What have we learned to suggest that moderate aortic stenosis (AS) may warrant more active risk stratification rather than routine surveillance alone?

Several studies have demonstrated that patients with moderate AS have an increased risk of adverse outcomes. The problem with those studies is that we never knew whether the moderate AS was causing adverse outcomes or whether it was an innocent bystander, with the outcomes instead driven by other comorbidities (eg, high cholesterol, diabetes, kidney disease) that may have actually contributed to the development of AS in the first place. With the recently published PROGRESS trial results, we now have evidence suggesting that treating moderate AS results in improved clinical outcomes.1 This is the first bit of evidence suggesting that adverse consequences associated with moderate AS are actually caused by the moderate AS.

Which patients with moderate AS concern you the most? What imaging, clinical, or hemodynamic features might suggest to you that the AS is clinically significant?

This is a nuanced question, and we are still learning. The highest-risk patients are those who would struggle to cope with the increased afterload of even moderate AS; in particular, those with a reduced ejection fraction are likely at highest risk. Having said that, we know from PROGRESS that there are several other hemodynamic and clinical factors that may make treating moderate AS advantageous. We are now tasked with intervening on moderate AS in a timely fashion before further rapid progression, clinical decompensation, or worsened cardiac damage occurs with the goal of improving clinical outcomes and quality of life (QOL).

Again, the highest-risk group is patients with a low ejection fraction, but patients with heart failure with preserved ejection fraction (HFpEF), symptomatic patients, those with a history of heart failure (HF) events, and those with abnormal diastolic cardiac function are less tolerant of moderate AS and might benefit from intervention. Atrial fibrillation (AF), when poorly controlled, can result from the reduced diastolic function and can also be a consequence of cardiac damage. Thus, AF also identifies an at-risk population.

Within the PROGRESS trial, a disproportionately high calcium score was also considered an at-risk feature for patients with moderate AS. This does not necessarily mean that these patients are less tolerant of the moderate AS, but I do think it’s an important factor because patients with a high calcium score are more likely to progress rapidly. It’s really about intervening now when you have the patient in front of you, even though the AS is moderate, as opposed to having them come back for follow-up in a year or two, during which time they may have progressed to very severe AS more quickly than expected or, alternatively, experienced progressive adverse cardiac remodeling.

How do you approach patients with moderate AS and HF in whom symptoms may have multiple causes? How do you determine if the valve disease is contributing to their symptoms?

Fundamentally, it does not matter what is causing symptoms. Let’s consider patients with severe AS, which frequently coexists with HF with either reduced or preserved ejection fraction. In the setting of severe AS, it doesn’t matter what is truly causing the symptoms or the HF. The patient has a class I indication for valve replacement regardless. We know that someone with reduced cardiac function is at increased risk of poor outcomes with severe AS. They simply are not going to have the reserve to overcome the increased afterload associated with severe AS, so aortic valve replacement (AVR) is indicated regardless of whether there are symptoms and regardless of whether reduced left ventricular (LV) ejection fraction is due to the AS or preceded severe AS.

Similarly, in patients with HFpEF and severe AS, patients who have an HF syndrome are, by definition, symptomatic. Diastolic dysfunction contributes to HFpEF, but again, these factors are additive and cannot be distinguished from symptoms relating to AS. Symptoms in AS are of course fundamentally caused by failure of the myocardium in the face of progressive AS, meaning that symptomatic AS is in fact “heart failure.” It is impossible to distinguish those symptoms from symptoms that are solely attributable to AS. Frankly, if a patient with AS has symptoms, it is prudent to assume that AS is the cause given that AS will result in rapid clinical decompensation (and potentially mortality) when valve replacement is delayed. Again, these comments apply specifically to severe AS.  

Moderate AS is a little bit more complex. We don’t have an established mortality benefit for treating moderate AS, so it is not as clear how urgently we need to intervene in these patients. What the PROGRESS trial has taught us is that treating moderate AS improves QOL and markedly reduces HF events, including escalation of HF medications and hospitalization for HF. This suggests that we should think about intervention for AS in the same manner that we think about the pillars of medical optimization—intervening on a structural lesion, namely the AS, may provide similar disease-modifying effects. Ultimately, treating moderate AS contributes to optimizing the patient’s care to prevent further progression of their HF syndrome. Given the less urgent need to treat moderate AS, it is reasonable to first optimize guideline-directed medical therapy and treat the patient’s HF with medications, especially in younger patients in whom valve durability may be a concern. If symptoms persist despite doing so, it would be prudent, based on results of the PROGRESS trial and the observed safety of transcatheter aortic valve replacement (TAVR), to then consider intervening.

