Aortic Stenosis
Definition
Aortic stenosis (AS) is progressive obstruction of left ventricular outflow from calcification and fibrosis of the aortic cusps, or from rheumatic involvement, producing a fixed afterload the ventricle cannot offload. Untreated it leads to myocardial dysfunction and death, and it is the commonest primary valve lesion referred for intervention in Europe and North America.
Two measurements organise everything that follows: the mean gradient and the transvalvular flow. Each grading pattern below is a combination of the two, and the question in each case is whether the ventricle is working against a fixed orifice or whether the measured area is an artefact of low flow.
Abbreviations: AS, aortic stenosis; AVA, aortic valve area; AVAi, indexed AVA; CCT, cardiac computed tomography; DSE, dobutamine stress echocardiography; GLS, global longitudinal strain; LVEF, left ventricular ejection fraction; PG, pressure gradient; SAVR, surgical aortic valve replacement; SVi, indexed stroke volume; TAVI, transcatheter aortic valve implantation; Vmax, peak aortic jet velocity.
Key Concepts
Epidemiology, Aetiology and Prognosis
- Calcific degeneration dominates in high-income countries and prevalence rises with age; rheumatic AS remains common in low- and middle-income countries and usually coexists with rheumatic mitral disease. (sources/vhd-esc-2025, rating: very high)
- Bicuspid aortic valve is the dominant morphology in younger patients reaching replacement, and more than half have associated aortic dilatation. (sources/vhd-esc-2025)
- Prognosis turns on velocity. Once Vmax reaches 4 m/s, event-free survival is only 30–50% at two years, yet sudden death under prospective surveillance is below 1% per year. Those two figures are why watchful waiting is safe month to month and untenable across years — and why the timing trials below were run at all. (sources/VHD-AHA-2020, rating: very high)
Valve Pathology and Mechanism
- The lesion begins as an endothelial disease. Shear stress disrupts the valvular endothelium; lipid infiltration, inflammation and oxidative stress follow; interstitial cells undergo osteogenic transformation; matrix disruption and extracellular-vesicle calcification complete it. Calcification starts in the fibrosa, usually at the noncoronary cusp first. (sources/vhd-mechanism-aha-2024, rating: very high)
- Initiation and progression behave as different diseases. Mendelian randomisation implicates Lp(a) and LDL-C in initiating CAVD, but Lp(a) does not track progression of established calcification. That dissociation is the standing explanation for the statin paradox — strong genetic evidence for lipids, no benefit from statins in SALTIRE. (sources/vhd-mechanism-aha-2024)
- Clonal haematopoiesis of indeterminate potential is present in CAVD, augments NLRP3 inflammasome signalling and predicts worse outcome after replacement. (sources/vhd-mechanism-aha-2024)
- The full cascade, the sex differences and the six ongoing progression trials — pelacarsen, colchicine, PCSK9 inhibition, evogliptin, ARBs and pioglitazone — are set out on concepts/CAVD-Mechanisms. None has yet changed the natural history.
Grading and Staging
| Pattern | Criteria | ACC/AHA stage | What it means |
|---|---|---|---|
| High-gradient severe | Mean PG ≥40 mmHg, Vmax ≥4.0 m/s, AVA ≤1 cm² (AVAi ≤0.6 cm²/m²) | D1 if symptomatic, C1 if not | Severe regardless of flow or LVEF; management rests on symptoms and risk, not on more measurement |
| Very severe | Vmax ≥5 m/s or mean PG ≥60 mmHg | Subset of C1/D1 | Weighs toward early intervention in the asymptomatic patient |
| Low-flow, low-gradient, reduced LVEF | AVA ≤1 cm², SVi ≤35 mL/m², mean PG <40 mmHg, LVEF <50% | D2 | True severe versus pseudo-severe is the whole question; needs DSE or CCT |
| Low-flow, low-gradient, preserved LVEF | AVA ≤1 cm², SVi ≤35 mL/m², mean PG <40 mmHg, LVEF ≥50% | D3 (paradoxical) | Often elderly women with small, stiff ventricles; multimodality workup before committing |
| Normal-flow, low-gradient, preserved LVEF | AVA ≤1 cm² with normal flow and mean PG <40 mmHg | Not staged | Usually moderate disease; surveillance unless multimodality confirms severe |
(sources/vhd-esc-2025, rating: very high; sources/VHD-AHA-2020, rating: very high)
Resolving Discordant Grading
