Poor Outcomes in HF Patients With Moderate Aortic Stenosis: Observational Data

(UPDATED) For patients with left ventricular systolic dysfunction, even moderate aortic stenosis (AS) is associated with significantly poorer long-term outcomes, including a higher risk of hospitalization for heart failure and mortality, new observational data show.

The study, published March 27, 2023, in the Journal of the American College of Cardiology, adds yet more evidence that moderate AS adversely affects clinical outcomes, raising questions about whether earlier aortic valve replacement could improve patient prognosis, according to co-lead investigators Kathleen R. Khan, MD, and Omar A. Khan, MD (Massachusetts General Hospital, Boston), and colleagues.

These results are particularly timely given that three ongoing clinical trials, including one that’s completed enrollment, are testing exactly that premise.

“Patients with severe AS and HFrEF [heart failure with reduced ejection fraction] represent a high-risk population that benefit greatly from AVR,” write Khan and colleagues. “Whether early AVR for moderate AS in the setting of HFrEF similarly interrupts maladaptive cardiac remodeling and improves clinical outcomes is of critical clinical importance and the focus of the ongoing TAVR UNLOAD clinical trial. The PROGRESS trial and the Evolut EXPAND TAVR II Pivotal Trial will also provide prospective data on the moderate AS population.”

Right now, valve replacement, including TAVI, is reserved solely for patients with symptomatic severe AS.

Senior investigator Sammy Elmariah, MD (University of California, San Francisco), said that after the expansion of TAVI from high- to low-risk patients with symptomatic severe AS, researchers began to think of populations where early treatment might make sense, one of those being patients with moderate AS and HFrEF.

“Of course, we all know that the underpinning of heart failure management in patients with reduced ejection universally depends on afterload reduction,” Elmariah told TCTMD. “It makes logical sense that patients with increased afterload related to moderate aortic stenosis may benefit from earlier intervention.”  

Philippe Généreux, MD (Atlantic Health System/Morristown Medical Center, NJ), one of the principal investigators of the ongoing PROGRESS trial, said the new findings aren’t surprising, noting there are accumulating data showing the negative impact of even moderate stenosis.

“People will often say, ‘It’s only moderate AS,’ but moderate AS is bad,” Généreux told TCTMD. “When you look at the valves, they’re very calcified, they’re restricted. It puts an extra load, more pressure on the left ventricle. Even mild aortic stenosis can be seen as bad especially since it progresses so fast. To have mild or moderate AS is not the same as having no aortic stenosis at all. I think we need to reevaluate the spectrum of AS severity because we’ve been very complacent in the past.”

In an editorial accompanying the study, Paul Sorajja, MD (Minnesota Heart Institute Foundation, MN), João Cavalcante, MD (Minneapolis Heart Institute at Abbott Northwestern Hospital, MN), and Josep Rodés-Cabau, MD (Quebec Heart and Lung Institute/Laval University, Canada), write that the ongoing randomized trials “are essential to understanding the optimal management of moderate aortic stenosis for our patients with heart failure. We have clear evidence of an association with impaired survival, but a causative role for moderate aortic stenosis is currently lacking.” 

More Days Alive Out of Hospital

The observational study included 9,133 patients with HFrEF treated at Massachusetts General Hospital between 2016 and 2019. Of these, 374 and 362 patients had moderate and severe AS, defined as an aortic valve area < 1.0 and 1.0-1.5 cm², respectively. The patients with AS were older than those without AS and had more comorbidities, including more diabetes, atrial fibrillation, CAD, chronic kidney disease, and concomitant aortic, mitral, and tricuspid regurgitation.

After a median follow-up of 3.1 years, rates of the primary composite endpoint—hospitalization for HF and all-cause mortality—were 62.0% and 62.7% in patients with moderate and severe AS (P = 0.68), significantly higher than in those without AS (49.5%; P < 0.0001). All-cause mortality rates were 42.2% and 45.9% in those with moderate and severe AS, again significantly higher than that seen in those without AS (25.8%; P < 0.0001).

In a propensity-matched analysis, patients with moderate AS had a higher risk of the combined endpoint of HF hospitalization/all-cause mortality (64.4% vs 59.9%; P < 0.05), as well as a higher risk of HF hospitalizations and all-cause mortality alone, when compared with those without AS. In a multivariate analysis, moderate AS was an independent predictor of HF hospitalization/all-cause death (HR 1.24; 95% CI 1.04-1.49).

Patients with moderate AS spent a median of 850.5 days alive out of hospital, which was similar to those with severe stenosis but significantly less than those without AS (732.5 and 1,162.0 days, respectively).

During the study period, just 12.2% of patients with moderate AS had their aortic valve replaced, either with surgery or TAVI. In contrast, 33.7% of those with severe AS underwent AVR. In multivariate analysis, AVR was associated with a 40% lower risk of all-cause mortality but was not associated with a reduction in the composite endpoint of HF hospitalization/all-cause mortality or HF hospitalization alone.  