What factors help predict progression from moderate to severe AS?

As I previously alluded to, patients with a high valve calcium score usually progress more rapidly. There are data to suggest that patients with bicuspid AS progress more rapidly as well. Otherwise, factors such as high cholesterol, diabetes, renal dysfunction, lipoprotein(a), and genetic causes have all been associated with AS, and to some extent with the rate of progression. Longitudinal data clearly outlining the rate of AS progression are somewhat limited.

What role do LV remodeling and myocardial damage play in risk stratification? How should these findings be considered alongside traditional measures of valve severity?

They are complementary—I do not think it is necessarily prudent, for example, to assume that adverse LV remodeling, whether it’s dilatation, hypertrophy, or decline in ejection fraction, must be related to AS for AVR to be justified. Any degree of adverse remodeling suggests that something is happening to the myocardium that is impeding the heart’s normal function and structure and, presumably, reducing its reserve to then cope with AS, regardless of whether it is causal or additive.

I think we should continue to grade AS severity based on valve area, gradient, and peak velocity, but we do need to be cognizant of how the myocardium looks because, ultimately, it’s the myocardium that will cause symptoms, HF, clinical decompensation, and mortality. We need to be mindful of myocardial function alongside valve severity because that’s ultimately what drives patient outcomes.

When echocardiographic measures of AS severity are discordant, how do you determine whether a patient has truly moderate disease? When do you turn to multimodality imaging to clarify severity?

It depends on the severity of AS we’re considering. For example, when the valve area suggests severe AS, with an area ≤ 1 cm2, but the mean gradient is < 40 mm Hg and peak velocity is < 4 m/s, that is consistent with low-gradient AS. A low stroke volume (stroke volume index < 35 mL/m2) of course suggests low-flow, low-gradient AS. In these patients, we don’t know whether they truly have severe or moderate, pseudosevere AS. In patients who also have a low ejection fraction (ie, classic low-flow, low-gradient AS), dobutamine stress echocardiography can be used to distinguish between truly severe or pseudosevere AS.

On the other hand, in patients with normal ejection fraction and paradoxical low-flow, low-gradient AS, I usually rely more on calcium score. If the calcium score is within the range to suggest severe AS, then that’s how we will ultimately grade the severity. If some degree of uncertainty remains, I rely on invasive hemodynamics, also keeping in mind that if a patient is hypertensive with paradoxical low-gradient AS, treating the elevated blood pressure acutely in the cath lab during invasive assessment can help determine AS severity. If the valve area increases when the patient is normotensive, the patient has moderate, pseudosevere AS.

In some regards, the PROGRESS trial has made this process easier. Now that there is merit to treating both moderate and severe AS, the distinction between the two may become less critical to patient care decisions. Ultimately, significant AS may simply be defined by the presence of symptoms or cardiac damage.

Having said that, there can also be discordant echocardiographic measures of moderate AS. There are patients with a valve area between 1 and 1.5 cm2, a mean gradient < 20 mm Hg, and a peak velocity < 3 m/s—essentially low-gradient, moderate AS.

This concept was reported in the TAVR UNLOAD trial, in which some patients with low ejection fraction underwent dobutamine stress echocardiography, similar to our approach in severe AS, to determine if augmenting cardiac output increased the valve area to help distinguish between truly moderate and pseudomoderate mild AS.2

The same complexities involving flow, valve area, and gradients exist at the lower end of the moderate AS spectrum, and they can help differentiate between patients with mild disease and those with moderate disease.

How can physicians reliably follow patients with moderate disease so progression and clinical deterioration are not missed? What does appropriate surveillance look like in practice?

Current guideline recommendations suggest that we should be performing surveillance echocardiography and clinical assessments every 1 to 2 years in patients with moderate AS and every 6 to 12 months in patients with severe AS.

Many of us are advocating for a more proactive approach to surveillance that goes beyond echocardiography and clinical assessment to optimize the patient’s medical condition and prepare for the inevitable need for valve replacement. Depending on the severity of AS and whether the patient is likely to undergo AVR within the next 6 to 12 months, this means completing the evaluation for AVR, whether surgical or transcatheter, so that rapid intervention can be safely performed once clinically needed. Specifically, this could include dental clearance, obtaining a TAVR CT if the patient is expected to undergo TAVR, arranging a surgical evaluation, or obtaining a coronary angiogram. The goal is to perform the AVR evaluation before the patient actually needs it, so that if symptoms develop, everything is in place to proceed as quickly as possible.

