Holding Out Hope for HF Shunts, Even After an FDA Panel’s Doubts

Late last year, a US Food and Drug Administration advisory panel resoundingly declined to endorse the first interatrial shunt system for patients with heart failure (HF). The decision was based on a neutral trial, albeit one with positive subgroup findings.

Recently, several experts who spoke with TCTMD said that, despite the regulatory setback, they still hold out hope that shunts have a future.

The Circulatory System Devices Panel, in December 2025, unanimously agreed that the benefits of the shunt in question—the hourglass-shaped Ventura (V-Wave, a Johnson & Johnson company), a permanent implant—didn’t outweigh the risks. Specifically, panel members weren’t convinced the device was effective when used in heart failure with reduced ejection fraction (HFrEF) patients who remained symptomatic despite use of guideline-directed medical therapy (GDMT).

Gregg Stone, MD (Icahn School of Medicine at Mount Sinai, New York), who led the RELIEVE-HF trial testing the Ventura shunt and who presented the findings, as well as a host of secondary analyses, to the FDA advisory panel, said the vote was disappointing but not altogether surprising.

“Speaking only for myself, I believe that there was enough data in patients with heart failure and reduced ejection fraction to warrant having it available to physicians to treat high-risk patients with symptomatic heart failure despite being on four to six medications,” he told TCTMD. “The data were very strong in that cohort. On the other hand, the overall trial did miss its primary endpoint.”

In RELIEVE-HF, implanting the shunt in a broad range of patients with HF did not significantly reduce the risk of the study’s primary effectiveness endpoint—a composite of all-cause mortality, left ventricular assist device/heart transplantation, HF hospitalizations, outpatient worsening HF events, and change in Kansas City Cardiomyopathy Questionnaire score—when compared with a sham procedure. The trial was surprising in that there was a benefit in patients with HFrEF but a signal of harm in those who had HF with preserved ejection fraction (HFpEF).  

The overall trial did miss its primary endpoint. Gregg Stone

The FDA panel, which was evaluating a potential approval only in patients with HFrEF, wasn’t convinced by the subgroup analysis and declined to endorse the shunt.

“I think that the FDA has been pretty clear in the past that when you have a neutral primary endpoint, the secondary analyses are hypothesis-generating,” Ryan Tedford, MD (Medical University of South Carolina, Charleston), an HF physician involved in shunt research, told TCTMD. “[The subgroup] certainly can form the justification or basis for another study but would not be sufficient for an approval of the device.”

Sheldon Litwin, MD (Medical University of South Carolina), who participated in RELIEVE-HF, as well as the REDUCE LAP-HF II study of a different interatrial shunt (Corvia Medical), said the FDA decision makes sense. RELIEVE-HF randomized 508 patients to treatment, but the benefit was observed only in a subgroup of 206 patients with HFrEF.

“My feeling was that we should confirm that,” Litwin told TCTMD, “so I’m not surprised by the FDA panel’s decision. I personally think that a follow-up study is the right thing to do.”

William Gray, MD (Main Line Health/Lankenau Institute for Medical Research, Wynnewood, PA), said he also would have been taken aback if the FDA panel recommended approval since the trial, even though it had been designated as a breakthrough device by the agency, was relatively small overall and smaller still when stratified by subgroups.

“Only about 100 patients received the device [in the HFrEF subgroup analysis],” he said. That’s a small number of people to represent a large, potentially addressable patient group and to predicate an interventional procedure that may carry risk, he added.  

Complexity of HF Phenotypes

Even with two neutral trials for two different shunts, and the FDA vote in December, the field remains highly competitive, with multiple companies developing devices and/or procedures—some that leave a permanent implant and some that don’t—in a range of sizes. However, as sources pointed out to TCTMD, interatrial shunting for HF is a complex area of medicine, with multiple variables—patient selection, shunt size, and type of shunt, among others—that factor into a successful therapy for appropriate patients.

The randomized, sham-controlled studies to date, one showing a potential benefit in HFrEF patients, the other showing a benefit in a subset of patients with HFpEF, highlight the challenges researchers and physicians face as they try to home in on who will benefit from this invasive technology. HFrEF and HFpEF are heterogeneous even within themselves, with HFrEF characterized by the underlying cause, structural changes, and disease severity. There are also three distinct phenotypes in HFpEF, each driven by underlying comorbidities such as hypertension, aging, obesity, pulmonary hypertension, and coronary artery disease.

For Gregg Fonarow, MD (University of California, Los Angeles), patient selection will be key to moving the field forward.

“There is heterogeneity in hemodynamic profiles and not all will benefit from left‑to‑right decompression,” he told TCTMD. “Future studies may need to focus on patients with documented exercise‑induced elevation in left atrial pressure, HFpEF subgroups with clear congestion phenotypes, and patients with recurrent hospitalizations despite optimized therapy.”

