For the HF patient with a gain in LVEF, there’s little guidance on how best to care for them or manage their GDMT.
Patients with newly diagnosed heart failure with reduced ejection fraction (HFrEF) commonly have improvements in their LVEF, but that doesn’t mean they can let their guard down, data from a large integrated healthcare delivery system show.
This group—deemed to have heart failure with improved ejection fraction (HFimpEF)—had significantly lower rates of worsening HF events and all-cause death compared with patients with persistent HFrEF, but the remaining risk was still substantial, researchers led by Kyung Min, MD (Kaiser Permanente San Francisco Medical Center, CA), report in a study published online ahead of the August 5, 2025, issue of JACC.
“We’ve seen from our data that improvement in left ventricular ejection fraction does not mean complete resolution of clinical risk,” Min told TCTMD. “These patients are still at high risk for coming back into the emergency department with worsening heart failure episodes.”
Unfortunately, there are limited data to inform how these patients should be managed. The latest US heart failure guidelines from 2022, which introduced HFimpEF into the classification of heart failure, contain a class 1 recommendation to continue guideline-directed medical therapy (GDMT) to prevent relapse and LV dysfunction, even if symptoms resolve. But that advice is based largely on the 51-patient TRED-HF trial, which included select patients with dilated cardiomyopathy who had their symptoms and HF improve. Treatment withdrawal was associated with high rates of relapse.
A major takeaway from the current study is that “heart failure with improved ejection fraction is a very common clinical entity, yet we have very little data about therapeutic escalation and therapeutic de-escalation in select patients,” senior author Ankeet Bhatt, MD (Kaiser Permanente San Francisco Medical Center and Kaiser Permanente Northern California, Pleasanton), told TCTMD. “It elucidated that data gap for us.”
These patients are still at high risk for coming back into the emergency department with worsening heart failure episodes. Kyung Min
To gain insights into patients with HFimpEF and how they’re treated in practice, Min, Bhatt, and colleagues examined data on 28,292 patients (mean age 69.4 years; 34.8% women) who were diagnosed with HFrEF (mean LVEF 31.1%) within the Kaiser Permanente Northern California system between 2013 and 2022. Within 1 year, 30.6% of patients moved into the HFimpEF category after having a follow-up EF above 40% and at least a 10% absolute improvement in LVEF. The mean gain in EF was 23%.
Patients who improved tended to be younger compared with the overall HFrEF cohort, with a mean age of 66. They had higher rates of atrial fibrillation/flutter but lower rates of other comorbidities and a smaller ventricular size compared with patients with persistent HFrEF at 1 year.
A year after the initial HFrEF diagnosis, the rate of a composite of worsening HF events and all-cause death was lower in the HFimpEF group than in the patients with persistent HFrEF (unadjusted 23.1 vs 45.0 per 100 person-years), with similar differences observed in the individual components of that endpoint and in all-cause hospitalization. The better outcomes in patients with HFimpEF remained after adjustment.
At the time of discharge following the initial HFrEF hospitalization, patients who would eventually move into the HFimpEF category received more than 50% of the maximum guideline doses of GDMT at rates similar to those who would have persistent HFrEF. Among patients in the HFimpEF group at 1 year, use of beta-blockers, ACE inhibitors/ARBs, and mineralocorticoid receptor antagonists “marginally decreased,” with small increases in use of angiotensin receptor-neprilysin inhibitors and sodium-glucose cotransporter 2 (SGLT2) inhibitors.
The researchers found that withdrawal of GDMT among patients with HFimpEF was “modestly associated” with greater risks of worsening HF events/death, all-cause hospitalization, and all-cause death, although only the difference in hospitalizations reached statistical significance.
“Whether certain phenotypes of HFimpEF are amenable to safe withdrawal of pharmacotherapy remains an understudied topic,” Min et al note.
Exploring the Heterogeneity of HFimpEF
Looking forward, Min indicated that it would be helpful to delve into the heterogeneous HFimpEF population to explore the likelihood of seeing improvements in ejection fraction across various etiologies and potential differences in outcomes.
Bhatt agreed, saying that “deciphering the heterogeneity of this clinical condition is where we need to go in terms of risk stratification, response to therapies, and then hopefully individualized implementation approaches.”
There’s a need for better ways to sort through the different subsets within HFimpEF, which may then lay the groundwork for additional trials, Bhatt indicated. “Given new tools and digital technologies, we may have the ability to better risk-stratify heart failure with improved EF and design larger, powered, clinical practice-changing trials in the near future.”
There seems to be something that’s unmeasured or uncaptured that would identify the people who go on to experience improvement, or remission, if you will. Andrew Perry
Andrew Perry, MD, who wrote an accompanying editorial with Lynne Stevenson, MD (both from Vanderbilt University Medical Center, Nashville, TN), told TCTMD an important component of this new study is that it affirms how common HFimpEF is in practice.
It also highlights that improvements in use of GDMT are not “the end of the story” when it comes to gains in cardiac function observed in the patients with HFimpEF, he said. Rates of medication use were only minimally different between the HFimpEF and persistent HFrEF.
“There seems to be something that’s unmeasured or uncaptured that would identify the people who go on to experience improvement, or remission, if you will,” Perry said. He cited differences in indicators of overall health and wellness, genetic predisposition, and HF etiology as possible contributors.
As for how to manage GDMT in patients with HFimpEF, Perry indicated that there is some uncertainty, pointing out that even though TRED-HF demonstrated that withdrawal of medications worsened clinical outcomes overall, there were trial participants who had drugs withdrawn without having poorer outcomes. “But how do you know who those people are up front?” he said.
On the other hand, there is some evidence—from an analysis of the DELIVER trial, for example—that additional GDMT can further improve outcomes even in patients who’ve already seen their LVEF improve on other drugs, Perry pointed out.
“I would be personally hesitant about withdrawing medications in the absence of a clearly reversible cause for their heart failure and, generally speaking, I would at least maintain [ongoing GDMT] and I might even consider increasing those doses,” he said.
In their editorial, Perry and Stevenson say that “the journey to improvement will hopefully include new lanes as GDMT continues to evolve beyond traditional triple therapy.
“It remains unlikely, however, that contemporary medications in any doses will be sufficient to achieve and maintain HFimpEF across the heterogeneous HF population,” they continue. “The imminent era of broad population data will hopefully guide integration of primary and secondary causes, therapeutic responses, and multidimensional surveillance beyond EF, in order to personalize regimens and improve prognosis along all journeys with HF.”