Moderate Aortic Stenosis
Definition
Moderate aortic stenosis is calcific or rheumatic valve obstruction that does not meet severe criteria: an aortic-valve area of about 1.0–1.5 cm², a peak velocity of 3.0–3.9 m/s, or a mean gradient of 20–39 mmHg. It is a heterogeneous label. Most patients follow an indolent course and are managed with echocardiographic surveillance every 1–2 years, but a subset with symptoms, adverse structural or functional features, or a high calcium score progresses rapidly and behaves like severe disease. The clinical problem is therefore not the average patient with moderate stenosis but the identification of the at-risk phenotype within it. (sources/TAVR-ModAS-NEJM-2026, rating: very high; sources/vhd-esc-2025, rating: very high)
Key Concepts
Grading and the Limits of the Label
| Parameter | Moderate range | Caveat |
|---|---|---|
| Aortic-valve area | 1.0–1.5 cm² | Area is flow-dependent; low-flow states can make moderate disease look severe and vice versa |
| Peak velocity | 3.0–3.9 m/s | Discordance between area and velocity is common and requires a full echocardiographic assessment |
| Mean gradient | 20–39 mmHg | Gradient depends on transvalvular flow and ventricular function |
- Concordance matters. In PROGRESS, 91.6% of evaluable patients had concordant moderate disease, with both area and velocity or gradient in the moderate range; the remainder had discordant measurements. (sources/TAVR-ModAS-NEJM-2026, rating: very high)
- Severity is frequently misclassified before referral. Approximately 16% of patients who consented to PROGRESS after being thought to have moderate stenosis by referring physicians were ultimately found to have lesser or greater disease severity. (sources/TAVR-ModAS-NEJM-2026)
- Calcium scoring helps arbitrate discordance. An aortic-valve calcium score above 1200 Agatston units in women or 2000 in men was one of the at-risk features used for trial entry, and the same thresholds are used to confirm severe disease in discordant grading. (sources/TAVR-ModAS-NEJM-2026; sources/vhd-esc-2025)
At-Risk Features
PROGRESS defined an at-risk feature as a symptom or structural or functional abnormality associated with faster progression or adverse cardiovascular outcomes. At least one was required for enrolment.
| Feature | Threshold or definition | Prevalence in PROGRESS |
|---|---|---|
| Valve-related symptoms | NYHA class ≥II, chest pain or syncope | 96.4% |
| Reduced LV ejection fraction | <60% | ~25% |
| Diastolic dysfunction | Grade ≥2 by ASE criteria | 43.8% |
| Low stroke volume index | <35 mL/m² | 8.4–14.2% |
| Recent atrial fibrillation | Persistent or paroxysmal within 6 months | 18.3–20.9% |
| Elevated natriuretic peptide | NT-proBNP >3× upper limit of normal | 20.6–22.1% |
| Elevated aortic-valve calcium | >1200 Agatston units in women, >2000 in men | 23.2–28.2% |
(sources/TAVR-ModAS-NEJM-2026, rating: very high)
- The features cluster. 72% of trial participants had at least two at-risk features and 40% had at least three.
- Symptoms dominate the phenotype. Most enrolled patients were symptomatic despite a preserved ejection fraction; the trial therefore tested earlier intervention in a population that was not truly asymptomatic.
- The features have not been prospectively validated as a treatment threshold. They were selected from observational data associating them with progression and events; PROGRESS validates the composite phenotype, not each individual criterion.
The PROGRESS Trial
| Element | Detail |
|---|---|
| Population | 744 patients aged ≥65 years with moderate AS and ≥1 at-risk feature; mean age 78.4; 39.1% women; STS-PROM 3.0%; 45.6% low surgical risk |
| Intervention | Transfemoral TAVR with SAPIEN 3, SAPIEN 3 Ultra or SAPIEN 3 Ultra RESILIA versus clinical surveillance |
| Primary endpoint | Death from any cause or heart-failure event, with or without hospitalization, at 2 years |
| Primary result | 16.7% versus 23.9%; risk difference −7.2 percentage points (95% CI −13.1 to −1.3); HR 0.67 (0.48–0.93); P=0.02 |
| Death | 8.4% versus 7.3%; HR 1.14 (0.67–1.93) |
| Heart-failure event | 10.8% versus 19.0%; HR 0.55 (0.37–0.81) |
| HF hospitalization | 7.4% versus 14.9%; difference −7.5 percentage points |
| Primary safety endpoint | 5.9% at 30 days, below the 13.7% performance goal (P<0.001) |
| Surveillance crossover | 27.0% underwent AVR by 1 year and 57.4% by 2 years; 81.4% for progression to severe AS |
(sources/TAVR-ModAS-NEJM-2026, rating: very high)
Secondary Outcomes and Health Status
- The hierarchical secondary endpoint narrowly missed significance. Death, stroke or HF hospitalization occurred in 17.7% versus 23.3% (difference −5.6 percentage points, 95% CI −11.5 to 0.3; P=0.07), so formal hierarchical testing stopped there.