- Dobutamine stress echocardiography separates true from pseudo-severe disease in the reduced-LVEF pattern: in true severe AS the gradient rises while AVA stays ≤0.7 cm², whereas in pseudo-severe AS the gradient does not change and AVA normalises to ≥1.2 cm². Flow reserve — a ≥20% rise in stroke volume — also stratifies operative risk. (sources/hemodynamics-circ-2012, rating: high)
- CCT calcium scoring addresses the preserved-LVEF pattern: above 2000 AU in men or 1200 AU in women is severe with roughly 85% sensitivity and specificity, and below 1600 AU in men or 800 AU in women makes severe AS unlikely. (sources/vhd-esc-2025)
- The female threshold is guideline-dependent. ESC 2025 sets it at 1200 AU and ACC/AHA 2020 at 1300 AU, so a woman scoring 1250 AU is severe in Europe and not severe in the United States. The difference is immaterial at the bedside, but it means the cut-off is not a single agreed number. (sources/VHD-AHA-2020)
- Nitroprusside challenge addresses the preserved-LVEF pattern: high peripheral resistance can suppress stroke volume and therefore the gradient, and lowering afterload uncovers a fixed area with a rising gradient. (sources/hemodynamics-circ-2012)
- Adjuncts to the decision: GLS worse than −15% identifies high-risk asymptomatic patients, and a natriuretic peptide above three times the age- and sex-corrected normal on repeated measurement is an adverse marker that also helps arbitrate the source of symptoms. (sources/vhd-esc-2025)
- Exercise testing confirms that the patient is genuinely asymptomatic and unmasks a fall in blood pressure of more than 20 mmHg — but it is contraindicated in symptomatic severe AS (Class III: Harm). (sources/VHD-AHA-2020)
- Screen for transthyretin amyloidosis with immunofixation and bone scintigraphy when it is suspected in an elderly patient; valve intervention remains beneficial when ATTR coexists. (sources/vhd-esc-2025)
Invasive Haemodynamic Assessment
- Catheterisation is reserved for discordance between Doppler and the clinical picture, or for very high flow (cardiac output >6.5 L/min). If the mean gradient is ≥40 mmHg and the findings agree, no further invasive data are needed. (sources/hemodynamics-circ-2012, rating: high)
- Technique: simultaneous LV and central aortic pressures through side-hole catheters, taking the mean gradient across the systolic ejection period. Peak-to-peak gradient is not physiological — peak LV and peak aortic pressure occur at different moments. (sources/hemodynamics-circ-2012)
- Never use femoral artery pressure as a surrogate for central aortic pressure. Peripheral amplification and iliofemoral stenosis distort the value, and the temporal delay distorts the mean. (sources/hemodynamics-circ-2012)
- Carabello sign: crossing a critical valve (AVA ≤0.7 cm²) with a 7–8F catheter adds obstruction, seen as a further pressure drop on pullback. Do not read it as valve pathology. (sources/hemodynamics-circ-2012)
- Gorlin formula (1951): A = F/(Cc × Cv × √2gh). The empirical discharge coefficient applies to the mitral valve only; both coefficients are assumed to equal 1 for the aortic valve, which is a theoretical impossibility. Valve area is therefore always read alongside gradient, contour and contractile state. (sources/hemodynamics-circ-2012)
- Hakki check: valve area = cardiac output / √mean gradient — a bedside way to verify a computer-generated Gorlin result. (sources/hemodynamics-circ-2012)
- Pressure contour separates the two obstructions: fixed valvular AS gives a parvus et tardus upstroke from the moment the valve opens; dynamic LVOTO in HCM gives a spike-and-dome, late-peaking LV pressure. (sources/hemodynamics-circ-2012)
- Braunwald–Brockenbrough sign: after a post-extrasystolic beat, pulse pressure rises in valvular AS and falls in HCM. (sources/hemodynamics-circ-2012)
- Output measurement is the weak link: assumed oxygen-consumption tables introduce up to 40% error in the Fick calculation, and thermodilution is unreliable in low output, tricuspid regurgitation, arrhythmia and shunts. (sources/hemodynamics-circ-2012)
Medical Therapy
- No therapy has been shown to delay progression; statins failed despite preclinical promise. (sources/vhd-esc-2025, rating: very high)
- Treat concomitant hypertension, preferring renin–angiotensin blockade with careful titration to avoid hypotension.
- In heart failure with reduced ejection fraction, start guideline-directed medical therapy before intervention and up-titrate after it.