I personally think that a follow-up study is the right thing to do. Sheldon Litwin

Stone believes that interatrial shunts, which are designed to decrease left atrial (LA) pressure and improve HF symptoms, might be most effective in HFrEF patients who are highly symptomatic despite taking maximally tolerated GDMT and/or device therapy.

“Those patients have a very poor prognosis and really need something to help them,” he said. “I do believe that atrial shunting may be that therapy. In the RELIEVE trial, more than 95% of those patients were in New York Heart Association class III, so really symptomatic walking across a room or starting to climb a flight of stairs. The benefit that those patients achieved—the stratified subgroup of HFrEF—was profound.”

If proven successful in responder-style HFrEF studies of those with class II or III symptoms, “that would be a boon for a very large population of patients that may need additional therapy not otherwise available,” added Gray.  

During the FDA advisory panel meeting, William Abraham, MD (The Ohio State University, Columbus, OH), who is the chief medical officer of V-Wave, said that lowering LA pressure in HFrEF is difficult to achieve with medical therapy and that there are no alternative treatments to improve outcomes in NYHA class III HFrEF patients optimized on medication.

Joanne Lindefeld, MD (Vanderbilt University Medical Center, Nashville, TN), who spoke on behalf of V-Wave during the FDA presentation, said that the number of patients with HF is expected to grow to 8.5 million by 2030, noting that HFrEF patients have a poor prognosis even with the introduction of several new drug classes. In the EMPEROR-Reduced trial with the sodium-glucose cotransporter 2 (SGLT2) inhibitor empagliflozin (Jardiance; Boehringer Ingelheim/Eli Lilly), nearly half of patients were hospitalized for HF or died from cardiovascular causes within 2 years despite being on GDMT.   

“The mortality rate has markedly improved with the four pillars of GDMT, but the survival probability is still unacceptable,” Lindenfeld said at the FDA meeting.

To TCTMD, Tedford said that while the benefit observed in HFrEF patients in RELIEVE-HF might have been the result of chance, there is biological plausibility behind the finding. In early January, investigators published an analysis showing reductions in LV end-diastolic volume and LV end-systolic volumes among the HFrEF patients who received the shunt.

“The HFrEF field is potentially a little bit crowded,” said Tedford. “We have a lot of medical therapies. We have a lot of devices that are available now and a lot that are currently being studied. That said, morbidity and mortality remains high and we can always use additional options, but it’s really going to depend on efficacy. In the HFrEF studies, if there’s a benefit [of interatrial shunting] on top of guideline-directed medical therapy, I think there is a place for it.” 

What Type of HF Patient Will Benefit?

Beyond the HFrEF population, some suspect there are at least some HFpEF patients who will also do well with a shunt.

In REDUCE LAP-HF II, the shunt device did not reduce the overall rate of HF events or improve health status in 1,072 patients with HFpEF. However, there was a benefit in a prespecified subgroup of patients who did not have evidence of pulmonary vascular disease (PVD) on an invasive exercise test, a group dubbed “responders.” Based on that finding, the company launched RESPONDER-HF, a 750-patient randomized trial of patients with HFpEF or mildly reduced EF (HFmrEF) that will exclude participants with hemodynamic evidence of PVD and cardiac rhythm devices. 

Litwin believes that in patients with HFpEF, careful selection based on hemodynamic measurements is needed in order to optimize the benefits and minimize the risks of creating a shunt. 

“I personally think the phenotype is the patient who has a relatively low LV filling pressure at rest, but a very high filling pressure during exercise,” he said. “What we want the shunt to do is to shunt only when people need it, which is when they’re walking. You want to have very little shunting the rest of the time because you don’t want to induce right-heart overload due to chronic left-to-right shunting.”

Morbidity and mortality remains high and we can always use additional options, but it’s really going to depend on efficacy. Ryan Tedford

Right now, Tedford said, the biggest unmet need is for patients with HFpEF or HFmrEF. He is national principal investigator of the randomized, sham-controlled ALT-FLOW II study with the Apture shunt (Edwards Lifesciences), which creates a channel between the left atrium and coronary sinus. The idea with this approach is that it allows for natural physiological flow from the left atrium to the coronary sinus and then into the right atrium, sparing the interatrial septum.

The sham-controlled ALLAY-HF trial is also testing interatrial shunting (ALV1 System; Alleviant) in approximately 500 patients with HFpEF/HFmrEF. Unlike V-Wave and Corvia’s shunt, the Alleviant system excises tissue in the septum and leaves no device behind.

Both Litwin and Gray said they have treated several patients as part of the prior, nonrandomized phases of trials who have benefited from shunting, with the doctors even removing some medications.