In their editorial, Sorajja, Cavalcante, and Rodés-Cabau say the big question for symptomatic moderate AS is its contribution to ventricular afterload and whether relief can be provided with valve replacement. They point out that that little can be said about the impact of AVR on clinical outcomes in the present retrospective series because so few patients with moderate AS were treated. Of those who were, 80% had surgery because of progression of AS to severe during follow-up as opposed to treatment at the time of initial AS diagnosis.

In some retrospective series, the risk of mortality with moderate AS is akin to that seen with severe AS, said Généreux, although he noted it can be difficult to assess the true severity of disease. “Some people with moderate AS have the phenotype where it behaves like severe AS, whereas some moderate AS [cases] are really more mild or moderate,” he said. Given that, retrospective studies need to be interpreted with caution, he said. “We need to be careful with our assumptions.”

Nicolas Van Mieghem, MD, PhD (Erasmus Medical Center, Rotterdam, Germany), one of the TAVR UNLOAD investigators, said the results of the study were timely given the widespread interest in “upstream management” of AS. He pointed out that moderate and severe AS trigger different behaviors with physicians, with watchful waiting the only available option for patients with moderate AS.

With clinical surveillance, patients are monitored every 6 or 12 months with imaging and a range of tests. One of the limitations of the new analysis is that moderate and severe AS were diagnosed with echocardiography, which is no longer the standard, said Van Mieghem. Today, diagnosis and classification of AS severity, as well as ongoing surveillance, involves multiple tests including echocardiography, dobutamine stress testing, calcium scoring with computed tomography, and biomarker assessments, such as B-type natriuretic peptide (BNP), he said.   

Aortic Stenosis in HF Patients

The 300-patient TAVR UNLOAD trial is focusing specifically on patients with moderate AS and advanced HF. The Edwards Lifesciences-sponsored PROGRESS study is testing whether TAVR is superior to clinical surveillance plus optimized guideline-directed medical therapy (GDMT) in patients with moderate AS, while EXPAND TAVR II trial is a similar trial conducted by Medtronic. The two latter studies include a mix of patients with and without left ventricular dysfunction.

“The collectivity of these trials will bring meaningful insights into the value of early intervention in moderate AS in patients with signs of cardiac damage or symptoms of heart failure,” said Généreux. “That’s what will really show if early intervention decreases mortality long-term, decreases hospitalizations, improves quality of life, and whether or not it’s a cost-effective strategy.”

To TCTMD, Van Mieghem said he suspects AVR to reduce the fixed afterload on the heart would be most beneficial in moderate AS patients who have depressed LV function and symptoms, which includes the patients studied in the new observational study and is the premise of TAVR UNLOAD.

The other studies—PROGRESS and EXPAND TAVR II—include patients with moderate AS and evidence of “cardiac damage/dysfunction,” a concept that removes ejection fraction from the equation. These patients might have mitral or tricuspid regurgitation, RV dysfunction, or even AF, said Van Mieghem, but not necessarily reduced EF. There is less observational data to support early treatment of moderate AS in that setting, although time will tell with the results of these two trials, he noted.  

Généreux pointed out that patients with AS and HFrEF already have two strikes against them, noting that adverse events are frequently driven by the LV dysfunction. “Whether the poor ejection fraction is due to the aortic stenosis or not, we know that these patients will be back at the hospital and are more likely to die,” he said. A big unknown is whether early intervention to treat moderate AS might already be too late in the patient with advanced HF, a question that will be answered with TAVR UNLOAD.

“On the opposite end, in what we’d call a more preventive approach with TAVI, they have a normal EF and moderate AS and if you intervene early enough, you may save the progression to cardiac damage,” said Généreux. “For me, the very interesting question is whether intervening before the ejection fraction is low will prevent events in the future, say in 5 or 10 years.”

For Elmariah, their new results suggest that the worse clinical outcomes and higher mortality in patients with HFrEF are attributable to the moderate AS, given that the comparator arm included patients with HFrEF.

“This is beyond the ejection fraction being down,” he said. “It suggests a causal nature of aortic stenosis in worsening that patient’s outcomes.” The hope is that if physicians can address moderate AS by performing TAVR, the patient’s prognosis could be improved to where it’s on par with those with HFrEF but without the added burden of AS, he added.

Elmariah added that while patients included in TAVR UNLOAD, PROGRESS, and TAVR EXPAND II are quite diverse, the goal in all of these trials is to identify a population where their risk might be modified by treating moderate AS. Given that the valve can be replaced with such low risk, it makes sense to explore any populations who might benefit from early intervention.

“The question then becomes what are the clear-cut risk factors that identify a patient who would benefit from that early intervention,” he said. “I think the low ejection is number one on the list. That’s a patient population that clearly has cardiac injury and is in need of help.”   

Sorajja, Cavalcante, and Rodés-Cabau—all of whom are involved in the ongoing TAVI trials—stress say that while it’s plausible that relieving the increased load from moderate AS could be clinically significant regardless of EF, “the outcomes from the randomized trials are needed to provide the necessary insights into the management strategies for these patients.”