Indications — Symptomatic Severe AS
| Scenario | ESC 2025 | Action |
|---|---|---|
| High-gradient severe AS | Class I | Intervene in every eligible patient with life expectancy >1 year |
| Low-flow, low-gradient, reduced LVEF, severe confirmed | Class I B | Intervene |
| Low-flow, low-gradient, reduced LVEF, pseudo-severe | — | No intervention; guideline-directed medical therapy |
| Low-flow, low-gradient, preserved LVEF, severe confirmed | Class IIa | Intervene after multimodality confirmation |
| Normal-flow, low-gradient, preserved LVEF | — | Usually moderate AS; surveillance unless multimodality confirms severe |
(sources/vhd-esc-2025, rating: very high)
Indications — Asymptomatic Severe AS
| Criterion | ESC 2025 | ACC/AHA 2020 |
|---|---|---|
| LVEF <50% without another cause | Class I | Class I B-NR |
| Confirmed asymptomatic, high-gradient, LVEF ≥50%, normal exercise test, low procedural risk | Class IIa (new 2025) | No equivalent — one named adverse feature is required |
| Vmax ≥5 m/s with low surgical risk | Adverse feature | Class IIa B-R |
| Progression ≥0.3 m/s per year with low risk | Adverse feature | Class IIa B-NR |
| Natriuretic peptide >3× age/sex normal on repeated measurement, low risk | Adverse feature | Class IIa B-NR |
| Exercise-induced symptoms or ≥10 mmHg blood pressure fall, low risk | — | Class IIa B-NR |
| LVEF falling to <60% on serial studies | Adverse feature (LVEF <55%) | Class IIb B-NR |
(sources/vhd-esc-2025, rating: very high; sources/VHD-AHA-2020, rating: very high)
ESC 2025 does not grade the adverse features separately: they reinforce the Class IIa rather than creating new indications, and each is a reason to move the balanced patient toward intervention.
Early Intervention in Asymptomatic AS — the Trial Evidence
| EARLY TAVR (NEJM 2025) | RECOVERY (NEJM 2026) | |
|---|---|---|
| Population | 901 patients; mean age 75.8; STS-PROM 1.8%; LVEF ≥50% (mean 67.4%); negative stress test in 90.6%; peak velocity 4.3 m/s | 145 patients with very severe AS (AVA ≤0.75 cm², Vmax ≥4.5 m/s); Korean high-volume centres; age >80 and major comorbidity excluded |
| Comparison | Transfemoral SAPIEN 3/3 Ultra versus guideline-based surveillance | Early SAVR versus conservative care |
| Follow-up | Median 3.8 years | Median 12 years |
| Primary endpoint | Death, stroke or unplanned cardiovascular hospitalisation: 26.8% vs 45.3%, HR 0.50 (0.40–0.63) | Operative mortality or cardiovascular death: 3% vs 24%, HR 0.10 (0.02–0.43) |
| All-cause mortality | 8.4% vs 9.2%, HR 0.93 — not significant | 15% vs 32%, HR 0.42 (0.21–0.86) |
| What drove the result | Unplanned hospitalisation, 20.9% vs 41.7% (HR 0.43) | Cardiovascular death; no heart-failure admission in the early-surgery arm |
| Principal limitation | 87.0% of the surveillance arm reached AVR at a median 11.1 months | Small, single-country, surgery only |
(sources/tavi-earlytavr-nejm-2025, rating: very high; sources/as-recovery-nejm-2026, rating: very high)
- EARLY TAVR's secondary outcomes pointed the same way: a favourable outcome at two years (alive, KCCQ ≥75 without a ≥10-point fall) in 86.6% vs 68.0%, and integrated LV/LA health in 48.1% vs 35.9% — subclinical cardiac damage accumulates during surveillance. (sources/tavi-earlytavr-nejm-2025)
- Surveillance was not benign: NT-proBNP rose from 298.6 to 462.2 pg/mL, KCCQ fell 14.8 points, the proportion with LVEF ≤60% rose from 12.7% to 20.7% and six-minute walk distance fell 46.4 m. Of those who converted to AVR, 39.2% did so with advanced symptoms, and 11 died first.
- RECOVERY's benefit did not attenuate — the Kaplan–Meier curves continued to diverge across ten years — and as-treated and per-protocol analyses agreed with intention-to-treat. (sources/as-recovery-nejm-2026)
- Read together, the two trials support intervening before symptoms in severe and very severe AS. Only EARLY TAVR tested a transcatheter strategy, and only RECOVERY followed patients long enough to see a mortality effect.