What Type of Safety Data Is Needed?

During the FDA panel debate over the approval of the V-Wave shunt, committee members were mostly convinced of its safety—the vote was 9 to 6 that the RELIEVE-HF trial demonstrated reasonable safety—if approved in patients who met the criteria for the proposed indication.

To TCTMD, Fonarow pointed out that there are still unresolved safety concerns, despite the short-term safety being considered acceptable by a “slim majority.” The biggest concern amongst the FDA advisory committee members was the balance between unloading the left atrium versus the burden to the right heart.

“There is the need for studies with longer follow‑up to assess remodeling and right‑sided effects that may develop over time,” said Fonarow.

An approval would mark the first for a therapy to reduce left-sided filling pressures by redistributing blood volume to the right heart, said Stone. Like the FDA experts, Stone said a key determinant of safety is whether the right heart can accommodate the increased blood volume. In the trials published to date, including both the single-arm studies and randomized, sham-controlled trials, experts say there does not appear to be an adverse impact on right-heart remodeling/dysfunction.

We have had some initial hints at benefit and hopefully can refine that application to fit populations we understand to benefit the most. William Gray

For example, in the responder patients enrolled in REDUCE LAP-HF II, the data suggest there are increases in right atrial and ventricle volumes out to about 6 months, but RV function then remains stable out to 3 years, said Litwin. In studies with the Apture device, there were no adverse effects on the right heart out to 2 years, said Gray. The 3-year clinical and echo results from ALT-FLOW I will be presented at the upcoming THT meeting in Boston, MA. 

“What we don’t know yet is what happens over 5-year or 10-year follow-up,” said Stone. “It’s possible, over a progressive length of time, that patients might develop long-term pulmonary vascular changes and reduction in right ventricular function.”

Litwin pointed out that patients were approximately 73 years old in the two existing randomized, sham-controlled trials and that life expectancy in the US is around 78 or 79 years. “If it’s still good at 5 years, I think we’re probably out of the woods,” he said. “I think a lot of patients would probably feel the same way. You get 5 years of symptomatic reduction or maybe being symptom-free—they’d probably make that trade-off.”    

There are already second-generation shunts in the works, such as those equipped with pressure monitoring, said Stone. These devices would theoretically respond to changes in right- and left-side hemodynamics over time, with the shunt lumen adjusting accordingly, he said.

Also looking to the future, Litwin said there has been some question about whether weight loss might obviate the need for interatrial shunting in patients with HFpEF. When people lose weight, said Litwin, right-atrial pressure declines more than LA pressure, at least based on the limited data available, which would reestablish the ability of the heart to shunt flow from left to right. In the SUMMIT trial testing tirzepatide (Eli Lilly) in obesity-related HFpEF, body mass index declined from 38 to 33 kg/m².

“Their heart failure didn’t go away, it just got better,” said Litwin. “I think what you could do is take patients and make them better candidates for a shunt,” he said. If patients remain symptomatic despite being optimized on a glucagon-like peptide-1 (GLP-1) receptor agonist, he added, “then a shunt would be safer and may work even better.”

Challenges Going Forward

Gray said that while the HF device field is robust, one of the remaining challenges is that there are multiple sponsors running different trials, each a little bit different and without much standardization in terms of shunt size, location of shunt, and phenotypic presentation. However, refinement of treatment and dialing in on the ideal population are part of the natural evolution of medicine, as was the case with cardiac resynchronization therapy (CRT), which took time to find an optimized population.

“We have had some initial hints at benefit and hopefully can refine that application to fit populations we understand to benefit the most,” said Gray.

There is the need for studies with longer follow‑up to assess remodeling and right‑sided effects that may develop over time. Gregg Fonarow

During the FDA discussions, committee member Clyde Yancy, MD (Northwestern Feinberg School of Medicine, Chicago, IL), stressed that there are a number of approved therapies and devices that alter the natural history of HF and that such interventions have cleared a high evidence bar.

“It’s always been driven by very high-quality, indisputable evidence,” he said. “That’s the standard that should continue. I am one that is still very clinically active, and I understand the necessity to instill hope in our patients, but it can’t be hope on a feigned basis. It has to be legitimate hope.”

While Yancy voted against approving the V-Wave shunt, he didn’t seem to see it as the end of the road. “This is still ‘version early,’” he said. “We need to continue to pursue this possibility. There may be benefit, and if we continue to work diligently like we have in heart failure for decades, we may be able to answer this question with more precision.”

Fonarow, for his part, told TCTMD that disruptive HF devices, including CRT, went through early periods of skepticism, device iteration, and trial redesign before being approved and recommended in the guidelines.

“Interatrial shunting is still in an early phase,” he said.