- Health status improved more with TAVR. A sustained ≥10-point KCCQ-OS improvement at 1 and 2 years occurred in 39.0% versus 23.2%; NYHA class III–IV at 2 years was 3.9% versus 10.9%.
- LV mass and diastolic function did not separate. LV mass index was 97.1 versus 101.1 g/m², and grade ≥2 diastolic dysfunction was 38.9% versus 36.5%.
- No difference in myocardial infarction or stroke was apparent.
Progression during Surveillance
- Patients in the surveillance arm who progressed to valve replacement lost ground while waiting: KCCQ-OS fell 9.3 points, peak velocity rose 0.5 m/s, mean gradient rose 10.5 mmHg and valve area fell 0.3 cm² from baseline to the time of replacement.
- Of the 66 surveillance patients with a heart-failure event contributing to the primary endpoint, 28 underwent delayed AVR with hospitalization for heart failure as treatment; 21 of those 28 had a class 1 indication.
- The trial therefore compares early TAVR with a surveillance strategy that eventually intervenes in most patients, not with indefinite medical management.
Procedural Safety
- Thirty-day mortality after index TAVR was 0%; stroke 0.8%; new permanent pacemaker 13.0%; moderate or greater paravalvular regurgitation 0%.
- After delayed AVR in the surveillance arm, thirty-day mortality was 0.5%, stroke 2.5% and new pacemaker 10.4%.
- The primary safety endpoint was met against a prespecified performance goal rather than against the surveillance arm, so it should be read as a benchmark comparison.
Relation to Other Trials
- TAVR UNLOAD enrolled patients with reduced LVEF (20–50%) and moderate AS and was not powered for its primary endpoint; PROGRESS enrolled a predominantly preserved-LVEF population selected by at-risk features. The discordance between the two trials is best explained by population and power, not by a contradiction in mechanism. (sources/TAVR-ModAS-NEJM-2026)
- EARLY TAVR tested asymptomatic severe AS, not moderate AS; its benefit was also driven by hospitalization rather than death. The two trials together shift the intervention threshold downward, but in different severity strata. See concepts/Aortic-Stenosis.
Contradictions / Open Questions
- The guidelines have not yet incorporated PROGRESS. Both ESC 2025 and ACC/AHA 2020 recommend surveillance for moderate AS, with valve replacement deferred until severe disease or another indication. PROGRESS now provides a positive randomized result in a selected at-risk subgroup, so the guideline position is expected to change but currently does not reflect the trial. (sources/TAVR-ModAS-NEJM-2026, rating: very high; sources/vhd-esc-2025, rating: very high; sources/VHD-AHA-2020, rating: very high)
- The primary endpoint is driven by heart-failure events, not death. The composite was positive (HR 0.67) while all-cause death was numerically higher with TAVR (HR 1.14). Whether earlier intervention changes survival is unanswered, and the mortality direction is a caution rather than a reassurance.
- The surveillance arm was not a true control. 57.4% underwent AVR within two years, mostly for progression to severe disease. The trial therefore estimates the benefit of earlier versus later intervention, and the true effect versus indefinite surveillance may be larger or smaller depending on how many patients would never progress.
- The at-risk features are a package, not a validated rule. The trial does not establish which single feature justifies intervention, whether combinations differ, or whether a patient with moderate AS and no at-risk feature benefits. Applying the result to every moderate-AS patient would treat a much larger and lower-risk population than was studied.
- The 16% misclassification rate is a systems problem. Referral physicians frequently misjudge severity, which means some patients labelled "moderate" are actually severe and some are milder. The trial supports referral to a dedicated valve clinic, but does not define how community practice should implement that.
- Bicuspid valves were included but not analysed as a subgroup. Bicuspid aortic valve was present in 7.4%, and the trial was not powered to determine whether the result applies equally to bicuspid and trileaflet anatomy.
- Durability and reintervention are unresolved. Earlier TAVR exposes younger patients to a bioprosthesis for longer; PROGRESS followed patients for two years, and the trade-off between earlier symptom benefit and lifetime valve durability cannot be settled within that horizon.
- The safety comparison is against a performance goal, not the control arm. The 5.9% thirty-day event rate was below the prespecified 13.7% goal, but the trial did not randomize safety outcomes against surveillance, so the comparison is indirect.
Connections
- Part of concepts/Aortic-Stenosis — the parent overview of grading, staging and indications
- Part of concepts/Valvular-Heart-Disease — the broader valve disease framework
- concepts/TAVI — the procedure, platforms and trial landscape
- concepts/HFpEF-Management — heart-failure events in a predominantly preserved-LVEF population
- entities/Bicuspid-Aortic-Valve — the younger-patient morphology within the moderate-AS population
- concepts/Structural-Valve-Deterioration — the durability trade-off of earlier intervention