Mode of Intervention
| Scenario | ESC 2025 | ACC/AHA 2020 |
|---|---|---|
| Age <65 years or life expectancy >20 years | Heart Team | Class I A: SAVR recommended |
| Age 65–80 years, transfemoral access suitable | Heart Team | Class I A: either, after shared decision-making |
| Age >80 years or life expectancy <10 years | Heart Team | Class I A: transfemoral TAVI in preference to SAVR |
| High or prohibitive surgical risk, predicted post-procedural survival >12 months | Heart Team | Class I A: TAVI regardless of age |
(sources/vhd-esc-2025, rating: very high; sources/VHD-AHA-2020, rating: very high)
ESC 2025 integrates anatomy, life expectancy, procedural risk and an informed patient preference into that decision, with durability data favouring SAVR at the younger end; ACC/AHA 2020 fixes the answer by age. (sources/vhd-esc-2025; sources/VHD-AHA-2020)
- Anatomy pushes toward surgery where the root is narrow, the coronary ostia are low, commissural alignment is unfavourable, the LVOT is heavily calcified or the valve is bicuspid. TAVI for bicuspid stenosis at high surgical risk is Class IIb B in ESC 2025 where the Heart Team judges the anatomy suitable. (sources/vhd-esc-2025)
- Prosthesis type by age: ESC 2025 recommends a bioprosthesis for most patients ≥65 years (aortic) or ≥70 (mitral), reserving a mechanical valve for those <60 (aortic) or <65 (mitral) with long life expectancy and no contraindication to anticoagulation. ACC/AHA 2020 sets the mechanical valve at <50 years, individualises 50–65 and calls a bioprosthesis reasonable above 65 — so a 55-year-old with an aortic indication is a judgement call in Europe and explicitly undecided in the United States. (sources/vhd-esc-2025; sources/VHD-AHA-2020)
- TAVI is non-inferior to SAVR in low-, intermediate- and high-risk symptomatic patients at mid-term follow-up. The trade-off is consistent across PARTNER 3 and DEDICATE-DZHK6: TAVI wins on new-onset atrial fibrillation, major bleeding and length of stay; surgery wins on pacemaker implantation, paravalvular leak and vascular complications. Trial-level detail is tabulated on concepts/TAVI. (sources/tavr-dedicate-nejm-2024, rating: very high; sources/tavr-partner3-5yr-nejm-2023, rating: very high)
- Small annulus (≤430 mm², up to a third of TAVI candidates and overwhelmingly women): SMART showed that a self-expanding supraannular valve produces a lower mean gradient (7.7 vs 15.7 mmHg), a larger effective orifice area (1.99 vs 1.50 cm²) and less prosthesis–patient mismatch at one year, at the cost of more pacemaker implantation (12.1% vs 7.8%), with the clinical composite non-inferior. See concepts/TAVI for the table and concepts/Structural-Valve-Deterioration for how dysfunction is defined. (sources/tavr-smart-nejm-2024, rating: very high)
Concomitant Disease at Other Cardiac Surgery
- Moderate AS when operating on another valve: SAVR is Class IIa. (sources/vhd-esc-2025, rating: very high)
- Moderate AS at the time of CABG or ascending aortic surgery: SAVR is Class IIb.
- Asymptomatic severe AS in a patient already undergoing cardiac surgery for another reason: replacement is Class I B-NR. (sources/VHD-AHA-2020)
Perioperative Management
- Severe symptomatic AS is an active cardiac condition: it requires cardiac evaluation before noncardiac surgery, and elective surgery should be deferred until the valve is treated. Balloon valvuloplasty is a bridge only when surgery cannot wait and TAVI cannot be arranged. (sources/periop-aha-2024, rating: very high)
- The graded recommendations, the rule for the asymptomatic patient and the local handbook thresholds are tabulated on concepts/Perioperative-Cardiovascular-Assessment.
- Haemodynamic goals during anaesthesia: avoid hypotension (MAP <65 mmHg), tachycardia and preload reduction, and preserve sinus rhythm and systemic vascular resistance so coronary perfusion survives the fixed obstruction.
