guided HF therapy has limitations. For example, “there is significant variability in echocardiography readings. If somebody has ejection fraction at 65%, it could actually be 58% or 72%,” Butler observed.
The number can also change over time in individual patients, especially those found with HFmrEF, which might label them at the middle of a journey from HFpEF to HFrEF as the disease progresses, or vice versa with successful therapy.
“The ejection fraction is very crude and only initiates discussion. It’s probably not sufficient to really precisely identify a phenotype and even less so to identify biology,” Yancy said.
In fact, there is a lot about the biology of HF-associated cardiomyopathy that LVEF doesn’t categorize. Neurohormonal activation is excessive in all HF, regardless of the metric, and myocyte damage, ventricular remodeling, and myocardial fibrosis can also exist across the range of LVEF.
“We have data with empagliflozin now across the entire spectrum of ejection fractions,” observed Packer. “And yes, there is attenuation on a population basis at 65% or greater, but on an individual basis, ejection fraction isn’t that reproducible. It’s a really variable biomarker.”
So given the current data, he said, “with respect to SGLT2 inhibition with empagliflozin, measurement of ejection fraction is not going to help you determine who to treat and who not to treat.”
“I think the field is going to undergo some changes,” Mann said in an interview. “I’m still a believer that LVEF gives you some information about the biology,” but some drugs, notably the SGLT2 inhibitors, are going to “erase the need” for LVEF in guiding therapy.
Continued Usefulness of LVEF
Speaking at the session on the Universal Definition of Heart Failure, Biykem Bozkurt, MD, PhD, Baylor College of Medicine, Houston, said the definition characterizes HF according to “symptoms or signs caused by a cardiac abnormality and corroborated by elevated natriuretic peptide levels or objective evidence of congestion.”
Establishing those features are the first steps for “enhancing appropriate diagnosis and optimization of GDMT [guideline-directed medical therapy], to achieve uniformity of care even in the setting of comorbidities,” she said.
Only then, characterization of LVEF as reduced, mid-range, or preserved will help “standardize and clarify the guideline-directed therapies,” said Bozkurt, who is lead author on the recent document that lays out the universal definition.
“It is critical for clinicians to classify patients according to these LVEF subgroups,” not by a single LVEF measurement alone, but in the context of other characteristics, such as ventricular mass and structural abnormalities, and the etiology of cardiomyopathy.
The universal definition, Starling interpreted, doesn’t eliminate LVEF as a component; rather, “it says if you have congestion and you have symptoms, you have heart failure. That’s the bottom line. And then the ejection fraction merely puts you into different buckets for a variety of different reasons.”
One of those considerations, at least for the quartet of drugs that target neurohormonal activation, is the attenuation of benefit with rising LVEF. And the reverse also appears true, in that the drugs are more effective as LVEF drops.
Although the FDA has indicated sacubitril–valsartan essentially regardless of LVEF, for example, the drug shows graded benefit in terms of CV death or HF hospitalization. “If you really look at bang for your buck, the lower your ejection fraction, the more the magnitude of effect,” Mann observed.
“Ejection fraction helps you to phenotype a group of people who respond more vigorously to neurohormonal antagonists,” he said. “Drugs that work directly in that pathway are more beneficial.”
But the SGLT2 inhibitors are “a whole different category,” he said. “The more data that come in on the SGLT2 inhibitors, they more they appear to be working outside of the traditional classical neurohormonal model that has really dominated new therapies for the last probably 20 to 25 years.”
How SGLT2 Inhibitors Seem Different
Benefit from SGLT2 inhibitors might still in some ways be related to LVEF, noted Bozkurt in her presentation. Both EMPEROR-Reduced and EMPEROR-Preserved overall met their composite primary endpoints. But in EMPEROR-Preserved, “there was heterogeneity for total heart failure hospitalizations, with the magnitude of benefit being significantly higher in mildly reduced LVEF” than in HF at more preserved LVEF levels. No such effect was seen for the primary endpoint.
The preferential effect on HF hospitalization at lower LVEF values, Mann proposed, might be due to the non-neurohormonal SGLT2 inhibitor’s mechanism of action. His “guess,” he said, is that the drugs “have beneficial effects on renal function, which would cut across all ejection fractions.” They seem to “make the kidneys better and help them deal with sick heart, whether it be HFpEF or HFrEF.”
That doesn’t mean LVEF isn’t helpful and shouldn’t be measured. “I don’t know that we need to throw out the baby with the bathwater [just] because the SGLT2 story has been so compelling,” Mann said. “I can tell you that the biology is different in HFrEF versus HFpEF. The former is characterized more by a remodeled, dilated ventricle; the latter by a ventricle that is more concentric and stiffer, he observed. “I do think they are different diseases.”
LVEF might also help guide the use of at least one member of the quartet, beta blockers, which “may be harmful in HFpEF” because they slow the heart rate, Mann observed. In general, “I don’t think beta blockers should necessarily be given for patients with HFpEF,” so LVEF “may help to fine-tune your management.”
There is “no specific evidence in HFpEF, however you define that, that a beta blocker will improve survival and reduce heart failure hospitalizations,” Starling said. Still, anyone with coronary disease, atrial fibrillation, or hypertension might well be prescribed beta blockers, since “they are approved for so many indications.” Most patients in any heart failure trial will be on the drugs as background therapy, he noted.
“If I were to look at a crystal ball,” Butler said in an interview, “I think that people are going to be much more vigilant and dogmatic about ejection fraction when it comes to devices. So, I don’t think that there’ll be a slippery slope for ICDs [implantable defibrillators] and biventricular pacemakers. They will still be very much tied to the traditional ejection fraction definition.”
But with respect to drug therapy, it’s becoming harder to justify the classification of mid-range LVEF, said Butler, who anticipates a potential consensus statement or expert recommendation, perhaps, for expansion of HFrEF up to an LVEF of 55% or 60%.
“People will still look differently at ejection fractions greater than 60%,” he said. But, “I think rather than three categories, it will become just two categories.”
Butler discloses consulting or serving on an advisory board for Abbott, Amgen, Applied Therapeutics, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, CVRx, G3 Pharma, Impulse Dynamics, Janssen, LivaNova, Luitpold, Medtronic, Merck, Novartis, Novo Nordisk, and Occlutech; and receiving honoraria for speaking from Novartis, AstraZeneca, Janssen, Boehringer Ingelheim, and Merck. Packer discloses consulting or serving on an advisory board for AstraZeneca, Boehringer Ingelheim, CSL Behring, Cytokinetics, Lilly, Moderna, Novartis, Salamandra, Actavis, Amgen, Bristol-Myers Squibb, Casana, and Relypsa. Bozkurt discloses consulting or serving on an advisory board for Abbott, Amgen, Vifor, Relypsa, Liva Nova, Sanofi-Aventis, scPharmaceuticals, and Respicardia. Yancy, Mann, and Starling had no disclosures.
Heart Failure Society of America (HFSA) Annual Scientific Meeting 2021. Digesting EMPEROR-Preserved: The Evolving Treatment of HFpEF and Use of SGLT2i, presented September 11, 2021. The Universal Definition of HF (Joint Session with ESC HFA and JHFS), presented September 12, 2021.