What have we learned from TAVR UNLOAD and the recently published and presented PROGRESS trial1,2 about the role of TAVR in moderate AS and the importance of patient selection? What questions remain to be answered by ongoing trials?

Both TAVR UNLOAD and PROGRESS strongly suggest that patients with moderate AS and at-risk features (eg, evidence of HF syndrome, whether with reduced or preserved ejection fraction; diastolic dysfunction; AF; high calcium score) benefit from valve intervention in terms of QOL improvement and reduction in HF events. As TAVR UNLOAD was a small trial, it was not able to show improvement in HF events or mortality. However, very robust, consistent improvements in QOL were demonstrated in both trials. We also hope to soon have results from the EXPAND TAVR II trial of moderate AS sponsored by Medtronic.

In terms of what we still need to learn, there will always be questions as to whether the criteria used to select patients in these trials are indicative of truly at-risk features and whether they all benefit from intervention or whether certain subgroups perhaps benefit more or less. The PROGRESS trial so far suggests consistent benefits across all subgroups.

We are also going to face capacity concerns for some of our programs, so there may be a need to further risk stratify patients. Learning which features are more indicative of urgent intervention versus which are less urgent will be an important next step.

As I alluded to before, we did not see a mortality benefit. There is an argument that perhaps the reason we are not seeing a mortality benefit in the PROGRESS and EARLY TAVR trials is that they are inherently TAVR trials. Specifically, a patient could only be enrolled after the TAVR evaluation was completed and the patient was deemed suitable for transfemoral TAVR. As a result, the conversion to AVR once symptoms developed or AS progressed to severe was accelerated. I think the delays in care that happen during routine clinical practice ultimately lead to adverse outcomes such as mortality. Because these were TAVR trials and the necessary data were collected ahead of time, we essentially operationalized the proactive management approach that many of us are now advocating for.

This is not what happens in the real world. In routine practice, clinicians wait for a patient to develop symptoms and then refer the patient to the valve team (so-called “watchful waiting”). The patient then undergoes dental clearance, a CT scan, cardiac cath, and a surgical evaluation. These assessments take time, so the patient may not undergo AVR until 6 months later. During this period, patients are living with symptoms while awaiting intervention, and unfortunately, many experience adverse outcomes, including death. In EARLY TAVR and PROGRESS,1,3 that delay lessened substantially because the full evaluation was completed before patients were randomized.

What do you think will ultimately change how we define and manage moderate AS (better identification of high-risk patients, new thresholds for intervention, etc)? What is one piece of clinical evidence you think could meaningfully change clinical practice?

Now that we see the clinical benefit of treating patients with moderate AS, the complexity and need to definitively distinguish between moderate and severe AS are lessened. Ultimately, if a patient is symptomatic, whether it is moderate or severe AS is less relevant. You can treat that patient knowing that you are helping them either way. This may streamline the evaluation a bit and reduce the time spent considering whether the AS is moderate or severe. This is a big step forward, and it will ultimately change our thought process to determining whether a patient has significant AS and less about the moderate/severe distinction.

1. Généreux P, Makkar RR, Pibarot P, et al; PROGRESS trial investigators. TAVR in patients with moderate aortic stenosis with at-risk features. N Engl J Med. Published online November 2, 2026. doi: 10.1056/NEJMoa2607326

2. Van Mieghem NM, Elmariah S, Spitzer E, et al. Transcatheter aortic valve replacement in patients with systolic heart failure and moderate aortic stenosis: TAVR UNLOAD. J Am Coll Cardiol. 2025;85:878-890. doi: 10.1016/j.jacc.2024.10.070

3. Généreux P, Schwartz A, Oldemeyer JB, et al; EARLY TAVR trial investigators. Transcatheter aortic-valve replacement for asymptomatic severe aortic stenosis. N Engl J Med. 2025;392:217-227. doi: 10.1056/NEJMoa2405880

Sammy Elmariah, MD, MPH
Division of Cardiology
University of California, San Francisco
San Francisco, California
sammy.elmariah@ucsf.edu
Disclosures: Consultant for Edwards Lifesciences; receives research grant funding from Edwards Lifesciences, Medtronic, and Abbott.