Contradictions / Open Questions
- ESC 2025 and ACC/AHA 2020 gate early intervention differently. ESC 2025 gives a single Class IIa for asymptomatic high-gradient AS with LVEF ≥50%, a normal exercise test and low procedural risk. ACC/AHA 2020 instead lists five separate Class IIa/IIb criteria — very severe disease, rapid progression, raised natriuretic peptide, exercise-induced symptoms and an LVEF fall below 60% — and requires one of them. The European route reaches an earlier decision for the typical patient; the American route demands a named adverse feature first. Both rest on the same four trials. (sources/vhd-esc-2025, rating: very high; sources/VHD-AHA-2020, rating: very high)
- The guidelines also differ on mode of intervention. ACC/AHA 2020 uses fixed age bands (SAVR below 65, either 65–80, transfemoral TAVI above 80), which are reproducible and auditable. ESC 2025 makes every case a Heart Team judgement, which is more flexible and cannot be audited. A 78-year-old with unfavourable anatomy is defaulted to TAVI under one framework and referred for a judgement under the other. (sources/VHD-AHA-2020; sources/vhd-esc-2025)
- EARLY TAVR's primary endpoint is a composite in which death carried no weight. The HR of 0.50 is driven by unplanned cardiovascular hospitalisation (HR 0.43), including conversions to AVR within six months counted as events; death (HR 0.93) and stroke (HR 0.62) were individually not significant. With 87% crossover at a median 11.1 months, the trial compares immediate with delayed TAVI, not TAVI with indefinite surveillance. (sources/tavi-earlytavr-nejm-2025, rating: very high)
- EARLY TAVR and RECOVERY disagree on mortality. All-cause mortality was neutral in EARLY TAVR (HR 0.93) and halved in RECOVERY (HR 0.42). Follow-up duration (3.8 versus 12 years), disease severity (Vmax 4.3 versus ≥4.5 m/s), modality and crossover rate plausibly explain the gap, but no trial has yet shown that early TAVI extends life — the mortality signal so far belongs to surgery in very severe disease. (sources/as-recovery-nejm-2026, rating: very high; sources/tavi-earlytavr-nejm-2025)
- RECOVERY will not transfer unchanged. A single-country, 145-patient trial in high-volume Korean centres that excluded patients over 80 and those with major comorbidity produced an NNT of 6. The same strategy applied to a frail 85-year-old with competing mortality risks yields a different arithmetic that no trial has tested. (sources/as-recovery-nejm-2026)
- The stroke direction in EARLY TAVR is unexplained. Stroke was 4.2% with early TAVI against 6.7% with surveillance (HR 0.62, 95% CI 0.35–1.10) — directionally favourable and underpowered. Fewer calcific emboli from an untreated valve and cumulative procedural exposure in the surveillance arm are both plausible and both untested. (sources/tavi-earlytavr-nejm-2025)
- Moderate AS with heart failure and reduced ejection fraction remains unproven. One small underpowered randomised trial supports intervention; it is not enough for a routine recommendation, and every guideline says so.
- Whether the asymptomatic result generalises to every low-risk patient is unsettled. EARLY TAVR enrolled a healthy group — negative stress test in 90.6%, STS-PROM 1.8%, mean peak velocity 4.3 m/s — in which 87% reached surgery within a year anyway. Applying the finding to a 60-year-old with a velocity of 4.1 m/s and no other feature would trade surveillance for decades of prosthesis exposure, and the durability data behind that trade extend only to five years.
Connections
- Part of concepts/Valvular-Heart-Disease — the parent overview
- concepts/TAVI — platform selection, procedural planning and the trial tables
- concepts/CAVD-Mechanisms — molecular pathogenesis and progression trials
- concepts/Structural-Valve-Deterioration — how bioprosthetic dysfunction is defined and measured
- concepts/PCI-Before-TAVI — coronary revascularisation before valve intervention
- concepts/Cerebral-Embolic-Protection-TAVI — embolic protection during TAVI
- entities/Bicuspid-Aortic-Valve — the younger-patient phenotype and aortic dilatation
- entities/ATTR-Amyloidosis — the amyloid phenotype that coexists with AS
- entities/Heart-Failure — systolic dysfunction before and after intervention
- concepts/Perioperative-Cardiovascular-Assessment — noncardiac surgery in AS
- concepts/RHC-Technique-and-Measurement — invasive haemodynamic technique and the artefacts that make it necessary
- concepts/Aortic-Regurgitation — the mixed-lesion differential
Sources
Guidelines and consensus statements
Mechanism
Diagnostic and haemodynamic assessment
Randomised trials